Current Protocols in Cytometry
Table of Contents Foreword Preface Chapter 1 Flow Cytometry Instrumentation Introduction Unit 1.1 Overview of Flow Cytometry Instrumentation Unit 1.2 Fluidics Unit 1.3 Standardization, Calibration, and Control in Flow Cytometry Unit 1.4 Establishing and Maintaining System Linearity Unit 1.5 Optical Filter Sets for Multiparameter Flow Cytometry Unit 1.6 Laser Beam Shaping and Spot Size Unit 1.7 High-Speed Cell Sorting Unit 1.8 Principles of Gating Unit 1.9 Lasers for Flow Cytometry Unit 1.10 Techniques for Flow Cytometer Alignment Unit 1.11 Flow Cytometers for Characterization of Microorganisms Unit 1.12 Principles of Resonance Energy Transfer Unit 1.13 Light Scatter: Detection and Usage Unit 1.14 Compensation in Flow Cytometry Unit 1.15 Time-Resolved Fluorescence Measurements Unit 1.16 Simultaneous Analysis of the Cyan, Green, and Yellow Fluorescent Proteins Unit 1.17 Plug Flow Cytometry Unit 1.18 Dynamic Thermoregulation of the Sample in Flow Cytometry Unit 1.19 Excitation and Emission Spectra of Common Dyes
Unit 1.20 Characterization of Flow Cytometer Instrument Sensitivity Unit 1.21 Separation Index: An Easy-to-Use Metric for Evaluation of Different Configurations on the Same Flow Cytometer
Chapter 2 Image Cytometry Instrumentation Introduction Unit 2.1 Contrast Enhancement in Light Microscopy Unit 2.2 Microscope Objectives Unit 2.3 Cameras Unit 2.4 Optical Filtering Systems for Wavelength Selection in Light Microscopy Unit 2.5 Digital Fluorescence Microscopy Unit 2.6 Calibration: Sampling Density and Spatial Resolution Unit 2.7 Microscope Alignment Unit 2.8 Confocal Microscopy: Principles and Practices Unit 2.9 Multiphoton Imaging Unit 2.10 Scanning Laser Cytometry Unit 2.11 Shading Correction: Compensation for Illumination and Sensor Inhomogeneities Unit 2.12 Photobleaching Measurements of Diffusion in Cell Membranes and Aqueous Cell Compartments Unit 2.13 Optimizing Laser Source Operation for Confocal and Multiphoton Laser Scanning Microscopy
Chapter 3 Safety Procedures and Quality Control Introduction Unit 3.1 Principles of Quality Control
Unit 3.2 Components of Quality Control Unit 3.3 Testing the Efficiency of Aerosol Containment During Cell Sorting Unit 3.4 Safe Use of Hazardous Chemicals Unit 3.5 Method for Visualizing Aerosol Contamination in Flow Sorters Unit 3.6 Standard Safety Practices for Sorting of Unfixed Cells
Chapter 4 Molecular and Cellular Probes Introduction Unit 4.1 Titering Antibodies Unit 4.2 Conjugation of Fluorochromes to Monoclonal Antibodies Unit 4.3 Nucleic Acid Probes Unit 4.4 Cellular Function Probes Unit 4.5 Spectroscopic Analysis Using DNA and RNA Fluorescent Probes
Chapter 5 Specimen Handling, Storage, and Preparation Introduction Unit 5.1 Handling, Storage, and Preparation of Human Blood Cells Unit 5.2 Handling, Storage, and Preparation of Human Tissues Unit 5.3 Flow Analysis and Sorting of Plant Chromosomes
Chapter 6 Phenotypic Analysis Introduction Unit 6.1 Quality Control in Phenotypic Analysis by Flow Cytometry Unit 6.2 Immunophenotyping
Unit 6.3 High-Sensitivity Immunofluorescence/Flow Cytometry: Detection of Cytokine Receptors and Other Low-Abundance Membrane Molecules Unit 6.4 Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells Unit 6.5 Immunophenotypic Analysis of Peripheral Blood Lymphocytes Unit 6.6 Immunophenotypic Analysis of Human Mast Cells by Flow Cytometry Unit 6.7 Measurement of CD40 Ligand (CD154) Expression on Resting and In Vitro–Activated T Cells Unit 6.8 Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques Unit 6.9 Immunophenotypic Identification, Enumeration, and Characterization of Human Peripheral Blood Dendritic Cells and Dendritic-Cell Precursors Unit 6.10 Immunophenotypic Analysis of Platelets Unit 6.11 Immunophenotypic Analysis of PNH Cells Unit 6.12 Quantitative Flow Cytometric Analysis of Membrane Antigen Expression Unit 6.13 Immunophenotyping Using a Laser Scanning Cytometer Unit 6.14 Enzymatic Amplification Staining for Cell Surface Antigens Unit 6.15 Whole Blood Analysis of Leukocyte-Platelet Aggregates Unit 6.16 Flow Cytometric Assessment of HLA Alloantibodies Unit 6.17 Enumeration of Fetal Red Blood Cells, F Cells, and F Reticulocytes in Human Blood Unit 6.18 Identification of Human Antigen-Specific T Cells Using MHC Class I and Class II Tetramers Unit 6.19 ZAP-70 Staining in Chronic Lymphocytic Leukemia Unit 6.20 Multiparameter Analysis of Intracellular Phosphoepitopes in Immunophenotyped Cell Populations by Flow Cytometry Unit 6.21 Ten-Color Immunophenotyping of Hematopoietic Cells Unit 6.22 Flow Cytometric Screening for the HLA-B27 Antigen on Peripheral Blood Lymphocytes Unit 6.23 Immunophenotyping of Plasma Cells
Unit 6.24 Flow Rate Calibration for Absolute Cell Counting Rationale and Design
Chapter 7 Nucleic Acid Analysis Introduction Unit 7.1 Overview of Nucleic Acid Analysis Unit 7.2 Quality Control in Nucleic Acid Analysis Unit 7.3 Differential Staining of DNA and RNA Unit 7.4 Analysis of DNA Content and DNA Strand Breaks for Detection of Apoptotic Cells Unit 7.5 DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis Unit 7.6 Analysis of Nuclear DNA Content and Ploidy in Higher Plants Unit 7.7 Analysis of DNA Content and BrdU Incorporation Unit 7.8 Analysis of DNA Denaturation Unit 7.9 Bivariate Analysis of DNA Content and Expression of Cyclin Proteins Unit 7.10 Flow Cytometric Analysis of Reticulated Platelets Unit 7.11 Assessment of Viability, Immunofluorescence, and DNA Content Unit 7.12 Flow Cytometric Analysis of RNA Synthesis by Detection of Bromouridine Incorporation Unit 7.13 Sperm Chromatin Structure Assay for Fertility Assessment Unit 7.14 Analysis of Cell Proliferation and Cell Survival by Continuous BrdU Labeling and Multivariate Flow Cytometry Unit 7.15 Ultraviolet-Induced Detection of Halogenated Pyrimidines (UVID) Unit 7.16 Analysis of DNA Content and Green Fluorescent Protein Expression Unit 7.17 Analysis of Viral Infection and Viral and Cellular DNA and Proteins by Flow Cytometry Unit 7.18 Apoptosis Signaling Pathways
Unit 7.19 Flow Cytometry of Apoptosis Unit 7.20 Analysis of Fine-Needle Aspirate Biopsies from Solid Tumors by Laser Scanning Cytometry (LSC) Unit 7.21 Measurement of Cytogenetic Damage in Rodent Blood with a Single-Laser Flow Cytometer Unit 7.22 Analysis of Tissue Imprints by Scanning Laser Cytometry Unit 7.23 Cell Cycle Analysis of Budding Yeast Using SYTOX Green Unit 7.24 Detection of Mitotic Cells Unit 7.25 DRAQ5 Labeling of Nuclear DNA in Live and Fixed Cells Unit 7.26 Assessment of Telomere Length, Phenotype, and DNA Content Unit 7.27 Detection of Histone H2AX Phosphorylation on Ser-139 as an Indicator of DNA Damage (DNA Double-Strand Breaks) Unit 7.28 RNA and DNA Aptamers in Cytomics Analysis Unit 7.29 Nuclear DNA Content Analysis of Plant Seeds by Flow Cytometry Unit 7.30 Estimation of Relative Nuclear DNA Content in Dehydrated Plant Tissues by Flow Cytometry Unit 7.31 Assessment of Cell Proliferation by 5-Bromodeoxyuridine (BrdU) Labeling for Multicolor Flow Cytometry
Chapter 8 Molecular Cytogenetics Introduction Unit 8.1 Overview of Fluorescence In Situ Hybridization Techniques for Molecular Cytogenetics Unit 8.2 Basic Preparative Techniques for Fluorescence In Situ Hybridization Unit 8.3 Probe Labeling and Fluorescence In Situ Hybridization Unit 8.4 Immunocytochemical Detection Unit 8.5 Processing and Staining of Cell and Tissue Material for Interphase Cytogenetics
Unit 8.6 Advanced Preparative Techniques to Establish Probes for Molecular Cytogenetics Unit 8.7 Combination DNA/RNA Fish and Immunophenotyping Unit 8.8 Single-Nucleotide Sequence Discrimination In Situ Using Padlock Probes Unit 8.9 Tyramide Signal Amplification (TSA) Systems for the Enhancement of ISH Signals in Cytogenetics Unit 8.10 Molecular Combing Unit 8.11 Principles and Applications of PRINS in Cytogenetics Unit 8.12 Comparative Genomic Hybridization (CGH)—Detection of Unbalanced Genetic Aberrations Using Conventional and Micro-Array Techniques Unit 8.13 Combined Immunofluorescence and FISH: New Prospects for Tumor Cell Detection/Identification
Chapter 9 Studies of Cell Function Introduction Unit 9.1 Overview of Functional Cell Assays Unit 9.2 Assessment of Cell Viability Unit 9.3 Flow Cytometric Measurement of Intracellular pH Unit 9.4 Analysis of Intracellular Organelles by Flow Cytometry or Microscopy Unit 9.5 Reporters of Gene Expression: Enzymatic Assays Unit 9.6 Estimation of Membrane Potential by Flow Cytometry Unit 9.7 Oxidative Metabolism Unit 9.8 Measurement of Intracellular Calcium Ions by Flow Cytometry Unit 9.9 Intracellular Cytokines Unit 9.10 Assays of Natural Killer (NK) Cell Ligation to Target Cells Unit 9.11 Flow Cytometric Analysis of Cell Division by Dye Dilution
Unit 9.12 Reporters of Gene Expression: Autofluorescent Proteins Unit 9.13 In Vitro Invasion Assays: Phagocytosis of the Extracellular Matrix Unit 9.14 Flow Cytometric Analysis of Mitochondrial Membrane Potential Using JC1 Unit 9.15 Multiparameter Analysis of Physiological Changes in Apoptosis Unit 9.16 Signal Transduction During Natural Killer Cell Activation Unit 9.17 Assessment of Surface Markers and Functionality of Dendritic Cells (DCs) Unit 9.18 Stem Cell Identification and Sorting Using the Hoechst 33342 Side Population (SP) Unit 9.19 Assessment of Phagocyte Functions by Flow Cytometry Unit 9.20 Flow Cytometric Analysis of Calcium Mobilization in Whole-Blood Platelets Unit 9.21 Flow Cytometric Analysis of Cytokine Responses in Stimulated Whole Blood: Simultaneous Quantitation of TNF-α-Secreting Cells and Soluble Cytokines Unit 9.22 Optimized Whole-Blood Assay for Measurement of ZAP-70 Protein Expression Unit 9.23 Flow Cytometry of the Side Population (SP)
Chapter 10 Data Processing and Analysis Introduction Unit 10.1 Data Management Unit 10.2 Data File Standard for Flow Cytometry, FCS 3.0 Unit 10.3 Listmode Data Processing Unit 10.4 Multidimensional Data Analysis in Immunophenotyping Unit 10.5 Two-Dimensional Image Processing and Analysis Unit 10.6 Data Presentation Unit 10.7 Data Analysis Through Modeling
Unit 10.8 Multivariate Analysis Unit 10.9 Detection and Location of Hybridization Domains on Chromosomes by Image Cytometry Unit 10.10 Three-Dimensional Image Visualization and Analysis Unit 10.11 Image Processing and 2-D Morphometry Unit 10.12 Dial-In Flow Cytometry Data Analysis Unit 10.13 The Application of Data Mining to Flow Cytometry Unit 10.14 Intensity Calibration and Shading Correction for Fluorescence Microscopes Unit 10.15 A Software Method for Color Compensation
Chapter 11 Microbiological Applications Introduction Unit 11.1 Overview of Flow Cytometry and Microbiology Unit 11.2 Flow Cytometry and Environmental Microbiology Unit 11.3 Estimation of Microbial Viability Using Flow Cytometry Unit 11.4 Sorting of Bacteria Unit 11.5 Detection of Borreliacidal Antibodies by Flow Cytometry Unit 11.6 Flow Cytometric Detection of Pathogenic E. coli in Food Unit 11.7 Mycobacterium tuberculosis Susceptibility Testing by Flow Cytometry Unit 11.8 Antibiotic Susceptibility Testing by Flow Cytometry Unit 11.9 Determination of Bacterial Biomass from Flow Cytometric Measurements of Forward Light Scatter Intensity Unit 11.10 Flow Cytometry of Yeasts Unit 11.11 Enumeration of Phytoplankton, Bacteria, and Viruses in Marine Samples Unit 11.12 DNA/RNA Analysis of Phytoplankton by Flow Cytometry
Unit 11.13 Cell Cycle Analysis of Yeasts Unit 11.14 Flow Cytometric Assessment of Drug Susceptibility in Leishmania infantum Promastigotes Unit 11.15 Resolution of Viable and Membrane-Compromised Free Bacteria in Aquatic Environments by Flow Cytometry Unit 11.16 Functional Assays of Oxidative Stress Using Genetically Engineered Escherichia coli Strains Unit 11.17 Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
Chapter 12 Cellular and Molecular Imaging Introduction Unit 12.1 Comparative Overview of Flow and Image Cytometry Unit 12.2 Basics of Digital Microscopy Unit 12.3 Modern Confocal Microscopy Unit 12.4 Time-Lapse Microscopy Approaches to Track Cell Cycle Progression at the Single-Cell Level Unit 12.5 Three-Dimensional Visualization of Blood and Lymphatic Vasculature in Tissue Whole Mounts Using Confocal Microscopy Unit 12.6 Quantitative Fluorescence In Situ Hybridization (QFISH) of Telomere Lengths in Tissue and Cells Unit 12.7 Detecting Protein–Protein Interactions with CFP-YFP FRET by Acceptor Photobleaching Unit 12.8 Measuring FRET in Flow Cytometry and Microscopy
Chapter 13 Multiplexed and Microparticle-Based Analyses Introduction Unit 13.1 Multiplexed Microsphere-Based Flow Cytometric Immunoassays Unit 13.2 Microsphere Surface Protein Determination Using Flow Cytometry
Unit 13.3 Use of Microsphere-Supported Phospholipid Membranes for Analysis of Protein-Lipid Interactions Unit 13.4 Multiplexed SNP Genotyping Using Primer Single-Base Extension (SBE) and Microsphere Arrays Unit 13.5 BeadCons: Detection of Nucleic Acid Sequences by Flow Cytometry Unit 13.6 Characterization of Nuclear Receptor Ligands by Multiplexed Peptide Interactions Unit 13.7 Detection of Gene Fusions in Acute Leukemia Using Bead Microarrays Unit 13.8 Reagents and Instruments for Multiplexed Analysis Using Microparticles
Appendix 1 Abbreviations and Useful Data 1A Abbreviations Used in this Manual 1B Common Conversion Factors
Appendix 2 Stock Solutions, Equipment, and Laboratory Guidelines 2A Common Stock Solutions, Buffers, and Media
Appendix 3 Commonly Used Techniques 3A Cell Counting 3B Techniques for Mammalian Cell Tissue Culture 3C Diagnosis and Treatment of Mycoplasma-Contaminated Cell Cultures 3D Wright-Giemsa and Nonspecific Esterase Staining of Cells 3E Techniques for Bacterial Cell Culture: Media Preparation and Bacteriological Tools 3F Growing Bacteria in Liquid Media 3G Growing Bacteria on Solid Media
3H Importing Biological Materials 3I Production of Polyclonal Antisera 3J Production of Monoclonal Antibodies 3K Enzymatic Amplification of DNA by PCR: Standard Procedures and Optimization
Appendix Suppliers Selected Suppliers of Reagents and Equipment
CHAPTER 1 Flow Cytometry Instrumentation INTRODUCTION
T
he purpose of this chapter is to acquaint readers with the instrumentation utilized in flow cytometry by describing various elements of the technology, and to provide detailed information on specific methods of using the instruments. Introducing the chapter is UNIT 1.1, an overview of the instrumentation involved in flow cytometry. This unit describes the various parts of a flow cytometer, explains the basics of how each component works, and presents a brief history of the development of flow cytometry instrumentation. Later units in the chapter describe specific aspects of the instrumentation. The fluidics system of a flow cytometer transports objects through the instrument, positioning them for accurate measurement. The various aspects of fluidics—from the sample tube to the interaction with the illumination beam, including flow analyzers and cell sorters— are described in UNIT 1.2. Calibration of an instrument for stability in measurement results (such that the same objects yield the same results day after day) requires frequent use of calibration particles. Calibration methods and sources of calibration particles are described in UNIT 1.3; this unit also includes a discussion of the difference between “standardization” and “calibration” and the need for both. Methods for establishing and maintaining the linearity of the system in order to make reproducible and accurate measurements are presented in UNIT 1.4. Another significant aspect of a flow cytometry system is the optical filters. Flow cytometers and flow sorters use light beams to excite fluorescent dyes; optical filters are used to separate the fluorescent light from the incident light. They are especially important with the current emphasis on multivariate analysis, which requires the simultaneous use of several dyes and light sources. UNIT 1.5 describes the basic principles of optical filters and discusses how to select a filter for a specific measurement in a cytometer. Most flow cytometers and sorters utilize laser beams as the excitation source; UNIT 1.6 discusses the various lasers that are in use and the various methods for delivering a laser beam to the objects being measured. A particularly valuable aspect of flow cytometry is its capacity for analyzing thousands of cells per second. For the detection and sorting of rare cells, special versions of the machines (high-speed sorters) have been developed. UNIT 1.7 discusses the principles and practices of operating a cell sorter to maximize its sorting rate, including the tradeoffs between speed and resolution. All commercial flow cytometers have a built-in capability for “gating,” a feature required for cell sorting. UNIT 1.8 describes the principles under which data are processed in real time and off-line to select subpopulations of cells (or cell organelles) with specific characteristics. Most flow cytometers utilize lasers as their illumination source. UNIT 1.9 describes the wide variety of laser types available and explains how they work. Included are pure and mixed-gas ion lasers as well as diode and other solid-state lasers (e.g., YAG lasers). The unit also discusses future directions, particularly in the development of new solid-state devices that might replace gas lasers. To ensure that high-quality, accurate data are obtained, flow instruments are checked for system performance each time they are used. UNIT 1.10 describes two levels of system Contributed by Bill Hyun Current Protocols in Cytometry (2003) 1.0.1-1.0.3 Copyright © 2003 by John Wiley & Sons, Inc.
Flow Cytometry Instrumentation
1.0.1 Supplement 26
alignment—routine alignment checks and complete alignment—that can be applied to a variety of instrumental configurations. The measurement of microbes is becoming an increasingly important application in flow cytometry. The small size of these organisms can create special difficulties that must be carefully addressed. UNIT 1.11 describes the problems encountered in making such measurements, the instrument requirements, and some approaches that can be taken to optimize the instruments for such measurements. This material is essential background for the protocols found in Chapter 11, Microbiological Applications. Fluorescence resonance energy transfer (FRET) is a technique that helps make possible the simultaneous use of three or more fluorescent dyes for the structural analysis of proteins in biological membranes. UNIT 1.12 describes the theory behind FRET, characterizes available parameters and instruments, discusses the method’s limitations, and presents a few examples of its applications. Light scatter was the first parameter measured in a flow cytometer. This highly useful parameter is probably the most widely used and least well understood by laboratories today. UNIT 1.13 describes in general terms the interaction of light with small particles, how the light is measured, and some current applications of this parameter in flow cytometry. In general, fluorescent dyes have relatively broad excitation and emission spectra, and when two or more are used simultaneously, the emission spectra will usually overlap to some extent. The measured fluorescence for a given dye can actually contain a significant contribution from any or all of the other dyes. To obtain accurate flow results it becomes necessary to compensate (correct) for this spectral overlap. UNIT 1.14 explains why compensation is necessary, how it is accomplished, and how it affects the visualization of data. The author addresses popular misconceptions that result in incorrect compensation and provides extensive discussion of suitable controls—cells, beads, and gates. The measurement of time-resolved fluorescence is a relatively new technique still under development in both flow and image cytometry, and promises to add a new dimension to multiparameter cytometry. Properly used, this technique can provide information about fluorophore/cell-interactions at the molecular level. UNIT 1.15 describes the theory behind fluorescence lifetime measurements and presents some recent applications. The simultaneous measurement of several fluorescent proteins allows one to monitor such things as gene expression and the intracellular localization of proteins. UNIT 1.16 presents a protocol for using a single laser operating at 458 nm to detect three fluorescent proteins simultaneously in a single cell. This is accomplished using a simple optical configuration and real-time compensation. Rapid screening of large combinatorial libraries, for instance of potentially valuable compounds in drug discovery tests, requires reliable automated sample-handling techniques. UNIT 1.17 describes one such technique, “plug flow cytometry,” in which precisely defined volumes of sample suspensions are injected at regular intervals into the flow stream. Many functions of living cells as well as molecular interactions are temperature dependent. UNIT 1.18 presents a protocol for using a Peltier module to control the temperature of a sample being processed in a flow cytometer/cell sorter. UNIT 1.19 covers in fine detail the photophysical and chemical properties of the common probes routinely employed in current flow cytometry. The spectral characteristics and use of example fluorescent dyes, probes, and labels for various functional assays and applications are described.
Introduction
1.0.2 Supplement 26
Current Protocols in Cytometry
Units covering other aspects of flow cytometry, including additional information about multispectral compensation in both flow and image cytometry and the use of micromirrors in new generation systems, will be presented in future updates to this chapter. Bill Hyun
Flow Cytometry Instrumentation
1.0.3 Current Protocols in Cytometry
Supplement 26
Overview of Flow Cytometry Instrumentation Flow cytometry is a technology in which a variety of measurements are made on cells, cell organelles, and other objects suspended in a liquid and flowing at rates of several thousand per second through a flow chamber. Flow sorting is an extension of this technology in which any single cell or object measured can be selectively removed from the suspension based on the measurements made. Flow cytometry is a very broadly applicable methodology. A brief list of applications that use flow cytometers includes: Disease diagnosis Chromosome karyotyping Cell function analysis Cancer therapy monitoring Detecting fetal cells Cell kinetics Identifying tumor cells Cytogenetics Fundamental cell biology. In a flow cytometer, cells in suspension are made to flow one at a time through a sensing region of a flow chamber (flow cell) where measurements are made. An example of an early flow cytometer is the Coulter counter (APPENDIX 3A). In this device, cells pass through a small orifice across which an electric current is flowing. As a cell enters the orifice, the flow of current is reduced because the cells are largely nonconducting. Electronic circuits detect the decrease in current and thus the presence of the cell. In this way the device can count the number of cells per second passing through the orifice, and because the volume flow rate can be measured one can determine the number of cells per milliliter of sample. The Coulter counter has been in use since 1949 and is still a mainstay of the clinical laboratory. Under the right conditions (e.g., size and length of orifice, current magnitude), the reduction in current through the orifice is proportional to the size (volume) of the cell, as demonstrated at the Los Alamos Scientific Laboratory in 1962. In modern flow cytometers, cells flow through a light beam rather than through a Coulter orifice; a Coulter orifice can, however, be included in these devices. Many different types of measurements can be made on the cells, based on the size and shape of the light beam and on the dyes used to stain components of interest. The light beam can come from arc lamps (e.g., mercury), as in early flow cytomeContributed by Phillip N. Dean Current Protocols in Cytometry (1997) 1.1.1-1.1.8 Copyright © 1997 by John Wiley & Sons, Inc.
UNIT 1.1
ters, or from lasers. Methods of measurement include absorption and scattering of the light beam by the cell, fluorescence of attached fluorescent dyes, and shape of the detected signal. Some of the properties and components that can be measured by a flow cytometer using these various methods are listed in Table 1.1.1. In principle, any component of a cell to which a fluorescent dye can be attached can be measured in a flow cytometer. If the binding of the dye is stoichiometric (i.e., amount of dye is proportional to amount of component) then the measurement can be quantitative and highly accurate (to within a few percent or better). Table 1.1.1 Properties and Components of Cells Measured in Flow Cytometry
Properties
Components
Cell diameter
DNA
Dye distribution Internal structure
Nuclear antigens Enzymes
Membrane potential Nuclear diameter
Protein RNA
Surface area Volume
Hormones Surface antigens
A flow cytometer is made up of several parts, as shown diagrammatically in Figure 1.1.1. All components of the system are necessary; the weakest part of the system defines its limitations. Other chapter units discuss the different parts of the system in detail. This overview describes the technology in general to give the reader a feeling for the interplay between the various parts of a flow cytometer. It also contains a brief history of the development of flow cytometry instrumentation.
CELL PREPARATION Objects to be measured must be suspended in a liquid. This is simple for blood cells, for example, but cells from tissue must be disaggregated and removed from any noncellular material. For most tissues this can be accomplished by procedures as simple as mincing the tissue with a knife and pulling cells through a 19-gauge needle into a syringe, followed by passing the cell suspension through a 200-mesh nylon screen. Details for such procedures are
Flow Cytometry Instrumentation
1.1.1
found elsewhere in this publication in units that deal with specific measurement and analysis protocols (e.g., see UNIT 5.2 for general procedures for handling, storage, and preparing human tissues and APPENDIX 3B for procedures for disaggregating cultured cell monolayers). After a single-cell suspension is obtained, the cells
are stained with dyes that bind to the specific features that are to be measured.
FLOW CHAMBER After staining, cells are made to flow one at a time through the interrogating light beam; a laser beam is illustrated in Figure 1.1.2. To
light source (UNIT 1.5 )
cell preparation
fluidics control
flow chamber (UNIT 1.2 )
(UNIT 1.2 )
sorter module (this unit, UNIT 1.2 )
detectors and signal processing (UNITS 1.3 &1.4 & Chapter 10 introduction)
analysis (UNIT 10.1)
display (UNIT 10.4)
Internet (UNIT 10.2 )
Figure 1.1.1 Schematic diagram of a complete flow cytometer system.
sample sheath
flow chamber
focusing lens
laser beam
Overview of Flow Cytometry Instrumentation
Figure 1.1.2 Longitudinal cross-sectional view of the flow chamber of a flow cytometer. The sample stream is surrounded by the sheath fluid which confines the cells (black dots) to the center of the chamber. The laser beam is focused onto the cell stream.
1.1.2 Current Protocols in Cytometry
obtain the best resolution, every cell must flow through the middle of the beam and be exposed to the same intensity of illuminating light. However, the laser beam has a Gaussian intensity distribution (i.e., the intensity is at a maximum in the center of the beam and decreases exponentially in the radial direction), and this puts a severe constraint on the stability of the flow stream. The system includes two features to alleviate this problem. (1) The beam leaving the laser has a circular cross section, and a long-focal-length cylinder lens is used to spread the beam in the horizontal direction and to produce a large depth of focus, resulting in a relatively large region of constant intensity in the center of the flow stream. (2) A “sheath” stream is introduced to the flow chamber. This sheath has a higher flow rate (∼5 ml/min) than the sample (∼100 µl/min), which serves to compress the sample stream and confine it to the center of the overall flow stream. This technique, called “hydrodynamic focusing,” is explained in more detail in UNIT 1.2. The end result is that cells are constrained to flow through an expanded laser beam in the center of the flow chamber. An additional constraint on the flow chamber is that it must be constructed of a material that will pass the excitation beam without appreciable scattering or absorption; this is usually accomplished through the use of quartz glass, which must be kept scrupulously clean. This is especially true when using ultraviolet light for excitation. The flow chamber can take many configurations. If a small orifice (e.g., sapphire jewel with a 70- to 100-µm hole) is placed at the
chamber exit, the flow stream will be compressed and will leave the chamber at high velocity. If the chamber is then vibrated at high frequency (e.g., 20,000 Hz), the stream will break up into uniform droplets and the flow cytometer will become a flow sorter. In this configuration measurements on cells can still be made in the chamber, although the time interval between cell detection and sorting can be relatively long. However, it is more common to pass the laser beam through the fluid stream just below the jewel before the stream breaks up (see Fig. 1.1.6). Then the interval between cell detection and sorting is shorter. In the latter configuration, the material requirements on the chamber are considerably reduced; the chamber becomes what is often called a sorter nozzle and can be constructed of ceramic materials. Because the hydrodynamic focusing does take place in the nozzle, in some sense the nozzle is a chamber. The sorting configuration is described in more detail later (see Sorting).
DETECTORS As a cell flows through the beam, light scattered by the cell and fluorescence light from dyes added to the cell are collected by light detectors, usually photomultipliers and photodiodes (see UNIT 1.4 for further discussion of photodetectors). These devices convert the light signal to an electrical signal that can be processed by the data processing and analysis unit. Photomultipliers, being very sensitive to light, are used where the light signal is weak (fluorescence), and photodiodes are used where the signal is strong (small-angle light scatter). The simplest flow cytometer would have per-
photomultiplier
filter pinhole photodiode collection lens laser beam flow chamber
Figure 1.1.3 Arrangement for a simple flow cytometer, containing a single fluorescence detector (photomultiplier) and a photodiode for detecting laser light scattered by a cell.
Flow Cytometry Instrumentation
1.1.3 Current Protocols in Cytometry
haps one photomultiplier and one photodiode, as shown in Figure 1.1.3. With the appropriate electronics system, this permits one to make two simultaneous measurements on a cell. As a cell flows through the beam it scatters some of the incident light, and the light scattering is typically detected by the photodiode, which is less sensitive than the photomultiplier. This continues as long as the cell is within the beam. Thus, the length of time a cell is in the beam (and the width of the electrical pulse produced) is proportional to the width of the cell. If the cell is also stained with a DNA-specific dye, the photomultiplier is used to measure the amount of fluorescent light emitted by the cell while it is in the light beam, producing a signal proportional to the DNA content of the cell. In Figure 1.1.3, an optical filter is shown that passes the fluorescent light and blocks the scattered excitation (laser) light. Thus, two measurements are made simultaneously. UNIT 1.5 contains a comprehensive discussion on how optical filters for flow cytometry are made and selected. By using a dichroic mirror (beam splitter) in front of the photomultiplier and incorporating a second photomultiplier with a different filter, as illustrated in Figure 1.1.4, three measurements can be made: e.g., DNA, total protein,
and narrow-angle light scatter. A dichroic mirror is one that reflects light below a specific wavelength and passes longer-wavelength light. The requirement for using more than one dye with this configuration is that both dyes excite at the same wavelength but emit at different wavelengths. The mirror is selected to separate the two emissions. Each detector also has a filter to block scattered excitation light. Fluorescent light is always emitted at a wavelength longer than that of the excitation light. Many flow cytometers today use two laser beams operating at different wavelengths to excite four or more dyes simultaneously. Figure 1.1.5 illustrates how this is done. The laser beams are separated vertically by ∼200 µm so that a cell flows through the two beams with a separation time of a few microseconds. Thus the two pairs of signals are separated in time, making it easier to resolve them. Each laser beam interaction point has its own pair of photomultipliers, dichroic mirror, and filter arrangement. In addition to measuring fluorescence, these detectors can be used to measure scattered light at 90°. The latter signal can help to distinguish cells with different internal structures. In principle, more detectors can be added to make even more measurements on each cell, with the limitation being the number of dye combina-
photomultiplier P2 filter P1
dichroic beam splitter
pinhole
photodiode collection lens laser beam flow chamber
Overview of Flow Cytometry Instrumentation
Figure 1.1.4 Arrangement for a flow cytometer with dual fluorescence detectors and a scatter detector. Light from two fluorescent probes is separated by the dichroic mirror and optical filters. With the appropriate filters, photomultiplier P1 can also be used to measure light scattered at 90° to the laser beam.
1.1.4 Current Protocols in Cytometry
tions that can be used. The combinations of excitation and emission spectra must be significantly different (see UNIT 1.5).
ANALYSIS All modern flow cytometers incorporate computers to monitor and in some cases to control the instrument, and to provide a capability for on-line analysis of instrument data. The computers are mostly Macintosh and IBMcompatible personal computers, which now have the power to perform virtually any kind of analysis desired. As the method of analysis required is not always known during an experiment, the computers are also used for off-line analysis. Software packages are available from the instrument manufacturers and from independent software companies. For more details on data processing and analysis, see Chapter 10. Flow cytometers are capable of producing enormous quantities of data very rapidly. This presents a challenge to the user, who must provide a means for storing the data in such a fashion that they can be recalled on demand. Because most data are stored in “listmode,” data files can be very large. “Listmode” means that every measurement on every cell is stored in a list. Thus, if five measurements are made on each of 50,000 cells, with a maximum value of 1024 per measurement (2 bytes), space has to be found for 500,000 bytes of information per sample. With the current development of
ever larger and less expensive storage devices such as read/write optical disk cartridges, this is not a major problem. A data file standard has been developed for the storage of flow cytometry data to make it possible for different laboratories to share data. This topic is discussed in UNIT 10.2. Sharing of data and the results of data analysis has become an important part of research; access to the Internet has become a desirable attribute of flow cytometer systems. To accomplish this one needs an Internet service provider and a “browser,” a computer program that provides access to other sites on the network. There are several browsers available, notably Mosaic, Netscape, and Internet Explorer. Many flow cytometry laboratories throughout the world have established sites on the Internet and made them available to other researchers in the field. A convenient location to begin a journey through the Internet is the home page of the International Society for Analytical Cytology (http://nucleus.immunol.washington.edu/ISAC. html), which contains links to most of these sites as well as to other sources dealing with both flow and image cytometry.
SORTING Principles A flow sorter is a cytometer with the additional capability of selectively removing from
beam 1 photomultiplier P1 beam 2 filter P2
beam splitter
flow chamber
P3
P4
half mirror
Figure 1.1.5 Flow cytometer with two excitation beams (lasers) that are separated vertically by 200 µm. A half mirror is used to direct fluorescent light from each beam interaction to a different pair of photomultipliers, each of which has a beam splitter and filter arrangement as in Figure 1.1.4. A photodiode could be added for each beam to permit a total of six measurements per cell.
Flow Cytometry Instrumentation
1.1.5 Current Protocols in Cytometry
the suspension of cells any selected cell flowing through it. The physical arrangement of a sorter is illustrated in Figure 1.1.6; the detailed fluidics are discussed in UNIT 1.2. Basically, the fluid containing the cells passes through a narrow orifice (∼100 µm in diameter) into the air. At the same time, the flow chamber is vibrated by the attached piezoelectric crystal, at frequencies on the order of 20,000 Hz. The vibration produces a disturbance in the ejected stream. The disturbance grows very rapidly, and the stream eventually breaks up into drops (i.e., 20,000 drops per second in the example given). The steady vibration causes the drops to be very uniform in size and spacing. Each cell that flows through the system will end up in a drop.
Measurements are made on the cells while they are either in the flow chamber or in the stream just below the orifice, before the disturbance of the stream has grown significantly. If the measurement result indicates that the cell is to be sorted, a voltage is applied to the stream just as the cell of choice reaches the end of the stream and a droplet is forming. When the drop separates from the stream it will carry electrical charge. The voltage on the stream is immediately reduced to zero so other drops will not be charged. Charges can be negative or positive, leading to the possibility of sorting two categories of cells simultaneously. As the drops continue to move downward they pass between two metal plates charged to a high voltage. Because
nozzle sapphire jewel laser beam
droplets
– –
deflection plates
–
charged droplets
– – collection beakers
Overview of Flow Cytometry Instrumentation
Figure 1.1.6 Diagram illustrating the principle of cell sorting. Cells flowing through the system are represented by small black dots. As cells to be sorted approach the end of the solid stream, a charge is applied to the stream. As the drop carrying the cell separates from the stream, the drop carries the charge. Passing between the high-voltage plates, charged drops containing desired cells are deflected into separate collection beakers. Deflection can be left or right, allowing for the simultaneous sorting of two classes of cells. In this illustration, two drops are sorted for each cell.
1.1.6 Current Protocols in Cytometry
the drops containing the selected cells are charged, they are deflected from the main stream of drops and collected in tubes or onto microscope slides for visual examination. Drops containing undesired cells are not charged and go directly into a waste tube.
Sort Purity Efficient and accurate sorting requires that the charge be applied just as the cell reaches the end of the stream. To compensate for variation in the flow velocity of the stream and to be certain the desired cell is sorted, typically two or three drops are sorted for each cell; Figure 1.1.6 shows two drops per cell. The sorting electronics are capable of detecting other cells
in the vicinity of the desired cell; if an unwanted cell might be sorted along with the desired cell, the sort is aborted. In some cases, particularly in the detection of very rare cells, one might want to accept some impurity in the sorted cell population to guarantee collection of the wanted cells. In that case, the sort purity requirement can be eased and the abort circuit disabled.
CHRONOLOGY OF FLOW CYTOMETRY DEVELOPMENT Table 1.1.2 is a list of significant events in the development of flow cytometry instrumentation. It is by no means comprehensive but illustrates the long history of the field. For a
Table 1.1.2
Development of Flow Cytometry
Year
Event
1934
Moldovan measures red blood cells in microscope with capillary flow and photodetector
1941
Kielland patents device like that developed by Moldovan
1947
Gucker uses Reynolds work to design laminar flow system for aerosols
1949
Coulter files for patent, “Means for Counting Particles Suspended in a Fluid”: the Coulter counter
1953
Crosland-Taylor designs system for aqueous suspension of cells flowing with a sheath: hydrodynamic focusing Parker and Horst apply for patent on device to do blood cell differentials using absorption of two colors of light
1956
“Model A” Coulter counter introduced
1962
Van Dilla uses Coulter counter to measure cell volume distribution
1965
Kamentsky measures UV absorbance and visible light scatter (500 cells/sec), generates bivariate plots Fulwyler develops electrostatic cell sorter, based on volume measurement Kamentsky develops fluidic switch cell sorter
1966
Van Dilla et al. introduce orthogonal measurements and laser excitation; prove DNA measurements were accurate; first to show discrete G1, S, and G2/M phase populations; show the possibility of quantitative cell kinetics studies
1968
Dittrich and Gohde patent microscope-based flow system with flow parallel to optic axis
1969
Hulett et al. introduce sorting based on cell fluorescence
1970
Wheeless et al. patent cell classification combining size and fluorescence
1971
Dittrich and Gohde introduce dual staining for DNA/protein, using ethidium bromide/FITC stains
1972
Wheeless et al. patent “slit-scan cytofluorometry” for automated cell recognition
1975
Gray et al. introduce flow karyotyping
1975 and beyond
Various investigators develop indexed sorting, high-speed sorting Flow Cytometry Instrumentation
1.1.7 Current Protocols in Cytometry
more complete discussion of the history, the interested reader is directed to Melamed et al. (1990).
SUMMARY Flow cytometry as a technology is still developing. New instruments with new or improved capabilities are constantly being introduced. This overview will be updated frequently to keep the reader apprised of developments in flow cytometry. A number of books provide excellent summaries of flow cytometry instrumentation (see Key References). For more details on particular techniques, the reader is referred to the articles in these books and to their extensive lists of references.
LITERATURE CITED Melamed, M.R., Lindmo, T., and Mendelsohn, M.L. (eds.) 1990. Flow Cytometry and Cell Sorting, 2nd ed. Wiley-Liss, New York.
Melamed, Lindmo, and Mendelsohn (eds.) 1990. See above. Second edition with all-new papers, producing true update of the first edition. Also contains extensive history of the field. Shapiro, H.M. 1995. Practical Flow Cytometry, 3rd ed. Wiley-Liss, New York. Comprises user’s reference manual for the laboratory. Also includes extensive list of current literature and list of key suppliers of instruments, parts, and reagents. Van Dilla, M.A., Dean, P.N., Laerum, O.D., and Melamed, M.R. 1985. Flow Cytometry: Instrumentation and Data Analysis. Academic Press, London. Contains papers by leading experts in the field on both subjects.
INTERNET RESOURCES http://nucleus.immunol.washington.edu/ISAC.html Homepage of the International Society for Analytical Cytology (ISAC).
KEY REFERENCES Melamed, M.R., Mullaney, P.F., and Mendelsohn, M.L. (eds.) 1979. Flow Cytometry and Cell Sorting, 1st ed. John Wiley & Sons, New York.
Contributed by Phillip N. Dean Livermore, California
First overall summary of the field, with many authors describing the state of the art as of 1979. Covers applications of the technology as well as instrumentation.
Overview of Flow Cytometry Instrumentation
1.1.8 Current Protocols in Cytometry
Fluidics
UNIT 1.2
Flow cytometry is so named because in this technique cells or subcellular particles suspended in a fluid are made to flow past sensors that take measurements from the cells. The primary function of the fluidics of a flow cytometric instrument is to deliver cells to the sensing area in single file and well aligned with the sensors. In a sorting flow cytometer, the fluidics must additionally be able to physically isolate cells chosen on the basis of measurements made by the instrument.
PRIMARY FLUIDIC FUNCTIONS Sample and Sheath Flow Figure 1.2.1 illustrates the main fluidic elements of a flow cytometer. Two primary lines feed fluid to the sensing region: the sample flow line and the sheath flow line. The sample flow line delivers the cell sample to the sensing region, and the sheath flow line provides a carrier fluid that helps position the sample flow and ultimately cells in the sensing region.
Hydrodynamic Focusing Most flow cytometry measurements are made optically, and it is important to keep particles well positioned in the flow stream in order to make accurate optical measurements. Flow cytometers use a fluidic method called hydrody-
namic focusing to control the position of particles in the flow (Fig. 1.2.2). In this technique, a flow of carrier fluid, called the sheath fluid, is established in the cytometer. The sheath fluid, which is usually normal physiological saline with perhaps a few additives, originates from a supply tank under pressure and flows through tubing to a sensing region where the detectors are located. Just before its arrival at the sensing region, the sheath fluid flows into a chamber of relatively large diameter, then out through a tapered conical section that reduces the diameter of the flow to the dimensions of the sensing region. The sample-containing fluid is introduced into the middle of this chamber through a tube positioned on the central axis of the sheath flow. Under laminar flow conditions, the sample and sheath fluids do not mix but join together to form a coaxial flow. This combined flow then passes through the tapered section, which reduces the diameter (and increases the velocity) of both the sheath and sample flows simultaneously before they reach the sensing region. This technique confines the cells to a very narrow central core so that the path cells follow through the sensing region is very consistent. The central area of combined flow that originated from the sample flow is called the sample core. It should be noted that the size of the sample
sample sheath pressure
sheath fluid
sample pressure
hydrodynamic focusing section sensing region
sheath tank
sample
sample tube
Figure 1.2.1 The basic flow system of a flow cytometer consists of sample and sheath forced under pressure through tubing to a hydrodynamic focusing section where the flows are combined in a coaxial flow prior to arrival at the sensing region. Flow Cytometry Instrumentation Contributed by Richard Stovel Current Protocols in Cytometry (1997) 1.2.1-1.2.7 Copyright © 1997 by John Wiley & Sons, Inc.
1.2.1 Supplement 1
sample sheath fluid sheath fluid
Figure 1.2.2 Flow cytometers use the principle of hydrodynamic focusing to align cells (represented here as black dots) in the center of the flow prior to their passage through the sensing region. The sample is injected into the middle of a sheath flow and the combined flow is reduced in diameter in a tapered section, forcing the cell into the center of the stream as shown.
to sensing region
core in relation to the diameter of the combined flow at the sensing area is a function of the relative sample and sheath flow rates, not the relative diameters of the sample and sheath introduction tubes. The effects of increasing or decreasing sample flow rate relative to sheath flow rate are shown in Figure 1.2.3. The reason that this point must be stressed is that the size of the sample core can be important for achieving data consistency. It is difficult optically to achieve spatially uniform illumination across the sensing region, so variation in particle position will result in variation of illumination intensity and therefore more variation in measurements due to position. Sample flow rates must be kept low relative to the sheath flow rate to maintain a narrow sample core and high data consistency.
Flow Cell
For most flow cytometers, the fluidic configuration of the sensing region can be classified as belonging to one of two types: jet-in-air or flow cell. Each type has its advantages and disadvantages. Some cytometers allow either type to be installed, and hybrids of the two types exist.
A second sensing configuration is the flow cell, in which the hydrodynamically focused coaxial flow passes through a transparent closed chamber wherein excitation light is brought into the chamber, and scattered and emitted light from the cells passes back out of the chamber to the detectors (Fig. 1.2.5). Generally, the inner diameters of flow cells are larger than the nozzles of jet-in-air systems, and the particles flow more slowly and spend more time in the sensing region. Higher-quality optical resolution can be achieved with flow-cell designs; moreover, the closed-flow system is advantageous when biohazardous samples are being handled.
Jet-in-Air
Hybrid
In a jet-in-air cytometer, the sheath and sample flow are combined in a nozzle, which tapers down to an orifice. The fluid in the nozzle is under enough pressure to form a continuous cylinder of fluid, or jet, as it emerges from the orifice (Fig. 1.2.4). Although some hydrodynamic focusing occurs in the tapered section, further focusing takes place as the fluid flow
Some cytometers combine advantages of both configuration types by sensing within a flow cell, then passing the flow through a jetforming orifice for droplet sorting. In this combination the exit section shown in Figure 1.2.5 is replaced by the orifice shown in Figure 1.2.4. This configuration provides a combination of high-quality optical measurement and droplet-
TYPES OF SENSING AREA
Fluidics
diameter narrows at the entrance to the orifice. The jet emerges from the orifice with the cells confined to the center of the jet. A light source, usually a laser, is focused onto the jet near the orifice, and optical elements collect the light scattered or emitted by cells as they pass through the light source and transmit this light to detectors, which convert it to an electronic signal. The jet-in-air configuration is well suited to droplet sorting (see discussion of Sorting).
1.2.2 Supplement 1
Current Protocols in Cytometry
A
sample sheath fluid sheath fluid
broad sample core
B
sample sheath fluid sheath fluid
narrow sample core
Figure 1.2.3 Diameter of the sample core is not dependent on the relative diameters of the sheath and sample tubes, but rather on the relative flow rates of the two fluids. (A) A sample at relatively high flow bulges after leaving the sample injection tube and is focused to a relatively broad sample core in the sensing region. (B) Low flow produces a narrow sample core that results in more accurate positioning of the cells in the sensing region.
sorting capabilities, but the sorting is less precise due to the longer transit time of the cells between sensing and sorting, and the biohazard containment advantage of a closed-flow system is lost.
Other Configurations In some cases flow cytometry techniques have been adapted to existing microscope technology by devising various flow configurations that use hydrodynamic focusing and pass cells through the focal plane of a microscope objective. One cytometer designed on these principles uses a jet-forming orifice to squirt the flow onto a microscope slide. The jet forms a continuous sheet of fluid on the slide, and flow conditions can be kept stable enough for cytometric measurements to be made through the microscope. Another type uses flow along the optical axis of the microscope; cells are measured as they pass axially through the focal plane of the microscope objective.
TYPES OF SAMPLE DELIVERY Standard Setup Samples of suspended cells are typically loaded into the flow cytometer in test tubes which are pushed on over an uptake tube that reaches to the bottom of the sample tube (see Fig. 1.2.1). The test tube pushes against a seal, and air pressure is applied through a port that projects through the seal. The pressure forces
the sample through the sample tubing to be combined with the sheath flow in the flow cell or nozzle. The pressure in the sample line at the point where the sample is injected into the sheath flow need be only marginally higher than the pressure in the sheath flow. Actual pressures applied at the sheath supply tank and the sample tube may differ by a larger amount due to pressure drops in the tubing and filters and differences in the height of the sample tube and sheath tank.
Alternative Techniques Other methods and devices, such as syringe pumps, are sometimes used to introduce samples into the cytometer. Automated delivery devices that can pick up and deliver samples sequentially from a rack of many samples are available. These are usually based on motordriven syringe pumps. There are also semiautomated pickup devices that are designed to pick up and deliver small-volume samples from microtiter plates. Special sample chambers are also available for certain specific purposes, such as mixing reagents with samples just prior to delivery for timed-reaction experiments.
POTENTIAL PROBLEMS AFFECTING SAMPLE FLOW Transient Flow Effects Liquid flowing through small tubing tends to travel slowly near the edges of the tubing,
Flow Cytometry Instrumentation
1.2.3 Current Protocols in Cytometry
Figure 1.2.4 A jet-in-air flow cytometer forces the coaxial flow of sheath and sample fluid through an orifice under sufficient pressure to form a jet. Illumination and detection take place in the jet after it has left the orifice.
orifice illumination
jet
due to drag by the walls, and faster in the center. The fluid develops a parabolic velocity profile in which the center of the flow travels twice as fast as the average flow. When a flow cytometric sample enters the sample tubing, the tubing is already filled with the sheath fluid that has been used to backflush the sample line. Because the outer fluid travels more slowly than the inner, the interface between sample and sheath fluid does not remain flat, but rather forms a curved surface; therefore, the first part of the sample to reach the sensing area will be what traveled in the center of the sample tubing, not the outside. This can lead to unusual effects. The first cells arriving at the sensing area may be more accurately positioned than following cells. Also, the first cells through have an opportunity to mix with and become diluted by the sheath fluid that was already in the tube. In situations where fluorescent dye is loosely associated with cells and the amount attached to cells depends on an equilibrium with the amount of free dye in the sample, this first-cell-through effect can cause measurements to vary over time until a stable sample flow that completely fills the sample tubing is achieved.
Settling
Fluidics
Cells will tend to settle out of suspension at a rate that varies with cell size and density. A typical settling rate for 10-mm-diameter lymphocytes is ~1 cm/hr. Thus, cells in a tube mounted on a cytometer for long periods require occasional resuspension. If the flow rate is very low, settling in the sample delivery tube can also be a concern.
The adherence properties, or stickiness, of cells influence their tendency to adhere to tubing walls or to each other. Differential settling of cells with different properties, in either the sample tube or the delivery tubing, can lead to changes in the distribution of cells reaching the sensing area.
Clogging Clogging of the small-diameter sensing region is frequently a problem in flow cytometry, especially in the case of jet-in-air systems, which have small jet-forming orifices; to avoid this problem, filtration of the cell sample is often advisable. Instruments with closed-flow sensing and no sorting orifice are more forgiving, because flow cells typically have a larger inner diameter than does the orifice of a sorting machine and therefore do not clog as easily. Filters are typically installed in the sheath fluid line both to guard against particulates that might clog the orifice or flow cell and to sterilize the fluid. Cell samples may or may not require filtration as they are introduced into the flow system of the cytometer, depending on the reliability of the sample preparation and the tendency of the cells to aggregate. If a filter is used, it can be a source of cross-contamination between samples; where high purity is needed, such as during reanalysis of sorted samples, the filter should be changed between samples.
Random Cell Arrival In a well-mixed cell suspension, the cells are, in principle, randomly distributed throughout the sample. Although it would be advanta-
1.2.4 Current Protocols in Cytometry
Figure 1.2.5 In a flow cell, the sensing region is contained within a chamber with transparent sides. Although not shown here, the cell usually has a square or rectangular cross section. The flow cell optical configuration may be combined with jet-in-air sorting by replacing the exit section with a jet-forming orifice.
illumination
transparent flow cell
geous to control the interval at which cells arrive at the sensing region, there is no way to do so, and in fact cell arrival time tends to follow a random distribution called the Poisson distribution. A number of effects may cause the actual arrival rate to deviate from the theoretical Poisson distribution: for instance, cells in the sample tube may not be mixed adequately, or may adhere to each other and tend to clump. The random arrival of cells at the sensing area presents problems for the signal processing electronics of the cytometer. A cell may arrive too soon after the previous cell so that the cytometer is not ready to measure it, or even worse, two cells may be so close together that the cytometer sees them as one. Inevitably, a certain percentage of measurements cannot be made satisfactorily. Such problems are alleviated if the average cell arrival rate is kept low, but this means that samples take longer to process.
SORTING The utility of flow cytometry in scientific experimentation is greatly enhanced by the ability of the instruments to isolate, or sort, cells on the basis of measurements made by the device. Some flow cytometers have this added capability and others do not. Most sorting flow cytometers use the electrostatic drop deflection method, which employs a jet-in-air configuration; some use other fluidic methods.
Electrostatic Drop Deflection A liquid jet in air emerges from its orifice as a column of fluid, but surface tension eventually causes it to break up into drops. If the jet is allowed to break up on its own, the sizes of the resulting drops will vary randomly over some range. In a sorting instrument, this breakup is brought under precise control by applying a periodic vibration at the orifice: this causes the breakup to occur in a very regular way so that the stream of drops is very uniform. The vibration is produced by means of a piezoelectric transducer attached to the nozzle: a periodic electrical signal applied to the transducer causes a small periodic variation in the diameter of the jet. Surface tension amplifies this variation as the jet progresses, and eventually the wave grows big enough to sever the jet, creating drops. The vibration may be applied either to the nozzle as a whole or to the fluid inside the nozzle. The main features of a sorting jet are illustrated in Figure 1.2.6. If the cell-sensing region of the cytometer is in the jet, a band of light is reflected and scattered perpendicular to the jet. The presence of a small surface wave on the jet causes this light band to be deflected up or down periodically by an amount that depends on the strength of the vibration that is applied to the jet. This effect causes additional optical noise that must be blocked in the measurement optical system usually accomplished with an obscuration bar. The vibration amplitude required to initi-
Flow Cytometry Instrumentation
1.2.5 Current Protocols in Cytometry
illumination
Figure 1.2.6 A jet-in-air sorter generates drops by vibrating the jet at a suitable frequency. Drops form with a uniform separation distance, called the drop wavelength.
satellite drop
drop wavelength
Fluidics
ate breakup of the jet by surface tension is quite small, however, and the optical noise problem is manageable. After measurements are made on a cell passing through the sensing region, the cytometer makes a decision on the basis of these measurements whether or not to sort the cell. The cytometer keeps track of the timing of desired cells, and when a desired cell arrives in the breakoff region of the jet, an electrical charging pulse is applied to the jet (which is an electrically conducting fluid) in such a way that the drop that contains the cell becomes charged and surrounding drops do not. The drop stream then passes through a strong, steady electric field that deflects the charged drops containing desired cells out of the stream of drops and into a collecting vessel. A drop may be either positively or negatively charged, allowing two populations to be sorted simultaneously, one to each side of the uncharged drop stream. In order to isolate chosen cells successfully, the rate at which cells arrive at the sensing region must be kept well below the drop formation rate. Otherwise, the probability of an unwanted cell being in the same drop as a wanted cell would be unacceptably high. Be-
cause sorting speed often limits the scope of experiments that can be performed with sorted cells, it is advantageous to generate drops at as high a frequency as possible. Drops are produced at the rate of the imposed vibration, but physical principles limit the size of the drops that can be produced by a jet of a particular diameter. For a cylinder of fluid, such as a sorting jet, the shortest section of fluid that can be made to form drops is a wavelength of about three times the diameter of the jet. A wavelength of 4.5 times the diameter is most favorable for drop formation. The frequency of the applied vibration must remain within these bounds. Given this limitation, there are only two ways to increase the rate of drop formation: to use a smaller-diameter jet (smaller orifice size) or to increase the velocity of the jet (higher nozzle pressure). Orifice size is limited by the tendency of small orifices to become clogged and by the size of the objects being sorted. Objects with diameters approaching that of the orifice will interfere with sorting in two ways: they will interfere with the regular formation of the surface wave at the orifice, and with the regular breakup of the jet at the breakoff point. Jet velocity is limited by the ability of cells to
1.2.6 Current Protocols in Cytometry
withstand the mechanical rigors of the sorting process and by the ability of the fluidic plumbing to withstand the higher pressures required. For historical reasons, commercial sorters have been limited to operating pressures of ~1 atm (15 psi) and nozzle sizes of ~70 to 100 µm, which meant that drop frequencies have been limited to ~25,000–35,000/sec. More recently, sorters have become available that operate at significantly higher pressure and correspondingly higher jet velocity and drop frequency. Sorting jets have the interesting property of forming a satellite drop between the main drops. Normally this smaller drop soon merges with an adjacent larger drop and is of little consequence. Under some circumstances, however, it may not merge, and a separate stream of smaller satellite drops may be formed; this can have the undesirable effects of interfering with sorting and causing an additional biohazard.
Closed-System Mechanical Sorting An alternative sorting method available in some commercial machines employs a mechanical actuator situated downstream from the sensing region of a closed-system flow cell. The actuator moves a collection tube into the flow of particles and picks off desired cells as they flow by. This method has the significant advantage of maintaining a closed flow path, which makes it more suitable for biohazardous experiments than jet-in-air methods. Its disadvantages include a slower sorting rate and a more dilute sorted fraction.
OTHER FLUIDIC FUNCTIONS In addition to its primary functions of delivering samples to the sensing region and combining sample and sheath flow for hydrodynamic focusing, the fluidic system of a flow cytometer generally has secondary functions built in. A boost function advances the sample from the sample tube to the sensing region at high speed at the start of a sample run to reduce the waiting time before the leading edge of the sample reaches the sensing region. This is done by switching the sample pressure momentarily to a higher level. Depending on the cytometer, this may be done either automatically or by the operator by means of a boost air-switch on the flow-control panel.
Another function that is often provided is a fill function, which allows the fluidic system to be filled more quickly after it has been drained. To do this, an extra port near the sensing region is opened to allow fluid from the sheath tank to enter the system at a much faster rate than if the sheath were flowing normally through the small-diameter sensing region. It is important to backflush the sample line between samples to remove all trace of the previous sample from the fluid system before introducing the next one. To do this, the uptake end of the sample line is left open to the atmosphere while the sheath flow and pressure is maintained at the sensing end. The resulting pressure differential causes sheath fluid to flow backward through the sample line, clearing out the remaining sample. Waste removal capability is provided to receive the sample after it has passed the sensing region and to remove fluid during backflushing. Waste fluids may drain into a waste tank under gravity or may be sucked into the waste lines by means of a vacuum applied to the waste tank. Jet-in-air systems usually have an aspirator to catch the unsorted droplet stream.
KEY REFERENCES Kachel,V., Fellner-Feldegg, H., and Menke, E. 1990. Hydrodynamic properties of flow cytometry instruments. In Flow Cytometry and Sorting (M. R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 27-44. Wiley-Liss, New York. Excellent review of published work in fluidics relevant to flow cytometry. Pinkel, D. and Stovel, R. 1985. Flow chambers and sample handling. In Flow Cytometry: Instrumentation and Data Analysis (M. A. Van Dilla, P.N. Dean, O.D. Laerum, and M.R. Melamed, eds.) pp. 77-128. Academic Press, London. Detailed treatment of both theoretical and practical flow cytometric fluidics. Shapiro, H.M. 1995. Practical Flow Cytometry, 3rd ed. Wiley-Liss, New York. Includes thorough description of fluidics as well as many other aspects of flow cytometry.
Contributed by Richard Stovel Stanford University Stanford, California
Flow Cytometry Instrumentation
1.2.7 Current Protocols in Cytometry
Standardization, Calibration, and Control in Flow Cytometry Standardization, control, and calibration provide different degrees of certainty about the data acquired with an instrument. Each process is aimed at assuring that results from the instrument have the quality required for the intended purpose (Horan et al., 1990; NCCLS, 1998a; Muirhead, 1993a,b; Schwartz and Fernandez-Repollet, 1993; Owens and Loken, 1995; Schwartz et al., 1996; Owens et al., 2000). The purpose may be an individual research experiment or a clinical result that determines the course of patient treatment. In the terminology used in this commentary, an instrument is standardized at certain time points and subsequently operated under quality control conditions (see UNIT 3.1–3.2). These processes maintain the instrument within predetermined bounds and ensure that results will vary only within certain limits. If a result is also calibrated when the instrument is standardized, then future results can be objectively and quantitatively compared with those from other laboratories. Quantitation of results should be considered. Most results from flow cytometers are expressed either in terms of “percent positive” or in qualitative terms such as “dim” or “bright.” These terms are relative: what is considered “negative,” “dim,” and “bright” in one laboratory may be quite different in another laboratory. When visualizing fluorescence using a fluorescence microscope, such relative terms are necessary. Flow cytometers can measure the amount of fluorescence and provide more objective criteria for expressing results. As flow cytometers are designed to measure particle characteristics (see UNIT 1.1 for an overview of flow cytometry), particles are the most common materials used to calibrate, control, and standardize the instruments. This commentary describes how various types of particles are used for these purposes. It also briefly reviews the status of standardization and quality control for flow cytometry (see Chapter 3 for further discussion of quality control). UNIT 1.4 covers calibration of detection system components (e.g., linear and logarithmic amplifiers) to ensure linearity of the flow cytometer response. The first section of this unit focuses on how the term “standard” has been used in flow
UNIT 1.3
cytometry (see Standards, Standardization, and Jargon). The intent is to alert readers of flow cytometry literature that they must always interpret critically how “standard” is being used in a particular context. The next section defines terms and also includes comments to put the term in context or to highlight issues (see Definitions). After providing extensive background on particle types and cautions (see Overview of Standardization in Flow Cytometry), this unit describes practical aspects of methods to standardize and calibrate flow cytometers (e.g., in terms of optical alignment, fluorescence and light scatter resolution, and sensitivity; see Standardization and Calibration section). Finally, suggestions are given for analyzing particles used as calibrators, including how to assign to fluorescent beads a value for molecules of equivalent soluble fluorochrome (MESF) and how to determine the inherent fluorescence coefficient of variation (CV) of a dim bead sample (see Characterizing Particles for Calibration and Control of a Flow Cytometer).
STANDARDS, STANDARDIZATION, AND JARGON It is common in flow cytometry to combine words that describe use of a particle with the word “standard.” Examples are “calibration standard” and “alignment standard” (Horan et al., 1990; Schwartz and Fernandez-Repollet, 1993; Shapiro, 2003; Schwartz et al., 1996). Rarely is there any indication of who has set the “standard” and by what authority or consensus. There can be many levels of “standards,” depending on the size and authority of the group that establishes them. For example, an individual laboratory or investigator may have standard practices or materials. A large clinical or research study may have standard practices and materials that are agreed to by all investigators involved in the study. A professional organization may establish standard methods or identify standard materials for specific purposes. If the word “standard” is not modified by a term such as “laboratory,” “clinical trial,” or “study XYZ,” it may imply something that is generally and widely accepted by acknowledged authorities. In that authoritative Flow Cytometry Instrumentation
Contributed by Robert A. Hoffman Current Protocols in Cytometry (2005) 1.3.1-1.3.21 C 2005 by John Wiley & Sons, Inc. Copyright
1.3.1 Supplement 32
sense, however, there are few “standards” in flow cytometry. Clear and common understanding of what is meant by a term is important, especially as flow cytometry is used by increasing numbers of investigators. The verb “standardize” means to cause to be without variation. Early use of the noun “standard” in flow cytometry seems to have been in the sense of a particle used to standardize (make consistent) one instrument in one laboratory (Fulwyler, 1979). This is much different from the authoritative sense of “standard.” In this commentary, other terms are used to describe more specifically what type of particle or material is being used for a particular purpose. For example, “calibration particle” or “calibrator” is used instead of “calibration standard,” and “alignment particle” rather than “alignment standard.”
DEFINITIONS
Standardization, Calibration, and Control in Flow Cytometry
Concern with terminology and its evolution is not just semantics, but reflects what has been important in flow cytometer technology and how the technology has grown and changed. More precise and generally accepted terminology should clarify communication and understanding among flow cytometrists as well as scientists in other fields. A very useful set of definitions that generally cover the field of clinical diagnostics is NCCLS publication NRSCL8-A, Terminology and Definitions for use in NCCLS Documents (NCCLS, 1998a). The definitions for limits below are from the NCCLS definitions. Additional definitions related to quantitative fluorescence cytometry are provided by Henderson et al. (1998). The definitions below should be considered a reasonable point along the way toward authoritative and broadly accepted and understood terminology. Some definitions include comments and references that may help put them in context. Accuracy: degree to which a measurement agrees with the true or expected value. Alignment particle: particle with uniform size, fluorescence, and light scatter characteristics that is used to check the alignment (or, in some instruments, adjust the alignment) of the excitation and emission optics in the flow cytometer. It is desirable that the alignment particle emits fluorescence in all detector channels, as this allows all channels to be checked simultaneously. Alignment of the optics is optimal when signals from the particles have maximum intensity and minimum variation or CV. The more uniform the particles, the better the degree to which small deviations from optimal
alignment can be detected. Optimal alignment is most critical for measuring DNA, because of the very low inherent variation in DNA content from cell to cell. Antibody binding capacity (ABC): number of antibodies of a particular type that can bind to a cell under saturating staining conditions. Autofluorescence: inherent fluorescence from a cell or particle to which no stain or fluorochrome has been added. Manufactured particles (such as plastic beads) can be prepared to have nearly the same autofluorescence as lymphocytes. Background (noise, fluorescence, scatter): signal present when no particles are flowing in the sample stream. Background noise is one factor that limits the sensitivity of fluorescence detection (see definitions of fluorescence sensitivity and light-scatter sensitivity below). Depending on how low are the signals that one is trying to detect in the sample, different factors are dominant contributors to the background. When no light is coming from the flow cell (e.g., lasers turned off), detector noise is the background limit. For photomultiplier tubes (PMTs), the detector background noise is called dark current and is due to random emission of electrons from the photocathode. For photodiodes and other solid-state detectors, which have no or low signal amplification, the limiting factor under best conditions is noise from the amplifier required to raise the signal to a useful level. Sources of fluorescence noise include Raman scatter from water and optical components; fluorescence from unbound fluorochrome, reagent, or contaminants in the sample or sheath stream; and fluorescence from optical components. Calibration: process of adjusting an instrument so that the analytical result is accurately expressed in some physical measure. Calibrator: material that has been manufactured or assayed to have known, measured values of one or more characteristics. The assayed values are provided with the material. Fluorescent manufactured particles can be assayed for diameter or for the amount of fluorescence they produce. A practical measure of particle fluorescence is the number of fluorochrome molecules in solution that produce the same amount of fluorescence as one bead (see definition of MESF). Coefficient of variation (CV): statistical measurement of the broadness of a distribution of values, usually defined as CV = σ /µ, where the standard deviation σ = [(xi − µ)2 / (N − 1)]1/2 , with the sum over N measurements of xi (where xi is the ith measurement of
1.3.2 Supplement 32
Current Protocols in Cytometry
variable x), and the mean µ = ( xi )/N. Shapiro (2003) gives an excellent discussion of CV and other, more robust statistics for flow cytometry. Another excellent reference for statistical methods is Bevington (1969). Control particle or material: stable material (e.g., sample of manufactured particles) that gives reproducible results when analyzed. Particles used to set up a flow cytometer are used as a control even if they do not have an assayed value assigned to a physical characteristic. Controls can be used to monitor the stability of an instrument and determine whether it is acceptably within calibration. A calibrator can be used as a control material, but a control material does not have to have an assigned value for a characteristic. Control sample: sample prepared in the same or nearly same way as a test or unknown sample and which should give an expected, predetermined result. In immunofluorescence analysis a positive control sample may use known cells (characterized for reactivity to a panel of antibodies) and the same antibody reagents as the test sample. A negative control sample may use the test cells but without antibody reagent or with an irrelevant antibody reagent. Fluorescence sensitivity: In flow cytometry there are two different aspects to the notion of sensitivity: threshold and resolution. The first has to do with the smallest amount of light that can be detected (Wood, 1993; Owens and Loken, 1995; Schwartz et al., 1996; Shapiro, 2003). This notion has also been given the name “detection threshold” (Schwartz et al., 1996). The second has to do with the ability to resolve dimly stained cells from unstained cells in a mixture (Brown et al., 1986; Horan et al., 1990; Shapiro, 2003). These concepts do not measure the same thing. The second notion incorporates a measure of the broadness of the fluorescence distributions for dim and unstained particles, not just the average fluorescence. Two instruments can have the same detection threshold but differ significantly in ability to resolve a dimly stained population. This is illustrated by example later (see Standardization and Calibration Section). 1. Degree to which a flow cytometer can measure dimly stained particles and distinguish them from a particle-free background (threshold). Threshold is important when the mean fluorescence of a dimly fluorescent population is measured. The greater the number of particles analyzed, the more accurately and precisely will the mean fluorescence be measured.
2. Degree to which a flow cytometer can distinguish unstained and dimly stained populations in a mixture of particles (resolution). Resolution is important for immunofluorescence analysis of subpopulations and is strongly affected by the measurement CVs for dim and unstained particles. Inherent sample CV: actual variability in the characteristics of a sample; for example, the actual variation in the amount of fluorochrome per bead in a sample of beads. Because the measurement process is not perfect and itself adds variation, the CV of the measured fluorescence will be greater than the inherent sample CV. The inherent CV of a sample can be estimated within a small uncertainty if the measurement variability added by the flow cytometer is well characterized (see Determining Inherent Fluorescence CV of a Dim Particle Sample). Light-scatter sensitivity: degree to which small particles can be detected above “particlefree” fluid. In practice, forward-scatter sensitivity is usually limited by optical noise caused by the excitation source, and side-scatter sensitivity is usually limited by submicron particles in the sheath fluid. Limit of detection: the lowest amount of analyte in a sample that can be detected but not quantified as an exact value. Limit of quantitation: the lowest amount of analyte in a sample that can be quantitatively determined with acceptable precision and accuracy under stated experimental conditions. Manufactured particles (beads, plastic beads, latex particles, microspheres, microbeads): particles made of synthetic polymers (plastics). Sizes range from submicron to over 100 µm, which generally covers the range of cells analyzed in flow cytometry. Most manufactured particles are made by bulk polymerization, but very uniform beads can be made employing the same droplet generation principle used for flow cytometric cell sorting (Fulwyler et al., 1973). Colored or fluorescent particles can be made by staining the beads with dyes or fluorochromes. Nonfluorescent beads, as well as many fluorescently stained beads, seem to be stable for many years. Two methods, namely, solvent (or “hard”) dying and surface staining, are used to stain particles. In solvent staining, non-water-soluble dyes are mixed with the particles in an organic solvent. The particles take up the dye and are then suspended in aqueous solution. The dye is trapped in the beads, which essentially become a “hard-dyed” plastic material. In some cases, hard-dyed particles can be synthesized
Flow Cytometry Instrumentation
1.3.3 Current Protocols in Cytometry
Supplement 32
Standardization, Calibration, and Control in Flow Cytometry
directly using fluorescent monomers (Rembaum, 1979). As most dyes or fluorochromes used to stain cells are water soluble, solvent staining cannot generally be used for them. When solvent staining is possible for watersoluble fluorochromes, the spectral characteristics can differ significantly from those of fluorochrome in aqueous solution. Surface staining allows many common fluorochromes— especially those used as tags on fluorescent antibodies—to be used for particle staining. In this case a chemical group on the particle surface (e.g., amino group) is covalently bound to a reactive group on the fluorochrome. MESF (molecules of equivalent soluble fluorochrome): measure of particle fluorescence in which the signal from a fluorescent particle is equal to that from a known number of molecules in solution. This is a practical measure because a known concentration of particles can be compared directly with a solution of fluorochrome in a spectrofluorometer (see Calibrating Particle Fluorescence in MESF). Nonfluorescent particle: particle whose fluorescence distribution is the same as that of a particle-free sample. In practice, the concept of nonfluorescence is dependent on the sensitivity of the instrument making the measurement. A particle that is not measurably fluorescent in one instrument may be so in a more sensitive instrument. Fluorescence (or other luminescence or Raman scatter) from otherwise unstained manufactured particles depends on the material and treatment with which the beads are made. With all other factors equal, the “fluorescent” signal from microbeads will be proportional to the volume of a single bead. Precision or reproducibility: degree to which repeated measurements of the same thing agree with each other. In flow cytometry, precision of a measurement is estimated by the CV obtained when measuring a sample of particles (biological or nonbiological) with very uniform characteristics. Resolution: degree to which a flow cytometry measurement parameter can distinguish two populations in a mixture of particles that differ in mean signal intensity. Fluorescence sensitivity (see above) can be considered a special case of fluorescence resolution for which the signals are very dim. Note that the resolution will appear different when data are acquired and/or displayed on a logarithmic rather than linear intensity scale. Depending on the maximum number of channels into which the signal intensity is acquired (e.g., 256 or 1024),
a logarithmic display of the data may not have sufficient resolution to display populations that can actually be resolved by the instrument. Standard: 1. noun. a. acknowledged measure of comparison for quantitative or qualitative value. b. something recognized as correct by common consent or by those most competent to decide. 2. adj. a. serving as a standard of measurement or value. b. commonly used and accepted as an authority. Standardize: a. cause to conform to a given standard. b. cause to be without variation. Test pulse–triggered background fluorescence: measurement of background fluorescence in a flow cytometer by using an electronic pulse to trigger the pulse detection electronics and acquire data from the fluorescence detector(s) (see UNIT 1.4). As no particle is present to emit light, the fluorescence signals acquired are due only to instrument background light and noise and thus establish the lowest signal that can be measured. The duration of a test pulse usually simulates a signal from a particle of typical size. Larger particles would have signals of longer duration and produce more background signal and noise. If equipped with a test pulse function, the flow cytometer can provide a measurement equivalent to running a sample of truly nonfluorescent particles. The background fluorescence distribution produced by a test pulse should provide a measure of the “detection threshold” described by Schwartz et al. (1996). In many instruments the test pulse signal produces a pulse of light from a light-emitting diode that is detected and processed by one detector. When the test pulse signal is applied only to the forward scatter detector, the response of all other detectors to background light and noise can be measured. When a sample is run under normal conditions, any signal from particles above this background and noise level actually comes from the particles. There is no guarantee, however, that the particle signal is from particle fluorescence; for example, light scattered by the particles may not be totally blocked by the optical filters, or in some cases the light scatter may actually induce the filter to fluoresce. The possibility of scatter-induced “fluorescence” signal can be checked by running unstained cells and looking for a signal in the fluorescence channel. Because such a signal can also come from autofluorescence, one should also look at the side scatter versus fluorescence histogram.
1.3.4 Supplement 32
Current Protocols in Cytometry
OVERVIEW OF STANDARDIZATION IN FLOW CYTOMETRY Standardization (see Definitions) is the foundation of flow cytometry and allows investigators to have confidence in instrument performance. This section surveys characteristics of particles used in flow cytometry, for example, to standardize immunofluorescence and to check alignment and measurement precision (see Types of Particles). Specific types of particles are compared. Standardization can be complicated, however, by factors other than particle type (see General Cautions for Using Particles in Standardization and Calibration; see What the Instrument Cannot Control: Sample, Reagent, and Data Analysis), but prospects for formalizing flow cytometry standards are encouraging (see Standard-Setting Organizations). The next section (see Standardization and Calibration) reviews various parameters of flow cytometers that can be standardized, such as resolution and sensitivity, and the final section (see Characterizing Particles for Calibration and Control of a Flow Cytometer) describes procedures and cautions for characterizing particles.
Types of Particles Manufactured particles and biological particles may be used to standardize flow cytometers. Beads may be spectrally matched to the fluorochromes used to stain cells, or they may simply fluoresce to a useful extent in the spectral range of interest. Spectrally matched beads allow standardization or even calibration across instruments that do not have exactly the same emission filters and/or excitation wavelengths. Biological particles may be stained with the same fluorochromes used in experiments to stain cells. Examples of data for some of the more common types of particles follow. Classification schemes for various types of particles used for standardization in flow cytometry have been proposed (Schwartz et al., 1996, 1998).
Comparison of spectrally matched and unmatched fluorescent particles Figures 1.3.1 and 1.3.2 show emission spectra from three types of particles: fluorochrometagged beads (CaliBRITE beads, BD Biosciences), stained with either fluorescein isothiocyanate (FITC) or phycoerythrin (PE); broad-spectrum hard-dyed beads (Rainbow beads, Spherotech); and glutaraldehyde-fixed chicken red blood cells (gCRBC, BioSure).
Figure 1.3.1A compares FITC-stained CaliBRITE beads with Rainbow beads and gCRBC, which are not spectrally matched to FITC. Figure 1.3.2A makes the same comparison with PE-stained CaliBRITE beads. Figure 1.3.1B compares quantitatively the fluorescence signal of each particle through filters of differing spectral bandwidth placed in front of PMT1, with data being normalized to the signal from FITC CaliBRITE beads for each filter. The gCRBC varied by about 35% over the range of filters used. Rainbow beads, however, varied by nearly 200% with the same filters. The differences in relative fluorescence with different filters should be considered when comparing different instruments. Even for a particular flow cytometer type or model, filters vary slightly due to manufacturing tolerances. Figure 1.3.2B shows a similar comparison for the relative fluorescence with different filters placed in front of PMT2. In this case, both gCRBC and Rainbow beads vary only slightly from PE-stained CaliBRITE beads. The fluorescence could be standardized with a maximum difference of 40% with any of these particles.
“Nonfluorescent” and autofluorescent particles Figure 1.3.3 shows emission spectra for particles with very low fluorescence. Unstained CaliBRITE beads have fluorescence comparable to autofluorescence from lymphocytes. Osmium-fixed chicken red blood cells (CRBC) had no fluorescence detectable above background in the fluorometer. Such “negative” particles are useful for estimating how well low level signals can be detected, as discussed later (see Sensitivity or Signal/Noise for Dim Fluorescence).
Comparison of particles for standardizing immunofluorescence analysis Figure 1.3.4 shows light-scatter dot plots (panel A) and green (515 to 545 nm) fluorescence histograms (panels B-F) for several types of particles used to standardize flow cytometers for immunofluorescence analysis. Fluorescence from the stained particles is in the range observed for immunofluorescence from cell-surface markers. All data were acquired using the same instrument settings, and panels A-D were obtained from the same sample acquisition of a mixture containing (1) unstained (autofluorescence) and FITC-stained CaliBRITE beads, shown in
Flow Cytometry Instrumentation
1.3.5 Current Protocols in Cytometry
Supplement 32
Figure 1.3.1 (A) Fluorescence emission spectra of FITC CaliBRITE beads (Becton Dickinson Immunocytometry), Rainbow beads RFP-30-5K (Spherotech), and glutaraldehyde-fixed chicken red blood cells (gCRBC, BioSure). Excitation at 488 nm was used. (B) Percentage of fluorescence signal through different optical filters for Rainbow beads and gCRBC normalized to signal from FITC CaliBRITE beads. Data in B are also scaled relative to the signal through the 505 to 525–nm filter.
Standardization, Calibration, and Control in Flow Cytometry
region R1 in panel A; (2) a combination of unstained and multiple levels of stained Rainbow beads, shown in region R2 in panel A; (3) gCRBC, shown in region R3; and (4) forwardscatter (FS) test pulses (no particle, R4 in panel A). Fluorescence histograms in Figure 1.3.4 are from unstained and FITC CaliBRITE beads (panel B), Rainbow beads (panel C), FS test pulses and gCRBC (panel D), Quantum 24 beads (Flow Cytometry Standards; panel E), and osmium- and glutaraldehyde-fixed CRBC (panel F). Panels B, D, and F illustrate different pairs of particles or signals at the low and high ranges of a scale for immunofluorescence. The Quantum 24 beads shown in Figure 1.3.4E had calibration values for the stained beads (upper four peaks in the histogram) of 4,201, 16,936, 37,466, and 65,797 fluorescein MESF (molecules of equivalent soluble fluorochrome).
Particles for aligning and checking measurement precision Figure 1.3.5 shows scatter and fluorescence data for a uniform 2.49-µm-diameter fluorescent bead that is useful for checking or adjusting optical alignment. All fluorescence CVs were <2%.
General Cautions for Using Particles in Standardization and Calibration There are two important factors to remember when using manufactured particles rather than cells in a flow cytometer. First, beads are not cells and do not necessarily scatter light as cells do. Second, fluorescence from a bead may be similar to that from a cell stained with a particular dye, but it is almost never identical. Regarding the first point, light scatter from beads usually differs greatly from scatter from cells of the same size. This is primarily due to
1.3.6 Supplement 32
Current Protocols in Cytometry
Figure 1.3.2 (A) Fluorescence emission spectra of PE CaliBRITE beads (Becton Dickinson Immunocytometry), Rainbow beads RFP-30-5K (Spherotech), and glutaraldehyde-fixed chicken red blood cells (gCRBC, BioSure). Excitation at 488 nm was used. (B) Percentage of fluorescence signal through different optical filters for Rainbow beads and gCRBC normalized to signal from PE CaliBRITE beads. Data in B are also scaled relative to the signal through the 564- to 606-nm filter.
differences in the optical refractive indexes of beads and cells. Beads with high water content and consequently low refractive index, such as Sephadex chromatography beads, give light scatter results similar to those from cells (Sharpless et al., 1977). However, such commercially available beads have a wide distribution of diameters and are difficult to use in standardizing an instrument. Regarding the second point, fluorochromes used in hard-dyed beads are rarely the same as those used to stain cells. Even when the fluorescence emission spectra of a hard-dyed bead and a fluorescently stained cell are closely matched, it will be rare that the excitation spectra also match unless bead and cell are stained with the same fluorochrome. Also,
one needs to be aware of a possible nonlinear relationship between the fluorescence signal and the intensity of light used to illuminate the particles (UNIT 1.4 discusses system linearity). The relative fluorescence of two different fluorochromes can differ considerably with the intensity of excitation light (Bohmer et al., 1985). Fluorochromes such as FITC and PE, which are used to tag antibodies, are available on surface-stained beads, and such beads closely match the fluorescence characteristics of cells stained with tagged antibodies. The surface-stained beads have the same excitation spectrum and sensitivity to excitation light intensity as do cells stained with the same fluorochromes.
Flow Cytometry Instrumentation
1.3.7 Current Protocols in Cytometry
Supplement 32
Figure 1.3.3 Fluorescence emission spectra of unstained CaliBRITE beads (Becton Dickinson Immunocytometry) and osmium-fixed chicken red blood cells (CRBC, BioSure). Excitation at 488 nm was used, and concentrations of the two types of particles were the same. No fluorescence from the osmium-fixed CRBC could be measured above background noise in the spectrofluorometer.
What the Instrument Cannot Control: Sample, Reagent, and Data Analysis It is important to keep in mind the factors that affect results but are beyond the control of the flow cytometer. A well-calibrated instrument and careful quality control cannot correct for samples and reagents that are not properly maintained, prepared, and used (see Chapter 4 for information on molecular and cellular probes; see Chapter 5 for specimen handling, storage, and preparation). Good data produced by the instrument cannot guarantee correct results if data analysis is wrong (see Chapter 10 on data processing and analysis). The flow cytometer hardware is only one part of the system that must work correctly to give good results. Owens and Loken (1995) provide an excellent and instructive introduction to the entire range of factors that affect results of flow cytometric analyses commonly performed in clinical laboratories.
Standard-Setting Organizations
Standardization, Calibration, and Control in Flow Cytometry
The Clinical Laboratory Standards Institute (formerly NCCLS), an international clinical laboratory standards–setting organization, has established several guidelines specifically for flow cytometry or that apply to flow cytometry. The list (at the time of this writing) includes: ILA24-A (Fluorescence Calibration and Quantitative Measurement of Fluorescence Intensity;
Approved Guideline; NCCLS, 2004); H42-A (Clinical Applications of Flow Cytometry: Quality Assurance and Immunophenotyping of Lymphocytes; Approved Guideline; NCCLS, 1998b); H43-A (Clinical Applications of Flow Cytometry: Immunophenotyping of Leukemic Cells; Approved Guideline); H44-A (Methods for Reticulocyte Counting– Flow Cytometry and Supravital Dyes; Approved Guideline); and H52-A (Fetal Red Cell Detection; Approved Guideline). The US National Institute of Standards and Technology (NIST) has developed a method for determining particle MESF (Gaigalas et al., 2001; Schwartz et al., 2002) and has developed a standard fluorescein solution (Standard Reference Material 1932). A standard set of fluorescein-labeled beads (Reference Material 8640) has also been developed. Practical issues to consider in using MESF and quantitating fluorescence have been addressed in detail by NIST (Wang et al., 2002). The NCCLS guideline H42-A (NCCLS, 1998b) takes a conservative approach in stating “There are at present no standards which can be used to check the accuracy of flow cytometric test results. Hence, verifying reproducibility of instrument performance is an essential element of daily quality assurance for the flow cytometry laboratory. Instrument performance must be monitored under the same conditions as are used to run test samples.”
1.3.8 Supplement 32
Current Protocols in Cytometry
Figure 1.3.4 Light scatter and green fluorescence distributions from several types of standardization particles or sources. All data were acquired with identical instrument settings, and pulse height was measured. (A) Forward scatter (FS) versus side scatter (SS) dot plot of mixture of beads and test pulse signals. Each type of particle or the test pulse is enclosed by a region in the dot plot: region R1 contains unstained and FITC CaliBRITE beads or Quantum 24 beads; R2 contains the Rainbow bead mixture RCP-30-5K; R3 contains glutaraldehyde-fixed chicken red blood cells (gCRBC). Region 4 contains forward-scatter test pulses, which allow background noise from all other parameters to be measured. (B) Green fluorescence histogram of unstained and FITC CaliBRITE beads. Unstained beads have nearly the same autofluorescence as lymphocytes. (C) Green fluorescence histogram of Rainbow bead mixture containing unstained beads and five levels of stained beads. (D) Green fluorescence histogram of background noise from test pulse–triggered acquisition (region R4 in panel A) and gCRBC (region R3 in panel A). (E) Green fluorescence from mixture of Quantum 24 FITC beads from Flow Cytometry Standards. (F) Green fluorescence histogram of osmium-fixed CRBC (low population) and gCRBC (high population).
The guideline proposes a two-step procedure for instrument quality assurance. First, establish that the instrument performance is acceptable at a particular point in time. Then, monitor performance with stable materials under testspecific instrument conditions. No specific criteria for calibration or control materials are given. The guideline instead provides a pro-
cess that a laboratory can follow by using materials recommended or supplied by the instrument manufacturer or by establishing independent criteria and materials. (Further discussion of the principles of quality control is given in UNIT 3.1; applications of quality assurance in phenotyping and in nucleic acid analysis are covered in UNIT 6.1 & UNIT 7.2)
Flow Cytometry Instrumentation
1.3.9 Current Protocols in Cytometry
Supplement 32
Figure 1.3.5 Scatter (FS and SS) dot plot (A) and fluorescence histograms (B-D) of 2.49-µmdiameter beads stained with Nile red.
STANDARDIZATION AND CALIBRATION
Standardization, Calibration, and Control in Flow Cytometry
A flow cytometer may be characterized or standardized by running samples, such as manufactured particles, that have some known properties. Controlling the parameters and characteristics provides consistent results over time from one instrument but, unless the instrument was calibrated, does not necessarily allow results to be quantitatively compared with those from other instruments. Flow cytometers have many parameters that can be standardized and controlled (e.g., see Optical Alignment, see Light Scatter and Particle Sizing, see Fluorescence and Light Scatter Resolution, see Measurement Response and Logarithmic Calibration, see Sensitivity or Signal/Noise for Dim Fluorescence, and see Spectral Overlap Compensation). This section also discusses aspects of standardization relating to specific applications (see DNA Measurements, see Sorting Purity and Recovery, see Standardization with a Particle in the Analysis Sample, and see Particle Concentration). The final subsection provides a summary (see What
Measurements Can Be Calibrated, and How Frequently Is Calibration Necessary?). Actual procedures and guidelines for characterizing beads follow (see Characterizing Particles for Calibration and Control of a Flow Cytometer).
Optical Alignment Optical alignment is most critically assessed and most easily optimized using particles with very uniform scatter and fluorescence. The objective of optical alignment is to center the sample stream in the light beam and simultaneously image the intersection of sample stream and light beam through the detection optics. At optimal alignment, signal pulses will have maximum amplitude and minimum width, and be most reproducible. In a histogram of the fluorescence or scatter, the distribution will be at its highest and most narrow. The most uniform particles are generally smaller (1 to 3 µm) than typical cells. Although much more difficult to make, larger uniform particles are available. Fluorescence CVs <2% can be expected for alignment particles (see Fig. 1.3.5).
1.3.10 Supplement 32
Current Protocols in Cytometry
Light Scatter and Particle Sizing Scatter Light scatter is a difficult parameter to standardize because it depends critically on the scatter angles measured and the geometry of the collection optics. The scattering of particles has strong nonlinear dependence on the angles measured, the refractive index of the particle, as well as particle size, internal structure, and content (Salzman et al., 1990; Shapiro, 2003; Doornbos et al., 1994). It is even possible at certain scatter angles to have a smaller signal from larger particles. The refractive index of most manufactured particles is much larger than that of cells, and beads and cells of the same size have different scatter intensities. For these reasons it is not advisable to use scatter for quantitative cell sizing. Scatter is primarily useful for discriminating cells based on relative scatter properties (e.g., discrimination of lymphocytes, monocytes, and granulocytes is a common application). Although it is difficult to standardize light scatter across instrument types using manufactured particles, it is possible to standardize within a particular instrument type and to monitor the relative performance of an instrument. The relative and absolute scatter intensities of two different size particles provide one way to standardize scatter. The ellipsoidal shape of fixed chicken red blood cells (CRBC) produces a characteristic light scatter pattern with two peaks in the forward scatter distribution. Region R3 in Figure 1.3.4A shows an example of the light scatter distribution for CRBC. Although the scatter pattern for CRBC may vary among different instrument designs, it may be useful for monitoring a particular instrument (Horan and Loken, 1985; Horan et al., 1990). At the present time, however, the most reliable approach is to use a biological sample of the same or similar type that is to be used for the flow cytometric analysis (NCCLS, 1998b; Owens and Loken, 1995).
Pulse width Pulse widths of fluorescence or scatter signals depend on both the height of the laser beam (in the direction of flow) and the size of the particle passing through the beam. When the particle diameter is at least as large as the beam height, it is possible to measure particle diameter accurately (Sharpless and Melamed, 1976; Sharpless et al., 1977; Eisert and Nezel, 1978; Leary et al., 1979; Shapiro, 2003). A mixture of two or more sizes of particles of
known, calibrated diameter can be used to calibrate this measurement.
Electronic cell volume Electronic cell volume (Schwartz et al., 1983; Kachel, 1990; Shapiro, 2003) is provided as a parameter in some commercially available flow cytometers. Microbeads are an excellent material for standardization and calibration of electronic cell volume. Accurate measurement of particle volume requires the particles to be electrically nonconductive. Thus, caution should be used when comparing results with fixed cells in which the plasma membrane has become permeable and electrically conductive.
Fluorescence and Light Scatter Resolution Resolution is usually estimated by running a sample of uniform particles and measuring the CV. For fluorescence measurements, CV is a good estimate of resolution because fluorescence signals are generally proportional to the amount of fluorochrome on the particle (i.e., the measurement and the characteristic of the particle are linearly related; see UNIT 1.4 on system linearity). Resolution of dimly stained particles is considered a special case (see Sensitivity or Signal/Noise for Dim Fluorescence). For light-scatter measurements, the CV obtained with one particle may not be a good measure of how well the cytometer may be able to resolve particles of different sizes, since the relationship between light-scatter signal and particle size is usually not linear.
Measurement Response and Logarithmic Calibration The measurement response of a flow cytometer can be determined in relative or absolute terms. Evaluating the relative response can be as simple as determining whether two identical particles stuck together give twice the signal as one. Alternatively, it may be as complex as determining the response of a logarithmic amplifier over a four-decade range. Absolute response calibrates the measurement (e.g., channel number in a histogram) in units such as molecules of equivalent soluble fluorochrome (MESF) or antibody binding capacity (ABC).
Linear and logarithmic response Testing the relative linearity of a measurement on a linear scale is conveniently done using small beads or other stained particles such
Flow Cytometry Instrumentation
1.3.11 Current Protocols in Cytometry
Supplement 32
as cell nuclei that contain aggregates. Measurement of the sample should then give histogram distribution mean channels that are multiples of the mean channel of a single particle. Measuring the response of a nominally logarithmic scale can be done in two ways. The first uses a known linear scale as reference and compares measurements on the linear and logarithmic scales (Muirhead et al., 1983; Horan et al., 1990). The second approach uses a mixture of particles with known, different intensities. The log response is determined by measuring the separation of the peaks on the log scale as the signal level is varied by changing the PMT voltage (Schmid et al., 1988). A more convenient version of the second approach uses a bead mixture containing a wide range of fluorescence levels whose relative intensities are known (e.g., see Fig. 1.3.4C). If the beads in the mixture are calibrated in terms of MESF, then the absolute log response may be determined (Schwartz and FernandezRepollet, 1993; Schwartz et al., 1996). Thorough discussion of establishing and maintaining system linearity appears in UNIT 1.4.
Secondary calibrators Cross-calibration of gCRBC or hard-dyed particles to surface-labeled beads calibrated in MESF can generally be done for any one instrument (Schwartz et al., 1996; see Characterizing Particles for Calibration and Control of a Flow Cytometer). To ensure that the secondary calibrator on an instrument is reliable, however, the emission filters and the wavelength and intensity of the excitation light must remain unchanged. See Figures 1.3.1 and 1.3.2 and cautions given by Schwartz et al. (1996). (Also see Types of Particles and discussion of comparison of spectrally matched and unmatched particles.)
Antibody binding capacity (ABC)
Standardization, Calibration, and Control in Flow Cytometry
A further step in immunofluorescence standardization and calibration is to express measurement results in terms of antibody binding capacity. Three approaches have been used to estimate ABC. Each approach has different critical technical requirements and potential sources of error. Although not, strictly speaking, a source of error, it must be kept in mind that different antibody clones with the same cluster designation (CD) can have different binding capacity. Particular examples of clone variability have been noted for CD4 (Davis et al., 1998) and CD34 (Serke et al., 1998). Therefore, if the three approaches to quantita-
tive ABC are to be compared, they should be compared with the same clone or with clones that are demonstrated to give the same ABC. In addition, the sample preparation method can affect the antibody binding and must be taken into consideration (Islam et al., 1995; Serke et al., 1998). The earliest approach (quantitative indirect immunofluorescence, or QIFI) uses a calibrated anti-mouse fluorescent second-step reagent (Poncelet and Carayon, 1985; Bikoue et al., 1996). The anti-mouse reagent can be calibrated using particles coated with known amounts of mouse immunoglobulin G (IgG). Polyvalent reagents are used in this method, however, and their reactivity with certain monoclonal mouse antibodies may produce artifacts. Bikoue et al. (1996) observed that different monoclonal antibodies to a molecule (e.g., CD8) can give different ABC values on the same cells. Whether this is due to differences in binding of the primary antibody to cells or to differences in binding of the secondary antibody to the primary antibody is not clear. Altered reactivity of the second-step reagent with fluorochrome-conjugated mouse monoclonal antibody is another possible variable. A second approach (Quantum Simply Cellular or QSC) uses particles coated with known amounts of polyvalent anti-mouse antibody (Schwartz and Fernandez-Repollet, 1993; Schwartz et al., 1996). The QSC method is designed to capture quantitatively any mouse monoclonal antibody independent of fluorochrome conjugation or IgG isotype. However, conjugates of the same monoclonal antibody with different fluorophores give different ABC values when calibrated with QSC beads, and different antibody clones directed against the same molecule can give differing results with QSC beads (Lenkei and Andersson, 1995). The QSC beads could, however, be calibrated to a specific monoclonal antibody reagent whose production is carefully controlled (Lenkei and Andersson, 1995; A. Schwartz, pers. comm., 1996). A third method uses antibody conjugates that have been prepared with a known MESF/antibody ratio and a flow cytometer that has been calibrated in MESF. Phycoerythrin is an attractive fluorochrome for this approach since antibody conjugates can be prepared with exactly one PE molecule per antibody. Initial experiments with this approach are promising (Davis et al., 1998; Iyer et al., 1998). Certain cell-surface markers may be useful as biological calibrators with a relatively
1.3.12 Supplement 32
Current Protocols in Cytometry
small variability and uncertainty. Although not as stable or reproducible as the amount of DNA per cell, the amount of CD4, CD45, and many other molecules on normal human lymphocytes is generally reproducible (Brown et al., 1986; Poncelet et al., 1991; Bikoue et al., 1996); thus, CD4 content may be a useful biological calibrator for immunofluorescence analogous to use of normal lymphocytes or other defined nucleated cells as biological calibrators for DNA quantitation (Hultin et al., 1998). The QIFI and QuantiBRITE methods have been found to be generally comparable (Lenkei et al., 1998; Serke et al., 1998) for ABC quantitation, but the QSC method frequently gives significantly different results from the other methods (Lenkei et al., 1998; Serke et al., 1998). Since the QuantiBRITE method uses direct fluorescence staining it can be easily used in a multicolor staining protocol. The general recommendation for comparing ABC across laboratories or over time is to use a single method along with the manufacturers’ recommended reagents.
Sensitivity or Signal/Noise for Dim Fluorescence The practical notion of fluorescence sensitivity involves the ability to resolve populations of dimly fluorescent particles. The factors affecting the ability to reliably detect dim fluorescence include electronic noise, the amount of fluorescent light collected, the efficiency with which the fluorescent light is converted into electrons in the detector, and the amount of background light that is present. Characterization of an instrument’s detection capability should include the effect of these factors. An overly simplified method for characterizing fluorescence sensitivity can give misleading results, as illustrated by the following example.
Resolution of dimly stained from unstained particles Three methods for assessing fluorescence sensitivity are illustrated in Figure 1.3.6. Data for each row of panels were obtained with exactly the same instrument conditions with a single sample that contained all the particle
Figure 1.3.6 Comparison of three methods for estimating green fluorescence sensitivity. Histograms are four-decade log with 256 channels per decade. Each column of histograms represents a different measure of fluorescence sensitivity. (A-E) Histograms of autofluorescent unstained CaliBRITE beads (low peak) and gCRBC (high peak). (F-J) Histograms of background noise from a test pulse trigger (low peak) and gCRBC (high peak). (K-O) Histograms of autofluorescent unstained CaliBRITE beads and dim Rainbow beads. Each row of histograms shows results for a different condition or perturbation on the optical system of the instrument (see Sensitivity or Signal/Noise for Dim Fluorescence).
Flow Cytometry Instrumentation
1.3.13 Current Protocols in Cytometry
Supplement 32
and test pulse events. The fluorescence histograms were gated to contain the events of interest by setting regions in a scatter dot plot, as in Figure 1.3.4A. The left column in Figure 1.3.6 (panels A-E) are histograms of unstained “autofluorescence” beads and glutaraldehydefixed CRBC (gCRBC). The middle column (panels F-J) shows histograms of background light (from test pulse–triggered fluorescence) and the same gCRBC. The right column (panels K-O) shows histograms of autofluorescence beads and dimly stained beads. The instrument condition for each row of histograms was varied by adding background light or reducing the amount of fluorescent light that reached the PMT.
Delta channel and detection threshold methods: Traditional and not very sensitive
Standardization, Calibration, and Control in Flow Cytometry
The “delta channel” approach measures the mean or median channel of autofluorescent or nonfluorescent particles (or test pulse– triggered background) and that of relatively bright particles. The approach is illustrated in Figure 1.3.6A-E, where the negative population is a bead with about the same autofluorescence as unstained lymphocytes and the bright population is gCRBC. With autofluorescence as the negative reference there is almost no difference in mean channel or delta channel for the different instrument conditions. There are considerable differences, however, in the ability of the instrument to resolve autofluorescent beads from dimly stained beads (Fig. 1.3.6K-O). For very dim populations, the delta channel method does not critically assess the ability to resolve unstained from dimly stained particles. This is primarily because the method does not take into account the broadness of the fluorescence distributions; rather, it considers only the mean or median channels. Resolution of the dimly stained population, however, is critically dependent on the broadness of the distributions, which is determined by background fluorescence and the amount of fluorescence signal reaching the PMT. If nonfluorescent particles or test pulse– triggered fluorescence is used as the negative population, one obtains additional information about sensitivity. For Figure 1.3.6F-J, the instrument was configured to have the same response for bright particles but different amounts of background signal. In the terminology used by Schwartz et al. (1996), the instrument had the same “window of analysis” for data in Figure 1.3.6F-J. Comparing
panels F-J with panels K-O in Figure 1.3.6, one sees an improved ability of this method to predict whether dimly stained particles will be resolved from unstained particles. The method is not perfect, however. Panels L and M have the same amount of background signal or, as expressed by Schwartz et al. (1996), the same “detection threshold,” but they differ noticeably in resolution of the autofluorescent and dimly stained particles. The same considerations hold for panels N and O of Figure 1.3.6, which have the same, intermediate amount of background signal. So the detection threshold method also fails to reliably measure differences in the ability to resolve dim populations from unstained.
Resolution of dimly stained from unstained particles The most direct measure of the ability to resolve unstained from dimly stained particles is simply to run a mixture of the particles and see if they are resolved. Panels K-O in Figure 1.3.6 illustrate this approach. This method at least guarantees unambiguously that the instrument is able to resolve a certain low level of fluorescence. Making this approach quantitative requires not only attention to the relative numbers of unstained and dimly stained particles, but also some limits on the inherent particle fluorescence CV. By calibrating the unstained and dimly stained particles in MESF, one could have a method that would standardize and calibrate sensitivity.
The Q and B method to characterize instrument sensitivity To address the quantitative characterization of fluorescence sensitivity, the underlying physics of optical detectors needs to be taken into account (Wood and Hoffman, 1998). A theoretical model for the CV of dimly fluorescent particles (Gaucher et al., 1988; Steen, 1992) can be developed based on optical detection efficiency, Q (number of photoelectrons per fluorochrome molecule analyzed), and background light, B. B can be expressed in units of the equivalent number of fluorochrome molecules that would produce that background. The studies by Gaucher et al. (1988) and Steen (1992) used flashes from a light-emitting diode (LED) to produce dim signals to a flow cytometer detector and used the CVs of the resulting signals to determine Q. The contribution to the CV from background light was determined by comparing CVs that were obtained from LED flashes with the laser
1.3.14 Supplement 32
Current Protocols in Cytometry
shining on the sample stream to those obtained with the laser blocked. A fluorescence intensity standard bead was used to calibrate the flow cytometer in MESF units. More recently, it has been shown that sets of beads with uniform but dim fluorescence could be used instead of LED flashes to determine Q and B (Chase and Hoffman, 1998; Wood, 1998). The intrinsic CVs of the dim beads could be determined by comparing their measured bead CVs with CVs of LED light flashes and CVs of identical but brightly stained beads (Chase and Hoffman, 1998). Thus, a set of beads stained at varying levels from dim to bright could be characterized for intrinsic properties and then used to measure Q and B in a flow cytometer. Q and B can then be determined by using a fluorescence intensity standard to calibrate the flow cytometer in terms of MESF and measuring the CVs (or SDs) of the dimly and brightly fluorescent beads. There are several ways to analyze the resulting bead data to determine Q and B (Chase and Hoffman, 1998; Wood, 1998). A robust method for determining Q is to correct the standard deviations of the beads for the illumination uniformity contribution to SD (this is determined from the CV of a bright bead). A plot of standard deviation squared (SD2 ) versus the mean bead intensity in MESF units gives a straight line whose slope is 1/Q. An estimate of B can be determined from the intercept of the line with the SD2 axis. The intercept is B/Q. Alternatively, and probably more accurately, B can be determined from the SD of a blank bead or noise distribution (Chase and Hoffman, 1998). An important reason for characterizing sensitivity in terms of Q and B is that these values can be used to predict the resolution of dimly fluorescent populations of cells (Chase and Hoffman, 1998). Q and B can also be used to determine the effect of background contributions from unbound fluorescent antibody or from spectral overlap on the resolution of populations. Pages 221 to 223 in Shapiro (2003) give a good overview of the method and and describe the utility of Q and B.
Spectral Overlap Compensation Measurement of fluorescence in a particular spectral range is not the same as measuring the fluorescence from a fluorochrome. One usually measures fluorescence in a spectral range that contains the emission peak of the fluorochrome, but emission from other fluorochromes used simultaneously to stain the cells may also overlap into the desired spectral
region. In Figures 1.3.1 and 1.3.2, the spectral overlap of FITC fluorescence extends into the range where PE has maximum emission at about 575 nm. Similarly, PE has a small amount of fluorescence in the range where FITC has peak emission. The effect of the spectral overlap is illustrated in Figure 1.3.7A, which shows a dot plot of yellow and green fluorescence from a mixture of unstained, FITCstained, and PE-stained beads. To make the dot-plot axes read in units of FITC and PE fluorescence rather than green and yellow fluorescence, the amount of spectral overlap from each fluorochrome can be subtracted (Bagwell and Adams, 1993). The method used to accomplish the subtraction is instrument dependent; ask the manufacturer for details. Since the percentage of FITC fluorescence in the yellow detector is always a constant fraction of the amount of FITC fluorescence in the green detector, the same percentage of green FITC fluorescence can be subtracted from the yellow signal no matter what the FITC signal is. For example, PE fluorescence = yellow fluorescence − f × green fluorescence, where the fraction f is a constant. Figure 1.3.7B shows the same sample as in Figure 1.3.7A but with spectral compensation applied. Spectral overlap compensation simply transforms the readout of the fluorescence from green and yellow to something more directly related to the analysis results: the amount of FITC and PE fluorescence. Correctly adjusting the fraction of compensation requires care, however, since small differences in the emission spectrum of the particle can have a large effect on the amount of compensation (Schwartz and Fernandez-Repollet, 1993; Schwartz et al., 1996). Manufactured particles stained with fluorochromes such as FITC or PE may not have exactly the same emission spectrum as cells labeled with the same fluorochromes. If beads are used to adjust compensation, it is always advisable to check compensation with labeled cells at least once for each new batch of beads (NCCLS, 1992; Schwartz and Fernandez-Repollet, 1993; Owens and Loken, 1995). Antibody capture beads are an alternative to using fluorochrome-stained beads or stained cells for compensation. It is advisable to stain the antibody capture beads with the same antibody conjugates that will be used to stain the cells. Antibody capture beads are especially useful as compensation control samples for multicolor analysis, since the unstained reference particle is identical for all fluorochromes and all fluorochromes are controlled with a
Flow Cytometry Instrumentation
1.3.15 Current Protocols in Cytometry
Supplement 32
Figure 1.3.7 Example of spectral overlap with FITC- and PE-stained beads. Three populations are shown in each dot plot: unstained beads, FITC-stained beads, and PE-stained beads, with unstained beads appearing in the lower left quadrant. (A) Uncompensated data with FITC-stained beads in the upper right quadrant with about 300 units of green fluorescence and 80 units of yellow fluorescence. PE-stained beads are in the upper left quadrant with about 8 units of green fluorescence and 800 units of yellow fluorescence. Unstained beads have about 6 units of green and 5 units of yellow fluorescence. (B) Same data after compensating for spectral overlap to make the dot-plot axes represent fluorescence from a specific fluorochrome rather than a color of fluorescence signal. Note that the FITC-stained beads in the lower right quadrant of B have the same mean “PE” fluorescence as the unstained beads, and the PE-stained beads in the upper left quadrant have the same amount of “FITC” fluorescence as unstained beads. FITC beads have no more PE signal than unstained beads, and PE beads have no more FITC signal than unstained beads.
stained population that is relatively uniform and bright. When more than two fluorochromes are used simultaneously to stain cells, compensation becomes significantly more complex. If the third and fourth fluorochromes in the sample have little or no spectral overlap, compensation can be set manually. However, if three or more fluorochromes all have significant spectral overlap among one another, correct compensation requires a matrix calculation that is practical only using software (Bagwell and Adams, 1993; Roederer, 2001). In instruments that use log amplifiers, a combination of hardware and software compensation may give the most accurate results (Baumgarth and Roederer, 2000). With compensation performed by software and with proper controls, accurately compensated results are possible and practical with ten or more fluorochromes (Baumgarth and Roederer, 2000).
DNA Measurements
Standardization, Calibration, and Control in Flow Cytometry
For DNA measurements, fluorescence linearity and resolution must be assured. Sample preparation and data analysis must also be carefully controlled. Reviews of standardization issues for DNA analysis include those by Dressler (1990), Bauer (1993), Darzynkiewicz
(1993), and Wheeless (1993); also see Chapter 7 for nucleic acid analysis. Approaches to performing instrument standardizations were discussed earlier (see Optical Alignment, see Fluorescence and Light Scatter Resolution, and see Measurement Response and Logarithmic Calibration: Linear and logarithmic response). Either fluorescent beads (see Fig. 1.3.5) or stained cells or nuclei can be used. For determination of abnormal DNA content it is important to use an internal staining control in the sample. Chicken erythrocytes, trout erythrocytes, and normal human cells have been used for internal controls. Calibrated measurement of the amount of DNA per cell has been reported (see Shapiro, 2003, p. 316, for a brief review), but sensitivity of the staining to chromatin structure should caution against overinterpretation of the results (Darzynkiewicz, 1993).
Sorting Purity and Recovery Various manufactured particles can be used to determine purity and recovery for sorting. A mixture of beads with different fluorochromes is typically used, but using only two types of stained particles can give results that are overly optimistic. This is because a mixture of two differently stained beads has “built-in” doublet detection. For example, if two different stained
1.3.16 Supplement 32
Current Protocols in Cytometry
beads with primary fluorescence in fluorescence channels 1 and 2, respectively, are measured together (coincident in time) the event is easily discriminated in a dot plot of channel 1 versus channel 2 fluorescence. This coincident event appears as doubly stained with signal in both fluorescence channels 1 and 2. A sort gate that was set only for a singly stained bead (i.e., in channel 1 only) would exclude these coincidences. If the real sample to be sorted has unstained cells that produce undetectable coincidences with stained cells, the sort electronics will either (1) detect the coincidence, abort the event, and reduce yield; or (2) miss the coincidence, sort the coincidence, and give lower purity. If one of the beads is unstained, however, it is not possible to detect a coincidence in a dot plot of channel 1 versus channel 2.
Standardization with a Particle in the Analysis Sample Adding the particle used for standardization or control to the sample to be analyzed gives an extra level of confidence to a flow cytometric analysis. Fluorescent beads are generally used for “standardization in tube” methods (Horan and Loken, 1985; Horan et al., 1990). The standardization or control particle must be sufficiently different in some measurement parameters to be distinguished from the cells in the sample. For example, the particles can be chosen to have lower forward scatter but much higher fluorescence than the cells in the sample. If a known number of particles is added to a known volume of sample, one can also use the particles to measure the concentration of cells (cells per volume) in the sample (see Particle Concentration).
Particle Concentration Particle concentration (i.e., the number of particles per unit volume) is becoming a more widely used measurement. This is largely motivated by the clinical need to measure the concentration of CD4 cells in HIV-positive individuals and in AIDS patients. In this context, the concept of “absolute count” has been used instead of cell concentration. The count is “absolute” in numbers per milliliter of the original blood sample, as calculated from the concentration of cells (e.g., CD4 T cells) for the sample analyzed and the known dilution of blood used in preparing the sample. Reference samples of known particle concentration can be used to standardize particle concentration measurements on a flow cytometer. For standardizing and calibrating the absolute count, a
controlled or calibrated dilution of the original sample must be made. Some instruments are capable of measuring particle concentration directly as they measure a fixed volume of sample. An alternative approach uses a known number, N, of reference particles added to a sample. The ratio, r, of sample particle events (e.g., CD4 T cells) to reference particle events is measured in the flow cytometer. Then the number of sample particles is computed from the product r × N (Stewart and Steinkamp, 1982).
What Measurements Can Be Calibrated, and How Frequently Is Calibration Necessary? Flow cytometry measurements that currently can be calibrated in at least some commercially available instruments are particle diameter or volume, numbers of fluorochromes (or MESF) per particle, antibody binding capacity (ABC), and particle concentration (e.g., particles per milliliter). Calibration need not be a daily practice, but the instrument must be quality controlled daily following calibration. Monitoring measurements with control material assure that the instrument is still in calibration as long as the measurements do not exceed acceptance limits determined by the application. If controls are out of range, it will be necessary to recalibrate. If the instrument is changed or serviced, it is usually advisable to recalibrate.
CHARACTERIZING PARTICLES FOR CALIBRATION AND CONTROL OF A FLOW CYTOMETER This section gives suggestions for how to assign fluorescence values to test beads (see Calibrating Particle Fluorescence in MESF), and for assignment and use of inherent fluorescence CV of a particle (see Determining Inherent Fluorescence CV of a Dim Particle Sample and see Measuring Signal to Noise from Dim Particles).
Calibrating Particle Fluorescence in MESF Assigning MESF to fluorescent particles Relatively bright beads or other particles of known concentration can be measured in terms of molecules of equivalent soluble fluorochrome (MESF) with a spectrofluorometer using a solution of fluorochrome as reference (Brown et al., 1986; Schwartz and FernandezRepollet, 1993). The spectrofluorometer is
Flow Cytometry Instrumentation
1.3.17 Current Protocols in Cytometry
Supplement 32
Standardization, Calibration, and Control in Flow Cytometry
adjusted for the excitation and emission of the fluorochrome to be measured. Depending on how close the excitation and fluorescence emission wavelengths are, it may be necessary to use a band-pass filter in the excitation light path and a high-pass or band-pass filter in the emission path to adequately reduce light scatter from the particles. The signal from a reference concentration of fluorochrome is measured, and then a suspension of the particles to be calibrated is measured under exactly the same conditions as the fluorochrome solution. The fluorescence of the particle suspension is expressed in terms of molecules of fluorochrome by comparing it with the reference fluorochrome solution. The particle concentration (particles/ml) is determined, correcting for doublets or aggregates if necessary. The MESF per particle is calculated as the fluorochrome concentration equivalent of the bead suspension divided by the particle concentration. The US National Institute of Standards and Technology (NIST) has published a series of papers (Gaigalas et al., 2001; Schwartz et al., 2002; Wang et al., 2002) detailing the fundamental scientific basis and reference methods for assigning MESF values to particles. NIST has also developed a standard fluorescein solution, Standard Reference Material 1932, and a set of FITC-stained beads, Reference Material 8640, which can be used to calibrate spectrofluorometers and flow cytometers. At low particle MESF the spectrofluorometer may become inaccurate or lack sensitivity. If the log response for the fluorescence channel of the flow cytometer has been carefully calibrated (see Measurement Response and Logarithmic Calibration; also see UNIT 1.4), then the brighter particles can be used to calibrate the upper range of the fluorescence channel in MESF per linear fluorescence unit on the histogram. Alternatively, the same process can be done using linear amplification. Dimmer particles can now be assayed on the calibrated flow cytometer and assigned MESF values.
Using hard-dyed particles as a secondary calibrator on one flow cytometer
Assigning relative intensity values to hard-dyed beads
The inherent fluorescence coefficient of variation (CV; see Definitions, Coefficient of variation) of a particle sample should be due only to variation in the amount of fluorochrome in the particles. Measurement limitations or noise in the flow cytometer broaden the measured CV (see Steen, 1992 and Chase et al., 1998, for more detailed discussion). The contribution to the CV from variation in illumination can be estimated by running a very bright, uniform “alignment” particle. For
A flow cytometer can also be used to measure and assign accurate relative fluorescence intensity values to hard-dyed particles such as Rainbow beads from Spherotech. For example, a flow cytometer with calibrated logarithmic or linear amplifier could be used to measure fluorescence from each peak in a fluorescence histogram for a mixture of particles. Figure 1.3.4C shows data for Spherotech Rainbow beads.
A mixture of stable, hard-dyed particles with a range of intensity levels and a sample of fluorochrome-labeled particles with known MESF values are used to calibrate a flow cytometer as follows: 1. Calibrate or verify calibration of the electronic response of the data acquisition electronics. Alternatively, a mixture of fluorochrome-labeled beads (e.g., labeled with FITC) of varying, known MESF can be used to calibrate the intensity scale (Schwartz et al., 1996). 2. Adjust the PMT voltage so the fluorochrome-labeled bead(s) of known MESF is in an appropriate histogram channel(s). This calibrates the histogram scale in MESF. 3. Run the hard-dyed bead mixture at the same PMT voltage as the fluorochromelabeled beads. 4. Use the histogram calibrated in MESF to assign MESF values to each of the harddyed bead populations. The hard-dyed beads are now a calibrator or “secondary standard” for this instrument and only this instrument. On subsequent days and at other PMT voltage settings the hard-dyed beads can be used to calibrate the fluorescence histogram in MESF.
Cautions using hard-dyed particles as MESF calibrators Because hard-dyed particles do not have exactly the same excitation and emission spectra as a fluorochrome used to stain cells, they will not necessarily have the same fluorescence intensity relative to that fluorochrome on another instrument. Caution and skepticism should be used in trying to assign to the hard-dyed beads a “global” MESF value that is valid for calibration in terms of MESF on all instruments (Schwartz et al., 1996).
Determining Inherent Fluorescence CV of a Dim Particle Sample
1.3.18 Supplement 32
Current Protocols in Cytometry
dimly fluorescent particles, background noise and photoelectron statistics become dominant contributors to the CV. To determine the inherent CV of a dim particle, where the total CV may be ≥10%, one must measure the background noise and photoelectron statistics. This can be done by using dim light flashes from a light-emitting diode (LED). Dim signals are created by using filters to attenuate the light from the LED or by simply holding the LED far from the detector. The inherent CV of the dim particles is found by subtracting (in quadrature) the noise CV from the total CV (Steen, 1992). The inherent CV of dim, hard-dyed beads can be determined in this way, effectively calibrating the particles in terms of population CV. When the particles are subsequently analyzed on any flow cytometer, the noise contribution from the measurement can be determined by the broadening of the CV. Essentially one works in reverse from what was explained above: the noise CV is determined by subtracting (in quadrature) the inherent particle CV from the total measured CV. For dim signals, this gives a measure of the fluorescence sensitivity in terms of conventional signal/noise.
Measuring Signal to Noise from Dim Particles In engineering, a standard definition of minimum resolvable signal is that for which the signal (S) and noise (N) are equal, that is, S/N = 1. In flow cytometry, it is customary to use CV rather than S/N, but the simple relation CV = N/S can be used to translate between the two measures. If the inherent CV of a dim particle is known (see Determining Inherent Fluorescence CV of a Dim Particle Sample), the system noise can be determined. The fluorescence of the dim particle is measured on the flow cytometer, and the CV of the resulting distribution is determined. Although a range of particle intensities might be required to measure system noise accurately, a single particle can be used to determine whether S/N at a particular MESF is above a required minimum. Noise can also be expressed in MESF, and the MESF level at which S/N = 1 or some other predetermined number is a measure of sensitivity. The instrument noise determined with the dim particle sample correlates with the ability to resolve dim particles from unstained particles. The other important factor is how large a signal is produced by unstained particles or cells. A convenient measure of the response to
truly nonfluorescent cells is provided by the test pulse mode available on many flow cytometers. The pulse detection electronics can be triggered by an electronically generated signal to a nonfluorescence parameter such as forward scatter. The fluorescence channels then measure the response only to background light and other noise sources.
CONCLUDING REMARKS Flow cytometry is a relatively young technology that has had rapid growth since the mid 1980s. It has moved from a technology that only a few hundred “initiated” experts understood and could use to become both a common laboratory tool and clinical diagnostic system. Improved standardization and calibration will help secure the value of this information-rich instrument and move it more rapidly into new biological and clinical applications.
Literature Cited Bagwell, C.B. and Adams, E.G. 1993. Fluorescence spectral overlap compensation for any number of flow cytometry parameters. Ann. N.Y. Acad. Sci. 677:167-184. Bauer, K.D. 1993. Quality control issues in DNA content flow cytometry. Ann. N.Y. Acad. Sci. 677:59-77. Baumgarth N. and Roederer M. 2000. A practical approach to multicolor flow cytometry for immunophenotyping. J. Immunol. Methods 243:77-97. Bevington, P. 1969. Data Reduction and Error Analysis for the Physical Sciences. McGraw-Hill, New York. Bikoue, A., George, F., Poncelet, P., Mutin, M., Janossy, G., and Sampol, J. 1996. Quantitative analysis of leukocyte membrane antigen expression: Normal adult values. Cytometry 26:137147. Bohmer, R.M., Papaioannou, J., and Ashcroft, R.G. 1985. Flow-cytometric determination of fluorescence ratios between differently stained particles is dependent on excitation intensity. J. Histochem. Cytochem. 33:974-976. Brown, M.C., Hoffman, R.A., and Kirchanski, S. 1986. Controls for flow cytometers in hematology and cellular immunology. Ann. N.Y. Acad. Sci. 468:93-103. Chase, E.S. and Hoffman, R.A. 1998. Resolution of dimly fluorescent particles: A practical measure of fluorescence sensitivity. Cytometry 33:267279. Darzynkiewicz, Z. 1993. The cell cycle: Application of flow cytometry in studies of cell reproduction. In Clinical Flow Cytometry: Principles and Application (K.D. Bauer, R.E. Duque, and T.V. Shankey, eds.) pp. 13-40. Williams and Wilkins, Baltimore.
Flow Cytometry Instrumentation
1.3.19 Current Protocols in Cytometry
Supplement 32
Davis, K.A., Abrams, B., Iyer, S.B., Hoffman, R.A., and Bishop, J.E. 1998. Determination of CD4 antigen density on cells: Role of antibody valency, avidity, clones and conjugation. Cytometry 33:197-205.
Kachel, V. 1990. Electrical resistance pulse sizing: Coulter sizing. In Flow Cytometry and Sorting, 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 45-80. Wiley-Liss, New York.
Doornbos, R.M., Deurloo, K.E., De Grooth, B.G., Sloot, P.M., and Greve, J. 1994. Lissajous-like patterns in scatter plots of calibration beads. Cytometry 16: 236-242.
Leary, J.F., Todd, P., Wood, J.C.S., and Jett, J.H. 1979. Laser flow cytometric light scatter and fluorescence pulse width and pulse rise time sizing of mammalian cells. J. Histochem. Cytochem. 27:315-320.
Dressler, L.G. 1990. Controls, standards, and histogram interpretation in DNA flow cytometry. Methods Cell Biol. 33:157-171. Eisert, W.G. and Nezel, M. 1978. Internal calibration to absolute values in flow-through particle size analysis. Rev. Sci. Instrum. 49:1617-1621.
Lenkei, R. and Andersson, B. 1995. Determination of the antibody binding capacity of lymphocyte membrane antigens by flow cytometry in 58 blood donors. J. Immunol. Methods 183:267277.
Fulwyler, M.J. 1979. Standards for flow cytometry. In Flow Cytometry and Sorting (M.R. Melamed, P.F. Mullaney, and M.L. Mendelsohn, eds.) pp. 351-358. John Wiley & Sons, New York.
Lenkei, R., Gratama, J.W., Rothe, G., Schmitz, G., D’hautcourt, J.L., Årekrans, A., Mandy, F., and Marti, G. 1998. Performance of calibration standards for antigen quantitation with flow cytometry. Cytometry 33:188-196.
Fulwyler, M.J., Perrings, J.D., and Cram, L.S. 1973. Production of uniform microspheres. Rev. Sci. Instrum. 44:204-206.
Loken, M.R., Parks, D.R., and Herzenberg, L.A.. 1977. Two-color immunofluorescence using a fluorescence-activated cell sorter. J. Histochem. Cytochem. 25:899-907.
Gaigalas, A.K., Li, L., Henderson, O., Vogt, R., Barr J., Marti, G., Weaver, J., and Schwartz, A. 2001. The development of fluorescence intensity standards. J. Res. Nat. Inst. Stand. Technol. 106:381389. Gaucher, J.C., Grunwald, D., and Frelat, G. 1988. Fluorescence response and sensitivity determination for ATC 3000 flow cytometer. Cytometry 9: 557-565. Henderson, L.O., Marti, G.E., Gaigalas, A., Hannon, W.H., and Vogt, R.F. 1998. Terminology and nomenclature for standardization in quantitative cytometry. Cytometry 33:97-105. Horan, P.K. and Loken, M.R. 1985. A practical guide for the use of flow systems. In Flow Cytometry: Instrumentation and Data Analysis (M.A. Van Dilla, P.N. Dean, O.D. Laerum, and M.R. Melamed, eds.) pp. 259-280. Academic Press, New York. Horan, P.K., Muirhead, K.A., and Slezak, S.E. 1990. Standards and controls in flow cytometry. In Flow Cytometry and Sorting, 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 397-414. Wiley-Liss, New York. Hultin, L.E., Matud, J.L., and Giorgi, J.V. 1998. Quantitation of CD38 activation antigen expression on CD8+ T cells in HIV-1 infection using CD4 expression on CD4+ T lymphocytes as a biological calibrator. Cytometry 33:123-132. Iyer, S.J., Hultin, L.E., Zawadzki, J.A., Davis, K.A., and Giorgi, J.V. 1998. Quantitation of CD38 expression using QuantiBRITE beads. Cytometry 33:206-212.
Standardization, Calibration, and Control in Flow Cytometry
Islam, D., Lindberg, A.A. and Christensson, B. 1995. Peripheral blood cell preparation influences the level of expression of leukocyte cell surface markers as assessed with quantitative multicolor flow cytometry. Cytometry 22:128134.
Muirhead, K.A. 1993a. Quality control for clinical flow cytometry. In Clinical Flow Cytometry: Principles and Application (K.D. Bauer, R.E. Duque, and T.V. Shankey, eds.) pp. 177-199. Williams and Wilkins, Baltimore. Muirhead, K.A. 1993b. Establishment of quality control procedures in clinical flow cytometry. Ann. N.Y. Acad. Sci. 677:1-20. Muirhead, K.A., Schmitt, T.C., and Muirhead, A.R. 1983. Determination of linear fluorescence intensities from flow cytometric data accumulated with logarithmic amplifiers. Cytometry 3:251256. NCCLS. 1998a. Terminology and Definitions for use in NCCLS Documents; Approved Standard NRSCL8-A. Clinical and Laboratory Standards Institute. Wayne, Pa. NCCLS. 1998b. Clinical Applications of Flow Cytometry: Quality Assurance and Immunophenotyping of Peripheral Blood Lymphocytes; Approved Guideline H42-A. Clinical and Laboratory Standards Institute. Wayne, Pa. NCCLS. 2004. Fluorescence Calibration and Quantitative Measurement of Fluorescence Intensity. Approved Guideline ILA24-A. Clinical and Laboratory Standards Institute. Wayne, Pa. Owens, M.A. and Loken, M.R. 1995. Flow Cytometry for Clinical Laboratory Practice: Quality Assurance for Quantitative Immunophenotyping. Wiley-Liss, New York. Owens, M.A., Vall, H.G., Hurley, A.A., and Wormsley, S.B. 2000. Validation and quality control of immunophenotyping in clinical flow cytometry. J. Immunol. Methods 243:33-50. Poncelet, P. and Carayon, P. 1985. Cytofluorometric quantification of cell-surface antigens by indirect immunofluorescence using monoclonal antibodies. J. Immunol. Methods 85:65-74.
1.3.20 Supplement 32
Current Protocols in Cytometry
Poncelet, P., Bikoue, A., Lavabre, T., Poinas, G., Parant, M., Duperray, O., and Sampol, J. 1991. Quantitative expression of human lymphocyte membrane antigens: Definition of normal densities measured in immunocytometry with the QIFI assay. Cytometry 5 (Suppl.):82-83. Purvis, N. and Stelzer, G. 1998. Multi-platform, multi-site instrumentation and reagent standardization. Cytometry 33:156-165. Rembaum, A. 1979. Microspheres as immunoreagents for cell identification. In Flow Cytometry and Sorting (M.R. Melamed, P.F. Mullaney, and M.L. Mendelsohn, eds.) p. 335. John Wiley & Sons, New York. Roederer M. 2001. Spectral compensation for flow cytometry: Visualization artifacts, limitations, and caveats. Cytometry 45:194-205. Salzman, G.C., Singham, S.B., Johnston, R.G., and Bohren, C.F. 1990. Light scattering and cytometry. In Flow Cytometry and Sorting, 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 81-107. Wiley-Liss, New York.
Serke, S., van Lessen A., and Huhn, D. 1998. Quantitative fluorescence flow cytometry: A comparison of the three techniques for direct and indirect immunofluorescence. Cytometry 33:179187. Shapiro, H. 2003. Practical Flow Cytometry, 4th ed. John Wiley & Sons, Hoboken, N.J.. Sharpless, T.K. and Melamed, M.R. 1976. Estimation of cell size from pulse shape in flow cytofluorometry. J. Histochem. Cytochem. 24:257264. Sharpless, T.K., Bartholdi, M., and Melamed, M.R. 1977. Size and refractive index dependence of simple forward-angle scattering measurements in a flow system using sharply focused illumination. J. Histochem. Cytochem. 24:257-264. Steen, H.B. 1992. Noise, sensitivity, and resolution of flow cytometers. Cytometry 13:822-830. Stewart, C.C. and Steinkamp, J.A. 1982. Quantitation of cell concentration using the flow cytometer. Cytometry 2:238-243.
Schmid, I., Schmid, P., and Giorgi, J. 1988. Conversion of logarithmic channel numbers into relative linear fluorescence intensity. Cytometry 9:533-538.
Wang, L., Gaigalas, A.K., Abbasi, F., Marti, G.E., Vogt, R.F., and Schwartz, A. 2002. Quantitating fluorescence intensity from fluorophores: Practical use of MESF values. J. Res. Nat. Ins. Stand. Technol. 107:339-353.
Schwartz, A. and Fernandez-Repollet, E. 1993. Development of clinical standards for flow cytometry. Ann. N.Y. Acad. Sci. 677:28-39.
Wheeless, L.L. 1993. The clinical utility of DNA cytometry. Ann. N.Y. Acad. Sci. 677:82-85.
Schwartz, A., Sugg, H., Ritter, T.W., and FernandezRepollet, E. 1983. Direct determination of cell diameter, surface area, and volume with an electronic volume sensing flow cytometer. Cytometry 3:456-458. Schwartz, A., Fernandez-Repollet, E., Vogt, R., and Gratama, J. 1996. Standardizing flow cytometry: Construction of a standardized fluorescence calibration plot using matching spectral calibrators. Cytometry 26:22-31. Schwartz, A., Marti, G.E., Poon, R., Gratama, J.W., and Fern´andez-Repollet. 1998. Standardizing flow cytometry: A classification system of fluorescence standards used for flow cytometry. Cytometry 33:106-114. Schwartz, A., Wang, L., Early, E., Gaigalas, A., Zhang, Y., Marti, G.E., and Vogt, R.F. 2002. Quantitating fluorescence intensity from fluorophore: The definition of MESF assignment. J. Res. Nat. Inst. Stand. Technol. 107:83-91.
Wood, J.C.S. 1993. Clinical flow cytometry instrumentation. In Clinical Flow Cytometry: Principles and Application (K.D. Bauer, R.E. Duque, and T.V. Shankey, eds.) pp. 71-92. Williams and Wilkins, Baltimore. Wood, J.C.S. 1998. Fundamental flow cytometer properties governing sensitivity and resolution. Cytometry 33:260-266. Wood, J.C.S. and Hoffman, R.A. 1998. Evaluating fluorescence sensitivity on flow cytometers: An overview. Cytometry 33:256-259.
Contributed by Robert A. Hoffman BD Biosciences San Jose, California
Flow Cytometry Instrumentation
1.3.21 Current Protocols in Cytometry
Supplement 32
Establishing and Maintaining System Linearity In order to quantitate the fluorescence or light scatter emission of a particle with a flow cytometer, it is necessary to have a flow cytometer system that is able to transform the received optical signal into an electrical signal. Accuracy of quantitation depends on how correctly the electrical signal amplitude represents the intensity of the received optical signal. Ideally, the electrical signal amplitude should be directly proportional to the received optical signal. Because of unavoidable offsets introduced in signal transformation by photodetectors, amplifiers, and other components, however, the electrical signal amplitude is in practice at least linearly related to the optical signal but not precisely proportional. In any event, it is important that the detection system be linear, as any deviation in linearity results in inaccurate measurements. Establishment and maintenance of system linearity are prerequisite to making accurate quantitative measurements with a flow cytometer. The many potential sources of offsets and nonlinearity in a flow cytometer system are found in all parts of the detection, amplification, and data acquisition subsystems. This commentary explores sources of system offsets and nonlinearity, and discusses ways they can be monitored and adjusted to improve system linearity. The goal of this unit is to help investigators evaluate performance of the hardware in flow cytometer detection systems. Calibrations involving factors over which investigators have more control, such as sample quality, reagents, and data analysis, are discussed elsewhere: UNIT 1.3 gives extensive guidelines for standardization, calibration, and control, including definition of terms; Chapter 3 covers quality control; and Chapter 10 deals with data processing and analysis. Tests for system linearity augment daily quality assurance assays performed to verify reproducibility of measurements (UNIT 1.3). Careful assessment of linearity and thorough characterization of the detection systems allow more meaningful comparisons of interassay results, whether obtained with the same instrument or a different cytometer.
DEFINITION OF LINEARITY A system is considered linear if the output is a linear function of the input. In the case of Contributed by James C. S. Wood Current Protocols in Cytometry (1997) 1.4.1-1.4.12 Copyright © 1997 by John Wiley & Sons, Inc.
a flow cytometer, fluorescence and light-scatter photons emitted by particles on illumination with a monochromatic light source are collected by a photodetector and serve as input. The input pulse of data involving light intensity versus time is reduced to a single output value. The output is a digital number which is intended to be linearly related to the number of photons collected by the photodetector. If the output of the flow cytometer were a direct reflection of the input, then the output values would be directly proportional to the input signal. If x photons produced an output value of 256, then it would be expected that 2x photons would produce a value of 512. If the output value is proportional to the input signal, then the ratio output value/input signal is constant. Mathematically, this is expressed as output value = k × input signal, where k is the constant of proportionality. This proportional response is illustrated as curve A in Figure 1.4.1. In the most general case, a linear system does not require output values to be proportional to the input values. A system can be linear without being proportional. A linear system that is not proportional may have some nonzero output for a zero input, but the output value will be still linearly related to the input value. Adding a constant offset to the equation of proportionality gives output value = k × input signal + offset, for which output values are no longer proportional to input values (see curve B in Fig. 1.4.1). For example, if the offset is 50, then in the previous example the output values for x and 2x photons would be 306 and 562, respectively. Thus, the ratio of the output values would be 1.84 instead of the expected 2.00. Because of the complex origin of an overall system offset, it is generally desirable to reduce to an insignificant value the contribution of each instrument subsystem to the system offset, to simplify data analysis. Thus, flow cytometer manufacturers design their systems to reduce the offset to essentially zero throughout the electronics, from detectors through data conversion in the data acquisition system. All real-life detection, amplification, and data acquisition systems, however, have at least a small constant offset. Any residual offset is typically apparent only at very low signal levels. In addition to electronic background noise, optical noise may be significant at lower signal
UNIT 1.4
Flow Cytometry Instrumentation
1.4.1
B
A
PMT signal
C
0 0
Fluorescence intensity
Figure 1.4.1 System response. Examples of proportional, linear, and nonlinear responses to a pulse of fluorescence light. Curve A shows a proportional relationship between pulse area and fluorescence intensity. Curve B shows a linear relationship but with an offset (there is a PMT signal even with no fluorescence light pulse). Curve C shows a nonlinear relationship between PMT signal and fluorescence signal at high levels of fluorescence intensity, including an offset.
levels and may contribute substantially to output signal as an offset. Thus, a detection system in a typical flow cytometer is more correctly described by the offset equation, particularly for low-level signals.
SOURCES OF NONLINEARITY IN A FLOW CYTOMETER Ensuring that a flow cytometer responds linearly requires proper choice of laser power, filters sets (UNIT 1.5), type of fluorochrome, and photomultiplier tube (PMT) voltage. This commentary includes guidelines for evaluating the linearity of detection system components, especially linear and logarithmic amplifiers (see Establishing System Linearity). Flow cytometer detection systems have many potential sources of nonlinearity: photodetectors (photomultipliers and photodiodes, which are affected by optical background noise), amplifiers, (linear amplifiers and logarithmic amplifiers), baseline restorers, peak sense and hold components, and analog-to-digital converters.
Photodetectors
Establishing and Maintaining System Linearity
Photodetectors are devices that are used to convert pulses of light into electrical pulses. The photodetectors used in flow cytometers are photomultipliers and photodiodes.
Photomultipliers Photomultiplier tubes (PMTs) combine the function of a photodetector and amplifier in a single device. Photons impinge on the photocathode material, which in turn generates photoelectrons. The photoelectrons are accelerated by a series of positively charged plates called dynodes, with each successive dynode being charged with a higher voltage than the last. As electrons encounter a dynode, more electrons are generated by a process called secondary emission. The resulting cascade of electrons reaches the anode and generates an electrical current. The current is converted to a voltage for subsequent processing by passing it through a resistor or a transimpedance amplifier. The number of electrons reaching the anode for each photoelectron generated represents the amplification factor of the PMT. Amplification factors in excess of 1,000,000 are possible. The charging voltage for all PMT dynodes and the anode is provided by a voltage divider chain and a high-voltage power supply. The magnitude of the charging voltage determines the gain of the PMT. Proper design of those system components ensures the linear response of the PMT, particularly for large-amplitude output pulses. The voltage divider is designed to maintain a stable voltage at each dynode with sufficient current to support the cascade of
1.4.2 Current Protocols in Cytometry
electrons. Variations in voltage across the chain of dynodes produce changes in PMT gain. Since the relationship between voltage and gain is nonlinear, small changes in voltage can produce large changes in gain. Enough current capacity must be available to accommodate the very bright light pulses. The quality of the high-voltage supply influences the overall stability of the voltage divider. Because the gain stability of the PMT depends on the stability of the voltage divider, the high-voltage supply powering the voltage divider must be very stable. There is a specific range of voltages over which the PMT/voltage divider circuit will produce linear output. Instrument manufacturers can provide details as to the optimal operating voltage range of PMTs in their instruments. Operation below the minimum operating voltage is most likely to arise when setting the linear amplifier to a high gain or when detecting a very bright fluorescence or light-scatter signal with the PMT voltage reduced. The first in-
stance (see Fig. 1.4.2A) is corrected by using a lower linear amplifier gain and increasing the voltage to provide more gain from the PMT. The second instance (see Fig. 1.4.2B) is corrected by using neutral density (ND) filters to reduce the amount of light reaching the PMT; again, the high voltage of the PMT is raised to compensate for the reduced light level. The other extreme of exceeding the maximum PMT operating voltage is likely when one tries to detect weakly fluorescent signals. As a note of caution, some flow cytometers may control the maximum PMT voltage so as not to damage the PMT; however, this control may not be available on all instruments, and users should consult the manufacturer to ascertain what the maximum voltage is for the PMT. The voltage should never be raised higher than this maximum voltage rating. The instrument manual or an instrument manufacturer applications specialist should be consulted for specific suggestions on applications and recommended PMT voltage settings.
A λ
PMT
λ
Pulse height
anode low voltage amplifier 100
Pulse height
anode correct voltage PMT
Light intensity
amplifier 2
Light intensity
B λ
PMT
λ
Pulse height
anode low voltage amplifier 1
anode correct voltage PMT
amplifier 1
Light intensity Pulse height Light intensity
ND filter
Figure 1.4.2 Proper settings for PMT anode voltage and amplifier gain. Panel A shows how the combination of low PMT anode voltage and high amplifier gain leads to a nonlinear response. Raising the PMT anode voltage and lowering the amplifier gain corrects the problem. Panel B shows how a low PMT anode voltage leads to a nonlinear response for large signals. Attenuating the light signal with an ND (neutral density) filter and increasing the PMT anode voltage produce a linear response.
Flow Cytometry Instrumentation
1.4.3 Current Protocols in Cytometry
Establishing and Maintaining System Linearity
The high-voltage power supply and the voltage divider need to supply sufficient current while maintaining a stable voltage on the PMT dynodes and anode to prevent changes in the PMT gain. Current demands are highest for very bright particles and for high throughput rates when there is insufficient time for the voltage divider chain to recover between particles. Manufacturers of flow cytometers are aware of these situations and design circuitry to provide sufficient current to the PMT for a wide range of applications. For investigators developing new applications, however, it is possible that the flow cytometer may be operated outside the designed operating range, and this may affect the data collected. For example, linearity may be affected by operating at higher than recommended flow rates or by analyzing heterogeneous particle populations with very bright particles.
the signal pulses and not be removed. To determine the relative intensities of two particles it may be necessary to determine the value of the noise-related offset. A simple way of determining the contribution of internal noise to the output signal involves covering the photodetector to exclude light and triggering the flow cytometer with an alternate signal. For example, to measure the internal noise level of the forward-light-scatter (FLS) diode sensor, one would cover the FLS diode and use the right-angle light scatter sensor to generate a trigger signal. With the gain set appropriately, if a distribution is generated in the FLS histogram, then it represents the contribution of the internal noise of the FLS photodiode to the output signal. If the histogram shows an accumulation in only the first channel, then the internal noise will not significantly affect output values.
Photodiodes Photodiodes are used in flow cytometers to measure bright fluorescence and light-scatter emissions. Photovoltaic, photoconductive, and avalanche diodes may be found in a flow cytometer. Photovoltaic diodes are used to detect the bright forward light scatter signal. Photoconductive diodes are used to detect the forward light scatter signal and the dimmer rightangle light scatter signal. Avalanche diodes are used to detect right-angle light scatter signals as well as bright fluorescence signals, e.g., fluorescently labeled DNA. Photodiode detection circuits are typically linear over a wide dynamic range and require no operator adjustment. Even though photodiodes are better than PMTs at collecting photons and converting them to electrical energy (i.e., have higher quantum efficiency), they have a high internal noise level. The internal noise level of photodiodes, called Schott noise, results from movement of thermal electrons within the diode and sets a bound to the lower detection limit. The electrical signal generated by photons colliding with the photodiode must exceed thermal noise levels to be detectable. In addition to reducing the usable sensitivity of the photodiode, internal noise increases the coefficient of variation (CV) of the measurement and introduces an offset that becomes significant at low signal levels. The detection system is still linear but is not proportional. Baseline restoration circuitry (see Baseline Restorers section) reduces some of the offset, but some noise may be of the same bandwidth as
Optical background noise Optical background noise does not affect the linearity of flow cytometer system measurements, but it does determine how proportional the system is. The received optical signal is a sum of the optical signal from the particle plus the optical background noise. Thus, the received signal intensity is not directly proportional to the intensity of the optical signal from the particle alone, particularly at lower particle optical signal levels. In addition, optical background noise increases the CV of the resultant signal (Steen, 1992; Pinkel and Steen, 1982). By designing instruments with AC amplifier coupling and baseline restorers, manufacturers can reduce the optical background contribution to the resultant signal. However, depending on the design of instrument components and the bandwidth of the noise, some or all of the optical noise still will be manifested as an offset in the final measurement value. The increase in the CV from the optical background noise remains despite reduction in the offset value (Steen, 1992). If the offset from the optical background noise is significant—for example, at low signal levels—then the data must be further processed before relative intensity comparisons can be made. This is a straightforward procedure for linear data but is complicated if the data are transformed by a nonlinear function (e.g., a logarithmic or power function). For linear data the offset needs only to be subtracted from each data value. Nonlinear transformation requires that the data be transformed back to linear intensity values, the noise removed, and the
1.4.4 Current Protocols in Cytometry
new values possibly retransformed. Such series of transformations can cause loss of resolution and diminish measurement accuracy in the data. It is better to reduce the optical background noise before it becomes part of the digitized data. The optical background noise can be reduced by proper filter selection and through proper alignment and maintenance of the instrument, particularly of the flow cell lenses and filters.
Amplifiers System linearity depends heavily on the proper operation of an instrument’s amplifiers. Every signal must be processed and/or amplified by an amplifier. Amplifiers have many internal adjustments to achieve linearity over a large dynamic range and require regular monitoring. It is important that the amplifiers be maintained according to the manufacturer’s recommended schedule. This may require adjustments following a manufacturer’s instructions and possible replacement if the adjustments are not successful in restoring correct operation of the amplifier. Linear amplifiers ideally convert input values of x and 2x into proportionally larger outputs of, say, 10x and 20x, respectively. Logarithmic amplifiers, on the other hand, produce outputs of 1x and 2x for inputs of 10x and 100x. Linear amplifiers are useful for studying populations with a narrow dynamic range because data analysis is simplified. Logarithmic amplifiers are better for populations with wide dynamic ranges, as they are more likely to detect all members. The choice of linear versus logarithmic amplifier depends on several factors: heterogeneity of cells to be studied, expected magnitude of the changes in light intensities, and the dynamic range of the amplifiers available. Linear amplifiers Linear amplifiers should (as the name implies) have a linear response, and, when adjusted properly for zero offset, they should even have a proportional response. If the signals measured were not pulses but constant or slowly varying voltage levels, then nonlinearities would be found only in poorly designed amplifiers. However, when amplifying pulses one must consider the amplifier bandwidth. The bandwidth of a linear amplifier plays the most important role in determining how the amplifier will respond to a pulse. A flow cytometer is composed of a number of linear amplifier integrated circuits called operational
amplifiers. There are many types of operational amplifiers, and a design engineer will choose one over others on the basis of a number of characteristics of each amplifier; relevant specifications can be obtained from the manufacturer. The small-signal and large-signal gain-bandwidth products, the most important characteristics for determining the linearity of pulse measurements, represent the extremes of performance of the operational amplifier. The large-signal gain-bandwidth product is the limiting feature because it is less than the smallsignal gain-bandwidth product, and it characterizes the response of the amplifier to large pulses at a gain of 1. The small-signal and large-signal gain-bandwidths may differ by as much as 10-fold. This is illustrated in Figure 1.4.3. Between the two extremes the gain-bandwidth of an amplifier is a nonlinear function of the pulse amplitude. A related characteristic is the slew rate, which is a measure of how fast the output of the amplifier can change. The large-signal gain-bandwidth product is measured in frequency units at a gain setting of 1. The gain-bandwidth for a typical amplifier in a flow cytometer is usually on the order of 10 megahertz (MHz). As the gain of the amplifier system is increased, the bandwidth of the system goes down proportionately. For example, for an amplifier with a gain-bandwidth product of 10 MHz, at a gain of 1 the bandwidth would be 10 MHz, but at a gain of 10 the bandwidth would be 1 MHz. Thus, the bandwidth of flow cytometer linear amplifiers is less at a gain of 100× than at a gain of 1×. The difference in bandwidth depends on the design of the linear amplifier system of the particular cytometer. A typical amplifier gain versus bandwidth curve is shown in Figure 1.4.4. In flow cytometry applications a wide range of pulse widths and amplitudes can be encountered, depending on excitation beam width, sheath stream velocity, and particle size (see UNIT 1.2 for a general description of flow cytometry fluidics). For a given amplitude wider pulses require a lower bandwidth than narrower pulses. Bandwidth requirements are related to the inverse of the pulse width. Given the pulseamplitude dependence of the gain-bandwidth product, there is a significant opportunity for nonlinearities to become apparent if the amplifiers are designed improperly or operated beyond the design constraints. This is illustrated in Figure 1.4.5. If the large-signal bandwidth at a given gain is inadequate for an application then a largeamplitude pulse would be rounded off and
Flow Cytometry Instrumentation
1.4.5 Current Protocols in Cytometry
would not reach the expected amplitude. Inappropriate rounding would introduce a nonlinearity that would become more noticeable as the pulse amplitudes approach the high end of the histogram scale. The nonlinearity is most likely to occur at high gains and with narrow pulses amplified to near full scale; however, it may still be present but less noticeable for other
combinations of pulse widths, pulse amplitudes, and gains. At low gains and with wider pulses this type of nonlinearity is less likely. In a well-designed commercial flow cytometer this type of nonlinearity is likely to be insignificant over the specified operating range of the instrument; however, if the instrument is operated outside the specified operating range,
Pulse height (V)
12
9
6
3
0 100
102
104 LSB 106 SSB 108 Bandwidth (Hz)
Figure 1.4.3 A typical pulse height versus bandwidth plot for an amplifier which could be used in a flow cytometer. The amplifier gain is set at unity. The large-signal bandwidth (LSB) is the maximum bandwidth at which the highest amplitude pulses are amplified. The small-signal bandwidth (SSB) is the bandwidth at which the smallest amplitude pulses are amplified.
108
Amplification
106
104
102
100 100
Establishing and Maintaining System Linearity
102
104 106 Bandwidth (Hz)
108
Figure 1.4.4 A typical gain versus bandwidth curve for an amplifier which might be used in a flow cytometer. Note how the bandwidth decreases as the amplification increases.
1.4.6 Current Protocols in Cytometry
then this type of nonlinearity would be more likely. Logarithmic amplifiers Analog logarithmic amplifiers share the same limitations as linear amplifiers with some significant additions. Logarithmic amplifiers use a semiconductor device as a nonlinear component to approximate the logarithmic function. For flow cytometry the logarithmic transformation is required to be accurate over four decades. This is a particularly challenging design problem, and usually at best one can ensure
only that the logarithmic transformation of the amplifier approximates a logarithmic function on the average and over a limited range. Thus, at any one point the amplifier may be inaccurate by a specified amount. Moreover, there are likely to be additional inaccuracies at the lower and upper regions of the logarithmic transformation due to design limitations. Accuracy of analog logarithmic amplifiers is affected by drifting of component values because of temperature and component aging, which cause a previously adjusted amplifier to move out of adjustment. As many adjustments
A
B
C
D
Pulse height
B A
D C Fluorescence peak intensity
Figure 1.4.5 Examples of dependence of amplifier performance on signal shape and amplifier bandwidth. The lower bandwidth requirements of pulses B and D allow the amplifier to linearly amplify the pulses. However, the narrower pulses A and C have higher bandwidth requirements and the large-signal bandwidth is exceeded by pulse A. The amplitude of pulse A would be lower than expected. On the other hand, the pulse C bandwidth requirement is within the small-signal bandwidth limit and this pulse is amplified correctly.
Flow Cytometry Instrumentation
1.4.7 Current Protocols in Cytometry
in a logarithmic amplifier are very sensitive, small changes in a component value due to temperature fluctuations or component aging can lead to large changes in the transformation response of the logarithmic amplifier. For quantitative measurements, logarithmic amplifiers need to be monitored on a time schedule recommended by the instrument manufacturer. To overcome the limitations of analog logarithmic amplifiers, some flow cytometer designers do the logarithmic transformation on the digitized data. The signals are amplified by linear amplifiers, digitized by a wide-dynamicrange analog-to-digital converter, and converted to logarithmic data by a digital lookup table or digital signal processor. Such logarithmic transformation of output values is not subject to the same limitations as when using analog logarithmic amplifiers on input signals (Auer et al., 1993)
Baseline Restorers Baseline restorers are designed to average out any offsets in analog processing of the data. They work by estimating the average amplitude of signal levels between pulses. Baseline restorers work best when the pulses are narrow and occur infrequently, because the percentage of time without pulses is maximized and the baseline restorer can more accurately determine the average amplitude of interpulse signal levels. If the pulses widen and/or the repetition rate increases, then the percentage of the time without pulses decreases. Depending on baseline restorer design, when the pulse-free fraction of time decreases below a certain point, then the baseline restorer cannot accurately estimate interpulse signal levels. In this case, the baseline restorer will overestimate interpulse signal levels and start overcorrecting the pulses, and the signal pulse amplitude will start to decrease in a rate-dependent manner. If there is a large amount of overcorrection, a negative offset will result, and a rate-dependent nonlinearity will be evident. Overcorrection is most commonly observed with increases in repetition rate. When testing for linearity at high repetition (sample) rates, it important to be sure that linearity is not dependent on a sample rate.
Peak Sense and Hold
Establishing and Maintaining System Linearity
Typically the peak sense and hold component does not contribute to system nonlinearity other than that usually attributed to a linear amplifier. If an instrument is modified to produce narrower pulses by changing the excitation optics (laser beam shape) or increasing
sheath pressure (velocity), however, then the peak sense and hold may not be able to track the sharper leading edge of the narrower pulses. Thus, the peak amplitude captured may not be equal to the pulse-height amplitude. Operators of modified instruments that produce narrower pulses are advised to consult the manufacturer or designer about the need to modify the peak sense and hold circuitry.
Analog-to-Digital Converters Analog-to-digital converters (ADCs) convert the output of the analog processing section into a digital equivalent for further processing and storage. Linearity of ADCs is described by the integral nonlinearity and differential nonlinearity. The integral nonlinearity determines the linearity of the whole ADC conversion scale. The differential nonlinearity is a measure of the uniformity of the histogram channels. Typically, ADCs in flow cytometers are sufficiently linear, and channel nonuniformity is <1%. If the ADC were to fail, however, then the ADC response would become nonlinear or histogram channel widths might change. Changing of histogram channel widths may cause gaps (or dropouts) in the histogram; in addition or instead, there may be unexpected peaks that are artifacts of the defective ADC. Correcting this problem involves replacing the ADC and possibly some of the support circuitry.
ESTABLISHING SYSTEM LINEARITY The easiest way to establish the linearity of a flow cytometer is to analyze particles that fluoresce or scatter light with known relative intensities. Absolute intensities of the particles need not be known; however, relative intensities must be known with an accuracy acceptable for the application for which the flow cytometer is being tested to run. The range of particle intensities is chosen to span the full histogram scale. By far the most common calibrators are fluorescent particles, because applications that require the linearity of the flow cytometer to be known are fluorescence based (e.g., quantitation of DNA, RNA, protein, and antigen, and determination of antigen density). The many types of fluorescence “standard” particles of biological or manufactured origin each have advantages and disadvantages. The biological particles used most commonly to determine the linearity of a flow cytometer are nucleated erythrocytes, typically of chicken or trout origin. Owing to their uniformity, significant difference in chromatin
1.4.8 Current Protocols in Cytometry
content from human cells, and widespread availability (Vindelov et al., 1983), chicken and trout erythrocytes are used extensively to verify the linearity of flow cytometers for DNA quantitation applications (Vindelov and Christensen, 1990). When used in standardizations, the particles are stained with the same DNA dye used in the DNA quantitation application and then run on the flow cytometer. The positions in the histogram of the populations of singlet, doublet, triplet, and quadruplet particle clusters are noted. The goal is to determine if the flow cytometer is both linear and proportional; thus, the expected result is for the doublet, triplet, and quadruplet particles to be two, three, and four times, respectively, the singlet peak location. Doing a linear regression of channel position versus number of particles in the multiplet would reveal any nonlinearity and/or offset (Ubezio and Andreoni, 1985). A disadvantage of nucleated erythrocytes as linearity calibrators is the difference in size among the singlet, doublet, triplet, and quadruplet clusters. As the particle clusters increase in size, the pulse width of the larger clusters would exceed that of a singlet particle, leading to a corresponding reduction in pulse bandwidth. A single particle with the same fluorescence intensity as a quadruplet cluster would have a higher pulse bandwidth than the cluster and thus may not be measured as having as much fluorescence intensity as an equivalent quadruplet cluster. The difference in bandwidth requirements between a singlet and a multiplet increases as the excitation beam size along the axis of particle travel decreases, because for narrower beams particle size is more a determinant of pulse width (Leary et al., 1979). Thus, a flow cytometer may seem linear when tested with particle systems using singlets, doublets, etc.; but the system may not be linear with a different set of particles. For example, the results obtained with a set of small-diameter particles may be different from the results obtained with a set of larger-diameter particles. The dimmer singlets require a wider bandwidth, but the small-signal bandwidth may be wide enough to accommodate the singlets. Quadruplets, which require a narrower bandwidth, may be accommodated by the narrower large-signal bandwidth, whereas a singlet particle of the same fluorescence intensity may require a bandwidth exceeding the large-signal bandwidth. Thus, the flow cytometer may seem linear with the wider brighter particles when it may not be when using smaller particles of the
same brightness. The important factor is the pulse width. If the pulse widths of singlet particles are significantly less than pulse widths of multiplets, then investigators need to consider the issue of amplifier bandwidth when interpreting results of linearity tests using singlet and multiplet biological particles. Another method of measuring the linearity of a linear amplifier system is to analyze two types of particles with different fluorescence intensities and measure the difference in pulse amplitudes at different PMT voltage settings or laser powers. This method has been used for biological particles (Vindelov and Chistensen, 1990) and microspheres (Bagwell et al., 1989). The difference in the means of the two particle populations M1 and M2 , where M2 is the particle population with the higher mean (M2 > M1), is plotted against the mean of the dimmer particle population. The relationship is given by (M2 − M1) = k × M1, where k is a constant related to the relative intensities of the two particles. Thus, a plot of (M2 − M1) versus M1 is expected to be linear and to intercept the origin (Bagwell et al., 1989; Vindelov and Chrisensen, 1990). If the resulting plot is not linear, then the instrument response is not linear and the instrument needs to be serviced. Unfortunately, this test for linearity is not complete, because a linear plot does not guarantee that the system is linear. Some nonlinear amplifiers, such as amplifiers that have an exponential transfer function (y = axb) will give a linear plot (Bagwell et al., 1989). With this caveat in mind, this method can be used to prove nonlinearity but not linearity. Manufactured fluorescent particles (i.e., fluorochrome-conjugated plastic beads) can be used in the same way as biological particles by locating the singlet, doublet, triplet, and quadruplet peak positions, but it is also possible to obtain manufactured particles of identical sizes with variable fluorescence intensities. Using plastic beads of the same size in multiple intensities avoids the problems of the influence of bandwidth in the measurement of system linearity and the need to use the particles in pairs. For a linear amplifier, a plot of the mean channel of each of the particle populations versus the relative or calibrated intensities should be a straight line. Linear regression analysis of the data can be done to determine the degree of correlation and the existence of an offset. If the linear amplifiers have no detectable offset, then the regression line should intercept the origin as well. If the plot is linear, then the amplifiers are linear. If the line does
Flow Cytometry Instrumentation
1.4.9 Current Protocols in Cytometry
Establishing and Maintaining System Linearity
not pass through the origin, then there is an offset in the linear amplifier or amplifiers. The amplifiers would be linear but not proportional. As long as the manufactured particles are approximately the same size as or smaller than the particles to be analyzed, then a flow cytometer that is determined to be linear with the manufactured particles should be linear for the application to be run. The converse is true as well: i.e., a flow cytometer that gives a nonlinear response with standard particles will also be nonlinear with actual samples. It is important to determine the linearity at the gain setting (and attenuation setting, if applicable) at which the application will be run, because the bandwidth of the amplifiers is different for each gain setting. In general, though, if the bandwidth at the highest gain setting is adequate then it probably will be adequate at the lower gain settings. However, because most linear amplifiers in flow cytometers are multistage, it is not easy to predict what the bandwidth would be at a given gain setting. Thus, it is better to test for linearity at the gain at which the application is to be analyzed. Although the discussion so far has focused on testing of linear amplifiers, similar methods can be applied to the testing of logarithmic amplifiers. Because of the wide dynamic range of logarithmic amplifiers, manufactured particles are favored over biological particles, as the manufactured particles can be fabricated to cover up to a 10,000:1 range commonly found in logarithmic amplifiers used in flow cytometers. If the scale of the logarithmic histogram is presented in linear units (e.g., 0.1 to 1000, 1 to 10,000) then the analysis techniques for linear amplifiers can be used to analyze the data. However, if the scale is in histogram channel units (e.g., 0 to 1024, 0 to 256), then the analysis techniques need to be changed to reflect the fact that the channel units represent the logarithm of the input values. For more detailed discussion of data analysis techniques, see Chapter 10. Two approaches are used specifically to evaluate the logarithmic transform function of logarithmic amplifiers. The first uses two types of microspheres with differing known relative fluorescence intensities. The difference between the mean intensities of the two populations is measured in histogram channel units at different PMT high-voltage settings. Initial high-voltage settings are chosen so that signals from the pair of microsphere populations are located near the bottom of the histogram, and the voltage is raised until the signals are near
the top. The intermediate voltage settings are chosen to cover the whole histogram range uniformly. Because the histogram channel units represent the logarithm of the input signal, taking the difference in the logarithmic domain is equivalent to dividing the means in the linear domain. Thus, taking the difference between the means is equivalent to determining the relative intensity of the one bead versus the other. As the relative intensities of the beads are invariant, this difference should remain constant regardless of the high-voltage setting. Any deviation would indicate a deviation from an ideal logarithmic transformation (Schmid et al., 1988). The second approach in evaluating the linearity of logarithmic amplifiers is to use a series of microspheres of known relative intensities. A plot of mean histogram channel versus the logarithm of relative microsphere intensity shows the relationship between microsphere intensity and histogram channel value (Muirhead et al., 1983; Parks et al., 1988; Schwarz et al., 1996). Because the histogram channel value represents the logarithm of the input signal, this plot is in reality a log-log plot (see Fig. 1.4.6). A straight line model of the log-log plot that is based on the relationship y = a(xb) used to fit the data, where y is the measured signal intensity, x is the actual particle intensity, b is the linearity factor, and the offset is zero (see line A in Fig. 1.4.6). Taking the logarithm of both sides of the equation gives log(y) = b × log(x) + log(a), which is the equation of a line in a log-log plot. The relationship between the histogram channel and log(y) is given by histogram channel = cpd × log(y), where y is scaled from 1 to the maximum intensity value (e.g., 10,000 for a four-decade logarithmic amplifier) and cpd is channels per decade. The slope of the line in Figure 1.4.6 is the product of the channels per decade (cpd) and linearity factor (b). The linearity factor will typically be unity but could be different because of nonlinearities in the linear amplifiers preceding the logarithmic amplifiers. If b is greater than or less than unity, then the effective dynamic range will decrease or increase, respectively. The slope of the line in the log-log plot represents the effective channels per decade, and power function– related nonlinearities before the logarithmic amplifier are masked. The last two methods involving pairs or series of microspheres can be used to evaluate the overall accuracy of the logarithmic transformation of a logarithmic amplifier and any linear amplifiers that precede it. Neither of
1.4.10 Current Protocols in Cytometry
1024
100
768
C
512
10
B 1
Channel number
Fluorescence intensity
1000
256 A
0.1
0 102
103
104
105
106
Fluorescence units
Figure 1.4.6 The log-log plot version of Figure 1.4.1 is commonly used to standardize logarithmic amplifiers. Curve A illustrates a proportional response. Curve B show the effect on an otherwise linear amplifier response of an offset from noise (optical and/or electronic). Curve C illustrates a nonlinearity at the high end of the histogram scale.
these methods, however, will determine the magnitude of the offset (if present). An offset may be apparent as a deviation from the ideal logarithmic transformation at the lower end of the histogram, but the two methods do not provide enough information to determine if the deviation is due to an inaccurate logarithmic transformation or to an offset (see Fig. 1.4.6, curves B and C). Although available from the data, the offset information is either hidden or ignored by these methods of data analysis. One way to detect and quantitate the offset is to use a set of multiple microspheres, as in the second method mentioned above that uses a series of particles, but to perform a linear regression of the converted (linear) intensities calculated from the histogram channel values versus the relative microsphere intensities, instead of a linear regression of log-log data. This method using a series of particles or particle multiplets is equivalent to that described above for linear amplifiers and will provide the same information about the degree of correlation and the magnitude of the offset. A high degree of correlation confirms an accurate logarithmic transformation. An alternative is adding the offset to the earlier equation, resulting in: histogram channel = cpd × log(axb + c) A nonlinear least-squares regression is needed to fit the four parameters. Because there are four variables in the new equation, four or more fluorescent microsphere populations
must be included in the test mixture. Either the linear regression or the nonlinear regression analysis would provide more complete characterization of the logarithmic transformation of the logarithmic amplifiers in a flow cytometer.
MAINTAINING SYSTEM LINEARITY Once the linearity of a flow cytometer system has been established, it is necessary to monitor and to maintain the system linearity. The tests described earlier (see Establishing System Linearity) should be done on a routine basis. Frequency of testing should be set so that any changes in the linearity performance of the flow system can be detected before the data to be collected are adversely affected. If one is just starting to monitor system linearity, monthly tests are recommended. Test intervals can then be changed depending on the historical performance of the flow cytometer being tested. Because of the stability of linear amplifiers, they should not need to be checked more than two or three times a year. Analog logarithmic amplifiers, however, will probably need to be checked monthly.
CONCLUSION When preparing to make quantitative measurements with a flow cytometer instrument, it is important to be able to measure and monitor not only the linearity and proportionality of
Flow Cytometry Instrumentation
1.4.11 Current Protocols in Cytometry
linear amplifiers, but also the accuracy of the logarithmic transformation and offsets of analog logarithmic amplifiers. This unit discusses both simple and more complete methods of measurement. Each method has merits and deficiencies; however, methods that use mixtures of multiple fluorescent microsphere populations with known relative fluorescence intensities provide more information with a reduced probability of introducing bandwidth-related artifacts. With these methods of analysis it is possible to monitor the performance of a flow cytometer instrument and preserve the value of the data for making comparisons between data sets collected at different times and between instruments.
LITERATURE CITED Auer, R.E., Starling, D., Weber, B., and Wood, J.C.S. 1993. A data acquisition system for flow cytometry with wide dynamic range analog to digital conversion and digital signal processing. Cytometry 6(Suppl.):146A. Bagwell, C.B., Baker, D., Whetstone, S., Munson, M., Hitchcox, S., Ault, K.A., and Lovett, E.J. 1989. A simple and rapid method of determining the linearity of a flow cytometer amplification aystem. Cytometry 10:689-694. Leary, J.F., Todd, P., Wood, J.C.S., and Jett, J.H. 1979. Laser flow cytometric light scatter and fluorescence pulse width and pulse rise-time sizing of mammalian cells .J. Histochem. Cytochem. 27:315-320. Muirhead, K.A., Schmitt, T.C., and Muirhead, A.R. 1983. Determination of linear fluorescence intensities from flow cytometric data accumulated with logarithmic amplifiers. Cytometry 3:251256.
Parks, D.R., Bigos, M., and Moore, W.A. 1988. Logarithmic amplifier function evaluation and procedures for logamp optimization and data correction. Cytometry 2(Suppl.):155. Pinkel, D. and Steen, H.B. 1982. Simple methods to determine and compare the sensitivity of flow cytometers. Cytometry 3:220-223. Schmid, I., Schmid, P., and Giorgi, J.V. 1988. Conversion of logarithmic channel numbers into relative linear fluorescence intensity. Cytometry 9:533-538. Schwarz, A., Repollet, E.F., Vogt, R., and Gratama, J.W. 1996. Standardizing flow cytometry: Construction of a standardized fluorescence calibration plot using matching spectral calibrators. Commun. Clin. Cytometry 26:22-31. Steen, H.B. 1992. Noise, sensitivity, and resolution of flow cytometers. Cytometry 13:822-830. Ubezio, P. and Andreoni, A. 1985. Linearity and noise sources in flow cytometry. Cytometry 6:109-115. Vindelov, L. and Christensen, I.J. 1990. A review of techniques and results obtained in one laboratory by an integrated system of methods designed for routine clinical flow cytometric DNA analysis. Cytometry 11:753-770. Vindelov, L.L., Christensen, I.J., and Nissen, N.I. 1983. Standardization of high resolution flow cytometric DNA analysis by the simultaneous use of chicken and trout red blood cells as internal reference standards. Cytometry 3:328-331.
Contributed by James C. S. Wood Coulter Corp. Miami, Florida
Establishing and Maintaining System Linearity
1.4.12 Current Protocols in Cytometry
Optical Filter Sets for Multiparameter Flow Cytometry One of the most important components of a fluorescence detection system is the optical filter set, which selects wavelengths of light for exciting fluorophores and for discriminating fluorescence signals. The striking growth of flow and image cytometry since the mid-1980s has resulted largely from parallel development of not only high-quality optical filter sets (see Desired Features of Filter Sets), but also multicolor fluorescent labels for tagging the increasing number of important monoclonal antibodies and DNA probes (UNITS 4.1-4.3) and powerful software for multiparameter analysis (UNIT 10.4). As a result, we are well into the age of multiparameter fluorescence analysis by flow cytometry. There are a variety of optical devices for selecting wavelengths of light. Monochrometers, acoustooptical tunable filters (AOTFs), and Fourier transform (FT)/Fourier interference (FI) devices (Farkas et al., 1996) have the advantage of providing tunable wavelengths. However, optical filters have gained wide acceptance because of the wavelength selectivity that can be achieved and because they easily can be fit into the optical path of a flow cytometer and are relatively inexpensive. This unit discusses the properties, compositions, and uses of optical filters in flow cytometry (in particular, see Spectral Characteristics of Optical Filters, see Optical Filter Systems for Flow Cytometry, see Construction of Optical Filters, and see Characteristics of Good Optical Filter Systems for Flow Cytometry). Such background information should allow investigators to select quality filters for modifying systems for new dyes and multicolor analyses. Additional details can be obtained from excellent technical information included in catalogs from several manufacturers (e.g., Corion, Chroma, and Omega Optical; see SUPPLIERS APPENDIX).
SPECTRAL CHARACTERISTICS OF OPTICAL FILTERS There are two major types of optical filters currently in use with flow cytometers: colored glass and interference filters. Colored glass filters work by absorbing unwanted wavelengths of light, passing only the desired fluorescent light. Interference filters work through
UNIT 1.5
reflecting light of specific wavelengths and passing other wavelengths (see the section on Construction of Optical Filters for more detail). Depending on their construction, interference filters can be of several types: short-wave pass, long-wave pass, and bandpass. When designed to operate at an angle to the incident light (typically at 45°), interference filters are usually called dichroic beam splitters or dichroic mirrors. “Dichroic” means two-color; such a filter separates the incident light into two different wavelength regions. Examples of these filters are presented later. Long-pass (LP) interference filters (see Construction of Optical Filters) have light transmission spectra as shown in Figure 1.5.1. They are specified according to the “cut-on” wavelength, which is located at half the maximal transmission of the filter. Short-pass (SP) filters transmit at short wavelengths. Transmission spectra for filters can be obtained on most scanning absorption spectrometers, although specialized instruments are required to obtain quantitative measurements of extreme levels of light blocking that can be incorporated into precision optical filters. Bandpass filters (Fig. 1.5.2) transmit light in a narrow range of wavelengths and are classified according to (1) the width of the pass band at half-maximal transmission and (2) the center wavelength of the pass band. For example, Figure 1.5.2 illustrates the pass band of a 610/60 bandpass interference filter. Some types of filters have multiple pass bands. Dichroic filters are designed to steer light of different colors along separate paths. Usually one color is separated and sent at 90° to the path of the remaining light, as illustrated in Figure 1.5.1. Such beam splitters, which are used in the publishing industry for color separation, are configured in a flow cytometry optical system as illustrated in Figure 1.5.3. Depending on filter design, either short or long wavelengths can be reflected. Dichroic filters are long-pass, short-pass, or wide-bandpass interference filters that are used in a 45° orientation and have transmission specifications quoted for the 45° orientation. The center wavelength and the width of the pass region change with angle, as explained later (see Construction of Optical Filters). Flow Cytometry Instrumentation
Contributed by Alan Waggoner Current Protocols in Cytometry (1997) 1.5.1-1.5.8 Copyright © 1997 by John Wiley & Sons, Inc.
1.5.1
100
25
Transmittance (%)
50
50
light
25
75 filter 100
0 100 75
0 dichroic beam splitter (also long-pass)
25
Reflectance (%)
75
0 long-pass interference filter
filter light
50
long λ
25
50 75
short λ 100
0 400
500
600
700
Wavelength λ (nm)
Figure 1.5.1 Transmission spectra of a long-pass interference filter placed at two orientations. Maximum filter transmittance is nearly 100%. In the top panel, the wavelength at half-maximal transmission is about 580 nm, so the filter is classified as a 580-nm long-pass interference filter. It passes red light. The bottom panel illustrates the transmission spectrum shift when the 580-nm long-pass filter is rotated 45°. In the latter orientation the filter can be used as a 520-nm long-pass dichroic filter for color separation, and the transmitted light appears orange.
OPTICAL FILTER SYSTEMS FOR FLOW CYTOMETRY
Optical Filter Sets for Multiparameter Flow Cytometry
Optical filters select the wavelength band to be measured by the detector. Occasionally, laser line rejection filters placed at the entrance of the filter set system may be used to minimize detection of scattered excitation photons. Dichroic filters, sometimes in pairs of long-pass and short-pass filters, separate light on the basis of wavelength, transmitting most of the light but reflecting photons in narrow energy bands or colors to photodetectors. Bandpass filters placed in front of the photodetectors ensure that the most spectrally relevant photons are analyzed. Usually, combinations of long-pass, short-pass, and bandpass filters are used, with an overall long-pass filter (which turns on just above the laser wavelength) to reject the scattered laser light. The fluorescence excitation light source available on most flow cytometers is a laser, but other instruments incorporate mercury arc lamps that emit in several narrow, distinct-
wavelength bands similar to laser lines. Instruments that deliver several laser excitation wavelengths simultaneously are becoming more prevalent, but most flow cytometers utilize a single argon ion laser emitting at 488 nm. A second laser line at 633 or 635 nm is becoming more common and indicates a trend for future systems. Powerful research instruments have UV excitation and perhaps a tunable dye laser that provides excitation at another wavelength, which can be selected, but they are expensive and require expert care. Many interesting and informative experiments require acquisition of multiparameter fluorescence data, usually from at least two differently colored fluorescent dyes, and sometimes from as many as five or six (e.g., immunophenotyping; UNITS 6.2 & 10.4). There are fluorescent reagents that allow three- and four-color data acquisition with a single-laser instrument. This unit focuses on filter sets for obtaining multiple parameter data from instruments containing a single laser, but the principles dis-
1.5.2 Current Protocols in Cytometry
100
Transmittance (%)
pass band center
50 width of pass band (at half-maximal transmittance)
0 540
560
580
600
620
640
660
680
Wavelength λ (nm)
Figure 1.5.2 Transmission spectrum of a bandpass interference filter. The center wavelength (top arrow) of the filter is ∼610 nm, and the bandwidth at 50% maximal transmittance is ~60 nm. Both specifications are required to define the filter (e.g., a 610/60 bandpass interference filter). The transmittance scale shown, however, does not show how effectively the filter blocks light outside the pass band. A logarithmic scale of 6 to 10 decades is required to define blocking ability for top performance filters. An optical density scale (which is logarithmic) of 6 to 10 decades will also suffice (see, e.g., Fig. 1.5.5).
dichroic
dichroic red PMT
red orange green
excitation source (blue)
green PMT
orange PMT
Figure 1.5.3 Typical filter setup for three-color flow cytometry using detection of fluorescein (green fluorescence), R-PE (orange fluorescence), and Cy5-PE (red fluorescence) as an example. The spectra of the fluorophores are shown in Figure 1.5.4. The first dichroic filter might be a 510-nm long-pass and the second a 550-nm long-pass filter. Filters directly in front of the photomultiplier tubes (PMTs) would be six-cavity (or greater) long-pass interference filters to block laser light. For example, in the green channel a 530/25 bandpass interference filter is used, in the orange channel a 575/30, and in the red channel a 670/40, where the first number is the pass band wavelength center and the second the bandwidth. There is some room for variation in the specifications, however, as the fluorescence peaks of the three dyes are well separated.
Flow Cytometry Instrumentation
1.5.3 Current Protocols in Cytometry
cussed can be applied to more complex instrumentation. The typical filter setup for three-color fluorescence data acquisition with a single argon ion laser operated at 488 nm is shown in Figure 1.5.3. Optical filters that isolate the individual fluorescence colors for detection by photomultiplier tubes (PMTs) act as a color separation system similar to that of the publication industry. Two long-pass dichroic beam splitters pick off the green and the orange signals, and the red fluorescence passes on to the third PMT. Because dichroic filters are not efficient at isolating particular wavelengths, each color detection channel requires a high-performance bandpass filter in front of the PMT to reject scattered laser excitation light and the bulk of the fluorescence of the other dyes. Transmission spectra of the filters in the typical threecolor system are shown in Figure 1.5.4. Notice that the respective pass bands are designed to overlap with the emission bands of the three dyes, fluorescein, phycoerythrin (PE), and PECy5, a complex of Cy5 bound to PE in which excitation energy is transferred to Cy5 with emission at a longer wavelength. Even with optimal selection of the bandwidth and central wavelength of the filters, there is spillover of
fluorescence of the shorter wavelength dyes into detection channels of the longer wavelength dyes. This is unavoidable because of the extended long wavelength tails of the dyes. Correction of data for spillover requires electronic or software compensation. For example, a fraction of the signal in the fluorescein channel is subtracted from the signal in the R-PE channel. The fraction subtracted is dependent on the shape of the spectrum of fluorescein and on the placement of the two detection bands.
CONSTRUCTION OF OPTICAL FILTERS Optical filters are fabricated in two ways: (1) by including light-absorbing molecules in glass (colored glass filters) or gelatin to filter out certain colors of light, and (2) by generating interference effects that block the passage of certain wavelengths. Filters made by the latter method are called interference filters. Colored glass filters have the advantages of being about one-fourth the price of interference filters and providing strong light blocking in certain spectral regions. The transition from blocking to maximal transmission in a colored glass filter is more gradual than for the bandpass interference filters, however, and the latter have
100 R-PE
PE-Cy5
Fluorescence
Transmission (%)
fluorescein
0 450
500
550
600
650
700
750
Wavelength λ (nm)
Optical Filter Sets for Multiparameter Flow Cytometry
Figure 1.5.4 Fluorescence spectra (dotted lines) of three fluorophores, fluorescein (green fluorescence, peak 525 nm), R-PE (orange fluorescence, peak 575 nm), and Cy5-PE (red fluorescence, peak 670 nm), superimposed on the transmission spectra (solid lines) of the bandpass filters used to collect the fluorescence signals. Notice the spillover of the fluorescein fluorescence into the orange R-PE detection channel. The intensity of the spillover is some fraction of the intensity in the main detection channel, so that spillover can be subtracted by electronic or software methods in the process called compensation. Adapted from Waggoner et al. (1993) with permission from New York Academy of Sciences.
1.5.4 Current Protocols in Cytometry
gradually taken over in applications that require the highest optical selectivity. Interference filters also predominate in multicolor detection applications because they can be made more wavelength selective. A further disadvantage of colored glass filters is that they often are weakly fluorescent, and as a result scattered excitation light may produce an unwanted fluorescence signal in the detection channel. However, interference filters are more susceptible than colored glass filters to physical damage, humidity, and high light intensities. Interference filters are prepared by creating partially reflective “cavities” (layers) with thicknesses equal to one-half the wavelength of light to be transmitted by the filter. The cavities are formed by layering dielectric materials, such as NaAlF or ZnS, on glass or quartz in a vacuum deposition chamber. The incoming light forms an in-phase standing wave between the two partially reflective walls of the cavities and is passed efficiently through the filter. Light that is not passed is usually reflected unless there are light-absorbing coatings on the filter for additional blocking of unwanted wavelengths. The slightly shorter or longer wavelengths, on the other hand, generate out-ofphase interference between the walls of the
cavities and are not transmitted. Filters must be properly oriented in the light path, or the wavelength selectivity is altered. If the filter is tipped from an orientation perpendicular to incoming light, a somewhat longer wavelength (different color) will be passed through the cavity because the cavity is thicker in the direction of incoming light. For this reason it is important to keep filters perpendicular to the beam of light (or 45° for dichroic filters). Figure 1.5.5 shows how the near-band light transmission is reduced and the pass band edges become steeper as additional cavities are added. Multicavity filters are ideal for transmitting the fluorescent light in a defined spectral region without passing light from the excitation source or from other fluorophores. High-performance multicavity filters are of great importance in multiparameter flow and image cytometry. The more cavities in a filter, however, the higher is the cost.
CHARACTERISTICS OF GOOD OPTICAL FILTER SYSTEMS FOR FLOW CYTOMETRY Optimal use of optical filters requires knowledge of the whole detection system (see UNIT 1.1 for an overview of flow cytometry). In
0 1
Optical density
2 two-cavity filter 3 4
three-cavity filter
5 6
four-cavity filter six-cavity filter
400
450
500
550
600
Wavelength λ (nm)
Figure 1.5.5 Schematic illustrating that steeper walls and greater blocking outside the pass band are obtained by increasing the number of cavities of the filter. Note that the vertical scale is logarithmic. A change of optical density of one unit means that transmission is reduced by a factor of 10; i.e., an optical density of 3 causes a reduction in transmission by a factor of 1000. Blocking can be achieved to >1 photon in 106.
Flow Cytometry Instrumentation
1.5.5 Current Protocols in Cytometry
flow cytometry, scattered light from the exciting laser could interfere with fluorescence detection if not for blocking by optical filters (see Rejection of Laser Excitation). Optimizing instrument sensitivity is another factor connected with filters (see Optimization of Light Collection and Minimization of Background). Finally, optical filters can be evaluated using a checklist (see Desired Features of Filter Sets), and they should be cared for properly (see Care and Maintenance of Optical Filters).
Rejection of Laser Excitation The blocking filters in flow cytometers must remove the vast majority of laser light that is scattered from the cells and stream. Removing scattered light is most difficult when capturing fluorescence from dyes that emit close to the laser line, such as fluorescein. Optimally, the pass band should be centered over the region of maximal fluorescence, but with fluorescein the blue edge of the bandpass filter would be very close to the laser line. An interference filter with six to nine cavities is essential to provide the steepest transition of the blue edge and to provide deep blocking at the wavelength of laser excitation. Bandwidth and band center have to be optimized for maximal light capture and minimal leakage of laser scattering. It is not useful, however, to go to go to more than nine cavities because at some point Raman light scattering from the laser-excited stream and cell autofluorescence dominate background levels of light reaching the photomultiplier tubes. The filters must be “blocked” so as not to transmit light at wavelengths distant from the primary pass band. Interference filters have various natural transmission bands at both ends of the spectrum that must be blocked by applying coatings to the filter which absorb light outside the pass band. Commercially available filters are often blocked into the UV and into the IR. Occasionally, a laser long-pass interference filter or a laser line rejection filter is placed at the entrance of the filter set system to ensure blockage of the scattered laser light.
Optimization of Light Collection and Minimization of Background
Optical Filter Sets for Multiparameter Flow Cytometry
Sensitivity in flow cytometry is a signal-to(background + noise) problem. Noise arises because of random arrival of photons at the detector and electronic noise in the detectoramplifier system. Background signal comes mainly from stray light that leaks through the filter or comes through the pass bands. Signal
is maximized by capturing every emitted photon from the fluorescent label associated with a particular detection channel (e.g., fluorescein photons in the green detection channel). Background is increased when photons from other sources are captured, such as those emitted by Raman scattering (light scattered that is of longer wavelength than the excitation wavelength because of energy absorption by the sample), Rayleigh scattering (light scattered that is of the same wavelength as the excitation light), endogenous fluorescent molecules (Aubin, 1984) in the cells, and other differently colored fluorescent labels with fluorescence that overlaps with the detection channel (Shapiro, 1995). As background photons usually have wavelengths throughout the spectrum, it does not make sense to use a bandpass filter which is so wide that it extends beyond the main part of the fluorescence spectrum of the fluorescent probe and captures photons arising mainly from background sources. There is an optimal pass band width and band center for maximizing signal to background. Although there are computerized and experimental approaches for optimization (Galbraith et al., 1989), close examination of the fluorescence spectra and filter transmission spectra should provide a good start. Experimenting with several bandpass filters and observing background and signal levels may lead to a practical solution. Filter manufacturers have traditionally been very helpful in exploring filter sets for new fluorescent probes.
Desired Features of Filter Sets There are a number of features required of a good filter set. Before placing an order, contact the manufacturer to obtain detailed specifications. The following are some of the features that should be reviewed. • Well-controlled pass band width, band center, and maximal transmission of all filters in a set. • Spectral transmission curves of all filters. If not provided by the manufacturer, the user should make the measurements on a spectrometer. • High rejection (blocking) outside the pass band of the filter; at least six orders of magnitude. • High rejection into the UV and IR ends of the spectrum. • Freedom from “pinholes” in filter materials, that would leak unwanted light.
1.5.6 Current Protocols in Cytometry
Table 1.5.1
Filter Specification for Flow Cytometry
Filter manufacturerb
Filter set
Zeiss Nikon Olympus Leitz Chroma Zeiss Nikon Olympus Leitz Chroma
01 or 02 UV-1A, UV-2A, or UV-2B U-Excitation A or A2 31000 10 B-2H, B-1H B + G520 or IB + G520 L3 or L3.1 31001
Propidium iodide Rhodamine
Zeiss Nikon
15 or 14 G-1B or G-2A
Texas red
Olympus Nikon Chroma Zeiss Nikon Olympus Leitz Zeiss Nikon Olympus Leitz
G-Excitation M2, N2, or N2.1 31002, 31004, 31005 17 or 10 B-2E, B-1E, B-2H, or B-1H B/G520 or IB/G520 L3 or L3.1 15 or 14 G-1B or G-2A G-Excitation M2, N2, or N2.1
Chroma Omega
51000 XP50
DAPI + propidium iodide DAPI + Texas red
Chroma Chroma
51002 51003
FITC + Texas red
Chroma Omega
51006 XP53
FITC + propidium iodide SpectrumGreen + SpectrumOrange
Chroma Zeiss
51005 23
SpectrumGreen + propidium iodide
Zeiss Nikon Olympus Leitz
23, 19, 16, 11, or 09 B-3A, B-2A, or B-1A B or IB H3, I2/3, or K3
SpectrumOrange + DAPI
Imagenetics
DAPI/IO4, IO2/IO4, or DAPI/IO4c/IO10c
Triple dye sets DAPI + FITC + TRITC
Chroma
61000
Chroma Chroma Omega Imagenetics
61001 61002 XP56 DAPI/IO4c/IO10c
Fluorochromesa Single dye sets DAPI
FITC
SpectrumGreen
SpectrumOrange
Dual dye sets DAPI + FITC
DAPI + FITC + propidium iodide DAPI + FITC + Texas red SpectrumGreen + SpectrumOrange + DAPI
aSpectrumGreen and SpectrumOrange are available from Life Technologies. bSee SUPPLIERS APPENDIX for addresses and phone numbers of manufacturers.
Flow Cytometry Instrumentation
1.5.7 Current Protocols in Cytometry
• Nonfluorescent glass, glue, coatings. • Resistance to degradation by high-intensity light and humidity. • Resistance to damage from handling, e.g., abrasion. • Clear labeling of central wavelength, pass band width, and part number.
Care and Maintenance of Optical Filters Colored glass filters are relatively robust but should be kept free of fingerprints while in use or storage. Clean with lens tissue as you would eyeglasses and handle around edges. Final cleaning with alcohol on a Q-tip cotton swab followed by wiping with lens tissue will remove most fluorescent materials that might be present. Colored glass filters should be very stable unless used directly in high-intensity light. These filters should be inspected visually for unevenness in color about every 6 months. Interference filters are usually sealed around the edges and can generally be easily handled. They can be cleaned with the Q-tip and lens tissue method provided that no cleaning liquid is allowed to reach the edges, where there may be access into the dielectric layers. They should never be immersed in any liquids. However, the edges of some dichroic filters may not be sealed, leaving the dielectric coating exposed. These filters are extremely fragile. Never directly touch the dielectric material, and do not use any liquids to attempt to clean the filter because the dielectric material may be released from the substrate. If you cannot store dichroic filters in a support that keeps the surface untouched, keep them protected between two sheets of lens paper in a solid box. Interference filters will degrade in high-intensity light but may also show changes if, for example, high humidity penetrates the dielectric layers. The best way to confirm the quality of an interference filter is to obtain a transmission spectrum of the filter for comparison against the original spectrum. Also, examine the filter under a lowpower microscope to look for pinhole defects in the dielectric material. Under low-humidity, low-light conditions in which the filter is free of contact with “the outside world,” interference filters can last more than 5 to 10 years.
CONCLUSIONS AND FUTURE There appears to be an unending appetite for gathering more information per cell. More multicolor fluorescent labels and biological probes are becoming available (Table 1.5.1), and as a result there is a need for continued advances in high-performance filters for multiparameter cytometry. Fourier transform (FT) techniques allow very rapid acquisition of emission spectra (Buican, 1990), and such devices would, in principle, be ideal in research instruments for capturing signals from newly developed fluorophores as well as the old standbys. As of 1996, however, such devices are in an experimental stage, and routine flow cytometers are likely to use filter sets for some time to come.
ACKNOWLEDGEMENT The author thanks Chuck Burke, Corion, for his help with the preparation of this unit.
LITERATURE CITED Aubin, J. 1984. Autofluorescence of viable cultured mammalian cells. J. Histochem. Cytochem. 27:36-43. Buican, T.N. 1990 Real-time Fourier transform spectroscopy of fluorescence imaging and flow cytometry. SPIE Proc. 1250:126-133. Farkas, D.L., Ballou, B.T., Fisher, G.W., Fishman, D., Garini, Y., Niu, W., and Wachaman, E.S. 1996. Microscopic and mesoscopic spectral bioimaging. SPIE Proc. 2687:200-209. Galbraith, W., Ernst, L.A., Taylor, D.L., and Waggoner, A.S. 1989. Multiparameter fluorescence and selection of optimal filter sets: Mathematics and computer program. SPIE Proc. 1063: 74122. Shapiro, H. 1995. Practical Flow Cytometry, 3rd ed. Wiley-Liss, New York. Waggoner, A.S., Ernst, L.A., Chen, C.H., and Rechtenwald, D.J. 1993. PE-Cy5: A new fluorescent antibody label for 3-color flow cytometry with a single laser. Ann. N.Y. Acad. Sci. 667:185193.
Contributed by Alan Waggoner Amersham Life Sciences Pittsburgh, Pennsylvania
Optical Filter Sets for Multiparameter Flow Cytometry
1.5.8 Current Protocols in Cytometry
Laser Beam Shaping and Spot Size The majority of fluorescence flow cytometers now in use derive illumination from one or more lasers. Air-cooled argon ion lasers operating at 488 nm are the most commonly used light sources, but helium-neon (He-Ne) lasers emitting at 633 nm and diode lasers emitting at 635 nm have come into more widespread use as red-excited antibody labels and other fluorescent probes have become available. Other lasers sources used in fluorescence flow cytometry include argon, krypton, and mixed-gas ion lasers operating at wavelengths from the ultraviolet to the infrared; helium-cadmium (HeCd) lasers emitting in the ultraviolet region at 325 nm and/or in the blue-violet region at 441 nm; and diode-pumped, frequency-doubled yttrium aluminum garnet (YAG) lasers emitting green light at 532 nm. A typical laser emits a beam on the order of 1 mm in diameter; efficient use of the laser light requires that this beam be shaped and focused to a smaller size to illuminate the cell stream passing through the flow cytometer. This unit will discuss spot size requirements and the methods by which optimal spot size is achieved.
CONFLICTING REQUIREMENTS: SENSITIVITY AND PRECISION To maximize the sensitivity of flow cytometric measurements, it is desirable to concentrate a substantial fraction of the intensity of the illuminating beam on the cell under observation, while illuminating as little as possible of the region surrounding the cell. Because most cells that are subjected to flow cytometry are <20 µm in diameter, it would in theory be advantageous to focus the illuminating beam to a spot not much bigger than this, which could be done using a single convex spherical lens. In practice, however, this approach typically compromises measurement precision, in part because of the vagaries of fluid flow and in part because of the intensity profile of the laser beam. To achieve a precision of a few percent in measurement of scatter and fluorescence signals from cells, it is generally necessary that illumination be uniform within that same percentage of variation. A typical laser-source flow cytometer employs a so-called orthogonal geometry, in which the axis of the illuminating beam, the center of the sample or core fluid stream, and the optical axis of the fluorescence and large-angle light scatter collection optics
Contributed by Howard M. Shapiro Current Protocols in Cytometry (1997) 1.6.1-1.6.5 Copyright © 1997 by John Wiley & Sons, Inc.
UNIT 1.6
are mutually perpendicular, with their intersection defining the observation or interrogation point. While the sample stream is flowing, cells will pass through the plane defined by the intersection of the axes of the illuminating beam and the collection lens. Under ideal conditions, each cell in the sample would be precisely centered on the axis of the core stream. In practice, however, there is always some variation in lateral position of cells in the core stream.
EFFECTS OF GAUSSIAN LASER BEAMS ON ILLUMINATION UNIFORMITY All of the lasers mentioned above except diode lasers emit beams that are radially symmetric, with intensity decreasing with distance from the axis of the beam. The most common intensity profile, associated with an emission mode termed transverse excitation mode (TEM00), is Gaussian; in other words, a plot of intensity along any line passing through the axis of the beam would yield a bell-shaped curve identical in shape to the normal or Gaussian distribution frequently encountered in statistics. The diameter, D, of a Gaussian beam is, by convention, defined as the distance between what are known as the 1/e or 1/e2 points. At the 1/e2 point the intensity falls to 0.135 times the intensity on-axis. When a laser beam is focused to a small spot by a convex lens, the diameter of the focal spot, or beam waist, is calculated with the aid of diffraction theory. The laser beam entering the lens can be thought of as collimated; in terms of geometric optics, it would be comprised entirely of parallel rays. Geometric optics predicts that parallel rays entering a convex lens will be focused to a point—i.e., a spot of infinitesimal dimension—at a distance of 1 focal length from the lens. Diffraction produces a spot of finite size; specifically, a convex lens of focal length F mm focuses a beam with diameter D mm to a spot with diameter d µm, where d ≅ (4/π)(λF/D) ≅ 1.27(λF/D). Here λ is the laser emission wavelength, in this instance expressed in units of micrometers (µm) instead of the more conventional nanometers (nm). This formula neglects lens aberrations and thus provides the size of what is called a diffraction limited spot; the formula is not, and cannot be, derived from geometric optics.
Flow Cytometry Instrumentation
1.6.1 Supplement 1
The intensity profile of the focal spot, like the intensity profile of the input laser beam, is Gaussian; the known shape of the Gaussian distribution permits relatively exact calculation of the variation in intensity to be expected within a core stream of a particular diameter, given the diameter of the focal spot. For example, in the original FACS instrument from Becton Dickinson (see SUPPLIERS APPENDIX), which focused the beam with a 125-mm-focal-length spherical lens, operation at 515 nm with a laser beam diameter of 1.5 mm produced a round focal spot 55 µm in diameter; at 458 nm, with the diameter of the input laser beam equal to 1.2 mm, the focal spot was 61 µm in diameter (Loken and Stall, 1982). Parenthetically, it should be noted that the different input beam diameters at different wavelengths result from diffraction effects within the laser itself; such differences are the rule, rather than the exception, in a laser in which emission wavelength is changed using a Littrow prism without changing the laser mirrors. Assuming the center of a 20-µm core stream runs along a diameter of a 60-µm spot, it is possible to calculate the variation in illumination over the width of the core. A table of the Gaussian distribution will show that the 1/e points, which in this case are 30 µm off axis, represent distances of 1.41 standard deviations on either side of the mean. Thus, a point 10 µm off axis is 0.47 standard deviations from the mean. At this point, beam irradiance is 0.79 times peak beam irradiance. What this means is that if the illuminating laser beam is focused to a 60-µm round spot, a particle traveling near the outside of a 20-µm core stream receives only 79% as much illumination as a particle traveling down the core axis. If fluorescence emission is a linear function of illumination intensity, two identical particles following these different trajectories will produce signals differing in amplitude by >20%, meaning that measurement precision will generally be unacceptable.
OBTAINING HIGH PRECISION FROM SEEMINGLY INADEQUATE SPOT SIZES USING HIGH LASER POWER
Laser Beam Shaping and Spot Size
In practice, high-precision fluorescence measurements have been made with beam geometries and core sizes that result in uneven illumination; this becomes possible when the illumination intensity is sufficiently high to produce photon saturation and bleaching of the dye being measured (van den Engh and Farmer,
1992). If the intensity at the edges of the core meets this criterion, the maximum possible fluorescence signal will be obtained from particles at the edge, and particles subjected to the higher illumination intensity on the core axis will not produce correspondingly stronger signals. In order to benefit from saturation, it is generally necessary to employ laser powers of hundreds of milliwatts for excitation; in most modern commercial instruments, in which 10 to 20 mW is a more typical figure, achieving high measurement precision requires a design in which illumination intensity is relatively uniform over the width of the core.
EFFECT OF INTERACTIONS BETWEEN CORE AND STREAM SIZE ON PRECISION Continuing with the analysis of the system just described, it is fairly obvious that one way in which this can be accomplished is to simply decrease the diameter of the core stream. If the 60-µm round spot is retained, and core diameter reduced to 10 µm, the edges of the core are now 5 µm off axis, at which point the irradiance is 95% of peak irradiance. With a further decrease of core diameter to 5 µm, illumination variations can be kept within 1% of peak value with a 60-µm spot. This is acceptable for analysis of particles such as bacteria or chromosomes with dimensions <5 µm; however, illumination of 10-µm cells in a 5-µm core will be less uniform and precision may be compromised. The second obvious way of increasing illumination uniformity is to increase the diameter of the focal spot. Tabulated values of the Gaussian distribution show that irradiance remains >99% of the peak or axial value within the central 8% of the beam width. A 125-µm beam diameter would produce <1% variation in illumination over the width of a 10-µm core stream.
USE OF ELLIPTICAL FOCAL SPOTS TO ENHANCE SAMPLE THROUGHPUT If the beam were focused to a 125-µm round spot, sample throughput would be substantially compromised. Because cells typically pass through a flow cytometer at velocities in the range of 2 to 5 m/sec, it will take 25 to 63 µsec for a cell to traverse a 125-µm beam, limiting maximum analysis rates to 16,000 to 40,000 cells/sec. However, although achieving uniform illumination may require that the beam be 125 µm wide, there is no reason for it to be 125 µm high. Because each cell passes through the entire profile of the beam in the direction of
1.6.2 Supplement 1
Current Protocols in Cytometry
flow, variations in intensity over the Gaussian profile of the laser beam along the axis of flow do not substantially compromise measurement precision. It therefore makes sense to use an elliptical rather than round focal spot with a large spot width (i.e., the axis dimension perpendicular to the direction of flow) and a smaller spot height (i.e., the axis dimension parallel to the direction of flow). A spot height of 20 µm allows cells to traverse the beam in 4 to10 µsec, raising the theoretical throughput limit to 100,000 to 250,000 cells/sec. Making the spot height smaller shortens cells’ dwell time in the beam. The amount of illumination reaching a cell during that dwell time remains the same, and therefore the amounts of light scattered by, and (in the absence of saturation and bleaching effects) fluorescence emitted from, the cell also remain the same: i.e., signal pulses are shorter and higher, but their integrated areas remain the same. If pulse height, rather than area, is measured, the signal-tonoise ratio may appear larger. Making the spot width smaller does increase the amount of illumination reaching a cell during its transit of the beam while decreasing the illumination of surrounding regions, and will increase signal pulse height and area, increasing signal to background but (again in the absence of saturation and bleaching effects) compromising precision because the variation in intensity over the width of the core stream will be increased. If the spot height is made smaller than a cell diameter—i.e., than ~5 µm—cells of different sizes will have measurably different transit times through the illuminating beam, yielding pulses of different widths at the detector(s), and pulse width can then be used to measure cell size. In the technique of slit-scanning flow cytometry, an extremely small spot height (~2 µm) is typically used, and the signal is digitized at very high rates, allowing a substantial amount of information about cell shape and structure to be extracted by computer analysis. Until recently, the processing electronics required for slit-scan flow cytometry were too complex and expensive to be widely used, but the hardware and software are now somewhat more accessible.
PRISMATIC EXPANSION VERSUS CROSSED CYLINDRICAL LENSES IN FORMATION OF ELLIPTICAL SPOTS Current conventional instruments use elliptical focal spots 5 to 10 µm high and ~100 µm wide. In Becton Dickinson’s FACScan, FAC-
Sort, and FACSCalibur instruments, an elliptical beam profile is obtained by using a prism to form an elliptical beam and a spherical lens to focus the beam to an elliptical spot. Instruments from Coulter and Ortho (see SUPPLIERS APPENDIX) and most laboratory-built systems, including this author’s “Cytomutts,” focus a round input beam to an elliptical spot using crossed cylindrical lenses of different focal lengths, each of which focuses the beam in only one dimension. The principle is illustrated in Figure 1.6.1. The shorter-focal-length lens determines the height of the focal spot; it is placed 1 focal length away from the sample stream, and focuses the beam in the dimension perpendicular to the plane of the paper, which explains why its curvature is visible in the side but not the top view. The other lens in this diagram is placed <1 focal length from the cell stream, so that its focal point is on the far side of the cuvette; placing the beam stop, which is a component of the forward-scatter collection optics, at this focal point permits collection of light scattered at smaller angles to the beam than would be possible if the focal point were coplanar with the cell stream. A beam stop is not shown in the figure. Focusing laser beams to extremely small spots—i.e., <5 µm—may be somewhat easier if a beam expander is used to increase the diameter, D, of the beam entering the focusing lens; this allows the use of a corresponding longer focal length focusing lens, placed farther from the cell stream. This is not done in most commercial laser-source flow cytometers.
BEAM SHAPING FOR MULTIPLE-LASER FLOW CYTOMETERS Illumination of a single cell stream with two or more laser beams almost invariably requires some compromises; in general, the focal length of a lens varies with wavelength, so using the same set of focusing optics for both beams, which is common practice in commercial instruments, mandates that at least one beam will be imperfectly focused. Multibeam illumination is generally used for excitation of multiple cell-bound probes with substantially different excitation spectra, and it is usually desirable to have the beam intersection points with the cell stream separated in space, so that cells traverse different beams at different times, facilitating separation of signals from different probes. This is achieved by separating the input beams slightly in space, typically on opposite sides of and at small angles to what would have been
Flow Cytometry Instrumentation
1.6.3 Current Protocols in Cytometry
Supplement 1
A
B
Figure 1.6.1 Formation of an elliptical focal spot by crossed cylindrical lenses. (A) Top and side views of beam-focusing optics and their effects on the laser beam in the flow cell; (B) Spot dimensions in a magnified front view of the flow-cell cuvette. Drawings are not to scale.
Laser Beam Shaping and Spot Size
the beam axis had a single beam been used. If the beams were kept parallel, they would be focused to the same spot; maintaining an angle between them results in separation of the focal spots. In the dual-beam chromosome sorters built at Lawrence Livermore and Los Alamos National Laboratories, input beams slightly separated in height have also been separated by a small angle in the plane perpendicular to the direction of flow, allowing each beam to be focused optimally through a separate set of crossed cylindrical lenses. The MoFlo instrument, now produced commercially by Cytomation (see SUPPLIERS APPENDIX) under license from Livermore, uses this focusing arrangement in dual-beam configurations. If only orthogonal (fluorescence and/or large-angle scatter) signals are of interest, it is possible to bring beams at different heights to a focus on the cell stream using separate pairs of crossed cylindrical focusing lenses located on opposite sides of the
flow cell; this has been done in some Cytomutts designed for analysis of chromosomes and bacteria. The formula used above for calculating focal-spot size neglects the optical effect of a round stream or capillary, which itself behaves as a cylindrical lens; this property results in the formation of a somewhat elliptical spot even when a spherical focusing lens is used. It should also be obvious that spherical and elliptical lenses can be combined in beam-focusing optics. If the input laser beam is itself elliptical, or nearly so, a spherical lens can be used to produce an elliptical spot; this is conveniently done when a single diode laser, which typically has a beam with an aspect ratio of around 4:1, is used as a light source. The optics required for arc lamp illumination in flow cytometry are somewhat more complicated than those used for laser illumination. Even when the same microscope objective is used for illumination and collection, as is
1.6.4 Supplement 1
Current Protocols in Cytometry
typically done in an epiilluminated fluorescence microscope or arc-source flow cytometer, the illumination optics also include a condenser lens for the lamp, and possibly a few additional lenses and diaphragms, in addition to optical filters for excitation and emission wavelength regions and a dichroic mirror. A discussion of these optics is available elsewhere (Shapiro, 1995).
LITERATURE CITED Loken, M.R. and Stall, A.M. 1982. Flow cytometry as an analytical and preparative tool in immunology. J. Immunol. Methods 50:R85-R112. Shapiro, H.M. 1995. Practical Flow Cytometry, 3rd ed. Wiley-Liss, New York. van den Engh, G. and Farmer, C. 1992. Photobleaching and photon saturation in flow cytometry. Cytometry 13:669-677.
Contributed by Howard M. Shapiro Howard M. Shapiro, M.D., P.C. West Newton, Massachusetts
Flow Cytometry Instrumentation
1.6.5 Current Protocols in Cytometry
Supplement 1
High-Speed Cell Sorting THE NEW IMPORTANCE OF HIGH-SPEED CELL SORTING Although cell sorting has been performed since the early 1970s, only recently has highspeed cell sorting become an important issue. For many years cell sorting was only performed in a limited way, and only a small fraction of cell sorters were actually used to sort. The reasons for this were numerous. First, cell sorters were inherently slower than other cell-separation methods and it was difficult or impossible to isolate enough cells for subsequent biochemical analyses using these older instruments. Cell sorting was primarily used to clone small numbers of live cells that could be expanded in tissue culture, to confirm cell types determined by flow-cytometric analysis, or to examine the purity of an aliquot of cells isolated by other cell-separation methods. Cell sorters are still used for these purposes. Second, cell sorters were large, expensive, and difficult to operate in comparison to most other cell separation methods. For the above reasons, people were (and many still are) more comfortable with, or can better afford, alternative cell separation methods. For an excellent review of a number of different cell separation methods see Kompala and Todd (1991). Although much of the above is still true, advances in other technologies and the necessity of using a multiparameter cell separation method for many biological applications now make high-speed cell sorting an attractive proposition. The invention and extensive application of the polymerase chain reaction (PCR) was a major advance that has drastically affected the importance of cell sorting. Now, when a single cell (or a very small number of cells) is sorted, specific sequences of DNA or RNA can be rapidly expanded. Instead of spending many hours or days trying to sort 108 cells, a researcher can sort a small number of cells in a few minutes and duplicate relevant portions of the nucleic acid by PCR into the equivalent of hundreds of millions of cells. This capability represents a very fortuitous wedding of two powerful technologies. Cell sorting can be used to isolate at high purity relatively rare cells, which can be expanded for sophisticated molecular characterizations. For many biological and clinical applications, cells need to be isolated on the basis of multiple quantitative properties. This need is ideally met by the
Contributed by James F. Leary Current Protocols in Cytometry (1997) 1.7.1-1.7.7 Copyright © 1997 by John Wiley & Sons, Inc.
UNIT 1.7
analytical power of flow cytometry, which provides the front-end selection mechanism for cell sorting. Cell sorting, in essence, represents real-time data classification with the added feature of actual physical isolation of the cell that has the properties of interest. Although the sort decision itself ultimately becomes a yes/no one, the information on which such a decision is based can be much more complex than: “is the cell within or without a set of regions defined by the multiparameter measurements.” Information about the probable identity of the cell itself and nearby cells can be combined with other information about the cell that is not measured by the cell sorter. This other information can be used as a weighting factor to determine how much weight to give to the flow-cytometric measurements.
SOME IMPORTANT ASPECTS OF CELL SORTING A number of factors, including fluidics (see are important to high-speed cell sorting. Fluidics obviously are a critical aspect in all cell sorters and particularly in those using charged droplets to sort cells electrostatically. The number of sorting units containing, for example, one, two, or three droplets per sorting decision per second, ultimately limits the rate at which cells can be sorted at a given purity. The droplet-formation rate that is used depends on a number of factors discussed in UNIT 1.2, including pressure and nozzle-orifice diameter. To operate at a higher droplet-formation rate, it is necessary to increase the pressure or decrease the orifice diameter, with a corresponding increase in drive frequency. Both tactics can have a significant effect on the cells sorted. Most cell types do not tolerate system pressures >40 to 50 psi, corresponding to flow velocities of 20 to 40 m/sec. High pressures can lead to explosive decompression of living cells exiting from the flow nozzle to an ambient pressure of one atmosphere (∼15 psi). In addition, the deceleration of the cells as they hit the sort collection device can cause significant damage to cellular morphology and even result in loss of viability. As the orifice diameter gets smaller, cells typically clog more frequently in the flow cell or undergo shearing. All of these factors have to be weighed against one other, and variations with cell type must be taken into account as well. Finally, the overall sorting will be only as UNIT 1.2),
Flow Cytometry Instrumentation
1.7.1 Supplement 1
fast as the ability of the flow cytometer to detect and measure the signals that will be presented to the sort logic circuits. The ultimate speed limitation is a combination of the deadtime of the instrument and the number of sorting units that can be generated per second, and is an important factor in all cell sorters.
THE EFFECTS OF INSTRUMENT DEADTIME ON CELL-SORTING SPEEDS
High-Speed Cell Sorting
One of the major reasons for interest in high-speed cell sorting is the need to isolate as many cells as possible of a minor (and possibly very rare) cell subpopulation. In order for a cell to be sorted, it must first be detected. The signal-processing electronics of a cell sorter can “see” only a certain number of cells per second. The minimal time taken by a flow cytometer/cell sorter to process a cell is known as its “system deadtime.” During this time period, the system is too busy processing one cell to pay attention to the next cell. Multiqueuing systems can reduce this deadtime by providing the next cell with a new set of signal-processing electronics if the previous cell is still being processed. Unless the sorting is done entirely in separate analog hardware, the system deadtime and sorting-unit length determine the ultimate cell-sorting speed limits of the instrument. If more cells are processed than can be seen with the instrument’s inherent system deadtime, the cell sorter will be effectively blind to what is passing through and will not be able to detect nearby contaminating cells that can compromise the purity of the sort. Most flow cytometers use cell light scatter as a default “system trigger” for obvious reasons. All cells must have a light-scatter signal, whereas some cells may not be fluorescent. Light scatter can also be used to reduce systemdeadtime effects produced by debris and cell clumps at the analog level prior to digital signal processing. When the sorter is digitizing everything (including the analog signals produced by debris), the accumulated deadtime can be quite large and many cells will not be detected. If unimportant and irrelevant signals can be eliminated at the analog level, then the sorter can spend its time processing just cells. However, the sort-decision hardware must know about the presence of neighboring debris or clumps if it is to take into account their effect on sort purity. Most flow cytometers have upper practical limits to analysis or sorting rates (typically 2,000 to 10,000 trigger events/sec, which trans-
lates to 2,000 to 10,000 total cells/sec if light scatter is used as a trigger signal). People have tried to get around the signal-processing rate limitation problem by using the fluorescent signal rather than light scatter as the system trigger (McCoy et al., 1991). In many cases this approach will work, but one should be aware of its limitations. The frequency of the rare cells in the sample will be completely lost; this will then have to be estimated by using overall cell concentration, sample volume flow rate, and time for collection of a number of fluorescent positive cells. Because the system trigger sees only fluorescence (and not the light scatter) of the rare cells, it is blind to the presence and identity of everything else, including coincident, nonfluorescent cells and debris, which will likely contaminate the sorted cells. If accurate original-frequency information is not important, or if an impure but enriched sort is adequate, this method can be very useful to researchers with conventional instrumentation. Instrument deadtime and number of sort units per second determine the ultimate rate at which cells can be processed. Conventional cell sorters typically have deadtimes of 10 to 25 µsec. This means that when a cell is detected above trigger threshold, the instrument will spend the next 10 to 25 µsec processing that cell and will be unavailable for the next cell or cells. If the instrument has multiple laser beams, the deadtime may be even greater. Many older instruments have a deadtime for the entire period that the cells take to pass between the two (or more) laser beams. More modern instrumentation allows, to various degrees, data on cells processed by the initial beams to be stored in analog delays, shift registers, or other devices, so that the entire system is not locked up in deadtime while those cells traverse the remaining beams, greatly improving the situation (Parson et al., 1985; Peters et al., 1985; van den Engh and Stokdijk, 1989; Leary et al., 1991; van Rotterdam et al., 1992; Leary and McLaughlin, 1995). These systems also keep information on the identity of the cell preceding the potential cell to be sorted, whereas simpler sort hardware in the past used analog comparators, which were essentially memoryless in that they only looked for the presence of a coincident cell following (not preceding) the cell of interest. Classical definitions of instrument deadtime provide an estimate of the number of cells that will be missed: n N= 1 − nd
1.7.2 Supplement 1
Current Protocols in Cytometry
where N is the expected rate (cells/sec), n is the observed rate, and d is the instrument deadtime in sec. Although this classical definition is often a good approximation, it is not rigorously correct, as it assumes equal spacing of signals rather than the actual Poisson arrival statistics of cells. Also, one should determine the deadtime of the instrument under the same conditions that are used in acquiring data for a particular application. It must be kept in mind that deadtimes may vary considerably depending on how the experiment is run. One needs to perform the deadtime experiments described below, customized to be as close as possible to the application in terms of number of parameters, signal levels, signal processing, gating, and number of cells (input to the instrument) per second. Another particularly important variable in many systems is the time taken for a given analog signal to return to a baseline of zero volts. Nonintegrated (“pulse height” as opposed to “pulse area”) signals—e.g., log peak pulse height or linear peak pulse height— generally will return to baseline much more quickly than either linear-amplified integrated or log-amplified integrated signals. Although integrated signals sometimes appear to give greater sensitivity, they can be costly in terms of instrument deadtime. They are also very susceptible to integration of noise, particularly if those integrated signals are log-amplified. Although instrument-deadtime experiments are fairly simple to perform, to obtain reliable data a high throughput rate of particles or cells is needed. For example, it is possible to use red blood cells and calculate (by any of a variety of measurements) the number of cells per unit volume. First, one must determine under the actual experimental conditions the volume of sample passing through the flow cytometer by weighing a volume of fluid in the sample vial before and after the run—over an interval of several (T) minutes depending on the accuracy of the result desired—on an analytical balance of sufficient accuracy. This will require a series of cell samples of concentrations from ∼5 × 106 to ∼1 × 108 cells/ml. Many flow cytometers use ∼30 to 60 µl/min of sample, but this may vary widely with the instrument model. Then one needs to record the total number of cells counted by the instrument in T minutes and compare that number with the number of cells that should have passed through the instrument over the same time, based on the product of the number of cells/ml times the sample volume used. Finally, the above equation is used to
calculate the deadtime. Rearranging that equation yields: N −n d= Nn where N and n are in cells/sec and the instrument deadtime d is in seconds. If, for example, on the basis of cell concentration and volume throughput, 20,000 cells/sec were expected with the particular system and only 15,000 cells/sec were observed, the instrument deadtime (under those conditions only) would be: 20,000 − 15,000 d= = 16.7 × 10 −6 20,000 × 15,000 = 1.67µ sec. This approximation is good enough for most uses. However, a more exact formulation, depending on Poisson statistics (Gross and Harris, 1985) as derived from queuing theory, needs to be used for more precise results. Poisson statistics describe the arrival statistics of cells to the excitation beam and to the sort droplets (Lindmo and Fundingsrud, 1981). For most people, analysis deadtimes should not be an issue provided the instrument is operated at less than ∼10,000 cells/sec, but can be a problem at higher rates. Some high-speed systems presently in research laboratories have, through a variety of methods, already reduced the deadtime for analysis of rare cells to <2 µsec (Leary et al., 1993), roughly the time it takes for cells to traverse a typical laser excitation beam. Obviously, one gains little from reducing the instrument deadtime below this value. A more common problem is the presence of an unwanted cell occurring within the sorting unit (one, two, or three drops).
THEORY AND PRACTICE OF HIGH-SPEED CELL SORTING In cell sorting, most people are pushing the limits of the instrument’s throughput. Sorting deadtime becomes a factor that affects anyone trying to sort rare cells. One can quickly estimate the amount of time it will take to isolate a number of cells of known frequency. For example, to sort 106 (1 million) rare cells of 1% frequency (i.e., 1 rare cell per 100 total cells) at a rate of 2000 cells/sec will require an investigator to spend almost 14 hr processing 100 million total cells. Even at 5000 cells/sec (the practical limit for many cell sorters, especially if anti-coincidence is required; see discussion of Effects of Anti-Coincidence on Sort Yields), the process would take >5 hr. If the cells must
Flow Cytometry Instrumentation
1.7.3 Current Protocols in Cytometry
Supplement 1
be maintained in a viable state, even 5 hr can sometimes represent a major problem, assuming furthermore that both the sorter and the experimenter can remain in a viable state for this period of time. If only small numbers are needed (potentially a single cell that can be cloned and expanded in tissue culture or whose DNA or RNA can be amplified by PCR), sorting becomes much more practical. A remaining problem is to be able to visualize enough rare cells and to isolate them from undesired cells and debris. Unfortunately, it is difficult to label rare cells specifically without labeling many other cells as well and labeling dead or damaged cells nonspecifically. Thus, to really do the job right requires processing of many controls and obtaining large enough distributions on each control to determine the optimal region to sort. If multiple parameters have been used in an attempt to separate the positive rare cells of interest from background, one now has a small number of cells spread out over many dimensions—the so-called problem of the “tyranny of dimensionality.”
HIGH-SPEED SORT STABILITY Maintaining sort stability over a period of hours in high-speed sorting experiments can be difficult. At high cell concentrations, maintaining a constant and stable droplet-breakoff po-
sition for the sort delay time can represent a considerable challenge. Additionally, the presence of cells in the neck of the droplet that is being broken off and charged causes a variation in charge on that sorted droplet. The result is that the sorted droplet path can vary with the varying charge; hence the sorted sample may be sprayed out over a larger area, making sort recovery difficult. For this reason some laboratories perform high-speed sorting as “straight-ahead sorting” whereby all droplets containing cells that are not of interest are charged, so as to effectively purge unwanted cells from the sample. Droplets containing desired cells are left uncharged and can be sorted more precisely with high recovery because they pass straight ahead into the collection vessel instead of having to be deflected into a sort recovery vessel located on the left or right (Fig. 1.7.1).
ENRICHMENT STRATEGIES EMPLOYING HIGH-SPEED SORTING Sorting of enriched but impure rare cell subpopulations is an interesting case to consider. Although at first it may appear counterintuitive, it is easy to show that cell sorters are actually far more efficient (in terms of enrichment factor) for isolating rare cell subpopula-
non-rare cell, “not-of-interest” (e.g., original frequency = 99.999%) rare cell, “of interest” (e.g., original frequency = 0.001%)
droplets containing cells “not-of-interest” − are charged, − deflected, and − − discarded
droplets containing rare cells “of interest” are uncharged and sorted straight-ahead for high-efficiency recovery
− −
to waste
High-Speed Cell Sorting
highly enriched rare cell subpopulation (e.g., final sorted frequency = 33.33%)
Figure 1.7.1 High-speed cell sorting using “straight-ahead” sorting. This example shows that enrichment of rare cells is actually more efficient than lower-speed sorting of non-rare cells. A high-speed sort of a 0.001% subpopulation of rare cells at 100,000 cells/sec yields an average of 3 cells/droplet. However, one of these cells is of interest, leading to a 30,000-fold enrichment, which may be based on multiple parameters.
1.7.4 Supplement 1
Current Protocols in Cytometry
tions at very high speeds than at low speeds, as shown in Figure 1.7.1. The more rare the cell of interest, the greater the potential enrichment factor that can be provided by a high-speed enrichment sort carried out using “straightahead” sorting.
HOW TO CALCULATE SORT PURITY A sample calculation of the Poisson arrival statistics of cells to sorting units as predicted by queuing theory is shown below. This calculation uses a simple “recursion relationship,” greatly simplifying the calculations and making them easy to perform with a hand-held scientific calculator. To keep the computations very simple, the trick is to compute the probability of zero cells occurring within the queue interval, P(0). Then the probabilities of finding one cell in the sort interval, P(1), two cells in the sort interval, P(2), etc., are easily calculated just by multiplying P(0) by the appropriate factors (Leary, 1994): λx − λ P( x ) = e x! where λ is the average number of cells per sort unit, x is the number of cells anticipated in the sorting unit, x! is “x factorial,” and e is the base of the natural logarithm. It follows from this that P(0) = e−λ, and using a recursion relationship to simplify calculations: λ P( x ) = P( x − 1) x Hence, the probability of 1 cell in the interval λ is: λ P(1) = P(0) 1 and the probability of 2 cells in the interval λ is: λ P(2) = P(1) 2 For example, assume that one is trying to process 5000 cells/sec with 2 droplets per sorting unit at a droplet frequency of 32,000 droplets/sec. The calculations for determining the probability of finding a cell within a 2-dropletssorting unit are as follows (Leary, 1994): λ = ft where f is the cell frequency (equal to 5000 cells/sec) and t is the time for the sorting unit of two droplets, equal to 2 × (1/32,000) or
0.0000626 sec. Hence λ = 5000 × 0.0000626 = 0.313 cells per sort unit and the probability of finding no cells within two sorted droplets is: P(0) = e−λ = e−0.313 = 0.731 meaning that 73.1% of these 2-drop sorting units are empty. Using the recursion relationship, the probability of finding 1 cell within these two droplets, P(1) is: 0.313 λ P(1) = P(0) = × 0.731 = 0.229 1 1 i.e., 22.9% of the sorting units contain 1 cell. Similarly, the probability of finding 2 cells within these 2 droplets is: 0.313 λ P(2) = P(1) = × 0.229 = 0.036 2 2 i.e., 3.6% of the sorting units contain 2 cells. To calculate the purity of the total sort, one must first calculate the purity of each sort unit—i.e., 1, 2, or 3 droplets/sorting unit, depending on the choice of the user and the capability of the instrument. The longer the sort unit, the greater the probability that two or more cells will be found in it. 3-droplet sorting is the easiest to set up and maintain, but it leads to the greatest amount of contamination. 2-droplet sorting is fairly easy to set up and maintain, yet leads to a considerable drop in contamination from cell coincidence in the sorting unit. For sorting at very high speeds, 1-droplet sorting can and should be used. However, it is not only difficult to set up but inherently unstable in terms of sort fluidic stability and sort-droplet break-off position. To calculate the overall purity of a sort, one needs to consider the probability of there being 2 cells, 3 cells, and so on in the sort interval being used. In the above example, the sort purity of rare cells is equal to the following: sort purity = 100 − 50 × P(2) − 67 × P(3) − 75 × P(4) − … because a sort unit containing 2 cells (one of which is the desired cell whereas the other is undesired) has a sort purity of only 50%; a sort unit containing 3 cells (1 desired and 2 undesired) has a sort purity of only 33% (hence a 67% contamination level); a sort unit containing 4 cells (1 desired and 3 undesired) has a sort purity of only 25% (hence a 75% contamination level); and so forth—the calculation can be continued to any desired level of accuracy. This very simple calculation considers the prob-
Flow Cytometry Instrumentation
1.7.5 Current Protocols in Cytometry
Supplement 1
ability of a sorting unit containing ≥2 cells of interest to be insignificant, which is true for rare cell sorting but not for the general case. For non-rare cell sorting, one needs to calculate the Poisson statistics of there being ≥2 cells of interest in the sort interval (Leary, 1994). Most conventional sorters have “anti-coincidence” circuitry intended to eliminate all sort units containing two or more cells. Depending on one’s sorting needs it may or may not be a good idea to use this feature. If the goal is to enrich a rare cell subpopulation and obtain as many rare cells as possible (and possibly process further by a second, slower-speed sort or by another method such as micromanipulation), it may be wiser not to use anti-coincidence as anti-coincidence sort rejection can drastically reduce the yield of rare cells as a function of sort rate. The problem with much of the conventional anti-coincidence circuitry is that when it is operating, nearby cells are rejected regardless of whether they are “friend” or “foe” (meaning desired or undesired, respectively). What this means in practical terms is that nearby desirable cells are thrown away, vastly lowering sort yield. Also, beyond a certain point (typically beyond a few thousand cells/sec), anti-coincidence can totally shut down cell sorting on some instruments. Again, these problems can be addressed by employing more
sophisticated “flexible” sorting strategies. Recently some commercial sorters have begun employing enrichment strategies that provide some balance between sort yield and purity. It is well worth exploring which of these options are available on a sorter of interest. A graph of sorting purity versus rate as a function of the number of droplets/sorting unit (with no anti-coincidence) calculated using the equations presented above is shown in Figure 1.7.2.
EFFECTS OF ANTI-COINCIDENCE ON SORT YIELDS The effect of an all-or-nothing “full” anticoincidence (which usually rejects all sorting units containing more than one cell, regardless of identity) is a sharp reduction in the yield of sorted cells at higher sorting rates. This is particularly true if the cell subpopulation to be sorted is not rare. If purity (and not yield) is the only concern, full anti-coincidence should be used with a total-cell parameter such as forward-angle light scatter chosen as the trigger input to the anti-coincidence circuitry. For this anti-coincidence circuitry to work properly, the cell rate must be reduced so that the average time between cells is greater than the deadtime of the anti-coincidence circuitry of the cell sorter being used. Although the optimal cell rate
100
Purity (%)
80
60
40 1-droplet 2-droplet
20
3-droplet 0 0
2
4
6
8
10
12
14
Cell rate (cells/sec × 104)
High-Speed Cell Sorting
Figure 1.7.2 Graph of sorting purity versus cell throughput rate as a function of the number of droplets per sorting unit, assuming no anti-coincidence. Clearly there is a major increase in sorting speed and purity to be gained by minimizing the number of droplets per sorting unit. However, droplet stability makes it difficult to maintain single-droplet sorting units.
1.7.6 Supplement 1
Current Protocols in Cytometry
can be calculated from deadtime rate and queuing-theory models (Gross and Harris, 1985), it is usually easier in most cases to determine this experimentally. High-speed impure-sorting enrichment can be accomplished by more sophisticated “flexible sorting” algorithms and hardware that allow the experimenter to optimize the yield/purity ratio based on probable identity and desirability of neighboring cells (Corio and Leary, 1993). Other non–droplet sorting methods not discussed in this unit, such as “zapper” elimination of unwanted cells (Herweijer et al., 1988; van Rotterdam et al., 1992; Keij et al., 1991; Keij, 1994), can also be described with queuing theory, although in those cases the queue length becomes the “zap” queue length rather than the sort-droplet interval.
Leary, J.F., Ellis, S.P., McLaughlin, S.R., Corio, M.A., Hespelt, S., Gram, J.G., and Burde, S. 1991. High-resolution separation of rare cell types. In Cell Separation Science and Technology, ACS Symposium Series no. 464 (D.S. Kompala and P. Todd, eds.) pp. 26-40. American Chemical Society, Washington, D.C.
LITERATURE CITED
McCoy, J.P., Jr., Chambers, W.H., Lakomy, R., Campbell, J.A., and Stewart, C.C. 1991. Sorting minor subpopulations of cells: Use of fluorescence as the triggering signal. Cytometry 12:268274.
Corio, M.A. and Leary, J.F. April 1993. System for Flexibly Sorting Particles. U.S. patent 5,199,576. Gross D. and Harris C.M. 1985. Fundamentals of Queuing Theory, 2nd ed. John Wiley & Sons, New York. Herweijer, H., Stokdijk, W., and Visser, J.W. 1988. High-speed photodamage cell selection using bromodeoxyuridine/Hoechst 33342 photosensitized cell killing. Cytometry 9:143-149. Keij, J.F. 1994. The Zapper. Ph.D. dissertation. TNO Health Research Institute, The Netherlands. Keij, J.F., van Rotterdam, A., Groenewegen, A.C., Stokdijk, W., and Visser, J.W.M. 1991. Coincidence in high-speed flow cytometry: Models and measurements. Cytometry 12:398-404. Kompala, D.S. and Todd, P. (eds.) 1991. Cell separation science and technology. ACS Symposium Series no. 464. American Chemical Society, Washington, D.C. Leary, J.F. 1994. Strategies for rare cell detection and isolation. Methods Cell Biol. 42:331-358. Leary, J.F., Corio, M.A., and McLaughlin, S.R. April 1993. System for High-Speed Measurement and Sorting of Particles. U.S. patent 5,204,884.
Leary, J.F. and McLaughlin, S.R. 1995. New technology for ultrasensitive detection and isolation of rare cells for clinical diagnostics and therapeutics. In Progress in Biomedical Optics: Proceedings of Ultrasensitive Instrumentation for DNA Sequencing and Biochemical Diagnostics, Vol. 2386 (G.E. Cohn, J.M. Lerner, K.J. Liddane, A. Scheeline, and S.A. Soper, eds.) pp. 150-163. SPIE/Optical Society of America, Bellingham, Washington. Lindmo, T. and Fundingsrud, K. 1981. Measurements of the distribution of time intervals between cell passages in flow cytometry as a method for the evaluation of sample preparation procedures. Cytometry 2:151-154.
Parson, J.D., Hiebert, R.D., and Martin, J.C. 1985. Active analog pipeline delays for high signal rates in multistation flow cytometers. Cytometry 6:388-391. Peters, D., Branscomb, E., Dean, P., Merrill, T., Pinkel, D., Van Dilla, M., and Gray, J. 1985. The LLNL high-speed sorter: Design features, operational characteristics, and biological utility. Cytometry 6:290-301. van den Engh, G. and Stokdijk, W. 1989. Parallel processing data acquisition system for multilaser flow cytometry and cell sorting. Cytometry 10:282-293. van Rotterdam, A., Keij, J., and Visser, J.W. 1992. Models for the electronic processing of flow cytometric data at high particle rates. Cytometry 13:149-154.
Contributed by James F. Leary University of Texas Medical Branch Galveston, Texas
Flow Cytometry Instrumentation
1.7.7 Current Protocols in Cytometry
Supplement 1
Principles of Gating Gating is an integral part of any flow cytometric analysis. From the discriminator or threshold circuitry used to prevent acquisition of low-level noise signals up through on-line and off-line gating to select specific subpopulations of cells or particles to be analyzed, the different levels of gating available in a flow cytometer provide a means to enhance and simplify analysis of heterogeneous cell or particle populations. It is becoming uncommon for any flow cytometry analysis requiring two or more parameters to lack some form of gating. Gating achieves data reduction by allowing the investigator to select populations with specific characteristics for display in one, two, or three dimensions and/or for further analysis. Complex gating schemes are becoming more commonplace as commercial flow cytometers provide the capabilities to collect multidimensional data of eight and more parameters. This unit describes the different types of gating typically available on a flow cytometer and the various means by which gating is accomplished. It begins with a description of the functional elements associated with gating in a flow cytometer, including brief descriptions of the electronics and software in these functional elements. Since gates are commonly combined to provide a more selective specification of a desired subpopulation, the logical or boolean mathematics used to combine gates is also discussed. The intent is to provide a basic understanding of how gating is accomplished in a flow cytometer and how it is used in flow cytometry applications. Specific gating strategies are introduced to help the investigator develop gating protocols for biological applications. Most of the flow cytometry applications that require gating protocols can be developed from a basic set of gating strategies. Use of these gating strategies is discussed in terms of common flow cytometry applications.
SCIENCE OF GATING Analog Threshold Gating A basic premise underlying any flow cytometric measurement is that the desired optical signal is significantly larger than spurious transient optical noise signals that can interfere with optical measurement. Photodetectors on a cytometer receive numerous bursts of photons
Contributed by James C. S. Wood Current Protocols in Cytometry (1998) 1.8.1-1.8.12 Copyright © 1998 by John Wiley & Sons, Inc.
UNIT 1.8 as the cell or particle suspension being processed is hydrodynamically focused and directed through the excitation light source. Some photon bursts occur randomly, unassociated with any particle or cell event, while other bursts are correlated with a cell or particle interacting with the excitation light source. Randomly occurring bursts contain no information about the sample suspension and contribute to what is considered as noise. Other sources of noise are photon bursts associated with particles or debris not intentionally a part of the cell or particle population under investigation. This optical noise is present in both light-scatter and fluorescence signals. Experiments can be designed and a flow cytometer adjusted to minimize both of these noise sources. Analog threshold gating is used to exclude these low-level optical noise signals from processing by the flow cytometer data acquisition electronics. Because low-level noise signals occur typically with a very high frequency, it is possible for the data acquisition electronics to be overwhelmed by these rapidly occurring signals. Many of the desired optical signals can be prevented from being processed by the data acquisition electronics because each optical signal takes a finite time to process. During this processing time, also known as the dead time of the data acquisition electronics, new optical signals are ignored. This can be readily observed on a flow cytometer by lowering the trigger or gating threshold and noting the reduction in the rate of acquisition of the desired brighter signals. Older flow cytometer systems had dead times on the order of a couple of hundred microseconds; however, newer systems have reduced the dead time to less than a few tens of microseconds through the use of analog buffer queues and faster electronics. Some research flow cytometers have reduced the dead time to a few microseconds (van den Engh and Stokdijk, 1989). Nevertheless, lowlevel noise signals can occur at rates in excess of one million events per second, so that even the faster flow cytometer instruments are susceptible to being overwhelmed. By suppressing the unwanted low-level signals, analog threshold gating frees up data acquisition electronics to process the intended brighter optical signals and improves the flow cytometer instrument throughput.
Flow Cytometry Instrumentation
1.8.1 Supplement 3
Principles of Gating
Analog threshold gating is typically done on only one optical parameter. A single analog comparator is used to detect when a pulse amplitude exceeds the threshold value. The threshold parameter can be either a light-scatter signal or a fluorescence signal. The choice of parameter depends on which parameter gives the best signal-to-noise ratio, i.e., the best separation between noise and signal. Forward light scatter is the most common signal used as it is typically a reliable indicator of when a cell or particle is intersecting the excitation beam. If the cells or particles are large enough and have a refractive index different enough from that of the suspending medium, then it is likely that they will have a strong enough forward lightscatter signal to be used as a trigger. Forward light scatter is particularly useful for applications in which the sample consists of particles ranging from nonfluorescent to brightly fluorescent. For those types of samples, the amplitude of the fluorescence signal may not always be large enough to trigger the threshold gate; thus, some cells or particles may not be counted. Using forward light scatter as a trigger ensures that all the particles or cells have their fluorescence quantified correctly. A fluorescence parameter is used as a trigger when all the particles have sufficient fluorescence intensity to trigger a threshold gate, e.g., DNA analysis. Also, if the amplitude of the forward light-scatter signal is very small, as when the particles are very small or have an index of refraction very nearly that of the suspending fluid, then a fluorescence parameter is a viable alternative trigger if the cells or particles can be stained brightly enough with a fluorescent dye. A good example of this scenario is bacteria stained with a fluorescent dye. Bacteria are typically very difficult to detect with forward light scatter in a nonepi-illuminated flow cytometer; thus, staining the bacteria with a fluorescent dye makes them bright enough to reliably trigger a fluorescence-signal threshold gate. The fluorescent dye can be a nucleic acid dye, a membrane potential dye, or an antibody-coupled dye, to name a few examples. The key is to pick a bright dye that does not bind significantly to debris or other contaminating particles. Another interesting way to use a fluorescence parameter as a trigger is for processing or sorting rare cells if the rare cells are the only brightly stained cells and a count of the other cells is not required. Gating on only the brightly stained rare cells means that flow cytometer
data acquisition would trigger just on the few bright cells of interest and ignore the rest. Few, if any, bright rare cells would be missed, making this a simple way to analyze and/or sort just the rare cell population (McCoy et al., 1991). Note, however, that the resulting sorted samples, although enriched, would have a low purity. Because the threshold gate is set high enough that only the brightly stained rare cells are analyzed, other contaminating cells and debris are ignored by the flow cytometer sorting–decision logic. In designing an application to be run on a flow cytometer, it is important to have a reliable means of triggering the flow cytometer so as to analyze all the desired cells or particles and minimize false triggering of the system by noise signals. Though forward light scatter is the most common signal used to trigger the threshold gate, fluorescence parameters can be used as well and are required in some applications. The key is to identify an optical signal with a high signal-to-noise ratio that is possessed by all the cells to be included in the analysis.
Window Gating Gating by setting a lower-limit threshold is not sufficient for many flow cytometer applications. For some applications, it is necessary to set an upper limit as well. In those instances only particles or cells of optical intensities within a specific range or window of intensities are intended to be analyzed. Flow cytometers use window gating to specify the range of acceptable optical intensities. For multiparameter experiments it is common to combine many window gates to define specific cell or particle subpopulations within a sample. Instrument design of gating There are many ways to achieve window gating within a flow cytometer. Some methods of window gating were developed for real-time gating applications, others for off-line (or nonreal-time) applications. Real-time window gating is accomplished by either high-speed analog or high-speed digital comparators. These circuits work very quickly, as the time between cells or particles can be very short and the number of missed cells or particles needs to be kept to a minimum. The off-line window gating is usually done in software, where it is possible to analyze the data in a less time-critical manner. For off-line analysis the data are already placed on a storage medium and no data will be missed since they are read only as quickly
1.8.2 Supplement 3
Current Protocols in Cytometry
as they are processed. The data can also be reanalyzed or replayed multiple times to try out different gating schemes. Real-time (on-line) gating Real-time gating occurs after the optical signals are converted to electronic signals but either before or after the electronic signals are converted to digital values by the analog-todigital converter (ADC). The former is accomplished by analog gating circuitry, the latter by look-up tables or high-speed digital signal processors. The choice of which method to use is dependent on the speed at which the gating decisions need to be made. Typically, the analog gating circuitry can provide the quickest gating decisions; however, the shape of the analog gating window is limited. For gating in two dimensions, analog gating windows can take the shape of squares, rectangles, or parallelograms. Digital gating windows require that the analog signal be converted to a digital value by an ADC; thus, digital gating is slower than analog gating but the shape of the gating window can now be a more complex amorphous shape. Bitmaps are examples of digital window gates.
Preprocessing Gates: Analog Gating Analog window gates typically work as prefilters by selecting specific signals to be converted by the ADC to digital values. Analog window gates cannot work directly with the raw pulses from the photodetectors and amplifiers. The pulses must be preprocessed by analog pulse–processing circuitry. Since it is either the pulse height or the pulse area which is of interest, either or both must be determined before being applied to the analog window gates. Analog pulse–processing circuitry consists of all or some of the following functions: (1) peak sense and hold, (2) active or passive integrator, and (3) sample and hold. The output of analog pulse–processing circuitry is a stretched pulse. It is equal in amplitude to the pulse height or area, and is held long enough for the analog window gates to give a result and the ADC to complete its conversion. The analog window gates are fast enough, however, to give a result as soon as the stretched pulse is stable, at which point the ADC is given the command to start a conversion. Thus the analog window gates can be used to determine, with minimal throughput penalty, whether the ADC should commence a conversion. Analog window gates were the first type of gates used in flow cytometry. When flow cy-
tometry was first being explored, digital processing was not a viable alternative because most flow cytometers used dedicated pulseheight analyzers. For those that did use computers, the computers were too slow to do any real-time gating. The first commercial flow cytometer to use gating, the Ortho Cytofluorograf, used analog window gating. It was not until the late 1970s that analog window gating started to be replaced by its digital equivalent in commercial instruments. The simplest form of an analog window gate consists of a pair of analog comparator integrated circuits. Together they form a limit or window comparator. One determines when the lower limit is exceeded, and the other determines when the upper limit is not exceeded. Their outputs are combined so that when the stretched pulse amplitude is between the lower and upper limit, both comparators indicate a true condition. The combined output indicates that the stretch pulse amplitude is within limits or the condition of being within the limits is true. Conversely, the output of the window comparator is false when the stretch pulse amplitude is outside the limits. This comparison takes much less than a microsecond, faster than most ADCs used in commercial flow cytometers today. A more recent application of analog window gates has been in rare-event analysis. With the exception of the most recent experimental research flow cytometers, flow cytometers cannot asynchronously process sample throughput rates in the 100,000 cells or particles per second and higher range. The combined dead time of the ADC and digital-processing electronics is too long and many events would be missed. In rare-event analysis, events of interest which need a full analysis occur at very low rates, whereas other events occur at very high rates— e.g., looking for fetal cells within maternal blood. In those instances it would be advantageous to preselect those events which meet specific criteria and let only those events be processed by the slower digital processing electronics. Analog window gating works well in this application as a preprocessor. After the pulse height or area is captured, an analog window gate can be used to determine whether the ADC conversion should be triggered and the converted pulse amplitude processed by the digital processor. This type of strategy has been used to analyze maternal blood for fetal cells at speeds in excess of 100,000 cells per second (Cupp et al., 1984).
Flow Cytometry Instrumentation
1.8.3 Current Protocols in Cytometry
Supplement 3
Postprocessing Gates: Digital Gating
Principles of Gating
Digital gating occurs after the analog stretch pulse has been converted to a digital value. The digital value can then be used by itself or combined with the digital values of other parameters and applied to a digital gating circuit. The simplest form of a digital gating circuit is a region. For a given parameter, two comparisons are made by digital comparators to determine whether the digital value is within the set limits. If two parameters are involved, then a bitmap is used. This is a series of memory locations which are either on or off; the digital values applied to it determine which location is accessed and output. Bitmaps may be used as a separate gating circuit or they may be used by a computer program to rapidly determine if a set of measured parameters places the cell or particle within the digital bitmap gate limits. A more sophisticated form of digital gating uses digital signal processing to act directly on the digital values, either singly or in combination, to determine if they fall within the specified limits; or the digital values are used to calculate a derived parameter (e.g., surface area, volume, or density) which is then directed to a digital window-gating function. Of these three approaches, the region and the bitmap approaches are the simplest and fastest; however, digital signal processors are becoming ever faster, so that they can be used in many flow cytometer window-gating applications. Regions apply to only one parameter. Each region consists of a lower and upper limit. The digital value for the specific parameter is compared to the lower and upper limits to determine if it falls within the designated limits. This is accomplished with two digital comparators that allow the lower and upper limits to be checked simultaneously. The user sets the lower and upper limits with a computer program by viewing a single-parameter histogram and using a pointing device to set lower and upper cursor locations to define the region. Bitmaps are the most common form of twoparameter digital window gating in flow cytometers. They utilize digital memory-integrated circuits to store the required gating decisions. The bitmaps consist of a series of memory locations, the number of which is determined by the needed resolution. Because bitmaps use two parameters at a time to make a gating decision, the required number of memory locations is 4096 for 64 × 64 channels, 16,384 for 128 × 128 channels, and so on. To create the bitmap, the user is shown a two-parameter histogram of representative data
for the parameters to which the digital gate is applied. A bitmap gate is input by the user using a pointing device to circumscribe the desired region with a polygon. The computer program then uses the polygon to set the values within the bitmap. The number of vertices allowed in the polygon is determined by the computer program. Digital signal processors are currently available with sufficient speed to complete most of the data processing required by a flow cytometer. Logarithmic transformation, spectral compensation, calculation of derived parameters, and window gating can be accomplished in a timely manner. Automatic setting of gates (autogating) is also possible. The flow cytometer samples data for a short period, then sets an elliptical bitmap gate around the targeted subpopulation. The flow cytometer then restarts data acquisition with the new gate. These features are already available in commercial instruments (Auer et al., 1993). Combined with better analog-pulse queuing and faster digitalsignal processor speeds, sample throughput speeds approaching 30,000 to 50,000 are conceivable with flow cytometers using all-digital processing. The availability of derived parameters will let the user set gates on more meaningful parameters such as protein density, volume, and binding sites. Sorting gates Use of threshold and window gates is an integral part of a particle or cell sorter application. The objective of such a sorter is to isolate a particular particle or cell type. Threshold and window gates are needed to identify the particles or cells to be sorted and those which are not to be sorted. These gates must be real-time gates because sorting decisions must be made during the time the particle or cell is in transit between the detection point and the droplet break-off point. Each particle or cell needs to be completely processed before the next particle or cell can be processed, so the time allowed for processing is actually shorter than the transit time. In the instances when a second or third particle or cell arrives too early, it is marked as an unknown and considered an unwanted coincident event. To achieve the highest purity sort results, the flow cytometer should have the threshold gate set low enough to trigger on all possible contaminating particles (see Fig. 1.8.1). If a particle passing through the excitation beam does not produce a signal that is bright enough to trigger the flow cytometer data acquisition electronics,
1.8.4 Supplement 3
Current Protocols in Cytometry
count
A
sorted region
TG
Signal amplitude
Signal amplitude
count
B
sorted region
TG
Signal amplitude
Signal amplitude
Unsorted
Sorted
Figure 1.8.1 Gating for sort purity. To achieve the expected sort purity, the threshold gate (TG) must be set low enough to trigger on all possible contaminating particles. (A) In this panel, the threshold is set higher than the amplitude of the dimmer contaminating particles. As a result, the dimmer particles are not processed by the sort-decision logic circuitry and appear as contaminants in the sorted sample. (B) In this panel, the trigger threshold is set below the dimmer particles, and they are now excluded in the sorted sample.
then it will be ignored by the sort-decision logic. It will not even be counted as an unwanted coincident event. Thus, it may be sorted inadvertently with a properly sorted particle or cell. If this occurs a significant number of times, then the purity of the sorted samples will be compromised. The sorted samples will contain the correctly sorted particles or cells plus the particles or debris which did not trigger the threshold gate. The threshold gate level needs to be adjusted low enough not to ignore particles or cells that would adversely affect the purity of the sorted samples.
The first generation of sorting flow cytometers used analog window gates to classify particles or cells; however, the current instruments now use digital window gates—i.e., regions and bitmaps. In order to maintain the required throughput, the sorting-decision logic takes data from the ADC down a separate data path from the data acquisition part of the flow cytometer. The sorting-decision logic consists of digital comparators to determine whether the particle is within a specific region and dedicated memory circuits that hold the information for the bitmaps. Two parameters at a time are
Flow Cytometry Instrumentation
1.8.5 Current Protocols in Cytometry
Supplement 3
Number of events
–
+
+
–
Number of events
Number of events
CD2+ fluorescence
Number of events
CD8+ fluorescence
–
+
CD2 CD4 CD8
– – –
+ – –
– + –
+ + –
– – +
+ – +
– + +
+ + +
PRISM parameter
CD4+ fluorescence
Figure 1.8.2 Gating with the PRISM parameter. Lymphocytes stained with CD2, CD4, and CD8 antibodies are easily analyzed with the PRISM parameter, which shows the number of lymphocytes binding each of the eight combinations of CD2, CD4, and CD8 antibodies.
Principles of Gating
used to access a bitmap memory circuit, the output of which determines whether the particle or cell is within the bitmap or not. In addition to bitmaps, there are some specialized gating tools available. The Coulter PRISM parameter can be used to simplify the analysis of immunology samples (Fig. 1.8.2). The PRISM parameter allows rapid negative/positive cell analysis for multiple colors of antibodies. For n parameters there are 2n combinatorial outcomes possible. The PRISM parameter shows the number of cells that are in each combination as 2n channels in a histogram. This can be used in combination with bitmaps
to simplify the gating equations for immunology applications. Off-line gating Off-line or analysis gating is used in processing list-mode data by flow cytometer analysis computer programs. Since the data reside on a storage medium readable by the computer system at whatever speed is required, the time limitations of real-time analysis are lifted. All of the gating functions used in realtime analysis are found in off-line analysis; however, more sophisticated and flexible gating schemes are available in off-line analysis
1.8.6 Supplement 3
Current Protocols in Cytometry
Table 1.8.1 Truth Table for AND Function
Input 1
Input 2
Result
0 0 1 1
0 1 0 1
0 0 0 1
Table 1.8.2
Truth Table for OR Function
Input 1
Input 2
Result
0 0 1 1
0 1 0 1
0 1 1 1
because there is more time available to process each stored event. To help visualize populations, gates may be adjusted and the data replayed. Coloring schemes can be added to identify subpopulations. Another technique is back gating, which is used to determine the number and distribution of a particular particle or cell subpopulation back onto a histogram, the axes of which are two gating parameters used to define a bitmap for the subpopulation. When multiple gates are used, back gating is useful to determine the effectiveness of a contributing gate in selecting the target particle or cell subpopulation. Completely automatic clustering of data is also possible using standard clustering algorithms or neural networks (Murphy, 1985; Frankel et al., 1993; Redelman, 1993). The data axes may be rotated and/or transformed to better separate the clusters (Leary et al., 1991). A more extensive array of calculated parameters is also possible and each could be used as a gating parameter. The amount of computational complexity that can be accomplished during off-line analysis is limited only by the speed of the computer and the patience of the investigator.
MATHEMATICS TO COMBINE GATES When more than two parameters are involved in a flow cytometry application, it is likely that more than one region or bitmap will be needed for satisfactory specification of a particular particle or cell subpopulation. The decision from each region or bitmap is a logical or boolean result, i.e., either it is true that the
event is within the region or bitmap or it is false. Some well-defined operations can be used to combine the boolean results. The primary operations are AND, OR, and NOT. Other operations such as XOR can be derived from the three basic operations. The AND operation is shown in a truth table, Table 1.8.1. An entry of 0 means false and an entry of 1 means true. Notice that if either input is false (0) then the output is false (0) as well. The AND operation bears a resemblance to multiplication because of this property. The equation for the AND function for two inputs A and B is expressed as A • B. The AND function can be used with more than two inputs, is commutative (A • B = B • A), and is associative (A • (B • C) = (A • B) • C = A • B • C). This function is often used when a particle or cell must meet the criterion of being in two or more bitmaps or regions at the same time. The OR function is also shown in a truth table, Table 1.8.2. If either input is true then the result of the OR function is true. The equation for the OR function for two inputs A and B is A + B. The OR function can have more than two inputs, is commutative (A + B = B + A), and is associative (A + (B + C) = (A + B) + C = A + B + C). The AND function is distributive over the OR function [A • (B + C) = (A • B) + (A • C)] and the OR function is distributive over the AND function [A + (B • C) = (A + B) • (A + C)]. The OR function is used to logically combine regions or bitmaps if a particle or cell can be in any or all of two or more regions or bitmaps. The NOT function is a unary operator. If the input is true then the output is false and vice
Flow Cytometry Instrumentation
1.8.7 Current Protocols in Cytometry
Supplement 3
versa. The form for the NOT function is ∼A. This operator is useful if a particle or cell can not be in a particular region or bitmap. For example, it may be easier to stain a contaminating cell population, so the NOT function would be used to exclude it from targeted subpopulation. For example, A ∼ B would be used if a particle or cell needed to be in bitmap A but not in bitmap B. Another useful pair of equations, known as De Morgan’s Theorems, are ∼(A • B) = ∼A + ∼B and ∼(A + B) = ∼A • ∼B. These are useful when it is necessary to switch from AND logic to OR logic and vice versa in order to simplify a gating equation. A particular flow cytometer or analysis program may not support the negation of calculated terms such as ∼(A • B) but can support the negation of individual terms, so that ∼A + ∼B could be used instead. De Morgan’s Theorems should also be kept in mind when combining regions. For example, if the investigator wishes to exclude everything within the intersection of two regions (A • B), then either ∼(A • B) or ∼A + ∼B should be used in the gating logic.
GATING STRATEGIES The following are some possible strategies that can be used in designing flow cytometry applications that require gates. In the discussion of each strategy, some example applications are mentioned. An application may use one or more of the following strategies.
Choosing an Acceptable Threshold or Trigger Gate
Principles of Gating
Choosing an acceptable threshold or trigger parameter and level is important because this determines what the flow cytometer considers to be a valid event and can affect the instrument throughput. The chosen parameter should produce a bright enough optical signal for all particles or cells that are to be analyzed, so they can be easily distinguished from the background noise and debris signals. The better the separation, i.e., the higher the signal-to-noise ratio, the easier it is to set the threshold level to minimize the loss of valid events and the inclusion of noise signals. Light-scatter parameters (most commonly forward or low-angle scatter) are routinely used to trigger flow cytometer data acquisition. Most eukaryotic cells are large enough and refractive enough to produce a bright light-scatter signal. For example, forward light scatter is a good choice for immunology applications in which cells can be weakly fluorescent (negative) or
strongly fluorescent (positive), and for reticulocyte analysis in which the reticulocyte fluorescence ranges from very dim to very bright. In both instances, forward light scatter serves as a reliable means of determining whether a cell has entered the excitation light beam. However, smaller cells, such as platelets and bacteria, and cells or particles that have the same refractive index as the suspending medium would be difficult to see with forward light scatter. For these applications, a fluorescence parameter may be a better choice as a trigger signal. Unfortunately, this may not always be a viable alternative, and compromises need to be made by judiciously setting the threshold level. Fluorescence parameters can serve either as equally acceptable or as alternative parameters to trigger flow cytometer data acquisition. For some applications, it does not matter whether the system is triggered on light scatter or on fluorescence parameters. For example, the fluorescence intensities encountered in DNA analysis applications are usually bright enough that fluorescence is just as reliable a trigger as a forward light-scatter signal. Bacteria can easily be detected with an appropriate nucleic acid or membrane potential dye, and antibodies coupled to fluorescent dyes can be used to detect a wide range of particles, including bacteria and platelets. Once the parameter for triggering flow cytometer data acquisition has been selected, the level of the trigger needs to be set. Ideally there would be a sufficient separation between the cells of interest and the background noise so the threshold level could have a wide range of acceptable values. However, in some instances, separation between noise and signals is minimal, so the threshold is more difficult to set. If it is set too high, some of the particles or cells of interest are missed. If it is set too low, then the data acquisition system spends more time analyzing noise and, as a result, some of the particles or cells of interest are missed. The investigator needs to set the threshold for the highest sample throughput of the particles of interest which minimizes missing of the particles or cells.
Light-Scatter Gating Use of forward and right-angle light scatter as gating parameters should be considered by every investigator in the development of a new flow cytometry application. This is a free pair of gating parameters because every particle or cell scatters light. Light-scatter intensity is sensitive to particle size, granularity, index of re-
1.8.8 Supplement 3
Current Protocols in Cytometry
Forward light scatter
granulocytes
monocytes lymphocytes
90° light scatter
Figure 1.8.3 Light-scatter gating is commonly used to identify lymphocytes, monocytes, and granulocytes in a whole blood preparation.
fraction, and shape, to name just a few possibilities. It is usually by serendipity that a helpful separation is found, but it is always worth a check. At a minimum, the light-scatter parameters can help exclude debris and multiplets of particles or cells from analysis. One of the more common uses of light scatter is the separation of lymphocytes, monocytes, and granulocytes (Fig. 1.8.3). Undoubtedly more possibilities remain to be discovered.
Doublet-Exclusion Gating Measurement of particle or cell parameters by flow cytometry is predicated on the assumption that these measurements are made on single cells. Doublets and larger multiplets are most commonly detected by comparing fluorescence pulse peak height to pulse area. Using either a two-parameter histogram of pulse peak intensity versus pulse area and setting a bitmap around the singlet particles along the diagonal, or a single-parameter histogram of the ratio of pulse peak intensity to pulse area and setting a region around the singlet particle peak (Fig. 1.8.4), the user can restrict the analysis to the singlet particles by gating on the singlet particle gates (Snow and Bauer, 1994). This type of doublet-exclusion gating has become a routine part of DNA quantitation analysis. Any investigator designing an assay for which doublets or multiplets would significantly affect the results should consider this type of doublet-exclusion gating. The above-mentioned pulse
peak height versus pulse area may be applicable, or the alternative methods using pulse width (time-of-flight) or forward-scatter intensity may need to be considered.
Inclusion versus Exclusion Gating The choice between inclusion and exclusion gating depends on how easy it is to label and isolate the particle or cell subpopulation of interest compared to the effort required to label and exclude contaminating particle or cell subpopulation(s). It is also contingent on which approach gives a more specific analysis of the targeted particle or cell subpopulation. Most of the time investigators choose to use inclusion gating, simply labeling the particles or cells with an appropriate marker specific for the subpopulation and setting a region or bitmap gate around the subpopulation cluster. This method works well when there is a high specificity for the targeted particle or cell population and low cross-reactivity with contaminating particle or cell subpopulations. In addition, the specificity of the analysis improves with the use of multiple gates. Examples of multiple gates are T-cell gating with CD3, CD4, and CD8 to identify all four T cell (CD3+) subsets and the enumeration of CD4+ cells using CD45, CD3, and CD4 (Mandy et al., 1992; Nicholson et al., 1993). Less frequently used than inclusion gating, exclusion gating is used to exclude a contaminating particle or cell population from an analy-
Flow Cytometry Instrumentation
1.8.9 Current Protocols in Cytometry
Supplement 3
sis. It can be useful when the marker for the contaminating particle or cell population is more specific than a marker for the targeted particle or cell subpopulation. It is also useful to confirm the specificity of a potential inclusion gating marker. It may also be combined with other inclusion markers for added selectivity. An example of exclusion gating application is the use of a fluorescently labeled antibody for monocytes to exclude monocytes from analysis of lymphocytes. Since the monocytes are not eliminated completely from the lym-
phocyte population by the light-scatter bitmap gate, the monocyte marker provides the needed added specificity to exclude the monocytes from contaminating the lymphocyte analysis. The advantages of inclusion gating versus exclusion gating depend on the application. Factors that need to be considered are the specificity of the analysis and the number of parameters that are needed to provide the required gating. The investigator should choose the simplest set of markers that achieves the desired results.
Peak amplitude
A
singlets
doublets
Integral pulse amplitude
B
Count
singlets
doublets
Peak amplitude/integral pulse amplitude
Principles of Gating
Figure 1.8.4 Gating for doublets and multiplets. Doublets and larger multiplets are commonly detected by comparing fluorescence pulse peak amplitude to the pulse integral amplitude. In (A) a two-parameter plot of peak versus integral pulse amplitudes is used to identify the singlet events which lie along the diagonal and the multiplet events which lie below and to the right of the diagonal. An alternative approach is shown in (B) using a single-parameter histogram of the ratio of peak and integral pulse amplitudes. The singlets are identified by the peak to the right and the multiplets are found to the left of the singlet peak.
1.8.10 Supplement 3
Current Protocols in Cytometry
Back Gating Back gating is useful for checking the effectiveness of a particular gate in a gating equation by gating a particle or cell subpopulation back onto a single- or dual-parameter histogram for analysis. This technique is particularly useful in checking the shape and positioning of gates to see if they are adjusted for best discrimination. For example, in DNA analysis the G2 and higher-ploidy populations can be back gated
onto the pulse height versus pulse area histogram to determine how effective the doublet correction bitmap is. Sometimes a bitmap cannot be adjusted; then back gating might be used to calculate a correction factor, as has been done to correct for monocyte contamination in the light-scatter bitmap gate used for lymphocyte analysis (Loken et al., 1990; Fig. 1.8.5). Back gating is also used with coloring schemes to identify the location of subpopula-
Figure 1.8.5 Back gating can be used to check the effectiveness of a particular gate in a gating equation. In this case, back gating is used to determine the level of monocyte contamination in the lymphocyte lightscatter gate. In panel (A) the light-scatter gate is set; in panel (B) the monocyte gate (CD14+) is set; in panel (C) the two gates are combined. The contaminating monocytes are identified as those cells which satisfy the lymphocyte lightscatter gate and are positive for CD14. The resulting histogram shows the small number of monocytes that would contaminate the lymphocyte analysis.
Flow Cytometry Instrumentation
1.8.11 Current Protocols in Cytometry
Supplement 3
tions in a particular histogram. For example, lymphocyte subpopulations can be back gated onto the forward scatter versus right-angle light-scatter histogram to see how their lightscatter properties differ. This form of analysis is potentially a very powerful off-line analysis technique.
CONCLUSION Gating is a powerful on-line and off-line tool for flow cytometry analysis; it will become increasingly important as the number of collected optical parameters expands and more calculated parameters become available. Gating is a powerful data reduction tool that can help the investigator isolate and identify particle or cell subpopulations. As computers inevitably become faster, more automation will become available, and there will probably be a merging of clustering techniques with gating.
LITERATURE CITED Auer, R.E., Starling, D., Weber, B., and Wood, J.C.S. 1993. A data acquisition system for flow cytometry with wide dynamic range analog to digital conversion and digital signal processing. Cytometry (Suppl.) 6:146A. Cupp, J.E., Leary, J.F., Cernichiari, E., Wood, J.C.S., and Doherty, R.A. 1984. Rare-event analysis methods for detection of fetal red blood cells in maternal blood. Cytometry 5:138-144. Frankel, D.S., Loken, M.R., Stelzer, G.T., Shults, K.E., and Bagwell, C.B. 1993. Neural network analysis of flow cytometric data for normal and leukemic bone marrow. Cytometry (Suppl.) 6:44. Leary, J.F., Ellis, S.P., and McLaughlin, S.R. 1991. 3-D autostereoscopic viewing of multidimensional data for principal component/biplot analysis and sorting. Cytometry (Suppl.) 5:134-135.
Mandy, F.F., Bergeron, M., Recktenwald, D., and Izaguirre C.A. 1992. A simultaneous three-color T cell subsets analysis with single laser flow cytometers using T cell gating protocol: Comparison with conventional two-color immunophenotyping method. J Immunol. Methods 156:151-156. McCoy, J.P. Jr., Chambers, W.H., Lakomy, R., Campbell, J.A., and Stewart, C.C. 1991. Sorting minor subpopulations of cells: Use of fluorescence as the triggering signal. Cytometry 12:268274. Murphy, R.F. 1985. Automated identification of subpopulations in flow cytometric list mode data using cluster analysis. Cytometry 6:302-309. Nicholson, J.K.A., Jones, B.M., and Hubbard, M. 1993. CD4 T-lymphocyte determinations on whole blood specimens using a single-tube three-color assay. Cytometry 14:685-689. Redelman, D. 1993. Improved procedures for training neural networks to analyze flow cytometric data. Cytometry (Suppl.) 6:43. Snow, C. and Bauer, K. 1994. Usefulness of DNA fluorescence ratioing in reducing DNA measurement artifacts caused by aggregates. Cytometry (Suppl.) 7:50. van den Engh, G. and Stokdijk, W. 1989. Parallel processing data acquisition system for multilaser flow cytometry and cell sorting. Cytometry 10:289-293.
KEY REFERENCE Shapiro, H.M. 1995. Practical Flow Cytometry, 3rd ed. Wiley-Liss, New York. A resource for more examples of the use of gating in flow cytometry applications.
Contributed by James C.S. Wood Coulter Corporation Miami, Florida
Loken, M.R., Brosnan, J.M., Bach, B.A., and Ault, K.A. 1990. Establishing optimal lymphocyte gates for immunophenotyping by flow cytometry. Cytometry 11:453-459.
Principles of Gating
1.8.12 Supplement 3
Current Protocols in Cytometry
Lasers for Flow Cytometry Most fluorescence flow cytometers use one or more lasers as light sources; in fact, the Bio/Physics Systems Cytofluorograf flow cytometer, introduced in 1970, was the first commercial product to incorporate an argon-ion laser, and most modern benchtop instruments feature similar air-cooled argon-ion lasers, emitting 10 to 25 mW of light at 488 nm. The process by which a laser beam, typically 0.5 to 2.0 mm in diameter, is focused to a smaller elliptical spot to illuminate the sample stream is discussed in UNIT 1.6; this unit covers lasers themselves and the qualities that make them suitable or unsuitable for use in flow cytometry. Almost all of the lasers discussed in this unit can be used for other laser-based imaging technologies such as scanning cytometry and confocal microscopy.
WHY LASERS OR ARC LAMPS ARE NEEDED AS LIGHT SOURCES FOR FLOW CYTOMETRY Flow cytometric measurements of scattered light and fluorescence emission are made during the few microseconds in which a cell passes through the illuminating beam. Basic physics dictates that the total number of photons scattered and emitted by a particle cannot exceed the number of photons incident upon it, and as even the most efficient collection optics capture light over a relatively small solid angle, only a fraction of this total can possibly contribute to a measured signal. As much as a third of the light collected in some spectral regions may be lost in the various dichroic mirrors and filters used to divert light to the detectors, and the detectors themselves convert no more than half (diodes) and typically less than one quarter (photomultiplier tubes) of the photons incident on their photocathodes to photoelectrons, which are the source of the signal currents further processed by flow cytometers’ electronics. It is thus necessary that the light source used in a flow cytometer be able to emit a relatively large number of photons per unit time, and further, that it be possible to direct or focus a substantial fraction of those photons into a small volume of the sample stream. The critical characteristics of a light source can be defined quantitatively in terms of its radiance, which is the power emitted from, transmitted through,
Contributed by Howard M. Shapiro Current Protocols in Cytometry (2004) 1.9.1-1.9.16 Copyright © 2004 by John Wiley & Sons, Inc.
UNIT 1.9
or reflected by a surface, per unit of its area, per unit solid angle; the units of radiance are W/m2/sr (watts per square meter per steradian). Extended light sources, a category that includes filament or arc lamps as well as the sun and other stars, can emit substantial amounts of energy, but their emission is typically dispersed in all directions, i.e., over the total solid angle of 4π sr, which represents the surface of a sphere. It is not physically possible to collect light from an extended source over a large solid angle and focus all of it through a smaller solid angle; thus, while a substantial fraction of the emission from such a source can be collected using parabolic or elliptical reflectors, the majority of the collected light cannot be delivered to the illumination region of a flow cytometer. Flow cytometers that do employ extended sources use high numerical aperture lenses to collect light from the source and to illuminate and collect light from cells in the sample stream. Arc lamps, which have relatively high radiance, are much better suited to flow cytometry than are filament lamps, which have substantially lower radiance. The seeming paradox that lower power (e.g., 100-W mercury) arc lamps are better sources for cytometers than filament lamps of substantially higher power finds an explanation in the smaller lamps’ higher radiance, which results from smaller dimensions of the arc itself. The output of a laser is typically a beam of relatively small diameter (0.5 to 2 mm) with a small angle of divergence; the radiance of a laser is, therefore, extremely high. The submilliwatt lasers used in supermarket bar code scanners emit more photons per square meter per steradian than does the surface of the sun. All of the power in a laser beam can be focused to a spot with dimensions as small as a few micrometers using a simple lens; for reasons discussed at length in UNIT 1.6, the optical systems of flow cytometers usually incorporate crossed cylindrical lenses of different focal lengths to produce an elliptical spot.
HOW LASERS WORK Stimulated Emission The word “laser” is an acronym for “light amplification by stimulated emission of radiation.” The physical process behind all lasers is stimulated or induced emission, a type of
Flow Cytometry Instrumentation
1.9.1 Supplement 27
nonradiative transition to 2nd electronic excited state
4th electronic excited state 3rd electronic excited state
laser transition
2nd electronic excited state (metastable)
1st electronic excited state nonradiative transition to ground state
ground state excitation by absorption
Figure 1.9.1 Transitions between electronic energy states in a typical laser.
Lasers for Flow Cytometry
fluorescence described by Einstein in the early 1900’s. In order for any kind of fluorescence to occur, the absorption of a photon must first excite the absorbing atom, ion, or molecule (hereafter, “molecule” will refer to “atom, ion, or molecule”), producing a transition from a lower to a higher electronic energy level. After a brief period of time, usually a few nanoseconds, the molecule typically returns to a lower energy state by emitting a photon with energy less than or equal to that of the absorbed photon. Under most circumstances, only a small fraction of the molecules in the illuminated material is in an excited state, and the photons emitted from different excited molecules are different in wavelength, phase, and polarization; what occurs is called spontaneous emission. As Einstein showed, however, once a molecule has been excited by absorption, the presence of a photon or photons of a particular energy in its vicinity increases the probability that it will emit a photon of the same energy (frequency or wavelength), phase, and polarization. Thus, photons can induce or stimulate the emission of like photons. Stimulated emission becomes more likely as the fraction of the molecules in excited states increases, and can become self-sustaining when there is a population inversion, i.e., when the excited molecules outnumber those in the lower energy state. In general, it is diffi-
cult to create population inversions for energy transitions between the lowest excited state and the ground state of a molecule, because the ground state is more favorable on thermodynamic grounds according to the Boltzmann law. Most practical lasers emit at a wavelength corresponding to the energy of a transition between a metastable higher energy excited state, i.e., one with a relatively long lifetime, and a lower energy excited state. The lasing medium is excited, or pumped, by electrical energy or by a high-intensity light source, causing the molecules in the medium to undergo transitions to excited states with energies equal to or higher than that of the metastable state; those at higher energies subsequently drop to the metastable state nonradiatively. Initially, spontaneous emission occurs at a particular laser wavelength as molecules drop from the metastable state to the lower excited state; thereafter, spontaneously emitted photons stimulate the emission of additional photons at that laser wavelength and the process continues. The population inversion required to sustain stimulated emission is maintained because molecules rapidly leave the lower energy state of the laser transition by thermodynamically favorable transitions to excited states of still lower energy or to the ground state. A diagram of the energy levels typically involved in laser action appears as Figure 1.9.1.
1.9.2 Supplement 27
Current Protocols in Cytometry
Gain, Resonators, and Optical Feedback While the “l” and the “a” in “laser” stand for “light amplification,” an operating laser is more like an amplifier that has been driven into oscillation by application of positive feedback. Most of us are familiar with an audio analog of this process, in which sound from a microphone placed in front of the speaker drives the audio amplifier into oscillation; the frequency of oscillation is a function of the distance between the microphone and speaker. The initiation of stimulated emission in a volume of a suitable material will not in and of itself produce the concentrated, low-divergence light beams that characterize lasers and on which so much of their utility depends. It will, instead, result in light emission in all directions, i.e., over a solid angle of 4π sr. This is so because, while the photons produced by stimulated emission travel in more or less the same direction as the stimulating photons, the spontaneously emitted photons responsible for the first round of stimulation do not have any directional preference. It is necessary, therefore, to perform some geometrical and optical manipulations in order to make a usable laser. First, the volume of lasing medium in which stimulated emission occurs is shaped to produce some directionality of emission. As just mentioned, spontaneously emitted photons are equally likely to be emitted in any given direction, and photons produced by stimulated emission, which follow the paths of the stimulating photons, will therefore also be equally likely to be emitted in any given direction. The probability that one photon will stimulate emission of others in the medium is proportional to the length of the path of the photon in the medium. If the medium were formed into a spherical shape, this average path length would be the same in all directions. The gain of the lasing medium, i.e., the number of stimulated photons emitted per unit distance per incident photon, depends predominantly on the quantum mechanical properties, i.e., energy levels and transition probabilities, of the medium. If the gain is high enough, and the intensity of excitation of the medium is sufficient, stimulated emission may be sustained in a spherical volume, but emission will be neither directional nor coherent. These conditions are, incidentally, believed to exist for carbon dioxide molecules in the atmosphere of Mars. If the lasing medium is shaped into a long, thin cylinder or rod, photons emitted parallel to or at small angles to the axis are more likely to
stimulate emission than photons emitted along or near the radius, because the path of the axial photons is substantially longer. Thus, the geometry of the medium will favor emission along the axis. Making the medium longer will, in general, increase the amount of power that can be obtained. In practical lasers, the directional property achieved by shaping the lasing medium is augmented by placing the medium inside a relatively rigid structure or resonator, with precisely aligned and spaced mirrors, highly reflective at the desired output wavelengths, mounted at opposite ends. Light emitted along the axis of the resonator is reflected back along the same path again and again, while light at increasingly larger angles to the axis is less and less efficiently reflected back through the medium. Since light produced by stimulated emission is identical in wavelength, phase, and direction to the stimulating light, most of the emission confined within the laser cavity, i.e., the space between the mirrors, will be concentrated along or very near its axis. Laser output is produced by making one of the mirrors able to transmit a small fraction of incident light; the amount of transmission permissible varies with the gain of the medium, which must be high enough to make up for the light lost by transmission outside the cavity and the light lost by absorption within the cavity. A schematic sketch of a laser is shown in Figure 1.9.2. The spacing between the mirrors is critical. If they are an even number of wavelengths apart, constructive interference will occur between the rays incident on and those reflected from the mirrors, maximizing output. If not, there will be destructive interference, which may be enough to prevent laser action entirely. In medium- and high-power lasers, wavelength selection within the spectral range attainable with a single set of mirrors is generally done by insertion of a Littrow prism in the cavity between the mirrors. The dispersion of the prism results in light of different wavelengths being refracted at different angles on passage through the prism. At any given position of the prism, only a relatively narrow range of possible emission wavelengths will be reflected along the axis of the laser cavity between the high-reflector and output-coupler mirrors. Gain in this selected wavelength range will be sufficient to maintain laser action; gain at wavelengths above and below the selected range will not. The emission wavelength is changed by changing the orientation of the prism; this usually involves a vertical angular adjustment. A
Flow Cytometry Instrumentation
1.9.3 Current Protocols in Cytometry
Supplement 27
pump energy source (electric or optical)
light out
high reflector mirror
lasing medium
output coupler mirror
Figure 1.9.2 Schematic diagram of a laser. Brewster windows used to produce polarized output would be placed between the medium and the mirrors; a Littrow prism or wedge used for wavelength selection is typically placed between the medium and the high reflector mirror.
Lasers for Flow Cytometry
dispersive element other than a prism, e.g., a grating, etalon, or optical filter, can also be used for wavelength selection. The resonator can be thought of as analogous to an organ pipe; the length of the pipe and the effective distance between the mirrors of the resonator determine the frequency of the standing wave sustained by the structure. In the case of the resonator, this characterizes what is described as the longitudinal mode of the laser. The energy profile of the beam itself, or the transverse mode of the laser, is determined by the geometry of the medium as well as by the geometric optics of the mirrors. If stimulated emission is confined to a volume close to the axis of the resonator, the laser will operate in what is called transverse electromagnetic mode zero zero (TEM00), and the intensity profile will be Gaussian, as described in detail in UNIT 1.6. As the effective cross-section of the medium increases, other transverse electromagnetic modes are superimposed on TEM00. These are individually undesirable in lasers designed for use in cytometry because they are, in general, not radially symmetric, but rather, multilobed, and are likely to produce nonuniform illumination. In many types of lasers, polarization is introduced into the beam by putting special windows between the medium and the end mirrors. The windows are placed at Brewster’s angle to the axis of the system and usually oriented such that light reflected from the surface of the window is directed upward (vertically). At Brewster’s angle (∼57° for glass), reflection from the window surface is minimized for vertically
polarized light, while a small percentage of horizontally polarized light is reflected out of the cavity. The slight difference in transmissions of the two polarizations is magnified many-fold by the feedback characteristic of the optical resonator structure, with the result that the laser output in a system with such Brewster windows is highly polarized (usually vertically), and typically in a ratio of at least 500:1.
Pump Energy and Efficiency To produce and maintain the population inversion necessary to sustain stimulated emission and laser action, energy must be injected from the outside. The method by which this pumping is achieved varies with the lasing medium used. In gas, ion, and metal-vapor lasers, electromagnetic energy is used to produce and in some cases confine the plasma that serves as the lasing medium. In pulsed lasers, in which the lasing medium might be a solution of fluorescent dye or a ruby or yttrium aluminum garnet (YAG) rod, light from a flashlamp is often the pump energy source. In a continuous wave (CW), as opposed to pulsed, dye or YAG laser, the output of a pump laser, typically an ion laser in the former case and a diode laser in the latter, is used. Diode lasers themselves are pumped by input of electric current. Since a substantial power density of excitation is typically necessary to produce a population inversion, for all laser types there is a threshold level of pump power below which laser action cannot be achieved. The efficiency of lasers varies greatly. An argon-ion laser emitting a watt or so of light
1.9.4 Supplement 27
Current Protocols in Cytometry
typically consumes ∼10 kW of electrical power while in operation; the overall efficiency of this system is therefore ∼0.01%. A CW dye laser, optically pumped with the 1-W output of the argon-ion laser, would typically emit a few hundred milliwatts; the efficiency of the dye laser, neglecting the power consumption of the pump laser, is typically 20% to 30%. Diode lasers are also relatively efficient. Less efficient lasers are more likely to require forced-air or water cooling, particularly when high power outputs are needed; this increases their size, complexity, and cost. Efficiency is strongly dependent on the gain of the laser, which may vary substantially for different laser lines. In the example of the argon laser given above, the same power input that could produce 1 W of visible output might be required to produce only 10 to 20 mW of UV output; the efficiency would then drop to between 0.0001% and 0.0002%.
Output Power Regulation and Laser Noise The ion lasers used in most flow cytometers can be operated in either a current control mode or a light control mode. In the current control mode, the power supply is regulated to deliver a constant current through the plasma tube, ostensibly producing a constant supply of pump energy. Light output remains constant as long as the mechanical and optical characteristics of the laser do not change substantially during operation. A perturbation such as a slight misalignment of a mirror will decrease light output, however, even though power supply current remains the same. In the light control mode, a feedback circuit measures light output and maintains it at a nearly constant level by varying the power supply current as necessary. Although this works well when the laser is emitting at a single wavelength, it is more difficult to keep power constant when emission of several lines occurs simultaneously, particularly if the gains differ considerably and/or there is competition between lines, which occurs when excited molecules in the medium can drop to lower energy states by two or more mechanisms associated with laser emission. The air-cooled argon-ion lasers in benchtop flow cytometers are operated in light control mode; so are most diode lasers, which are usually built with a light-sensing photodiode in the same package. In the case of diode lasers, the incorporation of a light-control feedback loop into the power supply is necessary to prevent the laser from overheating and self-de-
structing when it is first turned on. Heliumneon (He-Ne) and helium-cadmium (He-Cd) lasers typically do not incorporate light control circuits. Laser noise may originate from several sources. Poor power supply regulation results in light output fluctuations at the frequency of the line current used to run the power supply or at a multiple thereof; when a high-frequency switching circuit is used in the power supply, light noise also typically occurs at the switching frequency. In some lasers, particularly He-Ne and He-Cd lasers, noise may be found at frequencies in the range of a few hundred kilohertz, due either to radio frequency energy used to pump the lasing medium or to fluctuations in the medium itself. In most circumstances, the level of laser noise determines the minimum detectable signal level for scatter measurement channels. Although preamplifier electronics remove the steady DC component of the background noise produced by stray laser light, the AC component of the background, representing fluctuations around the DC level, is amplified along with the signals produced by particles passing through the illumination beam. To be detectable, the signal from a particle must be substantially above the level of the fluctuations; therefore, a laser with lower noise allows detection of smaller particles. Operation in the light control mode does not guarantee protection against noise. If the light output drops substantially enough for the supply current to rise to its maximum value, any further mechanical or optical deterioration cannot be compensated for by either the light control or the current control circuitry; thus, a change in light output will occur. One logical solution to laser noise is the incorporation of compensation for changes in source intensity in the signal-processing electronics, e.g., amplifiers with gain automatically controlled by a feedback circuit that senses variations in source output at the specific excitation wavelength(s) of interest. Such circuits have been used in several laboratory-built flow cytometers, but are not currently incorporated into commercial instruments.
Laser Peripherals: Harmonic Generation and Modulation The light produced by lasers, like all other light, has associated electric and magnetic fields, and because the radiance of a laser beam is substantially higher than that of an incandescent source, the associated electric field intensity may be high enough to produce nonlinear
Flow Cytometry Instrumentation
1.9.5 Current Protocols in Cytometry
Supplement 27
responses in certain materials. One notable application of these phenomena is in harmonic generation, in which nonlinear effects in crystals result in generation of light at two or more times the frequency of the incident light. Second harmonic generation, or frequency doubling of the 1064 nm YAG laser line, for example, produces emissions at 532 nm, while thirdharmonic generation, or frequency tripling, produces emissions at 355 nm. The same nonlinear crystals may also be used to produce output at the sum and/or difference of the frequencies of two incident beams; in the case of YAG lasers, frequency summing can produce emission at 473 nm. While a reasonably broad range of crystalline materials capable of harmonic generation is available, the range of wavelengths at which continuous (CW) output can be obtained is restricted. It is sometimes desirable to vary the output power of a laser more rapidly than can be accomplished by adjustments to the power supply. Modulation at frequencies up to several hundred megahertz is possible using electrooptic modulators, which incorporate crystals that change their refractive index as a function of an applied voltage. Acousto-optic modulators, which use sound waves to produce changes in density that affect the light transmission characteristics of a substrate, work at lower frequencies, generally below 100 MHz. The combination of a light sensor and a modulator, connected by a feedback circuit, can be used as a “noise eater,” providing the equivalent of light-regulated output for a laser into which it was not built.
LASERS USED IN FLOW CYTOMETRY
Lasers for Flow Cytometry
The vast majority of fluorescence flow cytometers do not offer the user a choice of light sources; they come equipped with air-cooled argon-ion lasers, emitting 10 to 25 mW at 488 nm. Older systems and some modern cell sorters used and use larger, water-cooled, multiwatt argon- and/or krypton-ion lasers. Various instruments have also used He-Ne, He-Cd, dye, diode, and solid-state lasers, with the latter two types becoming increasingly popular in newer cytometers. While the successful operation of large ion and dye lasers and of some He-Cd lasers requires some skill and training, fixedwavelength low-power argon, He-Ne, He-Cd, and solid-state lasers typically do not allow the end user to make any adjustments at all. Diode laser systems may or may not provide for power adjustment. Emission wavelengths of a variety
of lasers usable for cytometry are shown in Table 1.9.1.
Argon- and Krypton-Ion Lasers The lasing medium in an argon- or kryptonion laser is a plasma; when the system is turned on, a high-voltage pulse is used to ionize the gas, allowing it to conduct the relatively high electrical current that maintains the plasma discharge, pumping ionized gas atoms into excited states. Singly ionized argon is capable of several laser transitions at wavelengths ranging from the blue-violet (454 nm) to green (528 nm); the highest gain lines are those at 515 and 488 nm, with the 488-nm line predominating in smaller systems and the 515-nm line in larger ones. Doubly ionized argon can undergo laser transitions in the UV, ranging from 275 to 363 nm. The gain of the UV and blue-violet argon lines is relatively low; this means that while an air-cooled argon-ion laser with a plasma tube on the order of a foot long can be used to produce tens of milliwatts of output at 488 or 515 nm, consuming ∼1 kW of electrical power, a substantially larger device, typically with a plasma tube 3 or 4 feet long, must be used to generate UV and/or to produce more than a few milliwatts at 454 or 457 nm. In the larger ion lasers, a strong magnetic field generated by a solenoid is used to confine the plasma to the central or bore region of the plasma tube, with the necessary solenoid current further increasing power requirements. It is thus necessary to use water, flowing at a rate of several gallons/minute, to cool the laser. As previously mentioned, the efficiency of argon lasers is low; the optical power output is between 0.0001% and 0.01% of the electrical power input. The air-cooled argon-ion lasers in most commercial flow cytometers are equipped with fixed mirrors limiting the emission wavelength to 488 nm. Larger ion lasers installed in flow cytometers (and air-cooled models available from laser manufacturers) usually feature interchangeable mirrors and a Littrow prism assembly that permits wavelength selection. In addition to the strong blue-green and green lines at 488 and 515 nm, argon-ion lasers emit at 454, 457 (violet-blue), 465 (blue), 472, 476 (bluegreen), 496, and 501 (green) nm. Emission can also be obtained in the UV at 351 and 363 nm and in the green at 528 nm using specially coated mirrors; in addition, the largest highpower argon-ion lasers can produce some deep UV lines between 275 and 305 nm. The complexity of ion lasers and the difference in construction between low- and high-
1.9.6 Supplement 27
Current Protocols in Cytometry
Table 1.9.1 Cytometrya
Ar Ion
Discrete Emission Wavelengths (nm) of a Selection of Lasers Usable for
Kr Ion
He-Cd
He-Ne
Solid State
275.4 300.3 302.4 305.5 325.0 333.6 337.4 350.7 351.1 355 (Nd:YAG × 3)
354.5 356.4 363.8 406.7 413.1 415.4
430 (Cr:LiSAF) 441.6 454.5 457 (Nd:YVO4 × 2) 457.9 460 (semi) 468.0 465.8 472.7 476.2 476.5 482.5 488.0 496.5 501.7 514.5
488.0 (semi) 514.5 (Yb:YAG × 2) 520.8
528.7 530.9 532 (Nd:YAG × 2) 533.7 537.8 543.5 568.2 594.1 611.9 632.8 635.5 636.0 647.1 676.4 752.5 799.3 aDiode, dye, and some solid-state lasers are tunable over ranges of wavelengths. See text for details.
Flow Cytometry Instrumentation
1.9.7 Current Protocols in Cytometry
Supplement 27
Lasers for Flow Cytometry
power argon-ion lasers make the relationship between between price and power output highly nonlinear. A laser that emits 25 mW at 488 nm costs ∼$5000; six or more times that amount buys a laser that delivers anywhere from 10 to 30 times the power. If UV output and/or multiwatt visible output to pump a dye laser are not essential, a high-power, air-cooled argon-ion laser with a power output of over 100 mW at 488 or 515 nm may be a good choice; these are available for under $10,000. The plasma tube in an argon-ion laser typically lasts for several thousand hours of operation; replacement thereafter is likely to cost between one-third and one-half the original price of the system. When considering installation of a water-cooled ion laser, the costs of wiring a room with 208- to 220-V, 70- to 100-A, three-phase electrical service, and of providing cooling water and/or the recirculating cooling system mandated by environmental legislation almost everywhere, must be factored into the calculations. It is also advisable to verify that the laser will fit, or can be made to fit, onto the cytometer’s optical bench. Large argon-ion lasers can be equipped with special mirrors permitting simultaneous emission at 351/363 nm in the UV and at 488 nm and other visible wavelengths. This provides the dual-wavelength source needed, for example, to do simultaneous analysis of DNA, using the UV-excited blue fluorochrome Hoechst 33342, and surface antigens, using fluoresceinated antibodies. In order to emit in the UV, the laser must be run at very high current; since the laser is much more efficient at 488 nm than in the UV, power output at 488 nm is quite high. If the sensor in the light output regulator circuit responds to UV and visible light, or to light at 488 nm alone, there may be considerable fluctuation in UV power output. If the sensor’s optical bandwidth is restricted so that 488-nm light is blocked, the sensor and the regulation electronics then respond to fluctuations in UV power output. The relatively large changes in current that may be necessary to keep UV output stable can then result in large fluctuations in power output at 488 nm, which may make it preferable to use separate sources for UV and visible light. The noise characteristics of large and small ion lasers may be different, predominantly owing to differences in the types of power supply used. Most older large ion lasers used linear power supplies, in which most of the noise is at small multiples of the line frequency, typically below 1 kHz. Air-cooled ion lasers use
switching power supplies, which have higher noise levels, typically at the switching frequency of a few tens of kHz. The noise on the output of the larger lasers is typically specified as <0.2% RMS (root mean square), while that on the output of the smaller ones is specified as <1% RMS. (The percentage of RMS noise is the coefficient of variation of the output power.) As previously mentioned, laser noise may determine measurement sensitivity, particularly for scatter measurements. This, and not the difference in power output, explains why it is generally possible to detect scatter signals from smaller particles when a larger ion laser is used as a source than when a smaller laser is used in the same instrument. As noted below, small, energy-efficient solid-state lasers emitting at 488 to 492 nm have recently become available. They have been incorporated into some flow cytometers and will probably replace both air- and water-cooled argon-ion lasers as blue-green light sources in cytometers within a few years. Krypton-ion lasers first came into use in flow cytometry for two-color immunofluorescence measurements. The 568-nm line from a krypton-ion laser was used to detect antibodies labeled with Texas Red or XRITC, both of which are derivatives of rhodamine 101, while 488-nm excitation from an argon-ion laser was used for fluorescein excitation. Krypton-ion lasers have lower gain and lower efficiency than argon-ion lasers, but emit over a much broader spectral range; they can produce blue-green at 468, 476, and 482 nm, green at 520 and 530 nm, yellow at 568 nm, and red at 647 (the strongest krypton line) and 676 nm, all at once, explaining their popularity for laser light shows. With different mirror sets, k ry pto n- ion lasers c an also emit UV (337/350/356 nm), violet (406/413/415 nm), and infrared (752/799 nm) light. Unfortunately, the optimum values for gas pressure and solenoid magnetic field for krypton-ion laser operation are different for different lines. Since the gains at all lines are low, these parameters must be well controlled to maintain laser action. Krypton-ion laser (and UV argon-ion laser) mirrors need to be extremely clean, and if the optical alignment of the laser is not near-perfect in the visible, as indicated by maximum all-lines power output near the manufacturer’s specs, a user is unlikely to get an ion laser to work in the UV. Low gain at all lines means that krypton-ion laser plasma tubes run hotter and fail sooner than argon-ion laser tubes. There is also substantial competi-
1.9.8 Supplement 27
Current Protocols in Cytometry
tion among the visible krypton lines; for example, running in light control mode with optics that allow simultaneous yellow and red emission often results in alternating fluctuations in the yellow and red power outputs. Argon- and krypton-ion lasers both normally produce TEM00 (Gaussian) output beams. Mixed-gas lasers, filled with argon and krypton and capable of emission at any of the visible output wavelengths available from krypton as well as at the major argon lines, are also available; these may have multimode output, but can generally be used in flow cytometers because the output beam is not multilobed. Keeping the optics of an ion laser clean and in alignment is a challenge best met by following the manufacturer’s instructions. Mirrors and Brewster windows are generally cleaned with methanol. Acetone, which is a good cleaner for some mirrors and destroys others, should be used only if the manufacturer recommends it. Electronic or HPLC-grade solvents are generally free enough of contaminants to be safe to use for cleaning optics, but the slightest contamination of the solvent with grease, from fingers or elsewhere, will result in the deposition on laser optics of a residue, which can drastically reduce output. Changing mirrors and getting the laser to lase again is a tedious procedure which must be done a little bit differently for each manufacturer’s lasers. It is learned by doing, preferably with someone more knowledgeable in attendance, at least the first time, and may remain something of a black art even to experienced operators. Some of the newer large ion lasers are said to incorporate features that facilitate manipulation of the optics; the author has had no experience with these.
Helium-Neon Lasers He-Ne lasers, once the smallest, most efficient, least complicated, least expensive, and most commonly used lasers available, have been displaced from all of the above categories by diode lasers, but remain useful for flow cytometry. The plasma tube of a He-Ne laser is not unlike the tubing of a neon sign; it contains a mixture of helium and neon gases through which a relatively low electric current is passed, raising helium atoms to electronically excited states. Collisions between the helium and neon atoms transfer the excitation energy to the latter, which may drop to lower excited states via any of several laser transitions, ranging in wavelength from green (543 nm) to infrared (3.39 m). He-Ne lasers typically require only
air cooling by convection and use fixed mirrors. They are more efficient than argon-ion lasers, yielding 1 to 50 mW optical power for electrical power inputs of a few hundred watts or less. The most common He-Ne lasers emit red light at 633 nm; other visible wavelengths at which He-Ne lasers are now available include 543, 594, and 611 nm. The head assembly containing a He-Ne plasma tube is typically 1 to 2 feet long; the highest power He-Ne lasers may be twice that length. Almost all He-Ne lasers emit (Gaussian) TEM00 beams. Noise levels vary; a 0.8-mW system with 0.05% RMS noise was used for extinction measurements in some commercial systems in the 1970’s and 1980’s, but lasers with the >5-mW output generally needed for fluorescence measurement are generally specified for a 1.0% RMS noise level. Red (633-nm) He-Ne lasers used in cytometry are typically polarized. Green (543nm) systems are most often not polarized, because the extremely low gain of the green line, which limits power output to a few milliwatts, makes it difficult to introduce Brewster windows into the system without unacceptable light losses. Depending on power level, He-Ne lasers cost between a few hundred and a few thousand dollars; plasma tube lifetimes are typically 10,000 hr or more. One common application of red He-Ne lasers in fluorescence flow cytometry is in measurements of immunofluorescence, using antibodies conjugated to the phycobiliprotein allophycocyanin. In instruments with reasonably efficient light collection, diode lasers emitting in the 635- to 640-nm region, which are now available at power levels up to 35 mW, and which are incorporated in some commercial instruments, may be a better choice, although He-Ne lasers are to be preferred when constancy of wavelength is critical. The 543-nm He-Ne laser wavelength is useful for excitation of immunofluorescence from antibodies labeled with phycoerythrin (one form of which has an absorption maximum at 545 nm), and can also be used to excite DNA stains such as propidium. Green He-Ne lasers are likely to face increasing competition from 532-nm frequency-doubled YAG lasers in these applications; prices of the latter are expected to fall substantially as commercial applications increase. The 594-nm He-Ne laser is usable for excitation of Texas Red and of the Texas Red/allophycocyanin combination; it operates at the same wavelength at which dye lasers are often used for these purposes. Power outputs are now in the 5- to 10-mW range, which should be
Flow Cytometry Instrumentation
1.9.9 Current Protocols in Cytometry
Supplement 27
adequate for immunofluorescence work in instruments with efficient light collection. The major disadvantage of the 594-nm laser is the proximity of its lasing wavelength to the fluorescence emission region of phycoerythrin.
Helium-Cadmium Lasers
Lasers for Flow Cytometry
He-Cd lasers, which can emit in the blueviolet (5 to 200 mW at 441 nm) and in the UV (1-100 mW at 325 nm; 1-10 mW at 354 nm), can be useful sources for flow cytometry. Like He-Ne lasers, they plug into the wall and do not require water cooling; they need few or no adjustments and have long plasma tube lifetimes. They are also cheaper than most ion lasers, ranging in price from ∼$7,000 to over $20,000. The lasing medium is cadmium vapor; cadmium ions are typically excited by collision with excited helium atoms. The pressure of the cadmium and helium and the temperature of the medium must be carefully controlled to assure stable operation. He-Cd lasers are intermediate in efficiency between argon-ion and He-Ne lasers. They typically emit TEM00 (Gaussian) beams in the blue-violet, and TEM01* (donut-shaped) beams in the UV. Some models are linearly polarized; others are randomly polarized. Plasma tube lifetimes are on the order of a few thousand hours. The 325- and 354-nm UV lines available from He-Cd lasers are suitable for excitation of the fluorescence of the DNA stains DAPI, DIPI, and Hoechst 33342 and 33258, and of the calcium probe indo-1, as well as other UV-excited dyes. Parinaric acid, used for analysis of lipid peroxidation, is optimally excited near 325 nm and only marginally excited at the longer UV wavelengths (350 to 365 nm) available from argon- and krypton-ion and arc lamp sources. The 441-nm He-Cd line is useful for excitation of the chromomycin family of DNA stains (olivomycin, chromomycin A3, and mithramycin), but these dyes can also be excited by the 457-nm argon-ion laser line. He-Cd lasers emitting both 325 and 441 nm are available, with either wavelength selectable by switching filters; there is no significant competition between the UV and blue-violet lines. He-Cd lasers are made by Melles Griot in the U. S. and Kimmon in Japan. The He-Cd medium is also capable of laser transitions in the green and red spectral regions, making it possible to produce a “white-light” laser. This requires a design substantially different from most of those now available (offered by The COOKE Corporation), and there are probably more cost-effective ways of obtaining multiple
lines. The principal problem with He-Cd lasers is noise. The major noise component is plasma noise, at frequencies between 300 and 400 kHz. It is difficult to keep RMS noise levels much below 1.5% even when the laser is new; noise levels tend to increase thereafter, especially if the laser is left idle for long periods of time, because this leads to an irreversible increase in helium pressure in the plasma tube, which increases noise. This effect can be minimized by running the laser for a period of several hours at least once a week. Once the noise develops, however, it cannot be reduced by such regular operation. Noise is not a problem when the laser is used for ratiometric measurement of calcium based on measurement of indo-1 emission in two spectral regions; the noise appears in both the numerator and denominator of the ratio, and is therefore factored out. However, noise can and does interfere with precise measurement of DNA content using DAPI or the Hoechst dyes. The effects of noise have been reduced in laboratory-built systems using noise compensation electronics and electro-optic modulators. The former, while inexpensive, are not available in commercial flow cytometers. The author has had better luck with UV He-Cd lasers than with UV argon- or krypton-ion lasers, which in his experience also tend to be noisy and which cost substantially more and require more elaborate and expensive power and cooling arrangements. 325-nm He-Cd lasers are difficult to use in microscope-based instruments because the wavelength is very poorly transmitted through most optical glasses.
Dye Lasers Dye lasers are used predominantly as excitation sources for antibody labels based on rhodamine 101, e.g., Texas Red and XRITC. They also provide acceptable excitation for the phycobiliproteins phycocyanin and allophycocyanin and their tandem conjugates, and for cyanine dye labels such as Cy5. The dye lasers used in flow cytometry are capable of continuous (i.e., CW) operation; the lasing medium is a fluorescent dye, usually rhodamine 6G, dissolved in an organic solvent such as ethanol or ethylene glycol. Which dye is actually used depends on the wavelengths at which operation is desired; dyes now available for use with blue-green/green (457- to 515-nm) argon-ion pump lasers permit operation at wavelengths extending from 540 nm to >900 nm. Selection of the output wavelength for any given dye is
1.9.10 Supplement 27
Current Protocols in Cytometry
usually done with a wedge or filter rather than with a prism. A dye laser is tunable over a wider continuous range of output wavelengths than is a HeNe, ion, or He-Cd laser, because the laser transitions in a dye laser are between electronic energy states of a molecule, rather than between electronic energy states of an atom, as in the other types. Each electronic energy state of a molecule has associated with it a number of vibrational and rotational energy states. Excitation of dye molecules by absorption of light from the pump laser typically leaves the molecules in the first electronic excited state and one of the associated vibrational excited states, from which they drop, usually nonradiatively, to the vibrational ground state, producing a population inversion in the presence of a pump power input above threshold. Laser transitions are possible from the first excited electronic/ground vibrational state to any of the vibrational states associated with the electronic ground state. This means that laser action can be sustained over a range of wavelengths, made essentially continuous by the thermal broadening of absorption and emission spectra that is characteristic of all molecules above absolute zero. The dispersion angle of the tuning element determines the precise output wavelength. The dye in a CW dye laser is circulated through a nozzle, producing a flat-walled stream; circulation is necessary to minimize bleaching and for cooling by a heat exchanger. The volume of the lasing medium is quite small, because the gain is very high. CW dye lasers require minimal electrical power, most of which is used to operate the circulator pump. While some dyes bleach faster than others, all must be replaced after a few months’ operation. This, and the fact that CW dye lasers are relatively hard to keep aligned and may not maintain output power as stably as do ion lasers, have restricted their use as sources for flow cytometry. The threshold power required from the pump laser to achieve output from a rhodamine 6G dye laser is usually ∼700 mW; this is generally obtained from a multiwatt, water-cooled argon-ion laser. With 700-mW pump power input, over 100 mW of light can typically be obtained from rhodamine 6G at wavelengths selectable between ∼570 and 620 nm; while the 14% efficiency of the dye laser is admirable, the pump laser remains extremely inefficient. Dye lasers typically emit Gaussian beams. For Texas Red excitation, a dye laser could
be operated at ∼595 nm, but a longer wavelength, 605 to 610 nm, is now more often used; this is better for simultaneous excitation of Texas Red and allophycocyanin, and interferes less with phycoerythrin emission. The disadvantage of using dye lasers in dual-laser systems lies in the resulting inability to use either the dye laser or the pump laser as a UV source. While, in principle, one can get around this using dual-wavelength UV/visible mirrors in the argon pump laser, this solution requires great technical skill on the part of the user and might also involve frequent plasma tube replacements. CW dye lasers are now used relatively infrequently in flow cytometry; the few people with enough skill and money to operate and maintain them now seem to be switching to high-power 532-nm YAG lasers (see below) as pumps.
Diode Lasers Tens of millions of infrared diode lasers are now sold every year; they are incorporated into compact disc players, laser printers, and CDROM readers, and now cost only a few dollars each. Red diode lasers emitting anywhere between 630 and 685 nm are also widely used in laser pointers and bar code readers, and are only slightly more expensive. In recent years, red (635 to 640 nm) diode lasers have been incorporated into laboratory-built and commercial flow cytometers, usually as an alternative to 633-nm He-Ne lasers; in the future, it is likely that diode and solid-state lasers, which will be discussed in the next section, will largely replace ion, He-Ne, and He-Cd lasers as light sources for flow cytometry. Like transistors, diode lasers are made of materials classed as semiconductors. The light emission from semiconductors is not from excited atoms or ions, as is the case in ion, He-Ne, and He-Cd lasers, and not, strictly speaking, from excited molecules, as is the case in dye lasers; the electrons that are excited in a diode laser are “free” in a crystalline material. Such free electrons also occur in metals; the nuclei in a metal are packed relatively close together, and the electrons in the outermost shells are not tightly held by any given nucleus and may be excited from the so-called valence band to the so-called conduction band by ambient thermal energy. Energy transfer among electrons in a metal can occur fairly readily; this is what makes metals good conductors of electricity. Semiconductors are so named because, while they do not conduct electricity well when
Flow Cytometry Instrumentation
1.9.11 Current Protocols in Cytometry
Supplement 27
Lasers for Flow Cytometry
in an unperturbed state, they may become conductive in the presence of an applied electric field or of incident light. Their electronic structure differs critically from that of metals in that there is a substantial energy difference, or bandgap, between the valence and conduction bands, with almost all the electrons lying in the valence band under normal conditions. The application of an electrical current to an appropriately configured semiconductor can result in light emission as electrons relax from the conduction to the valence band; this type of spontaneous emission is what occurs in light-emitting diodes (LEDs). A laser diode is basically an LED, the geometry of which is tailored to provide a resonator structure that will support stimulated emission. The active regions of diode lasers typically have dimensions on the order of a few micrometers. They use either polished facets on the semiconductor material itself or adjacent structures of differing refractive index to perform the function of the mirrors used in larger lasers. Because the efficiency of diode lasers is extremely high, typically ∼20% to 30%, high reflectivity is not needed. Also, because hundreds of lasers can be produced from slices of a single semiconductor wafer, diode lasers are much less expensive than any other type of laser. The first material of which diode lasers were made is gallium aluminum arsenide (GaAlAs); this is a so-called ternary semiconductor (made up of three elements). The bandgap energy, and therefore the emission wavelength, is varied by changing the ratio of gallium to aluminum in the semiconductor material. The emission wavelengths theoretically achievable with GaAlAs lasers range from ∼650 nm, at which point the material is almost pure AlAs, to ∼900 nm, at which point the material is almost pure GaAs. The GaAlAs lasers made to date which emit below 750 nm are typically unstable, however, and succumb to thermal runaway within minutes to hours. The GaAlAs lasers now available in quantity emit at 750 to 780 nm; these are used in compact disc players, CD-ROM readers, and laser printers. Gallium indium phosphide (GaInP) lasers go down to ∼670 nm, providing up to 20 mW emission, and aluminum gallium indium phosphide (AlGaInP) devices are now available with a 635-nm emission wavelength and power outputs of 3 to 35 mW. 635-nm diode lasers are now available as light sources in commercial cytometers from a number of manufacturers. Unlike the other types previously discussed,
diode lasers do not normally emit anything approximating a Gaussian beam. The emitting surface of most of the diode lasers now in commercial production is a stripe ∼1 µm high and a few micrometers wide. The beam diverges more in the direction perpendicular to the long axis of the emitter surface than in the direction parallel to it. When a spherical or radially symmetric aspheric lens is used to collect light from the laser, the resultant collimated beam is asymmetric, and when focused to a small spot, often shows substantial intensity variations along one dimension or another. It is thus generally more difficult to achieve uniform illumination of the illumination volume of a flow cytometer using a diode laser than using lasers that emit Gaussian or nearGaussian beams. Further difficulties are introduced because diode lasers show some lot-tolot and unit-to-unit variation in beam geometry, simply because of slight differences in structure. In addition, output wavelength may vary by as much as several nanometers as a function of small differences in semiconductor composition, and diode lasers are also subject to wavelength variation with changes in temperature. Output within a desired wavelength range is, at present, best achieved by negotiating with the laser supplier for selected, tested units. Improved beam quality is now offered by firms such as Blue Sky Research, who mount a small cylindrical lens assembly inside the case of a laser diode, providing a radially symmetric, near-Gaussian output beam that can readily be focused using the same crossed cylindrical lenses used to derive flow cytometer illumination from other laser types (see UNIT 1.6). In addition to their small size (laser, power regulator, and collimating optics in a package less than 2 in. [5.08 cm] long and less than 1 in. [2.54 cm] in diameter), minimal power requirements (a 15-mW diode laser draws a fraction of a watt of electrical power), and low cost (a 15-mW, 635-nm diode laser with a circularized beam costs ∼$1,000, around half the cost of a He-Ne laser emitting 15 mW at 633 nm), diode lasers as typically supplied have extremely low noise. A photodiode that senses laser output is built into the case of most laser diodes; this is connected to a feedback-controlled power-supply regulator that converts a nominal 5- to 6-V DC input from a battery or DC supply to the lower voltage required by the diode and regulates current to prevent the laser from self-destructing. The laser effectively operates in light control mode, with <0.05% RMS noise; this makes diode lasers well suited for
1.9.12 Supplement 27
Current Protocols in Cytometry
extinction measurements. The same dyes can be used with 635- to 640-nm diode lasers as with 633-nm He-Ne laser. The list of dyes usable with 670-nm lasers is somewhat more restricted. These sources are usable for DNA content analysis using rhodamine 800 or oxazine 750, for immunofluorescence analysis using the cyanine label Cy5.5, and for membrane potential analysis using dibenzoDiIC1(5). The latter two dyes are not widely available commercially, nor are flow cytometers with 670-nm lasers. It is relatively difficult to build flow cytometers using 780-nm diode lasers, because it is difficult to see the beam when setting up the laser, and although scatter and extinction measurements can readily be done at this wavelength, few dyes suitable for 780-nm excitation have been well characterized for use in cytometry. Diode lasers are used for optical disc recording and playback, applications for which achieving a smaller focal spot size allows more information to be stored in and retrieved from the same area. Recordable CDs, written and read with 780-nm diode lasers, top out at around 700 Mb per disc; using a shorter-wavelength (650- to 660-nm) red diode laser instead of an IR laser allows a DVD to store 5.7 Gb on a side. Considerable money and effort have been expended in developing even shorterwavelength “violet” (∼405-nm) diode lasers, which extend optical storage capacities to >25 GB/disc. These lasers are now available with emission in the range from 370 nm (UV) to 445 nm (blue-violet), providing as much as 10 mW of UV and 30 mW of violet light. They are small and energy-efficient, although they typically require both temperature control and light control to maintain stable output. Violet laser diodes are good excitation sources for many materials that would otherwise require a krypton-ion laser for excitation. The list includes Cascade Blue and Cascade Yellow labels from Molecular Probes, monobromo- and monochlorobimane, both used for detection of intracellular glutathione, the cyan fluorescent reporter protein ECFP, and the DNA dyes mithramycin and chromomycin A3. Laser diodes operating at the short (370- to 400-nm) end of the range are effective excitation sources for DAPI (38% of maximum excitation at 395 nm) and usable with the Hoechst DNA dyes; they cannot be used with the calcium probe indo-1, but it may be possible to synthesize a similar calcium probe that would work with 370-nm excitation. The Hoechst dyes and DAPI can also be excited, albeit
suboptimally, by 405-nm laser diodes. By 2002, almost all major manufacturers of fluorescence flow cytometers were offering violet/UV diode lasers as excitation sources.
Other Solid-State Lasers In solid-state lasers using media such as ruby and neodymium (Nd)-doped YAG, a rod made from the lasing material is optically pumped. While flashlamps were originally the predominant pump sources for pulsed solid-state lasers, newer, high-power AlGaAs diode lasers, emitting near 800 nm, are increasingly being used for pumping Nd:YAG lasers; this technique and frequency doubling or summing have made it possible to produce small, energy-efficient CW solid-state lasers emitting green and blue light. Laser transitions in Nd can be exploited when this element, nominally in a triply ionized state, is used as a dopant in YAG, yttrium vanadate (YVO4), or yttrium lithium fluoride crystals, or in glass. The principal laser transition in Nd:YAG is at 1064 nm; slightly different wavelengths are obtained for Nd in materials other than YAG. The conversion efficiency of Nd:YAG, when optically pumped, may be ∼20%, and power outputs of tens of watts may be obtained when a Nd:YAG rod of suitable dimensions is pumped by an array of diode lasers emitting, in the aggregate, close to 100 W at 800 to 850 nm. Efficient frequency doubling can be obtained by placing the doubling crystal inside the laser cavity; power outputs of 5 W or more at 532 nm are now available from Nd:YAG and Nd:YVO4 lasers, requiring power inputs of less than 100 W from a 110-V AC line. Even high-powered doubled Nd:YAG lasers generally require no more than fan cooling; lower-power models may be cooled by convection. Lifetimes are in the thousands of hours, and although earlier models had noise problems, some current production systems specify <0.1% RMS noise. The output beam is typically Gaussian. Green Nd:YAG lasers with power outputs of 5 to 25 mW, suitable for flow cytometry, may cost between $5000 and $15,000, depending on the manufacturer. The less expensive lasers now available tend to be too noisy for use in flow cytometers, but prices of the better systems are expected to drop. The large (∼5-W) lasers, as mentioned above, make excellent pumps for dye lasers. Frequency-doubled, diode-pumped green Nd lasers can be used for excitation of tetramethylrhodamine, phycoerythrin and its tandem conjugates, ethidium, propidium, and 7-aminoactinomycin D, to name a few dyes.
Flow Cytometry Instrumentation
1.9.13 Current Protocols in Cytometry
Supplement 27
Lasers for Flow Cytometry
Because 532 nm excites much less autofluorescence in mammalian cells than does 488 nm, and is also closer to the excitation maximum of phycoerythrin, the substitution of a 532-nm laser for a 488-nm laser generally makes it possible for a flow cytometer to detect fewer phycoerythrin-labeled molecules bound to a cell. Tripled UV YAG lasers are potentially useful for excitation of DNA stains and calcium probes. They emit at the right wavelength (355 nm), but they operate only in pulsed modes. However, a technique called mode-locking allows the lasers to emit regularly spaced pulses at 80 to 100 MHz; this repetition rate is high enough that the laser behaves more or less as if it were a CW light source. Both Lightwave Electronics and Spectra-Physics have introduced mode-locked 355-nm Nd:YAG lasers, and BD Biosciences is offering them as sources in both benchtop systems and sorters. The lasers are expensive, possibly even more expensive than the UV ion lasers that represent the only alternative sources, and offer power levels of tens to hundreds of milliwatts. However, mode-locked Nd:YAG lasers share the modest power and cooling requirements of their CW cousins, and therefore do not require the expensive infrastructure needed to support UV ion lasers. An attempt to develop a solid-state CW UV laser for flow cytometry is ongoing at Light Age, a company that makes alexandrite lasers. Alexandrite, which is beryllium aluminum oxide containing chromium, emits between 700 and 850 nm, and can be pumped by 635- to 670-nm diode lasers. UV emission of 10 to 15 mW at ∼370 to 380 nm has been achieved from a frequency-doubled, diode-pumped alexandrite laser; one of Light Age’s prototypes was run for several months, and maintained a reasonably low optical noise level. The tunability of alexandrite lasers is another point in their favor. The lasers have been expensive, although probably no more expensive than UV ion lasers or mode-locked UV Nd:YAG lasers. Since experimental UV diode lasers have already achieved power outputs in the 100-mW range, demand for other solid-state lasers as UV sources for cytometry should eventually diminish. Alexandrite is an example of a vibronic crystalline laser material. Such materials behave similarly to laser dyes; laser transitions may occur between an excited electronic state and any vibrational state associated with a lower electronic energy state, allowing output
to be tuned over a broad range. Titanium-doped sapphire lasers can operate in pulsed or CW mode between 660 and 1180 nm. Pump energy can be supplied by a high-power green YAG laser. Pulsed Ti-sapphire lasers are used for multiple-photon excitation of fluorescent dyes and intrinsically fluorescent cellular constituents; their high cost (typically over $100,000) has, to date, limited their use. Cr:LiSAF lasers, in which the laser transitions occur in Cr3+ ions in a matrix of LiSrAlF6, have been doubled to 430 nm; a laser based on this technology was briefly offered for sale by Melles Griot, but violet diodes may be a better bet for this wavelength range. For substantially higher blue-violet power output, Melles Griot also offers 457nm CW lasers made by frequency-doubling the output of Nd:YVO4. These are fan-cooled and run on house current, drawing ∼75 W at the plug, and emit as much as 400 mW. Such lasers could replace the 457-nm water-cooled argon lasers now used in chromosome sorters. Although green Nd:YAG lasers are good excitation sources for many dyes and labels used in cytometry, they cannot excite fluorescein, which not only is still very popular as a label, but has also been derivatized into probes for a large number of structural and functional parameters. Fluorescein can be excited at wavelengths as high as 515 nm. The primary emission wavelength of ytterbium YAG (Yb:YAG) lasers is 1029 nm; they could be doubled to 514.5 nm and used for fluorescein excitation, although this has not yet been reported. Yb:YAG lasers have some advantages over Nd:YAG in terms of ease of pumping and stability, and this suggests that green Yb:YAG lasers might have fewer problems than green Nd:YAG lasers, but the emergence of 488-nm solid-state lasers (see below) has probably discouraged development of a green Yb:YAG for cytometry, and is also likely to set back development of 473-nm, frequency-summed Nd:YAG lasers, which have been offered in some cytometers. Doubling a semiconductor laser operating at 750 to 1000-nm will yield UV, violet, blue, or blue-green light. The process is inefficient; a laser capable of producing several hundred milliwatts is required to get a few milliwatts of visible light. The resulting laser system is, typically, considerably more complex and more expensive than a diode laser, because other components, notably an external mirror and a crystal of the material used for harmonic generation, must be incorporated into the system. Before 2000, there were a few doubled diode
1.9.14 Supplement 27
Current Protocols in Cytometry
lasers on the market, but they were neither powerful nor cheap. Coherent’s “Sapphire” laser, named for its blue output rather than its chemical makeup, was introduced to the cytometry community in May, 2000. The Sapphire is a frequency-doubled, diode-pumped semiconductor laser; Coherent, Inc., currently sells a 10-mW, 460-nm model and both 20-mW and 200-mW models emitting at 488 nm; output wavelengths are guaranteed to ±2 nm. BD Biosciences, Beckman Coulter, and DakoCytomation, among others, offer the 20-mW, 488-nm Sapphire in benchtop instruments; iCyt-Visionary Bioscience will retrofit the 200-mW, 488-nm Sapphire to sorters. These lasers are not inexpensive, but they are small in size, run on small amounts of house current, and reportedly feature long lifetimes (>10,000 hr). Another laser company, Novalux, offers another frequency-doubled semiconductor laser, the Protera; this will also deliver 10 to 20 mW at 488 nm, and is priced competitively with argon lasers in the same power range. The semiconductor lasers in both Coherent’s Sapphire and Novalux’s Protera are surface-emitting lasers which, unlike typical edge-emitting diodes, produce circular, near-Gaussian beams. The technology can yield relatively inexpensive laser sources at wavelengths ranging from UV to yellow. Recent issues of optical industry trade publications indicate that several other manufacturers, including Blue Sky Research, Spectra-Physics, and Uniphase, are now poised to bring blue-green solid-state lasers to market, and suggest that at least some of them will be cheaper than argon-ion sources of equivalent power. A solid-state laser can be made using an appropriately doped glass fiber as the lasing medium. This is an up-conversion rather than a frequency-doubled laser; pumping to the excited state is accomplished by two-photon absorption of light from a diode laser. Fiber lasers can be made to operate at various visible wavelengths; they also have excellent pointing stability, meaning that the beam has less tendency to undergo slight changes in direction than do beams from other types of lasers. This would be advantageous for flow cytometry. The German company Unique-m.o.d.e. AG is expected to have a 10-mW, 491-nm fiber laser on the market by the end of 2003.
classified as hazardous by the U. S. Bureau of Radiological Health. At power levels just above 800 µW, laser beams are potentially harmful to the eye; substantially higher powers, e.g., hundreds of milliwatts in visible and UV beams, can burn skin and set fire to paper and paint, among other things. Changing or cleaning the optics in some large, water-cooled ion lasers, if it involves removing the outer casing of the laser, may expose the operator to potentially lethal electric currents (tens of amperes at hundreds of volts). To date, as far as the author is aware, the only fatalities associated with laser use have been due to electric shock, and the number of severe eye injuries reported has been relatively small. Flow cytometers, like other apparatus that incorporate lasers, are required by law to be equipped with light shielding, and operators are required to be provided with safety goggles that prevent laser light from reaching the eye. While the users of most modern benchtop commercial flow cytometers are not generally exposed to the laser beam, people who operate larger instruments, particularly those incorporating large ion lasers with interchangeable mirrors, generally must deal with the laser when the light shields are removed. Unfortunately, adjustments to the laser and illumination optics are best made, and often only possible, when one can see the beam, which requires removing the goggles. In principle, one could use a video camera and/or a laser power meter to determine the presence and intensity of a laser beam while wearing goggles that block the laser wavelength; most people do not.
FUTURE DIRECTIONS The original version of this unit, published in 1998, suggested that “It would, obviously, be desirable to replace the ion, He-Ne, and He-Cd lasers now used in flow cytometry with diode or solid-state lasers, which would offer greater efficiency, smaller size, lower power consumption, lower noise, and it is to be hoped, lower cost.” The newest flow cytometers now offer diode and/or solid state light sources for UV, violet, blue-violet, blue-green, green, and red light, providing adequate excitation for almost all of the fluorescent probes and labels (see UNIT 1.19) now in widespread use. Prices for all of these light sources are trending down, which is a good sign.
LASER SAFETY
Lasers that emit more than 800 µW anywhere from the UV to the near infrared are
Flow Cytometry Instrumentation
1.9.15 Current Protocols in Cytometry
Supplement 27
KEY REFERENCES Harbison, J.P. and Nahory, R.E. 1997. Lasers: Harnessing the Atom’s Light. Scientific American Library, New York. Aimed at the interested layman, this book is beautifully illustrated and includes detailed discussion of the operation of semiconductor lasers. Hecht, J. 1992. The Laser Guidebook, 2nd ed. Tab Books (McGraw-Hill), Blue Ridge Summit, Pa. Hecht, J. 2001. Understanding Lasers: An EntryLevel Guide, 2nd ed. IEEE Press, New York.
Hecht’s books provide substantial technical detail about lasers in a manner accessible to readers without a strong background in physics and engineering. Shapiro, H.M. 2003. Practical Flow Cytometry, 4th ed. Wiley-Liss, New York. This book provides more particulars about specific flow cytometric applications of various lasers than appear here.
Contributed by Howard M. Shapiro West Newton, Massachusetts
Lasers for Flow Cytometry
1.9.16 Supplement 27
Current Protocols in Cytometry
Techniques for Flow Cytometer Alignment A wide variety of commercial and custom flow cytometers are used for biological and biomedical analyses. Users of flow cytometry instrumentation need to monitor the performance of their instrument on a daily basis to ensure that the system is operating within the desired specifications. Because the instruments in use today span several generations of development, it is not possible to provide a complete generic procedure for “How to Align a Flow Cytometer.” However, some general techniques that can be applied to a variety of instrumental configurations are presented in the discussion that follows. There are two levels of system alignment: routine alignment checks and complete alignment. It bears noting that the procedures used in a complete alignment are often extraordinary—that is, the steps required are not those used in general routine instrument operation. Hence, hazards not encountered in everyday operations are possible, including possible exposure to electrical power sources that have high current capacity and/or generate high voltages, and to intense light beams from lasers and/or arc lamps. The electrical hazards become even more of a concern when conducting fluids such as sheath fluid or laser cooling water are present. A first consideration is to differentiate between checking or verifying instrument alignment and active alignment of a cytometer. To verify the proper operation of an instrument, standard particles of some kind are analyzed (UNIT 1.3) using standard settings for laser power, PMT high voltage, amplifier gain, and so forth. When an instrument is initially installed, the field service technician will leave data sheets and/or computer files that describe the results of analysis of standard particles supplied by the instrument manufacturer to provide a benchmark. For many instruments, particularly nonsorting ones, the operator can use only “noninvasive” techniques to verify correct operation. For the more complex instruments, which have greater flexibility and more alignment options, more invasive techniques can be used. In these cases, the challenge in a complete instrument alignment is to bring the flow, excitation, and detection axes together in such a way that they intersect at, and are focused on, a common point in space. After accomplishing this, one must ensure that the proper optical
elements (such as filters, lenses, and apertures) are in place and that the electronics configuration is correct.
VERIFICATION OF ALIGNMENT For noninvasive approaches, a variety of tools are at the user’s disposal to assist in verifying instrument alignment. The first part of system verification or of an alignment procedure should be a visual inspection of the instrument and its control readouts. Knowing what a properly operating cytometer “looks like” is the starting point for checking performance.
Visual Inspection As a first step, the values of all control and status displays should be verified. These include laser power, sheath flow rate or pressure, sample flow rate or pressure, analysis rate, and any other indicators provided. These operating parameters should correspond to a standard set of conditions. In the flow cell of enclosed systems, the sample stream viewed through a microscope should be a centered thin dot at the laser beam intersection point. Swirls on the sides of the main sample stream are an indication of flow problems caused by excessive sample debris, a partial plug, or an obstruction in the flow channel. The sample stream should be of a reproducible size for standard settings of the sample and sheath flow rate controls. Sample stream diameters that are too large or too small indicate problems in the flow channel, such as partial clogs, empty sheath tank, faulty pressure settings, or other flow-related problems. In jet-in-air systems, the sample should be centered in the jet and the jet should enter the waste drain tube. Visual inspection of the laser beam path(s) includes determining where the laser beam is striking focusing lenses. No matter how meticulous one is in keeping the surfaces of lenses or filters clean, a small amount of dust and contaminants will always accumulate. Laser light scattered by these surfaces provides an indication of where the laser beam is striking. There will also be reflections from the lens surfaces, as well as from the surfaces of filters and of the flow cell. The positions at which these reflections strike the interior surfaces of the instrument also provide clues to misaligned
Contributed by James H. Jett, John C. Martin, and Robert C. Habbersett Current Protocols in Cytometry (1998) 1.10.1-1.10.6 Copyright © 1998 by John Wiley & Sons, Inc.
UNIT 1.10
Flow Cytometry Instrumentation
1.10.1 Supplement 6
parts in the laser beam path. Noting where the reflections appear in a properly operating instrument provides useful information for future alignments.
Alignment Particles
Techniques for Flow Cytometer Alignment
The next most obvious tool used in instrument alignment is a set of standard particles. Both fluorescent and nonfluorescent particles are used. For alignment of fluorescence detection channels, the goal is to achieve maximum signal strength with the lowest coefficient of variation (CV) possible. A minimum CV indicates that the system is optimally aligned. However, this might not always be desirable, as in cases where the goal is maximizing light collection efficiency—largest signal for a given high voltage and amplifier gain setting—rather than minimizing the CV. In some instruments, this conflict occurs in the forward light-scatter channel when alignment for the maximum signal does not correspond to the minimal CV. Thus, one might need to decide for a given experimental situation whether to optimize the alignment for maximum light collection efficiency or best resolution. The particles must have a better intrinsic CV than the best instrument response CV. For example, if the instrument is capable of 2% CV, the particles must have an intrinsic CV of ≤2%. In general, particles ≤2 µm in diameter have the best CV. Biological particles, such as propidium iodide– stained thymocytes, are also good reference particles for fluorescence. CVs as low as 0.5% have been observed on a variety of instruments using very small microspheres. Finally, in general, pulse-area measurements have better resolution than pulse-height measurements. The first thing to remember about light-scatter measurements of polymer particles is that these particles are not cells, and do not have the same indices of refraction or internal structures. These properties, along with size, affect the angular distribution of scattered light. The combination of these effects with the measurement geometry can lead to unexpected results, such as differences in the apparent size of particles with the same diameter but different indices of refraction. Under some conditions, therefore, measurement of particle size by light scatter can give confusing results. For these reasons, light scatter from cells is used primarily in an empirical manner for distinguishing among cell types. Absolute size measurements based on light scatter are extremely difficult to calibrate accurately and interpret properly. Nonfluorescent particles can be used to align light-scatter
detection channels and must have properties similar to those of particles used for the alignment of fluorescence detection channels—that is, their intrinsic CV should be less than is achievable by the light-scatter measurements. Most often, fluorescent particles are used for alignment of both fluorescence and light-scatter detection channels. Particles are also used to verify the linearity of fluorescence measurement channels. The first method is to use the locations of the peaks representing singlets and doublets in a pulsearea distribution. The mean values will give a two-point calibration curve that should pass through zero within the uncertainty of the measurements. This approach is valid only for pulse area, or measurements of the total fluorescence emitted as the particles or doublets pass through the laser beam. Doublet orientation and other effects may preclude drawing the same conclusions from pulse-amplitude distributions. See UNIT 1.4 for a complete discussion of linearity in flow cytometers. The second method of verifying system linearity is to use beads whose fluorophore loading has been calibrated. Calibrated beads are available from several of the manufacturers listed in Table 1.10.1. Verification of linearity and calibration of the measurement scale should be done according to the directions provided by the manufacturer.
Data Display/Analysis In keeping with the old axiom that a picture is worth a thousand words, examining histograms and other graphical data displays provides considerable information on instrument operating characteristics. Data display and analysis are discussed in depth in Chapter 10. However, they are also relevant to any discussion of instrument alignment as they are ultimately the main tools for assessing instrument operation: the various data displays, along with analysis of the data they contain, are the primary source of quantitative information. For homogeneous particles, univariate histograms of fluorescence intensity and light scatter should each have a symmetric peak with a low CV. A nonsymmetric peak is indicative of misalignment—for example, of the sample stream not being centered in the laser beam focal spot. Measurement statistics such as CV and signal intensity (peak mean value) and symmetry should return to the original setup values when the benchmark particles are reanalyzed in a standard configuration.
1.10.2 Supplement 6
Current Protocols in Cytometry
Table 1.10.1
Sources of Supplies for Flow Cytometer Alignmenta
Type of item
Company
URL
Alignment and calibration particles
Bangs Laboratories Beckman Coulter—U.S. Country Operations (Coulter) Becton Dickinson Immunocytometry Systems Duke Scientific Flow Cytometry Standards Molecular Probes Polysciences Seradyn Particle Technology Spherotech Coherent–Auburn Group, Instruments Division Melles Griot Molectron Detector Newport Melles Griot Photon Spiricon
http://www.bangslabs.com/ http://134.217.3.35/
Laser power meters
Laser beam profile measurement instruments
http://www.bdfacs.com/source_book/ http://www.dukesci.com/ http://www.fcstd.com http://www.probes.com/lit/microscopy/toc.html http://www.polysciences.com/pr/products.html http://www.seradyn.com/particletech.html http://www.spherotech.com/index.htm http://cid.cohr.com http://www.mellesgriot.com http://www.molectron.com http://www.newport.com http://www.mellesgriot.com http://www.photon-inc.com http://www.spiricon.com
aThese lists are illustrative, not complete, and do not represent endorsement by the authors of the suppliers of specific items. For full contact information,
see the SUPPLIERS APPENDIX.
Bivariate displays provide correlated information on two parameters at a time. These displays can be static or live. Live displays are often in the form of dot plots that show the location of the last N events recorded. As an instrument is adjusted, the tradeoff between resolution, intensity, and correlations can be observed on the dot plot displays in near real time.
Other Considerations The settings obtained during alignment should ideally be used when analyzing samples; otherwise results may diverge from expectations. For example, if the particles used for alignment are at a high concentration, and the sample concentration is much lower but the same analysis rate is desired, the obvious response is to increase the sample flow rate. Once the sample flow rate is changed, however, the sample stream diameter changes and the conclusions concerning alignment reached by analysis of the standard particles may no longer be valid. This is especially true of measurement resolution, which usually decreases with increasing sample stream diameter. Another instrument alignment problem is the conflict between optimal resolution for fluorescence measurements and for light-scatter measurements. It may not be possible to optimize both measurements simultaneously.
The operator must decide which parameter or parameters demand the best resolution for a particular set of measurements. If immunofluorescence is being measured along with light scatter, it may be prudent to optimize the lightscatter measurements to enhance discrimination between subpopulations while sacrificing some fluorescence resolution, since immunofluorescence distributions are broad, even on a logarithmic scale. On the other hand, if DNA measurements are to be made, conditions that yield the best fluorescence CV are indicated.
METHODS FOR COMPLETE SYSTEM ALIGNMENT When invasive alignment procedures are pursued, in addition to the techniques and considerations enumerated above, there are additional tools available that provide detailed information about instrument component operation. As stated earlier, in general, the endpoint of alignment procedures is to bring the three major axes of a cytometer—the excitation light path, the detection light path, and the sample stream—into a precise mutual intersection. Figure 1.10.1 illustrates some of the degrees of freedom (DOF) or adjustments possible in a cytometer. Many cytometers, both commercial and home-made, have numerous optical adjustments with varying degrees of freedom. To
Flow Cytometry Instrumentation
1.10.3 Current Protocols in Cytometry
Supplement 6
lens 2 z
laser beam
sample stream
0
x
lens 1
y
Figure 1.10.1 Schematic drawing of a basic flow cytometer. (For clarity, the flow chamber is not shown.) The clusters of three-way arrows indicate 3 translational degrees of freedom. Not indicated are possible rotational degrees of freedom, which can be especially useful for flow chamber alignment.
simplify alignment, most instruments will have constraints on some of these—if everything can be varied, the system will almost invariably slip off by small degrees into oblivion and performance will be degraded. Since there are so many possible adjustment permutations, only a general, simplified example is provided to elucidate the general principles. Usually one or more fixed apertures or restricted DOFs are incorporated into an instrument to constrain the physical system. For example, although in some instruments it is possible to move the entire detector assembly vertically, providing the fluorescence collection lens with two DOFs, in most instruments the lens can be moved only in or out to focus on the sample stream. The laser beams(s) may pass through, or be reflected by, various optical elements (prisms, lenses, or mirrors) but must, in the end, be focused on the sample stream–detector optical axis intersection. Optimal alignment is achieved by two related operations: iteration and walking. In both methods it is extremely helpful to observe the laser beam–cell stream– collection lens intersection point through a microscope. This usually makes it much easier to detect gross misalignment.
Iteration Techniques for Flow Cytometer Alignment
Iteration means adjusting the various optical DOFs—one at a time—while monitoring the raw signals on an oscilloscope for maximum
signal intensity and tightness. For example, after getting the main laser beam to pass through confining apertures and along the optical axis of any in-line lenses, one would vary each degree of freedom in a systematic and thorough manner. Once the laser beam is aligned vertically with the center of the collection lens, the flow chamber should be adjusted to align it such that the sample stream passes through the intersection of the other two axes. The laser beam focus lens is then adjusted for tightest focus of the laser beam on the sample stream, and the collection lens position is adjusted up/down and in/out while constantly monitoring signals from microspheres. Next, these various adjustments should be iterated slowly and carefully until the maximum signal is obtained. (The optimal order of alignment steps may vary with instrument type.)
Walking The second operation, walking, requires more careful and deliberate consideration: it is not enough merely to iterate the various adjustments provided—at least not all of the time. Alignment must be checked periodically to determine if walking of the alignment is necessary. Because all systems gradually go out of alignment or optimal configuration—and they certainly never improve by themselves—one must occasionally slightly deoptimize one DOF, for example the laser focus or the horizontal position of the sample stream, in order
1.10.4 Supplement 6
Current Protocols in Cytometry
to see if adjustment of complimentary or related DOFs will produce better performance (brighter or tighter signals). With practice and patience one will become familiar with the behavior of a particular instrument and learn which adjustments need frequent checking. If a laser beam wavelength is changed, corresponding adjustments of focusing and beamsteering lenses must be made. The focal point of a lens depends on the curvatures of its surfaces and its refractive index. Owing to dispersion effects the refractive index of optical materials changes with wavelength. Thus, for each excitation wavelength there is a different value of the focusing lens’ index of refraction and a correspondingly different focus position. Because of this effect, often referred to as chromatic aberration, whenever the excitation wavelength is changed, the focusing lens position must be adjusted to compensate for the shift in focal position. For example, when changing from 488 nm to 514.5 nm, the focal position shift for a 20-mm focal length lens is ∼40 µm. For changes in laser optics from visible to UV or vice versa, the laser beam path often needs radical realignment. Just remember the basics—that the laser beam should go along the optical axis of beamsteering lenses and through alignment apertures, without obstruction, to the intersection of the sample stream and the detection system optical axis.
Oscilloscope In keeping with the theme that a picture is worth a thousand words, an oscilloscope for monitoring detector pulse shapes is an invaluable tool. By observing amplified detector output pulses on the oscilloscope, one can determine whether the signals have the proper width and shape, indicating that the flow velocities and the laser beam–steering optics are correctly adjusted. Amplified detector pulses that have not been processed (e.g., integrated) should have a symmetric Gaussian shape. Nonsymmetric pulses are indicative of misalignment of the laser beam, sample stream, or detection optics or possibly problems with the laser mode. An oscilloscope can also be used to detect flow problems, sometimes before their effects appear in the collected data. If the scope trace speed is set low enough to see multiple pulses in a single trace, the pulses should appear at random intervals across each trace. Detector pulses that appear in bursts separated by short periods of inactivity can be an indication of
sample delivery problems. In making that judgment, one must be careful that the oscilloscope trigger level is properly adjusted; on some scopes, if the trigger level is too low, occasional bursts of traces can mimic the effect of erratic sample flow.
Laser Beam Analysis The principles of laser beam focusing and the use of lasers in flow cytometry are covered in UNITS 1.6 & 1.9. The first tool to use for analysis of the laser beam characteristics is visual inspection. Here safety is a significant issue. It can not be too strongly emphasized that when working with exposed laser beams one is engaging in procedures that are out of the ordinary, and therefore actions must be carefully planned with safety considerations in mind. Many institutions have a designated Laser Safety Officer, or the equivalent, who can serve as a valuable resource for information about safe operating procedures. This is especially important since in order to inspect or analyze a laser beam path, it is necessary to put something into the laser beam. Just this simple act significantly increases the likelihood of undesired and uncontrolled reflections of the beam. In general, one should attenuate the laser beam power to <1 mW when carrying out visual inspections and alignment. In instruments with low-power lasers, such as air-cooled argon-ion lasers or helium-neon lasers, power levels can be attenuated by inserting one or more neutral-density (ND) filters in the laser beam path. It is best to use reflective glass ND filters to avoid photodamage (hole burned through the filter) that can occur in gelatin ND filters. For higher-power lasers, a series of ND filters in combination with laser safety glasses can be used. Care must be taken to place these filters normal (orthogonal) to the laser beam to prevent refractive displacement of the beam. The desired operating mode of the laser should be a TEM00 mode, which yields a symmetric Gaussian energy distribution (described in detail in UNIT 1.6). A quick way to ascertain whether the laser is operating in TEM00 mode is to reflect the beam off a concave mirror onto a distant wall or the ceiling. This will increase the beam size and provide a general sense of the shape of the intensity profile. A common non-TEM00 mode is the so-called donut or TEM01 mode. By looking at the expanded beam projected onto a wall one can quickly see if there is a hole in the laser intensity, if the beam spot is asymmetric, or if the intensity pattern has the desired symmetry. Figure 1.10.2 illus-
Flow Cytometry Instrumentation
1.10.5 Current Protocols in Cytometry
Supplement 6
Laser intensity (arbitrary units)
1 0.8 0.6 0.4 0.2 0 –200
–150
–100
–50
0
50
100
150
200
Beam radius (arbitrary units)
Figure 1.10.2 Laser beam intensity as a function of distance from the beam center line for the TEM00 mode (solid line) and the TEM01 mode (dashed line).
trates intensity distributions for the two modes. These distributions are also the raw pulse shapes that one would observe on an oscilloscope for small particles (diameter ≤10% of the laser beam diameter) traversing a TEM00 laser beam or a TEM01 laser beam. A laser power meter is an invaluable tool for analyzing laser performance. A variety of laser power meters are available today. The power meter, in addition to being an independent measure of laser power levels, can be used to assess the condition of the laser from time histories of the output power for given tube-current settings. As a tube ages, the ratio of light output to tube current will decrease. A power meter can also be used to quantify the beam shape. By translating a sharp edge (e.g., a razor blade) across the laser beam in small defined steps and recording the measured power, one can not only record the shape of the laser beam but also determine its diameter. The laser intensity profile and beam size can also be measured directly by a laser beam profiler. These devices determine the laser beam intensity as a function of position by directly measuring the intensity. This is accomplished
either by positioning a large-area detector behind a moving slit (or slits) whose position scans back and forth, or by using a positionsensitive detector such as a CCD camera. The camera gives a two-dimensional measurement of the intensity distribution, whereas the first type (with a single moving slit) yields a one-dimensional intensity distribution and integrates the second dimension.
OTHER RESOURCES The techniques and tools for system alignment described above by no means constitute an exhaustive list. One can gain further detailed insights from manufacturer service representatives, from the Purdue cytometry bulletin board (information can be obtained from
[email protected]), and at flow cytometer user group meetings.
Contributed by James H. Jett, John C. Martin, and Robert C. Habbersett Los Alamos National Laboratory Los Alamos, New Mexico
The authors wish to acknowledge the support of the NIH National Center for Research Resources Grant no. RR01315 for the National Flow Cytometry Resource, and of the U.S. Department of Energy.
Techniques for Flow Cytometer Alignment
1.10.6 Supplement 6
Current Protocols in Cytometry
Flow Cytometers for Characterization of Microorganisms Flow cytometry of microorganisms presents several challenges that are not so often encountered in measurement of mammalian cells. Bacteria, probably the most important group of microorganisms with regard to flow cytometry, are typically some three orders of magnitude smaller than mammalian cells with regard to cell volume as well as DNA and protein content. The majority of the flow cytometers currently available were designed primarily to measure mammalian cells. Hence, flow cytometry of bacteria puts particular demands on instrument sensitivity for both light scattering and fluorescence, although the light scattering sensitivity of many instruments is especially insufficient for reliably measuring certain bacteria. The cell cycle of bacteria is radically different from that of mammalian cells in that cells may have anything from one to eight chromosomes depending on the state of growth (as determined by a number of factors, such as the supply of nutrients). Hence, DNA histograms of bacteria may yield a lot of additional information compared to what one can learn from corresponding data for mammalian cells. However, sufficiently specific staining of bacterial DNA may be hampered by the relatively large amount of bacterial RNA, especially under favorable growth conditions. As treatment of bacteria with RNase is not as straightforward as with mammalian cells, possibly due to restricted permeation of the cell wall by that enzyme, highly DNA-specific dyes may be required to obtain good results. This limits the selection of dyes and thereby puts demands on the excitation wavelengths that must be available. Dyes that have proved sufficiently DNA specific include DAPI, Hoechst 33342, and Hoechst 33258, all of which are excited in the UV range (∼350 to 370 nm; Bernander et al., 1998), and Mithramycin, which is excited at ∼430 nm (Steen et al., 1994). Hence, instruments that do not provide at least one of these wavelengths are not well suited for measurement of bacterial DNA content. In many types of studies it is desirable to relate the presence of some protein (measured by the use of a fluorescence-conjugated antibody) with the cell cycle and growth state (measured with a DNA-specific dye). With mammalian cells this is typically done by comContributed by Harald B. Steen Current Protocols in Cytometry (1999) 1.11.1-1.11.9 Copyright © 1999 by John Wiley & Sons, Inc.
UNIT 1.11
bining a fluorescently conjugated antibody with a DNA-specific dye that has its fluorescence at a higher wavelength than the antibody conjugate, in order to reduce the problem of spectral overlap. For example, a FITC-conjugated antibody might be used with propidium iodide (PI) to stain DNA (subsequent to treatment with RNase and/or removal of the cytoplasm). However, because DNA-specific dyes excited in the UV or deep blue must be used with bacteria, this option is not available, and often the weak antibody fluorescence can “drown” in the long-wavelength tail of the fluorescence spectrum of the very much stronger DNA-associated fluorescence. The best solution to this problem is a dual-laser instrument. Many species of bacteria are rod-shaped, and many other microorganisms do not even have axial symmetry. Nonspherical cells may cause orientation artifacts, where the fluorescence and light scattering signals depend on the orientation of the cell relative to the direction of the excitation light. This is especially problematic in instruments where the cells pass through a near-parallel beam of excitation light, as is the case in most laser-based flow cytometers (Pinkel and Stovel, 1985). Finally, the permeability of the bacterial cell wall is generally less than that of mammalian cell membranes, again limiting the selection of dyes that may be appropriate for certain applications. Gram-positive cells have an especially complex cell wall. The resulting problems are enhanced by the fact that some bacteria have highly efficient efflux pumps, so that even dyes that are able to permeate the cell wall cannot be used to stain viable cells. On the other hand, every problem related to the nature of the cell is also an opportunity to gain more insight.
FLOW CYTOMETER DESIGNS Laser-Based Instruments There are two basic designs on the market, employing lasers and arc lamps, respectively (see UNIT 1.1. for an overview of the principles of flow cytometry). Instruments with laser excitation exhibit an orthogonal configuration in which the laser beam, the fluid stream carrying the sample, and the optical axis of the fluorescence detection optics are all at right angles to
Flow Cytometry Instrumentation
1.11.1 Supplement 7
one another (Fig. 1.1.4). Forward scattering is detected immediately outside the laser beam, which is prevented from reaching the optics by means of a beam dump, while 90° scattering is detected through the fluorescence optics, and separated from the fluorescence by means of a dichroic mirror and a suitable optical band-pass filter. The fluorescence may be split into two or more spectral components by dichroic mirrors and optical filters for simultaneous measurement of a corresponding number of different chromophores. A photodiode may be used for the forward scatter detection, while the other detectors are usually photomultiplier tubes (PMTs). The laser is a monochromatic light source with sufficient intensity and stability for flow cytometry applications (i.e., power between 10 mW and a few watts, constant to within 1%). However, only the relatively large and expensive water-cooled lasers can be tuned to different wavelengths, and the set of wavelengths that can be obtained with each of this type of laser is limited (see UNIT 1.9 for a discussion of lasers used in flow cytometry). Furthermore, the orthogonal configuration used in most laserbased flow cytometers requires highly precise alignment of the three orthogonal axes of the system, making it susceptible to mechanical shock and vibration, which is a relevant consideration in instruments intended for field applications (see UNIT 1.10 for a discussion of flow cytometer alignment considerations). The alignment of such instruments, in particular some of the large water-cooled ones, may be quite tricky and time consuming. In instruments that are located in a laboratory, the reason for misalignment is nearly always malfunction of the nozzle or flow cell, resulting in a skewed sample flow that thereby misses the center of the excitation focus. Jet-in-air nozzles are particularly prone to this problem. Hence, indications of misalignment, which should always be confirmed by running a sample of test beads, should be attacked first by checking and cleaning the nozzle/flow chamber, and then by realigning it if necessary. Attempts to adjust the laser beam usually make things worse, and should be carried out only as a last resort. Adjustment of the detection optics is only for qualified service personnel.
Microscope-Based Instruments Flow Cytometers for Characterization of Microorganisms
The other main type of flow cytometer is the microscope-based instrument. These instruments employ epiillumination through an oilimmersion microscope lens in a configuration
similar to that of epiillumination fluorescence microscopes (Fig. 1.11.1; Steen, 1990). This implies that excitation and fluorescence detection are carried out through the same lens and therefore are inherently confocal. This feature greatly simplifies alignment and makes such designs less susceptible to mechanical shock and vibration. The use of immersion optics optimizes excitation intensity as well as fluorescence detection efficiency. Most microscope-based instruments employ high-pressure arc lamps containing mercury (Hg) or xenon (Xe) gases or a combination of the two, although the optical configuration of such instruments is readily adaptable to lasers as well. High-pressure arc lamps are essentially white light sources that can provide a wide range of excitation wavelengths by the use of appropriate band filters (see UNIT 1.5 for a discussion on selection of optical filters for flow cytometry). The excitation intensities that can be achieved at the major emission lines of mercury (365, 436, 546, and 578 nm) with a 100-W Hg or Xe-Hg lamp are 20 to 30 mW. In contrast, at the optimal excitation wavelength for FITC (470 to 490 nm), the intensity is only 2 to 3 mW (Steen, 1990). It may be tempting to try to increase excitation intensity by the use of a larger lamp (e.g., 500 or 1000 W). However, according to the laws of optics, the light density in the focus cannot exceed that of the source itself (i.e., the arc of the lamp), and the arc density of a 100-W lamp is as high as or higher than the density of the larger ones. In addition, a larger lamp would cause severe problems with heating and degradation of optics and filters. The effective intensity of arc lamps may be increased substantially (∼50-fold) by pulsing the lamp, but that requires significant modification to the electronics of the instrument (Steen and Sorensen, 1993). A significant disadvantage of arc lamps is their relatively short lifetime, ranging from 200 hr for Hg lamps to 2000 hr for some Xe lamps. Moreover, during its lifetime the intensity of the lamp declines by up to 50%. In addition, some lamps, especially some mercury lamps, have a tendency to instability caused by spurious arc movements. The effect of such flicker on the excitation intensity depends on details of the optical configuration, but is usually limited to a few percent. These problems can be reduced significantly by the use of some Xe-Hg lamps (e.g., Hamamatsu), which have a rated lifetime of 1000 hr and exhibit negligible flicker.
1.11.2 Supplement 7
Current Protocols in Cytometry
Light-scattering measurement in such instruments may be achieved by means of a darkfield configuration, as shown in Figure 1.11.1. This configuration facilitates separate detection of forward and large-angle light scattering (Steen, 1986). This type of optical configuration provides a sensitivity that is easily sufficient for precise measurement of even the smallest bacteria. Thus, it facilitates measurement of polystyrene particles to well below 0.2
µm (compared to ∼0.5 µm with laser-based systems), representing a ∼40-fold difference in the signal-to-noise ratio in light-scattering detection. Another advantage of the microscope-based flow cytometer is that the use of a high–numerical aperture (NA) lens for both the illumination and the collection of fluorescence essentially eliminates orientation artifacts associated with nonspherical cells.
A dichroic mirrors
ocular
dichroic long-pass mirror filter
PMT (red)
microscope objective (oil immersion, NA 1.3)
focus
band-pass filters PMT (yellow)
PMT (green)
measuring slit retractable mirror
band-pass excitation filter excitation slit
object plane
heat filter condensor lens arc lamp spherical mirror
B fluorescence microscope objective (oil immersion, NA 1.3)
illumination field dark field
45˚mirror
ocular telescope
low-angle detector
PMT
focus field stop incident light
measuring slit microscope objective (long working distance, NA 0.4)
short-pass filter retractable mirror
PMT
collection lenses large-angle detector
Figure 1.11.1 Optical outline of a microscope-based flow cytometer. (A) The epifluorescence side of the instrument. The excitation and measuring slits are essential for reducing optical background. The sample flow may be viewed through the ocular to facilitate the positioning of the flow cell (not shown) in the excitation focus. (B) The optical system for measurement of light scattering. A field stop inside the microscope objective gives rise to a shadow (i.e., a dark field), inside which is situated a microscope lens that forms an image of the sample flow on the measuring slit. The telescope behind the slit facilitates separate detection of the low-angle and large-angle components of the scattered light.
Flow Cytometry Instrumentation
1.11.3 Current Protocols in Cytometry
Supplement 7
Flow Cells
Flow Cytometers for Characterization of Microorganisms
The flow cell consists of a conical nozzle that either produces a jet in air, with a typical diameter between 50 and 100 µm, or leads into a rectangular tube (i.e., a closed flow cell), with a cross section of, for example, 200 × 300 µm. The jet-in-air flow cell is preferable for fast and reliable sorting of cells, but otherwise it has several significant disadvantages. The cylindrical surface of the jet gives rise to an enormous intensity of reflected light in the horizontal plane, which makes it necessary to shield the detection apertures by obscuration bars made of metal or some other opaque material. These bars cause additional scattered light in the region around the focus and may thereby increase the background level of the light scattering detectors. Furthermore, they reduce the effective aperture of the detection optics. Since the jet in air is not compatible with immersion optics, the numerical aperture of the detection optics is in practice limited to ∼0.6, and the effective aperture is reduced further by the obscuration bar, with the result that <5% of the fluorescence is collected by the optics. Finally, the jet in air sets a lower limit to the flow velocity, which in turn limits the increase in sensitivity that can be obtained by reducing flow velocity (see below). Thus, a nozzle diameter of 100 µm can produce a jet only when the velocity exceeds ∼3 m/sec (Kachel et al., 1990). The flow from a nozzle in air is much more susceptible to disturbance from impurities in the orifice than are closed flow cells. Because the orifice is usually smaller than the cross section of closed flow cells, it is more prone to clogging, especially when running cells with a tendency to aggregate, such as some strains of Streptococcus. The main disadvantages of the jet-in-air flow cell are avoided with closed flow cells. These devices facilitate the use of immersion detection optics with NA ∼1.2 and do not require obscuration bars to shield the detection system from scattered excitation light. Since light collection efficiency is proportional to NA2, a closed flow cell may facilitate a fluorescence collection efficiency of ∼20%. Closed flow cells also allow large reductions of the flow velocity (e.g., down to a few cm/sec). One type of microscope-based flow cytometer employs a jet-on-open-surface flow cell (Steen and Lindmo, 1979). This flow cell can be used with oil-immersion optics, exhibits very low inherent light scattering, and thereby facilitates highly sensitive light-scattering measurement.
The intensity profile of the excitation beam at focus is approximately Gaussian (bellshaped) in both laser- and microscope-based instruments. This implies that for all cells to be exposed to the same intensity they have to pass close to the center of the focus. For example, with a focus width (measured between the 1/e2 points of beam intensity) of 100 µm, the cells have to pass within ∼7 µm from the center of the focus in order to be exposed to the same intensity to within 1%, and within 10 µm to be within 2%. In order to steer the cells through the focus with such precision, flow cytometers utilize the principle of hydrodynamic focusing (see UNIT 1.2 for a discussion of the fluidics of flow cytometers). In practice, this is achieved by the use of a conical nozzle ending in an orifice that leads either to the open air (to produce a cylindrical jet) or to a tube (i.e., a closed flow cell). The cell suspension is introduced through a tube that ends in the center of the flow of sheath fluid in the conical part of the nozzle. Because the flow is laminar, the sample will remain confined to the central core of the flow as it converges toward the orifice. The diameter of this sample core, d, is the precision of the cell path through the focus, and is given by: d=D
w W
=2
w πv
where D is the diameter of the total flow, v is flow velocity, and w and W are the flow rates of the sample and total flow, respectively. A typical value of W is 10 ml/min. If the flow diameter is 200 µm (closed flow cell), a sample flow (w) of 10 µl/min gives ≈ 6 µm, which is the precision required to obtain 1% reproducibility in exposure in a 100-µm focus (see above). Note that this reproducibility applies only to cells much smaller than 6 µm. The equivalent of a cell radius must be added to the calculated value of d in order to achieve a realistic estimate. As seen from the above equation, decreasing the flow velocity in order to increase sensitivity (see below) increases the sample stream diameter, thus reducing the precision of the sample flow and hence the precision of measurement. On the other hand, reduced measuring precision is quite acceptable in many cases where biological variation is large anyway. In fact, the only biological application that requires a measuring precision of a few percent is measurement of cellular DNA content.
1.11.4 Supplement 7
Current Protocols in Cytometry
Sample Injection Two different types of systems are currently in use for injecting sample into the flow cell. In the pressure differential system, the sample flow is controlled by the pressure difference between the tube containing the sample and the sheath fluid container by means of a pressure regulator. This system has the advantage that it can be flushed out rapidly and efficiently so as to minimize carryover. Its disadvantage is that it does not facilitate a calibrated flow rate. Hence, the cell density of the sample can only be determined by indirect methods. The volumetric injection system employs a syringe to inject the sample at a calibrated rate under the control of the instrument computer, and thereby facilitates direct determination of cell density. Because it includes a valve in the sample line, it may require more flushing, especially with sticky samples.
SENSITIVITY AND RESOLUTION As noted above, sensitivity is a major concern in flow cytometry of bacteria. The sensitivity is limited primarily by optical noise. With adequate electronics, electronic noise is insignificant in comparison. The dark current of the PMTs is also of no concern in this connection. Optical noise is a basic phenomenon of light emission. It arises from the fact that emission of photons and emission of photoelectrons (released by the light signal in the light detectors) are stochastic processes. This implies that if identical particles pass through the same excitation focus, the number of photons emitted will vary from cell to cell with a coefficient of variation (i.e., relative standard deviation) equal to the square root of the average number of photons emitted. Hence, any light signal exhibits an inherent fluctuation that effectively constitutes noise. Photons give rise to a signal in the photodetector by causing emission of electrons from the photocathode of the detector (PMT). The emission of these photoelectrons is also a stochastic process and results in a corresponding variation in the electric signal induced by the light signal. In the flow cytometer, the signals from cells are superimposed on a constant background of light from several other sources: (1) light, especially scattered excitation light, leaking through barrier filters; (2) fluorescence from sources other than the cells, such as dyes dissolved in the sample fluid (e.g., by measurement of DNA) and cellular debris sticking to the inner surface of the flow cell; (3) Raman scattered light from the sheath
fluid, which is red-shifted relative to the excitation light, very much like fluorescence; (4) fluorescence and scattered light from optical components of the system; and (5) ambient light. It can be shown (Steen, 1992) that the sensitivity, expressed in terms of the lowest number of fluorescent molecules that can be detected, fL, is given by:
sensitivity =
1 fL
≈
aΦ e ix b
where b is the equivalent background (the number of fluorescent molecules that would be required to produce an intensity equal to that of the background), a is a constant, Φe is the photoelectron quantum efficiency of the detector (the number of photoelectrons released per photon reaching the detector; typically Φe ∼0.2), and ix is the excitation intensity (i.e., the number of photons per cm2 per sec in the excitation focus). Note that the lowest possible background is as important as the intensity of the excitation light. The constant, a, reflects the characteristics of the fluorochrome and the detection optics of the flow cytometer: a=
NA 2 T v
1
ε( λ ) I x (λ )d λ
where NA is the numerical aperture of the detection lens, T is the transmission of the optical pathway between that lens and the detector, v is the flow velocity, and the integral is the overlap between the absorption spectrum, ε(λ), of the dye and the spectrum of the excitation light, Ix(λ). The resolution of the flow cytometer, usually expressed as the relative standard variation of the measurement related to instrumental factors (CVp), is also limited primarily by photon noise:
CVp =
f +b aΦ e ix f 2
where f is the number of fluorescent molecules in the cell that yields the signal. The coefficient of variation measured for a given sample, CVm, results from the variation associated with the
Flow Cytometry Instrumentation
1.11.5 Current Protocols in Cytometry
Supplement 7
instrument, CVp, and the variation within the sample, CVs, according to the laws of statistics:
CVm2 = CVp2 + CVs2 Combining these two equations yields:
CVm2 =
f +b aΦ e ix f 2
+ CVs2
Essentially the same equations governing sensitivity and resolution apply to the sensitivity of light-scattering detection, except that f is replaced by a number that is a function of the parameters determining light-scattering characteristics of the cell, notably size and refractive index.
LIGHT-SCATTERING DETECTION The light scattering of microscopic particles is a complex phenomenon, so much so that the theory describing it, called Mie theory, cannot be used to obtain an exact, quantitative description of the light scattering of such intricate objects as biological cells. If, however, the cell is approximated by a spherical bag of liquid containing smaller objects with a well-defined
size distribution, the general features of the light scattering can be obtained, as shown in Figure 1.11.2. The absolute intensities increase with the refractive index of the contents of the cell, but the general features of the curves in Figure 1.11.2 remain the same. It may be seen that the smaller cell has relatively more light scattering at higher scattering angles. The intensity of the small particles relative to the large ones is shown as a function of scattering angle in Figure 1.11.3. Hence, when a cell is approximated by this simple model, it is seen that the ratio of light scattering at high scattering angles, LS2, and low scattering angles, LS1, will increase with the number of smaller objects. This fact is exploited to distinguish cells on the basis of granularity. For example, the value of LS2/LS1 is significantly higher for granulocytes than for lymphocytes. From Figure 1.11.3 it can be seen that this ratio remains approximately constant upwards from ∼15°, whereas the intensity falls off substantially (by ∼3 orders of magnitude) between 20° and 90° (Fig. 1.11.2). Hence, to obtain optimal discrimination of cells on the basis of granularity, LS2/LS1 can be measured at any angle above 15°. However, in order to obtain maximum signal, and thereby optimal sensitivity, LS1 should be measured at the smallest possible
Scattering intensity (relative scale)
100 10 –1 10 – 2 granula (1.2 µm)
10 – 3 10 – 4 10 – 5
contour (12 µm)
10 – 6 10 –7 10 – 8 0
20
40
60
80 100 120 140 160 180
Scattering angle (degrees)
Flow Cytometers for Characterization of Microorganisms
Figure 1.11.2 Light-scattering intensity versus scattering angle calculated from Mie theory for 12 µm spheres (contour) and for a size distribution of smaller particles (granula) with an average diameter of 1.2 µm. The refractive index of the large and small particles is 1.37 and 1.39, respectively. The curves are normalized to unity at zero scattering angle. The fine structure of the lower curve represents the diffraction pattern of the 12-µm particles.
1.11.6 Supplement 7
Current Protocols in Cytometry
angle (upwards from 1° or 2°), whereas LS2 should be measured around 20° rather than around 90°. Note that since the intensity of light scattering falls off quite rapidly with increasing scattering angle (Fig. 1.11.2), the upper limit of the range of scattering angles included in LS1 and LS2 is not important.
PRACTICAL CONSIDERATIONS Excitation Intensity Both sensitivity and resolution increase with the square root of the excitation intensity, ix, that is, with the square root of the number of photons absorbed in the dye from which the signal arises (see Sensitivity and Resolution, equations). The number of emitted photons is also proportional to the time the cell spends in the focus, which in turn is inversely proportional to flow velocity. Hence, rather than increasing excitation intensity, one can reduce flow velocity. Note, however, that a significant reduction of flow velocity gives a corresponding increase of the pulse length from the light detectors, and may require a modification of the time constant of the electronic signal amplifiers in order to achieve the full benefit of the reduction. The improvement in fluorescence sensitivity that can be achieved by reducing flow velocity is limited by bleaching to the same extent as by increasing the excitation intensity, while ground state depletion will not
increase. All of this discussion applies to light scattering as well. The excitation intensity may also be increased by concentrating the light into a smaller focus. However, this can only be done at the expense of measurement precision and reliability. Moreover, a narrower focus makes alignment more critical, and the precision of measurement cannot be maintained, especially with cells whose size exceeds a few percent of the width of the excitation focus.
Dye Saturation With fluorescence, there is a limit to the sensitivity one can achieve by increasing excitation intensity or decreasing flow velocity. This limitation is due to the photochemical degradation of the dye, called bleaching. For example, a FITC molecule passing with a velocity of 4 m/sec through a 488-nm Gaussian laser focus having a power of 1 W and a width of 100 µm will absorb ∼140 photons. Every time the molecule is excited there is a certain probability (quantum yield) that it will dissociate or react with some other molecule (oxygen in particular), and thereby lose its ability to emit fluorescence. For some molecules, like many porphyrins, that quantum yield approaches unity, which is to say that these molecules will not survive passage through a 1-W focus. Other chromophores may be quite stable and survive thousands of excitations.
Scattering intensity differential
104
103
102
101
100 0
30
60
90
120
150
180
Scattering angle (degrees)
Figure 1.11.3 The ratio between the curves in Figure 1.11.2 after smoothing of the diffraction fine structure.
Flow Cytometry Instrumentation
1.11.7 Current Protocols in Cytometry
Supplement 7
In addition to bleaching, high excitation intensities may also cause ground-state depletion of some dyes; that is, the high intensities keep a significant fraction of the dye molecules in an excited state at the same time. Thus, a large part of the excitation light at such intensities may be wasted, with the result that sensitivity and resolution will no longer improve with increasing excitation intensity according to the above equations. In fact, it presently becomes worse than that. Because the background will usually not be saturated, it continues to increase with ix, while the fluorescence levels off. The result is that, above a certain limit of excitation intensity, sensitivity and resolution will actually decrease. That limit depends on the dye and the flow velocity, but in general one should be cautious with laser powers >100 mW.
Background
Flow Cytometers for Characterization of Microorganisms
Sensitivity and resolution can also be improved by reducing the background. As seen from the equation for sensitivity, the improvement obtained by reducing background by a given factor is the same as that obtained by increasing the excitation intensity (e.g., by installing a larger laser). The background may often be reduced significantly at a quite moderate cost, if any cost at all. In some cases it may be as simple as reducing the ambient light that reaches the measuring region of the instrument. The excitation light source may emit some radiation in addition to the wavelength(s) used for excitation. Some lasers have a faint glow, which may contribute to the background. Arc lamps emit large intensities of infrared light, which may be picked up by PMTs. Hence, background may be reduced by installing better filters. In particular it is important to check that excitation light is not leaking through fluorescence emission filters. Infrared light (i.e., heat radiation) leaks through most filters, including the interference filters used most commonly in flow cytometers. Hence, in arc lamp–based instruments the background may be reduced significantly by introducing an efficient heat filter (for example, a short-pass filter with a cutoff at ∼700 nm). Filter quality can be checked by using two of the same kind of filter in order to detect possible transmission of light outside the transmission wavelength range of the filter. For example, filters may have secondary transmission bands far outside this region. This doubling of filters is bound to reduce the signal, but
at the same time it may improve the performance of the instrument with regard to resolution and sensitivity (see Checking Instrument Performance). If it does, one should try to find a better filter. Even the best microscope lenses exhibit some fluorescence. Background due to fluorescence from optical components will typically extend over the entire spectrum upwards from the excitation wavelength. Hence, in some cases a better signal-to-noise ratio may be obtained by the use of emission filters with a narrower transmission band. In general, band filters should be used rather than cutoff filters, especially for fluorescence detection.
Spectral Overlap Fluorescence sensitivity and resolution depend not only on instrument characteristics, but also on the dye and excitation wavelength. Therefore, sensitivity and resolution must be measured or specified separately for every dye and excitation wavelength. The specific combination of dye and excitation wavelength affects not only the spectral overlap between excitation wavelength and dye absorption spectrum, but also the background, generally so that background increases with decreasing wavelength.
Checking Instrument Performance According to the above equations, sensitivity and resolution depend on the same instrument parameters, and are therefore closely related. Hence, the CV measured for a given sample (CVm) may also be a good indication of the sensitivity. The instrument CV (CVp) may be determined quite simply by running a sample of monodisperse particles with an inherently low CVs and with relatively low fluorescence and/or light scattering—low enough that CVm is limited primarily by photon statistics (i.e., by CVp). If the measured CVm matches that obtained when the instrument is at its very best, it is a very strong indication that other performance parameters are good. If not, and the instrument otherwise appears to be working properly, the most probable cause is that the background is higher than normal, and one should check the various sources of background (see above). How can one know if CVm is limited by photon statistics? If the CVm is measured at a couple of different excitation intensities and the results are plotted as CVm2 versus 1/ix (on a relative scale), the data should fall on a straight line that extrapolates to the value of CVs.
1.11.8 Supplement 7
Current Protocols in Cytometry
Hence, the instrument resolution, CVp, can be calculated from the equation for CVm2. The excitation intensity can be varied by turning the appropriate knob on the power supply of the light source, or, if that is not possible (e.g., with lasers running at a fixed power or with high pressure arc lamps), by introducing neutral density filters in the excitation light path. This method may be used to determine the true variation of any sample, i.e., the value of CVs. For example, it can be used to determine the extent to which the width of a DNA histogram peak represents true variation within the sample or just reflects the sensitivity limit of the instrument.
LITERATURE CITED Bernander, R., Stokke, T., and Boye, E. 1998. Flow cytometry of bacterial cells: Comparison between different flow cytometers and different DNA stains. Cytometry 31:29-36. Kachel, V., Fellner-Feldegg, H., and Menke, E. 1990 Hydrodynamic properties of flow cytometric instruments. In Flow Cytometry and Sorting (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 27-44. Wiley-Liss, New York. Pinkel, D. and Stovel, R. 1985. Flow chambers and sample handling. In Flow Cytometry: Instrumentation and Data Handling (M.A. Van Dilla, P.N. Dean, O.D. Laerum, and M.R. Melamed, eds.) pp. 77-128. Academic Press, London.
Steen, H.B. 1986. Simultaneous separate detection of low angle and large angle light scattering in an arc lamp-based flow cytometer. Cytometry 7:445-449. Steen, H.B. 1990. Characteristics of flow cytometers. In Flow Cytometry and Sorting (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 11-25. Wiley-Liss, New York. Steen, H.B. 1992. Noise, sensitivity, and resolution of flow cytometers. Cytometry 13:822-830. Steen, H.B. and Lindmo, T. 1979. Flow cytometry: A high resolution instrument for everyone. Science 204:403-404. Steen, H.B. and Sorensen, O.I. 1993. Pulse modulation of the excitation light source boosts the sensitivity of an arc lamp-based flow cytometer. Cytometry 14:115-122. Steen, H.B., Jernes, M.W., Skarstad, K., and Boye, E. 1994. Staining and measurement of DNA in bacteria. In Methods in Cell Biology, Vol. 42, Part B (Z. Darzynkiewicz, J.P. Robinson, and H.A. Crissman, eds.) pp. 477-487. Academic Press, San Diego.
Contributed by Harald B. Steen Norsk Hydro Institute for Cancer Research Oslo, Norway
Flow Cytometry Instrumentation
1.11.9 Current Protocols in Cytometry
Supplement 7
Principles of Resonance Energy Transfer Fluorescence methods are widely used to investigate molecular details of various biochemical and biological processes because of the inherent sensitivity, specificity, and temporal resolution of fluorescence spectroscopy. The combination of fluorescence spectroscopy with flow and image cytometry has provided a solid basis for rapid and continuous development in these technologies. In order to utilize these techniques properly, cytometrists must be familiar with the working principles of the instruments and also with the basic concepts of fluorescence spectroscopy. This unit focuses on a special phenomenon of fluorescence spectroscopy, namely fluorescence resonance energy transfer (FRET). FRET is a radiationless process in which energy is transferred from an excited donor to an acceptor molecule under favorable spectral and orientational conditions. These conditions will be discussed in detail below. FRET processes during fluorescence measurements in flow and image cytometry can either compromise results or open new applications for these techniques. In order to distinguish between the adverse and beneficial effects of FRET, one must understand the theoretical background of the phenomenon. When multiple fluorescent probes are simultaneously applied, the possible cross-talk between fluorescent dyes (e.g., FRET processes) should be ruled out, or controlled if one wants to quantitate the cell-surface expression of various antigens at the same time. In contrast to this adverse effect, FRET can also be used to improve the spectral characteristics of fluorescent dyes and dye combinations, such as the tandem dyes in flow and image cytometry and FRET primers in DNA sequencing and the polymerase chain reaction. The driving force in these applications is the use of single-wavelength excitation while providing various dye combinations with a wide range of Stokes shifts to make possible the simultaneous detection of three or four fluorescent dyes. Combination of FRET with monoclonal antibodies has led to a boom in structural analysis of proteins in solutions and also in biological membranes. Analysis based on functional heterogeneity of leukocytes is accompanied by analysis based on specific expression of various cell-surface antigens. International workshops assign a “cluster of differentiation” (CD) nomenclature to these antigens, based on reactivity with groups of
monoclonal antibodies. Cell-surface mapping of CD molecules on immunocompetent cells has attracted more and more interest in the last two decades. With the help of FRET, molecular dimensions can be measured and determined in functioning, living cells, providing information that would be impossible to obtain with other classical approaches—e.g., with X-ray-crystallographic methods. This unit describes the theory behind FRET, characterizes available parameters and instruments, discusses limitations, and provides a few examples of the application of FRET.
THEORY OF FRET FRET was first observed by Perrin at the beginning of the century, but it was Theodor Förster who proposed a theory describing longrange dipole-dipole interactions between fluorescent molecules more than 50 years ago (Förster, 1946, 1948). He derived an equation that relates FRET efficiency to the spectroscopic parameters of fluorescent dyes. His ingenious discovery that fluorescence dipole-dipole interaction depends, in addition to orientation and other spectroscopic parameters, on the negative sixth power of the distance between the dipoles furnished one of the most sensitive methods for measuring atomic and molecular distances at the nanometer level. After the theoretical background of the FRET process was illuminated, it took decades before FRET technology gained wide application in chemistry, biochemistry, and cell biology. FRET is a physical process in which energy is transferred from an excited donor molecule to an acceptor molecule by means of intermolecular long-range dipole-dipole coupling. One of the most important factors influencing the strength of coupling is the distance between the donor and acceptor dye molecules. Energy transfer occurs in the 2- to 10-nm distance range with measurable efficiency, and these distances correlate well with macromolecular dimensions (Stryer, 1978). Energy transfer is nonradiative—i.e., the donor does not actually emit a photon and the acceptor does not absorb a photon. The so-called “trivial” radiative energy transfer has very low probability at low concentrations (<10−6 M) of the fluorescent probes. In order to explain the mechanism of fluorescence resonance energy transfer, let us consider a system with two different fluorophores
Contributed by János Szöll§si, Sándor Damjanovich, and László Mátyus Current Protocols in Cytometry (1999) 1.12.1-1.12.13 Copyright © 1999 by John Wiley & Sons, Inc.
UNIT 1.12
Flow Cytometry Instrumentation
1.12.1 Supplement 9
where the molecule with higher energy absorption is defined as the donor (D) and the one with lower energy absorption as the acceptor (A). If the donor is in an excited state, it will lose energy by internal conversion without emission, until it reaches the ground vibrational level of the first excited state (Kasha’s rule; see Fig. 1.12.1). If the donor emission energies overlap with the acceptor absorption energies, the following resonance can occur through weak coupling: D* + A
Û D + A*
The efficiency (E) of FRET is a quantitative measure of the number of quanta that are transferred from the donor to the acceptor and can be expressed as: E=
no. quanta transf. from donor to acceptor no. quanta abs. by donor Equation 1.12.2
According to the theory of Förster, the rate (kT) and efficiency (E) of energy transfer can also be written as: k T = const k F Jn −4 R −6 κ 2
Equation 1.12.1 Equation 1.12.3
where D and A denote the donor and the acceptor molecules in ground state, while D* and A* denote the first excited states of the fluorophores. The rate of the forward process is kT, while the rate of the inverse process is k–T. Since vibrational relaxation converts the excited acceptor to the ground vibrational level, the inverse process is highly unlikely to occur (Fig. 1.12.1). As a result, the donor molecules become quenched, while the acceptor molecules become excited and, under favorable conditions, can emit fluorescent light with their own quantum yield. This latter process is called sensitized emission (Fig. 1.12.2).
E=
kT kT + k F + k D
Equation 1.12.4
where kF is the rate constant of fluorescence emission of the donor and kD is the sum of the rate constants of all other deexcitation processes of the donor. R is the separation distance between the donor and acceptor molecules, and κ2 is an orientation factor which is a function of the relative orientation of the donor’s emis-
D* A*
KD
KT
KF
D
Principles of Resonance Energy Transfer
KFA
A
Figure 1.12.1 The energy balance of the FRET phenomenon. The donor molecule is excited to a higher vibrational level of the first excited state, from which it decays to the lowest vibrational level according to Kasha’s rule. From this state it can relax to ground state through fluorescence or internal conversion, or via energy transfer (KT arrow). Only those transitions that have the matching pair in the energy diagram of the acceptor take part in this process. The acceptor decays to the ground state through similar mechanisms.
1.12.2 Supplement 9
Current Protocols in Cytometry
sion dipole and the acceptor’s absorption dipole in space. Other parameters are n, the refractive index of the medium, and J, the spectral overlap integral, which is proportional to the overlap in the emission spectrum of the donor and the absorption spectrum of the acceptor:
J=
I
FD ( λ )ε A ( λ )λ4 dλ
I
FD ( λ )dλ
Equation 1.12.5
where FD(λ) is the fluorescence intensity of the donor at wavelength λ and εA(λ) is the molar extinction coefficient of the acceptor. For dipole-dipole vectorial interaction, the transition dipoles of the donor and the acceptor in space must be oriented favorably relative to each other as given by the following equation: κ = (cosα − 3cosβcosγ ) 2
the centers of the fluorophores and the transition moments of the donor and acceptor, respectively (Fig. 1.12.3). From theoretical considerations, κ2 is in the range between 0 and 4. Uncertainties in the value of κ2 cause the greatest error in distance determination by energy transfer. Fortunately R depends on (κ2)1/6, so that it changes only slightly over a wide range (e.g., 0.3 to 3) of κ2. Direct measurement of the value of κ is impossible; however, fluorescence anisotropy measurements on donor and acceptor molecules can be performed to limit possible values of the factor, but rarely do they eliminate all of the uncertainty (Dale, 1979). In addition, if the donor or the acceptor or both have a certain degree of rapid rotational freedom, κ2 becomes 2/3, due the random movement and orientation of the donor and the acceptor. This condition is usually satisfied for fluorophores attached to biomolecules at the cell surface (Dale, 1979). It can be shown that: E=
Equation 1.12.6
where α is the angle between the transition moments of the donor and the acceptor, and β and γ are the angles between the line joining
R −6 + R0−6
Equation 1.12.7
Aex
Aem
Fluorescence intensity
Dex Dem
R −6
400
450
500
550
600
650
700
Wavelength (nm)
Figure 1.12.2 Fluorescence excitation and emission spectra of a suitable FRET pair: fluorescein (donor) and tetramethylrhodamine (acceptor). The shaded area represents the overlap integral (J). The spectra are normalized for display purposes. The downward-pointing arrow indicates the quenching of the donor, whereas the upward-pointing arrow shows the sensitized emission of the acceptor. The amount of quenching and sensitized emission is distorted for demonstrational purposes.
Flow Cytometry Instrumentation
1.12.3 Current Protocols in Cytometry
Supplement 9
From Equations 1.12.3, 1.12.4, and 1.12.7 it follows that: kT =
τ R
1 R0
6
Equation 1.12.8
where τ is the donor’s lifetime in the absence of the acceptor, and R0 is the characteristic distance between the donor and the acceptor when the transfer efficiency is 50%. Also consider the equation: R0 = const ( Jκ 2 QD n −4 )
1
6
Equation 1.12.9
In this equation QD is the quantum efficiency of the donor in the absence of the acceptor. To observe effective transfer in the 2- to 10-nm range, the fluorescence emission spectrum of the donor and the absorption spectrum of the acceptor should overlap adequately, and both the quantum yield of the donor (QD) and absorption coefficient of the acceptor (εA) should be sufficiently high (QD ≥ 0.1 and εA ≥ 1000 M-1cm-1).
DETERMINATION OF FÖRSTER DISTANCE (R0) FRET efficiency measurement is most sensitive to distance variation when the separation of the donor and acceptor is close to the R0 (Förster) distance. Thus, when choosing a do-
nor-acceptor pair, the molecular dimensions of the system to be studied should be considered. Since there is no internal distance reference in fluorescence resonance energy transfer, all distances calculated from transfer efficiencies are relative to an R0 distance evaluated from spectroscopic properties of donor and acceptor. R0 thus provides a reference ruler in distance measurements. Even to estimate the adverse effect of FRET, the R0 value of the donor-acceptor pair in question must still be determined in order to determine the effective range of the FRET process. Due to error propagation, it is almost impossible to measure FRET efficiency of <5% accurately, so that the maximal distance at which FRET can be measured is 1.63 × R0. This relatively short distance range comes from the fact that FRET efficiency depends upon the inverse sixth power of the distance between the donor and acceptor (Fig. 1.12.4). Calculation of R0 for a donor-acceptor pair requires knowledge of: (1) the molar absorption coefficient and the absorption spectrum of the acceptor, (2) the fluorescence emission spectrum of the donor, and (3) the quantum yield (QD) of the donor. The absorption coefficient of the acceptor is usually given at least at one wavelength, and the absorption spectrum can easily be obtained using any commercially available spectrophotometer. The fluorescence emission spectrum of the donor can be determined with a spectrofluorimeter; however, the spectrum usually contains a wavelength-dependent instrument response. The true emission spectrum can be determined with instruments in which manufacturers provide instru-
D′
D β
Principles of Resonance Energy Transfer
α
γ A
R
Figure 1.12.3 The orientation of the emission dipole of the donor (D) and the absorption dipole of the acceptor (A). R is the distance between the two dye molecules. α is the angle of the two dipoles, β is the angle between the transition moment of the donor and the line joining the two dyes, while γ is the angle between the transition moment of the acceptor and the line joining the two dyes.
1.12.4 Supplement 9
Current Protocols in Cytometry
ment-response correction. Alternatively, reference compounds with known emission spectra can be used for calibrating the instrument. Similarly, QD can be determined using reference compounds with known quantum yields, such as quinine in 0.1 N H2SO4, with Q = 0.55, or fluorescein in 0.1 N NaOH, with Q = 0.92. With the knowledge of the absorption spectrum of the acceptor, and the true emission spectrum of the donor, the overlap integral (J) can be calculated according to Equation 1.12.5. After J and QD have been determined, R0 can be calculated, assuming n = 1.33 for aqueous solution and κ2 = 2/3 for random dipole orientation, according to Equation 1.12.9. Since quantum yield and spectral shape may be environment-sensitive, R0 distances may vary as solution conditions change. A selected list of R0 distances for donor-acceptor pairs that have been applied in flow and image cytometric measurements is shown in Table 1.12.1. A comprehensive and useful list of R0 values for over 70 donor-acceptor pairs is provided by Wu and Brand (1994). The largest R0 value reported for a single donor-acceptor pair is 8.0 nm for the
rhodamine B–malachite green dye pair (Yamazaki et al., 1990). The use of molecules that have clusters of acceptors with high molar absorption coefficient for each acceptor can extend the R0 value. Along this line Mathis reported an exceptionally large R0 of 9.0 nm, using europium cryptate as donor and allophycocyanin as acceptor (Mathis, 1993). When applying phycobillin proteins, however, it should be kept in mind that these molecules have bulky dimensions, which can interfere with the original goal, i.e., with accurate distance measurements.
HOW TO MEASURE FRET EFFICIENCY The energy transfer efficiency, as follows from the above formulas, can be determined in a number of different ways. Since energy is transferred from the excited donor to the acceptor, the lifetime (τ), quantum efficiency (Q), and fluorescence intensity (F) of the donor decrease, if the acceptor is present (Equation 1.12.10). As a consequence, the fluorescence
1.0
Energy-transfer efficiency
0.8
0.6
0.4
0.2
0.0 0.0
0.5
1.0
1.5
2.0
Distance (R0)
Figure 1.12.4 Distance dependence of the energy transfer efficiency. Distances are expressed in R0 units. The shaded area shows the useful distance range, where the energy transfer efficiency is between 0.95 and 0.05, meaning (0.61 × R0) – (1.63 × R0). Note that the curve is asymmetrical.
Flow Cytometry Instrumentation
1.12.5 Current Protocols in Cytometry
Supplement 9
intensity of the acceptor increases (sensitization) if the donor is present (Equation 1.12.11).
1− E =
τ AD τD
=
FDA FD
=
QDA QD
Equation 1.12.10
FAD FA
=1+
ε C E ε C D
D
A
A
Equation 1.12.11
In the above equations, the lower indices refer to the donor (D) or acceptor (A), while the upper indices indicate the presence of the donor
Table 1.12.1
(D) or the acceptor (A) in the system. CD and CA are the molar concentrations, while εD and εA are the molar absorption coefficients of the donor and the acceptor, respectively. E is the efficiency of the energy transfer. FRET can be determined by measuring the fluorescence characteristics of the donor or the acceptor. Fluorescence intensity, quantum yield, or fluorescence lifetime of the donor changes upon FRET. The simplest way to measure energy transfer is to determine the decrease in the fluorescence of the donor in the presence of the acceptor. The fractional decrease in the donor fluorescence with the acceptor present is equal to the efficiency of FRET. Fluorescence lifetime measurements are not widely applied for monitoring FRET efficiency because timeresolved fluorescence measurements require sophisticated and expensive instruments.
Characteristic R0 Values for Selected Donor-Acceptor Pairsa,b
Donor (λex/λem)
Acceptor (λex/λem)
R0 (nm)
References
FCA (400/470)
PI in DNA (540/620)
3.0
Szöll§si et al., 1978
IAF (490/515) DAPI (350/470)
DiI-C18 (546/565) EB (510/595)
3.5 3.7
Shahrokh et al., 1991 Maliwal et al., 1995
IAF (490/515) BFP (389/440)
TMR (557/576 GFP (488/511)
3.7 4.0
Taylor et al., 1981 Mahajan et al., 1998
IAEDANS (336/490) RLUC (-/475)
IAF (490/515) EYFP (480/530)
4.4 5.0
Gettins et al., 1990 Xu et al, 1999
Cy3 (554/568) CF (490/525)
Cy5 (649/666) Texas red (596/620)
5.0 5.1
Bastiaens and Jovin, 1996 Johnson et al., 1993
5(6)-CF (490/517) TMR (557/576)
5(6)-CF (490/517) Texas red (596/620)
5.1-5.7 5.2
Chen and Knutson, 1988 Ha et al., 1996
Fluorescein (490/525) C18-Rh (560/590)
EITC (525/545) C18-Rh (560/590)
5.4 5.5-5.8
Carraway et al., 1989 MacDonald, 1990
Fluorescein (490/525) NBD (450/530)
TMR (557/576) LRH (575/595)
5.6 5.6
Kosk-Kosicka et al., 1989 Wolf et al., 1992
AO in DNA (502/526) BPE (480-565/578)
Crystal violet (596/-) CY5 (649/666)
7.0 7.2
Maliwal et al., 1995 Ozinskas et al., 1993
Rhodamine B (540/625) TBP(Eu3+) (307/620)
MG (630/-) APC (650/661)
8.0 9.0
Yamazaki et al., 1990 Mathis, 1993
aλ /λ , wavelengths of excitation/emission in nm. ex em bAbbreviations: AO, acridine orange; APC, allophycocyanin; BFP, blue fluorescent protein; BPE, B-phycoerythrin; CFSE,
Principles of Resonance Energy Transfer
carboxyfluorescein succinimidyl ester; 5(6)-CF, 5(6)-carboxyfluorescein; C18-Rh, octadecylrhodamine B; Cy3, sulfoindocyanine dye Cy3.29-OSu; Cy5, sulfoindocyanine dye Cy5.29-OSu; DAPI, 4′,6-diamidino-2-phenyl indole; DiI-C18, 1,1′-dioctadecyl-3,3,3′,3′-tetramethyl-indocarbocyanine; EB, ethidium bromide; EITC, eosin-5′-isothiocyanate; EYFP, enhanced yellow fluorescent protein; FCA, fluorescamine; GFP, green fluorescent protein; IAEDANS, 5-((((2-iodoacetyl)amino)ethyl)amino)naphthalene-1-sulfonic acid; IAF, 5-iodoacetamidofluorescein; LRH, lissamine rhodamine; MG, malachite green; NBD, 7-nitro-2,1,3-benzoadiazol-4-yl (also known as 7-nitro-benz-2-oxa-1, 3-diazole); PI, propidium iodide; RLUC, Renilla luciferase; TBP(Eu3+), Eu3+ trisbipyridine diamine; TMR, tetramethyl rhodamine.
1.12.6 Supplement 9
Current Protocols in Cytometry
Although it is not necessary to have a fluorescent acceptor, the observation of the increased emission of the acceptor is an important confirmation of energy transfer, because it arises only from FRET, whereas donor quenching can arise from several trivial sources. In an ideal case, where a system is excited at the excitation maximum of the donor and the fluorescence detected at the emission maximum of the acceptor, we can detect fluorescence only if both donor and acceptor molecules are present and there is FRET between them. (Donoronly and acceptor-only samples will not produce measurable fluorescence intensities under these ideal conditions.) Generally, the donor has a tail in the emission spectrum contributing to the fluorescence intensities at the emission maximum of the acceptor. In addition, most acceptors have some absorption at the excitation wavelength of the donor. Hence, the calculation of energy transfer efficiency is more complicated due to these convolution factors. The spectral overlap between the donor’s fluorescence and the acceptor’s own fluorescence should be taken into account when calculating the FRET efficiency. In addition to methods involving the quenching of the donor and the sensitization of the acceptor, FRET efficiency can be determined using time-resolved or steady-state anisotropy measurements. These measurements entail quantitation of the anisotropy increase of the donor in the presence of an acceptor dye (Matkó et al., 1993; Damjanovich et al., 1997). On the other hand, anisotropy may decrease when FRET occurs between the same molecules in identical environments (homo-transfer), due to likely change in fluorophore orientation, whereas there is no change in fluorescence intensity or lifetime of the dyes. The efficiency of homo-transfer can only be determined with fluorescence anisotropy measurements. Another possibility for determining the energy transfer efficiency is based on the altered photobleaching rate of the donor in the presence of acceptor. Although photobleaching should usually be minimized, it can in some cases actually be exploited to measure FRET efficiency. Photobleaching of the donor occurs only when it is in the excited state; before deexcitation occurs there is some probability that photobleaching will remove that fluorophore from the excited state by destroying its molecular structure. The excited-state reactions are instrumental in photobleaching processes. The donor photobleaches more slowly if energy
transfer to an acceptor is occurring, since energy transfer is an alternative pathway for the excited-state relaxation. It can be shown that the fractional change in the photobleaching time constant is the same as the fractional change in the fluorescence lifetime of the donor (Jovin and Arndt-Jovin, 1989a,b). The efficiency of FRET can be determined by comparing the bleaching rate of donor fluorescence in the presence of acceptor to the bleaching rate of the donor in the absence of acceptor (Jovin and Arndt-Jovin, 1989a,b; Nagy et al., 1998a,b). A new variation of donor-photobleaching FRET has been introduced where the efficiency of FRET is determined from the kinetics of the photobleaching of the acceptor, which has been sensitized by a donor. This technique utilizes dye combinations in which the acceptor bleaches much faster than the donor does, even when only the donor is excited. This acceptor-photobleaching FRET allows the measurement of exceptionally accurate efficiencies (Mekler, 1997). In another approach, the acceptor is directly excited and bleached at its absorption maximum, and the intensity of the donor is compared before and after the photodestruction of the acceptor (Bastiaens et al., 1996).
CHOICE OF INSTRUMENT FOR FRET MEASUREMENT When designing a FRET study, the first question to consider is whether the biological problems are best addressed by spectrofluorimetry, flow cytometry, or imaging microscopy. Spectrofluorimetry is often much easier to implement than quantitative flow cytometry or imaging-microscopy studies, despite the advances in instrumentation and software. Usually, donor quenching or acceptor sensitization is measured in this approach. In spectrofluorimetry, the cells can be either in suspension or attached to a coverslip held at an angle to the incident beam. Average fluorescence intensity for thousands of cells can be rapidly measured by these methods. A complete set of samples for FRET efficiency determination should contain at least one sample that is unlabeled, two that are single-labeled (one labeled with donor only, and one labeled with acceptor only), and one that is double-labeled (labeled with donor and acceptor simultaneously). The measured fluorescence intensities must be corrected for light scattering and autofluorescence using the unlabeled sample. At the same time, the fluorescence intensities should be normalized to the same donor and
Flow Cytometry Instrumentation
1.12.7 Current Protocols in Cytometry
Supplement 9
Principles of Resonance Energy Transfer
acceptor concentration. For both corrections, very accurate sample preparation is required; cell concentration should be carefully controlled. Another possible source of error is the contribution to the specific fluorescence signal from unbound fluorophores and cell debris. This is very difficult, if not impossible, to control, especially if the fluorescent label has a low binding constant. Multiple washings decrease the contribution of free fluorophores to the fluorescence intensity, but unavoidably increase the amount of cell debris. Another problem is that some cells may become extremely bright during fluorescent labeling (e.g., dead cells in some immunofluorescence experiments). Some potential problems occurring in spectrofluorimetry could be less serious in flow cytometry or image microscopy. Distortion caused by dead cells, free dye, or debris can be avoided, and uncertainties in cell concentration do not cause a problem due to the cell-by-cell type measurement. These are easily excluded in analyses of flow cytometry data or in selecting fields for analysis by microscopy. A detailed comparison of energy transfer measurements in spectrofluorimetry and flow cytometry is described in several references (Szöll§si et al, 1984; Mátyus, 1992; Trón, 1994). Flow cytometry can provide quantitative measurements on thousands of individual cells, allowing convenient determination of the distribution of fluorescence intensities and energy transfer values in a population. To achieve a high rate of analysis, cells should be in suspension, meaning that cells growing attached to a substrate should be detached mechanically or enzymatically for flow cytometric measurements. These treatments can interfere with the cellular parameters to be investigated. In flow cytometry, donor quenching cannot be used to determine transfer efficiency on a cell-by-cell basis; because the expression levels of various proteins have broad distributions, the donor intensity cannot be measured in the absence and presence of acceptor on the same cell. (For mean values, donor quenching can also be used as in spectrofluorimetric measurements to provide a single mean FRET efficiency value for the whole cell population.) Sensitized emission of the acceptor, however, can be applied to calculate the transfer efficiency on a cell-bycell basis, since with multiple excitations the direct and sensitized emissions of the acceptor can be determined on the same cell. As with spectrofluorimetry, four samples are needed for FRET measurements. Unlabeled cells are used for autofluorescence correction, cells labeled
with donor only for determining spectral overlap correction factor, and cells labeled with acceptor only for determining the correction factor for direct excitation of the acceptor. In flow cytometric measurements, there are three unknown fluorescence parameters: the unquenched donor intensity, the nonenhanced acceptor intensity, and the efficiency of FRET. In order to determine these parameters, three independent signals for the same cell must be measured. The three independent fluorescence signals differ from each other in the wavelength of the excitation or the spectral range of the detection. Because of the different wavelength dependence of the absorption and emission spectra of the donor and acceptor, the corrected transfer signal can be evaluated from the three independently measured parameters, using correction factors determined with the help of single-labeled cells. In this case, biological variance in the level of expression of the investigated protein has no effect on the accuracy of the FRET measurement. The generated distribution histogram of FRET efficiency will provide information about the heterogeneity of the cell population with high statistical accuracy. If there is a change in the FRET efficiency upon stimulus, flow cytometry cannot provide information about the time course of a response in a single cell, but it can easily measure the time course of response in a population of cells. In addition, flow cytometry gives only one value for one cell; the intracellular heterogeneity in FRET efficiency cannot be studied. In many studies, however, the ability of flow cytometry to measure the properties of thousands of cells may be more valuable than detailed information on a smaller number of cells obtained by microscopy. Technical details of how to perform flow cytometric energy transfer measurements can be found in several recent reviews (Szöll§si and Damjanovich, 1994; Damjanovich et al., 1997; Szöll§si et al., 1998). The obvious advantage of image cytometry is that it yields spatial information at the single-cell level regarding FRET efficiency, a type of information that is not available through other approaches. Resolution of subcellular structures, and analysis of cells in situ, can be achieved only by imaging microscopy. Similarly, time-course measurements in single cells can be obtained only by microscopy. Low statistical accuracy is one of the disadvantages of image cytometric FRET measurements, since only a relatively small number of cells can be investigated within a reasonable time frame. The availability of modern digital imaging
1.12.8 Supplement 9
Current Protocols in Cytometry
cameras, as well as associated computer hardware and software, has resulted in rapidly increasing interest in FRET microscopy. These developments have enabled an easy and rapid measurement of fluorescence at up to 106 points of an image simultaneously. Nowadays, one of the most frequently applied approaches for FRET measurements is donor photobleaching. A major advantage of the photobleaching method is that it uses only a single excitation and emission wavelength. The bleaching rate of the donor in the absence of acceptor should be measured under the same experimental conditions as the double-labeled sample, because bleaching rates can vary significantly in different environments. If the efficiency of FRET is relatively low, if the acceptor is nonfluorescent, or if rapid photobleaching prevents measurements of stable fluorescence intensities, the photobleaching method may provide the only practical way to measure FRET efficiency. Because of the destructive nature of the photobleaching method, kinetic measurements of FRET efficiency cannot be performed on the same cell by this method. Intensity-based measurements using the sensitization of acceptor fluorescence can also be used to determine the FRET efficiency on a pixel-by-pixel basis without losing temporal resolution. A thorough comparative analysis of intensityand photobleaching-based FRET microscopic analyses has been carried out recently (Nagy et al., 1998b). It has been shown that the photobleaching-based method inherently overestimates the FRET efficiency, because this technique weights the FRET efficiency values in single pixels, which is different from the intensity-based energy transfer approach. As with the flow cytometric approach, donor quenching cannot be used to determine FRET efficiency on a pixel-by-pixel basis. However, the combination of donor quenching with acceptor photobleaching might overcome this problem, allowing the FRET efficiency to be determined on a pixel-by-pixel basis. It has already been mentioned that the availability of instruments for time-resolved FRET studies is limited because they are sophisticated and expensive. However, significant progress has resulted from the latest developments in this field. Time-resolved fluorescence measurements not only provide an easy way to obtain averaged lifetimes without the exact knowledge of donor concentration, but, more importantly, also give detailed structural information about the donor-acceptor system. The past decade has witnessed tremendous improvements
in this area. Picosecond and nanosecond technologies are considered mature now, and commercial instruments are available. Fluorescence decays can be detected by either the single-photon-counting or the phase-modulation method. Both approaches have successfully been applied in spectrofluorimetry; interesting results are summarized in Wu and Brand (1994). Although fluorescence-lifetime measurement based on phase modulation has been available in flow cytometry for years (Steinkamp, 1993), no FRET application has been published so far. Like spectrofluorimetry, both time-gated and phase-modulated lifetime measurements can be implemented in imaging microscopy. Several nanosecond time-resolved fluorescence images of a sample can be obtained at various delays after pulsed laser excitation of the microscope’s entire field of view. Lifetimes are calculated on a pixel-by-pixel basis from these time-resolved images, and the spatial variation of the lifetimes is then displayed. This technique has been used to detect endosomeendosome fusion in single cells (Oida et al., 1993). The other approach, using modulated illumination in microscopy, resulted in successful application of FRET studies for monitoring oligomerization of epidermal growth factor receptors (Gadella and Jovin, 1995).
LIMITATIONS OF FRET STUDIES For proper application of FRET it is important to understand its limitations as well as its advantages. The most serious drawback of FRET is its modest capacity for determining absolute distances. It is quite good at determining relative distances, namely, whether two points are getting closer or farther apart upon a stimulus. This is caused by the fact that FRET efficiency depends not only on the distance between the donor and acceptor, but also on the relative orientation (κ2) of the dyes (see discussion of Theory of FRET). Even when measuring relative distances, care must be taken to ensure that the orientation factor (κ2) does not change between the two systems under comparison. In addition, the system can be more complicated when a random conjugation of the fluorescent label is applied. For example, in a substantial fraction of experiments, fluorophore-conjugated monoclonal antibodies are used to label cell-surface antigens. According to common practice, covalent coupling reactions frequently occur between isothiocyanate or succinimidyl ester reactive groups of appropriate derivatives of the fluorophore and the
Flow Cytometry Instrumentation
1.12.9 Current Protocols in Cytometry
Supplement 9
Principles of Resonance Energy Transfer
ε-amino groups of lysine side chains of immunoglobulins. The number of exposed lysine side chains of comparable reactivity found in an antibody molecule usually exceeds one. Each antibody molecule may carry several fluorophores, and the labeled lysine side chains of the individual antibody molecules may be different. Reactive groups are often attached to the fluorophores via an n-carbon linker, with n typically ranging from 2 to 12. The linker often allows relatively free rotation of the dye, which minimizes uncertainty of κ2. It also minimizes quenching of the dye by the protein, especially if it is due to a hydrophobic environment. The linker, however, has the disadvantage of adding uncertainty to the exact position of the dye. In general, the minimal length that allows free rotation of the dyes and does not cause quenching is around six carbon atoms. Another problem is that FRET has very sharp distance dependence. For this reason, it is difficult to measure relatively long distances because the signal gets very weak. At the same time, energy transfer tends to be all or none; if the donor and acceptor are within a distance of 1.63 × R0, there is energy transfer, but if they are farther apart, energy is transferred with very little efficiency. When studying cells labeled with donor- and acceptor-conjugated monoclonal antibodies, averaging is performed at different levels. The first averaging follows from the random conjugation of the fluorescent label. An additional averaging is brought about by the eventual distribution of separation distances between the epitopes labeled with monoclonal antibodies. This multiple averaging, an inevitable consequence of the nonuniform stoichiometry, explains why FRET measurements are usually carried out, with different goals, on purified molecular systems and on the surface of the cytoplasmic membrane. In the former case, FRET efficiency values can be converted into absolute distances. Calculation of distance relationships from energy-transfer efficiencies is easy in the case of a single-donor/single-acceptor system if the localization and relative orientation of the fluorophores are known. However, if cell-membrane components are investigated, a two-dimensional restriction applies to the labeled molecules. Analytical solutions for randomly distributed donor and acceptor molecules and numerical solutions for non-random distributions have been elaborated by different groups (Wolber and Hudson, 1979; Dewey and Hammes, 1980; Snyder and Freire, 1982). In order to differentiate between random
and nonrandom distributions, energy transfer efficiencies have to be determined at different acceptor concentrations.
APPLICATIONS FRET is widely utilized for a variety of applications. In one series of studies, FRET is used to obtain structural information that is otherwise difficult to obtain. The major advantage of FRET for structural studies is that, due to the specificity of labeling, the experimental object can be investigated in situ and/or in vivo with little or no interference from the rest of the system. Even complex and heterogeneous systems can be studied this way. In cytometry, usually cells or cell-like objects, such as ghosts and liposomes, are investigated. FRET applications for liposome fusion or for localization of drugs and membrane proteins in liposomes are reviewed by Szöll§si and Damjanovich (1994). Several reviews are available concerning FRET measurements in biological membranes (Szöll§si and Damjanovich, 1994; Wu and Brand, 1994; Damjanovich et al., 1997; Szöll§si et al., 1998). These reviews deal with associations of various membrane proteins, structures of receptors, and conformational changes in transmembrane proteins evoked by ligands and membrane potential changes. In another series of studies, FRET is used as a tool for ensuring high sensitivity of various biological assays. The biotechnological applications of FRET are summarized in recently published reviews that thoroughly discuss working principles of FRET-based enzyme assays and immunoassays, as well as design of tandem dyes and FRET primers for DNA analysis (Clegg, 1995; Szöll§si et al., 1998). A detailed list of possible FRET applications is beyond the scope of this unit. Readers are referred to the reviews cited above to find useful examples of FRET studies. Here, three examples are mentioned, demonstrating new and interesting applications of FRET. Photobleaching FRET measurements have been used to monitor intercellular proximity in order to reveal spatial organization of interacting proteins in the contact region of two cells participating in cytolysis (Bacsó et al., 1996). Interactions between CD8 and major histocompatibility complex (MHC) class I molecules and between leukocyte function antigen 1 (LFA-1) and intercellular adhesion molecule 1 (ICAM-1) have been investigated using donor (fluorescein)– and acceptor (rhodamine)–labeled monoclonal antibodies. The geometry of the orientation of these proteins based on FRET
1.12.10 Supplement 9
Current Protocols in Cytometry
data was consistent with the observed blocking effects of monoclonal antibodies on the cytolytic activity of killer T lymphocytes (Bacsó et al., 1996). A steadily expanding new field in FRET studies is based on the application of green fluorescent protein (GFP) as a sensitive reporter. When two differently colored mutants of GFP, such as enhanced green fluorescent protein (EGFP) and enhanced blue fluorescent protein (EBFP), are covalently linked to different intracellular proteins, and these proteins interact with each other, FRET can be detected. Using this approach, spatial and temporal interaction of Bcl-2 and Bax proteins was studied at the single-cell level by monitoring FRET efficiency (Mahajan et al., 1998). A new extension of the application of GFP proteins in FRET studies has been achieved by combining GFP with luciferase. The bioluminescent resonance energy transfer (BRET) uses bioluminescent luciferase that is genetically fused to one candidate protein and a GFP mutant, enhanced yellow fluorescent protein (EYFP), fused to another protein of interest. If the two fusion proteins come close enough, resonance energy transfer can occur, resulting in changes in the spectrum of the bioluminescent emission. The BRET method was used to assay interactions between proteins encoded by the circadian clock genes kaiA and kaiB in cyanobacterium (Xu et al., 1999). Since Förster first established the analysis of FRET phenomena in 1946, the applications of FRET have increased enormously in various fields of research and biotechnology. Technical improvements in spectrofluorimeters, flow cytometers, and microscopes, as well as introduction of new fluorescent probes with better photophysical properties, open up new areas for employing the FRET method innovatively and successfully.
LITERATURE CITED Bacsó, Z., Bene, L., Bodnár, A., Matkó, J., and Damjanovich, S. 1996. A photobleaching energy transfer analysis of CD8/MHC-I and LFA1/ICAM-1 interactions in CTL-target cell conjugates. Immunol. Lett. 54:151-156. Bastiaens, P.I.H. and Jovin, T.M. 1996. Microspectroscopic imaging tracks the intracellular processing of a signal transduction protein: Fluorescent-labeled protein kinase CβI. Proc. Natl. Acad. Sci. U.S.A. 93:8407-8412.
cellular trafficking and state of the AB5 quaternary structure of cholera toxin. EMBO J. 15:4246-4253. Carraway, K.L. III., Koland, J.G., and Cerione, R.A. 1989. Visualization of epidermal growth factor (EGF) receptor aggregation in plasma membranes by fluorescence resonance energy transfer. J. Biol. Chem. 264:8699-8707. Chen, R.F. and Knutson, J.R. 1988. Mechanism of fluorescence concentration quenching of carboxyfluorescein in liposomes: Energy transfer to nonfluorescent dimers. Anal. Biochem. 172:6177. Clegg, R.M. 1995. Fluorescence resonance energy transfer. Curr. Opin. Biotechnol. 6:103-110. Dale, R.E., Eisinger, J., and Blumberg, W.E. 1979. The orientational freedom of molecular probes: The orientation factor in intramolecular energy transfer. Biophys. J. 26:161-194. Damjanovich, S., Gáspár, R., and Pieri, C. 1997. Dynamic receptor superstructure at plasma membrane. Q. Rev. Biophys. 30:67-106. Dewey, T.G. and Hammes, G.G. 1980. Calculation of fluorescence resonance energy transfer on surfaces. Biophys. J. 32:1023-1035. Förster, T. 1946. Energiewanderung und Fluoreszenz. Naturwissenschaften 6:166-175. Förster, T. 1948. Zwischenmolekulare Energiewanderung und Fluoreszenz. Ann. Phys. (Leipzig) 2:55-75. Gadella, T.W.J. and Jovin, T.M. 1995. Oligomerization of epidermal growth factor receptors on A431 cells studied by time-resolved fluorescence imaging microscopy: A stereochemical model for tyrosine kinase receptor activation. J. Cell Biol. 129:1543-1558. Gettins, P., Beechem, J.M., Crews, B.C., and Cunningham, L.W. 1990. Separation and localization of the four cysteine-949 residues in human β2macroglobulin using fluorescence energy transfer. Biochemistry 29:7747-7753. Ha, T., Enderle, Th., Odletree, D.F., Chemla, D.S., Selvin, P.R., and Weiss, S. 1996. Probing the interaction between two single molecules: Fluorescence resonance energy transfer between a single donor and a single acceptor. Proc. Natl. Acad. Sci. U.S.A. 93:6264-6268. Johnson, D.A., Leathers, V.L., Martinez, A.M., Walsh, D.A., and Fletcher, W.H. 1993. Fluorescence resonance energy transfer within a heterochromatic cAMP-dependent protein kinase holoenzyme under equilibrium conditions: New insights into the conformational changes that result in cAMP-dependent activation. Biochemistry 32:6402-6410. Jovin, T.M. and Arndt-Jovin, D.J. 1989a. Luminescence digital imaging microscopy. Ann. Rev. Biophys. Biophys. Chem. 18:271-308.
Bastiaens, P.I.H., Majoul, I.V., Verveer, P. J., Söling H.-D., and Jovin, T.M. 1996. Imaging the intraFlow Cytometry Instrumentation
1.12.11 Current Protocols in Cytometry
Supplement 9
Jovin, T.M. and Arndt-Jovin, D.J. 1989b. FRET microscopy: Digital imaging of fluorescence resonance energy transfer: Application in cell biology. In Microspectrofluorometry of Single Living Cells (E. Kohen, J.S. Ploem, and J.G. Hirschberg, eds.) pp. 99-117. Academic Press, Orlando. Kosk-Kosicka, D., Bzdega, T., and Wawrynow, A. 1989. Fluorescence energy transfer studies of purified erythrocyte Ca2+-ATPase Ca2+-regulated activation by oligomerization. J. Biol. Chem. 264:19495-19499. MacDonald, R.I. 1990. Characteristics of selfquenching of the fluorescence of lipid-conjugated rhodamine in membranes. J. Biol. Chem. 265:13533-13539. Mahajan, N.P., Linder, K., Berry, G., Gordon, G.W., Heim, R., and Herman, B. 1998. Bcl-2 and Bax interactions in mitochondria probed with green fluorescent protein and fluorescence resonance energy transfer. Nature Biotechnol. 6:547-552. Maliwal B.P., Kusba, J., and Lakowicz, J.R. 1995. Fluorescence energy transfer in one dimension: Frequency-domain fluorescence study of DNAfluorophore complexes. Biopolymers 35:245255.
Ozinskas, A.J., Malak, H., Joshi, J., Szmacinski, H., Britz, J., Thompson, R.B., Koen, P.A., and Lakowicz, J.R. 1993. Homogeneous model immunoassay of thyroxin by phase modulation fluorescence spectroscopy. Anal. Biochem. 213:264270. Shahrokh, Z., Verkman, A.S., and Shohet, S.B. 1991. Distance between skeletal protein 4.1 and the erythrocyte membrane bilayer measured by resonance energy transfer. J. Biol. Chem. 266:12082-12089. Snyder, B. and Freire, E. 1982. Fluorescence energy transfer in two dimensions: A numeric solution for random and nonrandom distributions. Biophys. J. 40:137-148. Steinkamp, J.A. and Crissman, H.A. 1993. Resolution of fluorescence signals from cells labeled with fluorochromes having different lifetimes by phase-sensitive flow cytometry. Cytometry 14:210-216. Stryer, L. 1978. Fluorescence energy transfer as a spectroscopic ruler. Annu. Rev. Biochem. 47:819-846.
Mathis, G. 1993. Rare earth cryptates and homogeneous fluoroimmunoassays with human sera. Clin. Chem. 39:1953-1959.
Szöll§si J. and Damjanovich, S. 1994. Mapping of membrane structures by energy transfer measurements. In Mobility and Proximity in Biological Membranes. (S. Damjanovich, J. Szöll§si, L. Trón, and M. Edidin, eds.) pp. 49-108. CRC Press, Boca Raton, Fla.
Matkó, J., Jenei, A., Mátyus, L., Ameloot, M., and Damjanovich, S. 1993. Mapping of cell surface protein-patterns by combined fluorescence anisotropy and energy transfer measurements. J. Photochem. Photobiol. B Biol. 19:69-73.
Szöll§si, J ., Sza bó, G., So mogyi, B., and Damjanovich, S. 1978. Simultaneous fluorescence labeling of human fibroblast cells with fluorescamine and propidium iodide. Acta Biochem. Biophys. Acad. Sci. Hung. 13:63-66.
Mátyus, L. 1992. Fluorescence resonance energy transfer measurements on cell surfaces: A spectroscopic tool for determining protein interactions. J. Photochem. Photobiol. B Biol. 12:323337.
Szöll§si, J., Trón, L., Damjanovich, S., Helliwell, S.H., Arndt-Jovin, D.J., and Jovin T.M. 1984. Fluorescence energy transfer measurements on cell surfaces: A critical comparison of steadystate and flow cytometric methods. Cytometry 5:210-216.
Mekler, V.M. 1997. A photochemical technique to enhance sensitivity of detection of fluorescence resonance energy transfer. Photochem. Photobiol. 59:615-620. Nagy, P., Bene, L., Balázs, M., Hyun, W.C., Lockett, S.J., Chiang, N.Y., Waldman, F., Feuerstein B.G., Damjanovich, S., and Szöll§si, J. 1998a. EGFinduced redistribution of erbB2 on breast tumor cells: Flow and image cytometric energy transfer measurements. Cytometry 32:120-131. Nagy, P., Vámosi, G., Bodnár, A., Lockett, S.J., and Szöll§si, J. 1998b. Intensity-based energy transfer measurements in digital imaging microscopy. Eur. Biophys. J. 27:377-389. Oida, T., Sako, Y., and Kusumi, A. 1993. Fluorescence lifetime imaging microscopy (flimscopy): Methodology development and application to studies of endosome fusion in single cells. Biophys. J. 64:676-685.
Szöll§si, J., Damjanovich, S., and Mátyus, L. 1998. Application of fluorescence resonance energy transfer in the clinical laboratory: Routine and research. Cytometry 34:159-179. Taylor, D.L., Reidler, J., Spudich, J.A., and Stryer, L. 1981. Detection of actin assembly by fluorescence energy transfer. J. Cell. Biol. 89:362-367. Trón, L. 1994. Experimental methods to measure fluorescence resonance energy transfer. In Mobility and Proximity in Biological Membranes. (S. Damjanovich, J. Szöll§si, L. Trón, and M. Edidin, eds.) pp. 1-47. CRC Press, Boca Raton, Fla. Wolber, P.K. and Hudson, B.S. 1979. An analytic solution to the Förster energy transfer problem in two dimensions. Biophys. J. 28:197-210. Wolf, D.E., Winiski, A.P., Ting, A.E., Bocian, K.M., and Pagano, R.E. 1992. Determination of the
Principles of Resonance Energy Transfer
1.12.12 Supplement 9
Current Protocols in Cytometry
transbilayer distribution of fluorescent lipid analogues by nonradiative fluorescence resonance energy transfer. Biochemistry 31:2865-2873. Wu, P. and Brand, L. 1994. Resonance energy transfer: Methods and applications. Anal. Biochem. 218:1-13. Xu, Y., Piston, D.W., and Johnson, C.H. 1999. A bioluminescence resonance energy transfer (BRET) system: Application to interacting circadian clock proteins. Proc. Natl. Acad. Sci. U.S.A. 96:151-156.
Yamazaki, I., Tamai, N., and Yamazaki, T. 1990. Electronic excitation transfer in organized molecular assemblies. J. Phys. Chem. 94:516-525.
Contributed by János Szöll§si, Sándor Damjanovich, and László Mátyus University Medical School of Debrecen Debrecen, Hungary
Flow Cytometry Instrumentation
1.12.13 Current Protocols in Cytometry
Supplement 9
Light Scatter: Detection and Usage Historically the oldest flow cytometric parameter, light scattering is a collective phenomenon that depends on all of the following: cell size, refractive index difference between the medium and the cell, refractive index differences within the cell, cell shape, and cell orientation in the laser beam. It also depends on scatter detector position with respect to the laser beam direction and scatter detector size. The uses of light scattering in flow cytometry include cell sizing and discrimination among cell types and cellular structural properties. The major use is as an auxiliary parameter in immunophenotyping. Because of the complexity of light scatter phenomena, care must be taken in the interpretation of flow cytometric scattering data. This unit begins with an introduction to the interaction of light waves with matter. A discussion of the fundamental mechanisms of scattering from small particles is followed by one on scattered light detection. Finally some applications of light scattering in flow cytometry are reviewed.
LIGHT WAVES Light can be treated as a wave with sinusoidally oscillating electric and magnetic fields. Here the focus is on the electric field, which is a transverse wave oscillating up and down as
UNIT 1.13
the light wave propagates to the left or right (Fig. 1.13.1). The lasers typically used in flow cytometers emit light at a fixed frequency or frequencies. The wavelength, which is the length of one full period of the sinusoidal wave, is inversely proportional to the frequency. The most commonly used wavelength in an argonion laser is 488 nm, which is in the blue-green region of the visible spectrum. The light wave in a laser is frequently vertically polarized, which means that the incident electric field oscillates up and down in the vertical plane.
SCATTERING What happens when the light wave strikes a cell? First consider the interaction between the light wave and an atom. When the light wave strikes an atom, the electrons surrounding the atomic nucleus oscillate in response to the incident electric field. This causes an oscillating net separation between the electrons and the atomic nucleus that produces an oscillating induced electric dipole. An electric dipole is two equal but opposite charges separated by some distance. The induced electric dipole has a dipole moment with a magnitude that is proportional to the incident electric field. The proportionality constant is the polarizability, which is a bulk property of the material through which the light wave is passing. The refractive
x
wavelength E
y z
Figure 1.13.1 The electric field of a light wave propagating along the z-axis. This wave is linearly polarized with the electric field in the x-z plane. The wavelength is the length of one full period of the wave. Contributed by Gary C. Salzman Current Protocols in Cytometry (1999) 1.13.1-1.13.8 Copyright © 1999 by John Wiley & Sons, Inc.
Flow Cytometry Instrumentation
1.13.1 Supplement 9
index of the particle is proportional to the square root of the polarizability. The oscillating electric dipole radiates a new light wave of the same frequency as the incident wave. This is the scattered light that is sensed by a photodetector. The incident light wave merely provides the electric field driving force that creates the oscillating induced dipole. The dipole example here is an atom; the dipole merely needs to be a particle much smaller than the wavelength of the incident light.
scattering plane for the solid line. The scatter intensity is in arbitrary units. These functions are given in Equations 1.13.1 and 1.13.2. For incident light polarized parallel to the scattering plane, the scattered light intensity ipar is given by
SCATTERING FROM A DIPOLE
where a is the particle radius, λ is the wavelength of light in the medium, r is the distance from the particle to the detector, θ is the scattering angle, and m is the particle refractive index relative to that of the medium. For incident light polarized perpendicular to the scattering plane, the scattered light intensity iper is given by
What does the scattered light intensity angular distribution look like for a single dipole? First some definition of terms is needed. Assume a right-handed coordinate system with x-, y-, and z-axes. The laser beam propagates down the positive z-axis, and the dipole is located at the origin. The z-axis and a line extending from the origin to the detector face define the scattering plane. Figure 1.13.2 shows the scattered light intensity for a dipole as a function of scattering angle. Remember that a dipole is a particle that is very small compared to the wavelength of the incident light. 0° is the forward direction, and 90° is the side or orthogonal scattering direction. Note that the scattered light intensity for the single dipole is the same in the forward and back directions. The incident light is polarized parallel to the scattering plane for the dashed line and perpendicular to the
i par =
4 6 2 16 π a m − 1
λr
4 2
2
m +2 2
2 cos θ
Equation 1.13.1
i per =
4 6 2 16 π a m − 1
λ4 r 2
2
m2 + 2
Equation 1.13.2
Equations 1.13.1 and 1.13.2 illustrate some significant properties of light scattering from small particles. The scattered light intensity increases as the sixth power of the radius. Because of their smaller size, bacteria scatter
Scatter intensity (arbitrary units)
1 × 10 –16 8 × 10 –17 6 × 10 –17 4 × 10 –17 2 × 10 –17
0 0
20
40
60
80
100 120 140
160 180
Scattering angle (degrees)
Light Scatter: Detection and Usage
Figure 1.13.2 Scattered light intensity angular distribution for a particle that is very small compared to the wavelength of the incident light. The particle is a homogeneous polystyrene sphere 0.488 nm in diameter, immersed in water, and illuminated with light at a wavelength of 488 nm. The incident light is polarized perpendicular to the scattering plane for the solid line and parallel to the scattering plane for the dashed line. The scatter intensity units are arbitrary; however, the values in Figures 1.13.2 to 1.13.4 are all in the same units.
1.13.2 Supplement 9
Current Protocols in Cytometry
much less light than do mammalian cells. The scatter intensity decreases as the fourth power of the wavelength. Shorter wavelengths scatter light more strongly. This is why the sky is blue. The polarization of the scattered light is given by
P=
i per − i par i per + i par
=
1 − cos 2 θ 1 + cos 2 θ
Equation 1.13.3
The polarization is the difference between the scattered light intensities with the incident light polarized perpendicular and the light polarized parallel to the scattering plane divided by the total scattered light intensity. Note that the scattered light intensity is strongly polarized at 90° to the laser beam (P = 1). One can observe this phenomenon by looking through polarized sunglasses at right angles to the sun. When the polarizer axis is vertical, the sky is bright; when the polarizer axis is horizontal, the sky is dark. Now what happens as more dipoles are added to make a larger particle?
SCATTERING FROM SEVERAL AND MANY DIPOLES When two dipoles are placed on a line and illuminated by light parallel to the line, both dipoles scatter the light in all directions. Only
in the forward direction are the scattered waves in phase with the incident light wave (Bohren, 1987). In other directions, the scattered waves are not necessarily in phase and may interfere destructively to reduce the scattered intensity. Figure 1.13.3 shows the angular distribution of scattered light intensity for a particle that is the same size as the wavelength of the incident light. Note that the scattering in the forward direction is much larger than that at other angles. The intensity is calculated by squaring the sum of the amplitudes of the waves scattered by the two dipoles. This trend continues for an even larger particle (diameter 10 times the wavelength), as shown in Figure 1.13.4. The actual scattered intensity angular distribution owing to one dipole must include illumination of the dipole with the original incident light as well as contributions from the light scattered by all the other dipoles. A cell can be described as a large collection of dipoles, with each dipole possibly having a different refractive index (polarizability). A theory called the coupled-dipole approximation has been developed to deal with complex arrangements of dipoles. This theory is summarized briefly in Salzman et al. (1990), which also contains many references to original work. Not much progress has been made in accurately explaining the scattered light angular distribution from a biological cell.
Scatter intensity (arbitrary units)
100
10
1
0.1
0.01 0
20
40
60 80 100 120 140 160 180 Scattering angle (degrees)
Figure 1.13.3 Scattered light intensity angular distribution for a particle with a diameter equal to the wavelength of the incident light (488 nm). The incident light is polarized perpendicular to the scattering plane for the solid line and parallel to the scattering plane for the dashed line. The scatter intensity units are arbitrary; however, Figures 1.13.2 to 1.13.4 are all in the same units.
Flow Cytometry Instrumentation
1.13.3 Current Protocols in Cytometry
Supplement 9
SCATTERING BY ANY OTHER NAME The basic mechanism of scattering has now been described. For homogeneous spheres and for some other regular particle shapes, the exact electromagnetic equations can be solved. Lorenz (1890) and Mie (1908) independently solved these equations to obtain scattered intensity angular distributions for homogeneous spheres. Their solutions are generally referred to as the Lorenz-Mie theory. Readable accounts of scattering by small particles are available (Van de Hulst, 1981; Bohren and Huffman, 1983). These general solutions are valid for homogeneous spheres of all sizes. Solutions for spheroids have been developed (Barber and Hill, 1990). The scattering from particles that are much smaller than the wavelength of the incident light is often called Rayleigh scattering (Rayleigh, 1871); however, the mechanism is really that described above. Similarly, diffraction theory is sometimes used to describe near forward angle scattering; again, the mechanism is the one described above. Scattering is scattering. Reflection and refraction of light at a surface can be described exactly by scattering theory.
FORWARD SCATTER DETECTION A forward scatter detector generally has, on the z-axis, a spherical lens that stares into the laser beam after the focused beam has passed through the flow chamber. An adjustable beamstop bar is usually located in front of the lens
to prevent the direct beam from entering the lens. The angular range intercepted by the stop bar has a significant effect on the amount of scattered light reaching the detector. Forward scatter measurements cannot be compared for different experiments if the bar is moved in the middle of a series of experiments. The lens focuses the collected scattered light onto a photodetector, which is often a photodiode, because the scattered light intensity in the forward direction is bright. The polar collection angle (measured from the z-axis) is usually in the range from 1° to 15°. The detected forward scatter intensity increases approximately as the volume or as the cross-sectional area of the particle increases, depending on the acceptance angular range of the detector. The functional relationship may or may not be monotonic, again depending on the acceptance angular range of the detector. The intensity of light scattered in the forward direction is less sensitive to particle refractive index than is the scattered intensity in other directions. Figure 1.13.5 shows calculated forward scatter detector values as a function of polystyrene sphere diameter for a FACSCalibur flow cytometer (Becton Dickinson Immunocytometry Systems [BDIS]). Note that the response function is not monotonic with increasing sphere diameter. A 5-µm-diameter polystyrene sphere has a slightly larger forward scatter detector response than does a 6-µm-diameter sphere. This is observed experimentally (pers. comm., Robert Hoffman, BDIS).
Scatter intensity (arbitrary units)
1010 109 108 107 106 105 104 103 102 101 0
Light Scatter: Detection and Usage
20
40
60 80 100 120 140 160 180 Scattering angle (degrees)
Figure 1.13.4 Scattered light intensity angular distribution for a particle with a diameter 10 times the wavelength of the incident light (488 nm). The incident light is polarized perpendicular to the scattering plane for the solid line and parallel to the scattering plane for the dashed line. The scatter intensity units are arbitrary; however, Figures 1.13.2 to 1.13.4 are all in the same units.
1.13.4 Supplement 9
Current Protocols in Cytometry
Forward scatter detector response (arbitrary units)
2.0 × 106
1.5 × 106
1.0 × 106
5.0 × 105
0.0 0
2
4
6
8
10
Sphere diameter (µm)
Side scatter detector response (arbitrary units)
Figure 1.13.5 Calculated forward scatter detector response as a function of particle diameter for a series of polystyrene spheres with refractive index 1.59 in water. The wavelength was 488 nm. The calculations are for a FACSCalibur flow cytometer. Note the nonmonotonic detector response.
5000
4000
3000
2000
1000
0 0
2
4
6
8
10
Sphere diameter (µm)
Figure 1.13.6 Calculated side scatter detector response as a function of particle diameter for a series of polystyrene spheres. The spheres are in water and have a refractive index of 1.59. The wavelength is 488 nm. The calculations are for a FACSCalibur flow cytometer.
Flow Cytometry Instrumentation
1.13.5 Current Protocols in Cytometry
Supplement 9
SIDE SCATTER DETECTION Detection of scattered light in the direction approximately perpendicular to the laser beam axis is used as an additional flow cytometric parameter to aid in discrimination among cell types in a heterogeneous sample. This detection is variously called side scatter, 90° scatter, orthogonal scatter, or right-angle scatter. Side scatter is more sensitive to refractive index variations than is forward scattering. The detected response increases approximately monotonically with particle cross-sectional area. It is also quite sensitive to internal variations in refractive index and to particle shape. The detector is usually a photomultiplier, because for mammalian cells the scattered light intensity near 90° is two orders of magnitude dimmer than that in the forward direction. Figure 1.13.6 shows calculated side-scatter detector values as a function of polystyrene sphere diameter for a FACSCalibur flow cytometer. Note that 4-µmdiameter and 5-µm-diameter spheres have the same side scatter detector response.
COMBINED FORWARD AND SIDE SCATTER DETECTION
Side scatter detector response (arbitrary units)
The presentation of bivariate data (sometimes called cytograms or dot plots) of forward versus side scatter has proved to be useful in the flow cytometric analysis of heterogeneous samples. Salzman et al. (1975) demonstrated a
crude human white blood cell differential using bivariate displays of forward versus 90° light scattering. The study also reported the first use in flow cytometry of logarithmic amplifiers. The amplifiers were required because of the large dynamic range of the signals. Hoffman et al. (1980) subsequently used this technique to set a gate around lymphocytes, so that fluorescence measurements could be made to detect only lymphocytes containing fluoresceinated monoclonal antibodies on their surfaces. Doornbos et al. (1994) have made extensive calculations and measurements to show the behavior of the forward and 90° scatter detectors as a function of calibration particle size. Figure 1.13.7 shows a calculated forward and side scatter detector bivariate dot plot for a FACSCalibur flow cytometer. The line connects the spheres at 1-µm-diameter intervals from 1 µm to 10 µm. Figure 1.13.8 shows a human leukocyte differential based on forward vs 90° light scattering. Figure 1.13.8A shows the forward scatter distribution and illustrates the overlap among the size distributions for lymphocytes, monocytes, and granulocytes. Figure 1.13.8B shows the side scatter distribution on a logarithmic scale. The granules in the granulocytes produce the highest side scatter for these cells. Figure 1.13.8C shows the bivariate dot plot of forward versus side scatter for human leukocytes.
5000
4000
3000
2000
1000
0 0.0
5.0 × 105
1.0 × 106
1.5 × 106
2.0 × 106
Forward scatter detector response (arbitrary units)
Light Scatter: Detection and Usage
Figure 1.13.7 Calculated bivariate dot plot of forward versus side scatter for a series of polystyrene beads. The wavelength is 488 nm. The calculations are for a FACSCalibur flow cytometer. The line connects the sphere scatter responses at 1-µm-diameter intervals from 1 µm to 10 µm. The nonmonotonic nature of the scatter detector responses is emphasized in this graph.
1.13.6 Supplement 9
Current Protocols in Cytometry
LITERATURE CITED
Light scattering is a useful flow cytometric parameter; however, it must be used carefully and the data interpreted conservatively. One cannot reliably calibrate a scattered-light detector using beads and then extend the results to cells, because the refractive indices of polystyrene beads (1.59) and mammalian cells (1.37) are significantly different and the scattering relationships are nonlinear.
Barber, P.W. and Hill, S.C. 1990. Light Scattering by Particles: Computational Methods. Advanced Series in Applied Physics, Vol. 3. World Scientific Publishing Corp., London. Bohren, C.F. 1987. Physics on a manure heap: More about black clouds. In Clouds in a Glass of Beer, pp. 139-142. John Wiley & Sons, New York. Bohren, C.F. and Huffman, D.R. 1983. Absorption and Scattering of Light by Small Particles. John Wiley & Sons, New York.
0 20 50 80 110 150 190
Number
CONCLUSION
200 0
100
Number
300
10 30 50 70 100 140 170 200 230 FS height
102 SS height
103
104
SS height 101 102 103
104
101
0
10
20
30 40 FS height
50
60
Figure 1.13.8 Scattering from human leukocytes. (A) Number of cells as a function of forward scatter intensity in a flow cytometer. (B) Number of cells as a function of side scatter intensity in a flow cytometer. (C) Bivariate dot plot showing forward versus side scatter for human leukocytes. The population with the highest side scatter intensity consists of granulocytes. The population with the lowest forward and side scatter intensities consists of lymphocytes. The other population consists of monocytes. (Data file kindly provided by Verity Software House and displayed with WinList 3D.)
Flow Cytometry Instrumentation
1.13.7 Current Protocols in Cytometry
Supplement 9
Doornbos, R.M.P., Hoekstra, A.G., Deurloo, K.E.I., De Grooth, B.G., Sloot, P.M.A., and Greve, J. 1994. Lissajous-like patterns in scatter plots of calibration beads. Cytometry 16:236-242. Hoffman, R.A., Kung, P.C., Hansen, W.P., and Goldstein, G. 1980. Simple and rapid measurement of human T lymphocytes and their subclasses in peripheral blood. Proc. Natl. Acad. Sci. U.S.A. 77:4914-4917. Lorenz, L.V. 1890. Upon the light reflected and refracted by a transparent sphere. Vidensk. Selsk. Skrifler 6:1-62. Mie, G. 1908. Considerations on the optics of turbid media, especially colloidal metal sols. Ann. Physik 25:377-442.
Salzman, G.C., Singham, S.B., Johnston, R.G., and Bohren, C.F. 1990. Light scattering and cytometry. In Flow Cytometry and Sorting, 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 81-107. Wiley-Liss, New York. Strutt, J.W. (Lord Rayleigh). 1871. On the light from the sky, its polarization and colour. Phil. Mag. 41:107-120. Van de Hulst, H.C. 1981. Light Scattering by Small Particles. Dover Publications, Inc., New York.
Contributed by Gary C. Salzman Los Alamos National Laboratory Los Alamos, New Mexico
Salzman, G.C., Crowell, J.M., Martin, J.C., Trujillo, T.T., Romero, A., Mullaney, P.F., and LaBauve, P.M. 1975. Cell classification by laser light scattering: Identification and separation of unstained leukocytes. Acta. Cytol. 19:374-377.
This work was performed under the auspices of the National Flow Cytometry Resource at Los Alamos National Laboratory and was supported by the U.S. Army Soldier Biological and Chemical Command, Aberdeen Proving Ground, Md. The author would like to thank Georgia Farris (Los Alamos) for collecting FACSCalibur data and Jianhong Wang (BDIS) and Robert Hoffman (BDIS) for providing optical data for the FACSCalibur.
Light Scatter: Detection and Usage
1.13.8 Supplement 9
Current Protocols in Cytometry
Compensation in Flow Cytometry The term “compensation,” as it applies to flow cytometric analysis, refers to the process of correcting for fluorescence spillover, i.e., removing the signal of any given fluorochrome from all detectors except the one devoted to measuring that dye. The process of compensation is relatively simple in theory—but there are many subtle aspects that render it much more complex in practice. Unfortunately, compensation in flow cytometry is perhaps one of the least understood processes accompanying data collection and analysis, perhaps because it is often described with the linear algebra elements needed for its computation, clouding the understanding of the fundamental process itself. Nevertheless, proper compensation is absolutely crucial for many aspects of flow cytometric analyses, particularly for the ever more popular antigen density determinations. Because compensation is often misunderstood, misapplied, and surrounded by so much incorrect mythology, many laboratories do not set compensation properly. One of the sections in this unit addresses some of the myths surrounding compensation. Figure 1.14.1 illustrates some of the common errors in setting proper compensation. This unit is devoted to compensation in flow cytometry: why is it necessary, how is it accomplished in hardware or in software, how compensation affects the visualization of data and, finally, how best to design an experiment to achieve proper compensation. By the end of this unit, the reader should be able to set correct compensation for experiments and to recognize improperly compensated data in publications and the impact of that error on the interpretation of the results. Although a complete description of compensation requires some necessary linear algebra, understanding this math is not necessary for understanding proper compensation; the equations in the two sections can be skimmed or skipped at will. A majority of flow cytometry users will not answer correctly all three of the questions posed in Figure 1.14.1. The correct answer to question 1 is “sample no. 2.” For the second question, an answer including the use of the dashed quadrant lines is incorrect! And finally, the correct answer for the third question is: “It is impossible to determine which panel is prop-
UNIT 1.14
erly compensated”—therefore, none of them can be said to be correct. If proper compensation is so important for analysis of flow cytometric data, yet so few people know how to set proper compensation, why has flow cytometry produced such good data for so long? The simple answer to this question is that most types of data analysis do not absolutely require proper compensation. Another answer is that the problems associated with incorrect compensation are not as profound in two- or three-color analysis. However, with five-color (or higher) analysis quickly becoming available to many laboratories, these problems, if allowed to persist, can doom experiments. For example, consider the graphs in Figure 1.14.2, which demonstrate that it is possible to compute subset frequencies properly even if the compensation settings are incorrect. However, these frequencies must be computed by comparison to a singly stained sample, not to a sample stained with isotype controls in every color! The correct gate can be set only on the sample stained with FITC-CD3 and PE-isotype control. Figure 1.14.2 also illustrates how proper compensation is absolutely necessary for correct antigen density measurements. Here, any uncorrected spillover will contribute artifactually to the measurement. In addition, proper compensation is necessary when it is important to distinguish dim populations from negative populations; undercompensation will result in overestimating the frequency of the dim cells while overcompensation will result in underestimating their frequency. Note that the ability to correctly calculate subset frequencies in improperly compensated data sets does not necessarily extend to threecolor (or higher) experiments. Later in this unit, an example of this failure is shown. In addition, the appropriate staining controls that aid in the analysis of data (even when compensation is not completely correct) are discussed. These controls, termed “fluorescence minus one” or FMO controls, can be critical to the success of a multi-color experiment.
BASIC COMPENSATION Compensation is the process by which fluorescence “spillover” between detectors is Flow Cytometry Instrumentation
Contributed by Mario Roederer Current Protocols in Cytometry (2002) 1.14.1-1.14.20 Copyright © 2002 by John Wiley & Sons, Inc.
1.14.1 Supplement 22
PE (isotype) fluorescence
A
1000
1
2
1000
100
100
10
10
1
1
1000
3
4
uncompensated
0.1 0.1
1
100
10 100 CD3-FITC
1000
10 1 0.1
PE (isotype) fluorescence
B
1
10
100
1000
1000
1
10
100
1
1000 1000 2
100
100
10
10
1
1
1000
3
4
100
uncompensated
0.1 0.1
1
10 100 CD3-FITC
1000
10 1 0.1 1
10
100
1000 1 CD3-FITC
10
100
1000
Figure 1.14.1 Can you determine which sample is appropriately compensated? In this hypothetical experiment, PBMC were stained with FITC-conjugated anti-CD3 and PE-conjugated isotype control. In both (A) and (B), the top graph shows the uncompensated data. Each of the numbered sections of the lower graphs represents the data as the compensation setting (between the FITC and PE detectors) is increased. In panel B, the data are shown for the same sample, with a much lower PMT voltage for the PE detector used during the collection. Answer three questions: (1) Which sample in panel A (no. 1, 2, 3, or 4) is properly compensated? (2) On what basis did you make this decision? (3) Which sample in panel B is properly compensated? See the introductory paragraphs of the text for the answers.
Compensation in Flow Cytometry
1.14.2 Supplement 22
Current Protocols in Cytometry
mathematically corrected. This procedure became necessary with the advent of two-color, one-laser analysis (Loken et al., 1977). Every fluorescent molecule emits light with a spectrum characteristic of that molecule. These emission spectra overlap, in some cases very significantly. For example, see the spectra shown in Figure 1.14.3 for fluorescein (FITC) and phycoerythrin (PE). A two-detector system can be designed to discriminate FITC fluorescence from PE fluorescence. The emission light is split according to wavelength and distributed to the detectors, each of which has a different filter that eliminates the light within all but a narrow region of the spectrum. Thus, the FITC fluorescence is predominant in the detector with a 530-nm filter; the PE fluorescence is predominant in the detector with a 575-nm filter. However, some FITC fluorescence appears in the PE detector because of the emission overlap of these two fluors. This signal is termed spillover, because it spills over from the FITC detector to the PE detector. Note that it is impossible to design filter sets that will detect emissions from only FITC or PE; spillover will always occur when this dye combination is used. Thus, whenever FITC is present, one will get a signal in the 530-nm band, and also some signal in the 575-nm band. Any PE present will also contribute to the 575-nm band. How then do we calculate how much of the 575-nm signal is from PE, and how much is from FITC? This is the process termed “compensation”: i.e., correcting the PE detector signal for the amount of FITC fluorescence in the PE band.
One-Way Compensation If a fluorescent dye emission is collected through two different light-collecting detectors that have different bandpass filters (i.e., detect light within different areas of the spectrum), one can always estimate how much emission is in one of these detectors based on how much is in the other. This is because the two signals will always vary proportionately (see Fig. 1.14.4). The constant proportionality means that we can exactly determine the area under the curve in the orange fluorescence detector (i.e., orange fluorescence signal) based on the area under the curve in the green fluorescence detector (green fluorescence signal). The ratio of these two values is computed based on a sample that has no fluorescent molecules except FITC. This sample is termed a “compensation sample” or a compensation control, since it is key for determining proper
compensation. For a typical flow cytometer, the FITC emission in the orange fluorescence detector is ∼15% that in the green fluorescence detector. Therefore, if we subtract 15% of the green fluorescence signal from the orange fluorescence signal, then no matter how much FITC is present, the “corrected” orange fluorescence signal will always be zero. That means that PE can now be added to the system, and the amount of signal in the orange fluorescence detector after subtracting 15% of the signal in the green fluorescence detector will represent the “true” PE fluorescence–irrespective of the presence or absence of FITC. This is “one-way” compensation: correction for the emission of a fluorophore in a second detector.
Two-Way (Pairwise) Compensation From the spectra in Figure 1.14.3, it is apparent that PE also has an emission signal in the FITC (green fluorescence) detector. By collecting a sample of cells stained only with PE, one determines this ratio, which is typically ∼2%. This is the compensation control for PE. But if some of the PE emission shows up in the green fluorescence detector, and one uses the green fluorescence signal to estimate the amount of FITC emission in in the orange signal, won’t this make it impossible to correct for the spillovers and compute the true fluorescences? An incomplete answer is to recognize that there are, in an n-color experiment, n unknowns (the amount of each dye present on a cell) with n measurements; this system can be solved exactly. Indeed, a bit of math shows that it is possible to do the compensation exactly (the reader may skip to the next section to avoid the equations). xF is defined as the amount of signal in n detector n that originates from fluorophore x. Therefore, fF1 is the amount of fluorescein signal in the green FITC detector; pF2 is the amount of PE signal in the orange fluorescence detector. Dn is the measured signal in detector n. For the time being, assume that there is no background autofluorescence; autofluorescence will be dealt with later. Therefore: f p D1 = F 1 + F1
Equation 1.14.1
Flow Cytometry Instrumentation
1.14.3 Current Protocols in Cytometry
Supplement 22
CD4 versus CD3 1000
100
100
10
10
1
1
0.1 1000
0.1 1000
100
100
10
10
1
1
0.1 1000
0.1 1000
PE CD4 fluorescence
PE (isotype) fluorescence
isotype versus CD3 1000
100 10 1
correctly compensated
10 1 0.1 1000
100
100
10
10
1
1
0.1 1000
0.1 1000
100
100
10
10
1
1
slightly overcompensated
overcompensated
0.1 0.1
1
10
100
1000
CD3-FITC fluorescence
Compensation in Flow Cytometry
under compensated
100
0.1 1000
0.1
uncompensated
Figure 1.14.2
0.1
1
10
100
1000
CD3-FITC fluorescence
(legend at right)
1.14.4 Supplement 22
Current Protocols in Cytometry
green
Fluorescence orange FITC
A B PE
C 500
D 550
600
650
Figure 1.14.3 Fluorescence emission spectra for FITC and PE. The emission spectrum (the wavelengths of light generated by excitation of these molecules) is shown for an excitation at 488 nm (the same as the argon-ion laser line). FITC emission is maximal at ∼515 nm; typically, a filter centered on 530 nm is used to collect the emitted light (shaded region). The emission of PE is farther red, with a maximum at ∼575 nm; typically, a filter centered on this emission maximum is used to collect PE. Note that FITC has some emission in the wavelength bands used to collect PE fluorescence (B); typically, the amount of light in the 575-nm band is ∼15% of that in the 530-nm band (A). The PE has very little emission in the 530-nm band (C), usually less than 2% of the emission in the 575-nm band (D).
Figure 1.14.2 (left) Incorrectly compensated data can often still be analyzed. These graphs represent data collections of PBMC stained with CD3-FITC and either CD4-PE or a PE-isotype control. The top graphs are uncompensated; the FITC into PE compensation setting is increased for each successive panel. The quadrant lines (dashed lines) represent what would be set based on an unstained sample (or a sample stained with isotype controls for both colors). The solid line represents a gate set based on the singly-stained isotype control from the left panels. Note that the computation of the percentage of CD3 cells that express CD4 (CD3+CD4+) can be correctly performed on any of the panels irrespective of the compensation setting—but only if the gate is set based on the singly-stained sample (left). The quadrant setting based on a complete isotype control stain (dashed lines) would result in incorrect frequencies. Only the correctly compensated sample (middle panels) shows the correct amount of CD4 fluorescence on the CD3+CD4− cells (i.e., no different than autofluorescence); the undercompensated samples have apparent CD4 fluorescence and the overcompensated samples have apparent negative fluorescence. Thus, antigen density measurements are particularly sensitive to improper compensation settings.
Flow Cytometry Instrumentation
1.14.5 Current Protocols in Cytometry
Supplement 22
Emission signal intensity
Fluorescence orange green
Fluorescence orange green
A A
B B
450
500
550
600
450 500 Wavelength (nm)
550
600
Figure 1.14.4 Emission spectra as a function of concentration (or amount) of dye. The FITC emission spectrum is shown for two samples; the right panel has twice the concentration of FITC as the left. The ratio of emissions in the two detector bands, r = B/A, is the same irrespective of the amount of dye. After this ratio is determined for one sample, the signal B in a different sample can be determined simply by measuring A and multiplying by r.
D2 = fF2 + pF2 Equation 1.14.2
The amount of fluorophore x is proportional to (where n is the primary detector used to collect that fluorophore’s emission); thus, to know how much FITC or PE there is in the sample, one needs to calculate fF1 and pF2. It is known (from Fig. 1.14.4) that the ratio of fF1 to fF2 is a constant, as is the ratio of pF2 to pF1, and these constants of proportionality are defined as follows: xF n
f f
S2 =
F2 f F1
Equation 1.14.3 p
F S1 = p 1 F2
p
where the sample fluorescence comes only from a single fluorophore. For example, for the FITC compensation control: D1 = fF1 + pF1 = fF1 + 0 = fF1 Equation 1.14.5
D2 = fF2 + pF2 = fF2 + 0 = fF2 Equation 1.14.6 f f
S2 =
F2 D = 2 D1 F1
f
Equation 1.14.7
Thus, the FITC spillover coefficient is the ratio of the signals measured in each detector for the FITC-stained sample. Similarly, the PE compensation control can be used to calculate: p
S1 =
F1
p
p
F2
=
D1 D2
Equation 1.14.8 Equation 1.14.4
Compensation in Flow Cytometry
Therefore, xSn is the spillover coefficient for fluorophore x between the primary detector for the fluorophore and detector n. This value is determined from the compensation controls,
Note that the values for D1 and D2 in Equation 1.14.8 will be different from those used in Equation 1.14.7, since they represent the signals measured for the PE compensation control.
1.14.6 Supplement 22
Current Protocols in Cytometry
Once the spillover coefficients are known, they can be used on the costained samples to calculate the “true” fluorescences xFn. Rewriting equations 1.14.5 and 1.14.6 above: D1 = fF1 + pF1 = fF1 +
F2 × pS1
p
%
Equation 1.14.9
D2 = fF2 + pF2 =
%
F1 × Sf 2 + pF2
f
Equation 1.14.10
These two equations have two unknowns (fF1 and pF2), which can be solved exactly:
F1 =
f
D1 − 1−
S1 × D2
p
S1 × S2
p
f
%
%
Equation 1.14.11
F2 =
p
S2 × D1
f
D2 − 1−
S1 × S2
p
f
%
%
Equation 1.14.12
PAIRWISE COMPENSATION ON FLOW CYTOMETERS: THREECOLOR COMPENSATION When performing compensation on an instrument, compensation controls are typically used to set the appropriate compensation coefficients. In the case of pairwise compensations (i.e., compensating for spillover between two detectors), the compensation coefficients are essentially the same as the spillover coefficients defined above. Thus, when compensating between FITC and PE, for example, one would use a FITC-stained sample to set the green fluorescence into orange fluorescence compensation to correct for FITC spillover into the PE detector, and then use a PE-stained sample to set the orange fluorescence into green fluorescence compensation. Most laboratories now perform three-color experiments, using a combination of fluorochromes such as FITC, PE, and PE-Cy5 or PerCP as the third color. There is a significant spillover from PE into the red fluorescence detector, and minor yet detectable spillover from the red fluorescence into the orange. Thus, most instruments provide for two pairwise compensations: between green fluorescence and orange fluorescence and between orange fluorescence and red fluorescence. This compensation is performed as a simple extension
of the two-color compensation noted above; i.e., with two sets of two-color compensation settings. For example, after performing the FITC-PE compensations, one would turn to PE-PE-Cy5. The PE compensation sample is used to set the orange fluorescence into red fluorescence compensation setting, and the PE-Cy5 compensation sample is used to set the red fluorescence into orange compensation setting. This process results in four compensation coefficients, two for each pairwise compensation. Note that this process does not directly correct for the amount of FITC fluorescence in the red fluorescence detector (nor vice versa), and is therefore an incomplete compensation. Is this a problem? Often, the answer is no, but one must be careful because it can become a significant problem. To determine if this is a problem, display the FITC versus PE-Cy5 detectors after setting compensation, for the FITC and the PE-Cy5 compensation controls. Spillover between the green and red fluorescence detectors means uncorrected compensation that can significantly affect the data. The only way that one will be able to correct for this will be to use software compensation after data collection.
COMPLETE COMPENSATION This section extends the equations given above to provide the complete solution to compensating any number of detectors. To avoid the mathematics, the reader may skip to the next section (see Autofluorescence). The general solution to compensating n detectors is an extension of the equations above. The procedure is much the same: for each fluorophore, the spillover coefficient between the primary detector for that fluorophore and every other detector being measured is determined based on the compensation control for that fluorophore. Extension of equation 1.14.1 to n fluorophores (again, ignoring the contribution of autofluorescence):
D1 =1F1 + 2F1 + 3F1 +K+ nF1 D2 = 1F2 + 2F2 + 3F2 +K+ nF2 M Dn = 1Fn + 2Fn + 3Fn +K+ nF
n
Equation 1.14.13
Flow Cytometry Instrumentation
1.14.7 Current Protocols in Cytometry
Supplement 22
Or, more simply: n
Dm = ∑ i Fm i =1
Equation 1.14.14
Likewise, we know that the contribution of fluorophore i to detector m will be equal to the contribution of fluorophore i in detector i multiplied by the spillover coefficient iSm. Note that, by definition, iSi = 1. We define the “true” fluorescence of fluorophore i as Ti: m
Fi = Ti × mSi
Equation 1.14.15
Therefore, we can rewrite equation 1.14.14 in terms of the spillover coefficients and the “true” fluorescences: n
Dm = ∑ (Ti × mSi ) i =1
Equation 1.14.16
In linear algebra terms, D is the vector of measured fluorescences, T is the vector of true fluorescences, and S is the matrix of spillover coefficients whose diagonal elements are all 1: D=T×S Equation 1.14.17
To determine the values in T, equation 1.14.17 can be solved by premultiplying by the inverted matrix of S: S −1 × D = S −1 × T × S Equation 1.14.18
S −1 × D = T Equation 1.14.19
Complete compensation, therefore, requires an inversion of the spillover coefficient matrix followed by multiplication by the measurement vector. The inverse of the spillover matrix is also termed the compensation matrix; the elements of the compensation matrix are the compensation coefficients. Note that the compensation coefficients are therefore different from the spillover coefficients! Compensation in Flow Cytometry
The spillover coefficients are closely related to the spectrum of a fluorophore: they convey the amount of a fluorophore’s emission in each of the detectors. The compensation matrix tells how much of each detector’s value must be subtracted in order to determine the final calculated true fluorescence. Table 1.14.1 and Table 1.14.2 illustrate the difference between these two for a simple three-color matrix, for an experiment utilizing the fluorophores FITC, PE, and PE-Cy5. The incomplete compensation that is typically done on instruments, using pairwise compensations between green/orange and orange/red, would be identical to this matrix with the exception that the corner (nondiagonal) elements would have values of 0.0. Note that while the spillover from FITC to PE-Cy5 is 4% (Table 1.14.1), the actual compensation value needed is only 0.17% (Table 1.14.2), because of the similarity between FITC and PE in terms of the ratio of orange to red signal (i.e., from Table 1.14.1, for FITC it is 0.04/0.18 = 0.22 and for PE it is 0.21). Thus, the compensation of orange into red for PE removes the FITC component from the red fluorescence. (Incidently, this this fortuitous relationship explains why most experiments do not require an explicit compensation between FITC and PE-Cy5—most of this compensation is taken care of by the combination of FITC to PE and PE to PE-Cy5.)
AUTOFLUORESCENCE Autofluorescence throws a small kink into compensation, but, as it turns out, does not change one’s ability to deconvolute spillovers. Cellular autofluorescence is present in all detectors to varying extents, and provides a background that varies from cell to cell. There are three ways to deal with autofluorescence. One way is to devote a single detector to autofluorescence measurement. Because the autofluorescence spectrum of similar cells is generally identical, autofluorescence can be treated as just one more type of fluorescent molecule. Then, by compensation, one can actually correct for the contribution of autofluorescence to all detectors. This process can significantly enhance sensitivity for detection of low-density antigens. Autofluorescence compensation is more fully described in published references by Roederer and Murphy (1986) and Alberti et al. (1987). This process works best for cell types that have a lot of autofluorescence (like cultured cell lines, fibroblasts, or large highly cytoplasmic
1.14.8 Supplement 22
Current Protocols in Cytometry
Table 1.14.1 Typical Spillover Matrix for a Three-Color Compensationa
Fluorophore FITC PE PE-Cy5
Detector Green
Orange
Red
1.000 0.009 0.005
0.180 1.000 0.029
0.040 0.213 1.000
aNote: The diagonal elements are 1, since the contribution of each
fluorophore to its cognate detector is defined to be 100%. In this table, the FITC into PE spillover is 18%; the PE into FITC spillover is 0.9%.
Table 1.14.2 Typical Compensation Coefficient Matrix for Three-Color Compensationa
Fluorophore FITC PE PE-Cy5
Detector Green
Orange
Red
1.0000 −0.0079 −0.0047
−0.1802 1.0000 −0.0281
−0.0017 −0.2126 1.0000
aNote: The matrix in Table 1.14.1 has been inverted. These values
correspond to what would be set on an instrument for performing pairwise compensation (after negation; i.e., the green to orange compensation is 18.02%, and to red it is 0.17%).
cells). It does not work very well for lymphocytes, nor does it work well for channels where the signal level is very low (like the red fluorescences). This is because the error in the ability to measure the autofluorescence exceeds the actual signal level; hence, the correction process does not help at all. The second way is an exact mathematical treatment of autofluorescence. This requires collection of one more control sample: an unstained sample. The mathematics required to do this calculation is no longer linear, but affine. This topic is dealt with in detail by Bagwell and Adams (1993). Note that the limitations of autofluorescence correction noted above (i.e., trying to work with cell types that have very low autofluorescence, like lymphocytes) apply to this method equally. The third way to deal with autofluorescence is to simply ignore it: applying the matrix algebra described above still works to make the detector measurements independent, such that only autofluorescence and the specific fluorescent molecule of interest contribute to each detector. The only effect that this has on the measurement is that the absolute amount of
autofluorescence is no longer comparable to uncompensated (and unstained) cells. However, this is virtually never a problem; the absolute amount of autofluorescence is in most cases irrelevant. In fact, the “true” fluorescence for a given cell population is then simply determined by subtracting the mean (or median) fluorescence intensity (MFI) for unstained cells from the MFI for stained cells, when both are collected with the same compensation settings; in fact, this “true” fluorescence is correct, independent of the presence or absence of autofluorescence. Standard compensation in the presence of autofluorescence, therefore, works just fine: the resulting values are independent of the presence of other reagents, and are proportional to the amounts of the fluorescent molecules present. Note, however, that if the autofluorescence of the stained cells in the compensation sample is different than that of the unstained cells in the compensation sample, then the computed compensation will be incorrect! Thus, one could not use CD14-FITC, which stains highly autofluorescent monocytes, to compensate Flow Cytometry Instrumentation
1.14.9 Current Protocols in Cytometry
Supplement 22
against unstained lymphocytes, which have low autofluorescence.
RESONANCE ENERGY TRANSFER (TANDEM) DYES
Compensation in Flow Cytometry
In the late 1980s, PE-Cy5 became the first resonance energy transfer (RET) dye to be regularly used in flow cytometric analysis. RET dyes are comprised of two covalently linked fluorophores having the property that the emission spectrum of one (the donor) overlaps with the excitation spectrum of the other (the acceptor). Under the appropriate conditions, the tandem dye has spectral properties that consist of a fusion of the properties of the individual dyes: an excitation spectrum that is the sum of the donor and acceptor but an emission spectrum that is principally that of the acceptor (UNIT 1.12). In the particular case of PE-Cy5, the tandem is excited by the 488-nm argon laser line, exciting the PE molecule, but emits light at ∼680 nm. The energy of excitation of the PE is transferred to the Cy5 before the PE itself emits light with its own characteristic spectrum, ∼575 nm. Cy5 itself is not excited by the 488-nm line directly. Were the efficiency of energy transfer between PE and Cy5 perfect, then the pair could be considered to be simply a well-characterized single fluor with a defined emission spectrum. However, the efficiency of transfer is never 100%, and, in fact, can vary significantly between different lots of Cy5PE. This variation has significant implications for compensation. Inefficient transfer from donor to acceptor means that the residual energy is emitted directly by the donor. In the case of the Cy5PE tandem, this appears as normal PE emission, ∼575 nm. The less efficient the tandem, the more 575-nm emission occurs. This is manifested by a greater spillover between the 680nm detector and the 575-nm detector, requiring greater compensation. Because of lot-to-lot variation of Cy5PE conjugates, this means that different compensation values can be necessary for each lot of a tandem. While the compensation may be correct for one tandem, another may be undercompensated or overcompensated. To date, manufacturers have achieved a quality control in the manufacture of tandem dyes such that most conjugates from a single manufacturer are very similar and require essentially the same compensation value. However, this is not the case for tandems from different manufacturers.
In general, when several different tandem conjugates in a single experiment are used, compensation controls should be generated for each in order to determine whether or not a single compensation value can be used for all tandems. If the tandems are sufficiently different that this is not the case, then the compensation must be adjusted individually for each different conjugated tandem. This problem can be a minor nuisance when only a single tandem type is used. However, in the last few years a number of different tandems have been introduced for use in flow cytometry. Almost certainly, it will be impossible to guarantee identical spectral properties for all different conjugates of these many tandems (e.g., PE-Texas Red, ECD, PE-Cy7, APC-Cy7). Therefore, it will be necessary to be able to apply compensation matrices specifically calculated for each lot of tandem used in any given panel. This can be achieved only with significant software support, either by selective modification of the hardware compensation values depending on the stains used for each tube, or by applying the appropriate coefficients during analysis (software compensation). In any case, it is important to remember that the use of tandem dyes in flow cytometry introduces special problems for compensation. Each different tandem used in an experiment must be carefully characterized to determine whether a common compensation value can be used or whether compensation values must be tailored to each conjugate.
COMPENSATION: EFFECT ON VISUALIZATION OF DATA Compensation has a profound impact on the visualization of data. After all, the goal of compensation is to remove the covariation in two measurements to provide a display for which the measurements are independent (and hopefully show “rectilinear” distributions—i.e., distributions are found vertically or horizontally displaced from unstained cells in the absence of co-expression of markers). However, this is often not the case, particularly as the far-red dyes (e.g., Cy7 tandems, APC tandems) are used more often. Indeed, compensation leads to visualizations of the data that appear wrong to most users (Fig. 1.14.5 and 1.14.6). A full description of this artifact is found in Roederer (2001); a summary is provided here. As illustrated in Figure 1.14.5, the process of compensation cannot change the “width” of the uncompensated distribution. Ideally, the uncompensated events would lie on a very tight
1.14.10 Supplement 22
Current Protocols in Cytometry
line extending at a 45° angle from the unstained population. In reality, the events fall about this line. This distribution away from the line is driven by errors in the measurement process. While there are small errors in the electronics that can contribute to this spread, the most significant source of error is photon-counting statistics (Roederer, 2001). Any fluorescence measurement is performed by integrating the number of photons collected from the cell as it passes through the laser. In the deeper-red channels, the number of events actually counted can be relatively small—a few dozen or hundred. A fundamental aspect of counting events is that the minimum error (standard deviation) associated with any count is equal to the square root of the number of counts. Therefore, a cell with 100 photon events at the detector has an associated error in this measurement of ±10%; i.e., running many cells with exactly the same number of fluorescent molecules through the system will give a distribution that has a width of at least 20% of the signal level. It is this counting error that leads to the spreading of the distribution in the uncompensated cells. Unfortunately, one aspect of the log-log display commonly used to view fluorescence data is that an apparently small distribution at highsignal levels becomes enormous when shifted down (by the linear compensation process) to low-signal levels. This is illustrated in Figure 1.14.5. It is important to be able to distinguish uncompensated (or under-compensated) data from properly compensated data. Note that the relationship between two detectors for uncompensated data (and under-compensated data) is linear, because of the proportional relationship of spillover (Fig. 1.14.4). In a log-log display, a linear relationship is a straight line with exactly a 45° slope. After compensation, the spreading in the distribution is related to the square root of the signal (because counting error is proportional to the square root of the signal). A square-root relationship in a log-log plot is a straight line with a 22.5° slope. Note that the slope of the right graphic in Figure 1.14.5 has a shallow slope (1:2 decade/decade), compared to the left graphic (1:1). This can aid in understanding whether data may be properly compensated. Finally, it is crucial to realize that the spread in the properly compensated distribution cannot be “corrected.” This spread arises from fundamental counting errors that contribute to the minimum possible measurement error. Fur-
thermore, this spread is a nonlinear relationship (square-root); compensation is a linear process. Thus, simply turning up the compensation value will not straighten out this distribution and will only lead to problems with data analysis. Again, for a full discussion of why overcompensating the data cannot help, see Roederer (2001).
GATING AND ANALYSIS CONTROLS: FMO CONTROLS Given that properly compensated data can show a spreading distribution into other detectors, it becomes apparent that discriminating positive from negative events is no longer a simple process of selecting a single threshold for positivity. The distribution shown in the previous discussion is reproduced in Figure 1.14.6A, for the purpose of asking the question: How do we identify cells that would be positive for an APC-Cy5.5 (y-axis) stain? For this purpose, the author introduces the concept of fluorescence-minus-one (FMO) gates. In trying to determine the optimal gate position for a given color, it should be apparent that the best control on which to base this position is one in which the same cells have been stained with everything except the one reagent of interest. After all, this is the most rigorous control as only one variable has changed between the two samples; therefore, any differences between the two distributions can be ascribed solely to the addition of the reagent of interest. This control sample, which has every fluorescence except one, is termed an FMO control; any gate based on this sample is an FMO gate. A one-dimensional (1-D) FMO gate is defined as the gate that best discriminates positive from negative events in the channel of interest when viewing a (1-D) histogram of that channel (i.e., at the high end of the fluorescence distribution of the FMO sample). A 2-D FMO gate would be that devised on a two-dimensional graphic of the parameter of interest versus another parameter for the FMO sample. Figure 1.14.6 contains several examples of FMO samples and gates to illustrate this process. In Figure 1.14.6A, the question was how to best identify positive events. Note that events that are below the putative isotype gate (a gate defined on a fully unstained sample) are easily identified as negative (open circles). However, at least some events that are stained only with CD57-APC would be considered positive with this gate—they rise over the gate because of the spreading in the distribution (see Fig 1.14.5).
Flow Cytometry Instrumentation
1.14.11 Current Protocols in Cytometry
Supplement 22
A
B Uncompensated
Compensated
104
104
1100 700
fluorescence
103
103 200
102
102
101
101
100 100
101
102
103
104
100 100
Primary fluorescence
101
102
103
104 (−200)
Figure 1.14.5 Measurement errors lead to spreading of properly compensated distributions. This example shows data collected for PBMC stained with APC-conjugated anti-CD57. This antigen shows a “smear” of events from negative to very bright, leading to an excellent visualization of compensation. (A). Uncompensated distribution of the APC signal versus the APC-Cy5.5 signal. There is considerable spectral overlap between these two channels; in addition, as they are in the far red, the number of actual photons counted is relatively small. The goal of compensation is to remove the contribution of the primary fluorochrome from the spillover channel (arrow). Note that compensation is a linear process, in which the amount to subtract from any given cell is based on the primary fluorescence value. Thus, every cell on the line of the arrow will have the same amount of fluorescence removed from it, an amount that is proportional to the primary fluorescence value (i.e., proportional to ∼500). Since the same amount has to be subtracted from every cell on the line, the vertical width of the uncompensated distribution cannot change in absolute amounts. The bottom of this distribution is at ∼700 fluorescence units, the top at ∼1100. Therefore, after compensation, the vertical width of this distribution must still be 400 units. (B) The same data, after correct compensation. Note that the vertical width of the distribution at the line is still 400 units, centered at ∼5 ± 200. All events below a value of 1 are forced onto the axis; hence the accumulation of a large number of events at the very bottom. The distribution must extend all the way up to 200. The width of this distribution is determined principally by photon-counting statistics (see text).
Compensation in Flow Cytometry
Therefore, a sample that is stained with everything except APC-Cy5.5 is used to define the limit of the negative population (FMO control). Note that the 1-D FMO gate (drawn on a histogram of APC-Cy5.5 for the FMO sample, shown as the upper dotted line in Fig. 1.14.6) would be accurate in that all events above this gate would truly be positive for APC-Cy5.5. However, this gate misses many events that are in fact positive—for example, the left-most event in the middle pair (shaded circle). The 2-D FMO gate, which curves upward with the spread of the negative events, can accurately discriminate the positive events from those events that are in the negative distribution and are likely negative.
Note that the APC-Cy5.5 fluorescence of the two central events is the same, yet one is clearly positive and the other is in the negative distribution. Thus, within the APC-Cy5.5 distribution, there are areas in which both positive and negative events can overlap. The multivariate approach to FMO gating will always be superior for identifying a greater fraction of positive events. Figure 1.14.6B illustrates the use of the FMO gating to most accurately enumerate CCR5-positive T cells. This example shows clearly that the threshold for discriminating CCR5 positivity will depend on the amount of CD8-APC-Cy7 fluorescence, the threshold for positive cells is higher for the CD8+ T cells than for the CD8– T cells. This example shows why
1.14.12 Supplement 22
Current Protocols in Cytometry
10
3
10
2
FMO gates: 2-D 1-D Isotype gate
101
B10
FMO control
5
Test sample
0.42% 10
CCR5-APC
APC-Cy5.5
A
100
5.39%
4
103 10
2
10
1
0.2%
1.8% 8.8%
19.9%
100 100
101
102
103
100
101
102
103
100
101
102
CD57-APC
C
10
103
104
100
101
102
103
104
105
CD8-APC-Cy7
Unstained control FITC PE Cy5PE Cy7PE 5
FMO control CD3
Fully stained CD3 CD4 CD8 CD45RO
CD8 CD45RO
104 103 10
FMO gate Isotype gate
2
PE
101 10 104
FMO gate
103
Isotype gate
102 101 100 100
101
102
103
104
100
101
102
103
104
100
101
102
103
104
FITC
Figure 1.14.6 Fluorescence Minus One (FMO) gates are an accurate way to identify positive versus negative events. See text for full discussion. (A) The distribution as shown in Figure 1.14.5 is reproduced here. In these two graphs, cells were stained only with CD57-APC. The isotype gate is that defined by an unstained (or fully isotype-stained) sample. The 1-D FMO gate is defined by the CD57-APC-stained sample, examining only the distribution of APC-Cy5.5 and setting the threshold above all events. The 2-D FMO gate is defined by the limit of APC-Cy5.5 distribution when viewing this two-dimensional graph. (B) Human PBMC were stained and gated for CD3 (not shown), CD8-APC-Cy7, with (right) and without (left) CCR5-APC. The expression of CCR5 is dim, and accurate discrimination of positive and negative events is necessary. Percentages show the fraction of CD8 T cells within each “positive” gate: the isotype gate (lower dotted line), the 1-D FMO gate (upper dotted line), and the 2-D FMO gate (solid polygon). (C) A four-color (single-laser-excited) staining combination is used to illustrate the utility of FMO gates. The goal in this illustration is only to identify the CD4 T cell population accurately. The stains used for each sample are listed in the table above the graphs. In the upper 3 graphs, the data are properly compensated. The lower 3 graphs represent exactly the same data, except that the PE-Cy7 into PE-Cy5 compensation setting is off by 20%. Note that the fact that this compensation was incorrect is not evident, as neither of these channels is viewed. The effect on the PE channel of this incorrect compensation is due to the interaction of compensation settings across channels (were compensation set by spillover values instead of compensation values, this would not occur; hence, the recent trend by manufacturers to provide control over compensation via the spillover domain). In this sample, the CD8 T cells (PE-Cy5+) have uncorrected fluorescence in the PE channel and show up as a separate, dull population. The “isotype” gate is that defined by the fully isotype-stained sample (far left); the FMO gate is that defined by cells stained with everything except CD4-PE. Without the FMO gate, it would be nearly impossible to know where to set the discriminating gate.
Flow Cytometry Instrumentation
1.14.13 Current Protocols in Cytometry
Supplement 22
full isotype-stained cells are inadequate and in the sample with no CCR5-APC, the isotypebased gate shows 8.8% positive events! The difference between a 1-D FMO gate (1.8%) and a 2-D FMO gate (5.4%) further shows that the multivariate FMO gate will always reveal more of the truly positive events. Finally, Figure 1.14.6C illustrates that FMO gates can even overcome incorrect compensation. In this simplistic example, the goal was to define an appropriate gate to distinguish CD4 T cells. As shown in the properly compensated samples, and like previous examples, only the FMO gate (in this case, a 1-D FMO gate) gives what is a proper division between the CD4 T cells and other cells. Importantly, the FMO gate (which is then in a different position) works for the data when there is incorrect compensation. It might be appreciated that there are as many FMO controls as there are colors in the experiment. However, this is typically not necessary. FMO controls are necessary only for those channels where accurate identification of the positive and negative populations is crucial, and where the separation between these populations is not great (i.e., dimly-expressed antigens or those that have continuous distributions from negative to bright). A typical staining panel might require only one or two FMO controls. Nonetheless, it is apparent that an unstained cell sample (or one stained with all isotype-control antibodies) is useless compared to an FMO control.
COMPENSATION CONTROLS: CELLS
Compensation in Flow Cytometry
No matter how compensation is to be set, the most critical factor is the selection of good compensation controls. To properly calculate the spillover coefficients necessary for the computation process, one must have samples that are stained to different extents for each fluorophore of interest (Fig. 1.14.7). A good strategy to adopt is to choose the brightest reagent among the ones being compensated to use as the compensation control. For regular immunophenotyping studies, one might alternatively select a very highly expressed antigen, such as CD8 or CD45, to stain for the compensation control. Figure 1.14.7 shows why it is necessary to have a positive and a negative population to set compensation properly: it is the ratio of the differences in the fluorescences between these two populations that is critical for defining the compensation. Strictly speaking, a true “negative” population is not necessary; it is only
necessary that two different levels of staining be present. Thus, in Figure 1.14.7, proper compensation could be set by using only the dim and the bright populations. Mathematically, the ratios of the differences between orange fluorescence and green fluorescence for each of the three pairs of populations in Figure 1.14.7 are identical. Thus, there is nothing special about a negative population with respect to the calculation of proper compensation. From Figure 1.14.7, one can also ascertain another crucial criterion for setting proper compensation: that the two populations used for setting compensation have the same autofluorescence and other background fluorescence values. Otherwise, the values for ∆green and ∆orange will be due not only to the fluorescent stain, but also to differences in background fluorescences between the two cell types. This would result in incorrect compensation. Ideally, the best compensation control is one that uniformly labels all cells in a population very brightly. These stained cells can then be mixed with unstained cells in equal numbers to generate two different peaks. For immunophenotyping of hematopoietic cells, CD45 represents an excellent choice. For each color in an experiment, label cells with CD45 conjugated to that fluorochrome. After washing the cells, mix them with equal numbers of unlabeled cells; this will be the compensation control. Although simple, this process is already more cumbersome than many laboratories are willing to perform. The next best compensation control is to use a bright stain that resolves two populations of cells within the same sample. CD8, which is highly expressed on CD8 T cells, is an ideal choice for immunophenotyping of blood cells. Note that if one chooses a reagent that specifically binds only to lymphocytes, one must first select a scatter gate for lymphocytes before setting compensation. For example, consider the use of conjugated CD8 as a compensation control for staining peripheral blood mononuclear cells (PBMC). CD8 T cells, which are lymphocytes, will be brightly stained. The unstained cells will include both lymphocytes and monocytes; thus, as a whole, their autofluorescence is significantly higher than that for lymphocytes alone. If one were to set compensation based on this ungated population, the result would be undercompensation. One approach taken by some laboratories for generating compensation controls is to stain a single sample of cells with antibodies that bind to exclusive populations within the sam-
1.14.14 Supplement 22
Current Protocols in Cytometry
Orange linear fluorescence
bright
dim unstained
orangebr orangedim
greendim greenbr
Green linear fluorescence
Figure 1.14.7 The most accurate compensation is achieved with the brightest compensation controls. This graph shows three populations of cells stained only with FITC: negatives, a dim population, and a bright population. Linear fluorescences illustrate the principle; the same principles hold true for logarithmic amplification. The correct spillover coefficient is the ratio of the ∆orange signal to the ∆green signal (for either the dim or the bright cells). In an ideal world, this ratio is the same for dim or for bright cells. However, the ∆orange and ∆green values will have an inherently greater proportionate error for the dim cells than for the bright cells; therefore, the spillover coefficient will be less accurately determined on the basis of the dim population than on the bright population. For logarithmic scaling, this problem is exacerbated because dim populations can have orders of magnitude less fluorescence than bright populations (remember, the error of the measurement varies with the square root of the absolute value of the measurement).
ple. For example, for compensating a threecolor analysis of PBMC, a single sample could be stained with reagents for CD19, CD8, and CD4. Since these reagents, for the most part, stain unique subsets of cells that have roughly the same autofluorescence, with CD8− NK cells providing the negative population, the single sample provides all the necessary populations. However, this forced economy of sample will inevitably lead to poor compensation at some point, for several potential reasons. First, there may not be enough of any given subset (CD19+, CD8+, CD4+, or triple negative) to provide a statistically accurate determination of the fluorescence distribution. Second, the autofluorescence of all these populations may not necessarily be the same for every cell sample. Third, CD19 and CD4 are relatively dim reagents; it is likely that other reagents in the panels are brighter than this (see discussion above regarding Fig. 1.14.7). Finally, this approach becomes untenable for experiments with more than three colors, or with cell samples for which it is impossible to
find enough distinct populations that can be uniquely labeled with reagents. There are too many potential problems with this method (simultaneous compensation stains on the same cell sample) to warrant its use; it should be discontinued. In summary, the ideal compensation control has a good representation of two or more populations of cells with as great a difference in fluorescence between these populations as possible. In addition, these populations must have the same autofluorescence (and other background fluorescence values). Finally, the ideal compensation control is labeled with only one of the colors of the experiment; there should be one compensation control for each color. Remember that with the use of tandem (RET) fluorophores like PE-Cy5, one compensation control may be needed for each different conjugate of the tandem in the experiment (see discussion of Resonance Energy Transfer Dyes). Flow Cytometry Instrumentation
1.14.15 Current Protocols in Cytometry
Supplement 22
COMPENSATION CONTROLS: BEADS
Compensation in Flow Cytometry
Recently, manufacturers have provided “compensation beads” that can be used to set up compensation. The only beads that work well for this purpose are those that bind to the actual reagents that are used in the experiment. Do not rely on the manufacturer to provide the actual fluorochrome; it may not spectrally match the one being used. Beads that are coated with antibodies, which bind to the reagents of use, provide an excellent compensation control, as the signals most closely reflect those obtained with cells. There are several advantages to using antibody-capture beads as compensation controls. (1) They will work equally well with any reagent being used (assuming that the bead captures the reagent). Since the beads capture the antibody, 100% of the capture beads will be fluorescent. If using an antibody that is expressed by a very small fraction of cells, it would be nearly impossible to use cells to compensate that reagent. (2) The beads are highly uniform. The precision with which the spillover can be calculated with beads is therefore very high. This is most important not in the primary detector channel, but the spillover channels. For example, a FITC compensationcontrol cell sample has a small amount of fluorescence in the PE-Cy5 detector—but this amount is not much above autofluorescence. To accurately compensate FITC out of the detector, however, the amount of spillover must be precisely determined—a difficult chore given the large variation in autofluorescence from cell to cell. Because the beads have essentially no autofluorescence and have a very uniform binding capacity, the spillover can be determined with great precision. (3) The beads, after labeling, are very stable and can be re-used for several experiments. (Note: tandems of PE, such as PE-Cy5 and PE-Cy7, exhibit changes in emission spectra—and therefore compensation requirements—with exposure to light; therefore, it is not recommended that compensation controls be used across experiments unless they are carefully stored in the dark and the cells are stained in the dark as well!) Antibody-capture beads have a couple of disadvantages. (1) They may not be as bright as some very bright reagents on cells. While the precision of the beads is very high, they probably should not be used to compensate samples that are more than ten-fold brighter than the bead itself. (2) They cannot be used for non-antibody reagents (such as live/dead discriminat-
ing markers) or for reagents that do not bind the capture beads. For reagents that fall under these categories, regular cell-stained compensation must be performed. Note that it is perfectly acceptable to use beads to compensate some colors and cells to compensate other colors in the same experiment.
FACTORS AFFECTING COMPENSATION VALUES In a perfect world, one could determine the emission spectrum for a given dye and use that to calculate the correct compensation value to set on the flow cytometer—never having to run compensation controls in the first place. Unfortunately, this isn’t the case. As shown previously, the spillover coefficients are determined by the ratio of the measured signals in two detectors for a compensation control (a singly stained sample). Therefore, anything that affects the signal level will directly affect the compensation values necessary to correct for the spillover. These include any number of factors, from instrument sensitivity—e.g., how well the optics are aligned and what filters and optical elements are in use—to photomultiplier sensitivity and electronic gains during signal processing. Until instrumentation advances to the point where these factors are internally calibrated and set each time during setup, they must be assumed to have varied from the previous experiment. Thus, compensation controls must be collected in order to determine the correct spillover coefficients for that experiment. A direct corollary of this is that should any of these factors change during an experiment, the compensation will no longer be correct. Therefore, once the compensation values are set, one cannot change PMT voltages, filters, or the optical focusing without risking that subsequent samples will no longer be properly compensated. Further, should one encounter a nozzle clog or other incident that requires any adjustments or change in apparent sensitivity, then one should reanalyze the compensation controls to make sure that the compensation values are set properly. As noted above, the tandems of PE exhibit time-dependent changes in emission spectra with exposure to room light. Therefore, it is critical that the compensation samples and the stained samples be treated identically with regards to light exposure. In general, it is advisable to keep stained cells under cover to block light. This aspect becomes crucial when the compensation samples are generated at a dif-
1.14.16 Supplement 22
Current Protocols in Cytometry
A 1000
Uncompensated
B
Undercompensated
3000
10
2000
103
4
10
PE
FS
100
102
PE (isotype)
1000
74.7 101
1 0
1
2
10 0.1
10
4
10
1
10
2
10
4
10
10
FITC
104
100
3
10
SS
Overcompensated
Correctly compensated
3
19.9
103
PE
10
14.5 2
10
79.6
1 101 0.1 0.1
84.7 1
1
10
100
0.1
1
10
100
CD3-FITC
1000
10
2
10
3
10
4
10
1
10
2
10
3
10
4
10
FITC
Figure 1.14.8 Proper compensation is set when the centers of positive and negative populations align. (A) In a hypothetical experiment, cells stained with CD3-FITC and PE isotype control were collected at different compensation settings. The horizontal line is drawn through the median of the population. The boxes indicate the analysis gates used when the median fluorescences were computed. Proper compensation is achieved when these centers align; note that the properly compensated positive population extends above the top of the negative population (i.e., above where an isotype gate would be set based on the negative population). (B) An example of using antibody capture beads to set compensation. On the top left panel, a forward- and side-scatter gate is drawn tightly around the main population to select only singlet events. The bottom panels show the distributions in the FITC and PE channels for beads labeled with FITC and PE antibodies, respectively; the beads represent a mixture of capture beads with identical but noncapturing beads as the blank. Gates are drawn around the positive and negative bead populations. Software can be used to automatically align the populations, or manually gate settings to align the medians. The top right panel shows a mixture of unlabeled, FITC-, and PE-captured beads. Note that the distribution of the beads is very small, allowing for precise determination of the compensation required. Also note that the distribution in the compensated channel is visually much larger (and similar to that of the blank beads). As shown in Figure 1.14.5, this apparent widening of the distribution is simply a visual artifact of moving the distribution from the bright to the dim area of a logarithmically scaled graph. In this case, because there are many photons being measured for each bead, photon-count- ing statistics has not widened the distribution beyond that of the background distribution. Numbers indicate the percentage of events displayed within each gate.
ferent time than the stained samples. Compensation samples stained on a different day must be carefully stored and judiciously used. Particular care must be taken to make sure that they are appropriate for the experiment. The effect of light exposure is to increase the compensation required between the primary (e.g., PECy5) detector and the PE detector. Note that the compensation required between PE-Cy5 and
APC will not change. The light sensitivity has not been observed for APC tandems.
COMPENSATION MYTHS Because of the subtleties that accompany the compensation process, and affect the visualization of compensated data in non-intuitive ways, it is not surprising that a number of myths have become prevalent regarding compensation. Flow Cytometry Instrumentation
1.14.17 Current Protocols in Cytometry
Supplement 22
Following are a few of the myths, with a short discussion of each one. 1. Compensation is dependent on brightness. In other words, the amount of compensation required depends on how bright a reagent is. This is incorrect. As illustrated in this unit, the degree of compensation is an intrinsic variable that relates to the spectrum of a fluorochrome. In fact, if compensation were brightness related, one would never be able to compensate properly. The source of this myth is that brighter compensation samples appear to require greater compensation settings. However, this is only because of the inability to properly judge proper compensation with the dimmer samples. They appear to be properly compensated when the compensation setting is still too low. This myth can be easily dispelled simply by showing
that a properly compensated bright sample results in proper compensation for dimmer samples as well. This also illustrates why it is important to choose the brightest stain possible for the compensation control. 2. Compensation introduces error. This, too, is incorrect. It is based on the observation that properly compensated data appear to spread out (as shown in Fig. 1.14.5). However, it should be noted that the error is already present in the data; compensation simply makes it more evident in the visualization process. In fact, compensation does not introduce any error, it only makes extant errors more evident. 3. One can not compensate properly with beads, or one can only compensate properly with the same cells that are used in the experiment. Remember that compensation is a property of the fluorochrome, not the cells. There-
1000 100
PE fluorescence
10 1 1000
100 10 1 0.1 1
10
100
1000
1
10
100
1000
Fluorescein fluorescence
Compensation in Flow Cytometry
Figure 1.14.9 Setting appropriate compensation. This is a peripheral blood mononuclear cell (PBMC) sample stained with a PE-conjugated reagent and collected at several different compensation values (into FITC). The top left panel is uncompensated, the top right is undercompensated, the bottom right is overcompensated, and the bottom left is properly compensated. The bold gray lines have been added for emphasis only; no quantitative relationship is implied by the shapes. In this example, the cells have been gated for lymphocytes, so the highly autofluorescent monocytes do not interfere with the setting. Note in particular that while the main PE+ population appears reasonably well compensated in all graphs where some compensation is set, the brightest cells clearly indicate the incorrect compensation level. This is a clear example of why only the brightest stain should be used to estimate correct compensation!
1.14.18 Supplement 22
Current Protocols in Cytometry
fore, proper compensation requires only that one can measure in each of the detectors the amount of fluorescence coming from a given fluorochrome. It is entirely appropriate to use monocytes for one compensation control, lymphocytes for another, and beads for a third. However, for each control sample, the positive and negative (or dim) gates must be applied such that the two gated populations would have the same autofluorescence if the sample were unstained. 4. The compensation value is too high/low [in absolute value], or one can not have compensation settings over 100%. In fact, there is no real meaning to the absolute value of the compensation setting. Remember that changing the PMT voltage will force a change in the compensation value, without changing the actual quality of the measurements in any way. Do not assign importance to the fact that one instrument has a given compensation setting of 30%, whereas another has the same setting as 15%; these absolute values are not comparable. For a given instrument that has been carefully calibrated, compensation values should be roughly comparable from day to day, but that is the extent to which the absolute values can be used. Likewise, there is no magical significance to a value of 100%; compensations of 200% can be used just as well as 2%. 5. One can use the same compensation settings every day. This time-saving approximation is no more than that: an approximation. For experiments where identification of relatively low-expression stains is important, or where more than three or four colors are involved, compensation controls should be generated concurrently with cell staining, and should be used to precisely set proper compensation.
THE PROTOCOL: HOW TO SET COMPENSATION This section is devoted to setting the compensation on the instrument manually. The use of software compensation obviates all these steps by automating the process and ensures that the compensation values are correct. Even when using software compensation, one should pay special attention to the discussion above regarding compensation (see Factors Affecting Compensation Values and Compensation Controls: Cells). To perform software compensation, follow the instructions given by the manufacturer. Proper compensation is set when, on average for a population of cells, there is no contribution of any given fluorophore into each of the
other detectors collecting fluorescence. For example, a population of cells stained with only FITC should have the same median PE fluorescence as would another aliquot of the same cells left unstained. Figure 1.14.8 illustrates this process. Starting with no compensation set, and viewing the dot display for the two parameters being compensated, the appropriate compensation control is slowly increased until the centers of the two populations are equal. Note that this is not necessarily when the tops of the two populations line up! The reader who does not understand this should refer to the section above (see Compensation: Effect on Visualization of Data). It can be difficult to determine when the centers of the populations are equal; once one has achieved approximately the correct setting, one should collect some events, and, using a software analysis package, calculate the medians of the two populations (high and low) to ensure that they are indeed equal. Note that it is preferable to use the median rather than the mean, since the median is a better estimator of the central tendency of a population, especially if any of the events are off-scale on either side. Figure 1.14.8 also illustrates another important facet. Before beginning to adjust any compensation values, one must ensure that the negative population is on-scale sufficiently so that its median is above the axis. Otherwise, it will be impossible to determine when one has set compensation properly. Note that if the positive cells are so bright that to keep them on-scale the voltage must be turned down to the point where negative cells are off-scale at the low end, then one must design a compensation sample having the positive and a dim population, both of which are on-scale. This can be accomplished, for example, by diluting the reagent used in the positive stain by 100-fold or more before staining the cells. Figure 1.14.9 is an illustration to help the reader learn how to view fluorescence plots to decide if compensation is proper or not. By mentally drawing a line through the centers of all the populations, even rare ones, one can often determine if samples are over- or undercompensated. Often in the literature one finds examples like the bottom-right panel in Figure 1.14.9, where the main population appears reasonably well compensated, but in fact the system is significantly overcompensated. Always use the brightest possible cells in determining appropriate compensation. Flow Cytometry Instrumentation
1.14.19 Current Protocols in Cytometry
Supplement 22
Four steps must be taken to ensure proper compensation. These steps should be taken with every experiment. 1. Make compensation controls that consist of cells with positive and negative (or dim) populations. The positive and negative populations must have the same autofluorescence as the negative populations. Use as bright a reagent as possible for each control. 2. Set the PMT voltages high enough that an entirely unstained population is completely off the lower axis for every parameter being measured. 3. Set an analysis gate such that only cells with the identical autofluorescence characteristics are being viewed (e.g., a lymphocyte gate). 4. Increase the compensation setting until the centers of the positive and negative populations (not the upper limits) are equal. Even after taking these steps, caution is still needed. Since most commercial instruments can perform only pairwise compensations, one should make sure that there are no significant uncorrected spillovers between other detector pairs (especially green fluorescence and red fluorescence). Do not change any instrument settings, or else the compensation settings will have to be reset by reanalyzing the compensation controls under the new settings. Finally, remember that different tandem dye lots, such as PE-Cy5 conjugates, can have significantly
different compensation requirements; make sure that the compensation setting is appropriate for all the reagents used in an experiment.
LITERATURE CITED Alberti, S., Parks, D.R., and Herzenberg, L.A. 1987. A single laser method for subtraction of cell autofluorescence in flow cytometry. Cytometry 8:114-119. Bagwell, C.B. and Adams, E.G. 1993. Fluorescence spectral overlap compensation for any number of flow cytometry parameters. Ann. N.Y. Acad. Sci. 677:167-184. Kantor, A. and Roederer, M. 1996. FACS analysis of leukocytes. In Handbook of Experimental Immunology, 5th ed. (L.A. Herzenberg, D.M. Weir, L.A. Herzenberg, and C. Blackwell, eds.) pp. 49.1-49.13. Blackwell Scientific, Cambridge. Loken, M.R., Parks, D.R., and Herzenberg, L.A. 1977. Two-color immunofluorescence using a fluorescence-activated cell sorter. J. Histochem. Cytochem. 25:899-907. Roederer, M. and Murphy, R.F. 1986. Cell-by-cell autofluorescence correction for low signal-tonoise systems: Application to epidermal growth factor endocytosis by 3T3 fibroblasts. Cytometry 7:558-565. Roederer, M. 2001. Spectral compensation for flow cytometry: Visualization artifacts, limitations, and caveats. Cytometry 45:194-205.
Contributed by Mario Roederer Vaccine Research Center, NIAID, NIH Bethesda, Maryland
Compensation in Flow Cytometry
1.14.20 Supplement 22
Current Protocols in Cytometry
Time-Resolved Fluorescence Measurements Time-resolved measurements of excitedstate lifetimes (fluorescence decay times) provide a means to discriminate among fluorescent markers and can be used to study the interaction of fluorescent markers with their cellular targets, with each other, and with the surrounding microenvironment. Time-resolved fluorescence measurements by flow cytometry (FCM) are important because they provide information about fluorophore-cell interactions at the molecular level. An advantage of fluorescence lifetime measurements is that lifetimes can be considered in some instances as absolute quantities. However, the lifetimes of fluorophores bound to cellular macromolecules can be influenced by physical and chemical factors near the binding site, such as solvent polarity, cations, pH, energy transfer, and excited-state reactions. Often such changes are accompanied by a change in the temporal nature of the fluorescence decay (e.g., single-exponential, multiexponential, or nonexponential). Therefore, it is expected that lifetime measurements can be used to probe cellular complexes and subcompartments. Table 1.15.1 lists the lifetimes of fluorescent markers used to measure cellular DNA, RNA, and protein, mitochondria, antibody labeling of cellular antigens, and the lifetime of cellular autofluorescence in Chinese hamster ovary (CHO) cells.
EXCITED-STATE LIFETIME MEASUREMENTS Fluorescence lifetime is defined as the characteristic time, ranging from a few hundredths to hundreds of nanoseconds (nsec), that a fluorophore molecule remains in an excited state prior to returning to the ground state. During the lifetime of the excited state the fluorophore can undergo conformational changes as well as interact with its local environment. If a uniform population of fluorescent molecules is excited with a brief pulse of excitation light, the decay of the fluorescence intensity as a function of time can be described by the exponential function:
I (t ) = I 0 e − t / τ Equation 1.15.1
where I(t) is the intensity measured at time t, I0 is the initial intensity immediately after the Contributed by John A. Steinkamp Current Protocols in Cytometry (2000) 1.15.1-1.15.16 Copyright © 2000 by John Wiley & Sons, Inc.
UNIT 1.15
excitation pulse, and τ is the fluorescence lifetime. The fluorescence lifetime is the time in which the fluorescence intensity decays to 1/e of the initial intensity. The decay of a fluorescent molecule in a uniform solvent is usually monoexponential, whereas the lifetime decay is often multiexponential in fluorophorelabeled cells, where multiple environments exist. A number of other deactivation or energy-depleting processes can compete with fluorescence for return of the excited-state electrons to the ground state. These include internal conversion, phosphorescence, and quenching. Other than fluorescence and phosphorescence, the processes for return of the excited-state electrons to the ground state represent nonfluorescent mechanisms.
Time- and Frequency-Domain Measurements There are two general methods for the timeresolved measurement of fluorescence lifetimes: time-domain analysis and frequency-domain analysis. In the time-domain method, the sample is excited by a series of short light pulses from a laser, a synchrotron radiation source, or a flash lamp, and the time evolution of the fluorescence emission is measured directly by time-correlated, single-photon counting or a high-speed digital-storage oscilloscope/sampling device (Demas, 1983; Lakowicz, 1983; O’Connor and Phillips, 1984). The photon detector is typically a conventional or multichannel-plate photomultiplier or an avalanche photodiode. In the time-correlated, single-photon counting method, the number of photodetector output pulses within a series of discrete sequential time intervals after the excitation pulse are counted and a computer-generated histogram approximating the fluorescence decay curve is obtained. The fluorescence decay time(s) is determined from these data through the use of a computer algorithm. Time-domain methods have been employed in fluorescence microscopy to measure the lifetime of fluorophores bound to cells (Vigo et al., 1987; Dix and Verkman, 1990; Keating and Wensel, 1991). Phase-modulation measurement is a frequency-domain alternative to the time-domain measurement method. Although the time- and frequency-domain methods are functionally equivalent in most aspects, phase-modulation spectroscopy uses different hardware and is
Flow Cytometry Instrumentation
1.15.1 Supplement 11
Table 1.15.1 Examples of Fluorescence Lifetimes and Corresponding Phase Shifts at Various Excitation Frequencies for Fluorochromes Used to Label Cellular Complexes and Cells
Fluorescent dye/compound Hoechst 33342 (DNA)c DAPI (DNA)b,c Acridine orange (cells)d Propidium iodided Propidium iodide (cells)d Ethidium bromidee Ethidium bromide (cells)d Ethidium bromide (DNA)e 7-AAD (DNA)b, f Pyronin Y (cells)d FITCg Fluoresceinh Rhodamine 123 (cells)i Phycoerythrin-avidind Texas Red–avidind CHO cells (autofluorescence)j
Excitation wavelength (nm)
Fluorescence lifetime (nsec)
Phase shifta at 10 MHz (degrees)
Phase shifta at 30 MHz (degrees)
Phase shifta at 50 MHz (degrees)
360 360 480 515 515 515 515 515 515 530 480 480 511 530 530 365
2.6 3.5 3(Gnb), 13(Rdb) 1.2 13.0 1.8 19.0 22.5 0.8 0.6, 2.3 3.6 4.7 2.0, 4.0 3.5 4.6 1.8
9.2 12.3 10.7, 39.2 4.3 39.2 6.5 50.0 54.6 2.9 2.2, 8.2 12.5 16.4 7.2, 14.1 12.6 16.1 6.4
26.1 33.4 29.5, 67.7 12.7 67.8 18.7 74.4 76.7 8.6 6.5, 23.4 34.0 41.5 20.6, 37.0 33.4 40.9 18.7
39.2 47.7 43.3, 76.2 20.6 76.2 29.5 80.5 81.9 14.1 10.7, 35.8 48.5 55.9 32.1, 51.5 47.7 55.3 29.5
aPhase shift equals arctan (ωτ), where ω= 2πφ is the angular frequency and τ is the fluorescence lifetime. bAbbreviations: 7-AAD, 7-aminoactinomycin D; DAPI, 4′, 6-diamidino-2-phenylindole; Gn, green; Rd, red. cLakowicz et al. 1997); dJ. Martin, unpub. observ.; eOlmsted and Kearns (1977); fBailey et al. (1994); gChen and Scott (1985); hSpencer and Weber (1969); iVilleneuve et al. (1996); jKönig et al. (1996).
somewhat more effective in analyzing the emission of multicomponent, exponentially decaying samples (Lakowicz, 1983). In contrast to pulsed techniques, which record the amplitude of the fluorescence decay directly, the phasemodulation method determines the phase and amplitude of the fluorescence emission relative to a periodically modulated excitation source. If the excitation source is sinusoidally modulated, then the phase and amplitude of the fluorescence emission are a function of both the input frequency (ω) and the fluorescence lifetime (τ). For a single-exponential decay lifetime, the phase angle (φ) with respect to a reference signal and the relative modulation index (m) are related to the input frequency by: tan φ = ωτ m = mem / mex = Time-Resolved Fluorescence Measurements
1 1 + (ωτ ) 2
Equation 1.15.2
where m is defined as the ratio of the peak-topeak amplitude of the AC component to the DC (average) level of the fluorescence emission (modulation of fluorescence, mem), divided by the ratio of the peak-to-peak amplitude of the AC component to the DC (average) level of the excitation (modulation of excitation, mex; Spencer and Weber, 1969). The fluorescence emission phase and amplitude, which have the same frequency content as the excitation, track the excitation for frequencies that are small relative to the inverse of the decay time; however, as the frequency is increased, a significant phase delay and decrease in amplitude are observed. By measuring phase and modulation over a range of frequencies, fluorescence lifetime(s) can be readily determined. Phase-modulation methods, even at fixed single-modulation frequency, have a number of advantages: (1) short (subnanosecond) lifetimes are easily measured; (2) phase and modulation measurements are rapid, requiring only seconds of data acquisition; and
1.15.2 Supplement 11
Current Protocols in Cytometry
(3) the technique of phase-sensitive detection of fluorescence increases the usefulness of phase-modulation methods in the analysis of heterogeneous samples. Progress in frequencydomain spectrofluorometric developments has been reviewed by Bright et al. (1990).
Fluorescence Lifetime Image Cytometry Fluorescence lifetime imaging provides a new way to unite functional and structural information for a more complete understanding of cellular processes by adding a new dimension to conventional fluorescence microscopy. Conventional fluorescence measurements are thus enhanced by the addition of lifetime imaging along with spatial lifetime measurement within the cell. The sensitivity of fluorescence lifetime to the microenvironment within the cell makes lifetime imaging useful in measuring numerous cellular features, and lifetime itself can serve as a contrast-enhancing mechanism. The utilization of fluorescence lifetimes enables numerous novel imaging applications that are particularly useful in cell biology. Fluorescence lifetime imaging cytometry has rapidly advanced from the early time-resolved fluorescence microscopy measurements, in which cellular information at selected points inside the cell was obtained. Fluorescence lifetime-resolved measurements were extended to obtain lifetime information across the entire cell using charge-coupled device (CCD) cameras equipped with gain-modulated image intensifiers to collect data simultaneously over the whole image, and using laser scanning microscopes to obtain time-resolved information on a point-by-point basis. Laser scanning microscopy has been used to obtain lifetime images with confocal detection, twophoton excitation, and time-dependent optical mixing. Near-field scanning optical microscopy has also been used to image fluorescence lifetimes. Both frequency-domain and time-domain techniques have been developed to implement time-resolved imaging of the full field simultaneously using a CCD, a gated image intensifier coupled to a camera, or a positionsensitive photomultiplier. A recent review by French et al. (1998) summarizes the developments of fluorescence lifetime imaging techniques for cytometry.
Fluorescence Lifetime Flow Cytometry Fluorescence lifetime flow cytometers (FLFCMs) have been described that are capable of measuring lifetimes by a number of
methods including phase shift using real-time analog phase-sensitive detection (Pinsky et al., 1993; Steinkamp et al., 1993); amplitude demodulation using a high-speed digital-storage oscilloscope followed by computer processing to determine lifetimes (Deka et al., 1994) and real-time analog electronics (Steinkamp et al., 1998); phase shift using a high-speed digital oscilloscope to record the data, followed by analysis using the fast Fourier transform (Beisker and Klocke, 1997); and phase shift and amplitude demodulation of heterodyned fluorescence emission signals acquired by a computer-controlled data acquisition system for subsequent lifetime calculation (Durack et al., 1998). The flow cytometric technology for combined fluorescence lifetime measurements by phase shift on fluorophore-labeled particles and on the surrounding fluorophore solution also has been described (Steinkamp and Keij, 1999b). Time-domain fluorescence lifetime measurements by flow cytometry have also been accomplished using a high-speed digitalstorage oscilloscope to record the fluorescence decay signals from fluorochrome-labeled cells excited with a pulsed laser (Deka and Steinkamp, 1996). Lifetimes were then calculated from stored data by interactive reconvolution analysis. Although analog signal-processing methods are currently more suitable for making frequency-domain lifetime measurements on a cell-by-cell basis in flow in real time, highspeed digital signal processing of fluorescence emission waveforms in real time is technically feasible and is under development (Parson et al., 1994). Fluorescence lifetimes can also be used to resolve overlapping spectral emissions from fluorescent probes based on their differences expressed as phase shifts using phase-sensitive detection (Lakowicz and Cherek, 1981). The flow cytometric resolution of signals from fluorescence emissions by phase-sensitive detection was first demonstrated on cells stained with propidium iodide (PI) and fluorescein isothiocyanate (FITC) for total cellular DNA and protein content, respectively (Steinkamp and Crissman, 1993). Although the PI and FITC fluorescence emission signals are readily separable by conventional FCM methods, they were separated electronically using a single photomultiplier tube (PMT), a long-pass (barrier) filter to block scattered laser excitation light, and two phase-sensitive detection channels, one for PI and the other for FITC. In addition, background interferences (e.g., autofluorescence, unbound fluorophores, nonspecific
Flow Cytometry Instrumentation
1.15.3 Current Protocols in Cytometry
Supplement 11
modulated fluorescence signal (frequency ω)
cell stream
Vem (max)
fluorescence detector
Vem (min) t
0 laser monitor
beam splitter
phase shift (φ) Vex (max)
0
Vex (min) t
0
t reference signal
driver signal
modulator
to signalprocessing electronics/ data acquistion system
laser beam
synthesized sine wave generator (frequency ω)
Figure 1.15.1 Conceptual diagram of the single-frequency fluorescence lifetime flow cytometer illustrating the laser excitation beam, modulator, modulated laser beam, beam splitter, laser monitor, cell sample stream, the cell stream/laser beam intersection point in the flow chamber, fluorescence detector, modulated fluorescence emission and reference signals, and synthesized sine-wave signal generator. The fluorescence signals and reference signal are input to (1) analog signal-processing electronics to give conventional FCM signals, fluorescence lifetimes, and phase-resolved fluorescence signals prior to acquisition by a computer-based data acquisition system, or (2) a data-acquisition system for digital signal processing. The acquired data are displayed as frequency distribution histograms and bivariate contours/dot plots.
fluorophore labeling, and Rayleigh scatter) may be reduced or eliminated by phase-sensitive detection (Steinkamp et al., 1997, 1999a; Steinkamp and Keij, 1999a).
THEORY OF FREQUENCYDOMAIN (SINGLE-FREQUENCY) LIFETIME MEASUREMENTS APPLIED TO FLOW CYTOMETRY Fluorochrome-labeled cells are analyzed as they pass across an optically focused continuous-wave (CW) laser beam with a Gaussianshaped intensity profile that is intensity modulated using a high-frequency (ω) sine wave (Fig. 1.15.1). The modulated laser excitation intensity, E(t), is expressed as: E (t ) = E0 [1 + m ex sin(ωt )] Equation 1.15.3
Time-Resolved Fluorescence Measurements
where E0 is the CW laser excitation intensity, mex is the excitation depth of modulation term, and t is time. The time-dependent fluorescence
emission signal, v(t), is a Gaussian-shaped, modulated pulse that results from the passage of the cell across the laser beam. It can be expressed in an approximate form as
v(t ) = V [1 + mem sin(ωt − φ s )] e
− a 2 ( t − t0 ) 2
Equation 1.15.4
where V is the signal intensity, φs and mem are the respective signal phase shift and modulation of fluorescence terms associated with a single fluorescence decay time (τ), and a is a term related to the laser beam height and to the velocity of a cell crossing the laser beam at time t0 (Zarrin et al., 1987). An exact derivation of this relationship has been given by Deka et al. (1994).
Conventional FCM Measurements The CW-excited (DC), low-frequency signal component, v(t)Lf, is extracted using lowpass filtering to remove the high-frequency signal component to give conventional FCM
1.15.4 Supplement 11
Current Protocols in Cytometry
fluorescence-intensity information expressed as
v(t ) Lf = Ve
− a 2 ( t − t0 ) 2
and illustrated below. The high-frequency modulated sine-wave signal component
obtained by band-pass filtering (center frequency f = ω/2π), is shifted in phase by an amount φ s = arctan(ωτ )
Vmem cos(φ s − φ R ) e
− a 2 ( t − t0 )2
2
where φR has been set to zero using light scatter signals (lifetime equal to zero). The phase-sensitive detection outputs are divided on a cellby-cell basis, which results in the vφ−90(t)/vφ(t) ratio expression. The ratio equals tanφs, which is directly proportional to the fluorescence decay time expressed as τ=
− a 2 ( t − t0 ) 2
Vmem cos( φ s ) e
Equation 1.15.9
Equation 1.15.7
relative to the excitation frequency and demodulated by a factor mem, and is processed by phase-sensitive detection to quantify fluorescence lifetime (as a parameter) and resolve fluorescence emission signals based on differences in lifetimes (expressed as phase shifts; Blair and Sydenham, 1975; Meade, 1982). The phase-sensitive detection process (by analog or digital signal-processing methods) multiplies the fluorescence emission signal (Equation 1.15.4) or the band-pass filtered fluorescence signal (Equation 1.15.6) by a sinewave reference signal, which is phase shifted (φR) with respect to the fluorescence signal on a cell-by-cell basis. The multiplied signals are filtered (low-pass) to remove the high-frequency signal components and the resulting outputs, vo(t), are Gaussian-shaped signals expressed as
− a 2 ( t − t0 ) 2
2
and vφ (t ) =
− a 2 ( t − t 0 )2
Equation 1.15.6
vo ( t ) =
Vmem sin( φ s ) e
vφ − 90 (t ) =
Equation 1.15.5
v(t ) = Vmem sin(ωt − φs ) e
(Meade, 1982), as illustrated below using analog signal processing. The two low-pass-filtered phase-sensitive detection outputs are:
tan φ s ω
=
[vφ − 90 / vφ ] ω
Equation 1.15.10
In addition to measuring the fluorescence lifetime by phase-sensitive detection, the corresponding signal intensity can be measured by taking the square root of the sum of the square of the vφ−90(t) and vφ(t) expressions in Equation 1.15.9 (Meade, 1982). Fluorescence lifetime may also be determined by measuring the relative depth of amplitude modulation (m) of the emission signal (mem) with respect to the excitation signal (mex). The relative modulation, or demodulation, factor m is determined from the ratio m=
mem mex
=
= cos φ s =
modulation of fluorescence modulation of excitation 1 1 + (ωτ )
2
2 Equation 1.15.11 Equation 1.15.8
Fluorescence Lifetime Measurements Fluorescence lifetime is measured by the two-phase phase comparator method using two reference sine-wave signals, 90 degrees out of phase with each other, that are each multiplied by the fluorescence emission signal from Equation 1.15.4 or Equation 1.15.6, and low-pass filtered to remove the high-frequency (ω) terms
In the steady-state system it is only necessary to measure the AC and DC fluorescence emission and excitation signal components and determine the relative modulation by ratio calculations (Spencer and Weber, 1969). In flow cytometry, the amplitude demodulation factor can be determined by measuring the maximum and minimum signal components at the peak height of the Gaussian-shaped fluorescence detector output signal and the AC and DC com-
Flow Cytometry Instrumentation
1.15.5 Current Protocols in Cytometry
Supplement 11
fluorescence signal (multifrequency content: m • ω, where m = 1, 2, 3...)
cell stream fluorescence detector
0
0
t
time
laser excitation monitor
beam splitter
t
to time- or frequency-domain measurement system for resolving fluorescence lifetime(s)
laser excitation signal
pulse train from a pulsed laser excitation source (frequency ω)
Figure 1.15.2 Conceptual diagram of a time- or frequency-domain fluorescence-lifetime flow cytometer illustrating the laser excitation pulse train, beam splitter, laser monitor, cell sample stream, the cell stream/laser beam intersection point in the flow chamber, fluorescence detector, and fluorescence and laser monitor signals. The fluorescence and laser excitation signals are input to a time-domain or multifrequency measurement system for resolving single- and multiexponential decay lifetimes.
ponents of the steady-state laser-excitation monitor signal (see Fig. 1.15.1), either by digital (Deka et al., 1994) or analog (Steinkamp et al., 1998) methods. The relative modulation is then expressed as
V V m=
max
− Vmin
max
+ Vmin
em
V V
max
− Vmin
max
+ Vmin
ex
Equation 1.15.12
The real-time ratio results in cosφs, which is proportional to the fluorescence decay lifetime expressed as τ=
−1 tan(cos m )
ω
Equation 1.15.13
Time-Resolved Fluorescence Measurements
in value. For modulation frequencies <0.5 MHz, the CW-excited, low-frequency signal component interferes with the 0.5 MHz highfrequency signal. The shortest measurable lifetime depends on the maximum highest modulation frequency usable and the bandwidth of the signal detection/processing electronics and data acquisition system. Lifetime measurement capabilities by FCM of a few tenths of nanoseconds have been reported (Pinksy et al., 1993; Steinkamp et al., 1993).
The longest lifetime that can presently be measured in phase-modulation flow cytometry depends on the lowest usable excitation frequency, which is ∼0.5 MHz. This corresponds to a 318-nsec lifetime calculated at a 45° phase shift, but in practice will be somewhat higher
Analysis of Heterogeneous Fluorescence Decays The above equations are derived on the assumption of a single-component exponential decay of fluorescence from a homogeneous emitting fluorophore population. This is often cited as the major shortcoming of the singlefrequency method, because the existence of a unique single-component decay is presupposed, but not demonstrated by the measurement. This is indeed true if only one of the two quantities—i.e., lifetime by phase shift or by amplitude demodulation—is measured. However, if both are measured, the existence of an exponential can be demonstrated. In the heterogeneous fluorophore population, the lifetime measured by the degree of amplitude demodulation will almost always be larger than the weighted average of the individual component
1.15.6 Supplement 11
Current Protocols in Cytometry
lifetime values, whereas the lifetime determined by phase shift will always be shorter than the weighted average (Spencer and Weber, 1969). It is only when there is a single exponential decay that both methods give the same result. Also, phase-shift measurements at two or more frequencies can be used to detect heterogeneous fluorescence decays (Deka et al., 1995). If the decay is multiexponential, measurement by time-domain methods (Demas, 1983; Lakowicz, 1983; O’Connor and Phillips, 1984) or by multifrequency methods over a wide range of frequencies (Jameson et al., 1984; Lakowicz et al., 1984) is required. Examples include using the multiharmonic frequency content from the pulse train of a modelocked laser (Alcala et al., 1985; Gratton and Barbieri, 1986; Laczko et al., 1990) as illustrated in Figure 1.15.2, or using a pulsed laser and time-resolved methods to quantify multiexponential decays from fluorochromelabeled cells by FCM (Deka and Steinkamp, 1996).
Phase-Resolved Separation of Fluorescence Emission Signals As applied to flow cytometry, the principle of phase suppression for separating two fluorescence emission signals having different lifetimes (i.e., phase shifts) is based on the theory of Veselova et al. (1970) for individually recording fluorescence spectra in systems containing two luminescent centers, and also on the phase-sensitive fluorescence spectroscopy work of Lakowicz and Cherek (1981). By use of the phase-sensitive detection process, two superimposed signals, each similar to Equation 1.15.4 or Equation 1.15.6 but having different phase shifts, are multiplied by a sine-wave reference signal (φR), and are low-pass filtered to remove the high-frequency signal components. The phase-sensitive detection output results in the expression vo ( t ) = +
Vm 1 em1 cos( φ 1 − φ R ) e
− a 2 ( t − t0 ) 2
2 V2 mem2 cos(φ 2 − φ R ) e
− a 2 ( t − t0 ) 2
2 Equation 1.15.14
where V1 and V2 are the signal intensities, mem1 and mem2 are the fluorescence modulation factors, and φ1 and φ2 are the phase shifts that result
when a cell stained with two fluorochromes, each having a different lifetime τ1 and τ2, is excited by a single modulated source. To resolve either of the two signals, the reference phase is shifted by an amount equal to π/2 + φ1 or −π/2 + φ2 degrees. This results in one signal being passed and the other being nulled. For example, if the reference phase is adjusted to equal −π/2 + φ2 degrees, the detector output is expressed as vo ( t ) =
V1 mem1 sin(φ 2 − φ1 ) e
− a 2 ( t − t0 ) 2
2 Equation 1.15.15
Similarly, if the reference phase is adjusted to equal π/2 + φ1 degrees, the output is expressed as vo ( t ) =
V2 mem2 sin( φ 2 − φ1 ) e
− a 2 ( t − t0 ) 2
2 Equation 1.15.16
When fluorescence signals are processed by two phase-sensitive detectors operating in parallel, contributions to the total fluorescence signal can be resolved by setting one detector reference to p/2 + f1 degrees and the other detector reference to -p/2 + f2 degrees (Steinkamp and Crissman, 1993).
FLFCM INSTRUMENTATION AND SIGNAL PROCESSING Ramsey and Hieftje (1983) have made a detailed study of the relative merits of various modulation and signal-processing schemes based on signal-to-noise ratio in shot noise limited luminescence measurement systems. Their conclusions were that high signal-tonoise ratios are achieved by (1) maximizing the measured radiant photon flux and observation time, (2) maximizing the ratio of the experimental bandwidth to the signal detector bandwidth, and (3) minimizing the number of frequency components in the excitation function. These criteria are best realized in flow cytometry by utilizing a single-frequency excitation source coupled with high-numerical-aperture fluorescence collection optics and bandlimiting electronic filtering of the fluorescence signals as described below. Flow Cytometry Instrumentation
1.15.7 Current Protocols in Cytometry
Supplement 11
cell stream
laser
modulator beam splitter
drive electronics
light scatter detector fluorescence detector modulated fluorescence conventional signal FCM signals
laser monitor driver signal
laser excitation signal
φ
phase shift 0
RF signal synthesizer
modulated forward flow chamber light scatter signals to low-pass filter lens
PS
reference signal
0
signal detection/ processing electronics
fluorescence lifetimes phase-resolved fluorescence signals
time
Figure 1.15.3 System block diagram of the single-frequency fluorescence-lifetime flow cytometer illustrating the laser, modulator/drive electronics, beam splitter, laser monitor, flow chamber, forward light scatter and fluorescence detectors, synthesized sine-wave signal generator, and signal detection/processing electronics. PS, power splitter; RF, radio frequency.
Laser Excitation Modulation and Fluorescence/Light Scatter Signal Detection
Time-Resolved Fluorescence Measurements
A laser serves as the excitation source, and the modulator is either a Pockels cell (an electro-optic device that utilizes the Pockels or Kerr effect) or an acousto-optic device (Fig. 1.15.3). Pulsed and mode-locked lasers also can be used as excitation sources for multifrequency and time-domain measurements. A low-phasenoise radio frequency (RF) signal synthesizer is used as the sine-wave generator for the modulator drive electronics and as the reference frequency source for homodyne signal detection. A second RF signal synthesizer (phase-locked to the first synthesizer) is required for heterodyne signal detection. A photomultiplier tube (PMT) or high-speed photodiode, with the anode connected to a high-speed operational amplifier (Op Amp) configured in the transimpedance mode, serves as a laser excitation monitor. The modulated laser beam is focused onto the cell stream in a flow chamber or in an emerging liquid jet from a flow nozzle of a flow cytometer. The modulated fluorescence emission and orthogonal light-scatter signals are detected by PMTs configured with high-speed Op Amps (transimpedance mode), which serve as preamplifiers. Forward-scattered light from
cells is also focused onto the photocathode of a PMT/preamplifier detector. The velocity of fluorophore-labeled cells crossing the laser beam and the dimensions of the focused Gaussian-shaped beam determine the time duration of modulated fluorescence and light-scatter signals. In the author’s phasesensitive flow cytometer (Steinkamp et al., 1993), fluorescence and light-scatter signals are 12 to 15 µsec in duration, which corresponds to an approximate bandwidth of 160 to 180 kHz. Other flow cytometers may have shorter or longer signal pulse durations and thus different bandwidths. Therefore, based on the signal pulse duration, the low-pass electronic filtering (analog or digital) bandwidth must be selected to pass the low-frequency signals of Equations 1.15.5, 1.15.9, 1.15.15, and 1.15.16 (for obtaining conventional FCM, fluorescence lifetime, and phase-resolved fluorescence measurements on fluorophore-labeled cells and particles) and reject the high-frequency, modulated signal components.
Homodyne Versus Heterodyne Signal Detection Signal homodyning relies on direct measurement of the phase shift by multiplying the modulated fluorescence signal of Equation 1.15.4 or Equation 1.15.6 with a sine wave or
1.15.8 Supplement 11
Current Protocols in Cytometry
inputs
A
modulated fluorescence emission signals
0
B
t
fluorescence signals
amplifier/ integrator or log amplifier
low-pass filter
0
modulated fluorescence emission signals phase-sensitive detector band-pass filter reference φref signal
C
outputs: to data acquisition system
multiplier
φ ref reference signal
0
t
cos(ωt-φ R) phase shifter
modulated fluorescence emission signals band-pass filter
phase-resolved fluorescence signals amplifier/ integrator or log amplifier
low-pass filter
t
PS
Vssinφs signal
V(φ– 90) phase-sensitive detector no. 1
0
amplifier/ integrator
cos(ωt – 90) quadrature phase hybrid
fluorescence lifetime signals ratio module
0
cos(ωt) amplifier/ integrator
phase-sensitive detector no. 2 V(φ)
t
t
Vscosφs signal
0
t
Figure 1.15.4 Block diagram of the analog signal-processing electronics for obtaining (A) conventional FCM signals by low-pass filtering, (B) phase-resolved signals by phase-sensitive detection, and (C) fluorescence lifetimes by the two-phase ratio method.
other suitable reference signal of the same frequency (or harmonics thereof), using any number of electronic devices (such as a double-balanced mixer). This is conceptually the simplest form of signal processing, and is performed using analog or digital methods. The main disadvantages are susceptibility to noise interference, requirements for high-frequency precision electronics, loss of phase resolution at high frequencies due to limited resolution of variable time delays, and processing normally implemented in the analog mode for high-frequency measurements. To avoid the need for high-frequency signalprocessing electronics and to better isolate the signal from noise interference, a frequency heterodyning technique has been developed for static spectrofluorometric frequency-domain lifetime measurements (Spencer and Weber, 1969) and is adaptable to flow cytometers (Pin-
sky and Ladasky, 1994; Durack et al., 1998). This technique works by mixing the fluorescence signal, at the detector PMT base or at an external mixer, with a second signal of different frequency—i.e., frequency heterodyning. The resulting difference frequency contains the same information as the original high-frequency modulated signal, but the difference can be set to any suitable lower value to suit the measurement conditions—e.g., signal-processing (digital) speed. In addition, frequency heterodyning can be used with either single- or multifrequency excitation schemes.
Analog and Digital Signal-Processing Approaches The analog signal-processing electronics illustrated in Figure 1.15.4 can be used with either the homodyne or heterodyne signalprocessing scheme. Fluorescence and light-
Flow Cytometry Instrumentation
1.15.9 Current Protocols in Cytometry
Supplement 11
Time-Resolved Fluorescence Measurements
scatter signals are demodulated using a lowpass electronic filter set at a bandwidth (e.g., 180 kHz) to obtain the conventional, CW-excited laser excitation signal(s) and reject the modulation frequency, followed by an amplifier/integrator or logarithmic amplifier (Fig. 1.15.4A). The phase-sensitive detection electronics (single-channel) for resolving fluorescence emission signals based on differences in lifetimes consists of a band-pass electrical filter having a center frequency corresponding to the laser modulation frequency, a passive multiplier, and a low-pass electronic filter with a bandwidth of 180 kHz (Fig. 1.15.4B). Switchable nsec delay lines are used to shift the phase of the reference signal with respect to the modulated fluorescence signal input. The phase-sensitive detector output signals (including logarithmic signals) are amplified or integrated. The two-phase ratio detector for making fluorescence lifetime measurements is shown in Figure 1.15.4C. A quadrature phase-hybrid module supplies two reference sine-wave signals, 90° out of phase with each other, to two phasesensitive detector (PSD) circuits for generating outputs V(φ − 90) and V(φ), the ratio of which is directly proportional to the lifetime. The conventional light scatter and fluorescence, phase-resolved fluorescence, and lifetime signals are recorded as listmode data for display as frequency distribution histograms or as bivariate dot/contour diagrams using a computerbased data acquisition system. By replacing the analog processing electronics with a computer-controlled digital signal-processing and acquisition system and heterodyne signal detection, lifetime data can be processed by software (Feddersen et al., 1989). The digital acquisition technology provides new capabilities for measuring excited-state lifetimes by both time- and frequency-domain methods. High-speed, computer-controlled digital-storage oscilloscopes, followed by postprocessing of acquired data, have been employed in flow cytometers for making frequency-domain (homodyne signal detection; Deka et al., 1994; Beisker and Klocke, 1997) and time-domain fluorescence lifetime measurements (Deka and Steinkamp, 1996). Direct processing of phase-modulation signals acquired by heterodyne signal detection has been recently reported (Durack et al., 1998). Modern high-speed digital signal-processing methods (Stearns and Hush, 1990; Marven and Ewers, 1996) offer a new approach for making excitedstate lifetime measurements on a cell-by-cell basis in real time.
Instrument Initialization For Lifetime Measurements The ability to quantify fluorescence decay times on cells labeled with fluorescent probes by direct phase-shift measurement in flow cytometry is illustrated below. The outputs of PSDs 1 and 2 (i.e., sinφs and cosφs output of Fig. 1.15.4C) are first initialized by removing the long-pass barrier filter in the fluorescence detector, adjusting the reference phase shift (φR) to zero (null), and maximizing the sinφs and cosφs output signals, respectively, using nonfluorescent microspheres. The barrier filter is then replaced and fluorescently labeled particles of known lifetime (e.g., Flow Check Fluorospheres, Coulter; lifetime ∼7.0 nsec) are analyzed at the same PMT and PSD amplifier/integrator gain settings. The ratio module gain is adjusted to center the Flow Check microspheres histogram, typically in channel 70 or 140, prior to analyzing labeled cell samples at fixed gain settings. Neutral density filters are used in the fluorescence detector to compensate for differences in light-scatter (nonfluorescent particles) and fluorescence (Flow Check microspheres and labeled cells) signal intensities when required to maintain the PMT voltage constant during lifetime measurement.
RECENT APPLICATIONS OF FLFCM Fluorescence Lifetime Measurements Figure 1.15.5A shows the autofluorescence lifetime histogram measured on cultured viable human lung fibroblasts (HLFs) suspended in phosphate-buffered saline (PBS) using a 488nm laser excitation wavelength, a 29-MHz modulation frequency, and an OG515 longpass (Melles Griot)/530-nm band-pass (Omega Optical) filter combination in the fluorescence detector. Autofluorescence lifetime histograms measured on ethanol-, methanol-, paraformaldehyde-, and formaldehyde-fixed HLFs are slightly increased compared to unfixed HLFs (data not shown). The problem this presents is that the broadened lifetime histogram coefficients of variation (CVs; standard deviations divided by the means) of viable and fixed cells partially overlap the lifetimes of FITC labeling of antibody probes, thus making it difficult to obtain phase-resolved fluorescence measurements to eliminate autofluorescence background in immunofluorescently labeled cells based on lifetime differences. To alleviate this problem, an approach in which low-concentration glutaraldehyde is used as a fixative for
1.15.10 Supplement 11
Current Protocols in Cytometry
A
autofluorescent human lung fibroblasts (HLFs)
C
DNA-binding fluorochromes (labeled ethanol-fixed CHO cells)
UV/29 MHz 325
1.7 nsec
Hoechst 33342
500
2.3 nsec
0 0
Number of cells
B
10 5 immunofluorescence markers (mouse anti-Thy-1.2)
0
1100
anti-Thy-1.2-FITC
488 nm/6 MHz propidium iodide
14.6 nsec
450 2.5 nsec
0
0
1500
488 nm/6 MHz ethidium bromide
anti-Thy-1.2-PE
450
19.5 nsec
1.3 nsec 0
0 514 nm/29 MHz anti-Thy-1.2-PE/Texas Red
450
0.5 nsec
1.8 nsec
0 0
5
1100
0 10 0 5 Fluorescence lifetime (nsec)
10
7–AAD
15
20
Figure 1.15.5 Fluorescence-lifetime frequency distribution histograms. (A) Cultured viable human lung fibroblasts (HLFs) suspended in phosphate-buffered saline (PBS). (B) Murine thymus cells labeled with anti-Thy-1.2 antibody conjugated to FITC, phycoerythrin (PE), and PE/Texas Red, and suspended in PBS according to the procedure in Steinkamp et al. (1999b). (C) Ethanol-fixed (70%) CHO cells stained with the DNA-binding fluorochromes Hoechst 33342 (0.5 µg/ml), propidium iodide (15.0 µg/ml), ethidium bromide (15.0 µg/ml), and 7-aminoactinomycin D (7-AAD; 12.7 µg/ml). Fixed CHO cells were treated with RNase prior to staining with propidium iodide and ethidium bromide.
phase-resolved, cell-surface immunofluorescence measurements has been developed. Compared to viable HLFs and HLFs fixed with the other fixatives, glutaraldehyde-fixed cells have considerably shorter fluorescence lifetimes, have smaller lifetime histogram CVs, and allow phase-resolved fluorescence measurement of FITC-labeled antibody probes (Steinkamp and Keij, 1999a). The measurement of fluorescence lifetimes on immunofluorescence cell-surface markers
to murine thymus cells labeled with the fluorophores FITC, phycoerythrin (PE), and the PE/Texas Red tandem fluorochrome conjugated to anti-Thy-1.2 antibody (denoted antiThy-1.2-FITC and so on) and suspended in PBS is illustrated in Figure 1.15.5B using a 488-nm excitation wavelength, a 29 MHz modulation frequency, and an OG515 longpass filter in the fluorescence detector. The modal channel values of the lifetime histograms are less than the expected values listed
Flow Cytometry Instrumentation
1.15.11 Current Protocols in Cytometry
Supplement 11
A
cell sample
to forward light scatter detector
modulated fluorescence signals
fluorescence band-pass detector filter
anti-Thy-1.2PE/Texas Red signals π/2 +φ (PI)
PSD no.1
φref1 RF signal generator
– π2 + φ(PE/Texas Red) phase shifter
φref2
driver signal modulator
PI signals
PSD no.2
laser beam
Phase-sensitive detector no. 1 output [φref1 = π/2 + φ (PI)]
B 900
Number of cells
PE/Texas Red– labeled thymus cells (controls)
nulled thymus cells
0 Phase-sensitive detector no. 2 output [φ REF2 = –π/2 + φ(PE/Texas Red)]
90 PI-labeled thymus cells (controls)
nulled thymus cells
0
0
255 0
255
Phase-resolved, log fluorescence intensity
Figure 1.15.6 Phase-resolved, log fluorescence analysis of PE/Texas Red and PI signals from murine thymus cells labeled with anti-Thy-1.2-PE/Texas Red or PI (“dead cells”; Steinkamp et al., 1999b) by two phase-sensitive detector (PSD) channels operating in parallel (A). In part (B), the reference phase shift of PSD channel no. 1 (φref1) was first adjusted [φref1 = π/2 + φ (PI)] to null signals from thymus control cells stained with PI and to pass cells labeled with anti-Thy-1.2PE/Texas Red. The reference phase of PSD channel no. 2 (φref2) was next adjusted [φref2 = −π/2 + φ(PE/Texas Red)] to null signals from thymus control cells labeled with anti-Thy-1.2-PE/Texas Red and to pass cells stained with PI.
Time-Resolved Fluorescence Measurements
in Table 1.15.1 (antibody solution measurements) and suggest fluorescence quenching. Studies using microspheres labeled with varying numbers of FITC molecules showed lifetime values ranging from 2.0 to 3.8 nsec (Steinkamp et al., 1996). To further demonstrate the FITC self-quenching phenomenon as a function of antibody labeling dilution and fluores-
cence-to-protein (F/P) ratio, fluorescence lifetime histograms were recorded on murine thymus cells labeled with anti-Thy-1.2-FITC having F/P ratios of 13.7 (maximum quenching), 8.7, and 4.5 (minimum quenching; Deka et al., 1996). Figure 1.15.5C shows the fluorescence lifetime histograms measured on CHO cells la-
1.15.12 Supplement 11
Current Protocols in Cytometry
anti-Thy-1.2PE/Texas Red labeled cells
Number of cells
1125
0 PSD output no.2 φ ref2 = –π/2 + φ(PE/Texas Red) 263
B
64
unlabeled Plpositive cells
PSD output no.2 (log Pl signal intensity)
PSD output no. 1 φ ref1 = π/2 + φ (PI)
A
0 Pl-positive cells
anti-Thy-1.2PE/Texas Red and PI-positive labeled cells anti-Thy-1.2PE/Texas Red labeled cells (PI-negative)
0
64 PSD output no.1 (log PE/Texas Red signal intensity)
0 0
255 Phase-resolved, log fluorescence intensity
Figure 1.15.7 (A) Phase-resolved, log fluorescence frequency distribution histograms recorded from the PSD no. 1 and no. 2 outputs of Figure 1.15.6A on murine thymus cells labeled with anti-Thy-1.1-PE/Texas Red and suspended in PBS containing PI, and (B) the corresponding bivariate contour diagram.
beled with DNA-specific fluorochromes using the excitation wavelengths and modulations frequencies shown in the figure and the following long-pass filters: GG400 (Melles Griot; Hoechst 33342), OG515 (PI and ethidium bromide), and OG550 (Melles Griot; 7-aminoactinomycin D). The peak modal channel lifetime value of the histogram recorded for Hoechst 33342 (A-T base-pair specificity) is slightly shorter than the reported value listed in Table 1.15.1. The lifetime values for PI and ethidium bromide (DNA intercalation, lacking base pair specificity) and 7-aminoactinomycin D (G-C base-pair preference) are essentially the same as those given in Table 1.15.1. The interactions of DNA-binding fluorochromes on the labeling of cells in the presence of deuterium oxide (Sailer et al., 1996, 1997a), the monitoring of the uptake of ellipticine and its fluorescence lifetime in relation to the cell cycle phase (Sailer et al., 1997b), the flow cytometric analysis of DNA-binding probes (Sailer et al., 1998a), and the measurement of differential changes in the fluorescence lifetime of DNA-bound ethidium bromide in apoptotic cells (Sailer et al., 1998b) all illustrate recent applications describing the capability to precisely measure the lifetimes of
DNA-labeling fluorophores and the changes that result from their binding to DNA.
Phase-Resolved Fluorescence Measurements An example illustrating phase-resolved measurements on murine thymus cells labeled with anti-Thy-1.2-PE/Texas Red and suspended in PBS containing PI (for labeling PIpositive damaged/dead cells) is shown in Figures 1.15.6 and 1.15.7, using a 488-nm laser excitation wavelength, a 10 MHz modulation frequency, an OG515 long-pass filter in the fluorescence detector, and two PSD channels operating in parallel. Since the fluorescence emission spectra of PE/Texas Red and PI completely overlap (Steinkamp et al., 1999b), the separation of PE/Texas Red and PI signals cannot be achieved using conventional methods employing electronic compensation (Loken et al., 1977). The PI labeling intensity is also ∼42 times greater than the anti-Thy-1.2-PE/Texas Red fluorescence. The measured fluorescence lifetime histograms of cells labeled individually with PE/Texas Red and PI are well separated (Fig. 1.15.5). Based on the lifetimes of PE/Texas Red (1.8 nsec) and PI (14.5 nsec), murine thymus cells were labeled separately
Flow Cytometry Instrumentation
1.15.13 Current Protocols in Cytometry
Supplement 11
with anti-Thy-1.2-PE/Texas Red and PI (controls), and the phase shifts of the two phase-sensitive detector channels (Fig. 1.15.6A) were adjusted to (1) null PE/Texas Red signals in PSD channel 2 output, and (2) null PI signals in PSD channel 1 output as illustrated in Figure 1.15.6B. Thymus cells labeled with anti-Thy1.2-PE/Texas Red and suspended in PBS containing PI were then analyzed at the same reference phase-shift and gain settings, and the phase-resolved histograms (Fig. 1.15.7A) and corresponding bivariate contour diagram (see Fig. 1.15.7B) were recorded. Approximately 89% of the total cell population was labeled only with anti-Thy-1.2-PE/Texas Red, 6% were only PI positive, 2% were positive for both PE/Texas Red and PI, and 3% were unlabeled null cells. These results demonstrate the ability to resolve highly overlapping fluorescence emissions differing in intensity by >40-fold based on differences in lifetimes. Along with the murine thymus cells labeled with anti-Thy-1.2-PE/Texas Red and stained with PI as described above, rat thymus cells were also labeled with the anti-Thy-1.1-PE antibody and suspended in PBS containing PI (partially overlapping emission spectra) for discriminating “dead cells,” and analyzed by phase-resolved methods to identify cells that were PI positive only, PE and PI positive, and PE positive only (Steinkamp and Keij, 1999b). Phase-sensitive detection has been used to eliminate autofluorescence from lung fibroblasts labeled with a cell-surface FITC-labeled antibody (Steinkamp and Keij, 1999a), and viable cells labeled with Hoechst 33342 and monobromobimane have been analyzed using phase-sensitive flow cytometry to determine relative DNA and glutathione content, respectively (Keij et al., 1999).
SUMMARY
Time-Resolved Fluorescence Measurements
Time-resolved fluorescence measurements by flow cytometry is so new that many of the potential applications have not been fully explored or developed. This technology will (1) add a new dimension to multiparameter flow cytometric analyses through the development of techniques for measuring the fluorescence lifetime of probes bound to macromolecular complexes in cells, (2) increase the range of fluorescent markers that can be used in multilabeling applications, and (3) reduce background interferences and thus enhance measurement precision to yield more accurate results. In the past, procedures were limited in some cases by the availability of fluorescent
markers with common excitation regions (so that a single laser excitation source could be used) and emission spectra that were sufficiently separated using optical color-separating filters. Because the lifetime-based sensing technology can separate fluorescence emissions electronically (and also optically), quantify fluorescence lifetimes directly, and make conventional flow cytometric measurements, it has a wide range of technically possible applications. The technology will significantly expand the researcher’s understanding of biological processes at the cellular, subcellular, and molecular levels. The author envisions that, through clinical and biomedical research, it will contribute to improving diagnoses and treatment, and to understanding the underlying mechanisms of human diseases. In addition, the technology can be adapted to commercial flow cytometry systems, where it can be used for virtually any clinical or research application involving the analysis of cells, cell function, or subcellular components through the use of fluorescent markers directed to specific targets.
LITERATURE CITED Alcala, J.R., Gratton, E., and Jameson, D.M. 1985. A multifrequency phase fluorometer using the harmonic content of a mode-locked laser. Anal. Instrum. 14:225-250. Bailey, S.A., Graves, D.E., and Eftink, M.R. 1994. Interaction of 7-aminoactinomycin D with single and double stranded oligonucleotides. Proc. SPIE 2137:462-468. Beisker, W. and Klocke, A. 1997. Fluorescence lifetime measurement in flow cytometry. Proc. SPIE 2982:436-446. Blair, D.P. and Sydenham, P.H. 1975. Phase sensitive detection as a means to recover signals buried in noise. J. Phys. E: Sci. Instrum. 8:621-627. Bright, F.V., Betts, T.A., and Litwiler, K.S. 1990. Advances in multifrequency phase and modulation fluorescence analysis. Anal. Chem. 21:389405. Chen, R.F. and Scott, C.H. 1985. Atlas of fluorescence spectra and lifetimes of dyes attached to protein. Anal. Lett. 18:393-421. Deka, C. and Steinkamp, J.A. 1996. Time-resolved fluorescence-decay measurement and analysis on single cells by flow cytometry. Appl. Opt. 35:4481-4489. Deka, C., Sklar, L.A., and Steinkamp, J.A. 1994. Fluorescence lifetime measurements in a flow cytometer by amplitude demodulation using digital data acquisition technique. Cytometry 17:94-101. Deka, C., Cram, L.A., Habbersett, R., Martin, J.C., Sklar, L.A., and Steinkamp, J.A. 1995. Simultaneous dual-frequency phase-sensitive flow cytometric measurements for rapid identification
1.15.14 Supplement 11
Current Protocols in Cytometry
of heterogeneous fluorescence decays in fluorochrome-labeled cells and particles. Cytometry 21:318-328. Deka, C., Lehnert, B.E., Lehnert, N.M., Jones, G.M., Sklar, L.A., and Steinkamp, J.A. 1996. Analysis of fluorescence lifetime and quenching of FITC-conjugated antibodies on cells by phase-sensitive flow cytometry. Cytometry 25:271-279.
Lakowicz, J.R., Laczko, G., Cherek, H., Gratton, E., and Limkeman, M. 1984. Analysis of fluorescence decay kinetics from variable-frequency phase shift and modulation data. Biophys. J. 46:463-477.
Demas, J.N. 1983. Excited State Lifetime Measurements. Academic Press, New York.
Lakowicz, J.R., Gryczynski, I., Malak, H., Schrader, M., Englehardt, P., Kano, H., and Hell, S.W. 1997. Time-resolved fluorescence spectroscopy and imaging of DNA labeled with DAPI and Hoechst 33342 using three-photon excitation. Biophys. J. 72:567-578.
Dix, J.A. and Verkman, A.S. 1990. Pyrene eximer mapping in cultured fibroblasts by ratio imaging and time-resolved microscopy. Biochemistry 29:1949-1953.
Loken, M.R., Parks, D.R., and Herzenberg, L.A. 1977. Two-color immunofluorescence using a fluorescence activated cell-sorter. J. Histochem.Cytochem. 25:899-907.
Durack, G., Yu, W., Mantulin, W., and Gratton, E. 1998. Fluorescence lifetime flow cytometry: A phase-sensitive detection system utilizing a cross correlation technique and digital signal processing. Cytometry Suppl. 9:39.
Marven, C. and Ewers, G. 1996. A Simple Approach to Digital Signal Processing. John Wiley & Sons, New York.
Feddersen, B.A., Piston, D.W., and Gratton, E. 1989. Digital parallel acquisition in frequency domain fluorometry. Rev. Sci. Instrum. 60:29292936.
O’Connor, D.V. and Phillips, D. 1984. Time-Correlated Single Photon Counting. Academic Press, London.
French, T., So, P.T.C., Dong, C.Y., Berland, K., and Gratton, E. 1998. Fluorescence lifetime imaging techniques for microscopy. Methods Cell Biol. 56:277-304. Gratton, E. and Barbieri, B. 1986. Multifrequency phase fluorometry using pulsed sources: Theory and applications. Spectroscopy 1:28-36. Jameson, D.M., Gratton, E., and Hall, R.D. 1984. The measurement and analysis of heterogeneous emissions by multifrequency phase and modulation fluorometry. Appl. Spectrosc. Rev. 20:55106. Keating, S.M. and Wensel, T.G. 1991. Nanosecond fluorescence microscopy: Emission kinetics of Fura-2 in single cells. Biophys. J. 59:1186-202. Keij, J.F., Bell-Prince, C., and Steinkamp, J.A. 1999. Simultaneous analysis of relative DNA and glutathione content in viable cells using phase-sensitive flow cytometry. Cytometry 35:48-54. König, K., So, P.T.C., Mantulin, W.W., Tromberg, B.J., and Gratton, E. 1996. Two-photon excited lifetime imaging of autofluorescence in cells during UVA and NIR photostress. J. Microsc. 183:197-204. Laczko, G., Gryczynski, I., Gryczynski, Z., Wiczk, W., Malak, H., and Lakowicz, J.R. 1990. A 10GHz frequency-domain fluorometer. Rev. Sci. Instrum. 61:2331-2337. Lakowicz, J.R. 1983. Principles of Fluorescence Spectroscopy. Plenum, New York. Lakowicz, J.R. and Cherek, H. 1981. Resolution of heterogeneous fluorescence from proteins and aromatic amino acids by phase-sensitive detection of fluorescence. J. Biol. Chem. 256:63486353.
Meade, M.L. 1982. Advances in lock-in amplifiers. J. Phys. E: Sci. Instrum. 15:395-403.
Olmsted, J. and Kearns, D.R. 1977. Mechanism of ethidium bromide fluorescence enhancement of binding to nucleic acids. Biochemistry 16:36473654. Parson, J.D., Deka, C., Habbersett, R.C., Martin, J.C., Naivar, M.A., Steinkamp, J.A., Wilder, M.E., and Jett, J.H. 1994. Digital signal processing of phase sensitive flow cytometry signals. Cytometry Suppl. 6:71. Pinsky, B.G. and Ladasky, J.J. 1994. Heterodyning of modulated pulses for fluorescence lifetime measurements in flow cytometry. Proc. SPIE 2137:794-799. Pinsky, B.G., Ladasky, J.J., Lakowicz, J.R., Berndt, K., and Hoffman, R.A. 1993. Phase-resolved fluorescence lifetime measurements for flow cytometry. Cytometry 14:123-135. Ramsey, J.M. and Hieftje, G.M. 1983. Signal-tonoise considerations in fluctuation analysis spectroscopic techniques. In New Directions in Molecular Luminescence (D. Eastwood, ed.) pp. 82-100. American Society for Testing and Materials, Philadelphia. Sailer, B.L., Nastasi, A.J., Valdez, J.G., Steinkamp, J.A., and Crissman, H.A. 1996. Interactions of intercalating fluorochromes with DNA analyzed by conventional and fluorescence lifetime flow cytometry utilizing deuterium oxide. Cytometry 25:164-172. Sailer, B.L., Nastasi, A.J., Valdez, J.G., Steinkamp, J.A., and Crissman, H.A. 1997a. Differential effects of deuterium oxide on the fluorescence lifetimes and intensities on dyes with different modes of binding to DNA. J. Histochem. Cytochem. 45:165-175.
Flow Cytometry Instrumentation
1.15.15 Current Protocols in Cytometry
Supplement 11
Sailer, B.L., Valdez, J.G., Steinkamp, J.A., Darzynkiewicz, Z., and Crissman, H.A. 1997b. Monitoring uptake of ellipticine and its fluorescence lifetime in relation to the cell cycle by flow cytometry. Exp. Cell Res. 236:259-267. Sailer, B.L., Steinkamp, J.A., and Crissman, H.A. 1998a. Flow cytometric fluorescence lifetime analysis of DNA-binding probes. Eur. J. Histochem. 42:19-28. Sailer, B.L., Valdez, J.G., Steinkamp, J.A., and Crissman, H.A. 1998b. Apoptosis induced with different cycle perturbing agents produces differential changes in the fluorescence lifetime of DNA-bound ethidium bromide. Cytometry 31:208-216. Spencer, R.D. and Weber, G. 1969. Measurements of subnanosecond fluorescence lifetimes with a cross-correlation phase fluorometer. Ann. N.Y. Acad. Sci. 158:361-376. Stearns, S.D. and Hush, D.R. 1990. Digital Signal Analysis. Prentice-Hall, Englewood Cliffs, N.J. Steinkamp, J.A. and Crissman, H.A. 1993. Resolution of fluorescence signals from cells labeled with fluorochromes having different lifetimes by phase-sensitive flow cytometry. Cytometry 14:210-216. Steinkamp, J.A. and Keij, J.F. 1999a. Elimination of light scatter interference in dual-laser flow cytometry by synchronous detection of emitted fluorescence: Theory and demonstration using simulated signals. Proc. SPIE 3604:170-176. Steinkamp, J.A. and Keij, J.F. 1999b. Fluorescence intensity and lifetime measurement of free and particle-bound fluorophore in a sample stream by phase-sensitive flow cytometry. Rev. Sci. Instrum. 70:4682-4688. Steinkamp, J.A., Yoshida, T.M., and Martin, J.C. 1993. Flow cytometer for resolving signals from heterogeneous fluorescence emissions and quantifying lifetime in fluorochrome-labeled cells/particles by phase-sensitive detection. Rev. Sci. Instrum. 64:3440-3450. Steinkamp, J.A., Deka, C., Lehnert, B.E., and Crissman, H.A. 1996. Fluorescence lifetime as a new parameter in analytical cytology measurements. Proc. SPIE 2678:221-230. Steinkamp, J.A., Lehnert, B.E., and Keij, J.F. 1997. Phase-sensitive detection as a means to recover
fluorescence signals from interfering backgrounds in analytical cytology measurements. Proc. SPIE 2982:447-455. Steinkamp, J.A., Parson, J.D., and Keij, J.F. 1998. Progress towards combined phase and amplitude demodulation fluorescence lifetime measurements by flow cytometry. Proc. SPIE 3260:236244. Steinkamp, J.A., Lehnert, N.M., Keij, J.F., and Lehnert, B.E. 1999a. Enhanced immunofluorescence measurement resolution of surface antigens on highly autofluorescent, glutaraldehydefixed cells by phase-sensitive flow cytometry. Cytometry. 37:275-283. Steinkamp, J.A., Lehnert, B.E., and Lehnert, N.M. 1999b. Discrimination of damaged/dead cells by propidium iodide uptake in immunofluorescently labeled populations analyzed by phasesensitive flow cytometry. J. Immunol. Methods 226:59-70. Veselova, T.V., Cherkasov, A.S., and Shirokov, V.I. 1970. Fluorometric method for individual recording of spectra in systems containing two types of luminescent centers. Opt. Spectrosc. 29:617-618. Vigo, J., Salmon, J.M., and Viallet, P. 1987. Quantitative microfluorometry of isolated living cells with pulsed excitation: Development of an effective and relatively inexpensive instrument. Rev. Sci. Instrum. 58:1433-1438. Villeneuve, L., Pal, P., Durocher, G., Migneault, D., Girard, D., Giasson, R., Balassy, A., Blanchard, L., and Gaboury, L. 1996. Spectroscopic and photophysical investigations on the nature of localization of rhodamine-123 and its dibromo derivative in different cell lines. J. Fluorescence 6:209-219. Zarrin, F., Bornhop, D.J., and Dovichi, N.J. 1987. Laser doppler velocimetry for particle size determination by light scatter within the sheath flow cuvette. Anal. Chem. 59:854-860.
Contributed by John A. Steinkamp Bioscience Division, Los Alamos National Laboratory Los Alamos, New Mexico
This work was performed at the Los Alamos National Laboratory under the joint sponsorship of the United States Department of Energy and the Los Alamos National Flow Cytometry Resource (National Institutes of Health Grants P41-RR013150 and R01-RR07855). The author wishes to thank Nancy M. Lehnert and Harry A. Crissman for their assistance in cell preparation and staining.
Time-Resolved Fluorescence Measurements
1.15.16 Supplement 11
Current Protocols in Cytometry
Simultaneous Analysis of the Cyan, Green, and Yellow Fluorescent Proteins
UNIT 1.16
The detection of fluorescent proteins in cells allows one to monitor gene expression, determine the intracellular localization of proteins, and identify transfected cells; the technique also has many other applications. When flow cytometry is used to detect the fluorescence signals, data from a large number of cells can be acquired, thereby improving the statistical accuracy of the quantitation. This makes the ability to analyze fluorescent protein–expressing cells by flow cytometry of particular interest to many types of researchers. Moreover, the recent introduction of novel proteins that exhibit different spectral properties allows multiple processes to be monitored in a single cell. While some fluorescence emissions are difficult to separate effectively by microscopy, the ability to compensate electronically in real time for the spectral overlap of the fluorescent signals makes flow cytometry an ideal method to detect expression of different fluorescent proteins in a single cell. SIMULTANEOUS ANALYSIS OF CYAN, GREEN, AND YELLOW FLUORESCENT PROTEINS
BASIC PROTOCOL
This protocol describes a method to detect and separate the fluorescent signals from three proteins (EGFP, EYFP, and ECFP) in transfected U-2 OS cells. Laser excitation at 458 nm is required. Each 15-cm plate of nearly confluent U-2 OS cells yields enough cells for ∼2 transfections. Each transfection produces two plates that can be harvested and analyzed together or separately. Eight transfections are required to produce a negative control as well as single, double, and triple fluorescent protein–expressing cells. Negative control 1. pSP72 Single-color controls 2. EGFP 3. EYFP 4. ECFP Dual-color controls 5. EGFP+EYFP 6. EGFP+ECFP 7. EYFP+ECFP Triple-color sample 8. EGFP+EYFP+ECFP. In addition, cells from individual transfections can be mixed prior to analysis to create unique combinations of cells individually expressing different fluorescent proteins. Mix A: transfections 2 and 3 (EGFP or EYFP) Mix B: transfections 2 and 4 (EGFP or ECFP) Mix C: transfections 3 and 4 (EYFP or ECFP) Mix D: transfections 2, 3, and 4 (EGFP or EYFP or ECFP). Contributed by Andrew J. Beavis and Robert F. Kalejta Current Protocols in Cytometry (2001) 1.16.1-1.16.12 Copyright © 2001 by John Wiley & Sons, Inc.
Flow Cytometry Instrumentation
1.16.1 Supplement 16
It is advisable that the negative control and single-transfected cells (transfections 1 through 4) be assessed in a separate, preliminary experiment to confirm the instrument setup prior to examining more complex samples (mixes A through D or transfections 5 through 8). However, these types of controls will have to be performed again when analyzing the double- and triple-labeled cells to ensure proper instrument setup. NOTE: All solutions and equipment coming into contact with living cells must be sterile, and aseptic technique should be used accordingly. NOTE: All culture incubations should be performed in a humidified 37°C, 5% CO2 incubator unless otherwise specified. Materials U-2 OS human osteosarcoma cells (ATCC) Complete medium with serum: DMEM plus 10% FBS, 100 U/ml penicillin, 100 µg/ml streptomycin Plasmids: pEGFPN-1, pEYFPN-1, and pECFPN-1 (Clontech) pSP72 (Promega) 1× PBS (Life Technologies or APPENDIX 2A) Hemacytometer Refrigerated centrifuge 0.4-cm electrode gap cuvettes BioRad gene pulser with capacitance extender 10-cm cell culture dishes 12 × 75–mm (3-ml) flow cytometry tubes Flow cytometer with 458-nm excitation and custom filter set for detection of three fluorescent proteins (e.g., FACSVantage; see Support Protocol) Additional reagents and equipment for trypsinizing cells (APPENDIX 3B) Transfect cells (day 1) 1. Trypsinize subconfluent plates of U-2 OS cells (APPENDIX complete DMEM medium with serum.
3B)
and collect cells in
2. Determine the number of cells/ml by counting with a hemacytometer (APPENDIX 3A) and adjust the concentration to 5 × 106 cells/ml by centrifuging the cells 5 min at 500 × g, 4°C and resuspending in the appropriate volume of complete medium with serum. 3. Mix 5 µg of each plasmid to be transfected with 700 µl cell suspension and place the mixture in a 4-cm electrode gap electroporation cuvette. 4. Apply a single pulse (960 µF, 220 V) using the BioRad gene pulser. 5. Remove the cells from the cuvette, dilute to 20 ml with serum-containing medium, and plate 10 ml/plate on two 10-cm cell culture dishes. Incubate ∼24 to 48 hr at 37°C, 5% CO2 in a humidified incubator. Harvest cells (day 2) 6. Approximately 24 to 48 hr after transfection, harvest cells with trypsin (APPENDIX 3B) and collect in complete medium. 7. Centrifuge cells from a single dish 5 min at 500 × g, 4°C and resuspend in 1 ml 1× PBS by pipetting up and down to create a single-cell suspension. Store on ice prior to analysis (<8 hr) in 12 × 75–mm (3-ml) flow cytometry tubes. Simultaneous Analysis of Three Fluorescent Proteins
Analyze single fluorescent protein–expressing cells 8. Analyze the cells using a flow cytometer equipped with a custom optical filter set for detection of three fluorescent proteins (see Support Protocol).
1.16.2 Supplement 16
Current Protocols in Cytometry
9. For control cells transfected with pSP72 (transfection 1), electronically gate the single cells from the FS/SS dot plot onto single-parameter histograms and bivariate plots of the fluorescent protein signals, i.e., EGFP versus ECFP; EGFP versus EYFP; ECFP versus EYFP. It is important to use mock-transfected (as opposed to untransfected) cells as the control since the transfection procedure utilized may result in changes in light scatter and autofluorescence. Also, be aware that light-scatter signals, particularly side scatter, may change in cells expressing high levels of fluorescent proteins.
10. Using logarithmic amplification, adjust the PMT detector sensitivity to position the autofluorescence signal for each parameter in the first log decade. 11. Determine the fluorescence signal for each fluorescent protein (Fig. 1.16.1) using single-color control cells that express each fluorescent protein separately (transfections 2 to 4). Set the electronic compensation as required (UNIT 1.14). An example of instrument settings for a FACSVantage SE is shown in Table 1.16.1.
12. Mix aliquots of the single-color samples (transfections 2 to 4) to make mixes A to C (cells individually expressing different pairs of the fluorescent proteins EGFP or ECFP, EGFP or EYFP, and ECFP or EYFP) and analyze using the current parameter settings. Verify that electronic compensation is set correctly so that the single-color positive cells are clearly resolved (Fig. 1.16.2). If required, position histogram markers and dot-plot quadrant markers for real-time statistics display during acquisition.
Analyze two and three fluorescent protein–expressing cells 13. If data for mock-transfected and single fluorescent protein–expressing cells have not been acquired (but only previously done in a pilot experiment), perform analysis again. Next, acquire data for the three samples co-transfected with pairs of the expression plasmids (transfections 5 to 7). Verify that the dual-color cells can be detected and resolved (Fig. 1.16.3). 14. Acquire data for samples expressing all three fluorescent proteins (mix D and transfection 8). Acquire a sufficient number of cells such that the subpopulation occurring at the lowest frequency will have enough events for accurate statistical analysis (Fig. 1.16.4). If required, cells may be purified by cell sorting (UNITS 1.1, 1.3, & 1.7).
Analyze data 15. Construct a dot plot for FS/SS and identify the single cells using electronic gating. Be sure to eliminate the possibility of doublets as they will produce errors in quantitating the cells expressing multiple fluorescent proteins.
16. Gate the cells onto single-parameter histograms. Create a histogram gate for each parameter to define the cells positive for expression of each fluorescent protein (Fig. 1.16.1). 17. Create Boolean gating logic (Table 1.16.2) to define the possible subpopulations expressing the different combinations of fluorescent proteins and assign a specific color to each subpopulation. Create three bivariate plots, gated on the FS/SS region, to display each two-color combination of fluorescent proteins using the multi-color mode. The exact position of any subpopulation of cells expressing a given combination of fluorescent proteins on any two-color plot can be visualized by the appearance of the dots of the corresponding color (Fig. 1.16.4). Flow Cytometry Instrumentation
1.16.3 Current Protocols in Cytometry
Supplement 16
300 ECFP
240 180 R3 120 60 0 100
101
102
103
104
300 EGFP
Frequency
240 180 R2
120 60 0 100
101
102
103
104
300 EYFP
240 180 R4 120 60 0 100
Simultaneous Analysis of Three Fluorescent Proteins
1.16.4 Supplement 16
101 102 103 Log fluorescence intensity
104
Figure 1.16.1 Histogram analysis of U-2 OS cells expressing EGFP, EYFP, or ECFP. U-2 OS cells were transfected with pECFPN-1 (ECFP), pEGFPN-1 (EGFP), or pEYFPN-1 (EYFP), cultured for 48 hr, and analyzed for fluorescence on a FACSVantage using an excitation wavelength of 458 nm at 50 mW. The ECFP fluorescence emission was separated with a 500 LPDi and collected as the cyan fluorescence parameter with a 480/30 BP filter. The EGFP (green fluorescence) and EYFP (yellow fluorescence) signals were separated with a 525 SPDi and collected with a 510/20 BP and a 550/30 BP, respectively. For each analysis, the 458-nm LP laser-blocking filter was placed in front of the BP filter. The single-parameter histograms indicate log fluorescence intensity on the x-axis and frequency or number of cells on the y-axis. Gating regions were drawn on each histogram to identify the cells that were positive for fluorescence for ECFP (R3), EGFP (R2), and EYFP (R4) compared to a mock-transfection control (not shown). Current Protocols in Cytometry
Table 1.16.1 Example FACS Vantage Instrument Settings for the Simultaneous Analysis of Three Fluorescent Proteins Expressed by U-2 OS Cells Using Single-Laser (458-nm) Flow Cytometry
Compensationb
Parametera
Fluorescent protein
PMT voltage (V) (-% Green fluorescence)
(-% Yellow fluorescence)
(-% Cyan fluorescence)
Green fluorescence Yellow fluorescence Cyan fluorescence
EGFP EYFP ECFP
525 525 600
44.6% — 2.1%
48.6% 24.7% —
— 50.7% 42.3%
aParameter settings were adjusted for cells expressing each of the fluorescent proteins to position fluorescence signals on scale. bCompensation was set using CellQuest software (version 3.3). This allows for the spectral overlap of the green fluorescence and cyan
fluorescence signals to be adjusted independently of the yellow fluorescence parameter.
104
A
104 103 EYFP
EYFP
103 102
101
102 103 EGFP
104
100 100
104
104
103
103 ECFP
ECFP
102 101
101 100 100
102
101
102 103 EGFP
104
100 100
104
103
103 ECFP
ECFP
102 103 EGFP
104
101
102 103 EGFP
104
101
102 103 EYFP
104
102
104
102
102 101
101 100 100
101
101
101 100 100
B
101
102 EYFP
103
104
100 100
Figure 1.16.2 Bivariate analysis of (A) uncompensated and (B) compensated data for mixed samples of U-2 OS cells expressing two fluorescent proteins. Mixed samples of cells expressing a single fluorescent protein were analyzed for fluorescence as described. Panel A shows dual-parameter dot plots illustrating the spectral overlap of EGFP, EYFP, and ECFP fluorescence (uncompensated data). When compensation is applied, as in panel B, the fluorescence emissions of these fluorescent proteins can be resolved. Current Protocols in Cytometry
Flow Cytometry Instrumentation
1.16.5 Supplement 16
104
EYFP
103 102 101 100 100
101
102
103
104
103
104
103
104
EGFP 104
ECFP
103 102 101 100 100
101
102 EGFP
104
ECFP
103 102 101 100 100
101
102 EYFP
Figure 1.16.3 Bivariate analysis of compensated data for cotransfected samples of U-2 OS cells expressing two fluorescent proteins. Two-color analysis (compensated) of cells co-transfected with expression plasmids for pairs of the fluorescent proteins indicates that cells expressing any two of the three fluorescent proteins simultaneously can be resolved from the single-positive cells for each pair of fluorescent proteins that are expressed in combination.
Simultaneous Analysis of Three Fluorescent Proteins
1.16.6 Supplement 16
Current Protocols in Cytometry
A
104
104
102 101
102
101
102 103 EGFP
104
100 100
102 101
101
100 100
C
103
103 ECFP
EYFP
103
B
ECFP
104
101
102 103 EGFP
104
100 100
101
102 103 EYFP
104
Figure 1.16.4 Simultaneous analysis of three fluorescent proteins expressed in U-2 OS cells using single-laser excitation at 458 nm. U-2 OS cells cotransfected with pEGFPN-1, pEYFPN-1, and pECFPN-1 were cultured for 48 hr and analyzed for fluorescence on a FACSVantage using an argon-ion laser tuned to 458 nm at 50 mW. Compensation for spectral overlap was set with CellQuest software (version 3.3) using cells expressing each fluorescent protein individually. This allows for the spectral overlap of the green fluorescence and cyan fluorescence signals to be adjusted independently of the yellow fluorescence parameter. The dual-parameter dot plots for (A) EGFP versus EYFP; (B) EGFP versus ECFP; and (C) EYFP versus ECFP are shown. The possible subpopulations can be identified by the multicolor gating analysis: EGFP(+)/ECFP(−)/ EYFP(−) = 9.5% [green]; EGFP(−)/ECFP(+)/EYFP(−) = 12.1% [cyan]; EGFP(−)/ECFP(−)/EYFP(+) = 2.6% [yellow]; EGFP(+)/ECFP(+)/EYFP(−) = 13.1% [magenta]; EGFP(+)/ECFP(−)/EYFP(+) = 2.1% [orange]; EGFP(−)/ECFP(+)/EYFP(+) = 2.5% [dark blue]; EGFP(+)/ECFP(+)/EYFP(+) = 11.5% [red]. See color plate.
Table 1.16.2 Boolean Gating Logic for Multicolor Mode Analysis of Three-Fluorescent Protein Data
Color
Label
Definitiona
White Green Cyan Yellow Magenta Orange Dark blue Red
G1 G2 G3 G4 G5 G6 G7 G8
R1 R1 and R2 and (not R3 and R4) R1 and R3 and (not R2 and R4) R1 and R4 and (not R2 and R3) R1 and R2 and R3 and (not R4) R1 and R2 and R4 and (not R3) R1 and R3 and R4 and (not R2) R1 and R2 and R3 and R4
aSingle cells are identified with region 1 (R1) from a dot plot of FS versus
SS (not shown). These cells are gated onto single-parameter histograms and the cells positive for expression of each fluorescent protein EGFP, ECFP, or EYFP can be identified with regions R2, R3, and R4, respectively. The regions can be combined using Boolean gating logic to create equations that define the subpopulations of cells that express one, two, or all three fluorescent proteins.
Flow Cytometry Instrumentation
1.16.7 Current Protocols in Cytometry
Supplement 16
SUPPORT PROTOCOL
CUSTOM CYTOMETER SETUP FOR DETECTION OF THREE FLUORESCENT PROTEINS This protocol allows the investigator to modify the optical configuration of a flow cytometer to separate and acquire the three fluorescent signals. This provides the ability to perform any experimental application requiring the analysis and quantitation of the expression of multiple fluorescent proteins on a flow cytometer with only routine, minor adjustments (Beavis and Kalejta, 1999). Filters should be mounted 24 hr in advance to allow time for adhesive to dry. Materials Methanol Suitable adhesive Fluorescent alignment particles Optical filters (Omega Optical) 500-nm LP dichroic 525-nm SP dichroic 480/30 BP 510/20 BP 550/30 BP 458 nm LP laser block 458/10 nm BP Lens paper Dichroic mirror mounts Band-pass filter holders Flow cytometer with tunable argon-ion laser (e.g., FACSVantage) Laser power meter 1. Clean all new optical filters using the “drop-and-drag” method. Fold a piece of lens paper in half, place on top of the optical filter, and dispense 1 or 2 drops of methanol onto the paper. Drag the paper over the optical filter to clean it. The dry section of paper will remove any excess methanol. It is important not to soak the paper or to get any solvents on the side of cavity filters.
2. Mount the dichroic filters onto the respective mirror mounts in the correct orientation (reflective surface away from the holder). To do this, place the dichroic filter with the reflective surface facing upwards onto the mirror mount and then place a small drop of suitable adhesive at three points around the side of the filter. Allow the adhesive to dry before using the filter. Band-pass filters may be mounted in their holders the same way or inserted into blank filter holders, depending upon the cytometer. Label all materials appropriately. The exact filter holders and dichroic mirror mounts will vary depending upon manufacturer and model of cytometer used.
3. Tune the primary excitation beam to 458 nm by adjusting the high reflector prism/mirror at the rear of the tunable argon-ion laser. Monitor the output wavelength by using a wavelength meter or by measuring the power output for each specific wavelength. Consult the laser manufacturer’s guide for the specific procedure for tuning the laser. Simultaneous Analysis of Three Fluorescent Proteins
1.16.8 Supplement 16
Current Protocols in Cytometry
EGFP green fluorescence
yellow fluorescence
550/30 BP
510/20 BP
525 SP dichroic
EYFP 480/30 BP
500 LP dichroic
cyan fluorescence
ECFP 458/10 BP
458 LP dichroic
SS
SIDE SCATTER
= photomultiplier tube = optical filter
Figure 1.16.5 Optical configuration for separation and detection of ECFP, EGFP, and EYFP fluorescence emissions. The emissions of the three fluorescent proteins are collected simultaneously. The ECFP fluorescence is separated with a 500-nm LP dichroic filter. This steers the ECFP fluorescence to the cyan fluorescence detector where it is collected with a 480/30 BP filter. The EGFP and EYFP signals pass through the 500-nm LP dichroic and are separated by the 525-nm SP dichroic filter. The EYFP signal is directed towards the yellow fluorescence detector where it is collected with a 550/30 BP, whereas the EGFP signal is collected by the green fluorescence detector with a 510/20 BP filter. A 458-nm LP laser-blocking filter is placed in front of the BP filters for the green fluorescence, yellow fluorescence, and cyan fluorescence photomultiplier tubes to eliminate any scattered laser light from the detector. If the side scatter signal (SS) is to be acquired from the 458-nm beam, a 458-nm LP dichroic is used to reflect the scattered laser light to the SS detector where it is collected with a 458/10 BP. This configuration conforms to the optical pathway used on standard FACSVantage cell sorter and with minor modifications should be applicable for use on other cytometers.
Flow Cytometry Instrumentation
1.16.9 Current Protocols in Cytometry
Supplement 16
4. Place the optical filters in the cytometer for separation and detection of the three fluorescent protein emissions. a. For a standard FACSVantage configuration, replace the 610 short-pass (SP)/half mirror assembly with the 500 long-pass (LP) dichroic and replace the 550 SP dichroic with the 525 SP dichroic. b. Replace the fluorescence 1 (green), fluorescence 2 (yellow), and fluorescence 3 (cyan) band-pass (BP) filters with the 510/20 BP, 550/30 BP, and 480/30 BP, respectively. c. Place a 458-nm LP laser-blocking filter in front of each band-pass for green fluorescence, yellow fluorescence, and cyan fluorescence. d. Replace the forward scatter (FS) and side scatter (SS) 488/10 BP filters with the 458/10 BP filters for collection of the 458-nm laser scatter signals. See Figure 1.16.5 for optical configuration. 5. Verify proper alignment using fluorescent particles that can be excited at 458 nm. COMMENTARY Background Information
Simultaneous Analysis of Three Fluorescent Proteins
The green fluorescent protein (GFP; reviewed in Tsien, 1998) has become a powerful tool for a multitude of applications in both mammalian and microbiological systems, including assessment of gene expression, protein trafficking, or localization and general cell labeling (Flach and Silver, 1994; Wang and Hazelrigg, 1994; Cole et al., 1996; Valdivia et al., 1996; Cormack et al., 1997; Okabe et al., 1997). The wild-type GFP (wtGFP) has a complex excitation spectrum with major and minor absorption peaks at 395 nm and 475 nm, respectively, but exhibits a single emission peak at 510 nm (Heim et al., 1994). The fluorescence of wtGFP can be observed by spectroscopy, fluorometry, and fluorescence microscopy, but its utility in flow cytometry is limited because its spectral properties do not match well with the optical configuration of most standard flow cytometers. However, site-directed mutations of the GFP fluorophore have yielded additional GFP molecules with enhanced spectral properties (Li et al., 1997; Wachter et al., 1998; Cubitt et al., 1999). One of the most useful mutants is the “red-shifted” GFP variant that results from a serine to threonine substitution at amino acid 65 and possesses increased extinction coefficient and quantum yield compared to wtGFP when excited at 488 nm (Heim et al., 1995; Cormack et al., 1996). This enhanced GFP (EGFP) is now widely used in flow cytometry for many different applications (Kalejta et al., 1997, 1999; Jiang and Hunter, 1998; Galbraith et al., 1999; Pestov et al., 1999). Recently, additional EGFP variants have been con-
structed that have excitation and emission maxima shifted to different regions of the visible spectrum while retaining high quantum yields. Two such mutants are the enhanced cyan fluorescent protein (ECFP) and the enhanced yellow fluorescent protein (EYFP). This unit presents a method for the simultaneous detection of ECFP, EGFP, and EYFP in living cells using single-laser excitation at 458 nm (Beavis and Kalejta, 1999). The Basic Protocol is applicable for use on flow cytometers equipped with a tunable multiline argon-ion laser and should increase the utility of these fluorescent proteins for all types of biological applications.
Critical Parameters The Basic Protocol describes a method to transfect U-2 OS cells with three fluorescent proteins and analyze them by flow cytometry. This cell type is commonly used for many applications, grows well, transfects easily, and yields single-cell suspensions upon trypsinization. However, individual applications will require employment of the appropriate cell line for the experiments. Thus, the proper method (electroporation, calcium-phosphate, lipid reagents, etc.), conditions, amount of DNA transfected, and time of analysis after transfection should be optimized for each individual cell type and experiment. These parameters can be defined empirically. Post-acquisition data analysis can be performed in a variety of ways, depending upon the exact aim of the experiment and the available software (CHAPTER 10).
1.16.10 Supplement 16
Current Protocols in Cytometry
Troubleshooting Transfection efficiency can be monitored by fluorescence microscopy to verify the expression of each of the fluorescent proteins and to estimate the percentage of positive cells. Expression should also be monitored by microscopy and flow cytometry to determine the optimal time after transfection to analyze the cells. It is difficult to generate single-cell suspensions of some cell types because the cells stick together and form clumps. This can often be avoided by adding protein (e.g., 0.5% BSA or 2% FBS) to the PBS used to resuspend the cells prior to analysis. If clumps of cells are still a problem, they can be removed by filtering the samples using a suitably sized mesh (e.g., 20to 50-µm size screen) and/or an in-line filter on the cytometer sample-uptake line.
Anticipated Results Typical results for the detection of ECFP, EGFP, and EYFP are shown in Fig. 1.16.4. All three fluorescent protein signals are resolved using the custom optical configuration, and spectral overlap could be compensated in real time. The number of positive cells and the level of fluorescence detected for each fluorescent protein will depend upon several factors, such as transfection efficiency, time of analysis after transfection, and the physical (half-life, cellular localization, etc.) and fluorescent properties (http://www.clontech.com/gfp/excitation.html) of the proteins. The excitation wavelength of 458 nm is optimal for ECFP but not EGFP (excitation maximum 490 nm) nor EYFP (excitation maximum 513 nm). Thus, there may be limitations, particularly when using EYFP, and in such cases it may be necessary to match the fluorescent protein to the expected expression level of that protein. When transfection is performed with multiple plasmids (e.g., EGFP and EYFP), not every cell that receives one plasmid will receive the other. Thus, while many cells will be double positive, some will be positive for only one of the fluorescent proteins. To help ensure that all the cells that receive one plasmid (e.g., EGFP) receive the other (e.g., EYFP), the cells can be transfected with molar excesses of one of the pair (Kalejta et al., 1997; e.g., 2 µg EGFP and 10 µg EYFP).
Time Considerations The adhesives used to fix the dichroic mirrors in place may need to be left overnight. Therefore, the mounting of the custom optical filter set should be performed ≥1 day prior to
the actual experiment. The cleaning of the optics should take 15 to 30 min. Cytometer setup will vary depending upon the familiarity of the operator with tuning the water-cooled lasers. Typically, changing from a 488-nm primary laser setup to a 458-nm primary laser setup and aligning the custom optics for the three fluorescent protein detection can take as little as 15 min. Transfections and cell culture may take 24 to 48 hr depending upon cell type, transfection method, and the expression plasmids used.
Literature Cited Beavis, A.J. and Kalejta, R.F. 1999. Simultaneous analysis of the cyan, yellow and green fluorescent proteins by flow cytometry using single-laser excitation at 458 nm. Cytometry 37:68-73. Cole, N.B., Smith, C.L., Sciaky, N., Teraski, M., Edidin, M., and Lippincott-Schwartz, J. 1996. Diffusional mobility of golgi proteins in membranes of living cells. Science 273:797-801. Cormack, B.P., Valdivia, R.H., and Falkow, S. 1996. FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173:33-38. Cormack, B.P., Bertram, G., Egerton, M., Gow, N.A., Falkow, S., and Brown, A.J. 1997. Yeastenhanced green fluorescent protein (yEGFP) a reporter of gene expression in Candida albicans. Microbiology 143:303-311. Cubitt, A.B., Woollenweber, L.A., and Heim, R. 1999. Understanding structure-function relationships in the Aequorea victoria green fluorescent protein. Methods Cell Biol. 58:19-30. Flach, J. and Silver, P.A. 1994. A yeast RNA-binding protein shuttles between the nucleus and cytoplasm. Mol. Cell. Biol. 14:8399-8407. Galbraith, D.W., Anderson, M.T., and Herzenberg, L.A. 1999. Flow cytometric analysis and FACS sorting of cells based on GFP accumulation. Methods Cell Biol. 58:315-341. Heim, R., Prasher, D.C., and Tsien, R.Y. 1994. Wavelength mutations and posttranslational autooxidation of green fluorescent protein. Proc. Natl. Acad. Sci. U.S.A. 91:12501-12504. Heim, R., Cubitt, A.B., and Tsien, R.Y. 1995. Improved green fluorescence. Nature 373:663-664. Jiang, W. and Hunter, T. 1998. Analysis of cell cycle profiles in transfected cells using a membranetargeted GFP. BioTechniques 24:348-354. Kalejta, R.F., Shenk, T., and Beavis, A.J. 1997. Use of a membrane-localized green fluorescent protein allows simultaneous identification of transfected cells and cell cycle analysis by flow cytometry. Cytometry 29:286-291. Kalejta, R.F., Brideau, A.D., Banfield, B.W., and Beavis, A.J. 1999. An integral membrane green fluorescent protein marker, Us9-GFP, is quantitatively retained in cells during propidium iodide-based cell cycle analysis by flow cytometry. Exp. Cell Res. 248:322-328.
Flow Cytometry Instrumentation
1.16.11 Current Protocols in Cytometry
Supplement 16
Li, X., Zhang, G., Ngo, N., Zhao, X., Kain, S.R., and Huang, C.C. 1997. Deletions of the Aequorea victoria green fluorescent protein define the minimal domain required for fluorescence. J. Biol. Chem. 272:28545-28549.
Wachter, R.M., Elsliger, M.A., Kallio, K., Hanson, G.T., and Remington, S.J. 1998. Structural basis of spectral shifts in the yellow-emission variants of green fluorescent protein. Structure 6:12671277.
Okabe, M., Ikawa, M., Kominami, K., Nakanishi, T., and Nishimune, Y. 1997. “Green Mice” as a source of ubiquitous green cells. FEBS Lett. 407:313-319.
Wang, S. and Hazelrigg, T. 1994. Implications for bcd mRNA localization from spatial distribution of exu protein in Drosophila oogenesis. Nature 369:400-403.
Pestov, D.G., Polonskaia, M., and Lau, L.F. 1999. Flow cytometric analysis of the cell cycle in transfected cells without cell fixation. BioTechniques 26:102-105. Tsien, R.Y. 1998. The green fluorescent protein. Annu. Rev. Biochem. 67:509-544.
Contributed by Andrew J. Beavis and Robert F. Kalejta Princeton University Princeton, New Jersey
Valdivia, R.H., Hromockyj, A.E., Monack, D., Ramakrishnan, L., and Falkow, S. 1996. Applications for green fluorescent protein (GFP) in the study of host-pathogen interactions. Gene 173:47-52.
Simultaneous Analysis of Three Fluorescent Proteins
1.16.12 Supplement 16
Current Protocols in Cytometry
Plug Flow Cytometry The flow cytometer is unique among biomedical analysis instruments in its ability to make simultaneous and multiple optical measurements on individual cells or particles at high rates. Fluorescent probes enable the flow cytometer to quantify an ever-increasing variety of cell-associated macromolecules and physiological responses. One of the most powerful aspects of flow cytometry has been the ability to rapidly screen large collections of cells or particles one by one for the presence of rare subpopulations and to subsequently purify such subpopulations by automated sorting. By contrast, flow cytometry has been slow to develop as an efficient means for large-scale screening operations involving multiple discrete suspensions of particles or reagents. Such a capability promises to benefit a number of areas of biological investigation. For example, modern drug discovery involves testing of cellular targets against millions of potentially valuable compounds that may bind cellular receptors to effect clinically therapeutic cellular responses. High-throughput flow cytometry represents an important multifactorial approach for screening large combinatorial libraries of such compounds. Recently, computer-controlled sample mixing and delivery systems have been adapted to provide flow cytometers with advanced sample-handling capabilities (reviewed in Nolan and Sklar, 1998; Nolan et al., 1999). The adaptation of flow injection analysis (FIA)-based techniques involving high-precision valves and stepper motor–driven syringes (Lindberg et al., 1993) has enabled unattended online bioprocess monitoring (Zhao et al., 1999) and execution of subsecond-resolution kinetics experiments (Sklar et al., 1998; Seamer et al., 1999). However, these sample-handling approaches have not been designed to facilitate high rates of throughput such as that required for largescale sample screening operations or short time-scale physiological measurements. The authors have recently developed an alternative FIA-based approach to automated sample handling for flow cytometry (Edwards et al., 1999). The concept is that there is a continuously flowing stream of fluid into which individual sample suspensions are sequentially inserted as a bolus or “plug” of precisely defined volume. The stream delivers the sample plugs, separated by empty volumes
UNIT 1.17 of fluid (the buffer of which the stream is composed), to the point of analysis in the laser beam. This unit describes basic elements and concepts of the plug-flow cytometry system, and representative applications. Hardware. The central element of the plug flow system is an eight-port, two-position reciprocating valve that is fitted with two sizematched sample loops (Fig. 1.17.1). Fluids may enter and leave the valve by two independent pathways, each with dedicated inlet and outlet ports. Each sample loop consists of a length of Teflon tubing inserted between the inlet and outlet ports of one of the pathways. Thus, during system operation, there are two continuously flowing streams, each of which enters and leaves the valve via dedicated ports and flows through one of the sample loops en route. The useful feature of the valve is that, under computer control, it can reversibly alternate the sample loop that is physically associated with a particular fluid pathway. For example, in the first valve position, sample loop A connects with pathway 1 and sample loop B connects with pathway 2 (Fig. 1.17.1A). When the valve switches to its second position, sample loop A is switched into pathway 2 and loop B into pathway 1 (Fig. 1.17.1B). Fluid trapped in sample loop A from pathway 1 is thereby inserted into pathway 2 and vice versa. In this way, the valve acts as a transducer by which fluid elements can be transferred between two independently flowing pathways with minimal effects upon flow characteristics in each pathway. In operation, there is continuous cycling of the valve so that as each loop inserts its sample particles into the fluid stream used for analysis, the other loop is being loaded with fresh sample particles (Fig. 1.17.1C-F). The other hardware elements of the system serve to control the characteristics of fluid flow in the two pathways. The pressurized fluidtransport pathway is dedicated to moving buffer fluid from a pressurized reservoir through the valve and into the flow cytometer sample stream. In this pathway, the flow is maintained at a rate compatible with good optical alignment of flowing sample particles with the laser interrogation point. In the system configuration illustrated in Figure 1.17.2A, the sample pressurization system of a Beckman Coulter Elite flow cytometer has been used to pressurize physiological saline buffer in a 25 × 100–mm
Contributed by Bruce S. Edwards and Larry A. Sklar Current Protocols in Cytometry (2001) 1.17.1-1.17.10 Copyright © 2001 by John Wiley & Sons, Inc.
Flow Cytometry Instrumentation
1.17.1 Supplement 17
polystyrene tube. This arrangement uses the sample pressurization controls of the flow cytometer to define the volumetric rate of buffer flow. The second fluid pathway is the sample-uptake pathway by which sample particle suspensions are introduced into the valve. To date, the authors have used two different methods to control fluid flow in this pathway. In the first (Fig. 1.17.2A), an automated stepper motor–
A
driven syringe is connected to the outlet port of the valve via a length of Teflon tubing. Another length of Teflon tubing connects the valve inlet port to the sample source. The sample is aspirated into one of the valve sample loops by applied negative pressure from the syringe. In the second method (Fig. 1.17.2B), a peristaltic pump controls the flow of sample particles to the valve. Sample is aspirated under negative pressure, then pushed into one of the valve
C Path 2
Pressurized fluid source
Sample source Sa m pl e
m
pl
e
To waste
B
op A
Sa
op
lo
Path 1
lo
Path 1
To flow cytometer
D Path 2 Position 1 Sample loop A in path 1 Sample loop B in path 2
B
E
Sa m pl e
p
op
B
op
m
lo
lo A
Sa
le
F
Position 2 Sample loop B in path 1 Sample loop A in path 2
Plug Flow Cytometry
Figure 1.17.1 (Legend is on next page.)
1.17.2 Supplement 17
Current Protocols in Cytometry
sample loops under positive pressure. Flexible tubing compatible with peristaltic pumping requirements is used to transport sample from source to valve. Software. All the hardware components of the system (i.e., valves, syringe pump, peristaltic pump, port switching device) are subject to computer control via an RS232 serial port interface. The primary functions of the systemcontrolling computer program are opening and closing an RS232 serial port and creating strings of ASCII characters (device command codes) to transmit via the serial port to the hardware devices. The program also detects and responds to characters transmitted from the devices back to the computer; these transmissions indicate device activity states and readiness to accept additional commands. ASCII command and response codes pertinent to device control were supplied by the device manufacturers. The authors used the Microsoft Visual C++ compiler to generate the controlling software code and the visual user interface. To control multiple devices from a single computer serial RS232 port, the authors used a 4-port Mini Smart Switch/Port Combiner (B&B Electronics, Ottawa, IL). A typical ASCII-character command string generated by the software contains two segments. The first consists of characters that are recognized by the Smart Switch and cause it to connect the computer serial port to the appropriate device. The second segment contains characters pertinent to controlling subsequent actions of the device. For example, the string of ASCII characters required for a syringe pump initialization command consists of the following 12 ASCII characters: ESC 3 A \ r / 1 Z 2 R \ r . The first 3 characters (ESC 3 A) signal the Smart Switch
to connect the computer serial port to subport A, to which the syringe pump interface is attached. The remaining characters signal the pump to carry out an initialization sequence of activities. Flow rate in the sample delivery pathway. As the flow rate of the sample stream increases, there is a corresponding increase in the diameter of the sample stream relative to the enclosing sheath stream and a greater probability of flowing sample particles straying off the optical alignment axis. As a rule of thumb, a volumetric sample stream flow rate of 0.5 to 1 µl/sec is conducive to good optical alignment. In plug flow cytometry, the fluid buffer-stream flow rate corresponds to the rate of sample-stream flow through the analysis point. It also governs the time required for eluting sample particles from the sample loop. With respect to the number of samples that can be analyzed per unit time, this is typically the rate-limiting step. The sample loop must be flushed with ∼3 volumes of buffer to completely elute sample particles. Thus, at a 1-µl/sec flow rate, sample elution from a 5-µl loop requires 15 sec. Faster sample throughput is achieved by increasing the flow rate, but with the consequence that the alignment of some sample particles will be sacrificed. High flow rates (>2 µl/sec) result in a broadening of the fluorescence intensity distribution of particles such that the misaligned particles form a left-shifted tail corresponding to reduced intensities. Light-scatter gating. In a sample-screening operation in which one expects a relatively large difference in fluorescence intensity between positive and negative samples, such skewing of the distribution will be of little consequence (especially if measurements are
Figure 1.17.1 A schematic of the interposition of sample loops in the two fluidic pathways of the reciprocating valve. A two-position eight-port valve (e.g., VICI C22Z, Valco Instruments) reciprocates between two positions. The entry and exit ports for each of the two fluid pathways that pass through the valve (path 1 and path 2) remain constant. (A) In valve position 1, fluid in path 1 passes through sample loop A while fluid in path 2 passes through sample loop B. (B) In valve position 2, the sample loops are switched so that each is in the alternate fluid pathway. In this fashion, a plug of fluid from each pathway is inserted into the alternate pathway. (C-F) Hypothetical sequence of events illustrating sequential sample plug formation and delivery to the flow cytometer. (C) A bolus of sample particles (gray band) has been moved from the sample source and through the valve so that a 5-µl fraction of sample is contained in one of the valve sample loops. (D) The valve has switched to its alternate position and the 5-µl sample plug has been partially moved out of the sample loop towards the flow cytometer. A second bolus of sample particles (black band) moves into the second sample loop as extraneous particles from the first sample move to waste. (E) The second sample is positioned for plug formation as the first sample plug is delivered to the flow cytometer. (F) After another valve switch, a third bolus of sample particles (gray band) moves into the first sample loop as a 5-µl plug from the second sample moves from the second sample loop to the flow cytometer.
Flow Cytometry Instrumentation
1.17.3 Current Protocols in Cytometry
Supplement 17
A
Sample chamber pressure outlet Union
TT
Sample insertion rod
Pressurized fluid
2 Position valve
TT
TT
Sample insertion rod
Syringe pump
Silicone tubing
Sample
Flow nozzle insertion rod Flow cytometer
B
Peristaltic tubing
Peristaltic pump
Waste
Sample
Figure 1.17.2 (Legend is on next page.) Plug Flow Cytometry
1.17.4 Supplement 17
Current Protocols in Cytometry
routinely made on a log scale). Alternatively, one may use the technique of light-scatter gating as a means to make precision measurements at high sample-flow rates. This technique involves first identifying the light-scatter profile of optimally aligned particles under low flowrate conditions and constructing an electronic light-scatter gate based on this profile. This light-scatter gate is then used to enable selective analysis of the fraction of particles that remain properly aligned under high flow-rate conditions. An example of such gating is illustrated in Figure 1.17.3, in which the light-scatter gate for fluorescent beads was established at a flow rate of 0.8 µl/sec (Fig. 1.17.3A). When plug flow was performed at a buffer flow rate of 3.6 µl/sec, there was a “comet tail” broadening of the bead light-scatter profile such that only 29% of beads fell within the gate (Fig. 1.17.3C). The linear-scale fluorescence intensity coefficient of variation (CV) of gated beads was 5.2% (Fig. 1.17.3D, filled histogram). This was larger than the CV of 3.3% for the corresponding profile of slow-flowing optimally aligned beads (Fig. 1.17.3B), but still in a range to permit measurements of reasonably high resolution. The broad left-shifted fluorescence distribution of ungated beads is also illustrated (Fig. 1.17.3D, open histogram). Thus, if it is acceptable, for screening purposes, to eliminate some of the particles from the analysis, then light-scatter gating represents a viable option to significantly improve assay performance under the suboptimal alignment conditions associated with rapid plug flow analysis. Note, however, that such analysis requires that the particles be a single population with relatively homogeneous light-scatter characteristics. Particles with several discrete light-scatter clusters might also be analyzed (i.e., by establishing a separate gate
for each cluster), but only if the ungated clusters do not overlap under high flow-rate conditions. Path length between valve sample loop and point of analysis. Laminar flow in a long straight tube produces a parabolic velocity profile (Kachel et al., 1990): Vr = 2Vavg [1 − (r/R)2] in which Vr is the flow velocity at a radial distance, r, from the tube axis, R is the tube radius, and Vavg is the average fluid flow velocity. The velocity is 2Vavg along the tube axis and progressively decreases with increasing radial distance from the axis. As a result of this velocity profile, particles along the tube axis move more rapidly than particles closer to the tube wall. As a sample plug moves along a tube, the distance between the fastest and slowest moving sample particles is therefore expected to progressively expand. The time required for a plug of sample particles to traverse the point of analysis is measured as the time interval observed between the leading edge and trailing tail of each sample plug. At a flow rate of 3.6 µl/sec, a sample plug of 5 µl is expected to require only 1.4 sec to traverse the point of analysis, if there is no sample dispersion. To determine the effect of travel distance on sample dispersion, the authors varied the length of tubing between the sample loop exit port and the point of analysis in the flow cytometer and measured the corresponding transit time of a 5-µl plug of fluorescent beads. By comparing the observed transit times to the 1.4 sec expected in the absence of dispersion, the authors estimated there to be a 10% to 12% increase in sample plug length per centimeter of travel. Thus, the sample exit port should be placed sufficiently close to the flow nozzle insertion rod to avoid overlap of adjacent sample plugs due to this magnitude of particle dispersion. The solution was to attach
Figure 1.17.2 System schematics for sample acquisition and delivery of sample plugs to the flow cytometer. (A) Samples are aspirated under the control of a Cavro XL3000 modular digital pump equipped with a 500-µl syringe barrel and attached to the sample-uptake fluid pathway. In the pressurized fluid transport pathway is a 25 × 100–mm Falcon polystyrene tube containing a fluid buffer solution (e.g., phosphate-buffered saline). The tube is sealed at the top with a standard sample cap assembly for the Elite flow cytometer (Beckman Coulter) and pressurized by attaching the silicone tubing part of the cap assembly to the cytometer sample chamber pressure outlet. A low-pressure union (Upchurch Scientific) couples the sample insertion rod of the cap assembly to a length of Teflon tubing that is joined in turn to an inlet port of the reciprocating valve. Inserted into the valve outlet port of the fluid transport pathway is another sample insertion rod. This is joined to the flow nozzle insertion rod by a 1-cm length of silicone tubing (sample tubing specified for conventional use with the Elite). TT, Teflon tubing of 0.065-in. o.d. and 0.01-in. i.d. (Upchurch Scientific). Each sample loop is a 10-cm length of this tubing and holds a fluid volume of 5 µl. (B) Use of a peristaltic pump in the sample uptake fluid pathway. A Gilson Minipulse 3 peristaltic pump (Gilson, Inc.) is used with 0.02-in. i.d. Gilson peristaltic tubing to aspirate and deliver samples to the reciprocating valve. Other system components are the same as in panel A.
Flow Cytometry Instrumentation
1.17.5 Current Protocols in Cytometry
Supplement 17
800
A
600
400 gate 200
Number of events
Side scatter
800
0
600
400
200
0 0
800
0 100 200 300 400 500 600 700 Fluorescence intensity
100 200 300 400 500 600 Forward scatter 800
C Number of events
Side scatter
B
600
400
200
D
600
400 ungated 200 gated
0
0 0
100 200 300 400 500 600 Forward scatter
0 100 200 300 400 500 600 700 Fluorescence intensity
Figure 1.17.3 Light-scatter gating to enable fluorescence analysis of the best optically aligned particles. Flow Check beads (Beckman Coulter) were analyzed at a volumetric sample-stream flow rate of 0.8 µl/sec to determine the light-scatter profile (forward scatter versus log side scatter) (A) and the green fluorescence intensity profile (B) of optimally aligned beads. The beads were then repetitively sampled by plug flow cytometry at a fluid transport stream flow rate of 3.6 µl/sec to determine the light scatter (C) and fluorescence intensity (D) profiles. The electronic light-scatter gate enclosing optimally aligned beads in panel A was used to gate the fluorescence intensity analysis in D (filled histogram) for comparison with the fluorescence profile of ungated beads (open histogram).
the reciprocating valve to an articulating-arm mounting system so that it could be positioned immediately above the flow nozzle (Fig. 1.17.4). Incorporation of a linear translation stage into the assembly was also helpful for precise positioning of the valve. Note, however, that the short (∼1 cm) length of silicone tubing that joins the valve sample insertion rod to the flow nozzle insertion rod (Figs. 1.17.2A and 1.17.4) serves the important function of isolating the flow nozzle from vibrations that result when the valve switches from one position to the other. If optical alignment varies in a periodic fashion, make sure that the rod extending from the valve outlet port Plug Flow Cytometry
is not in contact with the flow nozzle insertion rod. Sample carryover. Carryover of particles or soluble reagents from one sample plug to the next is potentially affected by a number of parameters. One source of contamination is particles from a preceding sample plug that are trapped in the dead volume of valve junctions that join sample loops to the sampling and fluid flow lines. These are probably the major source of the 2% to 4% particle carryover observed when sampling is done rapidly in a no-wash format (Edwards et al., 1999, 2001a). Such trapped particles may be cleared by including an intersample wash step during which the valve is reciprocated multiple times to flush
1.17.6 Supplement 17
Current Protocols in Cytometry
Figure 1.17.4 Positioning of the reciprocating valve above the flow nozzle. The valve is attached to an articulating-arm mounting system (Edmund Scientific), which is mounted on a 12 × 12–in. optical bench plate (Edmund Scientific). The plate rests atop the flow cytometer in proximity to the flow nozzle. A linear translation stage (Edmund Scientific) between the arm and the valve is used for fine positioning.
particles from the junctions. Another source of carryover is particles trapped in the sampling line. This is minimized by using appropriate Teflon tubing, adding surfactants or serum albumin to particle suspensions, and minimizing the number of junctions in the sampling line. Increasing the sampling-line fluid flow rate may also be helpful. In many cases (e.g., robust screening assays), low levels of carryover (<5%) will not
significantly affect the analysis. The use of nonparametric analysis statistics will also minimize effects of statistical outlier measurements resulting from carryover (Edwards et al., 2001a). Modifications/Options. The buffer reservoir can be increased to a larger volume to permit extended time before reservoir reloading. An external pressurization source may be used Flow Cytometry Instrumentation
1.17.7 Current Protocols in Cytometry
Supplement 17
Plug Flow Cytometry
(e.g., house air, pump, compressed nitrogen, etc.). Applications. This plug flow sampling system has several important features that promise to extend the capabilities and usefulness of flow cytometry. First, because sample plugs are of a precisely defined volume (5 µl in the present system), particle concentrations are directly determined from the total particle counts in each sample plug. Second, it is not necessary that the vessel containing the sample be pressurized, thus relieving the restrictions on sample vessel geometry encountered with most conventional flow cytometers. Third, the plug flow system serves as a fluidics transducer, which permits the flow cytometer to be interfaced with other fluidics-based instruments to enable novel applications. Direct cell or particle concentration determinations. There are circumstances in which it is desirable to determine absolute counts of cells or cell subsets in suspension. If the flow rate of the sample stream is known, cell concentrations may be determined by partitioning flow cytometry data into discrete time intervals and determining the cell counts and sample stream volumes pertinent to the time intervals. This approach requires that the sample stream flow rate be invariant or be measured periodically over the course of a series of measurements. Another approach is to incorporate fluorescent beads at a known concentration into each sample and to estimate sample volume by counting the number of beads over a specified time interval. However, this approach increases overall statistical variance by adding the statistical error associated with bead counting to the statistical error associated with sample particle counting. In plug flow cytometry, constant sample volume is an inherent aspect of the assay. Thus, determination of the number of cells or particles in each sample plug provides a direct readout of particle concentration in the source suspension. To illustrate the reproducibility of particle concentration determinations, a stirred suspension of fluorescent beads was sampled at rates ranging from 3.5 to 7 sample plugs per minute (Fig. 1.17.5). Multiple sample plugs were clearly resolved as discrete clusters of beads over each sampling time domain. The series of peaks at the bottom of each panel represent the temporal fluctuations in bead numbers as samples transited the point of analysis in the flow cytometer. CVs associated with bead concentration determinations ranged from 5% to 8%. Note also the uniformity of the fluorescence
intensity profiles in the repetitive samples (event clusters at the top of each panel). The authors have exploited this particlecounting capability to extend the accuracy of flow cytometric cell-cell adhesion assays (Edwards et al., 2000, 2001b). Examples of other potential applications include assays of the death or survival of cell subsets (since dyes such as propidium iodide fail to account for dead cells that have disintegrated), and cell subset proliferation. Interfacing flow cytometers with other analytical instruments. The ability to interface flow cytometers with other instruments offers the promise of greatly extending the breadth and quality of analytical measurements amenable to flow cytometry. Problems encountered in designing such interfaces with conventional flow cytometers include sample pressurization requirements and incompatible sample flow characteristics (e.g., flow rates too high, peristaltic flow, etc.). Recently, the authors have reported examples of two novel instrument interfaces made possible by the plug flow cytometry technology. One set of studies used an automated syringe to aspirate samples from cell mixtures exposed to uniform fluid shear in a cone-plate viscometer (Edwards et al., 2000, 2001b). This enabled use of the flow cytometer to measure quantitative differences in the sheardependent adhesion of eosinophils and neutrophils to P selectin (Edwards et al., 2000). In other studies (Edwards et al., 2001a,b), the authors described a series of novel applications made possible by interfacing the flow cytometer with an automated fluidics-based commercial pharmacology platform (HTPS, Axiom Biotechnologies). These applications included endpoint fluorescence assays of cells sampled from microplates at analysis rates of 9 to 10 cell samples per minute and assays of cellular calcium responses to soluble receptor ligands sampled from microplate wells at analysis rates of 3 to 4 samples per minute (Edwards et al., 2001a). The HTPS interface also enabled rapid (2 to 3 min) automated characterization of the dose-response profile of cells to ligand gradients spanning 3+ logs of concentration. These gradient assays exploited the multiparametric measurement capabilities of the flow cytometer to make novel measurements of the quantitative relationship between receptor occupancy and cell response (Edwards et al., 2001a). Setup time. In the authors’ experience, a baseline setup time of ∼10 to 15 min is required for plug flow cytometry. The valve is first positioned above the flow nozzle and the valve
1.17.8 Supplement 17
Current Protocols in Cytometry
A 103
fluorescence
102 101
number
100 0 0
30
60
90
120
150
30
60
90
120
150
30
60 90 Time (sec)
120
150
B 103
fluorescence
102 101
number
100 0 0
C 103
fluorescence
102 101
number
100 0 0
Figure 1.17.5 Repetitive sampling to characterize the precision of particle concentration determinations. A stirred suspension of Flow Check beads was sampled at 3.5 (A), 4.5 (B), and 7 (C) sample plugs/min. The mean ±SD number of beads in sample plug peaks (bottom of each panel) were 502 ± 43, 500 ± 35, and 847 ± 42, respectively. This indicated source concentrations of 1.0 × 105, 1.0 × 105, and 1.7 × 105 beads/ml in the respective sample suspensions. Dot clusters at the top of each panel represent the log fluorescence intensity profile of individual beads on the time axis.
insertion rod is joined to the nozzle insertion rod with silicone tubing. The sample pressurization tubing is then attached to the sample chamber pressure outlet. A bead suspension is placed in the pressurized fluid reservoir and flow is briefly established in order to perform optical alignment. The bead suspension is then replaced with the buffer fluid to be used and flow is reestablished for the subsequent initiation of sample acquisition. Additional setup time depends upon the application for which
the plug flow system is to be used. Setup of the cone-plate viscometer requires an additional 5 to 10 min. Setup for the automated pharmacology platform may require from 10 to 30 min, depending upon the complexity of the analysis to be performed. It is worthwhile noting that the automated pharmacology platform, plug flow valve, and flow cytometer are not integrated directly with one another via a central device controller. Rather, in these applications, the valve is synchronized to the timing of the
Flow Cytometry Instrumentation
1.17.9 Current Protocols in Cytometry
Supplement 17
autosampler, which requires additional setup time and empirical observation.
LITERATURE CITED Edwards, B.S., Kuckuck, F., and Sklar, L.A. 1999. Plug flow cytometry: An automated coupling device for rapid sequential flow cytometric sample analysis. Cytometry 37:156-159. Edwards, B.S., Curry, M.S., Tsuji, H., Larson, R.S, Brown, D., and Sklar, L.A. 2000. Expression of P-selectin at low site density promotes selective recruitment of eosinophils over neutrophils. J. Immunol. 165:404-410. Edwards, B.S., Kuckuck, F.W., Prossnitz, E.R., Ransom, J.T., and Sklar, L.A. 2001a. HTPS flow cytometry: A novel platform for automated high throughput drug discovery and characterization. J. Biomol. Screening 6:83-90. Edwards, B.S., Kuckuck, F.W., Prossnitz, E.R., Okun, A., Ransom, J.T., and Sklar, L.A. 2001b. Plug flow cytometry extends analytical capabilities in cell adhesion and receptor pharmacology. Cytometry 43:211-216. Kachel, V., Fellner-Feldegg, H., and Menke, E. 1990. Hydrodynamic properties of flow cytometry instruments. In Flow Cytometry and Sorting, 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 27-45. Wiley-Liss, New York. Lindberg, W., Ruzicka, J., and Christian, G.D. 1993. Flow injection flow cytometry: A new approach for sample and solution handling in flow cytometry. Cytometry 14:230-236. Nolan, J.P. and Sklar, L.A. 1998. The emergence of flow cytometry for sensitive, real-time measurements of molecular interactions. Nature Biotechnology 16:633-638.
Nolan, J.P., Lauer, S., Prossnitz, E.R., and Sklar, L.A. 1999. Flow cytometry: A versatile tool for all phases of drug discovery. Drug Discov. Today 4:173-180. Seamer, L.C., Kuckuck, F., and Sklar, L.A. 1999. Sheath fluid control to permit stable flow in rapid mix flow cytometry. Cytometry 35:75-79. Sklar, L.A., Seamer, L.C., Kuckuck, F., Posner, R., Prossnitz, E., Edwards, B., and Nolan, J.P. 1998. Sample handling for kinetics and molecular assembly in flow cytometry. S.P.I.E. Proc. 3256:144-153. Zhao, R., Natarajan, A., and Srienc, F. 1999. A flow injection flow cytometry system for on-line monitoring of bioreactors. Biotechnol. Bioeng. 62:609-617.
INTERNET RESOURCES http://www.beyondlogic.org/serial/serial.htm Interfacing the serial/RS232 port. A very informative series of articles by Craig Peacock relevant to the type of serial port programming used in the Plug Flow system.
Contributed by Bruce S. Edwards University of New Mexico Albuquerque, New Mexico Larry A. Sklar University of New Mexico Albuquerque, New Mexico and Los Alamos National Laboratory National Flow Cytometry Resource Los Alamos, New Mexico
Plug Flow Cytometry
1.17.10 Supplement 17
Current Protocols in Cytometry
Dynamic Thermoregulation of the Sample in Flow Cytometry
UNIT 1.18
Flow cytometry has long been a method of choice for the functional analysis of living cells and, more recently, for the analysis of molecular assemblies in cells or on microspheres. Temperature is an important parameter for all these applications. Both molecular interactions and their cellular consequences have temperature dependencies, and control of temperature is often used as an experimental tool to dissect molecular mechanisms. Thus, fine control of temperature is an important capability for any analytical platform. In flow cytometry, temperature has typically been controlled by using a circulating water bath to maintain a constant temperature in the sample chamber (Omann et al., 1985; Kelly, 1991). This approach has been adequate for holding samples at a given temperature, but does not allow for rapid changes in sample temperature. Recently, the authors have used Peltier modules to regulate the temperature of the sample line (Graves et al., 2001). Peltier modules are inexpensive, are capable of rapid temperature changes, have low power requirements, and are physically small. These advantages make Peltier modules the basis for a new generation of temperature-control capabilities for flow cytometry. FLOW CYTOMETRY MEASUREMENTS AS A FUNCTION OF TEMPERATURE
BASIC PROTOCOL
To effectively use temperature as an experimental parameter, it must be regulated in the experimental zone, its value recorded into the data file, and it must be tightly correlated with the measured optical parameters. The first two tasks are accomplished with appropriate hardware design (see Support Protocol 1), but the third task is complicated by the fact that current flow cytometry geometries limit the distance between the temperature regulation zone and the analysis point to several centimeters, which introduces a delay volume between the final point of thermoregulation and analysis. The delay volume can be precisely measured (see Support Protocol 2), and used to correctly correlate events with the actual temperature they experience in the thermoregulation zone via the use of standard flow cytometry analysis software and a spreadsheet program. Use of these programs, as described below, will allow accurate correlation of temperature and make it possible to use temperature as an experimental parameter in flow cytometry. Materials Sample Flow cytometer with time-stamp capability and equipped with a Peltier module–based thermoregulation unit (see Support Protocol 1) Flow cytometry analysis software capable of exporting the mean of a parameter as a function of time (e.g., IDYLK, National Flow Cytometry Resource, http://www.bdiv.lanl.gov/NFCR/; or FlowJo, TreeStar Software) Spreadsheet software (e.g., Excel, Microsoft; or Quattro Pro, Corel) Collect data 1. Program the syringe pump on the flow cytometer according to manufacturer’s instructions to deliver the sample at an appropriate flow rate. For most flow cytometers, 10 to 50 ìl/min is suitable. Check the instrument manual to verify the recommended flow rates.
2. Program the temperature controller with the desired temperature profile according to manufacturer’s instructions. Contributed by Steven W. Graves, Robert C. Habbersett, and John P. Nolan Current Protocols in Cytometry (2002) 1.18.1-1.18.16 Copyright © 2002 by John Wiley & Sons, Inc.
Flow Cytometry Instrumentation
1.18.1 Supplement 20
3. Set the temperature controller to hold the inline thermoregulation unit at the starting temperature of the profile. 4. Turn on the DC power supply to the inline thermoregulation unit and wait until the temperature controller indicates that the inline thermoregulation unit has come to the starting temperature. 5. Set the data acquisition software to collect, with time-stamping, the optical parameters of interest and the parameter that is set to record the temperature signal. 6. Load the sample into the loading loop of the sample-loading fluidics (see Support Protocol 1, steps 26 to 28) via the inlet port of the three-way valve. 7. With the flow cytometer delivering sheath, begin sample delivery by starting the syringe pump. 8. Start data acquisition and the temperature profile simultaneously. 9. Collect data files for an exact time duration. If the flow cytometer acquisition software allows the data file to be stored automatically after a specific collection time, use this feature to collect files of an exact duration. If the acquisition software does not have this feature, then use a timer to collect data files for a known duration, close the data file manually, and record the duration of the data file.
Manipulate data 10. Export the means of a bivariate plot of the optical parameter of interest versus time and the temperature voltage values versus time using data-analysis software. IDYLK on Windows platforms and FlowJo on Macintosh platforms are good choices of analysis software.
11. Import the means of fluorescence versus time, and temperature signal versus time into a spreadsheet program such as Excel (Table 1.18.1). 12. Correlate the time of each channel by dividing the duration of the experiment by the number of channels exported and creating a column of data representing the time value of each channel. In the example shown in Table 1.18.1, the experiment duration was 150 sec and the data were exported at a resolution of 64 channels. Therefore, each channel represents 150/64 sec or ∼2.3 sec per channel. Using these values, a time column was created that incrementally spanned from 2.3 to 150 sec.
13. Using the values for the temperature calibration curve of the system (see Support Protocol 2), create a column of real-time temperature values from the temperature voltage signal (Table 1.18.1). Shift the data to account for the delay volume 14. Using the values for the delay volume calculated for the system (see Support Protocol 3), calculate the delay time for the sample delivery rate used in the experiment. Simply divide the delay volume by the delivery rate of the sample to obtain the delay time. In the example in Table 1.18.1, the delay volume was 5.1 ìl and the flow rate was 50 ìl/min or 0.833 ìl/sec, resulting in a delay time of 6.1 sec.
15. Using the delay time, calculate the number of channels that the fluorescence signal needs to be offset. Dynamic Thermoregulation of the Sample in Flow Cytometry
In the example in Table 1.18.1, the delay time was 6.1 sec and there were 2.3 sec/channel. Therefore, the fluorescence data need to be shifted by 2.7 channels to accurately reflect the temperature of the sample as it passed through the inline thermoregulation device.
1.18.2 Supplement 20
Current Protocols in Cytometry
Display the data 16. Shift the fluorescence data down by the number of channels required to account for the delay. Because channels are discrete entities, the number of channels to shift is rounded to the nearest whole number, e.g., 2.7 is rounded up to 3. This rounding introduces a small error, Table 1.18.1 Exported Data in Spreadsheet Format
Time channel no.
Time (sec)
Mean channel of temperature signal
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52
2.3 4.7 7.0 9.4 11.7 14.1 16.4 18.8 21.1 23.4 25.8 28.1 30.5 32.8 35.2 37.5 39.8 42.2 44.5 46.9 49.2 51.6 53.9 56.3 58.6 60.9 63.3 65.6 68.0 70.3 72.7 75.0 77.3 79.7 82.0 84.4 86.7 89.1 91.4 93.8 96.1 98.4 100.8 103.1 105.5 107.8 110.2 112.5 114.8 117.2 119.5 121.9
14.5 14.5 14.3 14.7 14.6 14.4 14.1 14.0 14.3 14.5 14.4 14.7 15.6 16.4 17.1 17.9 18.7 19.6 20.4 21.2 22.0 22.9 23.6 24.4 25.2 26.0 26.9 27.7 28.4 29.3 30.1 31.0 31.7 32.4 33.2 34.0 34.8 35.6 36.4 37.2 38.0 38.9 39.7 40.4 41.1 42.0 42.7 43.7 44.3 44.5 44.5 44.5
Mean Converted Shifted mean channel of temperaturea channel of fluorescence fluorescenceb signal (°C) signal 43.3 42.8 43.0 42.9 42.6 43.2 43.3 43.4 43.1 43.7 42.8 43.0 43.5 43.1 43.6 42.7 43.5 43.6 43.8 43.8 44.6 43.5 42.8 43.1 42.4 42.4 41.9 42.9 41.8 41.4 41.7 41.3 40.0 39.6 38.4 39.5 37.8 37.1 36.0 35.3 34.6 33.5 33.5 33.0 33.7 33.4 33.2 32.5 32.7 32.6 33.1 32.7
31.2 31.3 30.9 31.6 31.5 31.0 30.4 30.2 30.8 31.3 31.0 31.8 33.6 35.2 36.7 38.4 40.2 42.0 43.6 45.4 47.0 48.9 50.5 52.1 53.7 55.5 57.4 59.0 60.6 62.4 64.1 65.9 67.5 68.9 70.6 72.3 74.0 75.7 77.4 79.1 80.8 82.5 84.3 85.7 87.4 89.2 90.8 92.7 94.1 94.4 94.5 94.5
42.9 42.6 43.2 43.3 43.4 43.1 43.7 42.8 43.0 43.5 43.1 43.6 42.7 43.5 43.6 43.8 43.8 44.6 43.5 42.8 43.1 42.4 42.4 41.9 42.9 41.8 41.4 41.7 41.3 40.0 39.6 38.4 39.5 37.8 37.1 36.0 35.3 34.6 33.5 33.5 33.0 33.7 33.4 33.2 32.5 32.7 32.6 33.1 32.7 33.1 32.5 32.6 continued
Flow Cytometry Instrumentation
1.18.3 Current Protocols in Cytometry
Supplement 20
Table 1.18.1 Exported Data in Spreadsheet Format, continued
Time channel no.
Time (sec)
Mean channel of temperature signal
53 54 55 56 57 58 59 60 61 62 63 64
124.2 126.6 128.9 131.3 133.6 135.9 138.3 140.6 143.0 145.3 147.7 150.0
44.5 44.5 44.5 44.5 44.5 44.5 44.5 44.5 44.5 44.5 44.5 44.5
Mean Converted Shifted mean channel of temperaturea channel of fluorescence fluorescenceb signal (°C) signal 33.1 32.5 32.6 32.7 32.7 31.9 32.7 32.1 32.0 32.4 32.3 34.0
94.5 94.5 94.5 94.5 94.5 94.5 94.5 94.5 94.5 94.5 94.5 94.5
32.7 32.7 31.9 32.7 32.1 32.0 32.4 32.3 34.0
aThe converted temperature value for each time channel is calculated from the calibration line derived in Support Protocol
1 (Fig. 1.18.5) and the values of the mean temperature signal.
Green fluorescence channel number
bThe sixth column is the mean fluorescence signal data shifted to account for the delay volume.
45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30
40
50
60
70
80
90
Temperature (°C)
Figure 1.18.1 Unshifted (filled circles) and shifted (open circles) fluorescence data plotted as a function of temperature.
and to minimize this error the data should be exported at the highest resolution available (4096 or 1024 where possible) prior to the shifting operation. In the higher-resolution files, each time channel will represent a smaller amount of time and therefore the rounding procedure will introduce less error.
17. Plot the data from Table 1.18.1 as the optical parameter of interest versus temperature. Dynamic Thermoregulation of the Sample in Flow Cytometry
The data from Table 1.18.1 can now be plotted in any number of ways using commercial graphing software. By plotting the optical parameter in its shifted or unshifted form against the temperature values, one can collect flow cytometry data in a temperature-dependent manner (Fig. 1.18.1).
1.18.4 Supplement 20
Current Protocols in Cytometry
CONSTRUCTION AND MOUNTING OF A PELTIER MODULE–BASED THERMOREGULATION UNIT
SUPPORT PROTOCOL 1
A basic Peltier module-based thermoregulation unit consists of two primary elements. The first is the inline temperature-regulation unit, which consists of a Peltier module, heat-transfer block, and cooling components. The second element is the temperature-control circuit, which consists of the polarity-switching circuit to provide power to the Peltier module and a process-control computer to provide temperature sensing and regulation functions. These elements need to be interfaced to the flow cytometer both fluidically and electronically. This requires modification of the sample fluid line and data acquisition system. Materials Copper block machined to 4 × 4 × 0.75 cm Titanium tubing (5-cm length, 1⁄16-in. o.d., 0.01-in. i.d., Upchurch) Silicone heat-sink compound (e.g., Dow Corning 340, Dow Corning) Tellurex Peltier module–based thermoregulation unit Melcor heat sink, model HX8-201 Plastic screws, washers, and clamps Three red banana jacks Three black banana jacks 12 V fan (Papst) Type J thermocouple (1⁄16-in. o.d. and 6-in. length, Omega Engineering) Watlow Electric Manufacturing model F4 ramping process controller 16-G electrical wire Crydom relay heat sinks, model HS-2 Four Crydom model D1D12 solid-state DC relays Four Newark Electronics model MUR1520 rectifiers, ON semiconductor 2 banana plug to banana plug black patch cords rated to 15 A 2 banana plug to banana plug red patch cords rated to 15 A Newark Electronics DC power supply, Tenma-72-6153 Flow cytometer Low-profile vented electronics case, model 94F4839 (Newark Electronics) Flexible silicone tubing (1⁄16-in. o.d. and 0.01-in. i.d.) PEEK tubing (0.030 in. × 1⁄16 in. × 200 cm, 912 µl; Upchurch) 3-way valve (Upchurch) Programmable syringe pump connected to syringe Construct the heat transfer block 1. Machine a copper block into a heat-transfer block that is 4 × 4 × 0.75 cm. 2. Implement the sample-line passage (centered vertically in the 0.75-cm side, ∼1.33 cm from the left edge, and 1⁄16 in. in diameter) lengthwise through the heat transfer block (Fig. 1.18.2A). 3. Drill the thermocouple hole (centered vertically in the 0.75-cm side, ∼2.66 cm from the left edge, and 1⁄16 in. in diameter) parallel to the sample line passage into the heat transfer block to a depth of 2 cm (Fig. 1.18.2A). 4. Coat the outside of the titanium tubing with a thin layer of heat-sink compound and insert it through the sample-line passage until 0.5 cm extends from each side of the heat-transfer block. 5. Coat both sides of the Peltier module with a thin layer of heat-sink compound and press it into place in the center of the top of the heat sink.
Flow Cytometry Instrumentation
1.18.5 Current Protocols in Cytometry
Supplement 20
6. Center the heat-transfer block on top of the Peltier module. 7. Using screws and plastic clamps, compress the heat-transfer block, Peltier module, and heat sink together. Plastic screws and washers can be used to accomplish this task. Alternatively, custom plastic clamps could be manufactured. It is important that the clamping mechanism not be heat conductive, or it will provide a heat path between the hot and cold sides of the Peltier module.
8. Attach a red and a black banana jack to the two wires extending from the Peltier module. 9. Using mounting hardware, clamp a 12-V fan to the bottom of the heat sink, so that it blows air into the bottom of the heat sink (Fig. 1.18.2B). 10. Insert the type J thermocouple into the thermocouple passage (Fig. 1.18.2A).
A
b
c
a d h
e
g f
B
f
e
b
a
k
l h q g p i j
o laser beam
Dynamic Thermoregulation of the Sample in Flow Cytometry
m n
Figure 1.18.2 (A) General schematic of the inline thermal regulation unit. (B) Detailed side view of the inline thermoregulation unit incorporated into a flow cytometer. Components of each panel are listed as follows: a: silicone rubber portion of delay line (∼63-mm length × 1.5875-mm o.d. × 0.254-mm i.d.); b: heat-transfer block (HTB; 40 mm × 40 mm × 7.5 mm); c: power lines for the Peltier module (PM); d: thermocouple (1.5875-mm o.d.); e: sample inlet line; f: titanium sample line (50-mm length × 1.5875-mm o.d. × 0.254-mm i.d.); g: heat sink; h: Peltier module; i: sheath stream inlet; j: flow nozzle; k: final heating point of sample; l: locking nut to hold nozzle to support bracket; m: jet in air; n: analysis point; o: fan; p: support bracket; q: stainless-steel portion of the delay line (50-mm length × 1.5875-mm o.d. × 0.254-mm i.d.).
1.18.6 Supplement 20
Current Protocols in Cytometry
Construct the temperature control circuit 11. Mount the F4 ramping process controller into the electronics case and wire it for AC power according to the manufacturer’s instructions. The process controller should be mounted into the far left front of the low-profile electronics case according to manufacturer’s instructions. This will provide space for the other components. Power must be provided to the process controller. This should be done according to manufacturer’s instructions. It is critical that this be performed in a safe manner. Proper electrical wiring procedures are beyond the scope of this protocol. Seek the assistance of a skilled electronics technician to complete the power wiring for the process controller.
12. Mount two red banana jacks ∼1-in. apart, one above the other, in the front of the case (Fig. 1.18.3). 13. Mount two black banana jacks even vertically with the red jacks and about ∼1 in. to the right of the red jacks. 14. Label the bottom set of jacks “DC power input” and the top set of jacks “DC power output.” 15. Mount the SSR heat sinks into the case. 16. Apply a thin coating of heat-sink compound to the base of the relays and mount the relays onto the heat sinks. The relays must be solid state and matched to the process-controller outputs and power requirements of the PM (∼8 amps for the Peltier module described above); in the arrangement described here, Crydom D1D12 relays are appropriate.
17. Connect the inputs of the heating-circuit relays (Fig. 1.18.4) to the process controller’s heating-control outputs according to manufacturer’s instructions. 18. Connect the inputs of the cooling circuit relays (Fig. 1.18.4) to the process controller’s cooling-control outputs according to manufacturer’s instructions. 19. Using the 16-G wire, connect the relay outputs to the power input and output jacks as shown in Figure 1.18.4. While doing this, it is important to incorporate the rectifiers inline, as shown (Fig. 1.18.4), to prevent accidental shorting of the system.
inline thermoregulation unit
syringe pump DC power supply
loading loop
temperature control circuit
37°C
laser flow cytometer
A/D converter
Figure 1.18.3 Overall schematic of the inline temperature control unit incorporated into a flow cytometer.
Flow Cytometry Instrumentation
1.18.7 Current Protocols in Cytometry
Supplement 20
20. Connect the positive wire of the fan to the positive control wire of the cooling circuit and connect the negative wire to the negative control wire of the cooling circuit (Fig. 1.18.4). Wiring the fan in this manner allows it to turn on when the cooling circuit is active.
21. Using the patch cords of the appropriate color, attach the DC power supply to the “DC power input” jacks and the Peltier module to the “DC power output” jacks. The DC power supply needs to be matched to the requirements of the Peltier module. In this case, the power supply was a Tenma-72-6153 laboratory DC power supply (Newark Electronics) set to provide 5 A at 15 V. A dedicated power supply of appropriate ratings could also be used, but a laboratory power supply allows more versatility in Peltier module choice.
22. Connect the thermocouple to the thermocouple input connection of the process controller. Mount the thermal regulation unit onto the flow cytometer 23. Mount the heat-transfer block and attached hardware onto the flow cytometer. The heat-transfer block needs to be mounted as close as possible to the flow cell or nozzle (Fig. 1.18.3) in order to minimize the sample delay volume between the temperature regulation point in the heat-transfer block and the analysis point in the flow cell/nozzle. Because each flow cytometer will have different geometries, creativity will be required at this step.
cooling control
to fan heating control SSR +
Watlow temperature controller
SSR +
DC + power supply −
Peltier module
+ SSR
+ SSR
heating control
cooling control
Dynamic Thermoregulation of the Sample in Flow Cytometry
Figure 1.18.4 The temperature control circuit (TCC). The heating and cooling control circuits are labeled. The boxes labeled SSR are the solid-state relays with indicated polarity. The filled triangles topped with a line represent the rectifiers.
1.18.8 Supplement 20
Current Protocols in Cytometry
24. Program the retransmit module of the temperature controller to output a signal proportional to the temperature reading (i.e., the process value) at voltage levels compatible with the analog-to-digital converters of the flow cytometer. The voltage from the retransmit module needs to be compatible with the input voltage requirements of the A/D board. Advanced process controllers, such as the F4, have programmable voltage outputs for their retransmit modules, allowing compatibility between the retransmit module and a variety of A/D systems.
25. Connect the output of the voltage retransmit module to the input of the analog-todigital converter for an unused channel of the flow cytometer data acquisition system. To effectively incorporate the process controller with an unused channel, the input port of that channel must be accessed. For many systems, this is an easy task, but in highly integrated systems, more effort maybe required. The user is cautioned that access to the A/D inputs may require significant modifications that could affect the warranty status of the instrument. Therefore, a skilled electronics technician should connect the retransmit module to the A/D input, and it is strongly recommended that the manufacturer of the instrument be contacted before modifications are attempted. Once the connection has been made, the temperature is recorded as a voltage level along with all other parameters. The A/D channel must be triggered (in a slave mode) to record the nearly DC temperature signal. (The voltage proportional to temperature will appear to be a slowly varying DC level, which requires that the A/D input be DC coupled.)
Connect the sample-driving fluidics 26. Using flexible silicone tubing (1⁄16-in. o.d. and 0.01-in. i.d.), attach the exit of the titanium sample line to inlet line of the nozzle (Fig. 1.18.2B). This tubing length needs to be as short as possible to minimize the delay volume, as well as thick-walled to resist expansion when the sample pressure is increased. The narrow diameter of the tubing also minimizes the delay volume. For low-pressure systems, this tubing can simply be pressed onto the titanium tubing and the tubing entering the nozzle. For higher-pressure systems, coupling devices may necessary.
27. Connect the inlet of the titanium sample line to the exit of the sample-loading loop using flexible silicone tubing (1⁄16-in. o.d. and 0.01-in. i.d.; Fig. 1.18.3). Again, in low-pressure systems, this tubing can simply be pressed onto the titanium tubing and the tubing entering the nozzle. For higher-pressure systems, coupling devices may be required.
28. Connect the inlet of the sample-loading loop (SPIRL-LNK, PEEK, 0.030 in. × 1⁄16 in. × 200 cm works well) to a three-way valve that is connected to the syringe of a programmable-rate syringe pump and to a loading port to allow introduction of sample (Fig. 1.18.3). CALIBRATING THE INLINE THERMOREGULATION UNIT For effective use of the inline thermoregulation unit on a daily basis, the temperature response of the process controller (in conjunction with the data acquisition system) must be determined for real temperature value extraction from the data. In addition, the delay volume of the system must be determined in order to calculate the time required for the sample to pass from the inline thermoregulation unit to the point of analysis (i.e., the “delay time”) at any sample delivery rate. With these items calibrated, the system can be used on a daily basis to monitor optical parameters as a function of temperature, whether the observed phenomena occur rapidly or slowly.
SUPPORT PROTOCOL 2
Flow Cytometry Instrumentation
1.18.9 Current Protocols in Cytometry
Supplement 20
Materials Microspheres 50 mM 2-[N-morpholino]ethanesulfonic acid (MES) buffer, pH 6.5 (see recipe) DNA oligomer ∼25 bases in length with a 5′ NH3 group 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDAC) 1× PBS buffer, pH 7 (APPENDIX 2A) DNA oligomer complementary to the specific DNA oligomer and with a 5′ fluorescent group attached Flow cytometer with time-stamp capability and equipped with a Peltier module–based thermoregulation unit (see Support Protocol 1) Flow cytometry analysis software capable of exporting the mean of a parameter as a function of time (e.g., IDYLK, National Flow Cytometry Resource, http://www.bdiv.lanl.gov/NFCR/; or FlowJo, TreeStar Software) Spreadsheet software Calibrate the temperature response 1. According to the manufacturer’s instructions, program the process controller to output the set value of the system rather than the process value of the system. Confirm that the retransmit module is configured to send a proportional signal compatible with the analog-to-digital converter and that the retransmitted signal is of the exact same magnitude as that sent for the process value in Support Protocol 1, step 23.
2. Program the process controller to have a set value of 30°C. 3. Using either test pulses or a concentrated sample of microspheres to trigger the system, collect a data file. 4. Using data-analysis software, obtain the mean channel response for the parameter that collects the temperature signal.
256
Temperature voltage signal channel number
224
y = 1.8994x − 1.3615 r 2 = 0.9999
192 160 128 96 64 32 0 25 30
35 40
45 50
55 60
65 70
75 80
85 90
95 100
Temperature (°C) Dynamic Thermoregulation of the Sample in Flow Cytometry
Figure 1.18.5 A calibration line for temperature response in channels versus actual temperature. The slope and intercept of this line can be used to convert temperature signal to actual temperature (Table 1.18.1).
1.18.10 Supplement 20
Current Protocols in Cytometry
5. Repeat steps 2 through 4 for temperatures of 40°, 50°, 60°, 70°, 80°, 90°, and 95°C. The calibration range should be appropriate for the temperature at which the experiment will be performed.
6. Plot the temperature means against the set temperature values and fit the data to a line (Fig. 1.18.5). The slope and intercept of this line can be used to calculate the temperature value of each channel.
7. Reset the process controller to output the process value rather than the set value. Calibrate the delay volume 8. Dilute microspheres to 1 × 108/ml in 100 µl MES buffer, pH 6.5, and 500 nM of appropriate DNA oligomer (with a 5′ amino group). To this solution, add 16 mg of solid 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDAC) in powder form. Vortex well until the EDAC is dissolved. 9. Incubate 1 hr at room temperature with gentle agitation. 10. Collect the microspheres by microcentrifuging for 2 min at 1500 × g, room temperature. Wash the microspheres with 250 µl of 1× PBS three times, collecting by centrifugation at the above conditions each time. 11. Suspend the microspheres at 1 × 108/ml in 1× PBS. 12. Combine 50 µl oligomer-conjugated microspheres with 50 nM of 5′ fluorescently tagged DNA oligomer in 50 µl of 1× PBS and incubate 1 hr at room temperature. 13. Dilute the annealed microspheres to 1 × 107/ml in 1× PBS. 14. Load the microspheres into the sample loop. 15. Program the process controller with a profile that holds 20 sec at 30°C, and then ramps linearly from 30° to 95°C at a ramp rate of 0.87°C/sec, followed by a 45-sec pause at 95°C. 16. Deliver the microspheres through the inline thermoregulation unit to the analysis point via a syringe pump at a delivery rate of 12.5 ml/min. 17. Collect and export data as in the Basic Protocol. 18. Repeat steps 16 and 17 at flow rates of 25 and 50 ml/min. 19. Repeat steps 16 through 18 using ramp rates of 0.36°, 0.54°, 0.72°, 1.08°, 1.44°, and 2.17°C/sec. 20. For each data file collected, normalize the fluorescence channel values by subtracting the average of the fluorescence mean values that occurred within the first 15 sec from every fluorescence mean value and then dividing every fluorescence mean value by the average of the fluorescence mean values that were collected in the last 15 sec. 21. Plot the normalized fluorescence values against time for the files collected at sample flow rates of 12.5, 25, and 50 ml/min at a ramp rate of 1.44°C/sec (Fig. 1.18.6A). 22. On the same graph, plot the temperature values against time for the files collected at sample flow rates of 12.5, 25, and 50 ml/min at a ramp rate of 1.44°C/sec (Fig. 1.18.6A). Flow Cytometry Instrumentation
1.18.11 Current Protocols in Cytometry
Supplement 20
1.0
95 12.5 µl/min 25.0 µl/min 50.0 µl/min
0.5
0.0
30
Time until 50% loss of binding (sec)
B
Temperature (°C)
Normalized fluorescence
A
0 10 20 30 40 50 60 70 80 90 100 110
55 50 45 40 35 30 25 20 15 10 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 1/flow rate (sec/µl)
Time (sec)
C
1.44 °C/sec 1.08 °C/sec 0.87 °C/sec
D 85
1.0 Temperature (°C)
Normalized fluorescence
90
2.17 deg/sec 1.44 deg/sec 1.08 deg/sec 0.87 deg/sec 0.72 deg/sec 0.54 deg/sec 0.36 deg/sec
0.5
80
12.5 µl/min 25.0 µl/min 50.0 µl/min
75 70 65 60 55
0.0
50 0 30
40
50
60
70
Temperature (°C)
80
90
100
0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50
Ramp rate (°C/sec)
Figure 1.18.6 Calculation of the delay volume between the ITRU and the analysis point. (A) The normalized fluorescence plotted as a function of time, and the voltage output of the TCU converted to temperature plotted as a function of time. The normalized fluorescence units are on the left-hand y-axis and the temperature scale is on the right-hand y-axis. The temperature parameter curves are increasing left to right and the fluorescence curves are decreasing from left to right. (B) The duration from the start of the temperature ramp until 50% of the normalized fluorescence was lost plotted versus 1/flow rate at three different ramp rates. The slopes of these lines represent the delay volume of the experiment as described in the results section. (C) Plot of the normalized fluorescence as a function of temperature at the point of heating for several different ramp rates. (D) Plot of the temperature when 50% of normalized fluorescence was lost versus the ramp rate at three different flow rates. The slopes of these lines represent the delay times at varying flow rates as described in the results section.
23. Plot the time from the start of temperature ramp until the fluorescence reaches a normalized value of 0.5 (calculated from Fig. 1.18.6A) as a function of the inverse of the flow rate (Fig. 1.18.6B). 24. Fit these points to a line, the slope of which represents the delay volume. 25. Repeat steps 21 through 24 for files with ramp rates of 0.87° and 1.08°C/sec (Fig. 1.18.6B). The slopes of these lines (Fig. 1.18.6B) represent the delay volume.
Dynamic Thermoregulation of the Sample in Flow Cytometry
26. Plot the normalized fluorescence versus temperature for all ramp rates at a sample delivery rate of 25 ml/min (Fig. 1.18.6C). The temperature values are derived as described in the Basic Protocol. Note that the fluorescence values used here are not shifted in any way.
1.18.12 Supplement 20
Current Protocols in Cytometry
27. Plot the temperature at which the normalized fluorescence reaches 0.5 versus ramp rate (Fig. 1.18.6D) 28. Repeat steps 26 and 27 for sample delivery rates of 12.5 and 50 ml/min. 29. Fit the data for each sample delivery rate to a line, the slope of which is the “delay time” for that flow rate. 30. Multiply the sample delivery rate by the slope of the corresponding fitted line to obtain the delay volume for the system. 31. Take the mean of all the values calculated for the delay volume of the system to obtain the best estimate of the delay volume of the system. The delay volume is more important to calculate than the delay time at any particular flow rate, because the delay volume can be used to calculate the delay time at any desired sample delivery rate.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
2-[N-morpholino]ethanesulfonic acid (MES) buffer, 50 mM, pH 6.5 Dissolve 1.95 g of 2-[N-morpholino]ethanesulfonic acid in deionized, distilled water. Bring the solution to pH 6.5 using a 2 N NaOH solution. COMMENTARY Background Information The Peltier module (PM) Peltier modules are thermoelectric modules that function on the Peltier effect, which is the heat pumping generated when current flows through properly matched semi-conductors, making it possible for these modules to be used for both heating and cooling (Rowe, 1995). There are many Peltier module producers, and readers are advised to examine the currently available units, as new developments can occur rapidly. When selecting a PM, the primary features of concern are the maximum temperature gradient that can be maintained from the hot side to the cool side (∆T, given in °C), the maximum wattage of heat transfer provided by the unit (Qmax, given in watts), and the dimensions of the unit. Each of the above parameters must be considered in light of experimental needs. Single-stage Peltier modules are sufficient to achieve ∆Ts of nearly 80°C, giving them a wide window of operations. For example, for DNA melting experiments from a microsphere, a single-stage Peltier module was used to provide sample temperature regulation from 30° to 95°C (Graves et al., 2001). This window can be shifted up and down easily between 0° and 100°C, which are biologically
relevant temperature limits. If a wider temperature window is desired, it is possible to use multistage Peltier modules, which can have ∆Ts >100°C. Qmax must be considered if a significant volume of sample needs to be heated, or if rapid changes in temperature are required, as both will place a significant heat load on the system. Larger Qmaxs will result in faster temperature changes. The dimensions of the Peltier module need to be considered, as they define the dimensions of the heat-transfer block (HTB) and the sample line that passes through the HTB. Larger or multiple Peltier modules will allow larger sections of sample line to be regulated. With the above considerations, one excellent Peltier module is the CZ1-1.4-1271.14 from Tellurex. This Peltier module is a single-stage, 4-cm-square unit with a ∆T of 79°C, and a Qmax of 78W. It has been successfully used to rapidly heat and cool sections of a flow cytometry sample line (Graves et al., 2001). Temperature control circuit The function of the control circuit is to provide power to and sense temperature from the inline thermoregulation unit. The direction of heat transfer across a Peltier module is determined by the polarity of the current running
Flow Cytometry Instrumentation
1.18.13 Current Protocols in Cytometry
Supplement 20
through it, and the temperature of the heattransfer block is regulated by rapidly switching the polarity of the power to the Peltier module. Commercially available controllers can regulate the temperature of the inline thermoregulation unit. However, these units are, in general, expensive and do not have advanced control capabilities such as ramping and time-based step control. With this in mind, it is simple to construct a Peltier-module temperature-control circuit with advanced functionality. The temperature-control circuit has two major elements: a process controller and a polarity switching circuit. The process controller Temperature control in other industries is very advanced; therefore, many versatile process-control computers on the market are capable of regulating the temperature of the inline thermoregulation unit. The process controller must have (minimally) two switched DC control circuits (one for heating and one for cooling) and provide a temperature level output signal. It will preferentially have internal programmability allowing for ramping, timebased steps, and external communication to allow for integration with other computers. An excellent controller that fulfills all these requirements is the series F4 1/4 DIN ramping controller (Watlow Electric).
Dynamic Thermoregulation of the Sample in Flow Cytometry
DNA melting experiments Dynamic sample thermoregulation is an important tool for the mechanistic analysis of many polynucleotide-modifying enzymes. Microsphere-based analysis of DNA-modifying enzyme products is an attractive alternative to gel-based methods (Nolan and Sklar, 1998). If oligomer pairs are used as substrates, modification generally results in a change in the melting temperature of the DNA. This change can be utilized to track the enzymatic functions. Furthermore, flow cytometry is an important tool for the analysis of genomic variation. It is currently being used for the identification of single-nucleotide polymorphisms (Cai et al., 2000), the most common form of human genetic variation. The second most common disease-causing form of genetic variation is expansion of triplet repeats within the genome (McMurray, 1999). Because it is possible to estimate the length of a piece of DNA based on its melting temperature (Ririe et al., 1997), expanded triplet repeats conceivably could be detected by the analysis of melting curves for an individual’s DNA. Dynamic, inline sample
thermoregulation will enable the calculation of the melting temperature of triplet-repeat DNA samples attached to microspheres. Coupled with the ability of flow cytometry to analyze a large array of microspheres simultaneously (Kettman et al., 1998), this will allow highthroughput screening for expanded triplet repeats. The above reasons make a protocol to measure the melting temperature of DNA duplexes via flow cytometry very valuable.
Critical Parameters and Troubleshooting The sample line The sample line must be heat conductive and preferably biologically inert. Titanium tubing is the best choice, as steel tubing is not inert, and polymer tubing (e.g., PEEK and Teflon) does not have high heat conductivity. Narrowdiameter tubing (in both outer and inner diameter) will ensure complete heat transfer through the sample. Tubing of 1⁄16-in. o.d. and 0.01-in. i.d. has been successfully used (Graves et al., 2001). However, while larger diameters of sample line have not been tested, their use may allow for greater volumes of sample to be temperature regulated. Delay and regulation volume In the above design, the delay volume is significant. Even at the fastest flow rates, the sample will experience a significant temporal delay between heating and analysis (∼6 sec) and will drift to ambient temperature for this duration prior to analysis. To make the instrument fully functional, this delay time must be taken into account. The delay time results from the volume of the line between the inline temperature regulation unit (ITRU) and the analysis point. Therefore, the parameters recorded for each particle analyzed by the flow cytometer are delayed relative to the recorded temperature that it experienced at the ITRU. The delay time is a function of the volume of the sample between the heat-transfer block’s final regulation point and the analysis point, which is defined as the “delay volume,” and the flow rate of the sample. The sample flow rate either is known because sample is delivered via a syringe pump or can be determined from the particle count rate of a well characterized sample. The delay volume can be estimated from the tubing specifications of the sample line and other fittings between the ITRU and the analysis point. However, direct measurement of the delay volume is desirable. For reactions that respond slowly
1.18.14 Supplement 20
Current Protocols in Cytometry
to temperature changes, the delay volume is of little importance. However, for fast responding systems it needs to be noted. Conversely, for slow-responding samples, longer incubation within the regulated volume than is possible through the small sample line described above may be necessary. By using a combination of multiple Peltier modules, curving sample lines, and larger diameters, it should be possible to regulate larger sections as needed for different experimental systems. Continuous triggering Continuous triggering of data acquisition is important to calibrate some aspects of the temperature data acquisition channel (see above). Some instruments have test triggering capabilities that can be set to trigger data acquisition in a preset fashion; if these are available, they should be used. However, it is also possible to trigger a system by using a concentrated solution of inexpensive microspheres where varying the concentration will change the trigger rate. Alternatively, the system can also be triggered continuously by reducing the trigger threshold until the noise of the system begins to trigger the acquisition system. Buffers The pH of some buffers is temperature dependent. Therefore, in experiments where pH sensitivity is critical, care should be taken in buffer selection. The delay volume The delay volume can be calculated in several different ways. First, it can simply be estimated from the known dimensions of the system. While this can be done quickly, it is unlikely to yield a highly accurate result. Second, it can be measured using other instruments such as a highly accurate scale. By this method, the fluidics past the thermoregulation unit can be weighed with and without water filling the lines and the difference in weight can then be used to calculate the volume of the lines. This would be tedious and difficult. Therefore, the authors devised a third method to calculate the delay volume using the instrument itself. In this method, a reaction that changes in a predictable manner is used to calculate the delay volume. In the example given above, the reaction was DNA melting, but it can be any reaction that predictably affects a fluorescence measurement in response to temperature changes. As shown above, by changing the sample delivery rate, the time for a predictable temperature
change to arrive can be extrapolated directly from the data. The important caveat is that the change that occurs must be essentially irreversible (DNA re-annealing is very slow and therefore, on the timescales shown here, DNA melting is essentially irreversible). Using varying ramp rates for the same reaction allows the delay time for a particular sample-delivery rate to be extracted from the data. In both cases, the chosen optical event was a 50% loss of fluorescence (this corresponds with the melting temperature of the oligomers), but it could have just as easily have been 25%, 75%, 100%, or any other percentage as long the value was a consistent one.
Anticipated Results A flow cytometer equipped with the Peltier module–based thermoregulation unit described here will be able to perform thermodynamic experiments for reactions that occur within a few seconds and that are irreversible on the same timescale. It will also be useful for temperature jump reactions, where an increase in temperature starts a desired reaction. The unit described here could be used in multiples or with a more complex fluidic design to easily provide a greater volume of thermoregulation, and allow slower reactions to be controlled for longer time frames. By following and expanding the basic design and experimental premises described above it will be possible to have dynamic thermoregulation of the sample throughout the flow cytometer.
Time Considerations Once the Peltier module–based thermoregulation unit has been constructed, the delay volume measured and the appropriate spreadsheet designed, it will take ≤1 min per data file to shift the data and obtain correlated temperature values. Spreadsheet design will take ∼1 day. Construction of the Peltier module–based thermoregulation unit can be broken down into the machining time (2 to 3 hr), electronic technician time (∼2 to 3 days), and interfacing the unit with the flow cytometer (1 to 2 days). Calculation of the delay volume including solution and microsphere preparation will take ∼3 days.
Literature Cited Cai, H., White, P.S., Torney, D., Deshpande, A., Wang, Z., Keller, R.A., Marrone, B., and Nolan, J.P. 2000. Flow cytometry-based minisequencing: A new platform for high-throughput singlenucleotide polymorphism scoring. Genomics 66:135-143.
Flow Cytometry Instrumentation
1.18.15 Current Protocols in Cytometry
Supplement 20
Graves, S.W., Habbersett, R.C., and Nolan, J.P. 2001. A dynamic inline sample thermoregulation unit for flow cytometry. Cytometry 43:23-30. Kelley, K.A. 1991 Very early detection of changes associated with cellular activation using a modified flow cytometer. Cytometry 12:464-468. Kettman, J. R., Davies, T., Chandler, D., Oliver, K.G., and Fulton, R.J.1998. Classification and properties of 64 multiplexed microsphere sets. Cytometry 33:234-243. McMurray, C.T. 1999. DNA secondary structure: A common and causative factor for expansion in human disease. Proc. Natl. Acad. Sci. U.S.A. 96:1823-1825. Nolan, J.P. and Sklar, L.A. 1998. The emergence of flow cytometry for sensitive, real-time measurements of molecular interactions. Nat. Biotechnol. 16:633-638.
Omann, G.M., Coppersmith, W., Finney, D.A., and Sklar, L.A. 1985. A convenient on-line device for reagent addition, sample mixing, and temperature control of cell suspensions in flow cytometry. Cytometry 6:69-73. Ririe, K.M., Rasmussen, R.P., and Wittwer, C.T. 1997. Product differentiation of DNA melting curves during the polymerase chain reaction. Anal. Biochem. 245:154-160. Rowe, D.M. 1995. CRC Handbook of Thermoelectrics. CRC Press, Boca Raton, Florida.
Contributed by Steven W. Graves, Robert C. Habbersett, and John P. Nolan Los Alamos National Laboratories Los Alamos, New Mexico
Dynamic Thermoregulation of the Sample in Flow Cytometry
1.18.16 Supplement 20
Current Protocols in Cytometry
Excitation and Emission Spectra of Common Dyes In cytometry, intrinsic cellular parameters are defined as those that can be measured without the use of an added reagent. Measurement of extrinsic parameters requires the use of added reagents, which are almost always referred to as probes. Fluorescent probes allow measurement of the widest variety of extrinsic cellular parameters. The list of probes now in common use includes dyes with relatively low molecular weights (typically <1000 kD), oligopeptides and oligonucleotides, and macromolecules, e.g., fluorescent proteins and the phycobiliproteins; the latter have molecular weights as high as 240,000 kD. The chemical properties of a probe determine the nature and specificity of its interaction with a cell and/or a specific molecular target. Some probes are themselves fluorescent; others are made fluorescent by the addition of a fluorescent label, permitting detection and quantification of the amount of cell-associated probe based on the amount of fluorescence measured from the label.
FLUORESCENCE BASICS In order for a molecule to fluoresce, it must first absorb energy in the form of a photon, raising an electron to a higher energy level, or excited state. Fluorescence occurs when the electron loses some or all of the absorbed energy by emission of a photon. The fluorescence lifetime, i.e., the period between excitation and emission, is typically on the order of a few nanoseconds for fluorescent organic dyes and proteins, but is notably longer for other materials. In almost all cases, some excitation energy is lost nonradiatively, and the emitted energy is less than the energy absorbed, meaning that the fluorescence emission is at a longer wavelength than the excitation. The difference between the principal excitation and emission maxima in a fluorescence spectrum is known as the Stokes shift; George Stokes first described fluorescence in the mid-1800s. Typical Stokes shifts are no more than a few tens of nanometers. Fluorescence is an intrinsically quantum mechanical process. The probability that a molecule will absorb photon energy is quantified by its extinction coefficient. The quantum yield and quantum efficiency of fluorescence are, respectively, the number and fraction or Contributed by Howard M. Shapiro Current Protocols in Cytometry (2003) 1.19.1-1.19.7 Copyright © 2003 by John Wiley & Sons, Inc.
UNIT 1.19
percentage of photons emitted per photon absorbed; they typically increase with the extinction coefficient, but are also dependent on the relative probabilities of the excited molecule’s losing energy by fluorescence emission and by nonradiative mechanisms. The quantum yields of some dyes used in cytometry are quite high, >0.5, but quantum yield, particularly for an organic molecule, is affected by the chemical environment in which the molecule finds itself. If an excited molecule that might otherwise fluoresce instead loses energy nonradiatively, by collision with solvent molecules for example, it is said to be quenched; once returned to the electronic ground state, it can be re-excited. However, light absorption may be followed by a change in molecular structure, making further cycles of fluorescence excitation and emission impossible; this is called (photo)bleaching. In general, it is possible to get only a finite number of cycles of excitation and emission out of each fluorescent molecule, or fluorophore, before photobleaching occurs. If two fluorophores A and B are in proximity, and the emission spectrum of A overlaps the excitation spectrum of B, nonradiative energy transfer between the two may occur, with the result that illumination at wavelengths that normally excite the donor, A, produces fluorescence from the acceptor, B. The likelihood of energy transfer increases with the degree of spectral overlap and diminishes with increasing distance between the fluorophores. Intramolecular energy transfer determines the fluorescence spectra of phycobiliproteins and their tandem conjugates.
FLUORESCENT PROBES AND LABELS EVOLVE WITH CYTOMETERS In recent years the trend in both flow and static cytometry has been toward measurement of an increasingly large number of characteristics for each cell subjected to analysis. Such multiparameter cytometry requires optical systems that separate fluorescence emission from cells into multiple spectral bands, and hardware and/or software fluorescence compensation to permit fluorescence contributions from probes with overlapping emission spectra to be resolved and quantified (UNITS 1.14 & 10.15). Flow Cytometry Instrumentation
1.19.1 Supplement 26
Excitation and Emission Spectra of Common Dyes
The best single reference on fluorescent probes is Haugland’s Handbook of Fluorescent Probes and Research Products (Haugland, 2002). This indispensable work, now in its ninth edition and available on CD-ROM and online as well as in print form, is the catalog of Molecular Probes, a subsidiary of Invitrogen and a major supplier. Practical Flow Cytometry (Shapiro, 2003) contains extensive discussions of fluorescence and fluorescent probes and their applications, and of the hardware and software used for multiparameter fluorescence measurements. The spectral characteristics of a representative sample of fluorescent dyes, probes, and labels, and the excitation wavelengths available from various light sources, are shown in Figure 1.19.1. Although there are thousands of compounds in suppliers’ catalogs, with some labels conjugated to any one of hundreds of different monoclonal antibodies, it is probable that a majority of the samples analyzed by flow cytometers contain at least one of the dyes, probes, or labels shown in Figure 1.19.1. Consideration of Figure 1.19.1 makes apparent the rationale for the use of multiple, spatially separated fluorescence excitation beams in multiparameter flow cytometry. Even when one can choose from a number of probes to select those with desired spectral characteristics, the use of multiple beams generally facilitates resolution of fluorescence signals from multiple probes. When a choice of probes is not available, multibeam flow cytometry may provide the only means of making correlated measurements of two or more parameters of interest. Most fluorescence flow cytometers use a single light source, i.e., an argon-ion laser emitting blue-green (488-nm) light. The majority of dual-laser instruments add a red (633- to 640nm) helium-neon or diode laser. UV (325- to 365-nm) excitation is common in instruments with arc-lamp sources and in larger benchtop flow cytometers and sorters. The newest instruments are offered with violet (395- to 415-nm) diode lasers, which provide excitation in a wavelength region previously available only from large, expensive, water-cooled kryptonion lasers. They can measure 488 nm–excited fluorescence in as many as six discrete spectral regions, typically centered near 525 nm (green), 575 nm (yellow), 610 nm (red-orange), 670 nm (red), 710 nm (far red), and 780 nm (near infrared), allowing discrimination of signals from cell-bound probes labeled with, e.g., fluorescein, phycoerythrin (PE), a PE-Texas
Red tandem conjugate, perCP, a perCP-Cy5.5 tandem, and a PE-Cy7 tandem. A three-beam cytometer with 488-nm, violet, and red excitation can add at least two channels for violet-excited fluorescence measurement (e.g., 430 and 530 nm for Cascade Blue and Cascade Yellow) and three for red-excited fluorescence measurement (e.g., 670 nm for allophycocyanin (APC), 710 nm for an APC-Cy5.5 tandem, and 780 nm for an APC-Cy7 tandem), making 11-color fluorescence measurements possible in a stock commercial apparatus. However, considerable skill is required in selecting probes, labels, and filters for such experiments, and in analysis of the data produced.
PROBES AND SPECTRA The first 19 spectra shown in Figure 1.19.1 are those of labels almost all widely used and readily available conjugated to antibodies and/or nucleic acid probes. The next twelve spectra are of nucleic acid dyes. Then come four spectra of fluorescent reporter proteins, and finally, spectra of indo-1, a reliable and widely used calcium probe, in the presence and absence of calcium ions. The list of light sources in the figure includes Nd:YAG and semiconductor (diode) lasers as well as argon- and krypton-ion lasers, He-Cd and He-Ne lasers, and the mercury arc lamp. Long vertical lines corresponding to the popular 488- and 633- to 635- nm excitation wavelengths are drawn through the spectra for orientation.
LOW-MOLECULAR-WEIGHT LABELS A number of UV-excited blue fluorescent labels are available. The first popular coumarin label was 7-amino-4-methylcoumarin-3-acetic acid, or AMCA, which excites maximally at ∼350 nm, and has an emission maximum near 455 nm. Molecular Probes’ Alexa 350 has a similar spectrum, but a higher quantum yield. Cascade Blue, a reactive derivative of pyrene, was also introduced by Molecular Probes; its excitation maximum is near 390 nm, with maximum emission at ∼415 nm. This dye and Molecular Probes’ Cascade Yellow (emissions maximum near 550 nm) are effectively excited by violet diode lasers. Fluorescein, which emits at ∼520 nm, is by far the most popular fluorescent label; its excitation maximum is very close to the popular 488-nm argon-ion laser wavelength, its quantum efficiency is high, and it had been in widespread use before flow cytometers became available. However, the effective quantum effi-
1.19.2 Supplement 26
Current Protocols in Cytometry
PROBE FLUORESCENCE SPECTRA excitation
>10% >50% max
emission
max >50% >10%
λ (nm) 300
350
400
450
500
550
600
650
700
750
800 850
AMCA/ALEXA 350 CASCADE BLUE CASCADE YELLOW CY2/DiOCn(3) FLUORESCEIN PHYCOERYTHRIN (PE) PE-TEXAS RED PE-CY5 PE-CY5.5 PE-CY7 PERCP PERCP-CY5.5 TETRAMETHYLRHODAMINE CY3/PKH26/DiICn(3) TEXAS RED/SR101 ALLOPHYCOCYANIN (APC) CY5/DiICn(5) CY5.5 APC-CY7 ETHIDIUM BROMIDE PROPIDIUM IODIDE (PI) HOECHST 33342 DAPI MITHRAMYCIN CHROMOMYCIN A3 ACRIDINE ORANGE (AO) TO-PRO-1/TOTO-1/TO PYRONIN Y 7-ACTINOMYCIN D (7-AAD) TO-PRO-3/TOTO-3 DRAQ5 ECFP EGFP EYFP DsRED INDO-1 (with Ca++) INDO-1 (no Ca++) λ (nm) 300 SOURCE
350
400
450
500
550
600
650
700
750
800 850
ARGON ION LASERS KRYPTON ION LASERS HELIUM-NEON LASERS HELIUM-CADMIUM LASERS NEODYMIUM YAG LASERS SEMICONDUCTOR LASERS MERCURY ARC LAMP
Figure 1.19.1 Probe fluorescence spectra and source emission wavelengths.
Flow Cytometry Instrumentation
1.19.3 Current Protocols in Cytometry
Supplement 26
Excitation and Emission Spectra of Common Dyes
ciency of fluorescein varies substantially with pH; other dyes (e.g., Alexa 488) share fluorescein’s spectral characteristics but not its environmental sensitivity, and thus may be better labels. Tetramethylrhodamine, while used to provide an antibody label distinguishable from fluorescein by fluorescence microscopy, is very poorly excited at 488 nm, and is not widely used in flow cytometry. It is suitable for use with instruments using green excitation sources, and emits at ∼570 nm. The first label combination widely used for two-color fluorescence flow cytometry paired fluorescein with rhodamine 101, conjugated in reactive form as an isothiocyanate (XRITC) or a sulfonyl chloride (Texas Red). Rhodamine 101 emits near 615 nm; it is best excited at wavelengths between 565 and 595 nm, and the fluorescein-rhodamine combination thus can be used only in instruments that have a suitable source such as a krypton or dye laser in addition to a 488-nm excitation source. Sulforhodamine 101 (SR101) is a rhodamine 101 derivative used as a stain for total cellular protein; in this application, there is usually enough dye bound per cell to get a fairly strong signal using 488-nm excitation. The “Cy dyes” are a series of reactive derivatives of symmetric cyanine dyes. Cy2, derived from oxacarbocyanine [DiOCn(3)], has absorption and emission spectral characteristics similar to those of fluorescein. Cy3, a derivative of indocarbocyanine [DiICn(3)], excites maximally at ∼545 nm, but can be excited adequately at 488 nm in some applications. The emission peak of Cy3 is at ∼565 nm; however, a substantial fraction of Cy3 emission is transmitted by the 575-nm bandpass filters typically used for PE detection. The “tracking dye” PKH26, which binds tenaciously within lipid membrane bilayers, is also an indocarbocyanine derivative, with long alkyl side chains. Symmetric cyanines with shorter alkyl side chains are used for cytoplasmic and mitochondrial membrane potential estimation (UNIT 9.6). Cy5, derived from indodicarbocyanine [DiICn(5)], absorbs maximally near 640 nm; it is very effectively excited by 633-nm He-Ne lasers or 635- to 640-nm diode lasers and emits at ∼660 nm. Cy5.5 is a reactive derivative of dibenzoindodicarbocyanine, with maximal absorption near 675 nm and maximal emission at 695 to 700 nm; its absorption at 633 nm is sufficient to make it feasible to use Cy5- and Cy5.5-labeled antibodies for two-color immunofluorescence analysis. Cy7 is a reactive derivative of indotricarbocyanine [DiICn(7)]; it
absorbs in the near infrared (∼750 nm) and emits at ∼770 nm, and is used primarily in tandem conjugates (see below). The Cy dyes, particularly Cy3 and Cy5, have become popular as labels for nucleic acids and oligonucleotides in applications such as gene array scanning and fluorescence in situ hybridization (FISH). However, antibodies labeled with Cy dyes seem to adhere to monocytes and to a lesser extent to granulocytes, resulting in low levels of irrelevant staining; antibody manufacturers have come up with various proprietary ways of minimizing such binding. Molecular Probes offers the Alexa dyes, a series of rhodamine-based labels with spectral characteristics similar to those of some of the more popular labels previously mentioned, e.g., AMCA, fluorescein, Texas Red, Cy3, Cy5, Cy5.5, and Cy7. However, the Alexa dyes have higher quantum yields, better photostability, and better charge characteristics, allowing more dye molecules to be put on a protein molecule, and do not appear to bind to monocytes. Alexa dyes are now being offered by several antibody manufacturers, both as direct labels and in tandem conjugates.
PHYCOBILIPROTEIN AND TANDEM CONJUGATE LABELS The phycobiliproteins play critical roles in the function of the photosynthetic apparatus in red algae and cyanobacteria, participating in nonradiative energy transfers that make bluegreen light energy available to chlorophyll. Phycoerythrins absorb blue-green and green light, and allophycocyanins orange and red light. These molecules are all highly fluorescent; each phycobiliprotein molecule contains a large number of bilin chromophores. The extinction coefficients of phycobiliproteins are extremely high, and the quantum yields are also high. Phycobiliproteins, and phycoerythrins in particular, are characterized by broad shoulders in their excitation spectra, allowing them to be excited effectively at wavelengths substantially below their emission maxima; with excitation at 488 nm, a phycoerythrin-labeled antibody molecule will emit several times as much fluorescence as a fluorescein-labeled antibody molecule. The peak absorption of R-phycoerythrin (RPE) is at 565 nm, with the emission maximum at 578 nm; R-PE can thus be used very effectively in combination with fluorescein for twocolor immunofluorescence flow cytometry using only a single 488-nm excitation beam. PElabeled antibodies have been widely available
1.19.4 Supplement 26
Current Protocols in Cytometry
since the late 1980s. Allophycocyanin (APC) exhibits high (∼75% of maximum) absorption in the 633- to 645-nm range in which red He-Ne and diode lasers operate. The absorption maximum of APC is at 650 nm; its emission maximum is at 660 nm. It is possible to prepare a tandem conjugate of PE covalently linked to APC in which energy transfer between the phycoerythrin donor and the allophycocyanin acceptor results in strong emission at 660 nm on excitation at wavelengths between 470 and 570 nm. The phycobiliprotein tandem conjugates now in the widest use incorporate only a single phycobiliprotein molecule, to which are conjugated several molecules of a lower-molecular-weight fluorochrome. The first conjugates prepared in this fashion incorporated phycoerythrin and Texas Red. While most of the emission from PETexas Red conjugates is in the 610- to 620-nm emission region of Texas Red, incomplete energy transfer results in some emission from the conjugates in the PE emission region at ∼580 nm, and PE itself has substantial emission in the 610- to 620-nm range. As a result, a lot of fluorescence compensation must be applied to separate the fluorescence signals from a PE-labeled antibody and another antibody labeled with a PE-Texas Red tandem conjugate. PE-Cy5 tandem conjugates comprise a single phycoerythrin molecule and several molecules of Cy5, and are preferable to PE-Texas Red conjugates as a third label for immunofluorescence analyses using 488-nm excitation; they emit at the emission maximum of Cy5, near 660 nm. Other phycoerythrin tandem conjugates are PE-Cy5.5 (emission maximum near 700 nm) and PE-Cy7 (emission maximum near 770 nm). Allophycocyanin tandem conjugates can be used in conjunction with APC itself for multicolor immunofluorescence measurements employing a red laser source. APC-Cy7 emits maximally near 770 nm, and APC-Cy5.5 near 700 nm. Peridinin chlorophyll protein, or PerCP, is a component of a dinoflagellate photosynthetic apparatus; it excites maximally near 490 nm and has a relatively sharp emission peak at ∼680 nm. The sharpness of the emission peak minimizes crosstalk between PerCP and PE, and therefore also minimizes the amount of fluorescence compensation needed. However, PerCP is relatively intolerant of high illumination power levels, because it may enter a longlived triplet state (the same problem is noted to a much lesser extent with PE and APC). This
problem is eliminated in a PerCP-Cy5.5 tandem conjugate, which has maximum emission near 700 nm. The probability that energy transfer between a donor and an acceptor will occur varies with the extent of overlap of the donor emission spectrum and the acceptor excitation spectrum. This diminishes markedly from PE-Texas Red to PE-Cy5 to PE-Cy5.5 to PE-Cy7. As a result, energy transfer in this series is progressively less efficient, and the longer wavelength–emitting tandems also exhibit more and more emission in the spectral range in which PE normally emits. This means that a substantial fraction of the PE chromophores will not donate energy to Cy7, and that there will not be as much emission from Cy7 in PE-Cy7 as there will from Texas Red in PE-Texas Red or from Cy5 in PE-Cy5. It is therefore inappropriate to use PE-Cy7 or other similarly inefficient labels to attempt to discriminate cells bearing small amounts of target ligands.
NUCLEIC ACID DYES DNA-Selective Dyes An ideal dye for measurement of DNA content would be DNA specific; i.e., it would form a fluorescent complex with DNA, but not with RNA or other macromolecules. It would also not exhibit any base or sequence preference; i.e., the fluorescence from a given number of dye molecules bound to a given number of base pairs’ worth of DNA would be the same, regardless of the relative proportions of A-T and G-C base pairs. A number of dyes are highly DNA selective. The UV-excited (excitation maximum at ∼350 nm), blue fluorescent (emission maximum at ∼450 nm) bisbenzimidazole dyes Hoechst 33342 and 33258 bind to sequences of A-T base pairs. The strong A-T base preference accounts for the popularity of Hoechst 33258 in combination with chromomycin A3, which has a G-C preference, for bivariate flow cytometric analysis of chromosomes. Hoechst 33342 is the only compound that has been extensively used for flow cytometric determination of DNA content in living cells. DAPI (4′,6-diamidino-2-phenylindole) is highly DNA selective, with fluorescence properties similar to those of the Hoechst dyes and a strong A-T base preference. In fixed or permeabilized cells, DAPI often yields DNA histograms with lower G0/G1 peak coefficients of variation (CVs) than are obtained using other dyes. Both DAPI and the Hoechst dyes can be
Flow Cytometry Instrumentation
1.19.5 Current Protocols in Cytometry
Supplement 26
excited by violet diode lasers, albeit somewhat inefficiently. The antitumor antibiotic mithramycin and the structurally related antibiotic chromomycin A3 are highly DNA selective and stain G-C rich regions of DNA. The excitation maxima of the DNA complexes of both dyes are at ∼440 nm; the maximum emission wavelength of the chromomycin A3 complex is at ∼555 nm, while that of the mithramycin complex is somewhat longer. 7-Aminoactinomycin D (7-AAD), a fluorescent analog of the antitumor antibiotic actinomycin D, is highly DNA selective, with a G-C base preference. The complex of this dye with DNA absorbs maximally near 550 nm and emits maximally near 660 nm; 7-AAD can be excited effectively at 488 nm. DRAQ5 is an anthraquinone dye which when bound to DNA has an excitation maximum near 650 nm and an emission maximum near 700 nm. It appears to be reasonably DNA selective, and produces reasonably good DNA content histograms when it is applied to intact cells, making it the only dye other than Hoechst 33342 usable for that purpose. Some manipulation of relative concentrations of dye and cells may be necessary to obtain the best-quality DNA content measurements. DRAQ5 can be excited at 488 nm as well as with red lasers.
Dyes that Stain Both DNA and RNA
Excitation and Emission Spectra of Common Dyes
Ethidium bromide and propidium iodide (PI) were among the first dyes used for DNA content determination by flow cytometry. Both are excited at 488 nm; the original rationale for the use of propidium lay in the fact that its emission maximum (∼615 nm) is 10 to 15 nm farther into the red region of the spectrum than that of ethidium, making it easier to separate red and green fluorescence signals from propidium and fluorescein using optical filters. Neither ethidium nor propidium is DNA selective; both form fluorescent complexes with double-stranded DNA or RNA. Neither dye exhibits a strong base preference. Asymmetric cyanine dyes have been widely used for nucleic acid staining since the mid1980s; thiazole orange (TO; 1,3′-dimethyl4,2′-quinothiacyanine) was developed for use in blood reticulocyte analysis in instruments using 488-nm light sources (UNIT 7.10). When bound to RNA, thiazole orange has an absorption maximum at 509 nm and an emission maximum at 533 nm, and its fluorescence quantum efficiency is increased ∼3000 times over that of the free dye. Thiazole blue (1,3′-di-
methyl-4,2′-quinothiacarbocyanine) also behaves as a DNA and RNA fluorochrome and when bound to nucleic acid has an excitation maximum at ∼640 nm, making it useful with red laser sources. Both thiazole orange and thiazole blue readily enter intact cells; Molecular Probes has developed derivatives of both dyes for a variety of applications. TO-PRO-1 and TO-PRO-3, respectively, share the thiazole orange and thiazole blue ring structure and spectrum, but are excluded from intact cells. TOTO-1 and TOTO3 are, essentially, dimers of TO-PRO-1 and TO-PRO-3; the binding affinity of these dyes for DNA is sufficiently high so that fragments labeled with TOTO-1 or YOYO-1 and with ethidium can be mixed and separated by electrophoresis.
Acridine Orange (AO) and Pyronin Y Under carefully controlled conditions, the blue-excited green fluorescence of acridine orange (AO) molecules intercalated into DNA can be used to provide good estimates of cellular DNA content, while the red metachromatic fluorescence of complexes of the dye with single-stranded RNA or denatured DNA, respectively, can provide indications of RNA content or of chromatin structure (UNITS 7.3 & 7.8). Measurements of RNA and DNA content define subcompartments of the cell cycle. Patterns of cellular DNA and RNA content observed using AO staining can also be demonstrated with other dyes, notably Hoechst 33342 and pyronin Y (UNIT 7.3). Pyronin Y, used as an RNA stain, can be excited at 488 nm, and detected using the 575-nm band-pass filter normally used for phycoerythrin.
FLUORESCENT REPORTER PROTEINS Genes encoding intrinsically fluorescent proteins can be co-transfected with other genes of interest to provide reporters of cell transfection status (UNIT 9.12; also see UNIT 1.16). The first protein to be used was green fluorescent protein (GFP) from the bioluminescent jellyfish Aequorea victoria. GFP is maximally excited at 395 nm, but can be excited moderately effectively at 488 nm, where its absorption is about one-third maximum; the emission spectrum has a sharp peak at 510 nm with a shoulder at 540 nm. A number of fluorescent proteins with spectral characteristics more suitable for flow cytometry have been developed. ECFP (cyan; excitation maximum 434 nm, emission maxi-
1.19.6 Supplement 26
Current Protocols in Cytometry
mum 477 nm), EGFP (green; excitation maximum 489 nm, emission maximum 508 nm), and EYFP (yellow-green; excitation maximum 514 nm, emission maximum 527 nm) are all produced by mutants of the Aequorea GFP gene. DsRed (orange; excitation maximum 558 nm, emission maximum 583 nm) is derived from a coral of the species Discosoma. Constructs needed to introduce the genes into cells are available from the Clontech division of BD Biosciences.
INDO-1: A PROBE OF CALCIUM CONCENTRATION Indo-1 is a selective calcium chelator, but does not significantly perturb cellular calcium metabolism. Its fluorescence is excited by UV light (325 to 365 nm) The free dye shows maximum emission at ∼480 nm, whereas the calcium chelate emits maximally at ∼405 nm. The ratio of emission intensities at 405 and 480 nm in cells loaded with indo-1 thus provides an indication of cytoplasmic [Ca++] (UNIT 9.8). Indo-1 is not excitable by long-wavelength (≥370 nm) UV light, although it may be possible to synthesize a related compound that could be used for calcium studies with the newly available 370-nm UV diode lasers.
SOMETHING COMPLETELY DIFFERENT: QUANTUM DOTS Semiconductor nanocrystals, better known as quantum dots, now appear to be becoming practical as labels. In semiconductors, absorption of a photon results in the loss of an electron by one atom of the material, while another atom nearby gains one. Some of the absorbed energy is subsequently lost by emission of a photon. In crystals smaller than ∼10 nm, the emission wavelength is determined primarily by the size of the crystal. The emission wavelength of a CdSe crystal with a diameter of 2.1 nm is ∼510 nm; that of a crystal of the same material with a 3.1 nm diameter is ∼560 nm. Larger crystals have longer emission wavelengths. Organic dyes typically have small Stokes shifts, and the excitation spectrum and emis-
sion spectrum of an organic dye often resemble mirror images of one another, with the latter having a “tail” extending toward longer wavelengths. The emission spectrum of a quantum is typically sharp and symmetric; the excitation spectra of nanocrystals are relatively independent of emission wavelength, with progressively shorter wavelengths increasingly effective for excitation. Excitation at 400 nm is typically at least twice that at 488 nm. This suggests that violet diode lasers will be useful for work with quantum dots, in either scanning or flow cytometers. Nanocrystals are much less susceptible to photobleaching than are organic dyes, and also have much higher absorption. Although the quantum efficiencies of nanocrystals and dyes are similar, the fluorescence from a nanocrystal is typically equivalent to the fluorescence from a dozen or more dye molecules. Some practical problems that have prevented widespread use of nanocrystal labels, such as hydrophobicity and conjugation, have been solved; some remain. Streptavidin-conjugated quantum dots with 605- and 655-nm emission wavelengths are now commercially available from Quantum Dot Corporation, but it is still too early to determine the extent to which the materials will change cytometric practice.
LITERATURE CITED Haugland, R.P. 2002. Handbook of Fluorescent Probes and Research Products, 9th ed. pp. 966. Molecular Probes, Eugene, Oreg. Shapiro, H.M. 2003. Practical Flow Cytometry, 4th ed. pp. 681. Wiley-Liss, New York.
INTERNET RESOURCES http://www.probes.com Location for the online version of the Handbook of Fluorescent Probes and Research Products.
Contributed by Howard M. Shapiro Howard M. Shapiro, M.D., P.C. West Newton, Massachusetts
Flow Cytometry Instrumentation
1.19.7 Current Protocols in Cytometry
Supplement 26
CHAPTER 2 Image Cytometry Instrumentation INTRODUCTION
T
his chapter is devoted to image cytometry instrumentation. In image cytometry a specimen or object of interest is placed on a glass slide or other rigid substrate and stained using fluorescence- or absorbance-based probes specific for the cellular substance or substances of interest. Analyses may be performed on tissue sections or individual cells. Image cytometry can be used to quantify probes (similar to flow cytometry) as well as to obtain morphometric and other information as permitted by its high optical resolution. Images may be acquired in two or three dimensions, permitting quantification and study of the distribution of substances throughout cells and tissues. The basic image cytometer consists of a microscope, a camera, a computer, and a monitor. Additional components such as a motorized stage, a motor for automated image focus, automated changing of objectives, and filter wheels for rapid selection of illumination and measurement wavelengths may be added depending upon the specific application(s). Instruments providing automated slide changing and even fully automated operation are also available.
Fundamental to any image cytometer is the optical microscope. The many types of instruments and techniques used for contrast enhancement in light microscopy are reviewed in UNIT 2.1. This unit also covers the basic concepts of light microscopy, including Köhler illumination, resolution, contrast, and numerical aperture (NA). Although brightfield and fluorescence microscopy are typically employed in quantitative image cytometry, other microscopy techniques are often used for image localization. Image cytometry is often referred to as high-resolution image cytometry because of the high-resolution optical features available for quantification and analysis. The resolution achievable in any microscope is determined by the NA of the optical system. Microscope objectives, the most crucial image-forming components of a microscope, are discussed in UNIT 2.2. The issues covered include aberrations in image formation and their correction, the many types of objectives available, and the advantages and limitations of each; a knowledge of the different types of objectives and their characteristics is critical to the selection of appropriate objectives for image cytometry. Cameras used in image cytometry are described in UNIT 2.3. The camera is the essential component responsible for converting the optical image into digital form for subsequent analysis. Modern cameras provide high sensitivity, low noise, and excellent linearity to permit the capture of large images while maintaining high resolution. Another important component of an image cytometer is the optical filter set used for wavelength selection. This is particularly true in quantitative fluorescence microscopy, where optical filters are used to select wavelengths of light for exciting fluorophore probes and for discriminating their fluorescence signals. Selection of appropriate filters is also critical in achieving adequate signal-to-noise ratios. Optical filtering systems for wavelength selection in fluorescence microscopy are discussed in UNIT 2.4. Biomedical research and clinical medicine make increasing use of image cytometry, especially digital fluorescence microscopy (UNIT 2.5). Quantification of intensity and Contributed by Leon L. Wheeless Current Protocols in Cytometry (2001) 2.0.1-2.0.2 Copyright © 2001 by John Wiley & Sons, Inc.
Image Cytometry Instrumentation
2.0.1 Supplement 15
intracellular location of fluorescently labeled antibodies and nucleic acid probes are of particular interest. To properly assess data obtained from a microscope system, it is necessary to know the system’s sampling density and spatial resolution. These important parameters are determined through the careful use of calibration standards and the appropriate image processing algorithms, as discussed in UNIT 2.6. Light microscopes will deliver optimal performance only if their various components are clean, properly installed, and aligned. UNIT 2.7 discusses the important topic of microscope alignment, including steps in setting up Köhler illumination. The maintenance of optical glass surfaces is also covered. If a microscopy specimen is thicker than the depth of focus of the objective lens, light coming from structures above and below the plane of focus will enter the detector (eye or camera). This out-of-focus light will tend to blur the image and make it difficult to resolve detail. Confocal microscopy, described in UNIT 2.8, utilizes specialized illuminating and light-collection optics to effectively eliminate the out-of-focus light. This important form of light microscopy permits high-resolution measurements within thick specimens. UNIT 2.9 discusses multiphoton imaging, a method for laser scanning fluorescence micros-
copy in which absorption is confined to the plane of focus. Very high laser intensities enable a molecule to absorb more than one photon within a very short coincidence time. The energy of the absorbed photons is equivalent to that of a single photon of shorter wavelength. Multiphoton imaging provides resolution similar to that of confocal microscopy, but with greatly reduced photobleaching owing to the lower energies (longer wavelengths) involved. Scanning laser cytometers provide low-resolution, multiparameter measurements of fluorescence emitted from and light scattered by cells or a fixed substrate. These instruments are useful in the analysis of small cellular samples, in low-resolution localization of materials and structures within cells, and in repeated analyses of the same cells over time. UNIT 2.10 describes the components and operation of these new instruments and gives an overview of their applications. UNIT 2.11 presents procedures to correct microscope images for shading that may be introduced in image formation. This shading, which can result from non-uniform illumination, non-uniform camera sensitivity, or dirt on optics, must be removed for image processing in quantitative microscopy. The unit begins with a mathematical model and develops techniques for shading correction in situations when one has only a single image with which to work as well as those in which calibration images may be recorded.
discusses the principles of fluorescence recovery after photobleaching (FRAP) and highlights recent work on molecular transport processes. FRAP is a well known technique for measuring mobilities driven by diffusion in membranes and liquid systems. The ability to spot photobleach with an intense laser beam and to measure the kinetics of recovery of fluorescence provides some unique information about lateral mobilities within the cell. Although not protocol driven, this unit provides an outstanding theoretical basis of the biophysics of FRAP and the principles of measuring kinetic changes in diffusion dynamics. UNIT 2.12
Future updates to this chapter will include a unit on flow versus image cytometry.
Introduction
Leon L. Wheeless
2.0.2 Supplement 15
Current Protocols in Cytometry
Contrast Enhancement in Light Microscopy Optical microscopes, which are among the oldest instruments of scientific discovery, continue to be key tools in both biomedical research and routine diagnosis. This remains true despite the development of a wide range of new imaging technologies, many with far greater resolution—ranging from electron microscopes to the multitude of scanning probe systems available today. The overall simplicity of optical microscopes, their minimally destructive impact, particularly on live specimens, and the many microscopy techniques available for enhancing or visualizing specific specimen features make this technique exceptionally useful in biomedical applications. In recent years, electronic imaging has both greatly enhanced the capabilities of light microscopes and placed ever-increasing demands on their optical performance. A number of purely optical methods can be used to enhance feature extraction from biological material for both visual observation and subsequent electronic image processing. In the present discussion, a short primer on image formation in basic transmitted-light bright-field microscopes and their resolution limits will be followed by a brief outline of how to optimize the imaging conditions by using what is known as Köhler illumination, and finally by discussion of alternatives to bright-field microscopy.
KÖHLER ILLUMINATION Köhler illumination is a method of focusing and centering a microscope’s light source and optics and setting its diaphragms to obtain the best image detail and contrast. It was pioneered in the late 1800s by August Köhler, a zoologist in Giessen, Germany, who in 1900 joined Carl Zeiss in Jena to head up the company’s microscopy group. Today, all manufacturers of highquality microscopes provide Köhler illumination in their instruments.
Basic Concepts Understanding the possible contrast enhancement techniques and appreciating their specific applications requires a clear grasp of the basic concepts of Köhler illumination. It is illustrative to follow the path taken by light through a microscope from source to final image (see Fig. 2.1.1). A single, off-axis source point is imaged by the collector into the front focal plane of the condenser (condenser aperture diaphragm), where an image of the light Contributed by H. Ernst Keller Current Protocols in Cytometry (1997) 2.1.1-2.1.11 Copyright © 1997 by John Wiley & Sons, Inc.
UNIT 2.1
source is formed. Every point of this source image is projected by the condenser as a parallel pencil into infinity through the specimen. These pencils uniformly distribute the light intensity of each source point across their diameter, which in turn is set by the field diaphragm. In the back focal plane of the objective, also called the objective exit pupil, the source is imaged again and relayed by the eyepiece to its exit pupil or eyepoint, which becomes the entrance pupil for the eye or imager. Four “sourceconjugated” planes in which a source image appears can be identified, the latter two of which are modulated by diffraction in the specimen. Superimposed over this illumination beam path is the image-forming beam path. The specimen is conjugated to the field diaphragm and imaged by the objective into the real intermediate image plane, then projected by the eyepiece into infinity and reimaged by the detector’s objective (eye lens or camera lens) onto the final sensor (retina, film, or camera). There are several advantages to this dual optical train. The field stop permits control over internal stray light and internal reflections by limiting the diameter of the illumination and imaging bundles to that needed for a given specimen field diameter. Contrast control can be achieved by controlling the condenser aperture, which directly influences the coherence and participation of diffracted wavefronts in the formation of the final image: closing the condenser diaphragm increases contrast and decreases resolution. An experienced microscopist will attempt to find the best compromise between resolution, contrast, and, to a lesser extent, depth of field. Most microscopes have built-in illuminators, so the steps for obtaining good Köhler illumination are simple: 1. Achieving sharp specimen focus; 2. Achieving substage or condenser focus and centering via the field diaphragm, and setting that diaphragm to cover a given objective’s and eyepiece’s field of view; and 3. Setting the condenser aperture diaphragm for best contrast, resolution, and depth of field, based on the nature of the specimen itself.
Resolution and Contrast The light path for Köhler illumination can be explained by geometric optics, a way of thinking about how light travels in a straight
Image Cytometry Instrumentation
2.1.1
A
B
final image exit pupil of microscope
eye lens
eyepiece real intermediate image plane
exit pupil of objective
objective specimen condenser condenser (aperture) diaphragm
luminous field diaphragm
collector
light source
Figure 2.1.1 Light path in Köhler illumination. (A) Path of the image-forming ray, with its four conjugated planes. (B) Illumination path, again with four directly related or conjugated planes.
Contrast Enhancement in Light Microscopy
line as a ray. To understand the limits of resolution and the contrast transfer from object to image, it is necessary to consider the electromagnetic wave nature of light, its wavelength (λ), and the interaction of wavefronts with the specimen’s structures. The theory of image formation through a microscope was developed by Ernst Abbe, who distinguished between two types of specimens: self-luminous objects (light sources, luminescence, or fluorescence), which follow the Rayleigh criterion, and illuminated objects, which follow the Abbe criterion. According to the Rayleigh criterion, which is based on diffraction in the objective,
the point-to-point resolution, δ, (in µm) is given by: δ=
0.61λ NA objective
In the case of an illuminated object, the Abbe criterion defines the resolution, δ, as δ=
λ 2NA
or
λ NA objective + NA condenser
where λ is the wavelength of the emitted or illuminating light and NA is the numerical aperture of objective or condenser, given by the
2.1.2 Current Protocols in Cytometry
sine of half its collection angle times the refractive index of the medium between specimen and objective. With a self-luminous object, diffraction in the objective itself causes the smallest image point to expand into what is known as an Airy disk (Fig. 2.1.2), whose diameter, D (in µml), will be:
D=
1.22 λ
NA With an illuminated object, diffraction is caused by the structural features of the specimen and their spacing. The diffraction angle (α) is determined by the wavelength and the spacing or distance (d) between features. sin α =
λ
d The diffraction angle that an objective is capable of collecting is directly related to its numerical aperture. Abbe proved that for a given structural spacing to be resolved, at least two orders of diffracted light produced by this spacing need to participate in the image forma-
tion. Interference between diffracted and nondiffracted wavefronts in the intermediate image plane resolves structural detail and determines the contrast at which the image is rendered.
Bright-Field Microscopy Bright-field microscopy is the optical technique most commonly used. With a microscope set up and adjusted for Köhler illumination, bright-field microscopy is ideally suited to the study of specimens whose features are clearly differentiated by differences in absorption. Such specimens are also called “amplitude specimens,” because they primarily change the intensity or amplitude of the illuminating light. Either inherent absorption or absorption induced by staining will change the gray level or color contrast if spectral differences in absorption exist. The contrast appears against a white, bright background. Histochemically stained tissue sections of all sorts, cytology and bacterial stains, and naturally absorbing specimens are best studied under a bright-field microscope.
A
Figure 2.1.2 (A) A typical Airy disk. (B) Computer graph of the intensity distribution of the Airy disk.
B
Image Cytometry Instrumentation
2.1.3 Current Protocols in Cytometry
optical terms, changes the mix of direct, nondiffracted wavefronts and diffracted light to produce pseudorelief and enhanced contrast. For more subtle enhancement, a similar effect can be achieved by decentering the light source. Care must be taken to retain even, uniform illumination over the field. Oblique illumination is a simple, inexpensive means to enhance contrast in unstained, transparent sections, sediments, or casts and can be a useful tool for finding focus in highly transparent specimens.
OTHER MICROSCOPY TECHNIQUES Oblique and Anaxial Illumination Just as the rising or setting sun will better reveal the topography and mountain ridges of a landscape than the noonday sun, obliquely illuminating a specimen with limited internal contrast can greatly enhance structural differences in optical density or refractive index and turn an otherwise flat or almost invisible object into an image of striking relief and apparent three-dimensionality with clearly enhanced contrast (Fig. 2.1.3). To obtain oblique illumination with some degree of reproducibility, a “turret condenser,” which allows the condenser aperture diaphragm to be shifted laterally, is helpful. This lateral displacement combined with the best setting for the aperture’s diameter varies with the objective’s numerical aperture and, in wave-
Dark-Field Illumination Dark-field illumination greatly enhances a microscope’s ability to detect minute structures or particles, often far below the theoretical limits of resolution—that is, even though the size and spacing of the structures cannot be resolved, their presence is obvious: they appear bright on a dark (black) background. This dark
A
B –1
0
+1
–1
–1
0
–1
0
+1
0
B´
+1 +1 G
B
Contrast Enhancement in Light Microscopy
Figure 2.1.3 Axial versus oblique illumination and the effects of aperture on resolution. (A) Low-aperture axial illumination will not resolve structures that generate diffraction angles 0/+1 or 0/−1. (B) Shifting the same aperture to the side permits the objective to collect diffraction order 0 and −1, resolves the structure, and, with only one side band of diffracted light participating, generates a relief effect.
2.1.4 Current Protocols in Cytometry
background is achieved by excluding all direct, nondiffracted light from the objective. Specifically, the dark-field condenser produces a hollow cone of illumination with an aperture higher than that of the objective. This can be accomplished by an annular diaphragm in the condenser aperture or by specific dark-field condensers, such as “paraboloid” (dry) or “cardioid” (oil) condensers (Fig 2.1.4). Objectives of high numerical aperture require a builtin iris or a funnel stop to reduce their numerical aperture below that of the condenser. Because the image is formed by diffracted light only, a contrast reversal takes place. As a result, darkfield microscopy is exceptionally sensitive to contamination; it is therefore imperative that
condenser, slide, and objective front lens be perfectly clean. Contaminants, bacteria, cell and urine casts, and blood, among others, lend themselves well to dark-field studies.
Hoffman Modulation Contrast and Varel Contrast Hoffman modulation and Varel contrast techniques are sophisticated methods for oblique illumination. Images generated by these methods exhibit a striking three-dimensional effect produced by converting directionally opposing specimen gradients in refractive index or thickness into opposing gray-level differences. The two techniques differ mainly in the geometry of a special attenuator for the
diffracted light direct light specimen
internal mirror
condenser
annular stop
Figure 2.1.4 The dark-field condenser’s hollow cone of illumination passes by the objective. Only light diffracted or refracted by the specimen is collected. Image Cytometry Instrumentation
2.1.5 Current Protocols in Cytometry
nondiffracted, zero-order direct light in the back focal plane of the objective and the corresponding illumination aperture in the condenser; the position of the attenuator determines the direction in which the gradients are best contrasted. In Hoffman modulation contrast a straight, bar-shaped attenuator or modulator is placed on the periphery of the objective’s aperture and absorbs ~85% of the direct light coming from a slit in the condenser aperture properly aligned to superimposition over the modulator (Fig. 2.1.5). In Varel contrast, an annular attenuator in the very outer back aperture of the objective absorbs ~85% of the direct light coming from a corresponding segment of an annulus in the condenser. Only direct light is attenuated, while diffracted light passes fully for a strikingly improved contrast generation. In both techniques, oblique brightfield can be added when specimens of relatively high inherent contrast are studied. Unstained live tissue and cell cultures in either glass or plastic vessels make ideal specimens for these techniques; the improved depth perception of the resulting images also facilitates micromanipulation or microinjection.
Differential Interference Contrast Differential interference contrast is the most sophisticated, most flexible, and potentially
most highly resolving technique available; it converts specimen gradients into gray-level differences and produces a striking pseudo-threedimensional effect. The system employs polarized light and special prisms called Wollaston prisms to produce two slightly sheared or separated wavefronts, which traverse the specimen (Fig. 2.1.6). The amount of shear is usually below the resolution of a given objective and is a function of the Wollaston prisms in the condenser and objective. Specimen gradients in refractive index or thickness result in an optical path difference between the two sheared wavefronts. When the wavefronts are recombined and made to oscillate in a common plane by an analyzer, different amounts of constructive or destructive interference produce distinct graylevel differences for opposing gradients, with the greatest contrast along the direction of shear. The system can be set to maximum contrast for any specific specimen gradient by adjusting one of the prisms or using a special compensator. It is often desirable to use the full objective and condenser apertures, particularly when using video-enhanced imaging to extract the very smallest contrast differences so as to detect and visualize intracellular organelles such as microtubules. Differential interference contrast (largely the version proposed by Nomarski) has contributed greatly to the study of live cells and tissues and is now an indispensable tool in develop-
3% transmittance 100% transmittance modular
15% transmittance
objective
specimen with gradients
condenser
slit aperture
Contrast Enhancement in Light Microscopy
Figure 2.1.5 Basic principle of modulation contrast.
2.1.6 Current Protocols in Cytometry
mental biology, physiology, neuroscience, and many other disciplines. Because it employs polarized light, plastic specimen vessels should not be used for this method, as they tend to show birefringence and depolarize the sheared wavefronts. To avoid strain or stress in the condenser and objective, it is important to use components recommended by the manufacturer.
Phase Contrast Phase contrast microscopy is designed for the study of thin, unstained sections or live
cultures—i.e., transparent specimens with minimal inherent contrast. Unlike amplitude or absorbing specimens, for which the diffracted wavefronts are phase-shifted by one-half of a wavelength, such so-called phase specimens generate shifts of only one-quarter of a wavelength. The interference conditions between diffracted and direct wavefronts are neither constructive nor destructive, and the image contrast is poor. The Dutch physicist Frits Zernicke won the Nobel prize for his proposal to add to condenser and objective elements that
intermediate image
analyzer (135 )
second Wollaston prism
objective
specimen
condenser
first Wollaston prism
polarizer (45 )
Figure 2.1.6 Principles of differential interference contrast. The separation between the two sheared beams is greatly exaggerated.
Image Cytometry Instrumentation
2.1.7 Current Protocols in Cytometry
shift the phase of nondiffracted, direct light by one-quarter of a wavelength, and at the same time attenuate the intensity so as to greatly enhance the interference conditions for the image rendition. The result is an image wherein “positive-phase” contrast areas of higher refractive index appear darker. Specific gray levels optically “stain” areas of specific refractive index and thickness. This is accomplished by
an illumination annulus in the aperture plane of the condenser along with a conjugate phase ring in the back focal plane of the objective that acts as both attenuator and phase shifter. A green filter further enhances the contrast (Fig. 2.1.7). In contrast to differential interference or oblique illumination techniques, which optically “stain” specimen gradients and generate
P´
λ/2
λ/4
P
R
Contrast Enhancement in Light Microscopy
Figure 2.1.7 Phase contrast. The annulus, R, in the condenser aperture is superimposed on the phase plate behind the objective, which is both an attenuator and a phase shifter.
2.1.8 Current Protocols in Cytometry
a pseudo-three-dimensional effect, phase contrast produces a two-dimensional image of index- and thickness-specific gray levels. The limitations of phase contrast are determined by the illumination aperture, a function of the condenser’s phase ring, and by the “halo” effect along steep specimen gradients, which limits the section thickness for phase contrast to ∼5 µm.
Reflection Interference Reflection interference microscopy looks at the interference pattern that is naturally present between a cell or tissue and its substrate (cover glass). Wavefronts reflected at the cell surface interfere with those reflected at the substrate and are one-half wavelength out of phase for those areas where the cell adheres to the cover glass, resulting in destructive interference and darkness (adhesion plaques). Using reasonably monochromatic incident light obtained by filtering light from a tungsten halogen or, better, a mercury lamp through a green filter, along with good Köhler illumination, produces a striking contrast that allows direct analysis of a cell’s proximity to the substrate on which it grows. Limiting the illumi-
nation aperture further contrasts intracellular features based on the varying path differences they generate. Video enhancement of the contrast can also considerably improve the results (see Key References for suggestions for further reading on this topic).
Polarized Light Microscopy For a wide range of biological materials, visualization can be considerably improved by using either simple polarized light illumination or polarization contrast, whereby a polarizer below the condenser (usually oriented “eastwest”) linearly polarizes incoming wavefronts. An analyzer behind the objective is oriented at 90o to the polarizer; without a specimen the field of view is dark. Specimens with distinct structural orientation, such as muscle, nerve, or bone tissue, are birefringent—i.e., they display different refractive indices in different directions. Depending on their thickness and orientation to the polarizer and analyzer, such specimens will alter the plane of vibration of incoming polarized light and change it to some form of elliptically polarized light, part of which will then be able to pass the analyzer. This results in bright specimen images on a dark back-
real image plane
barrier filter
source
dichromatic filter
exciting radiation
exciter filter
objective emitted radiation
object
Figure 2.1.8 Epifluorescence is made possible by a dichromatic filter that reflects the exciting radiation down on the specimen and allows the emitted radiation to pass upward to the eye.
Image Cytometry Instrumentation
2.1.9 Current Protocols in Cytometry
ground. When the full visible spectrum of light is used, specific wavelengths will be suppressed and others enhanced as a function of the path difference the specimen has generated between its two orthogonal vibration directions. This can result in vibrant interference colors, which in turn provide information about the specimen’s birefringence and thickness. A detailed discussion of polarized light microscopy—especially the quantitative aspects of the analysis of specimen birefringence and directional and structural orientation—would require far more in-depth treatment than can be contained in this brief overview; the reader is encouraged to consult the relevant literature (see Key References) for further information.
Fluorescence Microscopy One of the fastest-growing tools in biomedical microscopy is fluorescence. The exceptional sensitivity of this technique, combined with the ever-growing list of very specific protein markers and fluorophores covering a wide range of different colors that are available, have made it indispensable for qualitative and quantitative diagnosis. Autofluorescent or fluorophore-labeled specimens are excited with short-wavelength radiation and almost instantly convert some of the absorbed exciting radiation to emitted longer-wavelength fluorescence. In present-day microscopes, excitation is provided almost exclusively by either incident light or epiillumination. The light source is usually a mercury or xenon gas discharge lamp. Special filter/reflector combinations isolate the fluorophore’s specific emission wavelength from the exciting radiation to maximize the efficiency of both and thereby produce a bright, sharp fluorescent signal on a black background (Fig. 2.1.8). Ideally the microscope should be equipped with objectives of high numerical aperture, magnifications just high enough to see the areas of interest, and good chromatic correction. Much of the image capture is done with sensitive cameras either in real time or, for low light levels, with long-term signal integration. A multitude of publications, textbooks, and reprints detailing fluorescence microscopy are available in bookstores and from most of the major microscope manufacturers. For a more in-depth look into these interesting and useful microscope methods the reader is encouraged to contact these sources (see Key References). Contrast Enhancement in Light Microscopy
KEY REFERENCES General Microscopy Born, M. and Wolf, E. 1970. Principles of Optics. Pergamon Press, Elmsford, N.Y. Bradbury, S., Evennett, P.J., Haselmann, H., and Piller, H. 1989. Dictionary of Light Microscopy. Oxford University Press, Oxford. Herman, B. and Jacobsen, K. 1990. Optical Microscopy for Biology. Wiley-Liss, New York. Lacey, A.J., 1989. Light Microsopy in Biology: A Practical Approach. IRL Press, Oxford. Pawley, J. 1989. Handbook of Biological Confocal Microscopy. Plenum, New York. Pluta, M. 1988. Advanced Light Microscopy, Vol. I. Elsevier Science Publishing, New York. Pluta, M. 1989. Advanced Light Microscopy, Vol. II. Elsevier Science Publishing, New York. Pluta, M. 1992. Advanced Light Microscopy, Vol. III. Elsevier Science Publishing, New York. Spencer, M. 1982. Fundamentals of Light Microscopy. Cambridge University Press, Cambridge.
Oblique Illumination Techniques Ellis, G.W. 1981. Edge Enhancement of Phase Phenomena. U.S. Patent No. 4255014. Hoffman, R. 1975. The modulation contrast microscope. Nature 254:586-588. Kachar, B. 1985. Asymmetric illumination contrast. Science 22:766-768.
Hoffman Modulation Contrast Hoffman, R. and Gross, L. 1975. Modulation contrast microscopy. Appl. Opt. 14:1169-1176.
Differential Interference Contrast Allen, R.D., David, G.B., and Nomarski, G. 1969. The Zeiss-Nomarski differential interference equipment for transmitted light microscopy. Z. Wiss. Mikrosk. Mikrosk. Tech. 69:193-221. Francon, M. 1962. Progress in Microscopy. Row, Peterson, Evanston, Ill. Lang, W. 1979. Nomarski Differential Interference Contrast Microscopy. Carl Zeiss, Oberkochen, Germany. Padawer, J. 1968. The Nomarski interference contrast microscope. J. Roy. Miscrosc. Soc. 88: 305349.
Phase-Contrast Microscopy Francon, M. 1962. Progress in Microscopy. Row, Peterson, Evanston, Ill. Ross, K.F.A. 1967. Phase Contrast and Interference Microscopy for Cell Biologists. Edward Arnold, London. Zernicke, F. 1942. Phase contrast, a new method for the microscopic observation of transparent objects. Physics 9:686-693.
2.1.10 Current Protocols in Cytometry
Reflection Interference Microscopy Beck, K. and Bereiter-Hahn, J. 1981. Evaluation of reflection interference contrast images of living cells. Microsc. Acta 84:153-178. Gingell, D. and Todd, J. 1979. Interference reflection microscopy: A quantitative theory for image interpretation. Biophys. J. 26:507-526. Izzard, C.S. and Lochner, L.R. 1976. Cell to substrate contacts in living fibroblasts. J. Cell Sci. 21:129-159. Ploem, J.S. 1975. Reflection Contrast Microscopy as a Tool for Investigation of the Attachment of Living Cells to a Glass Surface. Blackwell Scientific, Oxford.
Polarized Light Microscopy Patzelt, W.J. 1985. Polarized Light Microscopy: Principles, Instruments, Applications. E. Leitz, Wetzlar, Germany. Shurcliffe, W.A. 1962. Polarized Light. Harvard University Press, Cambridge, Mass.
Waggoner, A.S., DeBiasio, R., Bright, G.R., Ernst, L.A., Conrad, P., Galbraith, W., and Taylor, D.L. 1989. Multiple spectral parameter microscopy. Methods Cell Biol. 30:449-478.
Photomicrography Delly, J.G. 1980. Photography through the Microscope. Kodak Publication P-2. Kodak, Rochester, N.Y. Loveland, R.P. 1981. Photomicrography: A Comprehensive Treatise. John Wiley & Sons, New York.
Video Microscopy Allen, R.D., Allen, N.S., and Travis, J.L. 1981. Video enhanced contrast, differential interference contrast microscopy. J. Cell Mot. 1:298302. Allen, R.D. and Allen, N.S. 1983. Video enhanced microscopy with a computer frame memory. J. Microsc. 128:3-7.
Shurcliffe, W. 1975. Polarized Light, Benchmark Papers in Optics. John Wiley & Sons, New York.
Inoué, S. 1981. Video image processing greatly enhances contrast quality and speed in polarization microscopy. J. Cell Biol. 89:346-356.
Fluorescence Microcopy
Inoué, S. 1986. Video Microscopy. Plenum, New York.
Bright, G.R. 1993. Multiparameter imaging on cellular function interference. In Fluorescence Probes for Biology Function of Living Cells: A Practical Guide (W.T. Mason and G. Rolf, eds.) pp. 204-215. Academic Press, San Diego. Herman, B. and Lemasters, J.J. 1993. Optical Microscopy, Emerging Methods and Applications. Academic Press, San Diego. Taylor, D.L. and Wang, Y.L. 1989a. Fluorescence Microscopy of Living Cells in Culture (Part A, Vol. 29). Academic Press, San Diego. Taylor, D.L. and Wang, Y.L. 1989b. Fluorescence Microscopy of Living Cells in Culture (Part B, Vol. 30). Academic Press, San Diego.
Inoué, S. 1987. Video microscopy of living cells and dynamic molecular assemblies. Appl. Optics 26:3219-3225. Inoué, S. 1988. Progress in video microscopy. Cell Motil. Cytoskeleton 10:13-17. Schotten, D. 1993. Electronic Light Microscopy. Wiley-Liss, New York. Weiss, D.G., Maile, W., and Wick, R. 1992. Video microscopy. In Light Microscopy in Biology (S.J. Lacey, ed.) pp. 221-278. IRL Press, Oxford.
Contributed by H. Ernst Keller Carl Zeiss, Inc. Thornwood, New York
Image Cytometry Instrumentation
2.1.11 Current Protocols in Cytometry
Microscope Objectives The basic task of a microscope is to provide enlarged images of small objects. Of all the optical components of the microscope, the objective is the most crucial to image formation. The microscope objective must provide: 1. Magnification: an image that is enlarged with respect to the specimen 2 Resolution: an image whose details are clearly separated 3. Contrast or visibility: an image with sufficient contrast that its details are readily distinguishable from each other and from background material in the field of view when viewed by the human eye or by a camera 4. Fidelity: an image that is a faithful reproduction of the original specimen, free from distortion (aberration) and spurious detail (artifact).
IMAGE FIDELITY Lens Aberrations The image of the specimen projected by the objective is an “optical replica” of the original, suspended in space along the optical axis ~10 mm below the top of the eyepiece tubes (Figs. 2.2.1 and 2.2.2). The objective is constructed to be as free as possible from aberrations that would degrade the magnified image representing the original specimen. Aside from providing magnification, the objective must pro-
UNIT 2.2 duce an image that is, at each and every point within it, a faithful mapping of the specimen in all the following respects: 1. Specimen point to image point 2. Specimen line to image line 3. Specimen shape to image shape 4. Specimen flatness to image flatness 5. Specimen color to image color. From the standpoint of geometric optics, simple glass lenses may exhibit two main types of aberrations: (1) aberrations related to wavelength or color, termed chromatic aberrations, that occur when white (polychromatic) light is used; and (2) aberrations that occur even when monochromatic light (light of a single color or wavelength) is used. It is important to understand that no objective, no matter how well-corrected for aberrations, can ever render an image “point”—i.e., infinitesimal in size, like a true point. Because of the wave nature of light and the fact that light is scattered (diffracted) as it traverses the minute pores and edges of a specimen, the best that even a perfectly corrected objective can do is to represent a specimen point as a tiny disk of light known as an Airy disk (Fig. 2.2.3).
Light Refraction and Refractive Index In the light microscope, light passes not only through the specimen but also through a series of glass lenses and accessories, and through air
eyepiece 10 mm plane of intermediate image at eyepiece diaphragm
objective
specimen
Figure 2.2.1 Schematic diagram of a compound microscope, illustrating the site of intermediate image formation at the plane of the eyepiece diaphragm. Reproduced from Abramowitz (1994) by courtesy of Olympus America. Image Cytometry Instrumentation Contributed by Mortimer Abramowitz and Marc M. Friedman Current Protocols in Cytometry (1997) 2.2.1-2.2.14 Copyright © 1997 by John Wiley & Sons, Inc.
2.2.1
or another medium such as oil. As light passes from one medium to another it is bent, or refracted, to a degree that depends on the refractive indices of the adjacent media. The refractive index, n, for a particular medium is defined as the ratio of the speed of light in a vacuum to the speed of light in that medium. The slower the speed of light in a medium, the greater its refractive index. For all practical purposes the refractive index of air is 1—that is, light travels at the same speed in air as in a vacuum. All other media have refractive indices >1. The refractive index of glass is ~1.5 and that of water is ~1.33. When light passes from a less dense medium (e.g., air; lower index of refraction) to a more dense medium (e.g., glass; higher index of refraction) at an angle other than 90o, the rays are bent (refracted) toward the perpendicular. Conversely, light rays are bent away from the perpendicular when travelling from a medium of higher refractive index to a medium of lower refractive index. These basic properties of light are central to the prac-
image
tical use of microscopes and to the design of microscope lenses, which typically must gather light passing through an aqueous specimen (e.g., a cell) and thence through a glass cover slip to air or another medium such as oil. For example, the refractive index of standard microscope immersion oil is carefully set at 1.515 so as to match that of the glass in the cover slip and the objective lens, to eliminate unwanted refraction. More detailed discussion of these properties may be found in the references listed at the end of this unit.
Chromatic Aberrations Chromatic aberrations result from the fact that glass used in microscope lenses exhibits different refractive indices at different wavelengths of light. Light of shorter wavelengths, at the blue end of the spectrum, is brought to a focus nearer to the back of the objective than light of longer wavelengths (Fig. 2.2.4). This effect is known as longitudinal chromatic aberration. Even when light of different wave-
Figure 2.2.2 Formation of a real image of a point by an ideal lens. Reproduced from Abramowitz (1994) by courtesy of Olympus America.
specimen
Figure 2.2.3 Airy disk diffraction image of a point, illustrating the formation of the image of a specimen point as a disk of light rather than a dimensionless fine point. Reproduced from Abramowitz (1994) by courtesy of Olympus America.
Microscope Objectives
2.2.2 Current Protocols in Cytometry
lengths is brought to a common focus by employing optical correction, the image of a point in the outer part of the field of view may be magnified to a greater extent with blue light than with red. This is known as chromatic difference of magnification, or lateral chromatic aberration, and causes the image of a point in white light to appear ringed by colors. These errors are typically corrected by employing glasses with differing optical properties, such as flint glass, crown glass, and low-dispersion glasses, in various lens elements within the objective. Manufacturers may use different design methods to correct for these aberrations. Older-style microscopes used compensating eyepieces to correct for chromatic difference of magnification. Olympus and Nikon now correct for chromatic difference of magnification within the objective itself, whereas Leica and
Zeiss accomplish this correction by means of a tube lens built into the microscope stand.
Aberrations in Monochromatic Light The most serious aberration in monochromatic light is spherical aberration. Light passing through the outer region of a convex lens is brought to a focus nearer to the back of the objective than light passing through the center of the lens (Fig. 2.2.5). The net result is a blurring of the image, because a single point in the specimen is represented by a series of points in the image, with each image point focused at a slightly different distance from the back of the objective. These and other monochromatic aberrations are corrected by designing required lens elements of different shapes and forming these into compound elements by cementing them into doublets and triplets. In practice,
red focus
green focus violet-blue focus
Figure 2.2.4 Chromatic aberration in an uncorrected lens, illustrating the existence of different focal planes for different wavelengths of light. Reproduced from Abramowitz (1994) by courtesy of Olympus America.
objective
white light
axial rays
peripheral rays
uncorrected lens
monochromatic light
Figure 2.2.5 Spherical aberration, or the difference of focus for light rays traversing the outer zones of a lens compared to light passing more centrally. Reproduced from Abramowitz (1994) by courtesy of Olympus America.
Image Cytometry Instrumentation
2.2.3 Current Protocols in Cytometry
eliminating spherical aberration is particularly difficult. Despite the excellent design of modern objectives, a microscopist may inadvertently introduce spherical aberration by using a too-thick layer of mounting medium below the cover slip and/or a cover glass of incorrect thickness (see discussion of Special Features later in this unit) or by employing improperly corrected optical accessories in the light path. Accessory image-forming lenses such as eyepieces, relay lenses, and tube lenses are used to enlarge, correct, or focus an image projected by the objective lens, but they must be carefully designed and properly used so as not to degrade the image. Spherical aberration is particularly troublesome when taking a series of optical sections through living cells or tissues for threedimensional reconstruction, in which dozens or hundreds of images may be recorded by successively focusing to great depths within a tissue, using intervals as small as 0.5 µm. Internal organelles encountered in the focal plane of each section may introduce, unpredictably, regions of differing indices of refraction that cannot be anticipated in the design of the lens. Other aberrations observed using monochromatic light include coma, astigmatism, distortion, and curvature of field. These errors
must be corrected in the design of the glass elements of the objective and are not under the control of the user. With coma, the image of a point appears comet-shaped (Fig. 2.2.6). With astigmatism, the image of a point may appear as a line with either a vertical or horizontal orientation (Fig. 2.2.7). Distortion (Fig. 2.2.8) causes images of parallel lines to bow outward (barrel distortion) or inward (pincushion distortion). With curvature of field (Fig. 2.2.9), an error inherent to curved lens elements, the image may be in focus in the center of the field of view but not at the edges, or vice versa. Modern objectives, even those from the less expensive series of the major manufacturers, are corrected to compensate for these aberrations.
PROPERTIES OF MICROSCOPE OBJECTIVES Numerical Aperture Numerical aperture (NA) is a term devised by Ernst Abbe in the late 19th century to describe the light-gathering ability of an objective. Figure 2.2.10 represents a set of longitudinal slices through the cone of light emanating from a point in the specimen and captured by each of three objective lenses. If the included
image plane
axis
objective
specimen point
Microscope Objectives
Figure 2.2.6 Diagrammatic representation showing the comet-like smearing of an image point from a lens that has not been corrected for coma aberration. Reproduced from Abramowitz (1994) by courtesy of Olympus America.
2.2.4 Current Protocols in Cytometry
half angle is designated µ, then the NA is defined as n × sin µ, where n is the refractive index of the medium between the cover slip and the objective. For dry objectives, with air between the cover slip and the objective, the maximum achievable NA is 0.95. Modern oilimmersion objectives offer a maximum NA of 1.4, whereas water-immersion objectives offer a maximum NA of 1.25.
Resolving Power The importance of achieving the highest possible NA becomes evident upon investigating the theoretical resolving power of a lens, which may be described as its ability to reveal a separation between two closely spaced objects. The theoretical resolving power is the maximum separation that is achievable, and depends on two factors: the wavelength of the
Ib
I
image plane Ia
principal axis
objective
specimen point
Figure 2.2.7 Astigmatism, an aberration in the outer lens zones caused by unequal magnification in the different azimuths. A specimen point appears in the image not as a point but as a line. Reproduced from Abramowitz (1994) by courtesy of Olympus America.
specimen
image with barrel distortion
image with pincushion distortion
Figure 2.2.8 Barrel and pincushion distortion of parallel lines in an uncorrected lens. Reproduced from Abramowitz (1994) by courtesy of Olympus America.
Image Cytometry Instrumentation
2.2.5 Current Protocols in Cytometry
illuminating light and the numerical aperture of the optical system (objective and condenser). The formula devised by Abbe for determining resolving power in transmitted light microscopy is: λ illuminating light D= NA objective + NA condenser where D is the distance between two close-lying objects that are nevertheless separable by
eye. The highest theoretical resolving power of the transmitted light microscope, using an objective and condenser each having an NA of 1.4, can reveal a separation of 0.2 µm using green light.
CONSTRUCTION AND TYPES OF MICROSCOPE OBJECTIVES The construction of microscope objectives must ensure that optical alignment of the glass elements within the casing of the objective is
curved image at image plane
objective
flat specimen
Figure 2.2.9 Curvature of field, in which the image of a flat specimen is curved and therefore not in a uniform plane of focus. Reproduced from Abramowitz (1994) by courtesy of Olympus America.
µ oil low-NA dry objective NA = n sin µ
Microscope Objectives
higher-NA dry objective
oil highest-NA oil-immersion objective
Figure 2.2.10 Illustration of the light gathering ability of three lenses with increasing numerical aperture (NA), where n is the refractive index of the medium between the objective and the cover slip and µ is the half-angle of the cone of light captured by the objective. Reproduced from Abramowitz (1994) by courtesy of Olympus America.
2.2.6 Current Protocols in Cytometry
highly accurate, and must provide maximum protection for exposed glass surfaces. Accurate mounting and spacing of the elements is critical to proper optical performance, and is typically accomplished using tiny metal spacer rings. The exposed front surface of the objective lens and the internal glass surfaces in contact with air spaces between lens elements are coated with special thin deposits that greatly reduce internal glare to maintain a high throughput of light. Objectives with very short working distances—e.g., those that must be brought very close to the specimen in order to focus properly—are often designed with a spring-loaded tip that retracts the front lens into the barrel if the objective is inadvertently brought into contact with the top of the cover glass. Objectives are classified and labeled according to their degree of optical correction. The most common, in order of increasing correction and cost, include achromats (marked Ach, or sometimes unmarked), semiapochromats or fluorites (Fl), and apochromats (Apo). An objective may also be inscribed with the designation Plan (e.g., Plan Apo) indicating correction for flatness of field, which ensures that the image remains in focus from one edge of the field of view to the other. Plan objectives are especially important when the entire field of view must be critically examined without refocus-
ing, and when recording images using photomicrography or video archiving. High-power objectives that have high NA and offer excellent correction (e.g., Plan Apos) may contain more than ten glass components, many of which are doublets or triplets (Fig. 2.2.11). The finished objective must possess both mechanical stability and heat stability within the normal temperature range encountered in the laboratory, and must tolerate normal handling. However, the large number of closely spaced and precisely positioned elements creates a practical problem: such objectives will not suffer the excesses tolerated by less complex objectives. Dropping or severely bumping a Plan Apo objective, or leaving it in a car trunk on a hot summer day, will likely result in cracking and/or loosening of the glass elements and may necessitate complete reassembly at the factory, or replacement. The metal jacket of an objective has inscriptions that provide the user with important information about it (Fig. 2.2.12). Written in a simple, almost universal code, the markings convey the magnification, tube length, type of optical correction (finite or infinity-corrected), numerical aperture, allowed cover glass thickness (a critical design criterion), focal length of the tube lens (for infinity systems), and a classification of the objective according to the ex-
Figure 2.2.11 Schematic, longitudinal section through a 60× Plan Apo objective lens, illustrating the numerous and complex elements of a highly corrected objective. Diagram reprinted courtesy of Olympus America.
Image Cytometry Instrumentation
2.2.7 Current Protocols in Cytometry
OLYMPUS plan apochromat numerical aperature
magnification
UPlanApo immersion fluid
40x /1.0 oil iris /0.17
infinity correction
corrected for coverglass thickness of 0.17 mm
manufacturer
color-coded band
1.0>
<0.5
special feature: dark-field iris
Figure 2.2.12 Diagrammatic representation of typical markings found on the case of an objective lens.
tent of correction. Also, it is now standard to inscribe a colored ring toward the front of the barrel that tells the user the magnification just at a quick glance (e.g., yellow for 10×, green for 20×, blue for 40×, and white for 100×). An additional colored or black ring may indicate a phase objective or an immersion objective. Some inscriptions may also be in color to signify that the objective is intended for specialized use, such as phase-contrast or polarization microscopy.
MODERN OBJECTIVES Achromats, Fluorites, and Apochromats
Microscope Objectives
Achromats are corrected to bring at least two colors of light, usually red and blue, to a common focus and are most highly corrected for spherical aberration in the apple-green portion of the spectrum. Thus, when a green filter is used in the illumination path, modern achromats will yield quite good images. Recently designed achromats may have chromatic correction for three wavelengths and spherical correction for two. They are well suited for use in the clinical laboratory and do a creditable job of providing good color photomicrographs. However, the more rigorous demands of research microscopy usually mandate the use of fluorites or apochromats. Fluorite objectives are corrected chromatically and spherically for at least two colors (blue and green) and more recently for a third (red). Plan fluorites yield very good images in
photomicrography using white light and typical laboratory stains, and make good allaround objectives for both routine and more demanding applications because they are wellcorrected, offer high numerical apertures, and—at about two to four times the cost of an achromat—are still significantly less expensive than Plan Apo objectives. Apochromats offer the highest available degree of correction and consequently are the most expensive objectives, typically about ten times the cost of an achromat and three to four times the cost of a fluorite. Because of its high degree of optical correction, an apochromat usually has a numerical aperture that is much higher than that of an achromat of comparable magnification but only somewhat higher than that of a comparable fluorite. Modern Apos are chromatically corrected for at least four wavelengths (violet, blue, green, and red) and spherically corrected for three or four wavelengths. Modern Apos and fluorites are typically Plancorrected because they are often used for demanding applications and image recording via film or video techniques. For those who wish to compare the quality of images produced by different types of objectives, manufacturers offer various test slides including a metallized slide with tiny pinholes, a slide with fine lines spaced close together, and a slide containing mounted diatoms whose different species have structural markings of a specific fineness or periodicity. In former years, the so-called “Abbe test plate” was used to test for chromatic as well as spherical aberration,
2.2.8 Current Protocols in Cytometry
and its use is described in standard microscope texts. For many users, a familiar stained slide with an appropriate specimen can serve as a suitable routine test object.
expert user may determine through rigorous testing that a specific objective from one manufacturer can be used on another manufacturer’s instrument to deliver superior performance for a specific application.
Optical Correction and Tube Length Objectives may be further classified as finite (or finite-tube-length) objectives or infinitycorrected objectives according to the way in which they are designed to project images. The barrels of finite-type objectives are marked with the mechanical tube length, defined as the distance in millimeters along the optical path from the opening of the nosepiece to the top of the observation tube where the eyepiece is inserted. This number is most commonly 160 (mm). Infinity-corrected objectives are inscribed with the symbol ∞ to signify such design; they may also be inscribed with the focal length of the tube lens employed in the system, which forms the intermediate image (Figs. 2.2.13 and 2.2.14). Finite objectives project an image that converges to a focus at the plane of the fixed diaphragm of the viewing eyepiece (Figs. 2.2.1, 2.2.15). If an optical accessory, such as a polarizing intermediate piece or fluorescence illuminator, is interposed between the back of the objective and the eyepiece, it must incorporate correcting lenses to return the image to proper focus at the prescribed position in the plane of the eyepiece diaphragm. Infinity-corrected objectives are designed to project the image of the specimen “to infinity,” rather than to a fixed plane within the eyepiece, and so light rays arriving from all azimuths emerge from the objective in parallel bundles (Fig. 2.2.14). To bring the image to focus at the plane of the eyepiece diaphragm, the microscope must incorporate a tube lens in the light path. Some manufacturers mount the tube lens within the body of the microscope, whereas others build the tube lens into the binocular or trinocular observation-tube head. With infinity correction, accessories interposed between the objective and the tube lens are far simpler to design and are far less prone to introduce aberrations. It is important to recognize that users should not employ finite objectives on an infinity-designed stand, nor use infinity-corrected objectives on finite-designed stands. Furthermore, it is not advisable to use infinity-corrected objectives interchangeably even among infinity-designed microscopes from different manufacturers, as the focal lengths of the tube lenses differ and the various chromatic corrections are achieved in different ways. Despite this, an
Immersion Objectives Most biological objectives, as opposed to reflected-light metallurgical objectives, are designed to properly correct for aberrations only when used with a cover-slipped specimen, as indicated by specific markings on the lens barrel (Fig. 2.2.12). Unless otherwise labeled, objectives are designed with the assumption that air will occupy the space between the front lens of the objective and the cover glass of the specimen. Such objectives are called dry objectives, and have a maximum achievable NA of 0.95. In order to increase the numerical aperture to a range of 1.0 to 1.4, an intermediate or immersion fluid with a specific refractive index must uniformly fill the space between the top of the cover glass and the front lens of the objective, and be in contact with both. Because resolving power is directly proportional to NA, achieving the highest resolution requires an immersion-type objective, and the usual immersion medium is an oil with a refractive index of 1.515, which is close to that of glass. Other immersion objectives for specific applications are designed for use with water or, less commonly, glycerin (glycerol) as the immersion medium. Water-immersion objectives designed for use with and without a cover slip have become increasingly important in fluorescence and confocal microscopy for the study of living cells, and for three-dimensional reconstruction of cells and tissues. These objectives help to avoid the introduction of severe spherical aberration that may be incurred when using different immersion media above and below the cover slip, as when observing cells and tissues (in water) with oil-immersion objectives. Spherical aberration becomes more pronounced as the user focuses deeper into the tissue and farther below the cover slip. Immersion objectives are always inscribed with the name of the required immersion medium, and will yield a very distorted image if used dry or with the wrong immersion fluid. Some objectives are designed and marked for use with multiple immersion media, such as water, oil, and glycerin, with a correction collar that the user must be sure to adjust to the proper setting for each medium. Air bubbles must be scrupulously avoided when using immersion fluids, as the contained air has a significantly different refractive index
Image Cytometry Instrumentation
2.2.9 Current Protocols in Cytometry
eyepiece diaphragm plane
tube lens
objective
specimen
front focal plane of objective
Figure 2.2.13 Diagrammatic representation of an infinity-corrected objective system. The diagram illustrates parallel beams emerging from different azimuths (shaded) of an infinity-corrected objective being brought to focus by a tube lens. Reproduced from Abramowitz (1994) by courtesy of Olympus America.
than the immersion medium. It is also important not to mix immersion fluids from different manufacturers even if they have the same refractive index, as they will likely have quite different viscosities and will therefore not mix together adequately, but form optically refractive swirls that will interfere significantly with proper image formation.
Special Features
Microscope Objectives
In some applications—such as when viewing cells grown in thick-walled culture vessels—it may be physically impossible for the objective to get close enough to the specimen to focus properly. Manufacturers have therefore designed dry “long-working-distance” (LWD) objectives, most often used on inverted microscopes that can bring a specimen into focus even when the distance to the specimen is ≥1 mm. Such objectives are usually inscribed as LWD or ULWD (ultra-long-working-distance)
and may be specially designed to compensate for a defined thickness of intervening glass or plastic. Although most biological objectives are designed for use with cover-slipped specimens, some objectives are optically corrected for use with non-cover-slipped specimens such as blood smears. If so designed, the objective will be inscribed with a “0” or “–” in place of a cover glass thickness value (e.g., 0.17). Typical inscriptions would therefore be “160/0” or “160/– ” for the finite type or “∞/0” or “∞/–” for the infinity type. If such an objective with a magnification >10× is used with a covered specimen, the image quality will be poor. At a magnification of ≤10×, objectives designed for use with and without cover slips may be used interchangeably for routine applications. Microscope cover glasses come in several thicknesses, as indicated by a number (e.g., 1, 1.5, or 2). Each number represents a defined
2.2.10 Current Protocols in Cytometry
eyepiece integrated observation tube
intermediate image
tube lens
infinity space
parallel light beam
intermediate attachments
objective
Figure 2.2.14 Infinity space: the distance between the back of an infinity-corrected objective and the tube lens (schematic). Reproduced from Abramowitz (1994) by courtesy of Olympus America.
thickness range. For example, #1.5 cover glasses are typically 0.16 to 0.19 mm in thickness. Typical dry biological objectives are designed and optically corrected for a cover-glass thickness of precisely 0.17 mm. Dry objectives with NA greater than ~0.75 will suffer noticeable image degradation if the cover glass differs even by a few hundredths of a millimeter from the specified thickness. Because cover glass thickness may vary by several hundredths of a millimeter even within a package, “high dry” (40× high-NA) objectives are available with an adjustable correction collar and scale that permits them to be adjusted for different cover glass thicknesses (e.g., from 0.11 to 0.23 mm). Turning the correction collar to match the actual thickness of the individual cover glass in use prevents the introduction of spherical aberration and its consequence, image degradation. Using too much mounting medium on the tissue will create an additional “cover glass–like” optical layer that must be added to the thickness of the cover glass to determine the total “effective” cover glass thickness. If the effective cover glass thickness is different from that specified for the objective, spherical aberration will be introduced into the image. For this reason, many experienced microscopists do not rely on the correction collar’s numbered scale
to set the proper correction. Rather, they choose a suitable area of the specimen and repeatedly refocus the microscope while moving the correction collar to different positions, finally reaching the setting that provides the best image. On objectives used for inverted tissue culture studies with flasks or other relatively thick culture vessels, the correction collar may have a range of correction from 0 to 2 mm; on standard upright microscope objectives, the range is usually from 0.11 to 0.22 mm.
OBJECTIVES FOR OTHER MICROSCOPY APPLICATIONS Phase Contrast For phase-contrast microscopy, an annular “phase plate” is installed by the manufacturer inside the back of the objective. This plate serves to “speed up” the undiffracted light passing through it and also to reduce its intensity. Phase specimens, such as unstained cells and tissues, are almost invisible in standard brightfield microscopy. The phase plate in the objective, when aligned with the annular opening of a phase condenser, optically renders small phase objects visible without the use of stains. Because phase-contrast observation is often done through glass or plastic culture vessels,
Image Cytometry Instrumentation
2.2.11 Current Protocols in Cytometry
eyepiece diaphragm plane
objective front focal plane of objective specimen
Figure 2.2.15 Objective system of finite tube length, showing the projection of the image by a finite objective to the intermediate image plane within the eyepiece tube. Reproduced from Abramowitz (1994) by courtesy of Olympus America.
some manufacturers offer interchangeable accessory lenses or “caps” that attach to the front lens of the objective (one set for use with plastic vessels, one set for glass vessels) to avoid distortion of images.
Polarization Techniques
Microscope Objectives
In polarization microscopy, it is important that the objective itself not contribute to the alteration of polarization effects induced by the specimen. Because glass that is physically strained affects polarized light, microscope manufacturers carefully select objectives in which the glass elements and their mountings are strain-free. The barrel of strain-free objectives supplied with polarizing microscopes is usually marked with a “P”, “SF,” or “POL” and is sometimes inscribed in a color different from the usual inscription color. Differential interference contrast (DIC) microscopy is also invaluable for making small phase objects readily visible. It has further advantages in that it (1) yields a pseudo-three-
dimensional image, in which the object appears shadowed—brighter on one side and darker on the other—displaying “elevations” and “depressions” within the specimen; (2) permits the use of high-NA optics; and (3) makes possible “optical staining” and “optical sectioning” of the specimen. In DIC microscopy, the distance from the back focal plane of the objective to the upper Wollaston prism (a special prism positioned above the objective) is usually critical, and microscope companies may therefore designate particular objectives for use in DIC microscopy. These objectives are relatively strainfree, because interference microscopy also involves the use of polarized light, and may be labeled DIC or NIC (for Nomarski interference contrast, a particular type of DIC).
Dark-Field Microscopy In transmitted-light dark-field microscopy, the illumination is directed obliquely so that the specimen appears bright on a dark background. For dark-field microscopy with high-NA objec-
2.2.12 Current Protocols in Cytometry
tives (≥1.00), the NA of the objective must be reduced below that of the oil darkfield condenser. Manufacturers therefore provide highNA objectives with built-in iris diaphragms (see Fig. 2.2.12). For dark-field use, the diaphragm is closed down to yield an NA below 1.1. For general use, the diaphragm must be fully open or optical performance will be degraded.
Ultraviolet (UV), Fluorescence, and Infrared (IR) Applications Standard glass objectives are relatively opaque to wavelengths in the lower UV range, below ∼380 nm. Special objectives are manufactured with special glasses to achieve greater transmission of these lower wavelengths, which are used to excite certain fluorescent dyes for measurement of intracellular ions. The cements used in complex lens elements for fluorescence microscopy are nonfluorescing, and the best fluorescence objectives are made using quartz optics. Other investigations may be carried out using longer, IR wavelengths (>750 nm), which offer poorer resolution (see Abbe’s equation in the discussion of Resolving Power) but greater depth of penetration into biological (and other) materials. Several companies offer objectives specially designed to more efficiently transmit wavelengths up to 1800 nm. The technical departments of the major microscope companies can provide transmission and spectral data for their objectives upon request to aid in selecting the proper objectives for special applications.
OTHER CONSIDERATIONS IN CHOOSING OBJECTIVES Other considerations may prove valuable in understanding the performance of objectives and in guiding the selection, purchase, and use of suitable objectives. Numerical aperture, the ability of the objective to capture a cone of light of wider angle, has a crucial effect on resolution. Although intuitively it may seem that resolving power should increase with increasing magnification, it can be shown that the ability to distinguish closely spaced details within a specimen is directly proportional to the twice the working NA. However, the use of objectives with higherthan-necessary magnification and NA for a given application can be detrimental not only because they are more expensive, but also because the specimen area observed within a field of view will be smaller and both the depth of
field (the vertical distance above and below the plane being observed that is still in acceptable focus) and working distance are shallower. When the finest specimen details need to be observed, high-NA objectives are required. High-NA objectives are also indicated when maximum throughput of light is needed. The light transmittance for an objective, using visible wavelengths, typically varies with the square of the NA of the objective. In reflectedlight and epifluorescence microscopy, light passes through the objective twice (first the illuminating light, and then the reflected or fluorescent signal), and so the intensity varies with the fourth power of the NA. In situations where the light level is low, NA is a critical factor in obtaining brighter images. A question often asked is why higher magnification cannot be achieved simply by using higher-magnification eyepieces with a given objective. Because of limitations due to the size of light waves themselves and the phenomenon of diffraction, higher and higher magnifications unaccompanied by increased NA will result in images that are less and less clear. The limiting factor in ensuring usable, as opposed to empty, magnification is the NA of the objective (more precisely, the average NA of the objective and the condenser). Eyepieces and accessory lenses are designed for use with certain objectives and condensers, and should not be switched to increase magnification except as recommended by the manufacturer. An oft-cited rule of thumb is that the user should limit the total optical magnification (the objective magnification multiplied by the eyepiece magnification and that of any other lenses) to between 500 and 1000 times the NA of the objective. At <500 times the NA, fine specimen details may not be perceivable by the eye; at >1000 times the NA, the likely result is empty magnification. In the favored method of Koehler illumination, the condenser diaphragm is partially closed down, slightly lowering the overall NA in order to improve contrast. Hence, a total magnification of ∼750 times the NA will usually produce excellent images with satisfactory contrast. All of the foregoing discussion of objective design, features, and performance assumes that the optics (and the rest of the microscope) remain forever in the pristine state in which they presumably arrived. Proper care of the objectives, including handling, storage, and cleaning, are essential prerequisites to keeping them in proper working order. The authors have often noted that the best microscopy is not necessarily performed by those with the best equipment,
Image Cytometry Instrumentation
2.2.13 Current Protocols in Cytometry
and quite often the performance of superior optics is profoundly or subtly degraded by a lack of care in choosing and maintaining the optics.
Delly, J.G. 1988. Photography Through The Microscope. Eastman Kodak, Rochester, N.Y. Inoue, S. 1986. Video Microscopy. Plenum Press, New York.
LITERATURE CITED
Leitz, E. 1938. The Microscope And Its Application. Ernst Leitz, Wetzlar, Germany.
Abramowitz, M. 1994. Optics: A Primer. Olympus America Inc., New York.
Mollring, F.K. 1976. Microscopy From The Very Beginning. Carl Zeiss, Oberkochen, Germany.
KEY REFERENCES
Spencer, M. 1982. Fundamentals of Light Microscopy. Cambridge University Press, Cambridge, UK.
Abramowitz, M. 1985. Microscope Basics and Beyond. Olympus Corporation, New York. Abramowitz, M. 1987. Contrast Methods in Microscopy: Transmitted Light. Olympus Corporation, New York. Abramowitz, M. 1993. Fluorescence Microscopy: The Essentials. Olympus America Inc., New York. Abramowitz, 1994. See above. Bradbury, S. 1984. An Introduction to the Optical Microscope. Oxford University Press, Oxford, UK.
Contributed by Mortimer Abramowitz Olympus America Inc. Melville, New York Marc M. Friedman AccuMed International Chicago, Illinois
Microscope Objectives
2.2.14 Current Protocols in Cytometry
Cameras One of the most critical components of an image cytometer is the camera, the unit that converts the optical image into electrical form so that it can be viewed on a TV monitor, recorded, or digitized for subsequent analysis. Any image degradation introduced at this stage will affect the quality and accuracy of the system’s output. Compared to those of the past, modern cameras offer an array of reasonably inexpensive alternatives. The proper choice of camera depends on the usage for which the instrument is designed. The camera should be matched both to the other system components and to the problems that will be addressed by the instrument. The camera is the eye of an image cytometer. It is called upon to convert a two-dimensional spatial distribution of light intensity into a corresponding electrical signal that is a faithful representation of the specimen. This, in turn, can be displayed, recorded, or processed. How accurately the camera can conduct this transformation affects the quality of results one can obtain from the instrument. Image sensing is a complex technology that harnesses a variety of phenomena. There are numerous sources of noise, distortion, and loss of resolution in the process. Any particular camera represents a series of design tradeoffs, and its performance will be higher in some areas and lower in others when compared with another unit. A poor choice of camera can severely limit the accuracy and usefulness of an image cytometry system. The camera with the best specifications or the highest price is not always the right one for a particular job. It is necessary to understand the fundamentals of camera phenomena in order to design an image cytometer or use it to best advantage. Fortunately, many of the image degradations that are introduced by camera shortcomings can be reduced or eliminated by subsequent image processing. Thus, the camera’s performance is interwoven with that of the software.
DEFINITIONS Image sensing involves three steps: sampling, transduction, and scanning. The image is divided up into an array (usually a rectangular grid) of small regions, called “picture elements” or “pixels,” that can be considered one at a time. Sampling is the process of measuring
Contributed by Kenneth R. Castleman Current Protocols in Cytometry (1998) 2.3.1-2.3.15 Copyright © 1998 by John Wiley & Sons, Inc.
UNIT 2.3 the light intensity at an individual pixel location. It requires a sampling aperture that defines the size and shape of the pixel (usually circular, square, or rectangular). Transduction is the process of converting the light intensity at a particular pixel location into a corresponding voltage. Scanning is the process of selectively addressing the picture elements in order. This creates the data stream that represents the image. If the image is to be processed digitally, it must be digitized. Quantization is the process of generating an integer that reflects the brightness of the image at a particular pixel location. Digitization is the process of sampling and quantizing an image. The degree to which a camera can reproduce small objects is its resolution. If it warps the objects in the image, this is distortion. Any undesirable additive components of the image are called noise.
IMAGE SENSING Light Sensing Light-sensing devices produce an electrical signal proportional to the intensity of light falling upon them. Different physical phenomena can be employed for this purpose, giving rise to different types of light sensors. Photoconductors, such as selenium, show a drop in their electrical resistance when exposed to light. Semiconductor devices made from pure silicon crystals generate free electrons in response to incident photons. Both these phenomena have been harnessed to sense images.
Photometry Photometry is the technology of quantifying light intensity, and there are many ways to do this. For example, photons of different wavelength have different energy. Thus, if incident light energy flux is measured, the spectrum of the light affects the intensity. Commonly used image sensors, however, merely count photons, so wavelength considerations do not directly affect the measured intensity. Although the sensors do have different sensitivities at different wavelengths, this is best accounted for separately. A quality image digitizer will produce an array wherein each gray level is proportional to the number of photons that landed on that pixel during the exposure time.
Image Cytometry Instrumentation
2.3.1 Supplement 5
The linearity of an image sensor specifies how accurately its output reflects the incident photon flux. Modern charge-coupled device (CCD) image sensor chips are quite linear over their entire range, and thus linearity is seldom a problem, as long as saturation (overload) of the sensor is avoided.
Scanning Conventions Video scanning conventions Figure 2.3.1 illustrates the Electronic Industries Association (EIA) RS-170 scanning convention, which is the standard for monochrome broadcast television in the United States (Fink, 1957; Fink and Christiansen, 1989; Hutson et al., 1990; Castleman, 1996). The beam scans the entire image in 525 horizontal scan lines, 30 times each second. The lines are not scanned in sequential order, however, because if the TV screen were to be refreshed at only a 30/sec rate, the eye would perceive an annoying flicker. Instead, an interlaced scanning convention is used to yield an apparent 60/sec refresh rate on the screen. Each frame is made up of two interlaced fields, each consisting of 262.5 lines. The first field of each frame scans all the odd-numbered lines, while the second field scans the intervening even-numbered lines. Interlacing yields a 60/sec field rate to minimize perceived flicker, while the 30/sec frame rate reduces the frequency bandwidth as required for broadcast television channels.
1 2 3 4
Each horizontal line scan requires 63.5 µsec, of which ~50 µsec (83%) is active, containing image information. Of the 525 lines per frame, 16 are lost in the vertical retrace of each field, leaving about 483 active lines per frame. The bandwidth of the standard video signal extends up to 4.5 megahertz (MHz), which allows 225 cycles, or about 550 pixels worth of information, across the active portion of each line. The NTSC (National Television Standards Committee) timing standard for color television in the USA differs only slightly from the RS-170 convention. It was designed to accommodate color transmission while maintaining compatibility with existing monochrome receivers. Different scanning conventions are used in other countries. For example, the CCIR (Comité Consultatif International des Radiocommunications) standard used in much of Europe employs a frame of 625 interlaced scan lines of about 768 pixels each and runs at 25 frames/sec. New broadcast scanning conventions, offering more lines per frame, more pixels per line, and higher image quality, are being developed in several countries under the name “high-definition television” (HDTV). Two proposals now under consideration in the USA are 1280 pixel × 720 line progressive (noninterlaced) scanning, and 1920 pixel × 1080 line interlaced scanning; both of these standards employ a 16:9 aspect ratio. One can use a video camera as an image digitizer simply by sampling the video signal with a fast analog-to-digital converter operat-
0
30 frames/sec 525 lines/frame 15,750 lines/sec 63.5 µsec/line 2 fields/frame 262.5 lines/field 60 fields/sec
50 µsec active/line 525
Cameras
524
483 active lines/frame
Figure 2.3.1 The RS-170 scanning convention, used for monochrome broadcast television transmission in the United States. The CCIR convention, used in much of Europe, employs 625 lines and operates at 25 frames/sec.
2.3.2 Supplement 5
Current Protocols in Cytometry
ing at ∼14 MHz. A frame grabber is a digitizer that stores this high-speed data stream in a solid-state memory and then feeds it out at a slower rate to a more permanent storage device, such as a disk drive. Other scanning conventions Cameras designed specifically for scientific image sensing can be made to scan by any set of timing rules. Typically scientific cameras have larger image formats (more scan lines and pixels per line), noninterlaced scanning, and slower readout rates (to reduce the noise associated with readout). They also may incorporate variable-length frame-integration periods (to increase sensitivity) and sensor-chip cooling (to reduce thermal noise).
Camera Performance Although cameras differ in the approach they use to sense an image, they can be compared on the basis of their performance. Image size An important parameter is the size of the image a camera produces. Image size is specified by the maximum numbers of scan lines and of pixels per line. Pixel size and spacing Two important characteristics are the size of the sampling aperture and the spacing between adjacent pixels. These parameters, specified at the image plane, scale down to the specimen plane by the magnification factor of the microscope. This is usually the objective power multiplied by any auxiliary magnification that is in place. The eyepieces normally do not contribute to this calculation. Resolution According to the Rayleigh criterion, one can just resolve (identify as separate) two point objects in a microscope image if they are separated by the distance δ = 0.61λ/NA, where NA is the numerical aperture of the objective and λ is the illumination wavelength (Castleman, 1996). The camera should be able to reproduce detail to this degree. Linearity The degree of linearity of the relationship between the input light intensity and the output signal amplitude is another important factor. Although the eye is not particularly critical in this department, the validity of subsequent pro-
cessing can be jeopardized by a nonlinear camera. Noise Finally, one of the most important characteristics of a camera is its noise level. If a uniformly gray image is presented to a camera, its output will show variations in gray level, even though the input brightness is constant across the image. Such noise introduced by the camera is a source of image degradation, and should be small relative to the contrast of the specimen. Requirements Whether or not a particular camera is adequate depends on the specific task at hand. In some applications, digitizing images with relatively few lines, pixels per line, or gray levels or with appreciable noise and nonlinearity may be sufficient. Image cytometry, however, normally requires a high-quality camera that is capable of sensing large images with many gray levels, good linearity, and a low noise level.
TYPES OF CAMERAS Historically, imaging tubes, such as the vidicon and its relatives, were the backbone of image cytometry. Currently, however, solidstate cameras generally offer more flexibility, better performance, and lower cost.
Tube-Type Cameras Vidicon construction Figure 2.3.2 illustrates the construction of the vidicon, a common type of television image-sensing tube. It is a cylindrical glass envelope containing an electron gun at one end and a target and faceplate at the other. The tube is surrounded by a yoke containing electromagnetic focus and beam deflection coils. The faceplate is coated on the inside with a thin layer of photoconductor over a thin transparent metal film, forming the target. Adjacent to the target (to the left in Fig. 2.3.2) is a positively charged fine wire screen called the mesh. A smaller positive charge is applied to the target. Vidicon operation A stream of electrons is projected from the electron gun, focused to a small spot on the target by the focus field, and steered across the target in a scanning pattern by the time-varying deflection field. Electrons decelerate after passing the mesh, and reach the target with approximately zero velocity. The moving electron
Image Cytometry Instrumentation
2.3.3 Current Protocols in Cytometry
Supplement 5
deflection coils electron gun
anode target faceplate
glass envelope electron beam
Figure 2.3.2 Vidicon camera tube construction.
beam deposits a layer of electrons on the inner surface of the photoconductor to balance the positive charge on the metal coating on the opposite side. Light striking an area of the photoconductor causes electrons to flow through, locally depleting the surface charge layer at that point. The optical image formed on the target then causes the photoconductor to leak electrons until an identical electron image is formed on the back of the target. Electrons will be present in dark areas and absent in light areas of the image. As the electron beam scans the target, it replaces the lost electrons, restoring a uniform surface charge. As the electrons are replaced, a current flows in the external circuit of the target. This current is proportional to the number of electrons required to restore the charge and therefore to the light intensity at that point. Current variations in the target circuit produce the video signal. The electron beam repeatedly scans the surface of the target, replacing the charge that bleeds away. The vidicon target is thus an integrating sensor, with the period of integration equal to the scanning frame rate. The vidicon family The photoconductor target of a standard vidicon is made of selenium photoconductor material. Relatives of the vidicon, with similarsounding names, differ mainly in the composition of the photoconductive target, and each excels in certain imaging characteristics.
CCD Cameras
Cameras
Silicon light sensors Pure silicon can be grown in large crystals in which each atom is covalently bonded to its six neighbors in a three-dimensional rectangu-
lar lattice structure. An incident photon can break one of these bonds, freeing an electron. A thin metal layer deposited on the surface of the silicon and charged with a positive voltage creates a potential well that collects and holds the electrons thus freed. Each potential well corresponds to one pixel in an array of sensors. A potential well can hold about 106 electrons on typical chips. Thermal energy also causes random bond breakage, creating thermal electrons that are indistinguishable from photoelectrons. This gives rise to dark current, current produced in the absence of light. Dark current is temperature sensitive, doubling for each 6°C increase in temperature. At the long integration times that are required for image sensing at low light levels, the wells can fill with thermal electrons before filling with photoelectrons. Cooling is often employed to reduce dark current and thereby extend the usable integration time. CCD construction CCD chips are manufactured on a light-sensitive crystalline silicon chip, as discussed above (Janesick and Elliot, 1992). A rectangular array of photodetector sites (potential wells) is built into the silicon substrate. Photoelectrons produced in the silicon are attracted to and held in the nearest potential well. By controlling the electrode voltages, they can be shifted as a charge packet from well to well until they reach an external terminal. CCD operation There are three architectures that can be employed for reading the accumulated charge out of CCD image sensor arrays. These are (1) full-frame architecture, (2) frame-transfer architecture, and (3) interline-transfer architecture (Fig. 2.3.3).
2.3.4 Supplement 5
Current Protocols in Cytometry
line shift
image array full-frame array
line shift
line shift
masked storage array
serial register
serial register
serial register
pixel shift full-frame CCD
pixel shift frame-transfer CCD
pixel shift interline transfer CCD
Figure 2.3.3 CCD chip operation using full-frame, frame-transfer, or interline transfer architecture.
Full-frame CCD. Following exposure, a full-frame CCD is shuttered to keep it in the dark during the readout process. It then shifts the charge image out of the bottom row of sensor wells, one pixel at a time. After the bottom row is empty, the charge in all rows is shifted down one row, and the bottom row is again shifted out. This process repeats until all rows have been shifted down and out. The device is then ready to integrate another image. Frame-transfer CCD. A frame-transfer CCD chip has a doubly long sensor array. The top half senses the image in the standard manner, while the storage array on the bottom is protected from incident light by an opaque mask. At the end of the integration period, the charge image that has accumulated in the sensing array is shifted rapidly, row by row, into the storage array. From there it is shifted out pixel by pixel in the standard manner, while the sensing array integrates the next image. Like interline transfer, this technique employs simultaneous integration and readout, making video-rate image sensing possible. Interline-transfer CCD. In an interline transfer CCD every second column of sensors is covered by an opaque mask. These columns of masked wells are used only in the readout process. After exposure, the charge packet in each exposed well is shifted into the adjacent masked well. This transfer requires very little time because all charge packets shift at once. While the exposed wells are accumulating the next image, the charge packets in the masked columns are being shifted out in the same way as in full-frame CCD. In an interline transfer sensor, the number of pixels per line is half the actual number of wells per row on the chip. No more than 50% of the chip area is light-sensi-
tive, because the masked columns cover half its surface. CCD performance Available in a variety of configurations, CCDs give rise to compact and rugged solidstate cameras for both television and image digitizing applications. They are free of geometric distortion and exhibit highly linear response to light. CCDs are therefore emerging as the device of choice for image cytometry. CCDs can be scanned at television rates (30 frames/sec) or much more slowly. Because they can integrate for periods of seconds to hours to capture low-light-level images, they are often used in fluorescence microscopy. Integration times longer than a few seconds require cooling the chip well below room temperature to reduce dark current, which would otherwise fill the wells with thermal electrons before photoelectrons had a chance to build up. Because of imperfections in the crystal lattice, dark current varies significantly from one pixel to the next, particularly in less expensive chips. In long-exposure images, this leaves a “starfield” of fixed-pattern noise due to the few pixels with abnormally high dark current. Because this pattern is stationary, it can be recorded and subtracted out, provided the offending pixels are not allowed to saturate. Defects in the crystal lattice can cause “dead pixels,” which will not hold or shift electrons. This can wipe out all or part of a column of pixels. CCD sensors are graded on the number of such defects, and the higher-grade chips are more expensive. Readout noise is random noise generated by the on-chip electronics. It ranges from a few to many electrons per pixel depending on chip
Image Cytometry Instrumentation
2.3.5 Current Protocols in Cytometry
Supplement 5
row scan
column scan
video signal
Figure 2.3.4 CID chip operation.
Cameras
design, and gets worse as the charge is read out at a faster rate. It is usually the dominant noise factor under short-exposure, low-light conditions where the dark current and photon noise components are small. Photon noise results from the quantum nature of light. The actual number of photons striking any particular pixel in any one exposure will be random. In general the photon noise component is the square root of the number of electrons that accumulate in a well (i.e., photoelectrons plus thermal electrons). It is usually the dominant noise source under high-exposure or high–dark current conditions. The charge developed at a particular pixel may be shifted as many as two thousand times, depending upon its location in the array. The charge transfer efficiency must be extremely high or significant numbers of photoelectrons will be lost in the readout process. Often, half or more of the available area of the sensor is covered by opaque charge-transfer circuitry, leaving gaps between the pixels and reducing the fill factor below the ideal of 100%. The chip can be coated with a thin layer of lenslets, each of which focuses the incoming light it receives onto the sensitive areas of one pixel. Overexposure of a CCD sensor can cause
blooming of the image as excess photoelectrons spread to adjacent pixels. Spectral sensitivity may also be a significant issue. Silicon sensors become less sensitive at the deep blue and ultraviolet end of the wavelength spectrum. This can be overcome by a lumigen coating, which absorbs the shortwavelength photons and then reemits the energy as longer-wavelength photons that the silicon can see. Dynamic range characterizes the performance of the chip at high light levels, where the wells can be filled with photoelectrons (rather than dark current) during a relatively short exposure. It is computed as pixel well capacity divided by readout noise level, both measured in electrons, and is usually expressed in dB. This parameter is independent of exposure conditions (light level and exposure time). The signal-to-noise ratio (SNR) can be computed as the number of photoelectrons received by a well divided by the total (photon plus readout) noise level; that is, SNR = (F × QE × te)/(Np2 + Nr2)1/2, where F is the incident photon flux, QE the quantum efficiency, te the exposure time, Nr the readout noise level, and Np = [(F × QE + DC)te]1/2 the photon noise that results from the statistical nature of light. The
2.3.6 Supplement 5
Current Protocols in Cytometry
SNR is quite dependent on light level and exposure time.
CID Cameras Charge injection device (CID) sensors (Williams and Carta, 1989; Kaplan, 1990) employ the photoelectronic properties of silicon as do CCDs, but they use a different method of readout. CID construction At each pixel site, the CID has two adjacent electrodes (Fig. 2.3.4) that are insulated from the silicon surface by a thin metal-oxide layer. Each pixel is connected to all the pixels in its column by one electrode, and to all the pixels in its row by the other electrode. Thus, a single pixel can be addressed by its row and column address. If all rows and columns of electrodes are held at a positive voltage, the entire chip accumulates a photoelectron image. CID operation When one electrode is driven to 0 V, the accumulated photoelectrons will shift under the second electrode, creating a current pulse in the external circuitry. The size of this current pulse reflects the amount of accumulated photoelectronic charge. Because the accumulated photoelectrons remain in the well after the shift, this is a nondestructive type of image readout: the pixel can be read repeatedly by shifting the charge back and forth between the electrodes. When both electrodes are driven to zero, the accumulated photoelectrons are injected into the underlying substrate, producing a current pulse in the external circuitry. The size of the pulse again reflects the amount of accumulated charge, but this process leaves the well empty. This destructive readout mode is used to prepare the chip for integrating another image. The circuitry on the chip controls the voltages on the row and column electrodes to effect image integration and destructive and nondestructive readout. Because the CID can address individual pixels in any order, subimages of any size can be read out at any speed. Nondestructive readout allows one to watch the image accumulate on the chip, which is useful when the length of the required integration period is unknown. CID performance CIDs are largely immune to blooming (charge spreading to adjacent pixels) and to radiation damage. Also, with nondestructive readout, the control program can monitor the
filling of the wells and selectively flush individual pixels that become full before the integration period is over. Because CIDs do not shift charge packets across the array, charge transfer efficiency is not a concern. Unlike the CCD, a small defect in the crystal lattice affects only one pixel. Also, essentially the entire surface area is light sensitive, leaving virtually no gaps between pixels. Even so, CIDs are considerably less light sensitive than similar CCDs.
APS Cameras An emerging new solid-state sensing technology is active pixel sensor (APS) cameras (Janesick and Elliot, 1992; Fossum, 1993, 1995). Like CCD cameras, these are fabricated on a silicon chip, but they use complementary metal-oxide-semiconductor (CMOS) integrated circuit technology. This allows the chip designers to embed on the sensor chip itself processing circuitry that normally exists elsewhere in the camera. Indeed, much of the circuitry that traditionally resides on various circuit boards in the camera can be fabricated directly on the image-sensor chip. Experimental APS camera chips have been developed with amplifiers at each pixel and with special noise-reducing readout circuitry and analog-to-digital (A-to-D) converters on the chip. Some APS chips contain circuitry that allows them to read out a rectangular subimage continuously and nondestructively. APS technology promises to reduce the cost of cameras in the future and perhaps to improve their performance as well.
IMAGE DIGITIZATION Before an image can be processed by computer, it must be converted into an array of numbers. This must be done in such a way that it does not destroy or significantly degrade the specimen content of interest. Both the number of gray levels in the grayscale and the number of pixels per row and column must be adequate for the tasks at hand. Binning is the technique of combining adjacent pixels in a sensor array to form larger pixels. For example, using 2 × 2 binning on a 1024 × 1024 sensor array with 6 × 6-µm pixels would produce a 512 × 512 image where the pixels were effectively 12 × 12 µm.
Noise Sources The time-varying electrical signal emerging from the image sensor is sampled and quantized by an analog-to-digital converter (ADC) cir-
Image Cytometry Instrumentation
2.3.7 Current Protocols in Cytometry
Supplement 5
Table 2.3.1
Front-Illuminated CCD Chips from Kodak
Well size (µm) Well capacity (e−) Frame size (pixels) Chip size (mm) Quantum efficiency at 550 nm Readout noise (e− rms) Dark current (e−/pixel/sec at 0°C) Type of transfer Fill factor Microlensing Dynamic range
KAF0400
KAF1000
KAF1300L
KAF1400
KAF1600
KAF4200
KAF6300
9×9 80,000 768 × 512 6.9 × 4.6 0.36
24 × 24 630,000 1,024 × 1,024 24.6 × 24.6 0.40
16 × 16 140,000 1,280 × 1,024 20.5 × 16.4 0.32
6.8 × 6.8 45,000 1,317 × 1,035 9.0 × 7.0 0.40
9×9 80,000 1,536 × 1,024 13.8 × 9.2 0.39
9×9 80,000 2,032 × 2,044 18.4 × 18.4 0.38
9×9 85,000 3,072 × 2,048 27.65 × 18.5 0.38
19 at 1 Mhz 18 at 1 Mhz
14 at 2 MHz
13 at 500 kHz 13 at 500 kHz 12 at 500 kHz 22 at 5 MHz
0.44
14.4
7.2
0.31
0.31
0.90
7.2
Full frame 100% No 72 dB
Full frame 100% No 82 dB
Full frame 100% No 80 dB
Full frame 100% No 72 dB
Full frame 100% No 72 dB
Full frame 100% No 76 dB
Full frame 100% No 72 dB
cuit. If B is the number of bits used in the quantization, the grayscale goes from zero to 2B − 1. Quantization can be viewed as a source of noise, because it alters the gray level at each pixel by a small random amount. The signalto-noise ratio (SNR) for quantization is the (full-scale) signal amplitude divided by the quantization noise level. For images with a Gaussian distribution of gray levels, the SNR, measured in decibels (dB), is 20 × log10[2B/σn] = 6B + 11, where B is the number of bits used in the quantization and σn is the standard deviation of the resulting quantization noise. Each 20 dB of SNR represents a factor of ten in the ratio. The commonly used eight-bit grayscale (B = 8, 2B = 256 gray levels, SNR = 59 dB = 891) is adequate for many image cytometry applications. Normally this quantization noise level (± 0.11% of full scale) is tolerable, but one should verify this and use ten or more bits of grayscale resolution if that is required by the application. The camera introduces other random noise components as well. Readout noise, introduced by the circuitry on the CCD chip, and photon noise, which results from the statistical nature of light, are discussed above (see CCD performance). In general, the different random noise sources combine in such a way that the overall noise level is the square root of the sum of the squares of their individual amplitudes.
Spatial Resolution Cameras
The well-known Shannon sampling theorem states that one can reconstruct, by proper
interpolation, a sinusoidal signal from equally spaced sample points if there are no fewer than two sample points per cycle of the sine wave (Castleman, 1996). If the sampling is done more sparsely, one can encounter the phenomenon of aliasing, which introduces Moiré patterns into the image. A microscope objective cannot pass image detail at frequencies higher than the optical cutoff frequency of fc = 2NA/λ, where NA is its numerical aperture and λ is the illumination wavelength (Castleman, 1996). Thus, aliasing can be avoided completely if the pixel spacing at the specimen is no larger than λ/4 NA. This is 1⁄8 µm for an objective with NA = 1 operating in green (λ = 500) light. For applications in which the specimen does not contain detail at the resolution limit of the objective lens, larger pixel spacing will suffice. However, even smaller pixel spacing may be required for accurate measurement of objects in the image or for optimal display of the image. In these cases, reduced pixel spacing can be achieved by interpolation of the image after it is digitized (Castleman, 1996).
AVAILABLE CCD CHIPS AND CAMERAS An impressive array of solid-state cameras, incorporating a variety of different CCD chips, is commercially available. These cover a wide range of cost and performance. In this section we tabulate some of the ones that are potentially most useful for cytometry. The list is by no means exhaustive, and the CCD camera situation is subject to rapid change.
2.3.8 Supplement 5
Current Protocols in Cytometry
The performance of a particular CCD camera depends on two major design factors: the choice of the CCD sensor itself, and the design of the supporting electronics in the camera. A poor quality chip in a well-designed camera, and a good chip embedded in poorly matched circuitry, will be equally disappointing. The circuitry in a well-designed camera will exploit the best characteristics of the sensor chip. Overall camera performance cannot exceed the limitations of either the chip or the electronics. Thus a particular CCD camera must be evaluated as a complete system. With that point made, we herein tabulate chip and camera characteristics separately for the sake of brevity.
CCD Chips Tables 2.3.1, 2.3.2, 2.3.3, 2.3.4, 2.3.5, 2.3.6, 2.3.7, and 2.3.8 have been compiled from manufacturers’ data. Achieving a consistent set of specifications is difficult because each chip maker chooses to specify chip characteristics differently. We include the commonly used
Table 2.3.2
CCD Cameras Table 2.3.9 presents specifications on several instrumentation quality CCD cameras. The terms used in the table are defined above (see Image Digitization). Many of the relevant specifications are not available from all manufacturers in a consistent form. Where entries are not shown, reliable data were unavailable at the time of publication. The very important sensitivity parameter, for example, is specified in so
Front-Illuminated Chips from Scientific Imaging Technologies, Inc.
Well size (µm) Well capacity (e−) Frame size (pixels) Chip size (mm) Quantum efficiency at 550 nm Readout noise (e− rms) Dark current (e−/pixel/sec at 0°C) Type of transfer Fill factor Dynamic range
Table 2.3.3
mechanical specifications and show the electrical parameters in common units. Well capacity and RMS readout noise are given in electrons, and dark current in electrons per second for a single pixel at 0°C. Dark current doubles for each 6°C increase in temperature, and vice versa. Dynamic range, computed as well capacity divided by readout noise, also appears in the tables. Signal-to-noise ratio is quite dependent on exposure conditions (light level and exposure time), and thus is not listed in the tables.
SI502AF
STAE01AF, SIA003AF
SIA002AB
24 × 24 325,000 1,100 × 165 26.4 × 4 0.30 20 at 1 Mhz 160 Full frame 100% 84 dB
24 × 24 325,000 1,024 × 1,024 24.6 × 24.6 0.30 20 at 1 Mhz 353 Full frame 100% 84 dB
15 × 15 100,000 4,096 × 2,048 30.7 × 30.7 0.30 7 at 45 kHz 10 Frame transfer 100% 83 dB
Front-Illuminated Chips from Thomson-CSF
Well size (µm) Well capacity (e−) Frame size (pixels) Chip size (mm) Quantum efficiency at 550 nm Readout noise (e− rms) Dark current (e−/pixel/sec at 0°C) Type of transfer Fill factor Dynamic range
TH7895M
TH7896M
TH7887
19 × 19 375,000 512 × 512 9.7 × 9.7 0.30 25 at 500 kHz 72 Full frame 100% 84 dB
19 × 19 375,000 1,024 × 1,024 19.5 × 19.5 0.35 25 at 500 kHz 90 Full frame 100% 84 dB
14 × 14 250,000 1,024 × 1,024 14.3 × 14.3 0.10
Frame transfer 100% 80 dB
Image Cytometry Instrumentation
2.3.9 Current Protocols in Cytometry
Supplement 5
overview of their cameras and the CCD chips used.
many different ways by different manufacturers that comparison is impossible, and it is thus conspicuously absent from this table. Price ranges are specified as “low” (<$10,000), “mid” ($10,000 to 20,000), and “high” (>$20,000). Photometrics cameras have three gain settings. Specifications are given here for the 1× gain setting, which matches the full scale range of the analog-to-digital converter (ADC) to the well capacity of a single pixel. The 4× gain mode, where one-quarter of full well capacity saturates the ADC, achieves greater sensitivity for use at low light levels. The 0.5× gain mode, when used with binning, increases the effective well size to improve the SNR at high light levels. Photometrics product literature (http://www.photomet.com) provides useful specifications and other helpful information about CCD cameras. Princeton Instruments’ catalog, available in Adobe Acrobat format (http://www.prinst. com/imagprod.htm), provides a comprehensive Table 2.3.4
Each image cytometry application deserves its own analysis of camera and digitizing requirements. When it is possible to select a specific camera, camera characteristics should be evaluated in light of the requirements of the planned experiments. In any case, the camera should be well matched to the problem and to the other system components. Although an inadequate camera can forestall success, camera overkill can waste resources that might be better applied elsewhere. Modern cameras are quite good in performance and reasonable in cost compared to those of the past, and will undoubtedly continue to improve. In addition, some of the image detail lost to camera-induced noise, distortion, and lack of resolution can be recovered with digital image processing.
Front-Illuminated Chips from EEV
Well size (µm) Well capacity (e−) Frame size (pixels) Chip size (mm) Quantum efficiency at 550 nm Readout noise (e− rms) Dark current (e−/pixel/sec at 0°C) Type of transfer Fill factor Dynamic range
Table 2.3.5
Cameras
CONCLUSION
EEV CCD37
EEV 02-06
EEV 05-20
15 × 15 165,000 512 × 512 7.7 × 7.7 0.26 23 at 2 Mhz 1,078 Frame transfer 100% 77 dB
22 × 22 500,000 288 × 384 12.7 × 8.4 0.32 28 at 1 Mhz 2,580 Frame transfer 100% 85 dB
22.5 × 22.5 500,000 576 × 770 12.9 × 17.3 0.30 22 at 500 kHz 2,580 Frame transfer 100% 87 dB
Front-Illuminated Chips from Texas Instruments and Loral
Well size (µm) Well capacity (e−) Frame size (pixels) Chip size (mm) Quantum efficiency at 550 nm Readout noise (e− rms) Dark current (e−/pixel/sec at 0°C) Type of transfer Fill factor Microlensing Dynamic range
T.I. TC-255
Loral CCD442A
10 × 10 50,000 320 × 240 3.2 × 2.4 0.4 28 97 Full frame 100% No 70 dB
15 × 15 150,000 2,048 × 2,048 30.7 × 30.7 9 at 100 kHz 102
84 dB
2.3.10 Supplement 5
Current Protocols in Cytometry
Table 2.3.6
Front-Illuminated Chips from EG&G Reticon
Well size (µm) Frame size (pixels) Chip size (mm) Type of transfer Fill factor Microlensing Dynamic range
Table 2.3.7
RL2005PAQ-11
RL2010PAQ-11
RL2020PAQ-11
14 × 14 512 × 512 7.2 × 7.2 Frame transfer 100% No 48 dB
14 × 14 1,024 × 1,024 14.3 × 14.3 Frame transfer 100% No 48 dB
14 × 14 2,048 × 2,048 28.6 × 28.6 Frame transfer 100% No 48 dB
Back-Illuminated Chips from Scientific Imaging Technologies, Inc.
Well size (µm) Well capacity (e−) Frame size (pixels) Chip size (mm) Quantum efficiency at 550 nm Readout noise (e− rms) Dark current (e−/pixel/sec at 0°C) Type of transfer Fill factor Microlensing Dynamic range
SIAE01AB, SI502AB SIA003AB
SIA002AB
24 × 24 325,000 1,100 × 165 26.4 × 4 0.8 20 at 1 Mhz 64 Full frame 100% No 84 dB
15 × 15 100,000 4,096 × 2,048 30.7 × 30.7 0.8 7 at 45 kHz 10 Frame transfer 100% No 83 dB
24 × 24 325,000 1,024 × 1,024 24.6 × 24.6 0.8 20 at 1 Mhz 353 Full frame 100% No 84 dB
Table 2.3.8 Back-Illuminated Chip from Princeton Instruments
P.I. 1000PB Well size (µm) Well capacity (e−) Frame size (pixels) Chip size (mm) Quantum efficiency at 550 nm Readout noise (e− rms) Dark current (e−/pixel/sec at 0°C) Type of transfer Fill factor Microlensing Dynamic range
15 × 15 60,000 1,000 × 800 15 × 12 0.67 20 at 500 kHz 97 Full frame 100% No 70 dB
Image Cytometry Instrumentation
2.3.11 Current Protocols in Cytometry
Supplement 5
Table 2.3.9
CCD Cameras
Photometrics SenSys 0400/1400/1600 Chip KAF0400/KAF1400/KAF1600 Binning options Arbitrary M × N Cooling method 10°C, thermoelectric A-to-D conversion 12 bits Frame readout time at rate 0.41/1.39/1.6 sec at 1 MHz Dynamic range 72.5/67/72 dB Price Mid-range Web site http://www.photomet.com Photometrics Quantix Chip Binning options Cooling method A-to-D conversion Frame readout time at rate Dynamic range Price
KAF1400 Arbitrary M × N −25°C, forced air; −35°C, liquid circulation 12 bits 0.3 sec at 5 MHz 67 dB High-range
Photometrics S300:1400/4200/502B/003B/003F/7895/7896 Chip KAF1400/KAF4200/SI502AB/SI003AB/SI003AF/TH7895M/ TH7896M Binning options Arbitrary M × N −25°C, forced air; −40°C, liquid circulation; −90°C, −110°C, LN2 Cooling method A-to-D conversion 12, 14, or 16 bits (chip dependent) Frame readout time at rate 2.8/8.7 sec at 500 kHz/1.4/5.4/5.4/1.4/5.5/5.5 sec at 200 kHz Dynamic range 70/77/89/86/86/87/88 dB Price Mid- to high-range Photometrics PXL:37/1000/1300L/1400/6300/003F/003B Chip CCD37/KAF1000/1300L/1400/6300/SI003AF/SI003AB Binning options Arbitrary M × N −25°C, liquid circulation Cooling method A-to-D conversion 12, 14, or 16 bits (chip dependent) Frame readout time at rate 0.16/0.58/0.74/2.6/3.9 sec at 2 mHz/1.2/1.4 sec at 1 MHz Dynamic range 77/82/82/75/75/79/76 dB Price High-range Princeton Instruments CCD-576E/CCD-770E/CCD-1242E Chip EEV 02-06/EEV 05-20/EEV 05-30 Binning options Flexible −35° to −130°C, various methods Cooling method A-to-D conversion 12, 14, or 16 bits Frame readout time at rate 0.286 sec at 1 MHz/1.8/1.5 sec at 500 kHz Dynamic range 85/87/86 dB Price Mid- to high-range Web site http://www.prinst.com/imagprod.htm continued
Cameras
2.3.12 Supplement 5
Current Protocols in Cytometry
Table 2.3.9
CCD Cameras, continued
Princeton Instruments CCD-512SF/CCD-1024SF/CCD-512SB/CCD-1024SB Chip SITe SI502FA/SIA003AF/SITe SI502AB/SIA003AB Binning options Flexible −30° to −140°C, various methods Cooling method A-to-D conversion 14 or 16 bits Frame readout time at rate 0.29/1.2/0.5 sec at 1 MHz/2.32 sec at 500 kHz Dynamic range 84/82/84/88 dB Price Mid- to high-range Princeton Instruments CCD-768K/CCD-1280K/CCD-1317K/CCD-1536K/CCD-2033K Chip KAF0400/KAF1300L/KAF1400/KAF1600/KAF4200/KAF6300 Binning options Flexible −35° to −60°C, various methods Cooling method A-to-D conversion 12 or 14 bits Frame readout time at rate 0.41/1.4/1.4/1.8/4.2/6.3 sec at 1 MHz Dynamic range 72/84/73/77/77/72 dB Price Mid- to high-range Princeton Instruments CCD-1000PB Chip P.I. 1000PB Binning options Flexible −35° to −55°C, various methods Cooling method A-to-D conversion 14 or 16 bits Dynamic range 70 dB Price Mid- to high-range PixelVision SV512/SV10K/SV165/SV20K Chip SITe SI502AB(F)/SIA003AB(F)/STAE01AB(F)/Loral CCD442A Binning options Yes Cooling method 40° or 70°C below ambient A-to-D conversion 16 bits Readout rate 100 kHz to 1 MHz Dynamic range 84 dB at 1 MHz Price Mid- to high-range Web site http://www.site-inc.com Hamamatsu C4880-10/20/30/31/40 Chip T.I. TC-215/S5466/SITe SI502A/SI003A/ICX-074 Binning options Yes Cooling method 5° to −120°C, using air, water, LN2 A-to-D conversion 12, 14, or 16 bits Price Mid- to high-range Web site http://www.hamamatsu.com Richter Enterprises Silicon Mountain Design 1M60-20 Chip Thompson TH7887 Binning options 2×2 A-to-D conversion 12 bits Readout rate 20 MHz Web site http://www.smd.com continued Image Cytometry Instrumentation
2.3.13 Current Protocols in Cytometry
Supplement 8
Table 2.3.9
CCD Cameras, continued
Spectra Source Orbis Chip Binning options Cooling A-to-D conversion Price Web site
Various EEV, Texas Instruments, Kodak, and SITe Yes −30° to −120°C 12 or 16 bits Mid- to high-range (some low-priced models) http://www.optics.org/spectrasource/producttoc.html
Capella Cooled CCD Chip Cooling A-to-D conversion Readout rate
Kodak KAF0400, Kodak KAF1300, EEV CCD37 Pelter, LN2 12 or 14 bits 500 kHz to 8 MHz
Diagnostic Instruments Inc. Spot (color) Chip Kodak KAF1400 A-to-D conversion 8 bits per color, liquid-crystal filter Web site http://www.diaginc.com/spotspec.htm Micro Photonics UltraPix Astrocam Chip Kodak KAF0400/KAF1600 Binning Yes Frame readout time 5 sec/20 sec Web site http://www.microphotonics.com/ccdupix.html EG&G Reticon LD2000 Chip Binning Readout rate A-to-D conversion
RL2005PAQ-11/RL2010PAQ-11/RL2020PAQ-11 Yes 33 MHz 10 bits
Integrated Scientific Imaging Chip KAF0400/KAF1600 Cooling 45°C below ambient A-to-D conversion 14 bits Note Integrated with filter wheel Optic PixCel 225 Chip Binning Cooling A-to-D conversion Web site
T.I. TC-255 2×2 35°C below ambient 14 or 16 bits http://www.optecinc.com/pixcel.htm
Xillix Microimager Chip Binning Readout rate A-to-D conversion Price Web site
Kodak KAF-1400 2×2 500 kHz to 8 MHz, up to 5 frames/sec 10 or 12 bits Low- to mid-range http://www.xillix.com/microimagerspecs.html
Cameras
2.3.14 Supplement 8
Current Protocols in Cytometry
LITERATURE CITED Castleman, K.R. 1996. Digital Image Processing. Prentice Hall, Englewood Cliffs, N.J. Fink, D.G. (ed.). 1957. Television Engineering Handbook. McGraw-Hill, New York. Fink, D.G. and Christiansen, D. 1989. Electronics Engineers Handbook. McGraw-Hill, New York. Fossum, E. R. 1993. Active pixel sensors: Are CCDs dinosaurs? Proc. SPIE 1900:2-14. Fossum, E.R. 1995. CMOS image sensors: Electronic camera on a chip. IEEE International Electron Devices Meeting Technical Digest, Dec. 10-13, 1995, Washington, D.C.
Hutson, G., Shepherd, P., and Brice, J. 1990. Colour Television Theory: System Principles, Engineering Practice and Applied Technology. McGrawHill, New York. Janesick, J. and Elliot, T. 1992. History and advancements of large area array scientific CCD imagers. Astronom. Soc. Pacif. Conf. Ser. 23:1-67. Kaplan, H. 1990. New jobs for charge-transfer devices. Photonics Spectra (Nov.). Williams B. and Carta, D. 1989. CID cameras: More than an alternative to CCDs. Adv. Imaging (Jan.).
Contributed by Kenneth R. Castleman Perceptive Scientific Instruments, Inc. League City, Texas
Image Cytometry Instrumentation
2.3.15 Current Protocols in Cytometry
Supplement 5
Optical Filtering Systems for Wavelength Selection in Light Microscopy The recent renaissance of light microscopy is based on a number of exciting technological developments that enable microscopic imaging of biological cells and tissues at greatly improved spatial, temporal, and spectral resolution (Taylor et al., 1997). Numerous disciplines (physics, chemistry, molecular biology, engineering, computer science) contribute to this enhanced performance of microscopic imaging. Fluorescence-based detection is one of the foremost methods, given its sensitivity and specificity. Its high quality is made possible by the ability to detect the emission corresponding to a single molecular species, based on selective labeling and equally selective optical detection, with simultaneously high spatial and wavelength resolution. This need has led to dramatic improvements in filter system design and implementation. With the concomitant development of sensitive fluorescent dyes (at multiple excitation wavelengths), multicolor imaging has become a powerful tool for analyzing structure and function of cells and tissues (Farkas, 2000). This unit reviews some of the main principles and developments in optical filtering, beginning with optical interference filters and colored glass filters that are the basis of color selection for the majority of commercial microscope systems. Table 2.4.1 gives a summary of the relative merits of the various waveTable 2.4.1
length selection techniques that will be discussed.
OPTICAL INTERFERENCE FILTERS AND COLORED GLASS FILTERS Optical interference filters and colored glass filters have gained wide acceptance over other devices for selecting wavelengths of light; they provide a small and relatively inexpensive way to achieve good wavelength selectivity. There are several types of optical filters that can be obtained. Long-pass filters have the light transmission spectra shown in Figure 2.4.1. They are specified according to their “cut-on” wavelength, which is located at half the maximal transmission of the filter. Short-pass filters transmit at short wavelengths. Transmission spectra for filters can be obtained on most scanning absorption spectrometers, although specialized instruments are required to obtain quantitative measurements of extreme levels of light blocking that can be incorporated into precision optical filters. Band-pass filters (Fig. 2.4.2) transmit light in a narrow range of wavelengths and are classified according to the width of the pass band at half-maximal transmission and the center wavelength of the pass region. Some types of filters have multiple pass bands.
Summary of the Relative Merits of the Various Wavelength Selection Techniques
Method/Tool
Main strengths
Main weaknesses
Filters
Excellent rejection Widely available Variable center wavelength
Fixed spectral characteristics Low
LCTF
UNIT 2.4
Reduced throughput Reduced rejection AOTF Variable center wavelength, Reduced throughput band-pass, and intensity Some image blur Fastest switching Reduced rejection Long acquisition time FTIS Good spectral resolution across large region High memory and processing requirements Grating imaging Only one exposure required High CCD requirements High memory and systems for all information processing requirements Gives spectral data across large wavelength region Prism imaging Gives spectral data across Requires scanning to get systems large wavelength region full x, y, and λ data
Relative cost
Moderate High
High
Moderate
Moderate
Image Cytometry Instrumentation Contributed by Alan S. Waggoner, Elliot S. Wachman, and Daniel L. Farkas Current Protocols in Cytometry (2001) 2.4.1-2.4.11 Copyright © 2001 by John Wiley & Sons, Inc.
2.4.1 Supplement 15
0
100 long-pass filter
75
25 50
light
25
filter
100 0
0 100 75
75
dichroic filter (also long-pass)
25
filter 50 25 0 400
long
light short 500
600 700 Wavelength
Reflectance (%)
Transmittance (%)
50
50 75 100
Figure 2.4.1 Transmission spectra for a long-pass interference filter. Maximum filter transmittance is nearly 100%. In the top panel, the wavelength at half-maximal transmission is ∼580 nm, so this filter is classified as a 580-nm long-pass interference filter. Manufacturers may add coatings to the filter that block the appearance of transmission in the wavelength range shorter than 450 nm. The bottom panel illustrates the transmission spectrum shift when the 580-nm long-pass filter of the top panel is rotated 45°. In the latter orientation the filter can be used as a 520-nm long-pass dichroic filter for color separation.
Transmittance (%)
100
pass-band center
pass-band width (at half-maximal transmittance)
0 540 560 580 600 620 640 660 680 Wavelength (nm)
Optical Filtering Systems for Wavelength Selection in Light Microscopy
Figure 2.4.2 Transmission spectrum of a band-pass interference filter. The center wavelength (top arrow) of this filter is ∼610 nm and the bandwidth at 50% maximal transmittance (between the two lower arrows) is ∼60 nm. Both specifications are required to define the filter, e.g., a 610/60 band-pass interference filter. The transmittance scale used in this figure does not show how effectively this filter blocks light outside the pass band. A logarithmic scale of 6 to 10 decades is required to define blocking ability for top-performance filters. An optical density scale (which is logarithmic) of 6 to 10 decades will also suffice (see for example Fig. 2.4.3).
2.4.2 Supplement 15
Current Protocols in Cytometry
Dichroic filters are designed to steer light of different colors along separate paths. Usually one color is separated and sent at 90° to the path of the remaining light, as illustrated in Figure 2.4.1. Such “beam-splitters” are used for color separation and are ordinarily included in a standard epifluorescence microscope filter set. Depending on filter design, either short or long wavelengths can be reflected. Dichroic filters are basically long-pass, short-pass, or wideband-pass interference filters which are used in a 45° orientation and which have transmission specifications designated for the 45° orientation. The center wavelength and width of the pass region change with angle, as explained later.
CONSTRUCTION OF OPTICAL FILTER COMPONENTS Optical filters are fabricated in basically two ways: (1) by including light-absorbing molecules in gelatin or glass (colored-glass filters) to filter out certain colors of light, and (2) by generating interference effects that block the passage of certain wavelengths. The latter are called interference filters. Colored glass and gelatin filters have the advantage of being inexpensive and having strong light blocking in certain spectral regions. The transition from blocking to maximal transmission in a colored-glass filter is more gradual than for the band-pass interference filters and the latter have gradually taken over for applications that require highest optical selectivity. Interference filters also predominate in multicolor detection applications because they can be made with sharper transitions between transmitting and blocking regions. Another disadvantage of colored glass filters is that they often are weakly fluorescent; as a result, scattered excitation light may produce an unwanted fluorescence signal in the detection channel. Interference filters are prepared by creating partially reflective “cavities” of a thickness 1⁄2 the wavelength of light to be transmitted by the filter. The cavities are formed by layering dielectric materials on glass or quartz in a vacuum-deposition chamber. The incoming light forms an “in-phase” standing wave between the two partially reflective walls of the cavities and is passed efficiently through the filter. The slightly shorter or longer wavelengths, on the other hand, generate “out-of-phase” interference between the walls of the cavities and are therefore not transmitted, but reflected. Filters must be properly oriented in the light path, or the wavelength selectivity is altered. If
the filter is tipped from an orientation perpendicular to incoming light, a somewhat longer wavelength (different color) will be passed through the cavity because the cavity is thicker in the direction of incoming light. For this reason it is important to keep filters perpendicular to the beam of light (or 45° for dichroics). Figure 2.4.3 shows how the near-band light transmission is reduced and the pass-band walls become steeper as additional cavities are added. Multi-cavity filters are ideal for transmitting the fluorescent light in a defined spectral region without passing light from the excitation source or from other fluorophores. Thus, these highperformance multi-cavity filters are of great importance in multiparameter flow and image cytometry. However, filters with more cavities are more expensive. The following are features found in a good interference filter: 1. Good quality control of band-pass width, band center, and maximal transmission. 2. A set of spectral curves delivered with each filter. 3. High blocking (at least 6 orders of magnitude) in the blocking region. 4. Blocking into the UV and IR. 5. Freedom from pinhole defects in layer material. 6. Nonfluorescent coatings (including the glass, glues, blocking agents, and antireflective coatings). 7. Resistance to degradation by high-intensity light and humidity. The filter should be examined every few months for pinhole defects and color change. 8. Low registration shift: for imaging, the image should not shift on the camera as filter sets are changed. 9. Resistance to damage in handling. 10. Clear labeling that contains pass-band width, central wavelength, and product number information. 11. Arrow indicating proper insertion direction in optical path (usually pointing in the same direction as the incoming light).
INTEGRATING OPTICAL FILTER COMPONENTS INTO THE MICROSCOPE SYSTEM Optimal use of optical filters requires a clear understanding of the entire excitation-detection system (Galbraith et al., 1989). All fluorescence microscopes use some form of the epifluorescence filter system shown in Figure 2.4.4. The light source is usually an arc lamp (mercury or xenon) or a laser. Because lasers
Image Cytometry Instrumentation
2.4.3 Current Protocols in Cytometry
Supplement 15
0
1
Optical density
2 2-cavity 3 4 3-cavity 5 6 6-cavity 400
450
500 Wavelength
4-cavity 550
600
Figure 2.4.3 Steeper walls and greater blocking outside the pass band are obtained by increasing the number of cavities of the filter. Note that the optical density (vertical axis) is logarithmic. Blocking can be achieved to greater than one photon per million.
emit at just a few wavelengths with very narrow wavelength bands, the excitation filter requirements for laser scanning microscopes are less stringent. Otherwise, considerations for filter set design are similar for arc-lamp- and laserbased microscopes. Each filter set must provide the optimal excitation wavelength band to reach the sample and allow the optimal emission wavelength band from the sample to reach the imaging camera (or eye). At the same time, the set should minimize light reaching the camera that is scattered excitation light, autofluorescence, or fluorescence from other excited fluorescent probes on the sample.
Single Fluorophore Imaging
Optical Filtering Systems for Wavelength Selection in Light Microscopy
Epifluorescence filter sets for single fluorophore detection are the simplest filter arrangement to set up. Still, there are several decisions that have to be made. First, how broad should the excitation filter bandwidth be and where should it be centered? If the excitation source is a laser, this decision is easy; a narrow bandpass filter centered at the laser wavelength should be used. For a mercury arc lamp, it is best to center the filter over one of the strong emission lines. If the absorption spectrum of the dye lies between mercury emission lines, the bandwidth of the filter should be wider to pass more photons within the absorption band of the dye. This is also the case for xenon arc
lamps, which have very little structure in their emission spectrum. It should be remembered that a neutral-density filter could be included in the excitation path to reduce illumination intensity if photobleaching is a problem. Generally, it is best to collect as much emission signal as possible, and often long-pass colored-glass filters are used as emission filters. For more challenging experiments where there is fluorescence background in the same wavelength region as the emission signal from the probe or label, it is better to take a smaller slice of the emission spectrum centered around the emission maximum of the dye by using a bandpass interference filter. Rejection of scattered excitation light becomes a challenge when capturing fluorescence from dyes (sometimes called fluorophores or fluorochromes) that emit close to the wavelength of excitation. Yet many of the best fluorophores—such as fluorescein, rhodamine, the BODIPYs, Cy2, Cy3, Cy5, and the Alexa dyes—have small Stokes shifts, i.e., separation between the longest-wavelength absorption band and the shortest-wavelength emission band (Fig. 2.4.5). In order to optimally excite and collect fluorescence from such fluorophores it is necessary for the excitation and emission filter pass-band regions to be positioned close to one another; however, the pass bands must not overlap or else scattered light will
2.4.4 Supplement 15
Current Protocols in Cytometry
Detector
excitation filter
emission (or barrier) filter
dichroic filter
Figure 2.4.4 Typical epifluorescence filter set used for detection of fluorescence with a microscope or other detection instrument. Excitation light from an arc, tungsten, or laser light source (solid line) is passed through an excitation filter which isolates the particular band of light for excitation. The excitation light is reflected 90° by a long-pass dichroic filter (with a cutoff halfway between the absorption maximum and the emission maximum of the fluorescent dye) and passes through a lens to illuminate the sample on a microscope slide. Fluorescence, which occurs at a longer wavelength, is collected by the lens and passes through the dichroic filter to the “barrier” or “emission filter.” The emission filter selects a band of fluorescence to be passed on to the detector.
dominate the signal received at the imaging camera, eye, or photomultiplier tube.
Multicolor-Multiprobe Imaging When the goal is to detect fluorescence signals from several different dyes with different excitation wavelengths, an efficient way to discriminate their signals is to use a separate epifluorescence filter set (as shown in Fig. 2.4.4) for each dye. This is a common procedure in microscopy. The image of each different color probe is acquired using a different filter set. Newer microscopes have multiple filter sets, sometimes under computer control. Multicolor fluorescence microscopy requires careful consideration of the spectral properties of the fluorescent probes. This has become a major challenge in light of the large number of new fluorescent labels and probes with different wavelengths that are designed for multiparameter analysis of cells. Composite images with five fluorescent probes have been published by DeBiasio et al. (1987). The problem is that even
the fluorescent dyes with the narrowest absorption and emission spectra have long tails (on the blue side of the absorption spectrum and the red side of the emission spectrum). Thus, each excitation wavelength used may excite more than one fluorophore, and each fluorophore may produce a signal that appears in fluorescence channels for longer-wavelength dyes. To choose filters with minimal fluorescence signal overlap, the first thing needed is a set of accurate emission spectra of the dyes that are used as probes or labels. The dyes must be in the same microenvironment as the biological experiment, or a similar one at least, because most dyes show wavelength shifts when they are transferred to solvents of different polarity. Once accurate emission spectra are obtained, the investigator should try to visualize what optimal bandwidths and central wavelengths should be. Usually the filter manufacturers will help with custom filter development, if they are provided with the absorption and emission spectra of the fluorophores used.
Image Cytometry Instrumentation
2.4.5 Current Protocols in Cytometry
Supplement 15
Absorbance, fluorescence, or transmittance
400
450
500
550
600
Wavelength (nm) Figure 2.4.5 Absorption and fluorescence (emission) spectra of fluorescein. Superimposed on the spectra are transmission spectra (dotted lines) of excitation and emission band-pass filters of an epifluorescence filter set that would be used to generate and detect the fluorescein emission. The transmission spectrum of the dichroic filter of this set is not shown, but it would be a long-pass filter with a half-maximal transmission centered between the absorption and fluorescence peaks of this dye. It appears that the pass band of this emission filter could be moved to a shorter wavelength to capture a greater part of the maximum fluorescence of the dye, or the excitation filter could be moved to a slightly longer wavelength to provide more excitation. This may be true, but it would be necessary to determine more closely the amount of overlap of the excitation and emission pass bands before moving the pass band so that the amount of scattered light reaching the detector is not increased substantially. A sensitive spectrophotometer (absorption/transmission) would be used for this determination.
Optical Filtering Systems for Wavelength Selection in Light Microscopy
Quantitatively designing the bandwidths and central wavelengths of all the epifluorescence excitation and emission filters to be used in a three- to six-fluorophore experiment requires accurate absorption and emission spectra of the fluorophores (in a molecular environment that simulates that of the sample), information on the distribution of excitation wavelength of the excitation source (if an arc lamp), and the wavelength profile of detector sensitivity. A computer program can be used to optimize the filter parameters for epifluorescence microscopy (Galbraith et al., 1989). Designing emission and dichroic filters for laser scanning microscopes is also a challenge and has been extensively discussed by Brelje et al. (1993). Again, filter manufactures and microscope companies should be consulted for help ( e.g., C hr oma Technology Corp.;
http://www.chroma.com; Omega Technology, Inc.; http://www.omegafilters.com). When constructing filter sets for multicolor imaging, the thickness and flatness of all of the dichroic and emission filters, as well as their orientation within the filter holder of the microscope, must be carefully controlled in order to minimize registration shifts of the different color images. Alternatively, the registration shifts can be measured with a sample that produces a pattern in each of the color channels. Correction factors can then be applied by software methods to register other acquired images (Galbraith and Farkas, 1993).
Single-Laser-Line Excitation with Multiple Fluorescence Signals: Laser Scanning Microscopes For multiple fluorescence signals, the balancing trick is to situate the emission band-pass
2.4.6 Supplement 15
Current Protocols in Cytometry
Fluorescence/Transmittance
488 nm exc.
450
500
550
600
650
Wavelength (nm) Figure 2.4.6 Fluorescence spectra (solid) of fluorescein (green fluorescence, 525-nm peak) and R-PE (orange fluorescence, 575-nm peak) superimposed on the transmission spectra (dotted) of the band-pass filters used to collect the fluorescence signals. Notice the spillover of the fluorescein fluorescence into the orange R-PE detection channel. The intensity of spillover is some fraction of the intensity in the main detection channel, so that the spillover can be subtracted by electronic or software methods. This process is called compensation (UNIT 1.14).
filters over fluorescence peaks of the dyes in such a way as to optimize light collection for each dye in its designated detection channel, while at the same time minimizing the detection of fluorescence signals from other dyes that spill over into that detection channel (Fig. 2.4.6). When it is not possible to exclude spillover signals with the filter sets, there are methods of removing spillover signals by electronic or software compensation, in which a portion of signal from one detector is subtracted from the signal from another detector. This is routinely done in two-color flow cytometry experiments where the fluorescein signal spills into the phycoerythrin channel (Fig. 2.4.6), but it can also be done with laser scanning microscopes.
Multi-Band-Pass Excitation with Multi-Band-Pass Emission A unique method for simultaneously allowing human (or color camera) visualization of two, three, or even four different color fluorophores became available with the development
of multi-band-pass filters (Pinkel et al., 1989). For example, a three-color multi-band-pass excitation filter contains three pass bands optimized for excitation at the absorption peaks of the three dyes for which the set has been designed (Fig. 2.4.7). Similarly, there is a tripleband-pass emission filter for collecting the three fluorescence signals. The three bands of the multi-band-pass emission filter lie between the excitation bands and are offset to longer wavelengths. The dichroic filter is also a tripleband-pass filter set at 45° and is designed to split each of the excitation-emission bands. These remarkable filters have become especially useful for imaging chromosome-specific probes and paints (Pinkel et al., 1989).
Multi-Band-Pass Emission with Excitation Changer In this hybrid system, the emission filter and dichroic filter are triple-band-pass filters, but the excitation filters are single-band-pass filters and are independently moved into position. With this setup, there is no possibility of regis-
Image Cytometry Instrumentation
2.4.7 Current Protocols in Cytometry
Supplement 15
Transmittance (%)
100
0 350
400
450
500
550
600
650
Wavelength Figure 2.4.7 Transmission spectrum of a triple-band-pass excitation filter (dotted) and a tripleband-pass emission filter (solid). Notice that the excitation and emission pass bands are located to excite and collect the fluorescence from three dyes (such as DAPI, fluorescein, and Texas Red) without scattered light reaching the eye or imaging camera.
tration shifts, which can occur when some epifluorescence filter sets are exchanged. A highresolution black-and-white imaging camera is used in this mode. Wavelength discrimination is not as good as with independent epifluorescence filter sets but is better than with dual multi-band-pass filter sets.
OTHER METHODS OF WAVELENGTH SELECTION While optical filters provide multi-wavelength capabilities, a number of more advanced methods additionally allow for a continuous scan of the wavelength domain, thus enabling spectral imaging microscopy (Farkas, 2000).
Liquid Crystal Tunable Filters
Optical Filtering Systems for Wavelength Selection in Light Microscopy
LCTFs consist of a number of liquid crystal layers, each of which passes a number of different frequencies; stacking them results in a single dominant transmission band, along with much smaller side-bands (and in some cases, unfortunately, additional major transmission
bands far from the desired spectral region). This concept is based on the Lyot birefringent interferometer. In addition to a computer, a small controller box is needed to drive an LCTF assembly. LCTFs can be switched from wavelength to wavelength in ∼50 msec, and are optically well behaved in that they do not seem to induce image distortion or shift. Each filter assembly can span ∼1 octave of wavelength (e.g., 400 to 800 nm) and their useful range can extend into the near IR. They can be introduced into either the illumination (excitation) or emission pathways, or both. Throughput is a problem, however, in that half the light corresponding to one polarization state is lost automatically, and peak transmission of the other half is limited. Overall throughput of a device approximating the bandwidth of the interference filter is ∼10%. In addition, focal-length variations can be introduced into the optical system as the LCTF is tuned from wavelength to wavelength. Finally, out-of-band rejection is not sufficient to prevent excitation light from leaking
2.4.8 Supplement 15
Current Protocols in Cytometry
into the emission channel without the use of a dichroic mirror or cross-polarization (Hoyt, 1996). Apart from these limitations, LCTFs are useful when multi-spectral microscopy is required, since their electronic tunability provides great flexibility in choice of center wavelengths for illumination and/or fluorescence excitation.
Acousto-Optic Tunable Filters In an AOTF, a radio frequency (RF) electronic signal is applied to an acoustic transducer attached to one face of the crystal. This generates a traveling acoustic wave in the crystal at the frequency of the applied RF signal. The resulting modulation of the crystal’s index of refraction acts as a sinusoidal phase grating for incoming light. Hence, when broad-band light is incident on an AOTF the narrow-band filtered light exiting the AOTF is angularly deflected away from the incident beam, as it would be by a conventional diffraction grating. The central wavelength of this filtered beam is determined by the acoustic frequency of the AOTF; this wavelength can be changed in less than 50 µsec to any other wavelength—several orders of magnitude faster than with other spectral filtering technologies. AOTFs have the ability to vary not only the wavelength, but also the bandwidth and the intensity of the transmitted light. Thus, experiments involving luminescence lifetimes or very rapid acquisition of multiple wavelengths are possible using this technology. Over the past few years, considerable progress has been made (Farkas et al., 1998) in overcoming the difficulties of using AOTFs for high-resolution imaging: (1) as with LCTFs, throughput is reduced since only one polarization state is typically used; (2) intrinsic image blur is present; and (3) out-of-band rejection is no greater than 10−2 to 10−3. Approaches and solutions to some of these problems have been described previously (Wachman et al., 1997), including use of both polarization beams to increase throughput, use of long-path-length imaging AOTF crystals to reduce blur, and transducer apodization (in the emission path) to improve out-of-band rejection. Nevertheless, additional optics (such as dichroics or rejection filters) are typically used to achieve adequate rejection for fluorescence measurements. Like LCTFs, AOTFs can be used for excitation and emission and give the benefit of continuous tunability. In addition, however, they have the capability for variable bandwidths,
which may be useful in multi-fluorophore applications. Finally, AOTFs are the technology of choice if sub-millisecond wavelength switching speed is required.
Fourier Transform Methods Fourier transform interference spectroscopy (FTIS) is a technique most widely used for generating spectral information in the infrared region. This method takes advantage of the principle that when light is allowed to interfere with itself at a number of optical path lengths, the resulting interferogram reflects its spectral constitution. Monochromatic light generates a pure cosine wave; multispectral light, which consists of a mixture of wavelengths, will generate interferograms that can be modeled as a sum of their respective cosine waves. An inverse Fourier transform of such an interferogram will regenerate the presence and intensities of all the contributing wavelengths. The classical Michelson interferometer, which uses two separate light paths to generate an interference pattern, is usable in the infrared region of the optical spectrum, because mechanical tolerances at these wavelengths are not relatively large. However, in the visible region, the required optical pathlength differences are much smaller, and vibration and mechanical imprecision become limiting. The Sagnac interferometer design, which directs the interfering light beams in opposite directions around a common path, overcomes many of these problems, and has been successfully adopted for use for visible light imaging spectroscopy by Applied Spectral Imaging (Garini et al., 1996). FTIS systems can be used to provide very rapid acquisition of emission spectra and have been used in experimental multicolor fluorescence flow cytometry and image cytometry (Buican, 1990; Cabib et al., 1996; Farkas et al., 1996). The optimal use of such a device occurs when several fluorophores can be simultaneously excited with a single laser wavelength but fluoresce at different wavelengths, as is true with fluorescein and phycoerythrin. In this case a long-pass filter can be used to remove laser scattering and the FTIS system can acquire the emission spectra. It is much more challenging to use an FTIS system when the fluorophores all absorb in different regions of the spectrum. In such a case, a double- or triple-dichroic filter cube (e.g., Fig. 2.4.7) is required in the optical path to ensure that each excitation band is excluded from the interferometer. Thus, instead of being able to collect light from a broad spectral region (and
Image Cytometry Instrumentation
2.4.9 Current Protocols in Cytometry
Supplement 15
thereby able to acquire detailed spectral shape information), the instrument is presented with light from several relatively narrow spectral regions inside of which it may be difficult to get enough data to distinguish spectrally similar dyes. Significant advances have been made in this area. It has been shown, for example, that six fluorophores with different spectra can be resolved by analyzing wavelength shifts that can be observed in the three pass-band regions of a triple-band-pass filter. A commercially available instrument (Applied Spectral Imaging) has been used with five fluorescent dyes (Garini et al., 1996) to resolve all chromosomes in human metaphase spreads that have been hybridized with chromosome paints (Schröck et al., 1996). Such instrumentation is also useful for multicolor immunofluorescence imaging (Farkas et al., 1998). There are advantages and drawbacks to Fourier transform techniques. As with tunable filters, and in contrast to fixed interference filters, wavelength ranges can be adjusted to match the spectral properties of the image. Compared to monochromator- or prism-based devices, FT spectroscopy enjoys a throughput advantage, as it does not require a narrow-slit aperture that would reduce the signal-to-noise ratio. Finally, FITS systems provide good spectral resolution (<10 nm) across a large spectral region. On the other hand, unlike a band-sequential device, an FT spectrometer is unable to increase its sensitivity (or imaging time) in different spectral regions to compensate for changes in collector sensitivity or light flux (Levenson et al., 1999). FTIS systems also require a long acquisition time (10 to 30 sec), a large working memory, and considerable processing time for converting an interferogram into a spectral data cube. Maximum image size is currently limited by these considerations to ∼500 × 500 pixels in the instruments currently mounted onto light microscopes. The cost of the commercially available units ($70,000 and up, exclusive of microscope) is also a consideration.
Use of diffraction grating systems for imaging is more complicated, but has also been successfully implemented (Descour et al., 1997). A computer-generated hologram disperser can be used to distribute various diffraction orders of the primary scene over a (much larger) 2-dimensional CCD array. The position and intensity of the resulting multiple image mosaic reflects the spectral content of the original scene. The spectral content of the image can then be reconstructed using an iterative multiplicative algebraic reconstruction algorithm. The advantage of this approach is that spatial and spectral information can be acquired with a single image exposure, without any necessity for scanning in either the x-y or wavelength domain. Thus, depending on signal intensity, the sampling can be accomplished at video rate or even faster. This advantage is “paid for” with much higher CCD real-estate requirements, since the CCD must capture not only the primary image but all its higher order diffraction images as well. With current CCDs, a 100 × 100-pixel primary image is all that can be achieved. In addition, the reconstruction of the spectral data cube from the dispersed images is a time-consuming, computationally intense procedure.
Prisms A form of microscope-based spectral imaging that resembles the techniques used in many so-called “push-broom” remote-sensing devices is the PARISS system (Lightform, Inc). In this emission-end device, fluorescent light passing through a prism is dispersed onto a CCD array. The imaged strip is stationary and the sample is moved beneath it. The resulting spectral information from each pixel along the strip being analyzed can be rapidly classified using neural-net software. The vertical and horizontal resolution of the unit is not as high as that of true imaging spectrometers, and a motorized stage is necessary to build up spectral information from a complete 2-D region. However, the unit is compact and relatively inexpensive.
Grating-based Systems
Optical Filtering Systems for Wavelength Selection in Light Microscopy
For multi-spectral filtering on the sample illumination part of a microscope, monochromator systems have many advantages. They are continuously tunable over a broad spectral range and can provide good light output (>5 mW). In addition, commercial systems exist which can switch wavelengths in <2 msec (Polychrome, T.I.L.L. Photonics).
CONCLUSION There appears to be an unending appetite for gathering more information per cell. More multicolor fluorescent labels and biological probes are becoming available, and as a result there is a need for continued advances in high performance filters for multicolor fluorescence microscopy.
2.4.10 Supplement 15
Current Protocols in Cytometry
Brelje, T.C., Wessendorf, M.W., and Sorenson, R. L. 1993. Multi-color laser scanning confocal immunofluorescence microscopy: Practical application and limitations. Methods Cell Biol. 38:98182.
Garini, Y., Katzir, N., Cabib, D., Buckwald, R.A., Soenksen, D.G., and Malik, Z. 1996. Spectral bio-imaging. In Fluorescence Imaging Spectroscopy and Microscopy, Chemical Analysis Series, Vol. 137 (X.F. Wang and B. Herman, eds.). John Wiley & Sons, New York.
Buican, T.N. 1990. Real-time Fourier transform spectroscopy of fluorescence imaging and flow cytometry. S.P.I.E. Proc. 1250:126-133.
Hoyt, C. 1996. Liquid crystal tunable filters clear the way for imaging multiprobe fluorescence. Biophotonics International, July/August. 49-51.
Cabib, D., Buchwald, R.A., Garini, Y., and Soenksen, D.G. 1996. Spatially resolved Fourier transform spectroscopy: A powerful tool for quantitative analytical microscopy. S.P.I.E. Proc. 2687:278-291.
Levenson, R.M., Balestreire, E.M., and Farkas, D.L. 1999. Spectral imaging: Prospects for pathology. In Applications of Optical Engineering to the Study of Cellular Pathology (E. Kohen, ed.), pp. 133-149.
DeBiasio, R., Bright, G., Ernst, L., Waggoner, A., and Taylor, D. 1987. Five parameter fluorescence imaging: Wound healing of Swiss 3T3 cells. J. Cell Biol. 105:1613-1622.
Pinkel, D., Gray, J., Segraves, R., Waldman, F., Trask, B., Yu, L.C., Eastman, D., and Dean, P. 1989. New technologies in cytometry. S.P.I.E. Proc. 1063:123-132.
Descour, M.R., Volin, C.E., Dereniak, E.L., Thome, K.J., Schumacher, A.B., Wilson, D.W., and Maker, P.D. 1997. Demonstration of a highspeed nonscanning imaging spectrometer. Optics Lett. 22:1271-1273.
Schröck, E., du Manoir, S., Veldman, T., Schoell, B., Wienberg, J., Ferguson-Smith, M.A., Ning, Y., Ledbetter, D.H., Bar-Am, I., Soenksen, D., Garini, Y., and Ried, T. 1996. Generation of a multicolor spectral karyotype of human chromosomes. Science 273:494-497.
LITERATURE CITED
Farkas, D.L. 2000. Cells and tissue structure and dynamics: A spectral imaging approach. In Methods in Cellular Imaging (A. Periasamy, ed.). Oxford University Press, New York. In press. Farkas, D.L., Ballou, B.T., Fisher, G.W., Fishman, D., Garini, Y., Niu, W., and Wachman, E.S. 1996. Microscopic and mesoscopic spectral bio-imaging. S.P.I.E. Proc. 2687:200-209. Farkas, D.L., Du, C., Fisher, G.W., Lau, C., Niu, W., Wachman, E.S., and Levenson, R.M. 1998. Noninvasive image acquisition and advanced processing in optical bioimaging. Comput. Med. Imaging Graphics 22:89-102. Galbraith, W. and Farkas, D.L. 1993. Remapping disparate images for coincidence. J. Microsc. 172:163-176. Galbraith, W., Ernst, L.A., Taylor, D.L., and Waggoner, A.S. 1989. Multiparameter fluorescence and selection of optimal filter sets: Mathematics and computer program. S.P.I.E. Proc. 1063:74-122.
Taylor, D.L., Burton, K., DeBiasio, R.L., Giuliano, K.A., Gough, A.H., Leonardo, T., Pollock, J.A., and Farkas, D.L. 1997. Automated light microscopy for the study of the brain: Cellular and molecular dynamics, development and tumorigenesis. Ann. N.Y. Acad. Sci. 820:208-228. Wachman, E.S., Niu, W., and Farkas, D.L. 1997. AOTF microscope for imaging with increased speed and spectral versatility. Biophys. J. 73:1215-1222.
Contributed by Alan S. Waggoner and Elliot S. Wachman Carnegie Mellon University Pittsburgh, Pennsylvania Daniel L. Farkas Carnegie Mellon University and University of Pittsburgh Pittsburgh, Pennsylvania
Image Cytometry Instrumentation
2.4.11 Current Protocols in Cytometry
Supplement 15
Digital Fluorescence Microscopy BACKGROUND INFORMATION The use of fluorescence microscopy in biomedical research has increased considerably in recent years. This trend is definitely related to the great potential of molecular research tools such as fluorescently labeled antibodies and nucleic acid probes. Fluorescent labels compare favorably with radioisotopes for many applications. With respect to cell analysis, fluorescence has replaced autoradiography almost completely. It provides good sensitivity and high multiplicity (defined as the number of fluorophores that can be detected simultaneously on the basis of their physicochemical properties). By applying specialized techniques (measurements of anisotropy or resonance energy transfer), one can obtain information on the microenvironment of the label or the spatial relationship of stained cellular molecules. Although visual examination by the human eye has sufficed for the majority of diagnostic tests, research applications often demand quantitative analysis of fluorescently stained cells. The obvious way to accomplish this is to scan the object of interest, measure the intensity point by point, and digitize the signals to generate an image. Such images can then be processed (corrected for errors or contrast enhanced) and analyzed to quantify morphological information as well as signal intensity.
UNIT 2.5
by the microscope, not the object itself, is scanned by a camera, for instance by a chargecoupled device (CCD) camera (Arndt-Jovin et al., 1985; Hiraoka et al., 1987).
Hardware Configuration The choice of hardware for fluorescence microscopy depends on the intensity and dynamics of the fluorescence. Dead and/or fixed cells may be visualized using slow-scan cameras that integrate the fluorescence over some time (seconds) to increase the sensitivity. Moving (e.g., live) cells require capture of images at video rate (25 to 30 images per second). Confocal microscopy is normally the method of choice when “out-of-focus” information has to be removed in real time or when three-dimensional images have to be constructed, although encouraging results are being obtained with so-called computational techniques. These methods (e.g., deconvolution) use mathematical procedures to remove the out-of-focus information from a consecutive stack of conventional images on the basis of the measured point-spread function of the microscope.
CHOICE OF CAMERA FOR IMAGE-PLANE SCANNERS This section describes the strategy for selecting a camera; for detailed technical discussion, see UNIT 2.3.
Scanning Methods There are two basic ways to scan a microscopic object: object-plane scanning and image-plane scanning. The earliest systems used were of the first type. In this approach mechanical stages were used to scan the object with steps of 0.25 or 0.5 µm at frequencies of typically 1000 Hz, and fluorescence intensity was measured at each point with a photomultiplier and digitized. With these systems it could take minutes to produce a picture. A later innovation was laser scanning systems, which were capable of scanning an object much faster by moving the laser beam instead of the object. Stage scanning became almost obsolete, despite its optical superiority (the constant axial illumination has fewer optical aberrations). Laser scanning is now frequently used in confocal imaging to remove image blur or to reconstruct three-dimensional images. The second type of scanning is image-plane scanning. In this approach the image produced Contributed by Hans J. Tanke Current Protocols in Cytometry (1997) 2.5.1-2.5.5 Copyright © 1997 by John Wiley & Sons, Inc.
Video-Rated Versus Slow-Scan Integrating Cameras Video-rated cameras are used to register changes in a fluorescence image, either relatively slow ones (e.g., when observing live, moving cells) or relatively fast ones (e.g., when observing changes in fluorescently stained intracellular molecules, such as intracellular calcium fluxes). This requires capturing multiple images per second to accurately describe the changes. High fluorescence intensity (enough photons) and/or a sensitive camera are required to produce images of sufficient quality. For many applications, image intensifiers are used to increase the photon flux (Shotton, 1993). In certain other situations, the images are in fact static but fast acquisition with cameras operating at high speed is indicated nevertheless. An example of such a case is when large numbers of cells have to be analyzed and the slide is moved automatically by a motorized
Image Cytometry Instrumentation
2.5.1
stage from one field to another. Bright-field images can easily be analyzed at video rate; for fluorescence, however, the scanning speed is limited by the number of photons emitted. Conditions for stationary imaging are different. The primary goal is to produce images with maximal “signal-to-noise” ratio (SNR). A “typical” CCD camera for analysis of cells and chromosomes stained by fluorescence in situ hybridization (FISH; UNIT 8.3) or immunofluorescence is capable of integrating the fluorescence emission for intervals varying from seconds to minutes. Such cameras are often cooled by Peltier elements or cryogenically to diminish the dark current during integration. Many of the cameras that are used for cell analysis are based on the Kodak KAF 1400 chip. This chip has 1348 × 1035 pixels with a pixel size of 6.8 × 6.8 µm and a full well capacity of approximately 32,000 electrons per pixel. The sensor has a Metachrome II coating to improve the sensitivity for light of shorter wavelengths (e.g., near-UV and blue; Aikens, 1990). For a complete description of cameras, see UNIT 2.3.
Color Versus Monochrome CCD Cameras
Digigtal Fluorescence Microscopy
A disadvantage of single-chip color CCD cameras is that their spatial resolution is lower in one direction than the other because the architecture of the chip requires that band-pass filters be used to detect the red, green, and blue light in parallel columns. Depending on the microscopic magnification, a small red dot in the microscopic image can be missed by the camera as it is imaged onto a column with a green or blue filter. Three-chip color CCD cameras capable of integrating the light have become commercially available. Particularly when prism optics are used for effective photon sorting and cooling is applied for reduction of noise, the performance of each of the single color channels is similar to that of monochrome cameras. HDTV (highdefinition television) technology used in broadcasting is expected to lead to further improvement in the performance of such cameras. The important advantage of capturing the red, green, and blue signals sequentially to create a color image using a monochrome camera and computerized filter wheels is the flexibility of this approach. This allows for the use of optical filters that specifically match the emission characteristics of the fluorophores studied. Furthermore, depending on the intensity of the signal, integration time can be optimized for each color separately.
Sampling an Image Ideally the recorded image should resemble the original image as closely as possible. This is determined by how the original image is sampled by the camera. Each camera has a number of pixels in the horizontal and vertical directions that register the image in one of the conjugated image planes of the microscope. The size of this image depends on the magnification. Thus, for a given magnification, the distance between the pixels of the camera corresponds to a certain distance in the object plane. If the pixels of the camera are square, the sampling density (that is, the number of pixels per distance in the object) is the same for the x and y axes. The optimal sampling density depends on the application.
MICROSCOPE HARDWARE Fluorescence Microscope Stand Because digital fluorescence microscopy (DFM) often aims at quantitative analysis of fluorescence images, requirements with respect to quality of the microscope hardware are high. Quantitative analysis will reveal errors such as chromatic or spheric aberration of lenses, uneven illumination of the object, unwanted excitation light (stray light), and autofluorescence of optics and filters in an often frustrating way. Because the light-collection efficiency of microscopes has increased considerably in the past decade, it is advisable to select modern stands (Tanke, 1989). Most commercial systems use infinity optics, which offers advantages for multimode quantitative microscopy. To improve sensitivity, well-stabilized bright light sources and well-corrected high-numerical-aperture (highNA) objectives with high transmission values in combination with suitable optical filters are recommended. In most modern microscopes simultaneous or sequential viewing of the image through the eyepieces and recording of the image with a CCD or other camera are achieved by means of a prism setup.
Objective Lens The objective lens plays a particularly important role in DFM. Its numerical aperture largely determines the detection sensitivity, achievable resolution, and depth of focus. In addition, chromatic aberrations may become more important in DFM than in conventional microscopy. Because light rays of shorter wavelength are refracted more than those of longer wavelength, each single, noncorrected
2.5.2 Current Protocols in Cytometry
lens has different focus points for blue, green, and red light. Therefore, microscope objective lenses are corrected and made “achromatic.” Modern fluorescence microscope objectives generally do not cause large aberrations, although their performances should be checked, especially in cases where colors over a wide spectral range (blue to infrared) are being analyzed. It is important to realize that whereas the human eye-brain combination has an adaptive ability to interpret images with information that is “out-of-focus” because of chromatic errors, this is obviously not the case for a CCD array. The collector lens in the lamp housing must also be chromatically corrected for DFM. If a chromatically imperfect collector lens is used in combination with multi-band-pass excitation filters (see UNIT 2.4), the focus levels of the various wavelength bands will be different, leading to different Köhler-type adjustment of the light. Consequently, the intensity of the excitation light will become dependent on the wavelength and the position in the microscopic field (Florijn et al., 1996). This is important in all cases where ratios of fluorescence emission are measured, such as multicolor FISH using ratio-labeled probes (UNIT 8.3; Nederlof et al., 1990; Dauwerse et al., 1992; Ried et al., 1992) or comparative genomic hybridization (Kallioniemi et al., 1992). The lens should be checked for residual chromatic errors using appropriate standards, such as beads containing defined amounts of two different fluorophores (see UNIT 1.3 for discussion on calibrating bead fluorescence).
Stage Control and Filter Wheels The microscope stage is moved along the x, y, and z (for focus) axes by computer-controlled stepping motors. Parameters to be considered are speed (important when large numbers of cells have to be analyzed automatically), range (when using a stage that carries multiple slides), and accuracy of relocation of cells (typically a few micrometers, depending on the magnification). Optical filters may be changed automatically. There are two ways to do this: (1) the entire cube containing all the filters may be moved; or (2) filters mounted on wheels positioned at both the excitation and emission sides may be operated independently. Switching filters may cause displacement of the image (“pixel shifts”) because the optical ray path is changed. Three strategies are possible to avoid or correct such shifts. One method is careful mechanical adjustment and optical
compensation. A second is mathematical adjustment of the digitized image: microscopy is performed using microspheres that can be excited using various filter combinations, then the shift is calculated from sequentially recorded images and used to correct all subsequent images. A third way to reduce image shift is to use fixed triple-band-pass dichroic mirrors and barrier filters in combination with changeable excitation filters. Triple-band-pass filters reflect and/or transmit multiple wavelength bands of light (for details, see UNIT 2.4).
Computer and Peripherals The main functions of the computer are camera control, operation of the automated microscope parts, and image analysis. Because frame grabbers (image acquisition units) are available for most types of computers (e.g., Macintosh, PC, and workstations), the choice of computer is determined by the availability of software for specific applications and by personal preference. Apart from standard peripherals such as storage devices and a (color) printer, a high-quality 21-in. color display is highly recommended.
ACQUISITION AND PROCESSING OF IMAGES Sources of Noise Any image obtained by a camera suffers from noise. Noise needs to be avoided or else its effects corrected for. Correction is relatively simple for additive systematic noise but virtually impossible for random noise in an image. Thus, all measures should be taken to avoid such noise as much as possible. Cooling cameras to suppress thermal noise is a good example (for a detailed discussion, see UNIT 2.3).
Image Acquisition Errors Aligning the microscope illumination system according to Köhler should theoretically result in the same amount of excitation light for each point of the object field; in addition, the efficacy of the light collection should be equal. In practice this is not always the case, although careful adjustments may reduce the variation to a few percent. To eliminate these deterministic variations in an image taken by a CCD camera, a flat-field correction is performed. That is, a background image is collected next to the object being processed. The average intensity of this image is a function of the local illumination intensity; to minimize the influence of small local bright
Image Cytometry Instrumentation
2.5.3 Current Protocols in Cytometry
or dark spots, the background image should not be too small. Also, a dark-current image is taken; this is defined as an image recorded with the same integration time but with the camera shutter closed. For each pixel, the corrected image (Icorrected) is obtained by applying the following equation to the data:
Icorrected =
Digigtal Fluorescence Microscopy
Iobject − Idark I backgr − Idark
The quality of the image can be expressed in the form of a number called the signal-tonoise ratio (SNR), determined as described in UNIT 2.3. Recommendations for the minimum SNR needed are difficult to give, as it strongly depends on the application. For example, images with low SNR may be good enough for detecting certain structures based on intensity thresholding (or other segmentation techniques) but unacceptable for accurately quantitating local light. A last important type of error that effects the acquisition of images for quantitative analysis is photodecomposition (fading). Fading can be operationally defined as reduction of the quantum efficiency as a result of excitation of the fluorophore. It is a complex phenomenon that depends on the type of fluorophore, its concentration and molecular environment, the intensity and nature of the excitation light, and the oxygen concentration of the embedding medium (Song et al., 1995). Fading can be reduced by including antifading agents in the embedding medium. The working mechanism of these reagents is often based on the scavenging of oxygen radicals that are produced during excitation and may react with the fluorophore. Using such reagents, fading of most fluorophores during the actual measurement is generally <10% for exposure times up to 30 sec, which are sufficient for many applications (Florijn et al., 1995). Exposure of objects to high-intensity excitation light prior to measurement should always be avoided. Manual searching and focusing of the object is preferably done using low excitation intensity (with neutral-density filters positioned at the excitation side) or, when feasible, using other means of imaging (e.g., phase contrast or Nomarski). The intensity of light needed for these latter types of illumination is so low that fading of the fluorophore is negligible. Note that in practice the effect of fading is minimized or partly averaged out when comparable images are taken with the same integration time. Using a cooled camera improves the SNR, which allows shorter inte-
gration times and thus reduces fading. Fading often determines the practical upper limit of the integration time: if the fluorophore does not emit anymore, continued integration is meaningless! If that is the case, though, one may wish to reconsider the merits of cooling to reduce the dark current for each application that requires only relatively short integration times (Vrolijk et al., 1994).
Troubleshooting and Maintenance The hardware described is relatively complex and the instrumentation should be tested at regular intervals. One of the most critical parts of the microscope is the illumination system. It is necessary to verify that the illumination is Köhler-type, not just when replacing the lamp bulb but at each measuring session; automated adjustment of the illumination system has recently been introduced in commercial systems. Optical filters and objective lenses need regular replacement: the high excitation intensity used in modern microscopy causes even high-quality interference filters to change in performance after some years, and objective lenses may develop autofluorescence upon prolonged use. A preventive measure that can help minimize this latter problem is closing the shutter in the excitation pathway between actual imaging sessions to prevent unnecessary exposure of the optics. Cameras can be checked for performance as described in UNIT 2.3. For cooled cameras, flaws in the vacuum system are occasionally observed after some years, and this can cause moisture (or ice crystals, when the system is cooled to −45°C) to condense on the chip. Performance can be restored by the manufacturer. Shielding of commercial slow-scan CCD cameras against electromagnetic radiation has improved over the past years. Despite that, severe damage to CCD sensors may occur as a result of voltage surges when high-pressure lamps are switched on. This is particularly a problem if the lamp and camera are on the same voltage line and if the lamp power cord is long and near the camera. Consequently, modern microscopes have high-pressure lamps with built-in starters. Nevertheless, it is still good procedure to switch on the lamp first and then the camera. The performance of the entire system can be checked using suitable standards (see UNIT 1.3). Microspheres with calibrated amounts of fluorophores can be used to determine the detection sensitivity and the accuracy of the meas-
2.5.4 Current Protocols in Cytometry
urement, and the results may then be compared with those obtained by flow cytometry, as an independent method.
LITERATURE CITED Aikens, R.S. 1990. CCD cameras for video microscopy. In Optical Microscopy for Biology (B. Herman and K. Jacobsen, eds.) pp. 207-218. Wiley-Liss, New York. Arndt-Jovin, D.J., Robert-Nicoud, M., Kaufman, S.J., and Jovin, J.M. 1985. Fluorescence digital imaging microscopy (DIM) in cell biology. Science 230:247-256. Dauwerse, J.G., Wiegant, J., Raap, A.K., Breuning, M.H., and Van Ommen, G.J.B. 1992. Multiple colors by fluorescence in situ hybridization using ratio-labelled DNA probes create a molecular karyotype. Hum. Mol. Genet. 1:593-598. Florijn, R.J., Slats, J., Tanke, H.J., and Raap, A.K. 1995. Analysis of antifading reagents for fluorescence microscopy. Cytometry 19:177-182 Florijn, R.J., Bonnet, J., Vrolijk, J., Raap, A.K., and Tanke, H.J. 1996. The effect of chromatic errors in microscopy on the visualization of multi-colour fluorescence in situ hybridization. Cytometry 23:8-14. Hiraoka, Y., Sedat, J.W., and Agard, D.A. 1987. The use of a charge-coupled device for quantitative optical microscopy of biological structures. Science 238:36-41. Kallioniemi, A., Kallioniemi, O.-P., Sudar, D., Rutovitz, D., Gray, J.W., Waldman, F., and
Pinkel, D. 1992. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 258:818-821. Nederlof, P.M., van der Flier, S., Wiegant, J., Raap, A.K., Tanke, H.J., Ploem, J.S., and Van der Ploeg, M. 1990. Multiple fluorescence in situ hybridization. Cytometry 11:126-131. Ried, T., Baldini, A., Rand, T., and Ward, D.C. 1992. Simultaneous visualization of seven different DNA probes by in situ hybridization using combinatorial fluorescence and digital imaging microscopy. Proc. Natl. Acad. Sci. U.S.A. 89:13881392. Shotton, D. 1993. Electronic Light Microscopy. Wiley-Liss, New York. Song, L., Hennink, E.J., Young, I.T., and Tanke, H.J. 1995. Photobleaching kinetics of fluorescein in quantitative fluorescence microscopy. Biophys. J. 68:2588-2600. Tanke, H.J. 1989. Does light microscopy have a future? J. Microsc. 155:405-418. Vrolijk, J., Sloos, W.C.R., Verwoerd, N.P., and Tanke, H.J. 1994. The applicability of a noncooled video-rated CCD camera for the detection of fluorescence in situ hybridization signals. Cytometry 15:2-11.
Contributed by Hans J. Tanke Leiden University Leiden, The Netherlands
Image Cytometry Instrumentation
2.5.5 Current Protocols in Cytometry
Calibration: Sampling Density and Spatial Resolution
UNIT 2.6
This unit presents a discussion of procedures for measuring the sampling density and spatial resolution of a quantitative microscope system. These two independent quantities are fundamental characteristics of a system that must be known in order to properly process and interpret digitized microscope images and measurements extracted from such images. Unless otherwise specified, it will be assumed in this discussion that two-dimensional (as opposed to three-dimensional) microscopy is under consideration—that is, images are being acquired as “simple” two-dimensional projections of what are inherently three-dimensional objects. Neither confocal microscopy nor three-dimensional microscopy based upon software reconstruction techniques are addressed. Throughout the discussion, the x direction will be referred to as the horizontal direction, the y direction as the vertical direction, and, if needed, the z direction—the third dimension—as the axial direction. Sampling density refers to the physical distance between pixels in the digitized microscope image. In general, it is necessary to measure the sampling distances ∆x and ∆y for each of the two independent spatial dimensions, x and y, in the imaging system. An example of the spatial sampling of a continuous image to produce a digitized image consisting of pixels is shown in Figure 2.6.1. Depending upon the type of image digitizing system used, sampling may take the form of either square sampling (where ∆x = ∆y) or rectangular sampling (where ∆x ≠ ∆y). Square sampling is preferable to rectangular sampling because it avoids the situation illustrated in Figure 2.6.2. Spatial resolution refers to the ability to resolve or distinguish two small or point-like objects separated by a given distance. The precise quantitative definition of spatial resolution is a complex issue whose origin dates back to the beginnings of scientific optical instrumentation, the first microscopes and telescopes. In modern instruments, where fundamental limits of optics and photon detection are readily achieved, the proper frame of reference for discussing spatial resolution is the point spread function (PSF) of an optical system and its Fourier transform, the optical transfer function (OTF). Why these two descriptions are so central to the issue of spatial resolution will not be discussed in detail here; for a detailed explanation, see O’Neill (1963), Born and Wolf (1980), Young (1989, 1996), and Castleman (1996).
Figure 2.6.1 An image of a fluorescently stained cell that has been digitized into individual pixels. The distance between two horizontal pixels in a row is ∆x and the distance between two vertical pixels in a column is ∆y.
Image Cytometry Instrumentation Contributed by Ian T. Young Current Protocols in Cytometry (1998) 2.6.1–2.6.15 Copyright © 1998 by John Wiley & Sons, Inc.
2.6.1 Supplement 5
analog object
digitizing raster
digital image
1:1 sampling
+
=
4:3 sampling
+
=
Figure 2.6.2 Square versus rectangular pixels. Unless corrective techniques are applied in software, a circular object (such as an erythrocyte) will appear as an elliptical object in the digitized representation when rectangular pixels are used.
In essence, the PSF describes the physical distribution of light produced by a very compact, very bright source of light as imaged through an optical system. In astronomy this could be the image of a star as seen through a telescope; in microscopy it could be the image of a fluorescence-labeled gene probe whose physical dimensions are much smaller than a wavelength of light. To predict how any object will look under conventional bright-field, fluorescence, or even confocal microscopy, optical theory states that an observed image is related to the original object by the convolution of the spatial distribution of the object intensities with the spatial distribution of the PSF intensities. A “good” PSF, therefore, would be one for which it is difficult to distinguish between the spatial distribution of the image intensities and the object intensities. An alternative description based on the OTF states that any “optical object” can be described as a weighted sum of sinusoidal distributions of light, where the sinusoids have varying spatial frequencies. The way an object “propagates” through an optical system depends on the OTF of that system. Thus, by knowing the OTF one can also predict how the image of an object will look. Formally, the spatial frequency amplitudes of the image will be given by the spatial frequency amplitudes of the object multiplied by the OTF, which is, itself, a function of the spatial frequencies. Going a step further, the PSF of an in-focus, circularly symmetric, diffraction-limited lens for incoherent, quasi-monochromatic illumination is given by:
PSF(r ) = 2
J1 (ar ) r
2
Equation 2.6.1
Calibration: Sampling Density and Spatial Resolution
where a = 2πNA/λ and J1(x) is a Bessel function of the first kind. The assumption that the PSF is circularly symmetric is reflected in the fact that it is simply a function of the radial distance, r. The assumptions that the lens is in focus, circularly symmetric, and diffraction limited lead to the Bessel function and the dependence on the numerical aperture, NA, of the lens. Finally, the assumption that the light is near monochromatic is the basis of the dependence on wavelength, λ. This PSF is pictured in Figure 2.6.3.
2.6.2 Supplement 5
Current Protocols in Cytometry
A
B Image intensity
1 0.8 0.6 0.4 0.2 r 0 0
2
4 6 Radial position
8
10
Figure 2.6.3 PSF(r) for a diffraction-limited lens with a = 1. The vertical axis is image intensity and the horizontal axis is radial position. (A) The two-dimensional PSF showing the well-known Airy structure. (B) Because the PSF depends solely on the radial distance r, it can be presented as a one-dimensional “slice” of the two-dimensional function.
The OTF, which is the two-dimensional Fourier transform of the PSF, is given by OTF(ωx,ωy) = F {PSF(x,y)} = F {PSF(r)} = OTF(ωr), where F (x) is the two-dimensional Fourier transform operation. The OTF is thus as follows:
%K 2 ω ω cos OTF(ω ) = & π 2a − 2a ! K' 0 −1
r
r
r
1−
ω 2a r
2
"# #$
ω r ≤ 2a ω r > 2a
Equation 2.6.2
The OTF is shown in Figure 2.6.4. Because the OTF is zero for radial frequencies greater than 2a, it is said to be band-limited to frequency 2a, with a highest frequency fc given by ωc /2π = fc = 2NA/λ. Because the PSF and the OTF are completely determined by the value of a, 2π NA/λ, it is possible, in principle, to characterize both of them by determining the NA and λ. But because one can never be sure that the optical system is in critical focus and that the lens meets the standard of being diffraction limited, in practice it is advisable to measure either the PSF or the OTF.
Image Cytometry Instrumentation
2.6.3 Current Protocols in Cytometry
Supplement 4
A
B 1
OTF
0.8 0.6 0.4 0.2 ω
0 0
0.5
1
1.5
2
Radial spatial frequency
Figure 2.6.4 OTF(ωr) for a diffraction-limited lens with a = 1. The vertical axis is the OTF and the horizontal axis is the radial spatial frequency. (A) Two-dimensional OTF showing the band-limited structure. (B) Because the OTF depends solely on the radial frequency, ωr , it can be presented as a one-dimensional “slice” of the two-dimensional function.
RELATING SAMPLING DENSITY AND SPATIAL RESOLUTION There is a well-known relation between desired sampling density and spatial resolution. If an image is band-limited, then according to the Nyquist sampling theorem (Oppenheim et al., 1983; Castleman, 1996; Young, 1996), the sampling frequency should be chosen to be more than twice the highest frequency in the image. The highest frequency that is associated with a microscope lens is obtained from Equation 2.6.2 and is given by ωc = 2a = 4πNA/λ. The sampling frequency is therefore given by: ω s > 4a =
8π • NA λ
Equation 2.6.3
This translates to a sampling density of: ∆x = ∆y <
0.25 • λ NA
Equation 2.6.4
Calibration: Sampling Density and Spatial Resolution
Note the resemblance of this equation to that for the well-known but rather arbitrary Rayleigh criterion, δ = 0.61λ/NA. The Rayleigh criterion, developed in the context of astronomy, was used to describe the minimum distance between two PSFs that could lead to the visual detection of two point-like objects instead of one larger object (Young, 1989). According to the equation above, for green light with λ = 500 nm and an NA = 0.5, the maximum sampling density that should be chosen is 250 nm/pixel. Determining whether
2.6.4 Supplement 4
Current Protocols in Cytometry
this condition is satisfied is one of the key reasons for using procedures to determine the actual sampling density. DETERMINING SAMPLING DENSITY There are two standard approaches to determining the sampling density. The first involves computing it based on knowledge of the pixel size and the optics. The second uses a calibration device—a stage micrometer—to measure the density. Computing the Sampling Density For a digitizing system with (pixel) center-to-center distances in the horizontal and vertical directions given by Dx and Dy, respectively, and a total optical magnification of M, the sampling density is:
∆ x = Dx / M and ∆ y = Dy / M Equation 2.6.5
Example 1
A specimen is observed through a 100× objective lens and a 1.25× “tube factor” using a camera based on a charge-coupled device (CCD) chip with a spacing of 6.8 µm in the horizontal and vertical directions. Thus: M = 100 • 1.25 = 125 Dx = Dy = 6.8 µm
(K → ∆ )K *
x
= ∆ y = 54.4 nm / pixel
Example 2
A specimen is observed through a 25× objective lens and a 2.25× “zoom lens” using a video camera based on a CCD chip with a center-to-center spacing of 14.5 µm in both the horizontal and vertical directions. The frame grabber that converts the video signal to the digitized format has equivalent distances of 19.33 µm in the horizontal direction and 14.5 µm in the vertical direction. (The term “equivalent distance” refers to distances measured in the plane of the specimen.) Thus: M = 25 • 2.25 = 56.25 Dx = 19.33 µm Dx = 14.5 µm
(K K) → ∆ KK ∆ *
x y
= 344 nm/pixel = 258 nm/pixel
This increase in the horizontal distance relative to the vertical distance occurs because the number of samples per line, the interval between samples, and the position in a horizontal video line at which the sampling starts are left to the discretion of the frame grabber’s manufacturer. Note that nonsquare sampling by the frame grabber can introduce nonsquare pixels. It is not enough to know that a camera has “square pixels” at the sensor level; the requirement of working with square pixels requires that the total system be considered. Further, because the output of a video camera is an analog video signal, the camera must have a certain smoothing or interpolation capability to “fill in” the brightness values between the discrete CCD pixels. If this smoothing is not done well, it can lead to artifacts in the digitized image produced by the frame grabber, artifacts that appear on the computer display as sinusoidal waves or Moiré patterns. In any event, one cannot assume that each sample in a digitized image corresponds to one pixel in the CCD camera. In order to determine if there are any “hidden sources” that could lead to nonsquare digitization, it is necessary to use calibration standards.
Image Cytometry Instrumentation
2.6.5 Current Protocols in Cytometry
Supplement 4
A
B
Figure 2.6.5 Stage micrometers. The horizontal lines go through a series of bars that have center-to-center spacings of 10 µm (A) and 2 µm (B).
Measuring the Sampling Density Stage micrometers are available from a variety of sources, including the major microscope manufacturers and other suppliers (e.g., Opto-Line Associates; see SUPPLIERS APPENDIX). The central feature of a stage micrometer that makes it appropriate for determining the sampling density is that it is marked with a periodic series of stripes at known distance intervals. Two such stage micrometers are shown in Figure 2.6.5. The line spacing can be measured using digital image processing techniques. The general idea is to identify those places on the bars where there is a transition either from black to white or from white to black. Then, by taking the distance between the transition positions over several periods of the bar pattern, it is possible to determine the average distance in pixels between the bars. This can then be converted to ∆x and ∆y. Example 1 The horizontal derivative is computed for the image in Figure 2.6.5A by application of a simple one-dimensional, “clipped derivative” filter: value = bin [ x + 1, y] − bin [ x − 1, y] bout [ x, y] =
%Kvalue if value > 0 &K 0 if value < 0 '
where b[x,y] is the brightness of the image at coordinate position [x,y], “in” refers to the input, or starting image data and “out” is the output data. This leads to a brightness representation for one line in Figure 2.6.5A as shown in Figure 2.6.6. The three maxima of the clipped derivative occur at the horizontal positions x = 23, 115, and 211. The distance from the first maximum to the last is known from the stage micrometer specification to be 2 × 10 µm = 20 µm. In pixels this is 211 − 23 = 188 pixels. The sampling frequency in the horizontal direction is 188 pixels/20 µm = 9.4 pixels/µm. The sampling density in the horizontal direction, ∆x, is 106 nm/pixel. Example 2 The horizontal derivative is computed for the image in Figure 2.6.5B by application of the same clipped derivative filter used in Example 1. This leads to a brightness representation for one line in Figure 2.6.5B as shown in Figure 2.6.7.
Calibration: Sampling Density and Spatial Resolution
The first maximum of the clipped derivative occurs at the horizontal position x = 109 and the last maximum at x = 191. The distance from the first maximum to the last is known from the stage micrometer specification to be 14 × 2 µm = 28 µm. In pixels this is 191 − 109 = 82 pixels. The sampling frequency in the horizontal direction is 82 pixels/28 µm = 2.9 pixels/µm. The sampling density in the horizontal direction, ∆x, is 341 nm/pixel.
2.6.6 Supplement 4
Current Protocols in Cytometry
250
Brightness
200 brightness data
150
derivative data
100 50 0 0
50
100
150
200
250
Horizontal position Figure 2.6.6 Brightness along a horizontal line in Figure 2.6.5A. The original data (thinner line) contain parts of three black bars with a center-to-center spacing of 10 µm. The clipped derivative data (thicker line) show peaks at the three transitions from dark to light.
Brightness
180 160
brightness data
140 120 100 80 60 40 20 0
derivative data
100
120
140
160
180
200
Horizontal position Figure 2.6.7 Brightness along a horizontal line in Figure 2.6.5B. The original data (thinner line) contain white bars with a center-to-center spacing of 2 µm. The clipped derivative data (thicker line) show peaks at the transitions from dark to light.
In both examples similar measurements must be made in the vertical direction to determine ∆y. This can be done by rotating either the calibration slide or the camera system through 90°. DETERMINING THE SPATIAL RESOLUTION Measuring the PSF or the OTF of a complete microscope imaging system is not easy. Based on the equations for OTF and sampling frequency, one might imagine that in order to completely specify the system it is sufficient to determine the NA and the relevant wavelength. Issues such as the proper (Nyquist) sampling density and the focusing of the image can, however, lead to a very different result when the OTF is actually measured, as illustrated in Figure 2.6.8. There are several standard ways to assess the OTF: using a step edge, a bar chart, or microbeads.
Image Cytometry Instrumentation
2.6.7 Current Protocols in Cytometry
Supplement 5
Step Edge Image Test charts, such as those shown in Figure 2.6.5B and Figure 2.6.9A, usually include an area where the intensity goes from an extended region of black to an extended region of white—a step edge denoted by u[x,y]. The resulting image is a convolution of the PSF and the step edge to produce the edge response: E[ x, y] = PSF[ x , y ] ⊗ u[ x, y ] Equation 2.6.6
Before the PSF (or OTF) can be calculated, the resulting image must first be corrected for any shading in the image that may be caused by nonuniform illumination, nonuniform camera sensitivity, and dark current. The correction procedure involves recording a “black” image with the camera shutter (or equivalent) closed and a “white” image with the camera focused on a blank (empty) field. The resulting correction, which starts with three digitized images, yields a corrected floating point image whose values are between 0.0 and 1.0: image corr [ x, y] =
imageorig [ x, y] − imageBLACK [ x, y] image WHITE [ x, y] − image BLACK [ x, y] Equation 2.6.7
1.00 Figure 2.6.8 As one moves away from critical focus, the OTF “sags.” These measurements were made with an oil-immersion objective with an NA of 1.4 and λ = 400 nm. According to Equation 2.6.2, maximum spatial frequency passed by the lens should be 7.0 cycles/µm.
0.00
0
10
B
C 8 µm
A
5 Frequency (cycles/µm )
8 µm
OTF(ω /2 ,z)
increasing z away from focus position
32
1 µm
Calibration: Sampling Density and Spatial Resolution
1 µm
Figure 2.6.9 (A) A 32-pixel-wide region is selected as the dark-to-light step edge response, E[x,y], from a shading-corrected image. (B) The region is enlarged in the horizontal direction by a factor of 8 using spline interpolation. (C) After computation of the derivative, the resulting set of impulse responses are aligned to produce the line response L[x,y].
2.6.8 Supplement 5
Current Protocols in Cytometry
Amplitude
A
B 500
600 500 400 300 200 100 0 –100
400 300 200 100
0 –100 0
0.6
1.2
1.8
x position (µm)
2.4
3.0
0
0.6
1.2
1.8
2.4
3.0
x position (µm)
Figure 2.6.10 The line response. (A) One row of data, L[x,y = yo], from Figure 2.6.9C before averaging. (B) The average of N rows of data produces L[x] where the SNR is improved by √ N .
An example of this type of corrected image is shown in Figure 2.6.9A. A 32-pixel-wide region centered over an edge is used for further processing. The image is then interpolated in the horizontal direction to a sample spacing eight times finer than the original, using a spline interpolation routine (Press et al., 1988). The result of such processing is shown in Figure 2.6.9B. This interpolation is done to ensure that the subsequent filtering operation will yield a satisfactory approximation to a derivative operation. A horizontal derivative is necessary to turn the edge response, E[x,y], in Figure 2.6.9B into a line response, L[x,y]. A one-dimensional derivative-of-Gaussian kernel with coefficient σ = 1.5 is convolved with the interpolated image along each horizontal line (Young and Van Vliet, 1995). Each horizontal line yields one (noisy) version of the line response. If the edges were perfectly aligned and free of defects, an ensemble average in the vertical direction would improve the signal-to-noise ratio (SNR) by the square root of the number of averaged lines. Unfortunately, due to the processes used to manufacture the step-edge test pattern, the alignment of the edge response (and thus the line response) is not perfect. This is clear in Figure 2.6.9B. To correct for this, each line is shifted so that the edges are aligned with one designated central pixel. Once again, noise produces uncertainty in the location of the central pixel. By applying a smoothing filter to the line and detecting the position of the maximum value, the uncertainty of the edge position is reduced (Canny, 1986). The result of this alignment procedure on the line image is shown in Figure 2.6.9C. A single row of data from Figure 2.6.9C is shown in Figure 2.6.10A. When the aligned data are averaged in the vertical direction, the result L[x], shown in Figure 2.6.10B, is produced. The noise reduction is evident. The Fourier transform of L[x] can now be computed (Press et al., 1988) to yield an estimate of the OTF in the horizontal direction. The result for the data shown in Figure 2.6.10B is given in Figure 2.6.11A. The fact that the OTF is not equal to 1.0 at a spatial frequency, ω, of 0 can be explained by a loss of light between the input illumination and the final measurement system. In general, however, it is very difficult to determine the exact amount of light input into the optical system and thus it is common practice to normalize the OTF such that its value is 1.0 at ω = 0. The example shown in Figure 2.6.11B represents OTF measurements taken in both the horizontal and vertical directions. The camera was intentionally used in frame integration mode, which meant that only one of the two fields per frame was used. (In ordinary video
Image Cytometry Instrumentation
2.6.9 Current Protocols in Cytometry
Supplement 4
A
1.00
OTF( /2ω , z)
0.80 0.60
0.40 0.20 0.00 0.0
B
1.0
2.0
3.0
4.0
5.0
1.00 total OTF (vertical)
OTF(ω /2 )
0.75
optical OTF total OTF (horizontal)
0.50 0.25 0.00 0.0
1.0
2.0
3.0
4.0
5.0
Frequency (cycles/µm) Figure 2.6.11 The optical transfer function. (A) The measured |OTF(ωx /2π)| in the horizontal direction for a complete system: illumination, microscope, (scientific, slow-scan) CCD camera, and digitizer. (B) The measured OTFs for a complete system, illumination, microscope, video CCD camera, and digitizer, in both the horizontal and vertical directions. The lens OTF is shown for comparison.
cameras a single image frame consists of two interlaced fields representing the odd-numbered lines in one field and the even-numbered lines in the other.) Using only one of the two fields means that the image sampling was reduced by a factor of two in the vertical direction. All three of the OTFs shown in Figure 2.6.11B have been normalized to the value 1.0 at ω = 0. If this had not been done, the vertical and horizontal OTFs would have been below the optical OTF for all spatial frequencies. Bar Chart Image The test chart shown in Figure 2.6.12 illustrates another mechanism that can be used to estimate the OTF. As the distance between the bars in an image becomes smaller, the local contrast decreases. The intensity along lines drawn through the bar patterns a, b, and c in Figure 2.6.12A shows decreasing contrast in Figure 2.6.12B as the spacing of the bars becomes smaller. The contrast, referred to as the contrast modulation transfer function, can be measured as: C( p) = Calibration: Sampling Density and Spatial Resolution
Imax ( p) − Imin ( p) Imax ( p) + Imin ( p)
Equation 2.6.8
2.6.10 Supplement 4
Current Protocols in Cytometry
A
B c Brightness
c
b
b
a a 20
30
40
50 60 70 Relative x position
80
90
Figure 2.6.12 Contrast change as a function of spatial frequency. (A) Fluorescence image taken with a bar chart using a complete system—illumination, microscope, video CCD camera, and digitizer. (B) The intensity along a line through each of the three indicated bar patterns. Bar a (500 line pairs/mm), p = 2 µm; bar b (315 line pairs/mm), p = 3.17 µm; bar c (251 line pairs/mm), p = 3.99 µm.
where p is the spacing between the bars and Imax(p) and Imin(p) are the (average) maximum and minimum intensities in a bar pattern of spacing p. Experimental determination of C(p) is relatively straightforward. Bar chart test patterns on a microscope slide are available from a number of commercial firms (e.g., Opto-Line Associates; see SUPPLIERS APPENDIX). By using transillumination or reflection illumination, or by placing the test pattern on top of a fluorescent standard such as uranyl glass, an image can be scanned and digitized. This image should be corrected for shading using the procedure described in Equation 2.6.7. Digital measurement of the contrast in the recorded image, at various positions corresponding to various bar spacings (Fig. 2.6.12), will then give a characterization of the contrast modulation transfer function for the complete system. Along each bar pattern, such as a, b, c in Figure 2.6.12B, the average difference and sum of the brightness intensities in the output image are measured at adjacent brightness maxima and minima to give [Imax(p) − Imin(p)] and [Imax(p) + Imin(p)]. The collection of measurements made at various values of p gives C(p), as shown in Figure 2.6.13. Again, it is important to remember that such measurements must be made in a variety of directions. Although the lenses in the complete system may be circularly symmetric, the scanning and digitizing system may not be. Using a known mathematical relationship between the contrast, as described in Equation 2.6.8, and the OTF(ω), it is possible to compute OTF(ω) from C(p) (Limansky, 1968; Young, 1989). The important condition is that the original bar chart test pattern has a 50% duty cycle at each value of p. That is, for any given periodic bar pattern, the width of the black bars should equal the width of the white bars. The relationship between C(p) and OTF(ω) is given by: C( P) =
2πn sin(nπ / 2) p n
∑ OTF
n odd
Equation 2.6.9
Measurements of C(p) should be made for the bar spacings {p = P, p = P/3, p = P/5, p = P/7,…}, where P is the fundamental period—that is, the longest bar spacing in the test pattern. If the bar pattern does not provide these spacings directly, they can be interpolated
Image Cytometry Instrumentation
2.6.11 Current Protocols in Cytometry
Supplement 4
1.0 C(s) OTF(ω)
Contrast, C OTF
0.8
0.6
0.4
0.2
0.0 0
100
200
300
400
500
Line pairs/mm, s Cycles/mm, ω
Figure 2.6.13 As the number of line-pairs/mm increases the contrast, C(s), decreases. The OTF(ω), derived from C(s), shows a similar behavior. A 12 × 12 matrix equation, as exemplified in the text below, was used to generate these data. Note that the period spacing p is inversely proportional to s.
from the values that are available (Press et al., 1988; see Fig. 2.6.13). The relationship between C(p) and OTF(ω), Equation 2.6.9, can be written in matrix form as:
C( p) "# 1 C( p / 3) # # C( p / 5) # # C( p / 7) # = ## • ## • ## ! • $ !
Calibration: Sampling Density and Spatial Resolution
OTF(ω ) "# •" # OTF(3ω )## •# OTF(5ω )# •# # # •# • OTF( 7ω )# ## ## • • ## •# • # ## •$ ! • $ 0
−1
1
3
+1
5
−1
7
+1
0
0
0
15 +1 5 −1 3
0 1
0 0 1
0 0 0 1
0 0 0 0
−1
1
0 1
3
13
−1
0
0
11
+1
0
0
9
−1
+1
17
−1
19
0
0
0
0
0 0 0 0
0 0 0 0
0 0
0
where ωo = 2π /P. The observable and/or measurable quantities are the contrasts {C(p), C(P/3), C(P/5),…}, and this relationship represents a sparse numerical matrix equation that can be solved directly for the values of the OTF at the indicated frequencies (Wolfram, 1991). If we denote the three matrices in the equation above by C, M, and OTF, then the solution for the OTF will be given by OTF = M−1C, where M is the matrix of numerical coefficients. These coefficients are constant—that is, they are not a function of the experimental data—and M−1 need be computed only once. For the 10 × 10 case drawn from above, the exact solution is given by:
2.6.12 Supplement 4
Current Protocols in Cytometry
OTF(ω ) "# OTF(3ω ) # ## 1 OTF(5ω ) # OTF( 7ω ) # ## OTF(9ω ) ## = OTF(11ω ) # # OTF(13ω ) # # OTF(15ω )# # ! OTF(17ω )# # !OTF(19ω )#$
0
0
+1
0
1
3
−1
5
+1
7
0
+1
0
0
0 0 0 1
0 0 0 0 1
0
0
+1
1
0 1
0 0 1
0 0
0
3
11
0
0
−1
13
−1
15 −1 5 +1 3
0 0 0 0 1
−1
17
0 0 0 0 0 0 0 1
0 0
"# C( p / 3) # +1 " 19# # 0 # C( p / 5) # # 0 # C( p / 7) # # ## 0 # # ( / 9 ) C p 0 # # • 0 # C( p / 11) # # # 0 # C( p / 13)# ## 0 # # C( p / 15)# 0 # # 1 #$ C( p / 17)# !C( p / 19)#$ C( p)
It is the specific use of a bar test pattern with a 50% duty cycle that makes it possible to find values of the OTF based upon values of the contrast modulation transfer function. The number of measurements that need to be included depends on the bar pattern as well as the expected maximum frequency in the OTF. According to the equations for PSF and OTF, a highest frequency of fc = 2NA/λ is expected for a diffraction-limited objective lens. If the smallest period of the bar spacing is Ps, the maximum frequency that can be deduced using this procedure and the matrix describing the relationship between C(p) and OTF(ω) is 1/Ps. In principle, this should be greater than 2NA/λ. If, for example, the system is limited by an objective lens with an NA of 0.7 and λ of 700 nm, then the lens is capable of resolving up to 2 cycles/µm (2000 cycles/mm). If the pattern has Ps = 2 µm, then the highest frequency from the chart, using this procedure, is only 0.5 cycles/µm (500 cycles/mm). This may not be as bad as it sounds. The resolution of the system can never be better than 2 cycles/µm. But due to poor electronics, noise, CCD chip design, and other elements in the total system, the resolution can certainly be worse. The results of using such a procedure on a specific microscope system (Young et al., 1982) are illustrated in Figure 2.6.13. Although the explanation might give the impression that the technique is quite complicated, the procedure for a given test chart is quite straightforward: 1. Grab a black image. 2. Grab a white (blank) image. 3. Grab an image containing the bar chart (as in Fig. 2.6.12A). 4. Correct for shading using Equation 2.6.7. 5. Measure C(p) for the collection of bar patterns that are available. 6. Plot C(p) or C(s = 1/p) from the longest period Po to the shortest period Ps (as in Fig. 2.6.13). 7. Evaluate C(p) at the values {p = Po , p = Po /3, p = Po /5, p = Po /7,…}, using interpolation as necessary. 8. Using the observed set {C(Po /n) | n = 1, 3, 5, 7,…} and the matrices above, solve for {OTF(fn = n/Po) | n = 1, 3, 5, 7,…}. 9. Plot OTF(fn) from the lowest frequency 1/Po to the highest frequency n/Po (as in Fig. 2.6.13). Image Cytometry Instrumentation
2.6.13 Current Protocols in Cytometry
Supplement 4
Microbead Image A third procedure for determining the PSF or the OTF of a complete system is based upon using a test object that is smaller than a wavelength of light in at least one dimension. Such an object, if it is a line, will produce the line spread function L[x,y], which—perpendicular to the line—will be the point spread function, PSF[x,y]. This approach is clearly related to the step-edge technique described above. If the test object is essentially a point source, then the result will be the “impulse” response of the entire system. But the impulse response is precisely the PSF. The point sources normally used for this purpose are microbeads, latex spheres that are manufactured to diameter specifications that can be significantly less than a wavelength of light. These spheres can be obtained with either fluorescent surface coatings or total fluorescent staining of the entire latex sphere (typical sources include Polysciences and Flow Cytometry Standards Corporation; see SUPPLIERS APPENDIX). The relationship between the three-dimensional image i[x, y, z], the total point-spread function PSF[x, y, z], and the sphere s[x, y, z] is given by:
i[ x, y, z] = psf [ x, y, z ] ⊗ s[ x, y, z] Equation 2.6.10
which in the Fourier domain becomes: I[ω x , ω y , ω z ] = OTF[ω x , ω y , ω z ] • S[ω x , ω y , ω z ] Equation 2.6.11
where the Fourier spectrum of a sphere is known (van Vliet, 1993). Because S[ωx,ωy,ωz] contains zeroes in the frequency domain and because the observed data I[ωx,ωy,ωz] are invariably noisy, direct inversion of this equation to solve for OTF[ωx,ωy,ωz] is not possible. As a result a number of nonlinear iterative procedures have been developed to estimate the OTF given S(…) and I(…). These procedures are not for the faint of heart. As the formalism and methodology for using them is beyond the scope of this unit, the reader is instead referred to the literature—in particular, Holmes and Liu (1991), Holmes (1992), Holmes et al. (1995), van den Voort and Strasters (1995), and van Kempen et al. (1996). CONCLUSIONS Measuring the sampling density and spatial resolution of a complete imaging system involves the careful use of calibration standards and appropriate image processing algorithms. A single well-constructed test slide is capable of providing images for both types of measurements. Measurement of sampling density will, in general, provide two or three numbers: ∆x, ∆y, and optionally ∆z. Measurement of spatial resolution requires the determination of a complete curve (function) known as the OTF, which can also be a function of one, two, or three independent variables. Because a microscope system always consists of more than just a circularly symmetric microscope lens, it is always necessary to make several measurements in at least the x and y directions to determine if the total system exhibits circular symmetry. Because of the advances in modern computer systems, none of the procedures described above requires more than a few seconds of computation on a personal computer. Calibration: Sampling Density and Spatial Resolution
2.6.14 Supplement 4
Current Protocols in Cytometry
LITERATURE CITED Born, M. and Wolf, E. 1980. Principles of Optics, 6th ed. Pergamon Press, Oxford, U.K. Canny, J. 1986. A computational approach to edge detection. IEEE Trans. Pattern Anal. Machine Intell. 8:679-698. Castleman, K.R. 1996. Digital Image Processing, 2nd ed. Prentice-Hall, Englewood Cliffs, N.J. Holmes, T.J. 1992. Blind deconvolution of quantum-limited incoherent imagery: Maximumlikelihood approach. J. Opt. Soc. Amer. A9:1052-1061. Holmes, T.J. and Liu, Y.-H. 1991. Acceleration of maximum-likelihood image restoration for fluorescence microscopy and other incoherent imagery. J. Opt. Soc. Amer. A-8:893-907. Holmes, T.J., Bhattacharyya, S., Cooper, J.A., Hanzel, D., Krishnamurti, V., Lin, W.C., Roysam, B., Szarowski, D.H., and Turner, J.N. 1995. Light microscopic images reconstructed by maximum-likelihood deconvolution. In Handbook of Confocal Microscopy (J.B. Pawley, ed.) pp. 389402. Plenum Press, New York. Limansky, I. 1968. Electronics Eng., pp. 50-55. O’Neill, E.L. 1963. Introduction to Statistical Optics. Addison-Wesley, Reading, Mass. Oppenheim, A.V., Willsky, A.S., and Young, I.T. 1983. Systems and Signals. Prentice-Hall, Englewood Cliffs, N.J. Press, W.H., Flannery, B.P., Teukolsky, S.A., and Veterling, W.T. 1988. Numerical Recipes in C. Cambridge University Press, Cambridge, U.K.
van den Voort, H.T.M. and Strasters, K.C. 1995. Restoration of confocal images for quantitative image analysis. J. Microsc. 178:165-181. van Kempen, G.M.P., van den Voort, H.T.M., Bauman, J.G.J., and Strasters, K.C. 1996. Comparing maximum likelihood estimation and constrained Tikhonov-Miller restoration. IEEE Eng. Med. Biol. 15:76-83. van Vliet, L.J. 1993. Grey-Scale Measurements in Multi-Dimensional Digitized Images. Ph.D. thesis, Delft University of Technology, Delft, The Netherlands. Wolfram, S. 1991. Mathematica: A System for Doing Mathematics by Computer, 2nd ed. Addison Wesley, Redwood City, Calif. Young, I.T. 1989. Image fidelity: Characterizing the imaging transfer function. Methods Cell Biol. 30:1-45. Young, I.T. 1996. Quantitative microscopy. IEEE Eng. Med. Biol. 15:59-66. Young, I.T. and van Vliet, L.J. 1995. Recursive implementation of the Gaussian filter. Signal Processing 44:139-151. Young, I.T., Balasubramanian, Dunbar, D.L., Peverini, R.L., and Bishop, R.P. 1982. SSAM: Solidstate automated microscope. IEEE Trans. Biomed. Eng. 29:70-82.
Contributed by Ian T. Young Delft University of Technology Delft, The Netherlands
Dr. Young wishes to acknowledge the support of the Netherlands Organization for Scientific Research (NOW) Grant 900-538-040, the Foundation for Technical Science (STW) Project 2987, the European Concerted Action for Automated Molecular Cytogenetic Analysis (CA-AMCA), and the Rolling Grants program of the Foundation for Fundamental Research in Matter (FOM).
Image Cytometry Instrumentation
2.6.15 Current Protocols in Cytometry
Supplement 4
Microscope Alignment
UNIT 2.7
The light microscope will deliver optimal performance only if its various components are properly installed and aligned. For upright microscopes from any of the major manufacturers, proper manufacture and assembly will ensure that the following conditions are met. 1. The microscope stage is firmly mounted to the focusing mechanism and is level. The stage should travel straight up or down and remain level when either the coarse or fine focus knob is rotated. 2. The substage condenser is properly installed and secured. The substage condenser should travel straight up or down when its elevating knob is rotated. 3. The binocular or trinocular observation tube is properly seated, and each user has: • adjusted the interocular distance between the eyepiece tubes (the fields of view from each tube should “fuse” into a single, circular field of view) and • adjusted the eyepieces so that any difference in acuity between the two eyes has been properly compensated for (see Eyepiece Adjustment). 4. The nosepiece and its complement of objectives have been designed so that the objectives are: • parfocal (i.e., little, if any, refocusing is required when changing objectives) and • parcentric (i.e., an object in the center of the field of view remains centered after changing objectives). Modern microscopes, including even moderately priced laboratory models with inexpensive achromatic objectives, are capable of producing excellent images providing that the optics are kept clean and free of obstructions. In this unit, the two light paths of the microscope’s optical system will be discussed: the image-forming path and the illumination path. The components of these optical paths must be properly aligned to achieve optimal image quality. More comprehensive descriptions and discussion of the microscope’s optical system may be found in Leitz (1938), Mollring (1976), Spencer (1982), Bradbury (1984), Abramowitz (1985, 1987, 1993, 1994), Inoue (1986), Delly (1988), and UNIT 2.2. THE IMAGE-FORMING SYSTEM The optical components of the image-forming system (Fig. 2.7.1) include the objective and the eyepiece. When properly installed and undamaged, these components are prealigned—the user need only keep them clean and free from contaminants such as dust, fingerprints, and immersion oil. Routine observance of the following practices will go a long way toward keeping the optical components clean. 1. Keep fingertips away from all lenses and optical glass surfaces. 2. At the end of each day’s use, carefully and properly remove contaminants (e.g., immersion oil, fingerprints) from all optical surfaces using the manufacturer’s recommendations (see Maintenance of Optical Glass Surfaces). 3. Keep the microscope covered when not in use, using the plastic dust cover supplied by the manufacturer. Image Cytometry Instrumentation Contributed by Marc M. Friedman and Mortimer Abramowitz Current Protocols in Cytometry (1997) 2.7.1-2.7.8 Copyright © 1997 by John Wiley & Sons, Inc.
2.7.1 Supplement 2
retina of the eye
eyepoint
image formed by objective (intermediate image plane)
specimen plane
condenser aperture diaphragm
field diaphragm
Figure 2.7.1 Image-forming ray paths in Köhler illumination are traced from two ends of the lamp filament. Conjugate planes are at the field diaphragm, specimen plane, intermediate image plane (entrance pupil of the eyepiece), and the retina of the eye. Modified with permission from Abramowitz (1985). Microscope Alignment
2.7.2 Supplement 2
Current Protocols in Cytometry
EYEPIECE ADJUSTMENT The microscope’s eyepieces provide the port through which information, in the form of an image, is transferred from the microscope to the user. Differences in visual acuity between users, and even between the eyes of the same user, require that the eyepieces be adjusted to accommodate each individual’s interocular distance and each eye’s visual acuity. Otherwise, the observed image quality may suffer dramatically and the user may experience discomfort, caused by eye strain, when using the microscope. To adjust the eyepieces, follow the procedure outlined below. It is intended for a typical microscope in which only the left eyepiece or eyepiece tube is adjustable, but is readily adaptable for microscopes with two adjustable eyepieces. 1. Select a low-magnification (e.g., 10×) dry objective and place a specimen slide on the stage. 2. With the left eye closed, look at the specimen through the right eyepiece, locate a well-defined target object near the center of the field of view, and bring it into sharp focus. 3. With the right eye now closed, rotate the diopter-adjustment ring on the left eyepiece alternately clockwise and counterclockwise until the target object is in sharp focus. 4. With both eyes open, touch up the focus of the left eyepiece using the diopter-adjustment ring. 5. Set the proper interpupillary distance for your eyes by grasping the bases of the eyepiece tubes (not the knurled adjustment ring) and moving them together or apart, as necessary. THE ILLUMINATION SYSTEM Proper adjustment of the microscope’s illumination system is a continuing and crucial requirement for observation, photomicrography, or electronic imaging. In typical modern microscopes the following components should be present and correctly aligned: 1. the light bulb • usually housed within a removable lamphouse; can be either precentered in the illumination path or centered by the user with knobs or screws built into the lamphouse 2. the collector lens • built into the base of the microscope and permanently aligned by the manufacturer 3. the variable field diaphragm • usually built into the base of the microscope, below the stage 4. the substage condenser • typically mounted beneath the stage in a bracket that can be raised or lowered independently of the stage by rotating a knurled knob • centered in the optical path using knobs or screws that extend from the housing 5. the variable aperture diaphragm • built into the substage condenser housing (optically, it is positioned at or near the front focal plane of the condenser) • opened or closed by means of a lever or knurled knob
Image Cytometry Instrumentation
2.7.3 Current Protocols in Cytometry
Supplement 2
Köhler Illumination Köhler illumination is provided by all manufacturers of modern laboratory microscopes because it can provide specimen illumination that is uniformly bright and free from glare, thus allowing the user to realize the microscope’s optimum performance. The manufacturers have designed the microscope so that the collector lens and any other optical components built into the base of the microscope will project an enlarged and focused image of the lamp filament onto the plane of the aperture diaphragm of a properly positioned substage condenser. Closing or opening the condenser diaphragm controls the angle of the light rays emerging from the condenser and reaching the specimen from all azimuths. The setting of the condenser’s aperture diaphragm, along with the aperture of the objective, determines the realized numerical aperture (NA) of the microscope “system.” As the condenser diaphragm is opened, the working NA of the microscope increases, resulting in greater resolving power and light transmittance. The field diaphragm in the base of the microscope controls only the width of the bundle of light rays reaching the condenser—it does not affect the optical resolution (NA) or the intensity of illumination. Proper adjustment of the field diaphragm (i.e., centered in the optical path and opened so as to lie just outside of the field of view) is important for preventing glare that can reduce contrast in the observed image. To achieve Köhler illumination, proper adjustment of the condenser and field diaphragms is critical. These adjustments must be made by the microscopist each time the microscope is used and each time the objective is changed. Steps in setting up Köhler illumination 1. Open the field and aperture diaphragms all the way. 2. Focus a typical specimen using a 10× objective, and adjust the transformer voltage so that the light intensity is comfortable for viewing. 3. Close down the field diaphragm most of the way. 4. While looking through the eyepieces, carefully raise or lower only the substage condenser (not the entire stage) until the polygon-shaped edge of the field diaphragm is inside the field of view and is sharply focused. 5. Using the small condenser-centering screws at the base of the substage holder, center the image of the field diaphragm in the field of view. Then open the field diaphragm until it just disappears from view. 6. Lift out one of the eyepieces and look down the observation tube at the back of the objective. 7. Slowly close the aperture diaphragm of the substage condenser and observe that the image of the condenser diaphragm is clearly visible at the back of the objective. If the illumination system does not have a frosted filter inserted in the light path, an image of the lamp filament will be visible. The filament image should be centered in the back aperture of the objective, either by the manufacturer (with precentered bulbs) or by utilizing a set of centering screws located on the lamphouse. The filament image should also fill, or nearly fill, the back aperture of the objective. 8. Adjust the aperture diaphragm so that it is open two-thirds to three-quarters of the way. This adjustment may vary according to the specimen: the aperture diaphragm might be opened nine-tenths of the way for a well-stained specimen. Microscope Alignment
9. Replace the eyepiece.
2.7.4 Supplement 2
Current Protocols in Cytometry
This completes the process of setting up Köhler illumination with the 10× objective. When the objective is changed during normal use of the microscope, the field diaphragm and the condenser diaphragm must be readjusted according to the procedure outlined above. For example, when switching from a 10× to a 20× objective: a. the field diaphragm should be closed down somewhat, because the specimen area being viewed is smaller, b. it should be recentered, and c. the aperture diaphragm should be opened somewhat to increase the numerical aperture of the condenser to match the higher NA of the 20× objective. NOTE: The substage condenser is properly aligned when the edge of the field diaphragm is sharpest. At this position, the edge may exhibit a slight blue tint, which should not vary in color around the edge. A highly corrected, achromatic-aplanatic condenser will yield an essentially colorless edge. Variation in color along the edge of the field diaphragm (e.g., from blue to red) indicates either that the diaphragm is tilted or, more likely, that some components in the illumination path are not properly aligned. For routine work this may not matter, but prior to critical work the microscope should be serviced. Conjugate planes To better understand Köhler illumination, it is helpful to separate the image-forming light path and the illumination light path, and to look at the various planes, or levels, in each of these paths. This is referred to as the analysis of the conjugate planes of these paths. By definition, an object that is in focus at one plane is also in focus at the other conjugate planes of that light path. Conjugate planes in the image-forming light path in Köhler illumination include: 1. the field diaphragm, 2. the focused specimen, 3. the intermediate image plane (i.e., the plane of the fixed diaphragm of the eyepiece), and 4. the retina of the eye or the film plane of the camera. Conjugate planes in the path of the illuminating light rays in Köhler illumination include: 1. 2. 3. 4.
the lamp filament, the condenser aperture diaphragm (at the front focal plane of the condenser), the back focal plane of the objective, and the eyepoint (also called the Ramsden disk) of the eyepiece, which is located ∼1⁄2 in. (∼1 cm) above the top lens of the eyepiece, at the point where the observer places the front of the eye during observation.
What are the advantages of Köhler illumination? 1. The image appears bright and evenly lighted, providing ideal illumination for observation and photomicrography. In older methods of illumination, the image of the lamp filament was focused at the specimen plane, thus partially obscuring the image of the specimen and also heating the specimen excessively. 2. Proper adjustment of the field diaphragm is readily achievable, which is important in controlling the width of the illuminating light beam to minimize glare. 3. The condenser aperture diaphragm can be varied to control the angle of the illuminating rays reaching the specimen and the image-forming optics, and thus controls
Image Cytometry Instrumentation
2.7.5 Current Protocols in Cytometry
Supplement 2
the working NA of the microscope. Leaving the condenser aperture wide open provides maximum theoretical resolution, but greatly reduces contrast and results in an inability to see fine detail. Closing the condenser diaphragm somewhat, to two-thirds or three-quarters of the area of the back focal plane of the objective (see discussion of steps in setting up Köhler illumination, above), strikes a reasonable compromise among resolution, contrast, and depth of field. A Note About Reflected Light Microscopy Köhler illumination is also important in reflected light brightfield microscopy (see Fig. 2.7.2) and epifluorescence microscopy. A good-quality reflected light illuminator has a field diaphragm and an aperture diaphragm. However, in a reflected light illuminator their positions are reversed—the aperture diaphragm is situated closest to the lamp filament, and the field diaphragm is closest to the specimen. In reflected light fluorescence, where the nature of the specimen often makes resolution less important than light capture, the aperture diaphragm is usually left wide open to permit capture of the most light. Conjugate planes are similar to those described for transmitted light. The microscope objective performs a dual role: that of condenser lens (for light traveling toward the specimen) and objective lens (for light returning from the specimen).
intermediate image plane FD′′
L1
L2
L3
half-mirror F F′ CAD
F′′
FD
objective
FD′ specimen
Figure 2.7.2 Köhler illumination for brightfield reflected light (diagrammatic for finite-tube-length objective system). F, filament light source; F′, image of light source at condenser aperture diaphragm (CAD); F′′, image of light source at back focal plane of objective; FD, field diaphragm; FD′ and FD′′, conjugate planes to FD; L1, L2, and L3, lenses of vertical illuminator. Reprinted with permission from Abramowitz (1990). Microscope Alignment
2.7.6 Supplement 2
Current Protocols in Cytometry
MAINTENANCE OF OPTICAL GLASS SURFACES The microscope’s optical surfaces are its most important but also its most delicate feature. Routine care of the optical surfaces as part of regular use will greatly help to preserve the microscope’s image quality. In the next two sections, several recommended routine practices are outlined, and guidelines are given for cleaning external optical glass surfaces. Do not attempt to clean internal optical or other microscope surfaces, or to clean stubborn contaminants (e.g., dried immersion oil) from external surfaces—call your local microscope service dealer! Routine Care 1. Keep the microscope covered when not in use. 2. Keep fingers and foreign materials away from optical surfaces. 3. Remove dust by blowing it from the the surface of the lens using a duster can of compressed air or nitrogen (purified and filtered). 4. Regularly check the external surface of components such as eyepieces and filters for contaminants. Dust, makeup, and body oil are common; blow off dust and, if necessary, clean as described below. 5. For immersion objectives, use only immersion oil recommended by the microscope manufacturer, and clean immersion oil from the objectives at the end of each day (check all objectives—dry objectives can become contaminated accidentally). Cleaning While a detailed treatment of cleaning supplies and methods is beyond the scope of this section, some discussion is warranted. Consult your microscope’s manufacturer or local dealer for any booklets or written materials that might be available. Many service dealers can provide expert advice and tips, as well as proper cleaning supplies, so don’t hesitate to consult them. Selection of the proper cleaning fluid is important, and there are many opinions about which cleaners and solvents are safe and effective. As noted above, the best place to start may be your local microscope service dealer. As a general rule, unless you are knowledgeable about their use and limitations, avoid toxic or dangerous (e.g., explosive) solvents such as acetone and ether, which may also damage painted and other microscope surfaces. Some references recommend a cleaning solution of 7 parts ether and 3 parts absolute ethanol, but substituting Freon for the ether is safer. Ethanol alone may also be used. While certain commercial cleaners such as Windex and Kodak Lens Cleaner can be used for the exposed surface of an eyepiece or objective, they may leave a slight film. In many cases, simply breathing onto the surface of the lens (followed by gentle wiping) provides a moisture layer that will lift off dust, dissolve organics, discharge static, and let the user see any dirt or imperfections on the surface. To wipe, use only optical-quality soft lens tissue or lint-free cloth that has been stored in resealable plastic bags to avoid dust contamination. Wooden, cotton-tipped applicators (individually wrapped) can be ordered from hospital supply companies. Avoid consumer cotton swabs and tissues, commercial lab-wipes (which may contain glue, wax, or dust), and camera lens cleaners with unknown ingredients. Exert only very gentle pressure while cleaning—the weight of the materials is sufficient, and additional pressure may cause trapped dust particles to scratch the lens. To clean a lens surface: 1. Wrap a layer of lens tissue around a cotton-tipped applicator. 2. Moisten it with cleaning solution and shake off the excess, which could seep into microscope parts.
Image Cytometry Instrumentation
2.7.7 Current Protocols in Cytometry
Supplement 2
3. Rest the tip of the applicator in the center of the glass surface to be cleaned. 4a. For an eyepiece or filter: Gently wipe the surface in an outward spiral pattern, from the center to the edge. 4b. For an objective lens: Rotate the lens (i.e., on a table top) while moving the applicator from the center toward the periphery. 5. Discard the used applicator and repeat with fresh materials, if necessary. Alternatively, moisten a piece of lens tissue with cleaning solution, cover the lens as if with a blanket, and gently pull the tissue along the surface and off of the lens. Gently remove any remaining fluid using fresh lens tissue. LITERATURE CITED Abramowitz, M. 1985. Microscope Basics and Beyond. Olympus Corporation, New York. Abramowitz, M. 1987. Contrast Methods in Microscopy: Transmitted Light. Olympus Corporation, New York. Abramowitz, M. 1990. Reflected Light Microscopy: An Overview. Olympus Corporation, New York. Abramowitz, M. 1993. Fluorescence Microscopy: The Essentials. Olympus America, New York. Abramowitz, M. 1994. Optics: A Primer. Olympus America, New York. Bradbury, S. 1984. An Introduction to the Optical Microscope. Royal Microscopical Society, Microscopy Handbooks 01. Oxford University Press, Oxford, UK. Delly, J.G. 1988. Photography Through the Microscope. Eastman Kodak, Rochester, New York. Inoue, S. 1986. Video Microscopy. Plenum Press, New York. Olympus B-Max Microscope Guide to Cleaning and Adjustment. Olympus America, New York. Leitz, E. 1938. The Microscope and Its Application. Ernst Leitz, Wetzlar, Germany. Mollring, F.K. 1976. Microscopy From the Very Beginning. Carl Zeiss, Oberkochen, Germany. Spencer, M. 1982. Fundamentals of Light Microscopy. IUPAB Biophysics Series, Cambridge University Press, Cambridge, UK.
Contributed by Marc M. Friedman Accumed International, Inc. Chicago, Illinois Mortimer Abramowitz Olympus America Inc. Mellville, New York
Microscope Alignment
2.7.8 Supplement 2
Current Protocols in Cytometry
Confocal Microscopy: Principles and Practices In light microscopy, the illuminating light passes through the specimen. The light is delivered as uniformly as possible over the field of view of the microscope. If the specimen is thicker than the depth of focus of the objective lens, light coming from structures above and below the plane of focus will also enter the detector (eye or camera). In fluorescence microscopy, any dye present in the specimen above and below the plane of focus will be stimulated and the fluorescent light will enter the detector. This light coming from out-of-focus structures will be added to that coming from the plane of focus and will tend to blur the image and make it difficult to resolve detail, especially where overlapping structures are present. Confocal microscopy is a form of light microscopy in which the illuminating light and the light-collection optics are focused on the same diffraction-limited spot in the specimen. Unlike the conventional microscope, the confocal instrument images only the one spot— rather than the entire field of view of the objective lens—onto the detector. To generate a complete image, the spot is moved over the specimen and the image built point by point. The most important aspect of confocal microscopy is that parts of the specimen not at the focal point contribute very little to the in-focus image. This effectively does away with the out-of-focus light problem. Because confocal microscopy allows for high-resolution meas-
urements within thick specimens, it has spurred considerable development in optical sectioning and three-dimensional reconstruction (see section on Applications, below). Other advantages of the confocal light microscope will also be discussed.
DESIGN Confocal microscopy is not a new technology. It was invented in 1955 by Marvin Minsky, a Harvard postdoctoral fellow who was trying to see the interconnections between neuronal cells (Minsky, 1957; 1988). The large number of stained cells present in the sample and the large amount of light scattered by them into the field of view blurred the image and made it impossible to resolve individual cells with a conventional light microscope. His device, patented in 1957, is illustrated in Figure 2.8.1. This is the simplest form of confocal microscope. Minsky used a zirconium arc light source focused onto a pinhole to provide the illumination. The condenser lens focuses an image of the illumination pinhole to a diffraction-limited spot in the specimen. The objective lens then forms an image of this spot in the specimen onto the pinhole in front of the photodetector, which measures the light passing through the pinhole. The two pinholes and the spot in the object are “confocal.” Any other point in the object is not focused on either pinhole and thus is not efficiently illuminated; moreover, most of the light scattered by the object would not
object
pinhole
UNIT 2.8
pinhole
light source
photodetector
condenser lens
objective lens
Figure 2.8.1 Confocal microscope design of Minsky. The condenser forms an image of the illumination pinhole onto a spot in the object. The objective lens forms an image of the spot in the object onto the detector pinhole. The two pinholes and the spot in the object are confocal. Any other spot in the object would be poorly illuminated and its emission would not pass efficiently through the pinhole. Contributed by Phillip N. Dean Current Protocols in Cytometry (1998) 2.8.1-2.8.12 Copyright © 1998 by John Wiley & Sons, Inc.
Image Cytometry Instrumentation
2.8.1 Supplement 5
pass through the second pinhole. This is illustrated by the dashed lines in Figure 2.8.1. To scan the object and form a complete image, Minsky moved the object line by line in an xy rectangular pattern. His detector was a lownoise photomultiplier and the image was viewed with a long-persistence radar scope. Scanning a field took 10 sec. Minsky also realized that there was more than one way to form the confocal image. The method illustrated in Figure 2.8.2 utilizes epiillumination, which is particularly well suited to fluorescence microscopy. This figure also shows both the use of a laser beam as the source of illumination and the discrimination against light coming from outside the focal plane (dotted and dashed lines). The dichroic mirror and filter are required to prevent scattered laser light from reaching the detector. The object is scanned using an x-y scanner to move the beam across the object, rather than moving the object through the beam. Advantages and disadvantages of the two scanning methods will be discussed later. As noted above, the confocal method is highly desirable since it rejects out-of-focus light. As illustrated in Figure 2.8.2, the illumi-
nation light passes through the object as two cones of light above and below the focal plane. Any material that will scatter the light or any fluorescent dye within the two cones will also cause light to be directed toward the detector. As this light does not come from the focal spot, however, it will not be focused onto the pinhole and will be strongly rejected. The end result is that the image received by the detector will be very little influenced by material outside the focal plane and will produce an image of a very thin slice of the object at a preset depth within it. In a conventional microscope, there is no pinhole and all of the light collected by the objective lens will be detected.
Resolution Because in confocal microscopy there are effectively two apertures, the resolution of the microscope, both laterally and axially, is greatly enhanced over that of a conventional microscope. In principle one gains a factor of 1.4 in both directions, as a diffraction-limited image is formed of a diffraction-limited spot in the specimen. By “resolution” is meant the minimum distance between two points in a specimen whose diffraction images can still be
dichroic mirror laser beam
photomultiplier
x
filter y x -y scanner
pinhole
microscope objective
focal plane
Confocal Microscopy: Principles and Practices
Figure 2.8.2 An epi-illuminated laser scanning confocal microscope. Rotating mirrors are inserted between the laser and the object to permit scanning of the object in three dimensions at high speed. Since the illuminating and fluorescent light both pass through the same lens and are reflected from the same scanner mirrors, only one pinhole is required.
2.8.2 Supplement 5
Current Protocols in Cytometry
resolved. The resolution depends on the wavelength of light used and on the numerical aperture (NA) of the lens, as well as on the index of refraction of the immersion medium for the axial resolution. The resolution of a confocal microscope is degraded, however, if pinholes larger than the spot size at the aperture are used. Having no pinhole yields the same image as a conventional microscope. Ideally, in a confocal microscope one can obtain resolutions on the order of 0.18 µm laterally and 0.6 µm axially for a 60× objective lens of NA 1.4 operating at 488 nm, assuming that the pinhole is of the minimum diameter. In practice these numbers are usually somewhat higher, typically 0.2 to 0.3 µm laterally and 0.7 µm axially. There is also a tradeoff between resolution and efficiency of light collection. For dimly fluorescing objects it might be more advantageous to give up some resolution and open the pinhole
slightly to collect more light and improve the contrast in the image. A full theoretical treatment of confocal microscopy can be found in Wilson (1990).
Examples of Confocal Effect Figures 2.8.3, 2.8.4, and 2.8.5 each contain two images of the same sample: the left image taken with the detector pinhole wide open (1600-µm diameter), simulating conventional microscopy, and the right image taken with an optimum pinhole for confocal operation (40µm diameter). In Figure 2.8.3, for example, showing lily pollen labeled with acridine orange, confocal operation provides a dramatic improvement. In the left (conventional microscopy) side of the figure individual cells cannot be distinguished; in the right (confocal) image both the cells and the distribution of the dye within them are visible.
Figure 2.8.3 Lily pollen labeled with acridine orange. The left image was taken with an aperture (pinhole) of 1600 µm diameter to simulate nonconfocal operation and the right image was taken with a 40-µm-diameter aperture. The improvement with confocal operation is dramatic. These images were provided through the courtesy of Edwin de Feijter, Insight Biomedical Imaging (Lansing, Mich.).
Image Cytometry Instrumentation
2.8.3 Current Protocols in Cytometry
Supplement 5
Utilizing color, particularly several color labels in combination, can improve the power of the confocal method. For example, the rod photoreceptors of Figure 2.8.5 were labeled with a monoclonal antibody to opsin, followed by a FITC-conjugated secondary antibody (green), in addition to the lectin wheat germ agglutinin conjugated to the fluorochrome Cy3 (red). In this case, the confocal image allows easy determination that the opsin is located on the surface of the rods.
TYPES OF CONFOCAL MICROSCOPES Two basic types of confocal microscopes are in use today: fluorescence and reflection. A third mode, transmission, which is the original one developed by Minsky (see Fig. 2.8.1), is seldom used owing to difficulties in maintaining optical alignment for the large number of components. Figure 2.8.2 shows the epi-illumination arrangement for fluorescence, in which the dichroic mirror is selected to pass the laser
Confocal Microscopy: Principles and Practices
light and reflect the longer-wavelength fluorescence light. The filter is selected to pass the fluorescence light and reject reflected laser light. As in flow cytometry, two fluorophores can be used simultaneously to mark two objects of interest. In such arrangements the filter is replaced with a second dichroic mirror to split the light emitted by the two fluorophores, and two photodetectors are used. The filters are included to provide additional selection power and to reject any stray laser light. This process has been extended by adding a second laser and additional dichroic mirrors and filters to the system to allow the simultaneous detection of four or more fluorochromes. If the dichroic mirror is replaced with a partially reflecting mirror (beamsplitter) and the filter is removed, then one can measure light scattered back from the object, as in reflection microscopy (Cogswell and Sheppard, 1990). Reflection microscopy is a valuable tool in the fields of metallurgy and silicon chip manufacturing. Applying confocal techniques has ex-
Figure 2.8.4 A live human teratocarcinoma cell labeled with the membrane probe NBD-phosphatidylcholine. Images were collected as in Figure 2.8.3. These images were provided through the courtesy of Edwin de Feijter, Insight Biomedical Imaging (Lansing, Mich.).
2.8.4 Supplement 5
Current Protocols in Cytometry
panded the usefulness of this technology in obtaining surface topography maps. Differential interference contrast techniques have also been successfully applied to confocal microscopy (Cogswell and Sheppard, 1992) and are proving useful in obtaining phase information from largely transparent objects.
METHODS OF SCANNING In confocal microscopy, the image is constructed one point at a time. To produce an image of a layer in the object, some type of scanning system must be employed. Currently three methods of scanning the object are in use.
Scanning Laser Beam In Figure 2.8.6 a scanner device is shown placed between the laser and the object. This device steers (scans) the laser beam in a rectangular pattern. The laser beam is directed onto a pair of oscillating galvanometer mirrors (Amos
et al., 1987). One mirror moves the beam across the object in the x direction to form the first line of the image. The second mirror is then moved incrementally in the y direction to permit a second scan in the x direction. This process is continued until an image is formed of a slice through the object. By changing the focal point and repeating the process one can construct a “stack” of images representing the entire specimen, which can then be processed for three-dimensional information. This process is discussed in more detail below. This type of scanning has several advantages: (1) a scan typically takes ≤1 sec; (2) the sample doesn’t move, which could be very important in examining live cells or objects mounted in liquid; and (3) the scanner can be attached to a conventional microscope without any modification to the microscope. The primary disadvantage of this scanning method is that the laser beam must pass through parts of
Figure 2.8.5 Rod photoreceptors from the frog Xenopus laevis labeled by the lectin wheat germ agglutinin conjugated to the fluorochrome Cy3. Images were collected as in Figure 2.8.3. These images were provided through the courtesy of Edwin de Feijter, Insight Biomedical Imaging (Lansing, Mich.).
Image Cytometry Instrumentation
2.8.5 Current Protocols in Cytometry
Supplement 5
the objective lens at a relatively long distance from the optical axis. To achieve optimum performance and the resolution promised by theory, the objective lens must be well corrected for flatness for the full width of the field and for chromatic aberrations for both the illuminating and emission wavelengths. This is a stringent requirement that is not easily met, and some loss of resolution is usually experienced. One frame per second is not “real time.” Two approaches have been taken to produce a device that will operate at video rates. In one device a galvanometer mirror is used to produce the vertical (y) scan at 60 Hz. The high speed x-scan is produced by an acousto-optical deflector (AOD) operating at ∼15 KHz (Draaijer and Houpt, 1987). The combination produces images at standard video rates. The problem with this system is that the diffraction efficiency of the AOD is poor and its deflection is wavelength dependent, making it a poor choice for fluorescence. It works well using only one wavelength—i.e., in reflection microscopy. The second video-rate device employs a single galvanometer mirror, with one side of the mirror used both to scan and descan the specimen (Brakenhoff and Visscher, 1990; Fig. 2.8.7). The other side of the mirror is used to scan the detector
scanning mirrors
dichroic mirror
(a camera) simultaneously to generate the confocal image. Variable slit or pinhole apertures in the illumination and detection paths are used to suppress out-of-focus light. If pinholes are used, the mirror must be rotated in two directions to scan the entire specimen. If a slit light source is used, the mirror needs to rotate in only one direction and images can be formed at video rates. However, uniformity of illumination intensity across the slit is required and is not always easy to achieve. Video scan rates can also be achieved in the more conventional point-scanning systems by increasing the speed of rotation of the galvanometer mirrors by operating them in resonance mode. The mirrors are not designed for such operation, however, and their long-term stability will be compromised. Any degradation in positioning of the mirrors will produce blurring in the image. In three-dimensional microscopy (taking sequential images at different depths in a specimen) there is another potentially serious problem that applies to all methods of scanning. The intense laser beam focused to a small spot subjects fluorescent dyes to very high excitation levels and can result in their photodestruction (photobleaching). Since the laser beam illuminates a cone in passing through the ob-
laser beam
detector pinhole
microscope objective
specimen
Confocal Microscopy: Principles and Practices
Figure 2.8.6 Schematic of operation of the xy scanner. Two galvanometer mirrors are used to scan the object. The mirror drivers are under computer control and scan a single line or a rectangular pattern. They can also be directed to a single spot for repeated measurements over time.
2.8.6 Supplement 5
Current Protocols in Cytometry
object lens mirror
dichroic mirror
laser
CCD camera
lens
slit mirror
mirror slit
Figure 2.8.7 A confocal microscope that includes a CCD camera. The front side of the mirror scans a slit shaped beam over the object, while the reverse side directs the reflected (or fluorescent) light onto a CCD camera, which collects the image.
ject, the entire volume of the object is exposed to the beam during the scan of each layer within it. Although the exposure is at a lower intensity than at the focal point, the damage is cumulative and the image for the last scan (at the bottom) of an object can potentially be bleached by 50% or more by the preceding scans. Although the photodestruction process is not well understood, it is thought to depend on the presence of molecular oxygen. Fading can be alleviated to some extent by adding antioxidants such as propyl gallate, hydroquinone, or p-phenylenediamine to the mounting medium. Also, some dyes (e.g., Texas red and rhodamine) are less affected by fading than others. If the objective of the measurement is qualitative in nature— e.g., location of specific structures—then fading might not be a large problem. If quantitative measurements are desired, however, fading can be a significant problem. A discussion of this problem and means of alleviating it can be found in Florijn et al. (1995).
Scanning Stage In scanning stage systems, the optics are all held fixed, and the object is scanned by moving the microscope stage. Some of the advantages of this method of scanning are: (1) only the central axis of the objective lens is used and effects of lens aberrations are minimized; lenses with high NA and long working distances can be used without degradation of signal; (2) all points in the image have identical optical properties; and (3) the field of analysis is limited only by the range of movement of the stage—i.e., the magnification of an image can be changed without changing lenses.
There are also disadvantages to this form of scanning, including: (1) to achieve optimum resolution in the xy plane, the stage assembly must maintain high mechanical precision and stability; (2) for fast scanning the object is moved with significant force, which can rearrange the internal structure of living cells, among other things; and (3) a longer time is required to move the stage physically than to scan the beam, resulting in longer image-generation times and consequently greater problems with fading. The latter could be a significant disadvantage if a time course of events were being studied.
Scanning Disk Neither the scanning beam nor the scanning stage system can easily produce a real-time image, with the possible exception of the scanning slit system. Normally the best the scanning beam can do is about one image per second. The scanning disk system was developed to provide the additional capability of forming an image in real time (at video rates). The technique is based on a design of Nipkow (1884), developed while he was working on a method of transmitting two-dimensional images electrically, a predecessor to television. In the modern version of the Nipkow disk, a series of pinholes is arranged as an Archimedes spiral (Boyde, 1985; Petran et al., 1968). The principle of operation is illustrated schematically in Figure 2.8.8. The confocal principle is the same as for the other forms of microscopy. Two pinholes on opposite sides of the disk are both imaged onto a point in the object. Illuminating light comes through one pinhole and emitted
Image Cytometry Instrumentation
2.8.7 Current Protocols in Cytometry
Supplement 5
light (scattered or fluorescence) is focused onto the other pinhole. In this configuration, instead of moving the object or the light source, the disk is rotated such that the pinholes trace a path across the object. To cover the entire field at a high rate, the disk contains 20,000 or more matched pinholes, typically ∼30 µm in diameter. The image is captured using conventional cameras such as the CCD (solid-state) type. Images can be collected at video rates (30 frames/sec) or higher. This type of instrument has two principal advantages: (1) tandem scanning confocal microscopy provides real-time imaging, at least in the sense of what humans can visually process; and (2) the object does not have to be moved. There are also several disadvantages. (1) Scanning disk confocal microscopes are usually used in reflection mode, because of the low efficiency of illumination. The light source is not focused onto a single pinhole but onto a large area of the disk so as to illuminate many holes simultaneously. The holes in the disk
consume <1% of the active area of the disk and provide low illumination levels. Because of this, scanning disk systems are not often used in fluorescence microscopy, although some investigators have used intensified CCD cameras to help alleviate this problem. (2) There are stringent demands on the mechanical system. The holes must be placed very accurately and the entire system must be very stable while the disk is spinning; if alignment is not perfect, scan lines can appear in the image, and any vibration can cause blurring of the image. (3) This system, like the laser scanning system, uses offaxis illumination and thus has the same strict requirement for superior optics. (4) In reflection mode, the greatest problem is unwanted scattered light from the illumination source. All surfaces in the system will reflect light, and because the optical efficiency is low, strong efforts must be made to block the scattered light. (5) Cameras do not have the dynamic range of photomultipliers.
lamp camera
spinning disk
beamsplitter top view of disk
objective lens focal plane
Confocal Microscopy: Principles and Practices
Figure 2.8.8 A tandem scanning confocal microscope. In this device the illuminating light is focused onto a small hole (pinhole) in a disk. The emitted light (scatter or fluorescence) is focused onto a complementary hole on the opposite side of the disk and into a camera, producing a confocal arrangement. The disk contains thousands of these hole pairs, of which only a few are shown. The disk is rotated at high speed such that the holes are scanned across the object, forming a complete confocal image of the scanned area at video rates.
2.8.8 Supplement 5
Current Protocols in Cytometry
A
B
C
Figure 2.8.9 Reconstruction of a human-hamster hybrid cell nucleus containing three human chromosomes, showing views at (A) 45°, (B) 90°, and (C) 135°.
To alleviate some of the mechanical problems with the scanning disk system, Kino (1989) developed a system in which detection of the scattered light takes place through the same pinhole used for illumination, as in epifluorescence microscopes. Kino also tilted the disk relative to the optic axis and used polarized light to discriminate further against the illumination light. His disk has 200,000 pinholes of 20- to 25-µm diameter, and rotates at ∼2000 rpm, producing 700 frames per sec.
APPLICATIONS Confocal microscopy provides a means for obtaining sharp images of a thin slice of a thick specimen with minimum background and interference from out-of-focus parts of the specimen. This has led and continues to lead to many applications in the biological sciences (Matsumoto, 1993). A major application is in optical sectioning, a process similar to using a microtome to obtain slices except that it is done optically. The procedure is to start at the top of the cell, take an image, move the microscope stage (cell) up one step, take another image, and so on, until the entire cell has been imaged. The resulting “stack” of images can be computerprocessed to reconstruct the object in three dimensions. Figure 2.8.9 shows a reconstruction of a human-hamster hybrid cell nucleus containing three human chromosomes. The nucleus was labeled with one dye and the chromosomes with another using in situ hybridization. Each layer in the stack of images was processed separately and the contour of each
dye region was drawn. The stack was then projected in a perspective three-dimensional array. By rotating and tilting the image, one can observe the nucleus from any angle; three orientations are shown. Creating stacks at slightly different rotation angles results in stereo pairs of images. The development of high-speed processors (including personal computers) permits rendering of these image stacks to produce realistic images with smooth surfaces. It is also possible to make quantitative measurements on the data, such as size, area, and location of all objects within the specimen. With the microscope stage and image-capture system under computer control, it is possible to “zoom” an image—i.e., limit scanning to a selected part of the field of view by changing the step size. The scan speed can be changed to permit light integration for weakly fluorescing specimens. Alternatively, multiple scans can be made and the results summed or averaged. It is also possible to scan a single line or even make repeated measurements at a single point. If a single line is scanned at each focal plane in the specimen, an xz (or yz) scan can be made, producing an image of a slice down through the specimen, a feature not available in conventional microscopy. In reflection mode, two methods have been developed to measure surface features of an object. These methods have been utilized primarily by the semiconductor industry to examine the surface of silicon chips, but they are also applicable to biological objects. Image Cytometry Instrumentation
2.8.9 Current Protocols in Cytometry
Supplement 5
Confocal Microscopy: Principles and Practices
The autofocus method. This method was developed to generate a surface map of an object. First, a stack of n images is taken, starting at the bottom of the object and proceeding to the top. Taking into account the step size (distance between images), sufficient images are collected to reveal the size features desired. Then, xz images are formed as described in the previous paragraph. For each x, the slice that contains the brightest pixel is found. That z value (slice number, 1 to n) is the height of that point in the object. The bright points will form a contour of the surface at that y value. This operation is repeated for each y. The end result is the height of the object for every point of the object. When plotted in a three-dimensional perspective view, the data form a map of the surface of the object. Extended-focus method. To create an extended-focus image, a stack of n images is collected as above. Then, a single image is formed in which each new pixel is the sum of all corresponding pixels in the stack: i.e., each pixel in the new image is the sum of n pixels. Since each image in the stack is sharply focused (there is no contribution from scattered and out-of-focus light), the result is a high-resolution image of the entire object with no sacrifice in lateral resolution. Both the autofocus and extended-focus methods will produce images with large depths of focus while maintaining xy resolution better than in conventional microscopy. The wide variety of fluorescent probes now available provides many applications for confocal microscopy. The parameters of interest would be the spatial distribution of the probe and the total fluorescence intensity (amount of material bound). Since a standard microscope is used in most commercial instruments, additional parameters that might be measured are depolarization, intensity ratios, emission spectra, and so on. Some targets for the probes are lipids, proteins, DNA, RNA, and organelles. With its capability to discriminate against outof-focus light, the confocal microscope makes it possible to study the location of organelles and macromolecules within cells and trace the movement of macromolecules therein. With its optical sectioning capability it allows for threedimensional reconstruction of cells with full three-dimensional viewing, including stereo pairs. Matsumoto (1993) and Cavanagh et al. (1993) contain extensive discussions of many applications of confocal microscopy in the biological sciences.
PRACTICAL CONSIDERATIONS To achieve the optimum resolution and contrast available with a confocal microscope, one must be even more careful in its use than in conventional microscopy. A few things to keep in mind are listed below. 1. Because the resolution of the optical system in the x, y, and z dimensions is determined by the lenses, always use the best lenses available under the precise conditions for which they were designed. This is always true for conventional as well as confocal microscopy. 2. Be aware that when using coherent light (e.g., from a laser) in reflection microscopy, interference effects can occur and dominate the optical signal. 3. To obtain the best resolution, immersion optics must be used. To avoid flexing of coverslips, use the smallest size feasible. 4. Use low-viscosity immersion oil where possible. 5. For high-speed data acquisition, store collected images onto video disk for later processing. 6. To minimize bleaching of the sample, first locate the object using a low-power objective and low laser power or an arc lamp. 7. To obtain the maximum advantage of confocal microscopy, always use the highestNA objective lens suitable for the application; the lateral resolution varies with 1/NA and the axial resolution varies with 1/NA2. A popular lens for fluorescence microscopy is a 63× 1.4 NA oil-immersion Plan Apochromat. 8. The method used in mounting a specimen can affect its three-dimensional shape. To avoid squashing of the specimen by the cover slip, use spacers to separate the coverslip from the specimen. Nail polish can be used to seal the coverslip to the microscope slide. 9. To achieve the ultimate resolution for any microscope system, keep the optics scrupulously clean at all times.
FUTURE DEVELOPMENTS Most developments in confocal microscopy in the future will probably be in applications to biological and medical research. A quite recent development, however, promises to improve the performance of confocal microscopes significantly, especially in regard to the bleaching problem: two-photon excitation of fluorescent dyes. In 1990, Denk et al. showed that it is possible to produce molecular excitation by the simultaneous absorption of two photons. When a dye molecule absorbs a photon, the energy of the photon causes the molecule to go to a higher
2.8.10 Supplement 5
Current Protocols in Cytometry
energy state. The molecule will rapidly return to its lower energy state by emitting a photon. As some energy is lost in the process, the emitted photon will have a longer wavelength than the incident one (fluorescence). For example, a photon in the ultraviolet (315 nm) will excite a Hoechst 33258 dye molecule, which will then emit a photon at ∼400 nm. If the incident photon flux is high enough, it becomes possible for a dye molecule to absorb two photons at once. If the dye molecule were to absorb two 630-nm photons at the same time, it would reach the same energy state as it would on absorbing a single 315-nm photon and would still emit one 400-nm photon. The critical point of this process is that the probability of absorbing two photons at once would be high only at the focal point of the 630-nm beam. Thus there would be no interaction of the beam with the specimen except at the focal point of the beam. This limits any dye excitation to the focal plane of the beam and results in depth resolution equivalent to that of the conventional confocal microscope, without the need for a pinhole. Including a pinhole, however, produces even better depth resolution, since it helps prevent scattered light from reaching the detector. The most significant effect of this process is that the dye will not absorb the 630-nm photons and there will be no photobleaching except at the point of excitation. The only disadvantage of the two-photon excitation process is the cost of the laser, which is prohibitive for most laboratories. Denk et al. (1990) used a colliding-pulse, mode-locked dye laser producing a stream of pulses with a pulse duration of ∼100 fsec at a repetition rate of ∼80 MHz. More recently, titanium:sapphire lasers pumped by argon-ion lasers have been used. These lasers are somewhat easier to operate but are still very expensive. The development and application of multiphoton excitation will be discussed in detail in a future unit. The field of confocal microscopy will benefit from the ongoing development of new fluorescent dyes for both flow and image cytometry, particularly ones that excite at longer wavelengths. In three-dimensional imaging particularly (via optical sectioning), the continued development of faster computers with larger storage capacities will make it possible to develop faster and better data-processing packages. There is still a demand for new objective lenses with higher numerical apertures, longer working distances, and higher transmission in the visible as well as UV ranges. Better correc-
tion for chromatic effects is needed, particularly for laser scanning systems; perhaps an objective lens will be developed that can be tuned to the wavelength being used.
LITERATURE CITED Amos, W.B., White, J.G., and Fordham, M. 1987. Use of confocal imaging in the study of biological structures. Appl. Opt. 26:3239-3243. Boyde, A. 1985. Stereoscopic images in confocal ( tan d e m s ca n n i n g ) micro scopy. Science 230:1270-1272. Brakenhoff, G.J., and Visscher, K. 1990. Novel confocal imaging and visualization techniques. In Transactions of the Royal Microscopical Society, Vol. 1. (H.Y. Elder, ed.) pp. 247-250. Adam Hilger, Bristol. Cavanagh, H.D., Petroll, W.M., and Jester, J.V. 1993. The application of confocal microscopy to the study of living systems. Neurosci. Biobehav. Rev. 17.4:483-498. Cogswell, C.J. and Sheppard, C.J.R. 1990. Confocal brightfield imaging techniques using an on-axis scanning optical microscope. In Confocal Microscopy (T. Wilson, ed.) pp. 213-243. Academic Press, London. Cogswell, C.J. and Sheppard, C.J.R. 1992. Confocal differential interference contrast (DIC) microscopy: Including a theoretical analysis of conventional and confocal DIC imaging. J. Microscopy 165:81-101. Denk, W., Strickler, J.H., and Webb, W.W. 1990. Two-photon laser scanning fluorescence microscopy. Science 248:73-76. Draaijer, A. and Houpt, P.M. 1987. A real-time confocal laser scanning microscope (CLSM). Proc. SPIE 809:85-88. Florijn, R.J., Slats, J., Tanke, H.J., and Rapp, A.K., 1995. Analysis of antifading reagents for fluorescence microscopy. Cytometry 19:177-182. Kino, G.S. 1989. Efficiency in Nipkow disk microscopes. In The Handbook of Biological Confocal Microscopy (J. Pawley, ed.) pp. 93-97. IMR Press, Madison, Wis. Matsumoto, B. 1993. Cell Biological Applications of Confocal Microscopy. Methods in Cell Biology, Volume 38. Academic Press, New York. Minsky, M. 1957. Microscopy Apparatus. U.S. Patent no. 3013467 (awarded 1961). Minsky, M. 1988. Memoir on inventing the confocal scanning microscope. Scanning 10:128-138. Nipkow, P. 1884. German Patent no. 30105. Petran, M., Hadravsky, M., Egger, D., and Galambos, R. 1968. Tandem-scanning reflected light microscope. J. Opt. Soc. Amer. 58:661-664. Wilson, T. (ed.) 1990. Confocal Microscopy. Academic Press, London.
Image Cytometry Instrumentation
2.8.11 Current Protocols in Cytometry
Supplement 5
KEY REFERENCES Elder, H.Y. (ed.) 1990. Transactions of the Royal Microscopical Society, Vol. 1. Adam Hilger, Bristol. Contains papers presented at Micro 90, a conference of the Society held in 1990 at Shortlands, Hammersmith, London. The 21 chapters cover virtually all aspects of microscopy. In particular, it contains excellent papers on the physics and configuration of confocal microscopes, as well as on methods of staining and preparation of samples and on three-dimensional image processing and applications. Matsumoto, B. 1993. See above. Contains 13 chapters with extensive discussions of a variety of applications that utilize confocal microscopy.
Pawley, J. (ed.) 1989. The Handbook of Biological Confocal Microscopy. IMR Press, Madison, Wis. Consists of 19 papers covering the practical aspects of confocal microscopy; an excellent general reference to the topic. Wilson, T. 1990. See above. Contains 15 chapters covering all aspects of confocal microscopy, primarily from the theoretical point of view.
Contributed by Phillip N. Dean Livermore, California
Confocal Microscopy: Principles and Practices
2.8.12 Supplement 5
Current Protocols in Cytometry
Multiphoton Imaging Two-photon imaging burst on the scene in 1990 (Denk et al., 1990) as a method for laser scanning fluorescence microscopy, in which the absorption is confined to the plane of focus. This restriction diminishes photobleaching (Song, 1996) and its deleterious cytotoxic products, such as triplet oxygen. In this respect, it is superior to confocal microscopy, where the product of intensity × dwell time (i.e., the timeintegrated dose) of radiation is the same all through the specimen. While the extraordinary property of the two-photon beam—production of a pinpoint of light just where required—is easily demonstrated (Fig. 2.9.1), the application of the method has been delayed by the high cost of suitable lasers; at present there are only exciting glimpses of its potential in the hands of leading groups, rather than full utilization. In April 1995, White’s group in Madison demonstrated clear quantitative evidence of three-photon imaging effects (Wokosin et al., 1997); this and much subsequent work makes the title of this unit, multiphoton imaging, appropriate. It has become clear that the advantages of multiphoton imaging are greater than initially supposed, and include the ability to penetrate deep specimens with the longerwavelength exciting radiation, to excite characteristic autofluorescence at the equivalent of ultraviolet without killing the cell, and to use non-focusing detection, which is both more efficient than confocal detection and totally insensitive to chromatic aberration in the objective lens. This latter means that multiphoton imaging is likely to become the method of choice for colocalization studies. On the other hand, the infrared wavelengths normally employed are by no means harmless to the specimen.
BASIC PRINCIPLES Multiphoton excitation has been recognized and used as a spectroscopic tool since 1961 (Kaiser and Garret, 1961). It is a process unknown in nature except in stars, but can readily be produced in laser beams, where, at very high intensities, more than one photon can be absorbed by a molecule within a very short coincidence time characteristic of the type of molecule. The energy of these photons is summed as if a single photon of shorter wavelength (higher energy) had been absorbed, and the subsequent emission process and emission
Contributed by Brad Amos Current Protocols in Cytometry (1998) 2.9.1-2.9.11 Copyright © 1998 by John Wiley & Sons, Inc.
UNIT 2.9 spectrum are identical to those in the normal single-photon situation. The probability of ordinary absorption (i.e., of a single photon) is proportional to the effective cross-section presented by the molecule; the unit of molecular absorbance has the dimensions of area, i.e., length squared. With twophoton excitation, the probability is proportional to two hits on a definite area (i.e., length to the fourth power) × the coincidence time. The unit is named the Gompert Meyer (GM) after the physicist who predicted such effects in the 1930s. 1 GM = 10−50 cm4 sec. With three-photon excitation, the unit has the dimensions of length to the sixth power × time squared. The laser scanning microscope of Denk et al. (1990) depends on the fact that the absorption is proportional to the square of the intensity in the two-photon case, because of the need for two photons to arrive in a short time interval. In the laser microscope, the exciting radiation is focused into a cone, with the result that the absorption falls off very rapidly away from the focus, according to an inverse fourth–power rule, as shown in Figure 2.9.1. The fall-off is even more rapid with three-photon and higher order processes. Analysis (Williams et al., 1994) yields the perhaps unexpected result that, provided the illuminated volume is uniformly filled with fluorophores, the total fluorescent emission is independent of the numerical aperture (N.A.) of the objective lens. Unlike confocal microscopy, multiphoton imaging is quite effective with low-magnification lenses of moderate N.A., provided wide-angle detection is used (see below). The resolution in a two-photon microscope is determined entirely by the restriction of excitation mentioned above. Williams et al. (1994) showed that the lateral resolution is given by 0.37λ/n sinθ and the axial by 0.32λ/n sin2 θ /2, where λ is the exciting wavelength and n sin θ is the numerical aperture of the objective lens. These expressions suggest approximately twice-better resolution than a conventional microscope, but this is counterbalanced by the need to use a twice-larger wavelength. In practice, the resolution in two-photon micrographs looks similar to that of conventional epifluorescence micrographs, or very slightly inferior if a 1047-nm wavelength is used. A combination of multiphoton excitation
Image Cytometry Instrumentation
2.9.1 Supplement 6
and confocal detection is often suggested as a means of improving resolution, but this is seldom used, because of the loss of signal observed when the confocal aperture is inserted. The failure of this combination is interesting:
Multiphoton Imaging
it is probably due to scattering of the emitted light in the specimen or to chromatic aberration. So far, no microscope objectives have been designed to be achromatic over a two-fold range of wavelength except the reflective
Figure 2.9.1 Demonstration of the difference between single- and two-photon excitation. The cuvette is filled with a solution of a dye, safranin O, which normally requires green light for excitation. Green light (543 nm) from a continuous-wave helium-neon laser is focused into the cuvette by the lens at upper right. It shows the expected pattern of a continuous cone, brightest near the focus and attenuated to the left. The lens at the lower left focuses an invisible 1046-nm infrared beam from a mode-locked Nd-doped yttrium lanthanum fluoride laser into the cuvette. Because of the two-photon absorption, excitation is confined to a tiny bright spot in the middle of the cuvette.
2.9.2 Supplement 6
Current Protocols in Cytometry
Schwartzchild type; these are unsuitable for high-resolution work because of their poor point-spread function.
LASERS The laser must have a wavelength approximately twice that of the single-photon absorption peak. There is a great advantage in bunching the energy of the laser into short pulses, because of the square or cube law relating fluorescence output to intensity. In fact, for a given average power, provided there is no saturation of the fluorophore (from ground-state depletion or intersystem crossing), the output is inversely proportional to pulse length (twophoton) or proportional to the inverse square of pulse length (three-photon). An optimum pulse length appears to be 50 to 100 fs. If such pulses could be seen in a parallel beam, they would appear as pancakes of light 100 µm thick. Longer (picosecond) pulses have been used for imaging, but these are definitely suboptimal and apparently have so far not produced threephoton images. It is also advantageous for the pulses to be at a high repetition rate, on the order of 50 MHz, to secure rapid cycling of the
fluorophore, which must execute tens of cycles per microsecond to provide a strong signal. Even with such a high repetition rate, the interval between pulses is several thousand times the pulse length. These requirements are satisfied only by mode-locked lasers, where the pulse interval is simply twice the traverse time of the laser cavity. Some two-photon images of bright specimens have been obtained with continuous wave (CW) lasers (Booth and Hell, 1998), but only with high powers (e.g., 210 mW, compared with the 1-5 mW normally used). Although the original work of Denk et al. (1990) was performed with a dye laser, the advantages of tunable titanium-doped sapphire (Ti-sapphire) lasers were realized early (Curley et al., 1992) and these are now the most commonly used. Figure 2.9.2 shows the tuning range of the Spectra Physics Tsunami system (690 to 990 nm), which is attainable by using a series of mirror sets. Originally, the Ti-sapphire laser was pumped by a large argon laser, which required a three-phase electric power supply and much cooling water. Now, a compact diode laser emitting 5 W at 532 nm is used
from laser objective
iris diaphragm
longpass dichroic
shortpass scanning 13 dichroic mirrors
external photomultiplier
scan head
specimen
A1
eyepiece mirror
A3
12 collector lens
external photomultiplier
tube lens I1 A2 inverted microscope
Figure 2.9.2 Diagram showing the difference between two types of detection of multiphoton-excited fluorescence. In semi-confocal detection, a photomultiplier is used inside a confocal scan head, receiving a descanned beam from the specimen. The confocal iris is opened as far as possible, but still limits the signal because of scattering and chromatic effects. An external detector receives non-descanned light through simple and efficient collector optics. In the diagram, A1, A2, and A3 represent aperture planes, while I1, I2, and I3 are image planes. For simplicity, light rays from only a single point in the specimen are shown. The scattering specimen is a multipoint emitter, and the emission beams have an envelope that expands rapidly with distance from the objective (becoming several centimeters wide at I2). The best collection efficiency is achieved by using a large collector lens, with the focal length chosen to image the objective exit pupil at A1 (the narrowest part of the envelope) on the detector aperture. Although both types of detector are shown here, the long-pass dichroic directs the bulk of the emitted light into the external photomultiplier.
Image Cytometry Instrumentation
2.9.3 Current Protocols in Cytometry
Supplement 6
as the pump; a standard single-phase 110- or 220-V power supply and a small recirculating cooler are adequate (total heat dissipation <0.5 kW). If the pump and Tsunami lasers are put in line, the length of the system is 2 m. Similar laser systems with both fixed and tunable wavelength are now available from Coherent. White and colleagues pioneered the use of a fixed-wavelength, mode-locked laser based on diode pumping of a crystal of neodymiumdoped yttrium lanthanum fluoride (Nd-YLF). This laser is commercially available from Microlase. The wavelength is 1046 nm, which falls just inside the second window of infrared transmission of water (Fig. 2.9.3). The name “shoebox” aptly describes the size of this laser, which has proved robust and particularly well suited for certain types of work with living cells (see below). Similar lasers based on Nd-doped yttrium aluminum garnet (Nd-YAG) have been used, for example by P.C. Cheng’s group. Other lasers have been used for multiphoton imaging, including the diode-pumped chromium-doped lithium strontium aluminum fluoride (Cr-LiSAF; Fig. 2.9.3; UNIT 1.9) laser, although the power of current systems is inconveniently low. New compact lasers are being developed.
SETTING UP A MULTIPHOTON LASER LAB Certain points need to be observed in setting up a multiphoton laser lab. All the mode-locked
lasers are sensitive to vibration, dust, and temperature fluctuations. The cleaning of laser mirrors, preferably by means of lens tissue moistened with dry, high-grade methanol, cannot be avoided, but will be required less often if a recirculating air filter (e.g., Filtaire 200, TAAB Laboratory Equipment) is used. The temperature should be kept to limits of ± 1°C. An anti-vibration table should be used. A Ti-sapphire laser system operated between 900 and 1000 nm must be flushed with dry nitrogen to remove oxygen and water vapor, both of which have absorption lines in this region. In view of the high powers involved (e.g., 5-W pump power and 500-mW output), laser safety must be the prime consideration in a multiphoton lab, particularly if the setup is homemade and lacks the interlocks and shielding provided by commercial suppliers (BioRad is currently the only manufacturer licensed to supply multiphoton imaging systems). An infrared beam at this power is either invisible or looks like a dim red safelight; however, a parallel beam, such as can arise by accidental specular reflection, destroys regions of the human retina rapidly and permanently. As well as observing all relevant regulations with regard to safety goggles, access, and beam height, users should check that inverted microscopes are permanently modified to ensure that the user cannot look down the optic axis with the objective swung out of the path.
1.00 Cr-LiSAF 860
Absorption coefficient (cm-1); dotted line)
0.50
Nd-YLF 1047
H2O
Ti-sapphire 0.10 0.05
0.01 700
800
900
1000
1100
Wavelength (nm)
Multiphoton Imaging
Figure 2.9.3 Relationship between the wavelengths of the principal types of mode-locked laser (solid line) and the absorption spectrum of water (dashed line). The water spectrum is replotted from Svoboda and Block (1984).
2.9.4 Supplement 6
Current Protocols in Cytometry
An infrared viewer (e.g., Find-R-Scope model 89000) is essential for setting up multiphoton systems. A cheap surveillance-type charge-coupled device (CCD) camera is equally good as an infrared viewer for checking the position of the beam on mirrors, and is even better when two hands are needed for adjustments. A power meter (e.g., Spectra Physics model 407a) is essential. With the Nd-YLF laser, a simple 20-MHz, two-channel oscilloscope is needed to check the modelocking. With tunable Ti-sapphire systems, a spectrum analyzer (e.g., Rees E201) is essential for measuring the output wavelength and checking the pulse width, though an experienced user might be able to make do with a simple diffraction grating blazed for infrared. (Wavelength is then indicated by diffraction angle, mode-locking by broadening of the diffracted spot.) It is convenient to divert a fraction of the beam to the spectrum analyzer by reflection off a plain glass plate on a swing-out mount. The plate should be at least 0.7 mm thick to avoid interference effects that superimpose a ripple on the spectrum. The laser manufacturers recommend or supply pulse compressors based either on gratings (Nd-YLF) or prisms (Ti-sapphire). Compensation for the effects of the microscope objective and other lenses by negative dispersion is necessary, especially with <50-fs pulses. Autocorrelation methods provide direct measurement of pulse length; it is even possible to measure this within the specimen (see Brakenhoff et al., 1995a), but this is not necessary in the course of most biological experiments.
THE SCAN HEAD In commercially available multiphoton equipment (Bio-Rad) the scan head design is based on that originally produced for confocal use. Fortunately, the scanning apparatus consists in this case of a reflective relay system that is intrinsically achromatic, but it is necessary for all the mirrors in the excitation path to have high infrared and visible reflectance. The normal internal photomultipliers can be and have been extensively used as detectors, with the variable iris fully open. In this mode of operation, the laser light is caused to scan over the specimen by oscillating mirrors; the emitted light is reflected in the reverse direction by the same set of mirrors, and is thus descanned before passing to the detector. Large improvements in signal strength can be gained by picking off the signal with a chromatic reflector before it passes back into the scan head (Fig. 2.9.2). This is termed “non-descanned detec-
tion.” Measurements with a Nd-YLF system and a Nikon 60× N.A. 1.4 objective show a minimum improvement factor of almost 3-fold with this strategy, even with a homogeneous non-scattering specimen (a solution of a fluorescent dye). The improvement factor is much greater with normal scattering specimens. It was found that one particular objective (20× Plan DIC N.A. 0.4) was fifteen times better when used with non-descanned detection, and this was traced to longitudinal chromatic aberration (this objective was never intended by Nikon to be achromatic over the range 500 to 1046 nm). A quantitative comparison of semi-confocal and non-descanned detection is given by Wokosin et al. (1998). Clearly, non-descanned detection is the method of choice, with the slight drawback that full room lights cannot be used during imaging, because the external photomultipliers discriminate less well against ambient light. In the commercial apparatus available from Bio-Rad (designed by the author), the longpass chromatic reflector (Fig. 2.9.2) can be swung out, permitting confocal operation. When it is in, the optics are optimized to collect the maximum signal by focusing the back focal plane of the objective onto the aperture of the special proprietary prismatic enhancer of the photomultiplier, in such a way that the signal varies minimally with scan position across the field. The enhancement depends on multiple reflection of the signal beam within the glass envelope of the photomultiplier tube (see Mansberg and Ohringer (1969) for details of this effect). Additional dichroic cubes can be inserted for division of the signal beam between multiple channels. For setting up the scan head, it is useful to have a pilot beam of visible light, which can be combined and introduced coaxially with the infrared beam. This facilitates the initial setup of the scan head and microscope and may be used for confocal imaging for comparative purposes. The installation engineer can, for example, insert a highly reflective specimen and then check the alignment of the signal beam path by direct inspection before switching to infrared. It is also useful to have a durable specimen that shows bright two-photon fluorescence with the Nd-YLF laser and throughout the tuning range of the Ti-sapphire laser. Such a specimen is imaged in Figure 2.9.4. With this, the scanned area of the specimen can be seen as a bright yellow spot plainly visible to the naked eye, a useful check that all is well with the mode-lock-
Image Cytometry Instrumentation
2.9.5 Current Protocols in Cytometry
Supplement 6
ing. The specimen shown in Figure 2.9.5 can be used to measure the longitudinal chromatic aberration of objectives between the infrared and visible wavelengths.
OF WHAT USE IS MULTIPHOTON IMAGING? A full assessment the value of this technology is not possible in this unit; instead, some established applications will be listed briefly. All fluorescent stains in common use in microscopy can be employed, depending on the laser type (see Table 2.9.1). Simultaneous excitation
Multiphoton Imaging
of multiple stains is easy. Provided the stains can be separated by their emission spectra, colocalization studies can be performed without fear of false single positives due to chromatic aberration, a common problem in confocal microscopy. As Figure 2.9.6 shows, multiphoton images look like confocal images, but can often be obtained from structures deeper in the specimen. The first few years of multiphoton imaging yielded few new biological results, but examples of work with living cells, which could
Figure 2.9.4 Test specimen. The bright matrix is a solution of Nile Red in a hydrophobic resin mountant (Fluoromount: BDH-Gurr). This shows bright yellow multiphoton fluorescence at all wavelengths available from Nddoped yttrium lanthanum fluoride or titanium-sapphire lasers. The black objects are nonfluorescent calcareous spicules added as focus markers. Such spicules can be obtained readily by boiling sea cucumbers (available in dried form from Chinese food stores) in concentrated potassium hydroxide. Field length = 250 µm.
2.9.6 Supplement 6
Current Protocols in Cytometry
not be carried out by other means, are steadily appearing. In developmental biology, White and colleagues (Squirell et al., 1996) used the Nd-YLF laser (1047 nm) and the dye MitoTracker (Molecular Probes) to study mammalian embryos. They collected repeated three-dimensional datasets from normally light-sensitive hamster embryos, and are confident that an imaged
Table 2.9.1
embryo will eventually be reared to term. With the nematode Caenorhabditis elegans, this group has shown that the dye FM464, used as a marker of cell outlines, can be imaged almost indefinitely, making lineage studies easier than ever before (White et al., 1997). In sea urchins, fate-mapping has been simplified by an elegant combination of optical uncaging of fluorescein and two-photon imaging of the subsequent de-
Some Dyes That Have Been Used for Multiphoton Microscopy
FM 464a GFP (wild type and S65T) Hoechst 33258 Hoechst 33342 INDO-1 Lucifer Yellow MitoTracker Rosaminea Nile Reda Oregon Greena Propidium iodidea Safranin Oa Thioflavin S Texas Reda TRITCa
AMCA Bodipy Cascade Blue Calcium Crimsona Calcium Greena Congo Reda Coumarin 307 Di-A, Di-I, Di-Oa Dansyl hydrazine DAPI (two- and three-photon)a Evans Bluea FITCa Flavin autofluorescence Fluo 3a aDyes that are well-excited by the YLF laser at 1046 nm.
546 nm light for scanned transmission image cyanoacrylate cement
coverslip
slide
aluminum film with holes
1047 nm light for scanned semi-confocal reflection image
Figure 2.9.5 Test specimen for measuring longitudinal chromatic aberration in objective lenses used in multiphoton systems. The aberration in microns is the distance between two focus positions: that for reflection imaging using the fundamental wavelength and that (usually closer to the objective) for scanned transmission imaging in green light of the holes in the aluminum film, using an internal (semi-confocal) photomultiplier for both measurements. The aluminum film is made in a vacuum evaporator and the holes by dabbing the film with a dry paintbrush.
Image Cytometry Instrumentation
2.9.7 Current Protocols in Cytometry
Supplement 6
Multiphoton Imaging
Figure 2.9.6 Comparison of the single-photon confocal image (above) and the two-photon image (below) of a waterflea (Ceriodaphnia sp.) fixed in ethanol and stained with eosin. The confocal image was obtained with 514-nm excitation; note that little structure is visible in the interior, probably because of absorption of green light by the exoskeleton. The two-photon image, obtained with 800-nm excitation from a titanium-sapphire laser, shows much internal structure, including muscles and lenses of the compound eye. Field width = 300 µm.
2.9.8 Supplement 6
Current Protocols in Cytometry
velopment (Summers et al., 1996). The S65T variant of GFP can be imaged with the YLF laser, possibly as a result of three-photon effects, but GFPs are more easily studied by two-photon with the Ti-sapphire laser. Potter et al., (1996) have shown that this technique, using GFP, is superior to confocal microscopy in studies of axonal growth in the visual system of Drosophila using GFP. Kohler et al. (1997) used GFP to demonstrate unsuspected connections between chloroplasts in Arabidopsis. Living brain tissue is difficult to image with a confocal microscope because of scattering. Using two-photon microscopy and nonfocusing detection, Denk, Svoboda, Juste, and coworkers have obtained images of unparalleled clarity, some using the Cr-LiSAF laser (Svoboda et al., 1997). They have been able to resolve synaptic spines at depths greater than 100 µm in rat and turtle preparations. Visible-range calcium-ion probes were used in the above work. Several groups have begun to exploit the precise three-dimensional positioning of the probe in two-photon excitation to improve measurement of calcium-ion level. Studies with the emission ratio dye indo-1 show differences in its behavior with single- and two-photon excitation. The dye fluoresces less brightly when calcium is bound to it, with the result that the isosbestic point in two-photonexcited emission is 405 nm (c.f. 450 nm in single-photon). The emission ratio is therefore best measured between 405 and 460 nm (Sako et al., 1996). Among the most exciting biological results are the two- and three-photon images of autofluorescence (i.e., emission from unstained tissue). These images are totally new: they could not be obtained before because of UV cytotoxicity, poor UV penetration, and high internal scattering and absorption. At Cornell, the neurotransmitter serotonin has been studied by three-photon excitation (Maiti et al., 1997), the redox state of the cornea has been measured (Piston et al., 1994), and elastin and collagen fibers have been imaged clearly in intact mammalian skin (W.R. Zipfel, pers. comm.).
PROBLEMS The biggest problem with multiphoton imaging is the high price of the laser, but technical problems also exist. A focused infrared beam has direct biological effects. In attempts to use 200-fs pulses at 640 nm to open nitrophenyl and dimethoxy nitrobenzyl cages, researchers have found that destruction of motor end plates occurs at radia-
tion doses too low for uncaging (D. Ogden and N. Kiskin, NIMR, pers. comm.). Experimental variation of power and time suggested twophoton rather than single-photon damage. In the author’s laboratory, both Ti-sapphire and YLF beams, at a power of ∼10 mW, can make graphite particles mounted on a microscope slide incandescent, so that they emit white light as they burn. Melanin granules, often present in embryos, are vaporized. Not all pigments are lethal, however: cyclosis of chloroplasts in Elodea continues for 90 min under continuous two-photon imaging by chlorophyll autofluorescence at 1047 nm, as it does also under confocal conditions with 514-nm excitation (unpub. observ.). The power of the infrared beam must be controlled carefully to suit the specimen. If reflective neutral density filters are used, they should be angled to avoid back-reflection of the beam into the laser, which can cause costly damage. A spectral problem is illustrated in Figure 2.9.7. Rhodamine and fluorescein cannot be separated by choosing appropriate wavelengths for two-photon excitation because the two-photon spectra overlap too much. This is particularly serious because the emission spectrum of fluorescein extends far into the red and overlaps with that of rhodamine, making separation impossible by emission windowing. Multiple labelling in two-photon requires new combinations of stains with nonoverlapping emission spectra.
THE FUTURE The rapid increase in the use of multiphoton lasers, which are probably now in use in 50 laboratories, guarantees that the present steady expansion of applications will continue. New dyes with enhanced multiphoton effects will undoubtedly emerge (see Cheng et al., 1998), perhaps from existing military work on eye protection. It is hoped that microscope designers will aid multiphoton development by adapting the microscope body for more efficient nonfocusing detection. For this form of detection, it would also be worth designing special objectives optimized for high N.A., flat field, and high transmission, but without any need for chromatic correction. Brakenhoff et al. (1995b) were the first to successfully implement two-photon, videorate scanning in a slit-scanning microscope. They showed that a pulse laser with a much lower repetition rate than those now used (ide-
Image Cytometry Instrumentation
2.9.9 Current Protocols in Cytometry
Supplement 6
Normalized linear scale
fluorescein
rhodamine B
700
800
900
1000
Wavelength (nm)
Figure 2.9.7 Two-photon absorption spectra of fluorescein and rhodamine. Data are from Xu and Webb (1996), replotted on a linear scale.
Multiphoton Imaging
ally one pulse per video line), but delivering similar average powers, used with a slit-scanning microscope can emit enough energy per pulse to produce two-photon excitation all along a line. Unfortunately, as soon as more than one point is illuminated in a scanning microscope, it becomes necessary to use a camera as detector, and cross-talk between pixels may spoil the image. Such cross-talk could arise from chromatic aberration, but a more serious cause is scattering of the emitted light within the specimen; indeed, Brakenhoff’s group found it necessary to use confocal detection. A multispot, multiphoton microscope has recently been described (Bewersdorff et al., 1998). Future work is likely to combine spectroscopic analysis with multiphoton excitation. Autofluorescence could clearly be analyzed this way. It remains to be seen how much more information can be gained from analysis of the autofluorescence of cells in the near-UV and blue regions. The full potential of multiphoton imaging will not be realized until the biochemical mechanisms by which cells protect themselves
from bright light are taken into account (Emerit, 1993). Anaerobic cells may turn out to be of particular value because of their light resistance. The Elodea example shows that it may be worth searching for species whose pigments do not make them infrared sensitive.
LITERATURE CITED Bewersdorff, J., Pick, R., and Hell, S.W. 1998. Multifocal and multiphoton microscopy. Optics Lett. In press. Booth, M.J. and Hell, S.W. 1998. Continuous-wave excitation, two-photon fluorescence microscopy exemplified with the 647-nm ArKr laser line. J. Microsc. 190:298-304. Brakenhoff, G.J., Muller, M., and Squier, J. 1995a. Femtosecond pulse-width control in microscopy by two-photon absorption correlation. J. Microsc. 179:253-260. Brakenhoff, G.J., Squier, J., Norris, T., Bliton, A.C., Wade, M.H., and Athey, B. 1995b. Real-time two-photon confocal microscopy using a femtosecond, amplified Ti-Sapphire system. J. Microsc. 181:253-259. Cheng, P.C., Pan, S.J., Shih, A., Kim, K.-S., Liou, W.S., and Park, M.S. 1998. Highly efficient upconverters for multiphoton fluorescence microscopy. J. Microsc. 189:199-212.
2.9.10 Supplement 6
Current Protocols in Cytometry
Curley, P.F., Ferguson, A.I., White, J.G., and Amos, W.B. 1992. Application of a femtosecond selfsustaining mode-locked Ti-sapphire laser to the field of laser scanning confocal microscopy. Optical Quantum Electr. 24:851-859. Denk, W., Strickler, J.H., and Webb, W.W. 1990. Two-photon laser scanning fluorescence microscopy. Science 248:73-76. Emerit, I. 1993. Reactive oxygen species, photosensitised skin reactions and the protective effect of superoxide dismutase. In Active Oxygens, Lipid Peroxisomes and Antioxidants. (K. Yagi, ed.) pp. 217-223. CRC Press, Boca Raton, Fla. Kaiser, W. and Garret, C.J.B. 1961. Two-photon excitation in CaF2:Eu2+. Physiol. Rev. Lett. 7:229-231. Kohler, R.H., Cao, J., Zipfel, W.R., Webb, W.W., and Hanson, M.R. 1997. Exchange of protein molecules through connections between higher plant plastids. Science 276:2039-2042. Maiti, S., Shear, J.B., Williams, R.M., Zipfel, W.R., and Webb, W.W. 1997. Measuring serotonin distribution in live cells with three-photon excitation. Science 275:530-532. Mansberg, H.P. and Ohringer, P. 1969. Design considerations for electronic and electromechanical flying-spot scanners. Ann. N. Y. Acad. Sci. 157:546. Piston, D.W., Masters, B.R., and Webb, W.W. 1994. Three-dimensionally resolved NAD(P)H cellular metabolic redox imaging in the in situ cornea with two-photon excitation laser scanning microscopy. J. Microsc. 178:20-27. Potter, S.M., Fraser, S.E., and Pine, J. 1996. The greatly reduced photodamage of two-photon microscopy enables extended three-dimensional time-lapse imaging of living neurones. Scanning 18.3:147. Sako,Y., Sekihato, A., Yanagisawa, Y., Yamamoto, M., Shimada, Y., Ozaki, K., and Kusumi, A. 1996. Comparison of two-photon laser scanning microscopy with UV-confocal laser scanning microscopy in three-dimensional calcium imaging using the fluorescence indicator INDO-1. J. Microsc. 185:9-20.
Song, L. 1996. Photobleaching kinetics of fluorescein in quantitative fluorescence microscopy. Doctoral Thesis. Leiden University, Leiden, The Netherlands. Squirrell, J.M., Wokosin, D.L., Centonze, V.E., White, J.G., and Bavister, B.D. 1996. Mitochondrial dynamics of living hamster embryos imaged by two-photon excitation. Mol. Biol. Cell 7:644c. Summers, R.G., Piston, D.W., Harris, K.M., and Morrill, J.B. 1996. The orientation of first cleavage in the sea urchin embryo, Lytechinus variegatus, does not specify the axes of bilateral symmetry. Dev. Biol. 175:177-183. Svoboda, K. and Block, S.M. 1984. Biological applications of optical forces. Annu. Rev. Biophys. Biomol. Struct. 23:247-285. Svoboda, K. , Denk, W., Kleinfeld, D., and Tank, D.W. 1997. In vitro dendritic calcium dynamics in neocortical pyramid neurones. Nature 385:161-165. White, J.G., Centonze, V., Wokosin, D., and Mohler, W. 1997. Using multiphoton microscopy for the study of embryogenesis. Proc. Microsc. Microanal. 3:307-308. Williams, R.M., Piston, D.W., and Webb, W.W. 1994. Two-photon molecular excitation provides intrinsic 3-dimensional resolution for laserbased microscopy and microphotochemistry. FASEB J. 8:804-813. Wokosin, D.L., Amos, W.B., and White, J.G. 1998. Detection sensitivity enhancements for fluorescence imaging with multi-photon excitation microscopy. I.E.E.E. Proceedings. In press. Wokosin, S.L., Centoze, V.E., Crittendon, S., and White, J.G. 1997. Three-photon excitation imaging of biological specimens using an all-solidstate laser. Bioimaging 4:208-214. Xu, C. and Webb, W.W. 1996. Measurement of two-photon excitation cross-sections of molecular fluorophores with data from 690 to 1050 nm. J. Opt. Soc. Amer. B. 13:481-491.
Contributed by Brad Amos MRC Laboratory of Molecular Biology Cambridge, United Kingdom
Image Cytometry Instrumentation
2.9.11 Current Protocols in Cytometry
Supplement 6
Scanning Laser Cytometry Flow cytometry has come into wide use at least in part because manufacturers have presented ready-made solutions to instrumentation problems that few users would have tried to solve on their own. However, there are many tasks to which flow cytometers are unsuited, such as measurement of attached cells, repeated measurements of a single cell over time, and localization of probes in or on cells. It is difficult for a flow cytometer to deal with a single sample comprising only a few hundred cells, and this problem is compounded when there is a need to analyze multiple samples of a few hundred cells each in plates that may contain as many as 1536 individual wells. Devices such as confocal microscopes and microscopebased imaging systems allow high-resolution analysis and provide flexibility, but typically have low sample throughput and often require a high level of interaction with a technically sophisticated operator. Scanning laser cytometers, which can make low-resolution, multiparameter optical measurements of light scattering and fluorescence similar to those made by flow cytometers, provide reasonably high sample throughput and may, depending on the purpose for which they are designed, either allow for considerable flexibility and operator interaction or permit analysis of large numbers of samples without any need for an operator in attendance.
OPERATING PRINCIPLES Optics and Mechanics: Light Collection and Detection In a typical flow cytometer, cells flow in suspension through the focus of a laser beam. In a scanning laser cytometer, the cells are in fixed locations on a specimen carrier, which may be a microscope slide or a capillary tube, placed on a mechanical stage in the focal plane of the laser beam. The stage is moved in one direction in this plane by computer-controlled motors. The beam is deflected in the perpendicular direction in the focal plane by a mirror mounted on the shaft of a low-inertia, computer-controlled galvanometer. In both flow and scanning laser cytometers, light scattered by and emitted from cells is collected by microscope lenses or their equivalent, passed through a system of dichroics and optical filters to select different wavelengths of interest, and
Contributed by Howard M. Shapiro Current Protocols in Cytometry (2004) 2.10.1-2.10.9 Copyright ©2004 by John Wiley & Sons, Inc.
UNIT 2.10
detected by photodiodes or photomultiplier tubes (PMTs); the electrical current outputs of the detectors are then converted into voltages by preamplifiers.
What Signals Represent In the flow cytometer, the output of a detector/preamplifier is a series of pulses, with the random spacing between the pulses reflecting the random arrival of cells at the measurement point; information about the amount of lightscattering or fluorescent material in the cell may be derived from the amplitude (height), integral (area), or width of the associated pulse. In most instruments, analog and hybrid electronics are used to capture and briefly store pulse height, area, and width values, and to initiate the digital conversion of these analog values to allow further analysis by a digital computer. Signals are digitized at irregular intervals, again reflecting random arrival of cells. In the scanning laser cytometer, the output of a detector/preamplifier is a raster scan of an area of the specimen carrier, and digitization is typically done at regular intervals, producing as an initial output a matrix of values, each of which represents one element or pixel of an image. Data from a single cell are represented by values of a number of contiguous pixels; the number may be on the order of 10 in lowerresolution systems and on the order of 100 in higher-resolution systems. However, although the effective pixel spacing in the image from a scanning laser cytometer may be as small as that in a fairly high-resolution digitized microscope image (∼0.5 µm), the spatial resolution is lower because the laser beam diameter at the focus is substantially larger than the pixel spacing.
Laser-Beam Characteristics Lasers are advantageous as light sources for cytometry because the entire output of the laser can be focused into a very small spot (see UNITS 1.6 & 1.10 for additional discussion). However, the fundamental physics of lasers dictates that the smaller the focal spot, the more rapidly the beam diameter diverges with distance along the beam axis on either side of the focal plane. This principle is exploited in confocal (UNIT 2.8) and multiphoton microscopy (UNIT 2.9), where spot sizes are typically ≤1 µm; the depth of focus of the spot is a fraction of a micrometer, and the
Image Cytometry Instrumentation
2.10.1 Supplement 28
CCD camera
scanner
scan lens
lasers Ar He-Ne
specimen slide
objective lens motorized stage
green orange PMTs red far red
scatter sensor white light source
Figure 2.10.1 Schematic of the optics of the LSC laser scanning cytometer (courtesy of CompuCyte Corporation).
Scanning Laser Cytometry
illumination intensity per unit area falls off rapidly above and below the focal plane due to the high beam divergence. This allows the collection of a high-resolution fluorescence image of a thin section of the specimen, and the reconstruction of three-dimensional detail from a succession of such images. The illumination optics of a flow cytometer are typically designed to form a focal spot 50 to 120 µm wide—to provide uniform illumination over the width of the core sample stream— and 5 to 50 µm high; shorter beam heights allow size information about cells to be derived from pulse height and width, but may complicate measurement of forward scatter at very small angles because of the large beam divergence. In the critical dimension, that of beam width, depth of focus is several hundred micrometers. Spatial resolution in a flow cytometer varies from low, when beam heights are small compared to cell sizes, to none, when beam heights are several times larger. In a scanning laser cytometer, the beam focal spot, or waist, size (typically 2.5 to 10 µm in diameter) is intermediate between that in a flow cytometer and that in a confocal microscope. Depth of focus is on the order of 100 µm, allowing reasonably uniform signals to be collected from a thin layer of cells on a slide or in a capillary. Although the image contains data points at intervals corresponding to the pixel spacing (0.5 to 4 µm), the effective optical resolution is lower, because the spot size (2.5 to 10 µm in various devices) is larger than the pixel spacing.
Representative Apparatus The optics of two representative instruments, the LSC Laser Scanning Cytometer (Kamentsky et al., 1997; CompuCyte Corporation) and the IMAGN 2000 (Dietz et al., 1996; Biometric Imaging, now part of BD Biosciences), are shown in schematic form in Figures 2.10.1 and 2.10.2, respectively (see SUPPLIERS APPENDIX for manufacturer contact information). The LSC, which is approximately the size of a research-grade benchtop flow cytometer, is one of the more elaborate scanning laser cytometers currently available, allowing measurements of green (530-nm), orange (580-nm), and red (610-nm) fluorescence to be made using standard argon-laser excitation at 488 nm; far-red (670-nm) and near-infrared (750nm) fluorescence can be measured using either 488-nm excitation or excitation at 633 nm from an optional He-Ne laser. Blue (460-nm) fluorescence can be measured using excitation from an optional 400-nm violet diode laser; a green (543-nm) He-Ne laser option is also available. The IMAGN 2000 and other systems developed by Biometric Imaging use 633-nm He-Ne laser excitation, and measure fluorescence at 670 and 690 nm; the focal-spot diameter of the laser beam is 8 µm, and the pixel spacing, determined by the scan-line digitization rate and the step size used for stage motion, is 4 µm. The Biometric instruments have lower optical resolution than does the LSC, but take proportionally less time to scan an area of a specimen.
2.10.2 Supplement 28
Current Protocols in Cytometry
long-pass filter focusing lens dichroic filter
aperture
laser line filter
dichroic filter He-Ne laser
scan mirror
PMT 0 PMT 1 (shorter λ) (longer λ)
capillary focusing objective fast scan axis slow scan axis
Figure 2.10.2 Schematic of the optics of the IMAGN 2000 volumetric capillary cytometer (courtesy of Biometric Imaging).
In the LSC, working at high magnification, a typical cell image might contain a hundred pixels; in a lower-resolution device, such as the IMAGN 2000, a cell image may contain only ten to twenty pixels. Higher-resolution images are appropriate and necessary for such analytical tasks as localization and counting of fluorescent spots from hybridized chromosome probes in nuclei; lower-resolution images are adequate for such applications as detection and quantification of surface immunofluorescence. A scan-line image from an IMAGN 2000 is shown in Figure 2.10.3. In the LSC, the beam from a laser (typically an air-cooled 488-nm argon-ion laser or 633nm He-Ne laser) is reflected from a galvanometer-driven mirror through a set of lenses that maintain its divergence and diameter, and then from a long-pass dichroic filter through a microscope objective, producing a spot 2.5 to 10 µm in diameter (depending on the magnification of the objective) in the plane of the specimen. The galvanometer is driven by a “sawtooth” signal, whose amplitude alternately increases linearly with time and rapidly returns to its initial value. During the linear portion of the signal, the focused laser spot travels along a line in the specimen plane. The length of this line varies between 171 and 685 µm, depending on objective magnification. Fluorescence from objects in the path of the beam is collected by the microscope objective, passes through the
dichroic, is further separated into spectral bands by additional dichroics, and is then detected by PMTs. The photomultiplier signal is digitized at regular intervals, yielding a series of values that correspond to pixel intensities at intervals of 0.5 µm along the scan line. At the end of each scan line, the microscope stage or specimen carrier is moved 0.5 µm in a direction perpendicular to that of the scan, and the process is repeated. Thus, the apparatus produces digitized images of fluorescence at one or more wavelengths from a rectangular area of the specimen. The LSC also incorporates a forward-scatter measurement, using the microscope’s substage condenser to collect light and a photodiode to detect scattered light, with an obscuration bar similar to that typically used in a flow cytometer positioned to block the transmitted laser beam. Large-angle scatter measurements have been implemented in experimental devices but are currently not available in production instruments.
Volumetric Capillary Cytometry A scanning laser cytometer scans a known area of a specimen. If the thickness of the specimen is uniform and known, the volume corresponding to the scanned area can be calculated, and a count of cells or other objects per unit volume of specimen can thus be derived. This technique represents a high-tech analog of
Image Cytometry Instrumentation
2.10.3 Current Protocols in Cytometry
Supplement 28
Figure 2.10.3 Image data represented as oscilloscope traces of serial scan lines from the IMAGN 2000 volumetric capillary cytometer. Fluorescent cells are represented by peaks; the “plateau” on which the peaks lie represents fluorescence from free dye in the capillary. Courtesy of Biometric Imaging.
cell counting by visual observation using a hemacytometer, a specially designed slide with rulings demarcating the area to be observed and provision for maintaining a constant thickness of sample between the slide and its coverslip (APPENDIX 3A). The IMAGN 2000 performs a variety of clinical cell-counting procedures in capillaries a few tens of millimeters long, with rectangular cross-sectional dimensions ≤1 mm on a side; the technique has been described as volumetric capillary cytometry, or VCC (Dietz et al., 1996). Making the depth of focus of the laser spot somewhat smaller allows a “virtual volume” to be defined, effectively employing the methodology of confocal microscopy at the lower resolution of scanning laser cytometry.
DATA ANALYSIS Image Segmentation Scanning Laser Cytometry
While the data collected in a flow cytometer unambiguously present values of one or more
parameters associated with each cell analyzed, data from a scanning laser cytometer, like cell image data, must be processed to determine which collections of contiguous pixel values belong to a particular cell or intracellular region. The process of image segmentation by which this is accomplished in the LSC can be appreciated with the aid of Figure 2.10.4, which shows fluorescence image data from lymphocytes stained with propidium iodide; the cell nuclei are visible as dark areas indicating relatively high-intensity fluorescence. Three concentric contour lines are drawn around each nucleus. The contours are drawn automatically by the analysis software, taking into account three values provided by the operator. The first value is the threshold level. Points above the threshold level will be considered to be inside a cell or a cellular structure of interest, in this case the nucleus. The innermost contour line, the threshold contour, defines collections of contiguous points falling above threshold level.
2.10.4 Supplement 28
Current Protocols in Cytometry
Figure 2.10.4 Digitized image of propidium-stained lymphocyte nuclei produced by the LSC, showing threshold, integration, and background contours (see text). Courtesy of CompuCyte Corporation.
However, generally some of the area of the structure of interest is likely to be represented by exterior pixels of somewhat lower intensity than those in the center. If one attempted to calculate the integral of nuclear fluorescence by summing intensity values for points within the threshold contour, the exterior pixels of the nucleus would be excluded, and the calculated fluorescence value would be too low. On the other hand, lowering the threshold value in an attempt to include all points in the nucleus within the threshold contour would greatly increase the probability that two nuclei close to one another would be identified as a single nucleus, or that particles other than nuclei would be identified as nuclei. In principle (and in the author’s practice, some 35 years ago!), it would be possible for the operator to trace the desired contour for integration on each nucleus in an image; however, this would reduce sample throughput to one or two cells per minute. The software used with the LSC provides a compromise that allows fairly accurate fluorescence measurement with substantially higher sample throughput. The operator defines a pixel spacing between the threshold contour and an integration contour; this contour, between the innermost and the outermost, is then automatically drawn by the software and used to compute the integral of the parameter of interest for each cell. This allows pixels within the structure of interest that may be below threshold level to be included.
In order to obtain accurate values of measured parameters, it is necessary to subtract background values; these may be different in different areas of the specimen, due, for example, to variations in background light intensity and the concentration of interfering substances. The operator may therefore set pixel spacing to be used between the integration contour and the outermost contour, which is the background contour. Points along this contour are used to compute the background value to be subtracted from each pixel during the integration process. The operator may also elect to omit local background measurement, and, instead, define a single background level or select an area that can be used to derive a background level for the entire specimen.
Derivation of Parameter Values for Individual Cells The software for the LSC computes values of a number of parameters associated with each cell, and can store these data in files compatible with the Flow Cytometry Standard (Seamer et al., 1997; UNIT 10.2). Some of the data are directly analogous to those obtained by flow cytometry. The integrated value—i.e., the sum of the fluorescence or scatter values for each pixel within the integration contour—corresponds to the pulse integral or area in flow cytometry, while the maximum pixel value within the integration contour corresponds to the pulse height. The area within the threshold contour can be con-
Image Cytometry Instrumentation
2.10.5 Current Protocols in Cytometry
Supplement 28
Scanning Laser Cytometry
Figure 2.10.5 Identification of micronuclei in cells. See text for details. Courtesy of CompuCyte Corporation.
2.10.6 Supplement 28
Current Protocols in Cytometry
position
y
x position Figure 2.10.6 An x-y plot of positions of single nuclei obtained from an image of a skin section analyzed with the LSC, providing a cartoon of the tissue architecture, clearly illustrating the locations of hair follicles. Courtesy of CompuCyte Corporation.
sidered analogous to pulse width as a measure of size. The LSC, like many flow cytometers, can also record the time at which a cell was measured. Among the data the LSC can routinely collect that provide information unobtainable by flow cytometry are the perimeter of the integration contour, which, in combination with the area, can provide information about shape, and the x and y coordinates of each cell, which give its position on the slide. Using position information, data in several files can be merged, allowing cell-by-cell correlation of information obtained from scans of the same specimen done at different times and with different reagents.
Gating and Data Manipulation Data from a scanning laser cytometer can be manipulated in much the same manner as that from a flow cytometer. For example, clusters of cells can be defined and counted on the basis of the presence and amount of a fluorescent antibody bound on or in the cells, allowing immunophenotyping to be performed. This is accomplished on the LSC by operator interaction; the IMAGN 2000, designed for walkaway clinical use, automates the process for such special-purpose applications as counting CD4and CD8-positive T lymphocytes in peripheral blood (O’Gorman et al., 1997). The LSC can be used to perform DNA content analysis on cells, with a gate set on a plot of maximum pixel versus nuclear fluorescence integral allowing
most doublets to be excluded, as would be done in flow cytometry using pulse height and integral values. Figure 2.10.5 illustrates the use of the LSC in detecting micronuclei, an important application in toxicology. The top panel shows a segmented image of a field of cells stained with the far-red fluorescent DNA dye DRAQ5 (UNIT 7.25); note that the binucleate cell in the center of the field also contains two micronuclei. Micronuclei (and double nuclei) within cells are segmented separately, with the result that the DRAQ5 fluorescence histogram in the left half of the middle panel shows a small population of objects with low DNA content representing the micronuclei. Information about the position of cells and nuclei is stored in the system, allowing regions of interest containing the micronuclei to be defined (right half of middle panel). The bottom panel of Figure 2.10.5 shows a “gallery” of cells bearing micronuclei. Position information can also be used as shown in Figure 2.10.6, to provide a map of the locations of nuclei or other cellular structures, effectively reconstructing tissue architecture.
COMMERCIALLY AVAILABLE SYSTEMS AND APPLICATIONS Acumen Bioscience Ltd. The Acumen Explorer HTS is a multiple-detector laser scanning system using a 488-nm argon-ion laser source; it is designed for highImage Cytometry Instrumentation
2.10.7 Current Protocols in Cytometry
Supplement 28
throughput screening of material in 96-, 384or 1536-well plates.
Applied Biosystems The FMAT (fluorometric microvolume assay technology) 8100 HTS System, developed by Biometric Imaging, identifies, counts, and reads fluorescence from cells or beads in multiwell plates; it includes a robotic plate loader. The light source is a 633-nm He-Ne laser. The instrument is designed for high-throughput screening.
Chemunex SA The ChemScan RDI is designed to detect microorganisms in water, food, cosmetics, or pharmaceuticals; samples are stained with a fluorogenic enzyme substrate and scanned on a filter (Baudart et al., 2002; de Roubin et al., 2002). The light source is an argon-ion laser.
BD Biosciences The IMAGN 2000, developed by Biometric Imaging, is a small tabletop system designed for clinical use. It was originally conceived for automated analysis of CD4- and CD8-positive T lymphocytes in peripheral blood using a disposable cartridge in which all necessary reagents are included; the apparatus itself contains a centrifuge in which sample staining and dilution are done. The report of a clinical trial of this apparatus (O’Gorman et al., 1997) is of general interest because of its superior statistical methodology, which differs substantially from that typically employed for comparisons between analytical methods. The IMAGN 2000 can also be used for analysis of blood stem cells (Chapple et al., 2000) and of residual white blood cells in packed red cells or platelets (Barnett et al., 2001), among other applications. The IMAGN 2000 is not intended for assay development by the end user and provides no software facilities other than those usable for assays offered by the manufacturer. Although the instrument was withdrawn from the market some years ago by BD Biosciences, it is once again being sold in Europe and the U.K, primarily for blood banking and transplant-related applications, and will be available in the U.S. following FDA reclearance.
CompuCyte Corporation
Scanning Laser Cytometry
The CompuCyte line now includes the LSC and the newer iCyte and iCys systems. The LSC is intended for more interactive operation in research and clinical laboratories, and provides
flexible software for analysis and application development. It can measure fluorescence in four wavelength regions and forward-scattered light; CCD cameras are provided to facilitate setup of the instrument and to allow collection of bright-field and fluorescence images of cells, but relatively high-resolution scatter-based images are now also available. The iCyte and iCys are built around inverted microscopes, facilitating examination of specimens in Petri dishes and multiwell plates. Like the LSC, the iCyte has four PMT fluorescence detectors and a photodiode scatter sensor; it can be equipped with 488-nm argon, 633-nm HeNe, and 400-nm diode lasers and is equipped for both bright-field scatter imaging and fluorescence pseudocolor imaging. The iCyte has autofocus capability. Robotic specimen handling for microtiter plates or slide holders is optional. The iCys is similar but is available with fewer fluorescence detectors and a single (488-nm) laser. A number of publications (Darzynkiewicz et al., 1999, 2001; Kamentsky, 2001; Tarnok and Gerstner, 2002; UNITS 6.13 & 7.22) describe the broad range of applications of the LSC (and the newer CompuCyte instruments) in cell biology and pathology; a bibliography and additional information are available from the manufacturer. Although the CompuCyte systems cannot achieve the high sample throughput rates routinely obtainable with flow cytometry, their abilities to deal with small cellular samples, perform repeated analyses of the same cells over time, and localize materials and structures within cells and tissues, while maintaining a relatively friendly user interface, can provide researchers and clinicians with the means to do experiments that would not be feasible using either flow cytometry or higher-resolution, lower-throughput imaging systems. The computer-controlled stage and associated software allow a slide to be removed, processed, and replaced, after which individual cells can be relocated even if the position of the slide on the stage has changed slightly. This provides a powerful means for performing sequential analyses; for example, mitochondrial membrane potential and/or annexin V binding can be examined to detect signs of apoptosis in living (or at least unfixed) cells, after which cells can be permeabilized or fixed, allowing the DNA content of apoptotic and nonapoptotic cells to be determined and compared on a cellby-cell basis using propidium iodide. Software capabilities have been greatly expanded during
2.10.8 Supplement 28
Current Protocols in Cytometry
the time that the instruments have been on the market; newer random sampling and scan compression procedures provide higher throughput for tissue analysis and for counting of confluent cells. However, no instrument is infinitely flexible; potential users should, as always, consult the manufacturers to determine whether hardware and software suitable for a particular application are or will be available.
LITERATURE CITED Barnett, D., Goodfellow, K., Ginnever, J., Granger, V., Whitby, L., and Reilly, J.T. 2001. Low level leucocyte counting: A critical variable in the validation of leucodepleted blood transfusion components as highlighted by an external quality assessment study. Clin. Lab. Haematol. 23:4351. Baudart, J., Coallier, J., Laurent, P., and Prevost, M. 2002. Rapid and sensitive enumeration of viable diluted cells of members of the family enterobacteriaceae in freshwater and drinking water. Appl. Environ. Microbiol. 68:5057-5063. Chapple, P., Prince, H.M., Wall, D., Filshie, R., Haylock, D., Quinn, M., Bretell, M., and Venter, D. 2000. Comparison of three methods of CD34+ cell enumeration in peripheral blood: Dual-platform ISHAGE protocol versus singleplatform, versus microvolume fluorimetry. Cytotherapy 2:371-376. Darzynkiewicz, Z., Bedner, E., Li, X., Gorczyca, W., and Melamed, M.R. 1999. Laser-scanning cytometry: A new instrumentation with many applications. Exp. Cell Res. 249:1-12. Darzynkiewicz, Z., Smolewski, P., and Bedner, E. 2001. Use of flow and laser scanning cytometry to study mechanisms regulating cell cycle and controlling cell death. Clin. Lab. Med. 21:857873.
de Roubin, M.-R., Pharamond, J.-S., Zanelli, F., Poty, F., Houdart, S., Laurent, F., Drocourt, J.-L., and Van Poucke, S. 2002. Application of laser scanning cytometry followed by epifluorescent and differential interference contrast microscopy for the detection and enumeration of Cryptosporidium and Giardia in raw and potable waters. J. Appl. Microbiol. 93:599-607; Erratum in J. Appl. Microbiol. 2003; 94:158. Dietz, L.J., Dubrow, R.S., Manian, B.S., and Sizto, N.L. 1996. Volumetric capillary cytometry: A new method for absolute cell enumeration. Cytometry 23:177-186. Kamentsky, L.A., Burger, D.E., Gershman, R.J., Kamentsky, L.D., and Luther, E. 1997. Slidebased laser scanning cytometry. Acta Cytol. 41:123-143. Kamentsky, L. A. 2001. Laser scanning cytometry. Methods Cell Biol. 63:51-87. O’Gorman, M.R.G., Gelman, R., Site Investigators, and the NIAID New CD4 Technologies Focus Group. 1997. Inter- and intra-institutional evaluation of automated volumetric capillary cytometry for the quantitation of CD4- and CD8positive T-lymphocytes in the peripheral blood of persons infected with human immunodeficiency virus. Clin. Diag. Lab. Immunol. 4:173-179. Seamer, L.C., Bagwell, C.B., Barden, L., Redelman, D., Salzman, G.C., Wood, J.C.S., and Murphy, R.F. 1997. Proposed new data file standard for flow cytometry, version FCS 3.0. Cytometry 28:118-122. Smith, P. J., Wiltshire, M., Davies, S., Patterson, L. H., and Hoy, T. 1999. A novel cell permeant and far red-fluorescing DNA probe, DRAQ5, for blood cell discrimination by flow cytometry. J. Immunol. Methods 229:131-139. Tarnok, A. and Gerstner, A.O. 2002. Clinical applications of laser scanning cytometry. Cytometry 50:133-143.
Contributed by Howard M. Shapiro, M.D., P.C. West Newton, Massachusetts The author thanks Bala Manian and Kim Mulcahy (formerly of Biometric Imaging) and Kurt Barkalow, Elena Holden, Anne Byrne, Louis Kamentsky, and Ed Luther of CompuCyte Corporation.
Image Cytometry Instrumentation
2.10.9 Current Protocols in Cytometry
Supplement 28
Shading Correction: Compensation for Illumination and Sensor Inhomogeneities
UNIT 2.11
This unit develops procedures to correct an image for the shading that can be introduced in the image-forming procedure. The method by which microscope images are produced—the interaction between objects in real space, the illumination, and the camera— frequently leads to situations in which the image exhibits significant shading across the field of view. In some cases the image might be bright in the center and decrease in brightness as one goes toward the edge of the field of view. In other cases the image might be darker on the left side and lighter on the right side. The shading might be caused by nonuniform illumination, nonuniform camera sensitivity, or even dirt and dust on glass (lens) surfaces. In general this shading effect is undesirable. Its elimination is frequently necessary for subsequent processing, especially when quantitative microscopy is the final goal. A mathematical model for the shading effect is based upon the simple schematic diagram shown in Figure 2.11.1. The illumination over the microscope field of view, Iill(x,y), usually interacts in a multiplicative way with the biological object, a(x,y), to produce the image, b(x,y): b( x, y ) = I ill ( x, y ) • a ( x, y ) Equation 2.11.1
with the object representing various microscope imaging modalities such as: r ( x, y ) a ( x, y ) = 10− OD( x , y ) c( x, y )
reflectance model absorption model fluorescence model
Equation 2.11.2
where at position (x,y), r(x,y) is the reflectance, OD(x,y) is the optical density, and c(x,y) is proportional to the concentration of fluorescent material. Parenthetically, note that this
a (x, y )
microscope system
b (x, y )
Iill (x, y )
camera system
gain [m, n ]
c [m, n ]
offset [m, n ]
Figure 2.11.1 A simplified model of a microscope/camera system for producing digital images.
Image Cytometry Instrumentation
Contributed by Ian T. Young
2.11.1
Current Protocols in Cytometry (2000) 2.11.1-2.11.12 Copyright © 2000 by John Wiley & Sons, Inc.
Supplement 14
fluorescence model holds only for low concentrations. The camera may then contribute gain and offset terms, so that: c[ m, n ] = gain[ m, n ] • b[ m, n ] + offset [ m, n ] = gain[ m, n] • I ill [ m, n] • a[ m, n] + offset[ m, n ] Equation 2.11.3
Note that both the camera gain and offset can vary as a function of position, thereby contributing to the problem of shading. In general, it is assumed (and borne out in practice) that Iill[m,n], gain[m,n], and offset[m,n] are slowly varying compared to a[m,n]. The step of going from the continuous microscope image in (x,y) to the digitized image [m,n] in a computer memory is described in UNIT 2.6, Young (1989), and Castleman (1996). FLAT AND NON-FLAT FIELDS If an image has a hypothetically constant illumination: Iill(x,y) = Io and the gain and offset of the camera are also constant for all positions: gain[m,n] = Go and offset[m,n] = Oo then the brightness c[m,n] recorded in the image should be the same at all pixel positions. It is referred to as a “flat” field of view or, simply, a flat field. If the illumination is not constant across the field of view or the camera gain and offset vary from position to position, or both, then this will give rise to a non-flat field, which is called shading. A measure of the shading in an image is defined in the following way. Adjust the illumination properly (e.g., as in Köhler illumination; UNIT 2.1) and focus the microscope objective on a plane (z = z0) where objects are to be examined. This can be accomplished by focusing on an object of interest such as an interphase nucleus and then moving the microscope stage in a lateral (x,y) direction to an empty field of view. One or more digital images of the empty field should then be recorded. If L multiple images are recorded (e.g., L = 16), then an average image can be calculated as: I avg [m, n ] =
1 L
L
∑ I [m, n] l
l =1
Equation 2.11.4
where Il[m,n] is the lth recorded image. If only one image has been recorded, then a similar smoothing effect can be achieved by fitting a two-dimensional, second-order curve to the recorded image. This quadratic curve should be of the form: Shading Correction: Compensation for Illumination and Sensor Inhomogeneities
I q [m, n ] = β1 + β2 < m + β3 < n + β4 < m < n + β5 < m 2 + β6 < n 2 Equation 2.11.5
2.11.2 Supplement 14
Current Protocols in Cytometry
The fitting can be based upon minimizing the squared error between the one recorded image I[m,n] of size M × N pixels, and the model given in Equation 2.11.5, using the standard expression:
sq_err =
1 M
M
N
∑∑ ( I [m, n] − I
[m, n])
2
q
m =1 n =1
Equation 2.11.6
Of course one could choose to use Iavg[m,n] instead of I[m,n] in the above fitting procedure, but for most applications where there is sufficient signal (light), the procedure described above is sufficient. The details of finding the coefficients {βi | i=1,6} in Equation 2.11.5 are described in v.d. Doel et al. (1998). In either case—one image or many—the intention is to reduce the effects of random noise that may be contained in an incidental image through either (multi-image) averaging or (polynomial) smoothing. The amount of shading in the image is then defined as: shading =
I max − I min ×100% I max
Equation 2.11.7
where Imax and Imin are the maximum and minimum values, respectively, of the image brightness in the “smoothed” image. To illustrate the applicability of this procedure as well as typical amounts of shading, Table 2.11.1 presents the results of this procedure applied to three microscope/camera systems. The details of the configurations, procedures, and experiments that led to these results can be found in v.d. Doel et al. (1998). The root mean square (rms) error is related to Equation 2.11.6 as 100% × (sq_err)1/2/I max. The residual 1% to 2% error in Table 2.11.1 is essentially the noise level in a single recorded image. The rms error is so small that Equation 2.11.5 can be regarded as providing an excellent description of the shading in a clear-field microscope image. It is interesting to note that while a shading of 25% is an objectively large amount, subjectively it is barely perceived by a human observer. According to the Weber-Fechner model, human perceptual response goes as the logarithm of the stimulus, meaning that a large change in stimulus is required to produce a noticeable change in response (Frisby, 1980; Murch,
Table 2.11.1 Shading and rms Error Fit Values for Several Microscopes and Imaging Modalities
Shading (%)
rms error (%)
Microscope L-1 Microscope Z-1
4.52 17.48
1.96 1.64
Microscope Z-2 Fluorescence Microscope L-1
16.31 6.85
1.71 2.05
Microscope Z-1 Microscope Z-2
27.70 16.67
1.53 1.69
Mode
Microscope
Absorption
Image Cytometry Instrumentation
2.11.3 Current Protocols in Cytometry
Supplement 15
shading = 0%
12.5%
25%
50%
Figure 2.11.2 A homogeneous field is contaminated by varying amounts of shading ranging from 0% (flat) to 50%. The resulting gray-level images are shown by the square images and the quadratic surfaces from Equation 2.11.5 are shown by the graph images. The amount of shading, according to Equation 2.11.7, is shown by percentage.
1973; Stockham, 1972; Young et al., 1998). This is illustrated in Fig. 2.11.2 for a “presumably” homogeneous field of maximum intensity 100 that is contaminated by various amounts of shading. SHADING CORRECTION Two cases for the determination of a[m,n] starting from c[m,n] are distinguished. In both cases there is an intent to estimate the shading terms {gain[m,n]•Iill[m,n]} and {offset[m,n]}. In the first case—which is referred to as the a posteriori estimate—an assumption is made that only the recorded image c[m,n] is available with which to work. In the second case—which is referred to as the a priori estimate—the assumption is made that the recording of two additional “calibration” images, in addition to recording of the images that contain the objects of interest, is possible. An explanation of the various procedures is illustrated by two examples shown in Figure 2.11.3. The first is based on one-dimensional synthetic data that show a clear shading trend as well as multiple well-defined “objects.” The second example is the image of a fluorescently labeled latex microsphere observed in absorption mode with a Nikon Optiphot microscope and Nikon PlanApo 60×, NA=1.40 lens and digitized with a Cohu 4810 CCD camera (Cohu Inc.) and a Data Translation QuickCapture frame grabber (Data Translation Inc.). The microspheres were obtained from Flow Cytometry Standards Corporation (FCSC); the sample shown in Fig. 2.11.3, panel B, comes from a population that is characterized by the manufacturer as having an average diameter of 5.8 µm and a CV (coefficient of variation = σ/µ) of 2%, where µ is the average diameter of the population and σ is the standard deviation of the population. Shading Correction: Compensation for Illumination and Sensor Inhomogeneities
A Posteriori Estimate In this case, an attempt to extract the shading estimate from c[m,n] is made. This situation arises when a recorded image that contains objects of interest is presented and there is no
2.11.4 Supplement 15
Current Protocols in Cytometry
B A brightness
250
shaded image
200 150 100 50 0
0
50 100 150 Horizontal position
200
Figure 2.11.3 Sample images (A) 1-D synthetic image with shading and four “objects.” (B) 2-D absorption image of a 5.8-µm latex microsphere. The image is 256 × 256 pixels. Notice the quadratic shading (brightest in the center), the left-to-right camera sensitivity change (indicated by the arrow), and the dirt particle below the microsphere.
opportunity to use the microscope/camera system to obtain the “calibration” images. The most common corrective possibilities are the following. Linear filtering Using a linear smoothing filter, a computed smoothed version of c[m,n], where the smoothing is large compared to the size of the objects in the image, is made. Thus, if the object in a specific dimension is ~25% of the image size in that dimension, then the smoothing window in that dimension should be ≥25% of the image size. This smoothed version is intended to be an estimate of the background of the image. The smoothed version is subtracted from c[m,n]—to eliminate the shading—and lastly the desired average brightness value is restored. In linear filter theory, smoothing filters are generally referred to as low-pass filters. The process is expressed by the formula: aˆ[m, n] = c[m, n] − LowPass {c[m, n]} + constant Equation 2.11.8
where {â[m,n]} is the estimate of a[m,n]. The two most common low-pass filters for this type of application are the uniform filter and the Gaussian filter. Both are described in detail in Young et al. (1998). For the uniform filter one chooses the appropriate filter size and for the Gaussian filter one chooses the appropriate Gaussian width, σ. Since objects in an image rarely have a preferred direction, it is common to choose filters that have either square (uniform case) or circular (Gaussian case) shapes as opposed to, say, rectangular or elliptical. Homomorphic filtering If the offset[m,n] = 0 in Equation 2.11.3, then c[m,n] consists solely of multiplicative terms. Further, the term {gain[m,n]•Iill[m,n]} is slowly varying while a[m,n] presumably is not. Therefore, take the logarithm of c[m,n] to produce two terms, one of which is slowly varying (low frequency) and one of which is rapidly varying (high frequency). Shading is suppressed by high-pass filtering the logarithm of c[m,n], and then taking the exponent
Image Cytometry Instrumentation
2.11.5 Current Protocols in Cytometry
Supplement 14
(inverse logarithm) to restore the image. This procedure is based on homomorphic filtering as developed by Oppenheim, Shafer, and Stockham (Oppenheim et al., 1968; Stockham, 1972), and is expressed in the formula:
1. 2. 3. 4.
c[ m, n] = gain[ m, n] • Iill [ m, n] • a[ m, n] ln {c[ m, n]} = ln gain[ m, n] • I ill [ m, n] + ln a
[ m, n]
rapidly varying slowly varying HighPass {ln {c[ m, n]}} ≈ ln {a[ m, n]} aˆ[ m, n] = exp HighPass {ln {c[ m, n]}}
{
}
Equation 2.11.9
This approach is mentioned for the sake of completeness, and while the method for choosing the high-pass filter should be specified, the results that will be shown below indicate that this technique does not really warrant further detailed exploration. Morphological filtering Again, a smoothed version of c[m,n] is computed, where the smoothing is large compared to the size of the objects in the image, but this time “morphological” smoothing is used. This smoothed version is the estimate of the background of the image. The smooth version is subtracted from c[m,n] and then the desired average brightness value is restored. This is expressed in the formula: aˆ[m, n] = c[m, n] − MorphSmooth {c[m, n]} + constant Equation 2.11.10
It is clear that this approach is the same as that described in Equation 2.11.8, but linear smoothing filters are now replaced by morphological smoothing filters. While this may sound difficult, implementation is simplicity itself. Without going into the theory or the details of morphological filters (for details see Young et al., 1998; Giardina and Dougherty, 1988; Heijmans, 1994), the two basic operations needed are the maximum filter and the minimum filter.
Shading Correction: Compensation for Illumination and Sensor Inhomogeneities
12 16 31
32 12 27
45 35 35
82 63 56
77 52 44
maximum
82
12 16 31
32 12 27
45 35 35
82 63 56
77 52 44
minimum
12
input window (5 × 3)
output window (5 × 3)
Figure 2.11.4 The minimum and maximum filters applied to an input window (J=5, K=3) to determine the appropriate pixel value in the output window. Note that the surrounding values in the output window are not shown, as they require values in the input window that are not specified in this example.
2.11.6 Supplement 14
Current Protocols in Cytometry
In the maximum filter, defined over a window of J × K pixels where both J and K are considered to be of odd size (e.g., 5 × 3), the value in the output image corresponding to the center pixel in the input window is the maximum brightness value found in the input window. In the minimum filter, defined over a similar J × K window of pixels, the value in the output image corresponding to the center pixel in the input window is the minimum brightness value found in the input window. The maximum and minimum filters are examples, respectively, of the morphological filters that are called dilation and erosion. In formula, these filters are defined as: D(A) = max
[ j , k ]∈W
A) {a [m − j, n − k ]} = max( W
Equation 2.11.11
E (A) = min
[ j ,k ]∈W
A) {a [m − j, n − k ]} = min( W
Equation 2.11.12
where the maximum or minimum is computed over those pixels [j,k] in the window, W. This is illustrated in Fig. 2.11.4. When these filters are applied to the one-dimensional example shown in Fig. 2.11.3, panel A, using a window of size 15 × 1, the result is shown in Fig. 2.11.5. Finally, the morphological smoothing filter described in Equation 2.11.10 can be defined as:
(
(
MorphSmooth( A) = min max max ( min ( A ))
))
Equation 2.11.13
250
dilation “original”
brightness
200 150 100
erosion
50 0
0
50 100 150 Horizontal position
200
Figure 2.11.5 Applying the dilation filter (Equation 2.11.11) and the erosion filter (Equation 2.11.12) to the original data shown in Fig. 2.11.3, panel A.
Image Cytometry Instrumentation
2.11.7 Current Protocols in Cytometry
Supplement 14
A
B
The basic minimum/erosion operation
C
1. Read image into A 2. grey_erosion A B 35 3. grey_dilation B C 35 4. grey_dilation C C 35 5. grey_erosion C D 35 6. sub_im A D B 7. rescale image B
35 35 35 35
1 1 1 1
The basic maximum/dilation operation /* put input image into memory A */ /* perform minimum from A→B */ /* perform maximum from B→C */ /* perform maximum from C→C */ /* perform minimum from C→D */ /* correction from equation (10) */ /* final brightness scaling */
Figure 2.11.6 Recipe for implementing the morphological approach to shading correction. The algorithms are capable of handling up to 3-D images. The example is drawn from the software package SCIL-Image. (A) The basic minimum/erosion operation configuration. (B) The basic maximum/dilation operation configuration. (C) A short program to implement Equations 2.11.10 and 2.11.13.
where all the filter operations are applied over the same J × K filter window W. While Equation 2.11.13 may seem cryptic, the actual implementation is straightforward. Assuming that two “subroutines” are available, one for the minimum filter and one for the maximum filter, they are applied in a sequential fashion as shown in Fig. 2.11.6. The reason that this is a smoothing operation will not be developed here. The reader is, instead, referred to the references on morphological filtering (Young et al., 1998; Giardina and Dougherty, 1988; Heijmans, 1994). A Priori Estimate If it is possible to record test (calibration) images through the camera’s system, then the most appropriate technique for the removal of shading effects is to record two images: BLACK[m,n] and WHITE[m,n]. The BLACK image is generated by blocking all optical ports, closing the camera shutter, or turning off all illumination sources (including room lights). All this is equivalent to setting Iill(•) = 0 in Equation 2.11.1, which leads to b[m,n] = 0. Carrying this through to Equation 2.11.3 leads to c[m,n] = offset[m,n], which is called BLACK[m,n] because the “lights have been turned off.” The WHITE image is generated by either the average clear field image Iavg(•) given in Equation 2.11.4 or the quadratic fit image Iq(•) given in Equation 2.11.5. Either procedure gives a smoothed estimate of WHITE[m,n] = gain[m,n]•Iill[m,n] + offset[m,n] from Equation 2.11.3 with a[m,n] nominally set to one. The correction procedure then leads to the “calibration” filter: Shading Correction: Compensation for Illumination and Sensor Inhomogeneities
2.11.8 Supplement 14
Current Protocols in Cytometry
A
250
shaded image
brightness
200 150 100 50 0
0
50
100
200
150
Horizontal position
B
C
linear filtering
250
300 brightness
brightness
300 200 150 100 50 0
homomorphic filtering
250 200 150 100 50
0
50
100
150
0
200
0
Horizontal position
D
300
E
morphological filtering
300
150
200
“calibration” filtering
250 brightness
brightness
100
Horizontal position
250 200 150 100 50 0
50
200 150 100 50
0
50
100
150
200
Horizontal position
0
0
50
100
150
200
Horizontal position
Figure 2.11.7 Result of applying the various shading-correction algorithms. (A) Original contaminated with shading. (B) Correction with low-pass filtering. (C) Correction with logarithmic filtering. (D) Correction with maximum/minimum filtering. (E) Correction with BLACK and WHITE images. The final result (E) is identical to the original image (not shown) before shading was used to contaminate the original to produce image (A).
aˆ [m, n ] =
c [m, n ] − BLACK [m, n]
WHITE [m, n ] − BLACK [ m, n] Equation 2.11.14
The result {â}[m,n] is a corrected floating point image whose values are between 0.0 and 1.0. This can be rescaled to a more conventional brightness range by multiplying, for example, by 255 to produce an image in the range 0 to 255. As modern digital computers
Image Cytometry Instrumentation
2.11.9 Current Protocols in Cytometry
Supplement 14
original
brightness
250 200 150 100 50 0
Pixel
linear
brightness
250 200 150 100 50
“calibration”
Shading Correction: Compensation for Illumination and Sensor Inhomogeneities
brightness
morphological
brightness
0
300 250 200 150 100 50 0
300 250 200 150 100 50 0
Pixel
Pixel
Pixel
Figure 2.11.8 Result of applying the a posteriori (i.e., linear filtering and morphological filtering) and a priori (i.e., calibration filtering) shading-correction algorithms to an image of a latex microsphere. The right-most column shows the same horizontal line (row) of brightness taken through the middle of each microsphere image.
2.11.10 Supplement 14
Current Protocols in Cytometry
are quite capable of efficiently handling floating point storage as well as computation, one is always free to simply use the floating point representation in subsequent work. RESULTS The efficacy of the various procedures described above will be demonstrated here. First, examination of the synthetic one-dimensional image will be described. Without a priori information—specifically, where the BLACK and WHITE images are not available—options are limited to the results shown in Fig. 2.11.7, panels B through D. With the availability of the BLACK and WHITE images, the final result is identical to the original, uncontaminated image of four objects on a uniform background. This can be seen in Figure 2.11.7, panel E. It should be clear from this example that the homomorphic filtering approach, while highly appropriate in certain applications (Oppenheim et al., 1968), does not yield useful results in this context. Of the two remaining procedures—low-pass filtering and morphological filtering—the nonlinear morphological filtering yields results that are, in general, superior to the linear filtering approach. Linear filtering, by its very nature, tends to produce the “Gibbs overshoots” (i.e., overestimates and underestimates) that can be seen in Figure 2.11.7, panel B. The only parameters that need to be specified in selecting the correct morphological filter are the sizes J × K of the filter window in Equations 2.11.10 through 2.11.13. The leftmost and rightmost objects in Figure 2.11.7, panel A, both have a width of about 20 pixels and are the biggest of the four objects in this one-dimensional image. This means that the window size of the morphological should be chosen to be larger than 20, for example, J = 25 (an odd number) and K = 1. Finally, Fig. 2.11.8 displays the results when these procedures are applied to a true two-dimensional image. In particular, when “calibration” filtering is applied, all the distortions listed in the caption of Fig. 2.11.3, panel B, have been eliminated, including the dirt particle below the microsphere. CONCLUSIONS It is not always possible to record calibration images. Images are sometimes collected at remote sites and have to be analyzed “as is.” When this is the case, then the morphological filtering approach gives the best performance. When the calibration images BLACK and WHITE are available, then the choice is clear. Shading correction should be performed on the basis of “calibration” filtering using as much of the a priori information as is available. LITERATURE CITED Castleman, K.R. 1996. Digital Image Processing (2nd ed.). Prentice-Hall, Englewood Cliffs, N.J. Frisby, J.P. 1980. Seeing: Illusion, Brain and Mind. Oxford University Press, Oxford. Giardina, C.R. and Dougherty, E.R. 1988. Morphological Methods in Image and Signal Processing. Prentice-Hall, Englewood Cliffs, N.J. Heijmans, H.J.A.M. 1994. Morphological Image Operators. Academic Press, Boston. Murch, G.M. 1973. Visual and Auditory Perception. Bobbs-Merrill, New York. Oppenheim, A.V., Schafer, R.W., and Stockham, T.G. Jr. 1968. Non-linear filtering of multiplied and convolved signals. Proc. IEEE 56:1264-1291. Stockham, T.G. 1972. Image processing in the context of a visual model. Proc. IEEE 60:828-842.
Image Cytometry Instrumentation
2.11.11 Current Protocols in Cytometry
Supplement 14
v.d. Doel, L.R., Klein, A.D., Ellenberger, S.L., Netten, H., Boddeke, F.R., van Vliet, L.J., and Young, I.T. 1998. Quantitative evaluation of light microscopes based on image processing techniques. Bioimaging 6:138-149. Young, I.T. 1989. Image fidelity: Characterizing the imaging transfer function. In Fluorescence Microscopy of Living Cells in Culture: Quantitative Fluorescence Microscopy–Imaging and Spectroscopy (D.L. Taylor and Y.L. Wang, eds.) pp. 1-45. Academic Press, San Diego. Young, I.T., Gerbrands, J.J., and van Vliet, L.J. 1998. Image processing fundamentals. In The Digital Signal Processing Handbook (V.K. Madisetti and D.B. Williams, eds.) pp. 51.1-51.81. CRC Press in cooperation with IEEE Press, Boca Raton, Fla.
Contributed by Ian T. Young Delft University of Technology Delft, The Netherlands
Dr. Young wishes to acknowledge the support of the Rolling Grants program of the Foundation for Fundamental Research in Matter (FOM) and the Delft InterFaculty Research Center (DIOC) program “Intelligent Molecular Diagnostic Systems.”
Shading Correction: Compensation for Illumination and Sensor Inhomogeneities
2.11.12 Supplement 14
Current Protocols in Cytometry
Photobleaching Measurements of Diffusion in Cell Membranes and Aqueous Cell Compartments
UNIT 2.12
In the pioneer work of hybridizing mouse and human cultured cell lines to produce xenogene hybrids, Frye and Edidin (1970) were the first to observe lateral movement of membrane proteins on fused heterokaryons. Since this first observation, the lateral mobility of many membrane components, including lipids and proteins, has been clearly established. Significant progress has been made during the past few years in understanding the molecular interactions that direct transport processes in the plasma membrane and in the aqueous compartments of cells. The progress has been driven by the fact that various biophysical, biochemical, and molecular approaches have been applied to study divergent model systems and cell types. This unit assembles a number of significant techniques and models related to the principle of fluorescence recovery after photobleaching (FRAP). This technique has been extensively used to study the diffusion of fluorochrome-labeled molecules and small particles in the cell membrane, in aqueous compartments of living cells, and in artificial membranes and solutions. This overview focuses on the principles of FRAP technique and highlights recent directions of work on molecular transport processes. It is beyond the scope of this unit to review the technical aspects of FRAP and to overview the models and principles controlling the molecular transport in the membrane and aqueous compartments. The goal of this unit is to highlight the impact of modern molecular mobility measurements in the field of analytical cytology. Fluorescence recovery after photobleaching (FRAP, or alternatively, fluorescence photobleaching recovery, FPR) is a well-known method for measuring mobilities driven by diffusion in membranes and liquid systems. In spot photobleaching, fluorophore species in a small region are irreversibly bleached with a short, intense laser beam, and the subsequent kinetics of the fluorescence recovery in the very same bleached volume are recorded. The information derived from the measured recovery curves includes diffusion coefficient and the “immobile fraction” of the detected species. FRAP is probably the most widely used method for measuring lateral mobilities in living cells. Attention has been focused on the following areas. Developing new methods and instrumentation. The classic FRAP method (Axelrod et al., 1976) and derived techniques such as pattern photobleaching (Smith and McConnell, 1978; Miehlich and Gaub, 1993), scanning microphotolysis (Kubitschek et al., 1994; Wedekind et al., 1994), video FRAP (Tsay and Jacobson, 1991), total internal-reflection FRAP (Thompson et al., 1981; Swaminathan et al., 1996; McKiernan et al., 1997), continuous fluorescence microphotolysis (Peters and Beck, 1983; Dietrich et al., 1997), photoactivation of fluorescence (McGrath et al., 1998a), and microsecond FRAP (Kao and Verkman, 1996; Seksek et al., 1997) are based on creating a gradient of the distribution of fluorochrome-labeled components in a predetermined region. The diffusion characteristics of the component(s) are extracted from the decay of dissipation of the concentration gradient. The majority of these techniques require dedicated instrumentation. A schematic diagram of a conventional FRAP instrument is shown in Figure 2.12.1. Establishing a solid theoretical basis for modeling lateral mobility, and quantitation of the observed mobilities in terms of diffusion constant (D) and fractional recovery (Φ). A spot photobleaching data analysis derived by Axelrod et al. (1976) and Yguerabide et al. (1982) is outlined below. The principle of the conventional FRAP method is shown in Figure 2.12.2. Fluorescence intensity measured in a small spot on the sample gives an Contributed by György Lustyik Current Protocols in Cytometry (2001) 2.12.1-2.12.15 Copyright © 2001 by John Wiley & Sons, Inc.
Image Cytometry Instrumentation
2.12.1 Supplement 16
Photon counter
PMT Computer Fluorescence microscope Optical flats OF
Laser
OF Shutter Beam splitter Specimen
Figure 2.12.1 Schematic diagram of a conventional FRAP instrument. The basic unit of the optical system is a fluorescence microscope. A laser beam is used for both fluorescence excitation and photobleaching. The attenuated excitation beam and the high-intensity main beam illuminate the very same region of the specimen. Attenuation can be achieved with neutral-density filters or with a beam splitter. On the figure, the attenuated beam is achieved by 4-fold reflection on a pair of precisely aligned optical flats. The shutter excludes the main beam when it is closed, and allows the beam to pass when it is open. OF, optical filter; PMT, photomultiplier.
Photobleaching Measurements of Diffusion in Cellular Compartments
intensity signal (Fo) that is proportional to the average concentration of the fluorescent compound within the spot. If the spot is exposed to an intense pulse of light, the fluorescent compound within the spot undergoes an irreversible photochemical decomposition (photobleaching). Measuring the fluorescence intensity in the very same spot results in a recovery curve, F(t), which is due to the transport (driven by the Brownian motion) of unbleached compounds from the nonirradiated region of the sample. The complete analysis of Axelrod et al. (1976) extracts the diffusion coefficient (D) and the fractional recovery (Φ) from the observed recovery curve for “normal” diffusion, uniform flow, and a combination of these processes. The information derived is the average of the investigated spot and the recorded time of the recovery curve. As a result, the spatial and temporal resolutions of the technique are limited by the spot size and the recovery time, respectively. Recognition that spatial heterogeneities exist in plasma membranes has led to the development of new photobleaching principles and techniques such as spatial Fourier analysis (Tsay and Jacobson, 1991) and confocal scanning light microscopic methods (Blonk et al., 1993; Kubitschek et al., 1994; Gribbon and Hardingham, 1998). The advances in detection of individual molecules in optical microscopy have made it possible to record their spatial diffusion and develop a method, technically related to FRAP. In recent years the field of optical single-molecule detection and imaging (Erdmann et al., 1999) based on this new technology has experienced exponential growth. The nearfield scanning optical microscope (NSOM) can fulfill the challenging technical requirements
2.12.2 Supplement 16
Current Protocols in Cytometry
Bleaching
Fluorescence intensity
FO F(∞) Recovery
Mobile fraction: Φ = [F(∞) − F(0)]/[F° − F(0)] F(0)
t=0 Time
Figure 2.12.2 A typical recording of a FRAP measurement. The fluorescence intensity (Fo) measured in a small region of the sample (typically 3 to 4 µm in diameter) is proportional to the average concentration of fluorochrome in that region. At t = 0, the small region is illuminated with a strong light pulse (bleaching). The fluorochrome in the region undergoes an irreversible, light-induced photo-decomposition and the post-bleaching fluorescence intensity drops to a lower level (F(0)). The fluorescence intensity shows a recovery because intact fluorochrome molecules diffuse into the bleached area due to the Brownian motion. The F(∞) intensity of full recovery at long time is lower than F o if the fluorochrome-labeled diffusing compound is partly immobile.
such as single-molecule detection sensitivity and high spatial resolution beyond the diffraction limit of the light combined with simultaneous temporal resolution in the millisecond range (Garcia-Parajo et al., 1999). Revealing the importance of mobility characteristics of molecules and particles. Since its development almost 25 years ago, the FRAP technique has contributed significantly to our understanding of the dynamics of membrane lipids and proteins. The fluid-mosaic model of the cell membrane (Singer and Nicolson, 1971) inspired numerous experimental and theoretical studies to determine the diffusion coefficient and the mobile fraction in natural membranes and model systems. The measurements yielded estimates for the diffusion constant of proteins ranging from D ≈ 1-2 × 10−9 cm2/sec to D ≈ 10−12 cm2/sec. Perhaps the two most important observations were that (1) the diffusion constant (D) of membrane proteins was considerably lower than expected theoretically from the hydrodynamic model of Saffman and Delbrück (1975); and (2) the diffusible fraction of proteins (Φ) was generally <100%. Lipid mobility in the plasma membrane was found to be virtually complete over the distance of FRAP resolution (a few micrometers) with a diffusion constant on the order of D ≈ 10−8 cm2/sec. Although early FRAP experiments were not able to detect submicroscopic lipid inhomogeneities, the measurements suggested that the lipids in membranes were not mixed homogeneously but rather segregated in domains (Wolf et al., 1981; Edidin, 1993, 1997). It was concluded that protein diffusion, which is too slow to be determined by lipid viscosity, is somewhat restricted by interactions between membrane proteins and coupling or anchoring to the cytoskeletal or other immobile structural element (Tank et al., 1982; Wu et al., 1982).
Image Cytometry Instrumentation
2.12.3 Current Protocols in Cytometry
Supplement 15
FRAP has also been used to examine complex transport processes such as nucleocytoplasmic transport (Schindler and Jiang, 1986), distribution of cytoplasmic actin between filamentous and monomeric phases (McGrath et al., 1998a,b), and movement of oligodeoxynucleotides and macromolecular probes inside the nucleus (Seksek et al., 1997; Politz et al., 1998). A series of experiments have demonstrated in hepatocytes and other cells that the diffusion coefficient and mobile fraction of membrane proteins decline with age (Lustyik et al., 1987; Zs.-Nagy et al., 1995, 1999; Kitani, 1999). In recent experiments, gap-junctional intercellular communication was investigated using the FRAP principle (Carruba et al., 1999; Himpens et al., 1999); individual cells previously loaded with a fluorescent dye were photobleached and the recovery by gap-junctional transport from the neighboring unbleached cells was monitored. These examples indicate that direct observation of the rate of molecular transport may be of great interest in various fields of analytical cytology. The principles of the major technique that can provide this type of information in both living cells and model systems are reviewed below. DIFFUSION IN THE MEMBRANE The principle of FRAP is intrinsically statistical, based on averaging over a large number of molecular events. A mathematical model describing the relation between the observed fluorescence recovery and the diffusion coefficient on membranous structures (i.e., mobility of molecular particles in the case of two-dimensional systems) was first published by Axelrod et al. (1976). Analysis of the recovery curve is based on several assumptions. 1. Before measurements, the cellular components are labeled with fluorescent dye molecules. The fluorescent components are uniformly distributed on the cell surface. The concentration of fluorophore is Co. 2. A strong laser beam of Gaussian intensity profile, I(r), bleaches fluorescent molecules in a small, radially symmetrical spot. I (r ) = (2 Po / πw2 )exp( −2r 2 / w2 ) Equation 2.12.1
where Po is the laser power; w is the radius of the Gaussian beam at e−2 intensity, and r is the distance from the center of the beam. 3. Photobleaching is a first-order reaction and is irreversible. dC (r, τ) = −αI (r )C (r , τ)d τ Equation 2.12.2
where αI(r) is the rate constant of the photobleaching reaction and C(r,τ) is the concentration of the unbleached fluorophore at radius r at time τ. If T is the bleaching time, at the beginning of the recovery phase (t = 0) the concentration profile of the unbleached fluorophore is: Photobleaching Measurements of Diffusion in Cellular Compartments
C (r , 0) = Co exp[−αI (r )T ]
Equation 2.12.3
2.12.4 Supplement 15
Current Protocols in Cytometry
4. The diffusion in the membrane can be described by two-dimensional Fick’s law. In the radially symmetrical case:
δ / δt C(r, t) = D [δ2/δr 2 +1/ r δ / δr]C(r, t ) Equation 2.12.4
In order to obtain the recovery curve, this differential equation should be solved with appropriate initial and boundary conditions. The initial concentration profile of the fluorescent compound (initial condition) is given by Equation 2.12.3, and the boundary condition by C(∞,t) = Co. The CK(r,t) solution of Equation 2.12.4 depends on the amount of bleach, K: K = αI (0)T i.e., C(0,0) / Co = exp(−K ) Equation 2.12.5
If r and φ are polar coordinates, the fluorescence intensity dF(r,φ,t) generated in the pixel defined by (r,φ) at time t is: dF (r , φ, t ) = (Q / A)CK (r , t ) I (r )r drd φ Equation 2.12.6
where Q is the product of all quantum efficiencies, A is the attenuation factor of the monitoring laser beam: Im(r) = AI(r), and r,φ are polar coordinates. From Equation 2.12.6, the fluorescence recovery curve is given as: F (t ) = F o
∞
( − K )n
∑ n![1 + n(1 + 2t / τ n=0
D )]
Equation 2.12.7
where Fo defines the fluorescence before the bleaching, and n is the positive integer. The K amount of bleaching can be expressed as: F (0) / F o = K −1 [1 − exp(− K )] Equation 2.12.8
τD is the “characteristic” diffusion time that can be computed from the experimentally observed F(t) recovery curve. The diffusion coefficient, D, can be calculated as: D = w 2 / 4τ D Equation 2.12.9
Image Cytometry Instrumentation
2.12.5 Current Protocols in Cytometry
Supplement 15
If all fluorophores are mobile in the bleached area, the asymptote F(t = ∞) of the F(t) recovery curve approaches Fo. If F(∞) < Fo, the fraction of the mobile compound is Φ = [F (∞ ) − F (0)]/[F o − F (0)] Equation 2.12.10
LINEARIZATION METHODS The evaluation of the lateral diffusion coefficient and fraction of mobile compounds of membrane proteins from Equations 2.12.7 to 2.12.10 is often complicated because the determination of the F(0) and F(∞) intensity values is limited by technical difficulties. Uncertainties in the F(0) intensity immediately after photobleaching and in the F(∞) intensity after full recovery are mostly due to instrumental instability and/or cell or tissue movement during the measurement. In order to minimize these problems, a linearization method for precise analysis of FRAP data was developed (Barisas and Leuther, 1979; Yguerabide et al., 1982; Soumpasis, 1983) and used successfully by a number of research groups. The method is based on the observation that the reciprocal function, defined by Equation 2.12.11, R(t ) = F (∞) /[F (∞) − F (t )] Equation 2.12.11
is a linear function of time, i.e., it can be written in the form R(t) = at + b (a is the slope and b is the intercept of the best-fitting linear equation). The intercept-to-slope ratio b/a of the linear plot is equal to the t1/2 time corresponding to 50% recovery. The diffusion coefficient (D) can be calculated from the relation between t1/2 and D. D = βw2 / 4t 12 Equation 2.12.12
where β is a parameter that depends on the amount of bleaching, K. The values of β were determined by Yguerabide et al. (1982) and are tabulated as functions of the percent bleach (P) and amount of bleaching (K) in Table 2.12.1. Table 2.12.1 Values of the β Parameter Versus % Bleach (P) and Amount of Bleach (K)a
Photobleaching Measurements of Diffusion in Cellular Compartments
%Bleach (P)
Amount of bleach (K)
β
% Bleach (P)
Amount of bleach (K)
β
5 10 15 20 25 30 35 40
0.100 0.225 0.340 0.475 0.610 0.765 0.935 1.14
1.00 1.01 1.03 1.04 1.06 1.075 1.09 1.11
45 50 55 60 65 70 80 85
1.35 1.60 1.88 2.20 2.62 3.20 5.00 6.80
1.13 1.15 1.18 1.22 1.26 1.30 1.45 1.59
aYguerabide et al. (1982). Copyright 1982 by the Biophysical Society.
2.12.6 Supplement 15
Current Protocols in Cytometry
Thus, the lateral diffusion coefficient (D) and the mobile fraction can be determined by linear fit to the reciprocal function R(t) (see Equation 2.12.11). The calculation of the experimental values of R(t), however, requires the value of F(∞), which is not known precisely from the recovery curve. The correct value of F(∞) can be closely approximated by testing the closeness of fit of the straight line to the calculated values of the R(t) data. The approximation is based on the observation that R(t) deviates significantly from a straight line when F(∞) is in error. The closeness of fit can be measured with the reduced chi-square given by: χ2 / N =
1 N −2
N
∑ W [ R (t ) − R i
i
2 ci (t )]
i =1
Equation 2.12.13
where N is the number of data points, Ri(t) is the ith data point, Rci(t) is the value of the ith data point calculated from the least-square straight line, and Wi is the weighting function defined by the expression: Wi = [F (∞ ) − Fi (t )]4 / F (∞ )2 Fi (t ) Equation 2.12.14
The correct value of F(∞) is that one at which the χ2/N reduced chi-square is a minimum. The F(0) intensity immediately after photobleaching can be determined from F (0) = F (∞)<[ R(0) − 1]/ R(0) Equation 2.12.15
MULTI-COMPONENT FRAP MODELS The above methods suffer from the drawback that they cannot solve for two or more diffusing components. If Equation 2.12.7 is written in the form F(t) = Fof(t), and Φ is the mobile fraction, the recovery function of the one-component model takes the form F (t ) = ΦF o f (t ) + (1 − Φ )F (0) Equation 2.12.16
and the multi-component model is described by N
F (t ) =
∑ α [Φ F i
i
o
fi (t ) + (1 + Φ i )F (0)]
i =1
Equation 2.12.17
where αi is the fraction of the ith component. Image Cytometry Instrumentation
2.12.7 Current Protocols in Cytometry
Supplement 15
Substituting Equation 2.12.8 for F(0) and using Σαi = 1, the multi-component model becomes N
F (t ) = F o
∑
αi Φ i fi (t ) + F o K −1[1 − exp(−K )](1 −
i =1
N
αi Φ i ) ∑ i =1
Equation 2.12.18
Assuming the number of diffusible components to be N, this equation has 3N parameters: D1, D2, … DN; Φ1, Φ2, … ΦN; α1, α2, … αN-1; and K. However, the αiΦι parameters cannot be separated and determined uniquely, i.e., the component fractions and the mobile fractions cannot be extracted uniquely from photobleaching recovery curves. Substituting αiΦι = ξi reduces the number of parameters to 2N+1: N
F (t ) = F o
∑
ξi fi (t ) + F o K −1[1 − exp(−K )](1 −
i =1
N
ξi ) ∑ i =1
Equation 2.12.19
To date, a system containing two distinct diffusing species has been studied (Gordon et al., 1995) by fitting photobleaching recovery curves with Equation 2.12.19. The 5-parameter (D1, D2, ξ1, ξ2, and K) curve-fitting algorithm based on the Marquardt method (Bevington, 1969) was capable of extracting one- and two-component D-values and the weighted mobile fractions (ξ1, ξ2). The performance of the method was demonstrated with cultured mouse 3T3 fibroblasts labeled with (1) the lipid dye rhodamine-PE, (2) rhodamine-labeled anti-GP80 antibody, and (3) double labeled for the GP80 protein and lipid diffusion measurements. An essentially similar two-component model has been applied to measure D-values for two diffusing lipid components in the plasma membrane of yeast by Greenberg and Axelrod (1993). Because the two-component fit requires a significantly higher signal level than the one-component fit, to achieve a given level of confidence in determining two-component recoveries may require collection and analysis of multiple recovery curves. ANOMALOUS DIFFUSION
Photobleaching Measurements of Diffusion in Cellular Compartments
The diffusion of a compound is “normal” or “simple” if its transport can be described by a single diffusion coefficient in a homogeneous medium. Due to the complexity of cellular systems and nonidealities in the measuring instrumentation, it is often invalid to apply the exact theories of simple diffusion in biological objects such as the cell membrane and the aqueous phase of the cytoplasm (Feder et al., 1996; Schutz et al., 1997; Periasamy and Verkman, 1998). The physical mechanisms resulting in anomalous diffusion include interactions with impermeable domains (Saxton, 1982, 1994), the “archipelago effect” (Saxton, 1982), percolation in systems of coexisting rigid and fluid phases (Coelho et al., 1997), and the presence of a continuous distribution of diffusion coefficients (Periasamy and Verkman, 1998). Most of these mechanisms are related to the microscopic and submicroscopic structural inhomogeneities of the cell membrane. In the classic description of the fluid-mosaic model of the cell membrane (Singer and Nicolson, 1971; Saffman and Delbrück, 1975), the lipid bilayer was pictured as a “lipid sea,” emphasizing the free diffusional mobility of lipid components within the lipid matrix. However, lateral inhomogeneities and molecular segregations have been detected in both synthetic lipid bilayers and cell membranes. The proposed models based on these observations suggest that lipid domains may be formed either by lipid-protein or lipid-lipid interactions (Tocanne et al., 1994; Welti and Glaser, 1994).
2.12.8 Supplement 15
Current Protocols in Cytometry
Organization of membrane lipids into domains may have great biological significance. The size, distribution, and fraction of the membrane domains and the kinetics of the domain formation can strongly affect the membrane function, e.g., by excluding diffusible particles from certain compartments, or increasing the concentrations of interacting molecules in particular regions. Feder et al. (1996) suggested that the frequently observed low apparent mobile fraction is due to spatially restricted diffusion in small fluid domains rather than to physical interaction such as anchorage of the diffusing compounds. They proposed a membrane model in which the diffusing molecules display a random movement in an array of potential energy traps. Small unrecoverable fluid domains surrounded by the randomly dispersed, continuously changing solid domains may form such traps. Monte Carlo simulations by Coelho et al. (1997) seem to support this model. The simulated data indicated that totally unrecoverable fluid domains are present even at the fluid area fraction far above the percolation threshold (the fluid fraction at which the liquid phase becomes disconnected). FRAP experiments give averages for the diffusion coefficient and immobile fraction over an area that scales with the resolution of the light microscope. Due to this resolution limit, classic FRAP measurements of the lateral mobility of lipid components can detect domains several hundred nanometers or larger in size. The mechanisms for domain formation emphasized in most review papers suggest that the lipid domains in most cell membranes are smaller, consisting of tens to a few thousand transiently clustered phospholipid molecules (Edidin, 1997). The recently developed method of single-particle tracking (Erdmann et al., 1999) is one that provides the required positional accuracy to image the small-scale lateral diffusion. However, new approaches to FRAP methods for measuring anomalous diffusion may also yield accurate results for understanding the principles of domain formation. A novel strategy elaborated by Periasamy and Verkman (1998) to analyze FRAP data is based on a model defined by the nonlinear relation of mean-square displacement versus time. This type of approach may represent a link between classical FRAP investigations and measurements based on single-particle tracking. The diffusion of a fluorescent component is anomalous if the mean-square displacement 〈x2〉 in single-particle analysis varies with time in a nonlinear manner. A possible form to describe the relation between 〈x2〉 and D in two dimensions is: x 2 = 4 Dt δ Equation 2.12.20
For simple diffusion δ = 1, i.e., the average length of displacement is proportional to the time of observation. Periasamy and Verkman (1998) elaborated regression procedures utilizing the maximum entropy method to extract the continuous distribution of diffusion coefficients α(D) and the function of time-dependent diffusion coefficients D(t). The key feature of the developed analysis procedure was the introduction of the basis recovery function, f(Dt). Derived from Equations 2.12.7 and 2.12.9: F (t ) = f (K , t / τD ) = f (K , d = 4 Dt / w2 ) Equation 2.12.21 Image Cytometry Instrumentation
2.12.9 Current Protocols in Cytometry
Supplement 15
For a fixed K and w, F(t) is a function of y = Dt. The f(y) = f(Dt) basis function is identical for any diffusing compound and matrix. The basis function approach is based on experimentally derived f(Dt) functions measured on reference samples and test samples under identical experimental conditions. One of the restrictions on the use of the method includes maintaining the same amount of bleaching on both the reference and test samples (the basis function is defined for constant K and w in Equation 2.12.21), which may be a challenging technical requirement to achieve. It can be an additional restriction that the theoretical model of the method is valid for equilibrium conditions, but living cells are thermodynamically open systems. DIFFUSION IN AQUEOUS COMPARTMENTS In addition to studying the diffusion of fluorescent species in membranes, FRAP has been extensively used to study fluorophore diffusion in solutions and aqueous cellular compartments. Mobility of molecules and molecular aggregates in the cytoplasm has been the topic of investigations since the early observations of cytoplasmic streaming. In the early 1980s, the techniques of microinjection and FRAP were combined to determine translational diffusion in cells (Wojcieszyn et al., 1981; Jacobson and Wojcieszyn, 1984; Luby-Phelps et al., 1986). The diffusion coefficient of macromolecules of various sizes ranging from ∼40 kD to 450 kD over the distance of a few micrometers varies between D ≈ 3-4 × 10−6 cm2/sec for BSA and 3-4 × 10−8 cm2/sec for actin. These values are ∼30 to 150 times lower than the corresponding D-values in water. In contrast, small molecules such as the spin labels move in the cytoplasm about half as fast as in water (Mastro and Keith, 1984; Gershon et al., 1985). The theory describing the relation between the observed recovery function and diffusion in a three-dimensional (3-D) system is much more complicated than in the two-dimensional (2-D) case. Most 3-D recovery curve fittings are approximations of the 2-D approach. Blonk et al. (1993) recently designed a confocal scanning light microscopic system that allows reduction of the 3-D diffusion model to the classical 2-D solution of Axelrod et al. (1976). The model describes a radially symmetrical bleaching pattern around the axial z-direction, and assumes the laser beam profile to be Gaussian in both the axial z-direction and the radial r-direction. The 3-D equivalent of Fick’s equation (Equation 2.12.4) in a cylindrical coordinate system has an analytical solution resulting in a somewhat more complex function than Equation 2.12.7. The mathematical fitting of the recovery pattern was achieved using the classical Marquardt procedure (Marquardt, 1963). In regards to the technical design of the system, the measurements were performed with a commercial laser scanning confocal microscope equipped with a software-controlled neutral-density filter accessory. Like other “confocal FRAP” adaptations (Gribbon and Hardingham, 1998) this method was validated with model systems.
Photobleaching Measurements of Diffusion in Cellular Compartments
The study of the architecture of the cytoplasm has been paralleled by the study of how molecules move in the cytoplasm. FRAP studies provided significant evidence for the conclusion that the cytoplasm is a matrix displaying a considerable organization rather than being a homogeneous, viscous medium. The measurements clearly demonstrated that the cytoplasm of living cells contains barriers restricting the free diffusion of macromolecules in a size-dependent manner. The organized cytoplasm may restrict macromolecular diffusion through retardation of molecular mobility by steric obstruction and/or binding to the matrix elements (Jacobson and Wojcieszyn, 1984). The diffusion constant measured in the cytoplasm of several cell lines was larger in growing cells and in cells treated with cytochalasin B and other compounds that inhibit actin polymerization and microfilament assembly.
2.12.10 Supplement 15
Current Protocols in Cytometry
As already discussed, the FRAP technique relies on ensemble averaging. The novel single-molecule tracking method allows the analysis of single trajectories of individual molecules (Peters et al., 1999; Weber et al., 1999). Although it is a challenging task to develop powerful methods for the analysis of the results, single-molecule imaging techniques have great potential to reveal details of lateral mobility in aqueous cellular compartments. SINGLE MOLECULE DETECTION In FRAP, the diffusion-driven restoration of fluorescence as intact fluorophores move into a previously photobleached volume element is followed in time. This principle was used to study the diffusion of membrane macromolecules for several years. Recently a new technique, complementary to FRAP and known as single-particle tracking (SPT), has been introduced to measure the motion of single molecular species or small clusters of proteins (Gelles et al., 1988; Saxton and Jacobson, 1997). Today, SPT is a well-established technique based on a broad range of practical realization of the method. An advantage of this method is that it can provide information regarding molecular interaction and dynamics, which is virtually impossible to obtain from ensemble-averaged investigations. The high spatial resolution and the ability to provide excellent temporal resolution will certainly make this technique popular in the future. As pointed out by Erdmann et al. (1999), “In light of the outstanding advances in ultrasensitive detection and spectroscopy, the 1990s can truly be called the Decade of the Single Molecule.” PRACTICAL GUIDELINES Technical principles, instrumentation: Most of the FRAP instruments were developed by laboratories applying the FRAP technique in their research program. As a result, a great variety of instrumentation has been built since the early seventies when the first FRAP devices were developed. In classical FRAP instruments, the translational diffusion coefficient was measured by bleaching fluorescent molecules in a small region and measuring the fluorescence recovery in the bleached area with a highly attenuated beam. In most instruments, the intense bleaching beam and the low intensity monitoring beam are provided by the same light source. The preferred light source is a laser from which the beam is directed into a microscope, and illuminates the sample through the microscope objective. Before being directed into the microscope, the laser beam may be split into a high intensity bleaching beam and a low intensity monitoring beam by a dual beam splitter (Figure 2.12.1), or the high intensity pulse may be generated by removing a neutral density filter from the path of the monitoring beam. Both arrangements are equipped with a mechanical device, a shutter that opens/closes the path of the bleaching beam or a filter controller, that moves a neutral density filter in and out of the beam path. Both solutions are relatively slow, limiting the bleach/monitor transition time to a minimum of a few milliseconds. This limitation reduces the applicability of the FRAP instruments built with mechanical shutters or filter withdrawal mechanisms to measure the diffusion constant for fast diffusions (D >10−6 cm2/sec), such as the diffusion of small molecules in free solutions. The transition time of the intensity modulation should be shorter by at least an order of magnitude than the characteristic diffusion time (τD). The τD time depends on the diffusion coefficient to be determined and the size of the bleaching spot. Typically, the characteristic diffusion time for a circular spot with a Gaussian intensity profile and radius, w, is τD = w2/4D, where D is the diffusion coefficient of the compound of interest. A much shorter transition time, typically a few microseconds, can be achieved by using acousto-optical modulators to switch between the bleaching and the monitoring intensities; these modulators have become very popular in newer instruments. The transition time limit is entirely eliminated by the continuous photobleaching method that does not
Image Cytometry Instrumentation
2.12.11 Current Protocols in Cytometry
Supplement 16
use two different beam intensities for bleaching and monitoring. In this method, the sample is illuminated continuously, causing continuous bleaching in the illuminated spot. The decay of the fluorescence intensity is perturbed by the influx of the intact (unbleached) molecules from the surroundings, which (using the appropriate mathematical model) allows the calculation of the diffusion coefficient. Photobleaching phenomena, fluorophores: The FRAP technique is based entirely on the fluorescence photobleaching phenomenon. Photobleaching is a decomposition process caused by a chemical reaction between the excited dye molecules and the molecular oxygen dissolved in the surrounding environment. In a typical FRAP experiment, the percentage of bleaching (the proportion of dye molecules undergoing photodecomposition) should be ≥40% to 70% to obtain a recovery curve that can be analyzed properly. The rate of bleaching depends on various parameters such as the photon density of the light beam in the bleaching region, the molecular oxygen concentration, temperature, some physico-chemical properties of the medium, dye concentration, and the photophysical characteristics (absorption coefficient, quantum yield) of the dye to be bleached. The required amount of bleaching is generally achieved by changing the intensity of the bleaching beam and/or the bleach time. Because the mathematical model of the FRAP process requires that the diffusion be negligible during the bleaching, the bleach time should be restricted to <5% to 6% of the characteristic diffusion time. FRAP generally assumes that photobleaching is an irreversible process, and that recovery is fully due to the movement of intact dye molecules into the bleaching spot. However, reversible bleaching, most probably due to triplet-singlet transition, may occur under certain experimental conditions, increasing the rate of fluorescence recovery shortly after the bleaching. Recovery from reversible bleaching is a relatively fast process; therefore its contribution to the recovery kinetics is most prominent on the fastest diffusing components. Bleaching due to the monitoring beam cannot be avoided completely, and it may cause significant deterioration of the recovery kinetics. Bleaching during recovery can be lowered by decreasing the intensity of the monitoring light, but the parallel decrease of the signal-to-noise ratio limits the use of very low intensities. The effect of bleaching due to the monitoring light is especially significant late in the recovery period, when it may lead to miscalculation of the mobile/immobile fraction of the diffusible compound. In many cases, the effect of bleaching during recovery must be corrected mathematically. Without correction, most FRAP models result in lower diffusion coefficients and higher percentage of immobile fraction than the real values. On the other hand, reversible bleaching tends to increase the calculated value of D. In order to fulfill these criteria, an ideal dye for FRAP must have suitable optical and photophysical properties and must provide good targeting specificity. (1) The photosensitivity of the dye needs to be high enough to allow the use of short bleaching time. (2) At the same time, the photosensitivity cannot be too high, in order to keep the bleaching during recovery phenomenon as low as possible. (3) It is an advantage if the dye can be efficiently excited and bleached with the 488-nm beam of the frequently used argon-ion lasers.
Photobleaching Measurements of Diffusion in Cellular Compartments
The most frequently used fluorescent dye in FRAP experiments has been fluorescein isothiocyanate (FITC). FITC is a hydrophilic dye that has several advantages making it especially suitable for FRAP, ideal photosensitivity/photostability, ease of labeling macromolecules, and small size. The last means that the FITC does not significantly alter the properties of labeled macromolecules. Other fluorophores such as phycoerythrin and BCEF are also suitable for FRAP experiments. The lipophylic dyes, dialkylcarbocyanine
2.12.12 Supplement 16
Current Protocols in Cytometry
and diphenihexatriene derivatives, are probably the most frequently used lipid probes for FRAP purposes. Recent experiments have shown that green fluorescent protein (GFP) can be used as a powerful noninvasive probe to quantify diffusibility in various cellular compartments (Swaminathan et al., 1997; Dayel et al., 1999). The excellent targeting specificity and the good photophysical properties of GFP make this one of the most promising fluorescent probes for future FRAP experiments. CONCLUSIONS The mobility of molecules in the cell membrane, cytoplasm, and nucleoplasm is of great interest in many fields of analytical cytology. Diffusion of molecules is relevant for various physiological processes in the cell membrane, cytoplasmic compartment, and organelles, as well as in transport of particular reporting molecules that participate in intercellular communication. For these reasons sophisticated techniques have been developed to investigate the diffusion characteristics of fluorochrome-labeled compounds in model systems and native cells. Routine flow cytometry techniques are not yet available for measuring molecular diffusion properties; therefore these photobleaching experiments may be important complements to conventional flow cytometric investigations. Introduction of new principles and/or technical development of the optical and signal detection systems may result in new technologies that make flow cytometric instrumentation capable of measuring diffusion characteristics of particles in living cells. LITERATURE CITED Axelrod, D., Koppel, D.E., Schlessinger, J., Elson, E., and Webb, W.W. 1976. Mobility measurements by analysis of fluorescence photobleaching recovery kinetics. Biophys. J. 16:1055-1069. Barisas, G. and Leuther, M.D. 1979. Fluorescence photobleaching recovery measurement of protein absolute diffusion coefficients. Biophys. J. 10:221-229. Bevington, P.R. 1969. Data Reduction and Error Analysis for the Physical Sciences. McGraw-Hill, New York. Blonk, J.C.G., Don, A., Van Aalst, H., and Birmingham, J.J. 1993. Fluorescence photobleaching recovery in confocal scanning light microscopy. J. Microsc. 169:363-374. Carruba, G., Webber, M.M., Bello-Deocampo, D., Amodio, R., Notarbartolo, M., Deocampo, N.D., Trosko, J.E., and Castagnetta, L.A. 1999. Laser scanning analysis of cell-cell communication in cultured human prostate tumor cells. Anal. Quant. Cytol. Histol. 21:54-58. Coelho, F.P., Vaz, W.L., and Melo, E. 1997. Phase topology and percolation in two-component lipid bilayers: A Monte Carlo approach. Biophys. J. 72:1501-1511. Dayel, M.J., Hom, E.F.Y., and Verkman, A.S. 1999. Diffusion of green fluorescence protein in the aqueous phase lumen of endoplasmic reticulum. Biophys. J. 76:2843-2851. Dietrich, C., Merkel, R., and Tampé, R. 1997. Diffusion measurement of fluorescence-labeled amphiphilic molecules with a standard fluorescence microscope. Biophys. J. 72:1701-1710. Edidin, M. 1993. Patches and fences: Probing for plasma membrane domains. J. Cell Sci. Suppl. 17:165-169. Edidin, M. 1997. Lipid microdomains in cell surface membranes. Curr. Opin. Struct. Biol. 7:528-532. Erdmann, R., Enderlein, J., and Seidel, C. 1999. Single molecule detection and ultrasensitive analysis in the life sciences. Cytometry 36:161-264. Feder, T.J., Brust-Mascher, I., Slattery, J.P., Baird, B., and Webb, W.W. 1996. Constrained diffusion or immobile fraction on cell surfaces: A new interpretation. Biophys. J. 70:2367-2373. Frye, L.D. and Edidin, M. 1970. The rapid intermixing of cell surface antigens after formation of mouse-human heterokaryons. J. Cell Sci. 7:319-335. Garcia-Parajo, M.F., Veerman, J.-A., Segers-Nolten, G.M.J., de Grooth, B.G., Greve, J., and van Hulst, N.F. 1999. Visualizing individual green fluorescence proteins with a near field optical microscope. Cytometry 36:239-246. Gelles, J., Schnapp, B.J., and Sheetz, M.P. 1988. Tracking kinesis-driven movements with nanometre-scale precision. Nature 331:450-453. Gershon, N.D., Porter, K.R., and Trus, B.L. 1985. The cytoplasmic matrix: Its volume and surface area and the diffusion of molecules through it. Proc. Natl. Acad. Sci. U.S.A. 82:5030-5034.
Image Cytometry Instrumentation
2.12.13 Current Protocols in Cytometry
Supplement 16
Gordon, G.W., Chazotte, B., Wang, X.F., and Herman, B. 1995. Analysis of simulated and experimental fluorescence recovery after photobleaching. Data for two diffusing components. Biophys. J. 68:766-778. Greenberg, M.L. and Axelrod, D. 1993. Anomalously slow mobility of fluorescent lipid probes in the plasma membrane of the yeast. J. Membr. Biol. 131:115-127. Gribbon, P. and Hardingham, T.E. 1998. Macromolecular diffusion of biological polymers measured by confocal fluorescence after photobleaching. Biophys. J. 75:1032-1039. Himpens, B., Stalmans, P., Gomez, P., Malfait, M., and Vereecke, J. 1999. Intra- and intercellular Ca2+ signaling in retinal pigment epithelial cells during mechanical stimulation. F.A.S.E.B. J. 13:S63-S68. Jacobson, K. and Wojcieszyn, J. 1984. The translational mobility of substances within the cytoplasmic matrix. Proc. Natl. Acad. Sci. U.S.A. 81:6747-6751. Kao, H.P. and Verkman, A.S. 1996. Construction and performance of a photobleaching recovery apparatus with microsecond time resolution. Biophys. Chem. 59:203-210. Kitani, K. 1999. Lateral mobility of proteins and lipids of cell surface membranes during aging: Do the data support “The Membrane Hypothesis of Aging”? Mech. Aging Dev. 107:299-322. Kubitschek, U., Wedekind, P., and Peters, R. 1994. Lateral diffusion measurement at high spatial resolution by scanning microphotolysis in a confocal microscope. Biophys. J. 67:948-956. Luby-Phelps, K., Taylor, D.L., and Lanni, F. 1986. Probing the structure of cytoplasm. J. Cell Biol. 102:2015-2022. Lustyik, G., Kitani, K., and Ohta, M. 1987. The mobility of concanavalin A receptors and surface immunoglobulins on rat hepatocyte plasma membranes. Biochim. Biophys. Acta 896:57-63. Marquardt, D.W. 1963. An algorithm for least squares estimation of nonlinear parameters. J. Soc. Ind. Appl. Math. 2:431-441. Mastro, A.M. and Keith, A.D. 1984. Diffusion in the aqueous compartment. J. Cell Biol. 99:180s-187s. McGrath, J.L., Hartwig, J.H., Tardy, Y., and Dewey, C.F. Jr. 1998a. Measuring actin dynamics in endothelial cells. Microsc. Res. Tech. 43:385-394. McGrath, J.L., Tardy, Y., Dewey, C.F. Jr., Meister, J.J., and Hartwig, J.H. 1998b. Simultaneous measurements of actin filament turnover, filament fraction, and monomer diffusion in endothelial cells. Biophys. J. 75:2070-2078. McKiernan, A.E., MacDonald, R.I., MacDonald, R.C., and Axelrod, D. 1997. Cytoskeletal protein binding kinetics at planar phospholipid membranes. Biophys. J. 73:1987-1998. Miehlich, R. and Gaub, H. 1993. Holographic pattern photobleaching apparatus for measurement of lateral transport processes at interfaces: Design and performance. Rev. Sci. Instrum. 64:2632-2638. Periasamy, N. and Verkman, A.S. 1998. Analysis of fluorophore diffusion by continuous distributions of diffusion coefficients: Application to photobleaching measurements of multicomponent and anomalous diffusion. Biophys. J. 75:557-567. Peters, I.M., van Kooyk, Y., van Vliet, S.J., de Grooth, B.G., Figdor, C.G., and Greve, J. 1999. 3D single-particle tracking and optical trap measurements on adhesion proteins. Cytometry 36:189-194. Peters, R. and Beck, K. 1983. Translational diffusion in phospholipid monolayers measured by fluorescence microphotolysis. Proc. Natl. Acad. Sci. U.S.A. 80:7183-7187. Politz, J.C., Browne, E.S., Wolf, D.E., and Pederson, T. 1998. Intranuclear diffusion and hybridization state of oligonucleotides measured by fluorescence correlation spectroscopy in living cells. Proc. Natl. Acad. Sci. U.S.A. 95:6043-6048. Saffman, P.G. and Delbrück, M. 1975. Brownian motion in biological membranes. Proc. Natl. Acad. Sci. U.S.A. 72:3111-3113. Saxton, M.J. 1982. Lateral diffusion in an archipelago. Effects of impermeable patches on diffusion in a cell membrane. Biophys. J. 39:165-173. Saxton, M.J. 1994. Anomalous diffusion due to obstacles: A Monte Carlo study. Biophys. J. 66:394-401. Saxton, M.J. and Jacobson, K. 1997. Single particle tracking: Applications to membrane dynamics. Annu. Rev. Biomol. Struct. 26:373-399. Schindler, M. and Jiang, L.-W. 1986. Nuclear actin and myosin as control elements in nucleocytoplasmic transport. J. Cell Biol. 102:859-862. Schutz, G.J., Schindler, H., and Schmidt, T. 1997. Single-molecule microscopy on model membranes reveals anomalous diffusion. Biophys. J. 73:1073-1080. Photobleaching Measurements of Diffusion in Cellular Compartments
Seksek, O., Biwersi, J., and Verkman, A.S. 1997. Translational diffusion of macromolecule-sized solutes in cytoplasm and nucleus. J. Cell Biol. 138:131-142. Singer, S.J. and Nicolson, G.L. 1971. The fluid mosaic model of the structure of the cell membranes. Science 175:720-731.
2.12.14 Supplement 16
Current Protocols in Cytometry
Smith, B.A. and McConnell, H.M. 1978. Determination of molecular motion in membranes using pattern photobleaching. Proc. Natl. Acad. Sci. U.S.A. 75:2759-2763. Soumpasis, D.M. 1983. Theoretical analysis of fluorescence photobleaching recovery experiments. Biophys. J. 41:95-97. Swaminathan, R., Bicknese, S., Periasamy, N., and Verkman, A.S. 1996. Cytoplasmic viscosity near the cell plasma membrane: Translational diffusion of a small fluorescent solute measured by total internal reflection-fluorescence photobleaching recovery. Biophys. J. 71:1140-1151. Swaminathan, R., Hoang, C.P., and Verkman, A.S. 1997. Photobleaching recovery and anisotropy decay of green fluorescent protein GFP-S65T in solution and cells: Cytoplasmic viscosity probed by green fluorescent protein translational and rotational diffusion. Biophys. J. 72:1900-1907. Tank, D.W., Wu, E.-S., and Webb, W.W. 1982. Enhanced molecular diffusibility in muscle membrane blebs: Release of lateral constraints. J. Cell Biol. 92:207-212. Thompson, N.L., Burghardt, T.P., and Axelrod, D. 1981. Measuring surface dynamics of biomolecules by total internal reflection fluorescence with photobleaching recovery of correlation spectroscopy. Biophys. J. 33:435-454. Tocanne, J.-F., Cezanne, L., Lopez, A., Piknova, B., Schram, V., Tournier, J.-F., and Welby, M. 1994. Lipid domains and lipid-protein interactions in biological membranes. Chem. Phys. Lipids 73:139-158. Tsay, T.-T. and Jacobson, K.A. 1991. Spatial Fourier analysis of video photobleaching measurements. Principles and optimization. Biophys. J. 60:360-368. Weber, M.A., Stracke, F., and Meixner, A.J. 1999. Dynamics of single dye molecules observed by confocal imaging and spectroscopy. Cytometry 36:217-223. Wedekind, P., Kubitschek, U., and Peters, R. 1994. Scanning microphotolysis: A new photobleaching technique based on fast intensity modulation of a scanned laser beam and confocal imaging. J. Microsc. 176:22-33. Welti, R. and Glaser, M. 1994. Lipid domains in model and biological membranes. Chem. Phys. Lipids 73:121-137. Wojcieszyn, J.W., Schlegel, R.A., Wu, S.L., and Jacobson, K.A. 1981. Diffusion of injected macromolecules within the cytoplasm of living cells. Proc. Natl. Acad. Sci. U.S.A. 78:4407-4410. Wolf, D.E., Edidin, M., and Hendyside, A.H. 1981. Changes in the organization of mouse egg plasma membrane upon fertilization and first cleavage: Indications from the lateral diffusion rates of fluorescent lipid analogs. Dev. Biol. 85:195-198. Wu, E.-S., Tank, D.W., and Webb, W.W. 1982. Unconstrained lateral diffusion of concanavalin A receptors on bulbous lymphocytes. Proc. Natl. Acad. Sci. U.S.A. 79:4962-4966. Yguerabide, J., Schmidt, J.A., and Yguerabide, E.E. 1982. Lateral mobility in membranes as detected by fluorescence recovery after photobleaching. Biophys. J. 40:69-75. Zs-Nagy, I., Zhang, X., Kitani, K., and Nonomura, Y. 1995. The influence of dystrophin on lateral diffusion of proteins in sarcolemma of L-185 and C2 myoblasts and mature striated muscle cells of rats and mice, as measured by FRAP technique. Biochem. Biophys. Res. Commun. 215:67-74. Zs-Nagy, I., Tanaka, S., and Kitani, K. 1999. Age-dependence of the lateral diffusion coefficient of concanavalin-A receptors in the plasma membrane of ex vivo prepared brain cortical nerve cells of BN/BiRijHsd rats. Exp. Brain Res. 124:233-240.
Contributed by György Lustyik University of Pécs, Faculty of Medicine Pécs, Hungary
Image Cytometry Instrumentation
2.12.15 Current Protocols in Cytometry
Supplement 16
CHAPTER 3 Safety Procedures and Quality Control INTRODUCTION
T
his chapter is designed to formalize what should be considered the minimum procedures for safety and quality control in flow cytometry and image analysis. Quality control was once the final process in manufacturing and process control, as well as the last hurdle faced by scientists as technology became more complex and more decisions were automated. Times have changed: without clear guidelines, well-documented procedures, and certified reagents, it is no longer possible to satisfy the relevant certification authorities. As a result, quality control must now be considered a primary rather than secondary objective. For clinical laboratories, where results are used directly for patient evaluation, diagnosis, and treatment, there can be no room for error or for poorly documented assay systems. Methods must be detailed and clearly documented, reagents must conform to strict standards, and issues such as reagent expiration dates must be considered. In a research laboratory, the same strict guidelines will ensure quality results. Moreover, in the past several years, as more decisions have come to be made based upon results of tests run on semiautomated high-technology instruments, instrument and reagent manufacturers have changed their practices as they have faced increasing regulation in the manufacture of their products.
Issues covered in this chapter include the historical basis for quality control and the scientific rationales that have driven changes in quality control procedures. Examples of quality control monitoring procedures are suggested in UNIT 3.1 and sample graphs for use in the laboratory are provided in UNIT 3.2. UNIT 3.3 presents techniques for testing the aerosol containment of cell sorters, to help minimize biohazards for workers engaged in sorting potential human pathogens. With the continuing increase in the sorting of viable human cells, it is important for cytometrists to be aware of the potential dangers. These procedures should be employed in all laboratories where such work is done, and every cytometry technician should be required to read this unit before being asked to sort viable human material. It should also be kept in mind that pathogens are not the only potential danger. Many commonly used fluorescent dyes and biological reagents are hazardous as well (UNIT 3.4). UNIT 3.5 presents an easy and inexpensive alternative method of detecting aerosol contamination that produces immediate results. A simple suspension of a commercially available resin which fluoresces orange is run like a normal sample for sorting. Contamination is detected by visualization with a black light source and by examination of slides under a fluorescent microscope. Knowledge of microbiology is not needed. Safety in the laboratory is an area of increasing concern for workers and regulators alike. Many of the reagents and fluorochromes encountered in cytometry are toxic, carcinogenic, mutagenic, and/or teratogenic; new materials with undetermined health and safety properties appear constantly. UNIT 3.4 outlines basic minimal safety concerns and procedures for laboratory workers: chemical storage, appropriate facilities, and personal protective equipment. Extensive tables list known hazards for laboratory chemicals, chemical incompatibilities, and chemical resistance of commonly used laboratory gloves. Protocols outline disposal methods for a number of hazardous chemicals and biological stains, decontamination methods, and detection methods for verification. The Literature Cited contains a lengthy list of reference texts on various aspects of laboratory safety. Safety Procedures and Quality Control Contributed by J. Paul Robinson Current Protocols in Cytometry (2002) 3.0.1-3.0.2 Copyright © 2002 by John Wiley & Sons, Inc.
3.0.1 Supplement 19
As the chapter expands, an attempt will be made to keep users of cytometry-related instrumentation abreast of the latest regulations, quality control procedures, and methods for monitoring these procedures in both research and clinical environments. Particular emphasis will be laid on current quality control requirements for clinical cytometry. J. Paul Robinson
Introduction
3.0.2 Supplement 19
Current Protocols in Cytometry
Principles of Quality Control The following is a general overview of the principles of quality control in the clinical laboratory arena. Please refer to UNITS 1.1 & 1.3 for definitions of terms and flow cytometry–specific designations. Here, the reader will become aware of historical clinical laboratory procedures and how they apply to cytometry. A basic premise of quality control is that the reported laboratory values should correspond to the correct or expected values. To examine this in more analytical terms, let us use specimens for which the true (expected) value of an analyte is known. The graph of reported known values should then be a straight line with a slope of 1.00. In other words, the reported values of those specimens analyzed by the laboratory should correspond exactly to the expected values. However, all procedures are subject to a variety of analytical biases or inaccuracies. More specifically, a multitude of analytical procedures are subject either to constant or proportional bias, or both. Proportional bias is when the reported values are higher or lower than the expected values by an amount proportional to the concentration of the analyte. Constant bias is when the reported values are higher or lower than the expected values by a constant amount at all concentrations of an analyte. The goal is for a laboratory to define an “operational line” (Grannis et al., 1972), because, for a particular procedure, every laboratory has results that tend to fall along some line. For each analytical procedure, each laboratory customarily operates with a certain degree of bias that causes its results to be distributed along some operational line (Fig. 3.1.1). One primary objective of quality assurance programs is to determine a laboratory’s operational line for an analytical method. If an operational slope of 1.00 cannot be attained, then the laboratory’s customary operational line should at least be maintained in a reproducible manner. In addition to analytical biases, laboratory analyses are also subject to imprecision, or random variability. Typically, results are distributed between constant limits. The knowledge of how various sources of analytical bias and variability affect the accuracy and precision of the operational line can be most helpful in identifying and correcting analytical problems as they arise. Several studies have indicated that as many as 2% of all clinical laboratory analyses may be erroneous as a result of mistakes—defined Contributed by Anne A. Hurley Current Protocols in Cytometry (1997) 3.1.1-3.1.2 Copyright © 1997 by John Wiley & Sons, Inc.
as human error rather than deficiencies in the analytical system (Grannis et al., 1972; Ladenson, 1975; Whitehurst et al., 1975). Because specimen processing involves a number of steps, mistakes may occur at numerous points (see Chapter 5). Control specimens, especially samples obtained from the same pool, must be used in monitoring analytical bias and variability. This concept, introduced nearly four decades ago, is still the most widely applied quality control technique (Henry and Segalove, 1952; Henry, 1959). A number of early studies (Belk and Sunderman, 1947; Wootton and King, 1953) examined the performance of laboratories with respect to the same solutions of analytes. The results revealed clear evidence of substantial systematic differences between laboratories. Levey and Jennings (1950) showed that variability in the measured values could be documented when samples of the same serum pool were analyzed in a single laboratory over a period of time. These early studies established the important principle that a laboratory’s analyses could be compared with those of other laboratories, or with its own prior analyses, by use of quality control samples. The studies of Belk led directly to the development of interlaboratory comparison programs (Dorsey, 1975) and those of Levey led to the development of intralaboratory quality control programs. However, even with analytes of a more “defined” nature such as those in clinical chemistry, these programs have not been developed without much difficulty. The effectiveness of quality control programs requires that the samples be essentially equivalent to one another, that the analytes be stable in storage over some time, and that the material be available in sufficient quantities to be used by one laboratory or many laboratories over a long period of time. Therefore, quality control can be divided into two major types: internal (intralaboratory) quality control and external (interlaboratory) quality control. Virtually any assay can be charted along an operational line to follow how controls are responding in relation to the expected value. However, it is important to select for monitoring those instrument or reagent parameters that are relevant—i.e., that will make a significant enough difference in results to cause poor patient care. For flow cytometry, such parameters should include: PMT voltage for each fluorescence channel; side scatter (SSC); FSC (for-
UNIT 3.1
Safety Procedures and Quality Control
3.1.1
Analyzed value
A
Analyzed value
C
B
50 40 30 20 10
D
50 40 30 20 10
10 20 30 40 Expected value
50
10 20 30 40 Expected value
50
Figure 3.1.1 Concept of operational line. In each graph, the solid line represents an operational slope of 1.00 and the dashed line represents the laboratory’s values (i.e., the operational line). (A) When a laboratory’s reported (analyzed) values for a series of specimens are plotted against the expected values, the data should fall along the straight-line slope of 1.00. (B) In this graph the dashed line demonstrates the effects of a proportional bias in which the reported values are higher than the expected values. (C) This graph illustrates the effect of a constant bias, where the reported values are higher than the expected values by a constant amount at all concentrations of an analyte. (D) Because many procedures are subject to either constant or proportional biases, or both, this graph demonstrates how combined constant and proportional biases may affect the correlation of reported and expected values.
ward scatter) gains; compensation settings; separation between “negative” and “positive” populations; sensitivity; CV (coefficient of variation) of control material in the fluorescence channels; and mean channel/fluorescence intensity of known cells or particles.
LITERATURE CITED Belk, W.P. and Sunderman, F.W. 1947. A survey on the accuracy of chemical analysis in clinical laboratories. Am. J. Clin. Path. 17:854. Dorsey, D.B. 1975. The evaluation of proficiency testing in the USA. In Proceedings of the Second National Conference on Proficiency Testing. Information Services, Bethesda, Md. Grannis, G.F., Gruemer, H.D., Lott, J.A., Edison, J.A., and McCabe, W.C. 1972. Proficiency evaluation of clinical chemistry laboratories. Clin. Chem. 18:222. Principles of Quality Control
Henry, R.J. 1959. Use of the control chart in clinical chemistry. Clin. Chem. 5:309.
Henry, R.J. and Segalove, M. 1952. The running of standards in clinical chemistry and the use of the control chart. J. Clin. Path. 5:305. Ladenson, J.H. 1975. Patients as their own controls: Use of the computer to identify laboratory error. Clin. Chem. 21:1648. Levey, S. and Jennings, E.R. 1950. The use of control charts in the clinical laboratory. Am. J. Clin. Path. 20:1059. Whitehurst, P., DiSilvio, T.V., and Boadjian, G. 1975. Evaluation of discrepancies in patient’s results: An aspect of computer-assisted quality control. Clin. Chem. 21:87. Wootton, I.D.P. and King, E.J. 1953. Normal values for blood constituents: Inter-hospital differencs. Lancet 1:470.
Contributed by Anne A. Hurley Comprehensive Cytometric Consulting Ballwin, Missouri
3.1.2 Current Protocols in Cytometry
Components of Quality Control Quality control within the clinical laboratory should be thought of as a system for assessing the quality of total laboratory performance. An effective system of quality control was developed by Sax et al. (1967), modified by Allen et al. (1969), and further refined by Grannis et al. (1972). In this system, each operator in the clinical laboratory routinely includes in each analytical run quality control specimens whose expected values are known. The purpose of the control specimens is to aid the operator in deciding whether or not the analytical system is producing reliable results for that particular assay. Additionally, other quality control specimens—which may be commercially available controls or samples of
UNIT 3.2
special diagnostic significance—should be interspersed randomly among the clinical specimens. Data from all of these types of specimens are assessed on a daily basis. Although the system just outlined may appear suitable only for larger laboratories with a very clinical focus, rather than cytometry laboratories, the basic concept of having a regular review of the quality control data and parameters using known or unknown specimens is applicable to all laboratories. Generally speaking, control materials should target the values of analytes at or near the clinical decision levels. Effective quality control depends on having control specimens that are highly reproducible. Many different
A +2SD +1SD mean –1SD –2SD
B
trend
+2SD +1SD mean –1SD –2SD
C
shift
+2SD +1SD mean –1SD –2SD
Figure 3.2.1 Examples of Levey-Jennings control charts. (A) Normal distribution about a mean. (B) Trend in which there is a progressive drift of the reported values from the mean value. (C) Shift in which there is an abrupt change from the reported mean value.
Contributed by Anne A. Hurley Current Protocols in Cytometry (1997) 3.2.1-3.2.4 Copyright © 1997 by John Wiley & Sons, Inc.
Safety Procedures and Quality Control
3.2.1
( ) +2SD +1SD mean –1SD –2SD
Figure 3.2.2 Example of Westgard single rule. The run is rejected as “out of control” because the value in parentheses is outside the ±2SD limit.
statistical techniques can be applied to help decide when control data indicate patient analyses are “in control” or “out of control.” Basically, these techniques allow for decision criteria that determine whether to accept or reject an analytical run. In short, it is of interest to know the probability for rejection when there are no analytical errors (false rejection) and the probability of detecting certain analytical errors, particularly those that would invalidate the medical usefulness of a test result. As both random and systematic errors can occur, it is necessary to assess the probabilities for detecting both kinds of errors. The performance of a quality control test can be optimized for detecting frequency of errors in the same manner that performance of a diagnostic test can be optimized for detecting prevalence of disease. A predictive value model for a quality control test has been described by Westgard and Groth (1983). In this model, diagnostic sensitivity (DSens) describes how often a test gives a positive result when a patient has the problem of interest. Diagnostic sensitivity is defined as DS ens =
TP TP + FN
where TP represents the number of true positives and FN represents the number of false negatives. Diagnostic specificity (DSpec) describes how often a patient not having the problems gives a negative test result. Diagnostic specificity is defined as
DS pec =
Components of Quality Control
TN TN + FP
where TN represents the number of true negatives and FP represents the number of false positives.
In addition to the concepts of diagnostic sensitivity and specificity, the concepts of predictive value and efficiency can be applied. The frequency of errors occurring with an analytical method is analogous to the prevalence of disease in the application of a diagnostic test. Using these and other more formal assessments, it is possible to judge the suitability of a control technique for achieving a specified level of quality. There are basically five steps to follow when assessing both image and flow cytometry systems performance. 1. Choose relevant instrument parameters to monitor (see UNIT 3.1). 2. Select the best quality control materials for each instrument parameter. 3. Establish or validate expected values of control materials and determine frequency of testing. 4. Set quality control acceptability or control rules. 5. Document and monitor results for change from expected variability. Although tabular records are certainly used, control charts with graphical displays are easier to interpret. These charts generally display the calculated result (or analytical result) as a function of time. The Levey-Jennings chart (Levey and Jennings, 1950) has been the most widely used (Fig. 3.2.1). Basically, there are two variations on interpreting and using Levey-Jennings charts. The first is called the Westgard single-rule QC procedure. This states that a run is rejected as “out of control” when one control measurement in the group exceeds the control limits of ±2SD (i.e., twice the standard deviation; Fig. 3.2.2). Another variation, the Westgard multi-rule QC procedure, states that run is rejected if: (1) two consecutive control measurements exceed the same limit—e.g., +2SD or −2SD; (2) four
3.2.2 Current Protocols in Cytometry
A blank Levey-Jennings chart. Figure 3.2.3
–2SD
–1SD
mean
+1SD
+2SD
consecutive control measurements exceed the same mean—e.g.,+1SD or −1SD; or (3) ten consecutive control measurements fall on one side of the mean. The Westgard multi-rule technique can uncover both random and systematic errors. In flow and image cytometry, these more classical quality control analyses are only starting to be applied to instruments and data. In most cases at this time, a Levey-Jennings graph of pertinent data is sufficient for understanding how well the cytometry system is performing. A blank Levey-Jennings chart is included (Fig. 3.2.3) for laboratory use. Some of the more critical variables in flow cytometric analysis do not easily lend themselves to mathematical reasoning as discussed above. These include specimen collection, transport and storage, sample preparation and storage, measurement, data analysis, and reporting and interpretation. For these variables, a standardized procedure or reference book should be established by the laboratory to specify these procedures and provide criteria for accepting or rejecting the specimen or data at every step. Obviously, the best quality control system in terms of such items as optics, gains, linearity, and compensation will still provide poor results if the sample considerations and interpretation are not standardized. For example, it is critical to set up accept/reject criteria for specimens (e.g., based on the age of the sample and whether it was refrigerated or heated) as a very first step. It is equally important to establish a procedure manual so all samples are processed in the same manner by laboratory personnel. An example of how this would work would be to ensure that test tubes are covered in foil or put in a dark place during the incubation stage so that quenching is less likely to occur. These issues have been recognized as extremely important to the integrity of the data and have been addressed extensively in several documents, listed in Table 3.2.1. The ultimate purpose of any method of quality control is to allow the laboratory to make the critical decisions of whether or not to release results to the requesting clinicians. Quality control programs provide the objective database from which these decisions may be made in confidence, and it is through making the correct decisions that the clinician/user is assured of the quality of the laboratory result.
Safety Procedures and Quality Control
3.2.3 Current Protocols in Cytometry
Table 3.2.1
Documents Relevant to Quality Control in Flow Cytometry
Document
Publisher
CAP Flow Cytometry 1994 Inspection Checklist (Section XI) Catalog #B-234.
College of American Pathologists Commission on Laboratory Accreditation 325 Waukegan Road Northfield, IL 60093
NCCLS Document C24-A. Internal Quality Control Testing: Principles and Definitions. Approved Guideline, Vol 11 (No 6).
National Committee for Clinical Laboratory Standards 771 East Lancaster Avenue Villanova, PA 19085
NCCLS Document H42-T. Clinical Applications of Flow Cytometry: Quality Assurance and Immunophenotyping of Peripheral Blood Lymphocytes. Tentative Guideline, Vol 12 (No 6).
National Committee for Clinical Laboratory Standards 771 East Lancaster Avenue Villanova, PA 19085
1994 Revised Guidelines for the Performance of CD4+ T Cell Determinations in Persons with Human Immunodeficiency Virus (HIV) Infections. MMWR 1994; 43 (no. RR-3) pp. 1-24
Centers for Disease Control and Prevention 1600 Clifton Road N.E. Atlanta, GA 30333
LITERATURE CITED Allen, J.R., Earp, R., Farrell, C.E., and Gruemer, H.D. 1969. Analytical bias in a quality control scheme. Clin. Chem. 15:1039. Grannis, G.F., Gruemer, H.D., Lott, J.A., Edison, J.A., and McCabe, W.C. 1972. Proficiency evaluation of clinical chemistry laboratories. Clin. Chem. 18:222. Levey, S. and Jennings, E.R. 1950. The use of control charts in the clinical laboratory. Am. J. Clin. Pathol. 20:1059.
Sax, S.M., Dorman, L., Lebenson, D.D., and Moore J.J. 1967. Design and operation of an expanded system of quality control. Clin. Chem. 13:825. Westgard, J.O. and Groth, T. 1983. A predictive value model for quality control: Effects of the prevalence of errors on the performance of control procedures. Am. J. Clin. Pathol. 80(1):49-56.
Contributed by Anne A. Hurley Comprehensive Cytometric Consulting Ballwin, Missouri
Components of Quality Control
3.2.4 Current Protocols in Cytometry
Testing the Efficiency of Aerosol Containment During Cell Sorting
UNIT 3.3
Deflected-droplet (jet-in-air) flow sorters (see UNIT 1.2) are able to separate cell populations on the basis of their physical properties or by exploiting differences in cell-surface and intracellular structures and molecular expression. Generally, in order to identify a population of interest on the instrument, cells are incubated with fluorescently labeled antibodies or ligands or with fluorescent dyes that interact with cellular components. Flow sorters inject these cells into a fluid jet that emerges under pressure from a rapidly vibrating nozzle. This serves to break up the stream into droplets. If the droplets contain cells that meet preselected criteria for the desired cell types, they become electrostatically charged and are deflected into receptacles. Thus, the production of small droplets and microdroplets (i.e., aerosols) is part of the normal operation of a cell sorter. In addition, secondary aerosols are formed when the undeflected center stream and the deflected side streams splash into receptacles. Production of aerosols can increase substantially during failure modes of the cell sorter— e.g., when a partially clogged sort nozzle causes a deflection in the stream, which then strikes an obstacle, or when air is present in the fluidics system. If aerosols are released into the room, they pose a potential hazard to the sorter operator and to others who may be present in the room, because many dyes that are used to identify cell subpopulations are toxins, carcinogens, or mutagens. Furthermore, it is often necessary for assessment of cellular function to sort viable biologic specimens that can harbor known pathogens (e.g., hepatitis or human immunodeficiency viruses) or unknown organisms that could potentially be transmitted through inhalation of aerosol droplets. Most newer models of commercially available cell sorters incorporate features designed to decrease the production of aerosols and prevent their escape into the room—e.g., an evacuated catch tube for the central, undeflected sort stream and an enclosed sort-sample compartment. However, the effectiveness of aerosol-containment measures on flow sorters must be verified in order to minimize hazard exposure for laboratory personnel. Testing of aerosol containment and the establishment of a safety protocol may also be necessary to obtain approval from an institutional biosafety committee and/or regulatory agency for cell sorting of potentially biohazardous specimens. This unit describes two procedures for the assessment of aerosol containment on jet-in-air cell sorters. In both of these procedures, lytic T4 bacteriophage is run through the instrument at high concentrations to tag aerosol droplets. The instrument is tested in regular sorting mode and in a failure mode simulating a condition that can arise from a clogged sort nozzle, air bubbles in the fluidic system, or other instrument malfunctions. In the first procedure, petri dishes with confluent lawns of T4-susceptible Escherichia coli are placed in and near the flow sorter at points where aerosols are formed and where they could potentially escape. Aerosols are detected by plaque formation in the E. coli lawn, which results when T4 bacteriophage lands on and lyses the E. coli bacteria (see Basic Protocol). The second procedure uses an air sampler for collecting room air and depositing it upon lawns of E. coli during the T4 aerosol containment test (see Alternate Protocol). Preparation of the T4 bacteriophage stock solution is also described (see Support Protocol 1), as well as initiation of the E. coli broth culture and generation of bacterial lawns (see Support Protocol 2), and titration of the T4 bacteriophage stock solution and determination of the quantity of plaque-forming T4 bacteriophage emerging from the cell sorter per minute (see Support Protocol 3). Some of the biological materials for the aerosol containment test take >1 day to prepare and procedures have to be done on consecutive days; therefore, timing is important (see Time Considerations). The Contributed by Ingrid Schmid, Lance E. Hultin, and John Ferbas Current Protocols in Cytometry (1997) 3.3.1-3.3.15 Copyright © 1997 by John Wiley & Sons, Inc.
Safety Procedures and Quality Control
3.3.1 Supplement 1
aerosol-containment test should be repeated every one to three months or whenever repairs or upgrades have been done on the instrument that may affect aerosol escape—e.g., replacement of the sort-chamber door or installation of a new drive head. Preparation of the microbial materials used in this unit is of intermediate complexity; therefore experience in microbiology or advice from a microbiologist will facilitate success. The test itself is fairly simple, but requires expertise in cell sorting. Note that fluidic-switching (or closed-flow-cell) sorters (e.g., FACSort from Becton Dickinson Immunocytometry Systems or Partec sorter from Partec Instruments) do not need to be tested because they do not generate aerosols during the sorting process. BASIC PROTOCOL
TESTING OF AEROSOL CONTAINMENT IN FLOW-SORTING SYSTEMS BY MONITORING GRAVITATIONAL DEPOSITION OF DROPLETS This protocol assesses containment of aerosol droplets that rapidly settle through gravitational forces (see Background Information for discussion of the significance of droplet size and mobility) on a flow sorter in regular sorting mode and in failure mode. Materials Dilution broth (see recipe) T4 bacteriophage stock solution (see Support Protocol 1), appropriately titered (see Support Protocol 3) 12 × 75–mm culture tubes (Falcon) Electronic balance Deflected-droplet (jet-in-air) cell sorter to be tested (e.g., FACStar and FACSvantage from Becton Dickinson, MoFlo from Cytomation, EPICS series from Coulter, and Ortho Systems from Ortho Diagnostics; see UNIT 1.2) Petri dishes containing confluent E. coli lawns (see Support Protocol 2) and dish with bottom agar only as control 35° to 37°C warm room or incubator Set up the cell sorter at fast flow rate 1. Weigh a 12 × 75–mm culture tube containing 2 to 3 ml dilution broth on an electronic balance. 2. Place the tube onto the sample-introduction port of the cell sorter and run dilution broth through the system for 10 min. Reweigh the tube to determine consumption of the dilution broth, calculate the flow rate of the instrument at the setting used, then adjust the setting to a high cell-sorting flow rate. 1 mg dilution broth equals 1 ìl. Note that the consumption of the liquid will depend on the instrument and the size of the sort nozzle used. For instance, a fast flow rate for a FACStar cell sorter equipped with a 70-ìm nozzle will be 30 to 40 ìl per min; an average sort flow rate would be ∼20 ìl per min. It is advisable to use a fast flow rate for the containment test to assess the efficiency of aerosol-control measures at high performance levels of the cell sorter. For standardization of the aerosol-containment test, it is important to adjust the instrument to a similar flow rate each time the test is repeated.
Perform the aerosol-containment test in regular sorting mode 3. Distribute six petri dishes containing confluent E. coli lawns (with lids still in place) on the cell sorter as follows: two within the sorting chamber close to the sorting streams; two more immediately outside the sorting-chamber door; and two near the instrument wherever aerosols could potentially escape. Testing Aerosol Containment During Cell Sorting
Each dish must be carefully labeled to identify its location during the test. Initially, it may be advisable to prepare more petri dishes with E. coli lawns and place them at various additional locations—e.g., on the cell sorter near the exhaust lines and in the
3.3.2 Supplement 1
Current Protocols in Cytometry
vicinity of gaps between the sample-compartment housing and the instrument, as well as near the flow sorter and at other locations within the room. If aerosol containment is found to be efficient, six petri dishes will suffice for repeat experiments. It is important to place petri dishes inside the sorting chamber, as these will serve as positive experimental controls. However, it may be necessary to jury-rig a platform to raise the dishes to the level of the sort-stream collection vials. Petri dishes should be handled carefully and kept level at all times, as the soft agar containing E. coli is easily damaged or dislodged, thus making T4 bacteriophage plaque counting unreliable or impossible.
4. Place a 12 × 75–mm culture tube containing 3 to 4 ml T4 bacteriophage stock solution on the sample-introduction port and set the fluid switch to the Run position. Place two collection vials into the sort-sample holder. More than 4 ml T4 bacteriophage stock solution may be needed, depending on the rate of sample consumption.
5. Remove lids from all petri dishes, close the sort-chamber door, and start the mock sort, generating right- and left-deflected streams using instrument-specific software. Run the cell sorter in this mode for 2 hr, performing step 6 each time the collection vials are full. Set up the instrument with a similar drop-drive frequency each time the test is repeated. Set up the mock sort with ∼1,000 sort decisions/sec right and ∼1000 sort decisions/sec left, which will simulate a higher-than-average sort rate (a typical rate is ∼100 to 400 sort decisions/sec each side) in order to assess the efficiency of aerosol containment at high performance levels of the cell sorter. The 2 hr run simulates an average sort time used for many applications. Make sure that the side streams are not hitting the petri dishes, because this will destroy the confluent E. coli lawns.
6. When the collection vials are full, turn off the sort logic and set the fluidic control to Off. Wait at least 3 min for aerosol clearance, then open the sort-chamber door and replace the full vials with empty ones. Close the door and continue the sort. Repeat as necessary during the total 2-hr sort time. The time needed for clearance of aerosol can be checked with bottled smoke (Lab Safety Supply). Opening the door too soon can result in escape of T4 bacteriophage and subsequent plaque formation on the petri dishes that are placed outside near the sort chamber. For added protection against false-positive T4 plaques forming on these petri dishes, it is advisable to put their lids on before the door is opened. Remember to take the lids off again when the mock sort is continued.
7. After the final 2-hr sort in regular sorting mode, stop the instrument as in step 6, waiting at least 3 min before opening the sort chamber door. Replace the lid on each of the petri dishes and remove them from the cell sorter. Perform the aerosol-containment test in failure mode 8. Prepare and distribute six new petri dishes with E. coli lawns as described in step 3. Prepare the instrument for delivery of T4 bacteriophage as in step 4, using the same flow rate as for the regular sorting mode. 9. Remove lids from all petri dishes, close the sort-chamber door, and initiate the failure-mode mock sort by keeping the deflection voltage on and turning off the drop drive, which leads to maximum fanning of the streams. Run the cell sorter in this mode for 15 min. In the case of an actual instrument failure, e.g., a partially clogged sort nozzle, a sort is usually stopped within <1 min; therefore the 15-min mock failure mode is assessing aerosol containment under extreme failure conditions. An alternative failure mode can be generated by directing the undeflected center stream towards the wall of the catch tube; however, it is difficult to obtain similar conditions for repeat experiments using this method.
Safety Procedures and Quality Control
3.3.3 Current Protocols in Cytometry
Supplement 1
Figure 3.3.1 Example of plaque formation on a confluent E. coli lawn. This petri dish was exposed to the sorting streams, which were near the upper right area of the dish. Note that the density of plaque formation is greatest near the aerosol source.
10. After 15 min, replace the lids on the petri dishes located near the sort-chamber door and stop the sort as described in step 6, waiting 3 min before opening the sort-chamber door. Replace the lids on the rest of the petri dishes and remove them from the cell sorter. 11. Collect all the petri dishes that were used during the regular-sort-mode and failuremode tests and incubate them 18 hr at 35° to 37°C in a warm room or incubator. Include one petri dish without E. coli and one dish with an E. coli lawn that was not used for the instrument test as experimental controls. 12. Count the plaques on each petri dish formed by T4 bacteriophage that have landed on the E. coli lawn. Plaques are recognizable as circular clear areas in the bacterial lawn, resulting from bacteriophage lysis of E. coli (Fig 3.3.1). Note that petri dishes placed inside the sorting chamber should always show plaques. The dishes placed outside the instrument should be free of plaques if aerosol containment on the cell sorter is efficient (see Anticipated Results and see Troubleshooting). If the test shows that aerosol containment was incomplete, modify the flow sorter to achieve complete aerosol containment. This may involve installation of a vacuum pump for evacuation of the sort chamber or sealing gaps between the sort chamber and the surrounding environment. Contact the manufacturer of the instrument for advice. Repeat the containment test until satisfactory results are achieved. ALTERNATE PROTOCOL
TESTING OF AEROSOL CONTAINMENT IN FLOW-SORTING SYSTEMS BY USE OF AN AIR SAMPLER In contrast to the Basic Protocol, which relies on gravitational forces for the settling of larger aerosol droplets, this alternate protocol tests for the presence of submicrometersized aerosol droplets (see Background Information for the significance of droplet size and mobility). Because these droplets may remain suspended in room air for a prolonged time, a commercially available air sampler is used to direct the air onto the E. coli–inoculated petri dishes. The air sampler should be equipped with a two-port manifold so that air immediately adjacent to the sorting chamber can be simultaneously tested with room air at greater distance—e.g., ∼1 m from the sort chamber. This protocol is performed in conjunction with the Basic Protocol; however, it may be performed separately if desired.
Testing Aerosol Containment During Cell Sorting
Additional Materials (also see Basic Protocol) Andersen single-stage air sampler (model #N6IACFM; Grayseby-Andersen) with two-port manifold
3.3.4 Supplement 1
Current Protocols in Cytometry
1. Set up the air sampler according to the manufacturer’s instructions and place one uncovered petri dish with confluent E. coli lawn in the sampling stage. Sample room air for 10 min. This dish will serve as the negative control.
2. Place two new uncovered petri dishes with confluent E. coli lawns in the sampling stages. Place one stage immediately outside the sorting chamber, preferably at a location near the sorter operator. Place the other at a distance from the instrument— e.g., 1 m away. 3. Sample the room air for the last 10 min of the regular-sorting-mode test (see Basic Protocol, steps 4 to 6). If microdroplets have escaped they will be at their highest concentration during this time interval. Under standard operating conditions, testing room air for 10 min is equivalent to sampling 0.3 m3 of air. The sampling time may be increased if desired; however, depending on the prevalence of spores in the room, the petri dishes may overgrow with airborne contaminants such as fungi.
4. Remove the petri dishes from the air sampler and replace their lids. 5. Place two new uncovered petri dishes with confluent E. coli lawns in the sampling stages and situate the stages as described in step 2. 6. Sample the room air for the last 10 min of the failure-mode test (see Basic Protocol, steps 8 to 9). 7. Place one more uncovered petri dish with confluent E. coli lawn into the sampling stage. Re-initiate the failure-mode mock sort (see Basic Protocol, steps 8 to 9). Open the sort-sample door and sample room air close to the open door for 10 min. This dish will serve as the positive control.
8. Collect and incubate all petri dishes, then count plaques (see Basic Protocol, steps 11 to 12). PREPARATION OF T4 BACTERIOPHAGE STOCK This protocol describes generation of the T4 bacteriophage suspension that is used for testing aerosol containment in the cell sorter (see Basic Protocol and Alternate Protocol). T4 bacteriophage is propagated in a log-phase culture of E. coli and leads to lysis of infected bacteria. Remaining uninfected bacteria are lysed through addition of chloroform and removed by centrifugation.
SUPPORT PROTOCOL 1
Materials T4-susceptible E. coli (ATCC #11303) Nutrient broth (see recipe) T4 bacteriophage, lyophilized (ATCC #11303-B4) Chloroform 25- and 250-ml Erlenmeyer flasks 10.2 × 10.2–cm sterile 12-ply gauze pads (e.g., Johnson & Johnson Steri-Pads) 35° to 37°C warm room and orbital shaker or shaking incubator 50-ml conical polypropylene centrifuge tubes, sterile Sorvall Omnispin R centrifuge with 42011-type rotor (or equivalent) NOTE: All reagents and equipment coming into contact with live cells must be sterile, and proper sterile technique should be followed accordingly.
Safety Procedures and Quality Control
3.3.5 Current Protocols in Cytometry
Supplement 1
1. Rehydrate one vial of T4-susceptible E. coli with 0.3 to 0.4 ml nutrient broth. Mix well, then transfer 0.1 ml of the mixture to a sterile 25-ml Erlenmeyer flask containing 10 ml nutrient broth. Insert a sterile gauze pad into the opening of the flask and secure it with tape. Good aeration during incubation is important; therefore do not fill the Erlenmeyer flask to more than 1⁄3 to 1⁄2 of capacity. If additional E. coli is required, several individual flasks should be set up. The remaining rehydrated E. coli suspension can be stored indefinitely frozen at −20°C. Before freezing, a streak of this suspension can be put onto a nutrient agar slant (Becton Dickinson), incubated 18 hr at 35° to 37°C in a warm room or incubator, then stored for at least six months at 4°C. This slant culture can be used for initiation of the E. coli culture used in generating the confluent lawns (see Support Protocol 2).
2. Incubate the flask 18 hr at 35° to 37°C in a warm room on an orbital shaker at 150 rpm or in a shaking incubator at an equivalent temperature and speed. After this incubation period the culture should appear turbid.
3. Transfer 0.5-ml aliquots of the E. coli broth culture into sterile 250-ml Erlenmeyer flasks, each containing 50 ml nutrient broth. Plug each flask with a sterile gauze pad and incubate 18 hr as in step 2. At the end of the incubation period, the cultures should be in log phase. The number of flask cultures will depend on how much T4 bacteriophage stock will be needed. The T4 bacteriophage stock suspension can be kept indefinitely; therefore it is advisable to initiate several flasks.
4. Rehydrate one vial of lyophilized T4 bacteriophage with 0.5 ml nutrient broth. Mix well and inoculate each of the log-phase cultures of E. coli (from step 3) with 0.1 ml T4 bacteriophage suspension. The remaining rehydrated T4 bacteriophage suspension can be stored for at least six months at 4°C and used for a repeat expansion of T4 by inoculation of a log culture of E. coli. This is important in case the resulting T4 bacteriophage stock suspension’s titer is too low (<1 × 109 plaque-forming units (pfu)/ml; see Support Protocol 3) for it to be used for testing of aerosol containment on the cell sorter.
5. Incubate the E. coli/T4 bacteriophage cultures as in step 2 until lysis of E. coli is complete as indicated by clearing of the turbid suspension (∼12 hr). Usually, this will take 12 hr; however, the cultures may be incubated overnight if they are started in the afternoon.
6. Transfer the broth from the flasks into sterile 50-ml polypropylene tubes. Add 12 drops chloroform to each tube and shake vigorously. Centrifuge 20 min at 2000 × g, room temperature, to remove cellular debris. 7. Transfer the supernatant, which contains the T4 bacteriophage stock suspension, to separate tubes (filled as close to the brim as possible). Cap, seal with Parafilm, and store indefinitely at 4°C. It is important to keep the air space over the suspension to a minimum to preserve T4 bacteriophage activity. Before the stock suspension is used for aerosol testing on the cell sorter, assess its titer by serial dilution as described below (see Support Protocol 3) to verify that it exceeds 1 × 109 pfu/ml.
Testing Aerosol Containment During Cell Sorting
3.3.6 Supplement 1
Current Protocols in Cytometry
PREPARATION OF E. COLI LAWNS First, this protocol outlines the preparation of agar plates in petri dishes using liquified bottom agar. Next, a log-phase broth culture of E. coli is initiated and poured onto the surface of the agar plates. Finally, the plates are incubated at 35° to 37°C to generate confluent E. coli lawns.
SUPPORT PROTOCOL 2
Materials Bottom agar (see recipe) Soft agar (see recipe) 40° to 50°C water bath 100-mm petri dishes Additional reagents and equipment for growing E. coli in nutrient broth (see Support Protocol 1) NOTE: All reagents and equipment coming into contact with live cells must be sterile, and proper sterile technique should be followed accordingly. 1. Liquify bottom agar in a 40° to 50°C water bath. Using a sterile pipet, put ∼20 ml bottom agar into each 100-mm petri dish to be used for plating. Leave the covers ajar until the agar has solidified (to prevent accumulation of moisture), then close the lids. Alternatively, bottom agar can be liquified by heating on a hot plate or in a microwave oven. However, its temperature has to be carefully monitored to prevent boiling over. The number of petri dishes to be prepared will depend on how many are needed for the aerosol-containment test. Preparation of 29 dishes will provide twelve for aerosol-containment testing via gravitational settling (six for testing of the regular sorting mode and six for testing of the failure sorting mode; see Basic Protocol), six for the aerosol-containment test using an air sampler (see Alternate Protocol), three each for the two titrations that are done on the T4 bacteriophage stock suspension (see Support Protocol 3), and five extra dishes as controls and replacements in case some plates are accidentally damaged. The petri dishes containing bottom agar alone may be stored upside down ≤6 months at 4°C. They may be placed in a plastic bag to prevent dehydration. If the dishes have been stored for a prolonged period of time, it is advisable to inspect them before use for contamination or cracks that are due to drying out. Alternatively, ready-to-use nutrient agar plates are commercially available, but in the authors’ experience have proven unsatisfactory for this assay because such plates result in uneven plaque formation—i.e., circular clear plaques alternating with circular opaque plaques.
2. Prepare log-phase cultures of E. coli in nutrient broth (see Support Protocol 1, steps 1 to 3). Alternatively (and preferably) the broth culture can be initiated from a nutrient agar slant stored at 4°C (see Support Protocol 1, step 1 annotation). Transfer E. coli to a 250-ml Erlenmeyer flask containing 50 to 100 ml nutrient broth by touching the surface of the slant with an inoculating loop; do not scrape excessive material from the slant. Incubate culture 18 hr on an orbital shaker in a warm room at 35° to 37°C.
3. Heat ∼110 ml soft agar in a 40° to 50°C water bath. Alternatively, heat soft agar on a hot plate or in a microwave oven, using an adequately sized container to avoid boiling over. The temperature must be monitored carefully with an ethanol-sterilized thermometer because E. coli will be killed if added to soft agar that is warmer than 50°C.
4. Add ∼2 ml of the E. coli broth culture to the soft agar and mix by swirling gently. Add 4 ml of the mixture to each petri dish containing bottom agar. Let the petri dishes cool with the lids ajar. Make sure that the mixture covers the entire surface of the dish. Work swiftly because the soft agar will solidify quite rapidly.
Safety Procedures and Quality Control
3.3.7 Current Protocols in Cytometry
Supplement 1
5. Close the lids and incubate the dishes 2 to 4 hr at 35° to 37°C. If the petri dishes cannot be used immediately after ∼4 hr of incubation at 35° to 37°C they can be stored up to 6 hr at 4°C, then used for the instrument test. When the petri dishes are ready to be used, their surface will have a slightly opaque appearance, which indicates the formation of E. coli lawns. Do not let the E. coli lawns overgrow, because this may impede optimal plaque formation. Handle dishes carefully and do not invert them, because the soft agar is easily dislodged or damaged. SUPPORT PROTOCOL 3
TITRATION OF THE T4 BACTERIOPHAGE SUSPENSION TO BE USED IN THE AEROSOL-CONTAINMENT TEST In this protocol, the number of plaque-forming units (pfu) per ml of the T4 stock suspension is determined by serial dilution. This titration establishes whether the concentration of T4 bacteriophage is high enough for the suspension to be used in the aerosolcontainment test on the cell sorter (see Basic Protocol and Alternate Protocol) and whether loss of activity has occurred in stored T4 stocks. In a second titration, the throughput (in pfu/min) of viable T4 bacteriophage particles on the cell sorter is measured by using serial dilution, to insure that adequate numbers of infectious T4 are present in the stream leaving the sort nozzle. This is important because shearing stress during sorting and certain sheath fluids used for cell sorting—e.g., PBS—may adversely affect T4 infectivity. Both titrations utilize petri dishes with E. coli lawns as the detection system. Materials Dilution broth (see recipe) T4 bacteriophage stock suspension (see Support Protocol 1) Petri dishes containing bottom agar (pre-prepared; see Support Protocol 2, step 1) Hanks’ balanced salt solution (HBSS; APPENDIX 2A) or other sheath fluid for flow cytometry that will maintain viability of T4 bacteriophage 12 × 75–mm culture tubes 15-ml culture tubes Cell sorter to be tested (see Basic Protocol 1) Additional reagents and equipment for preparing E. coli lawns (see Support Protocol 2) and for determining and adjusting flow rate of cell sorter (see Basic Protocol, steps 1 to 2) NOTE: All reagents and equipment coming into contact with live cells must be sterile, and proper sterile technique should be followed accordingly. Titrate T4 bacteriophage stock suspension 1. Place 0.9 ml dilution broth in each of nine labeled 12 × 75–mm culture tubes. Add 0.1 ml of the T4 bacteriophage stock suspension to the first tube and mix well. Make serial dilutions by transferring 0.1 ml of the first dilution to the next tube, mixing well, then transferring 0.1 ml from that tube to the next one, and so on. Preferably, mix by pipetting up and down; do not vortex extensively because T4 bacteriophage will lose infectivity from excessive shearing forces.
2. Combine a broth culture of E. coli with soft agar (see Support Protocol 2, steps 2 to 4). Using a sterile pipet, place 4 ml of this mixture into each of three 15-ml culture tubes. Testing Aerosol Containment During Cell Sorting
Work swiftly at this and the next two steps, because the warm soft agar will cool rapidly, after which it can no longer be evenly poured.
3.3.8 Supplement 1
Current Protocols in Cytometry
T4 bacteriophage dilutions
0.1 ml
0.1 ml
0.1 ml
liquified soft agar + E. coli
4 ml
4 ml
4 ml
petri dishes containing bottom agar Figure 3.3.2 Plating of serial dilutions of T4 bacteriophage to determine titer.
3. Quickly add 0.1 ml each of the last three (10−7, 10−8, and 10−9) T4 bacteriophage dilutions to the soft agar/E. coli mixture in each of the three culture tubes, respectively, and mix well. Initially, it may be advisable to plate all T4 dilutions. Usually, the higher concentrations of the T4 bacteriophage stock suspension are not used, because they will lead to confluent lysis of the E. coli lawns or will create too many plaques for accurate counting. However, if the titer of the stock suspension is low, some or all of the lower dilutions will have to be plated. Otherwise, discard the six tubes containing the lower dilutions of T4.
4. Quickly pour the entire contents of each of the three tubes onto petri dishes containing bottom agar, making sure that the mixture covers the entire surface of the dish. Let the petri dishes cool with the lids ajar until the soft agar has solidified, to avoid accumulation of moisture. For illustration of steps 2 to 4 see Figure 3.3.2.
5. Close the lids and incubate the petri dishes 18 hr at 35°C to 37°C for plaque formation. Handle the dishes carefully because the soft agar is easily dislodged or damaged.
6. Count the plaques on each petri dish. Select a dish with an intermediate number of plaques to calculate T4 pfu/ml. Perform the calculation (see Anticipated Results). Dishes with a number of plaques between 10 and 100 are usually selected for the calculation. Note that the final dilution factor that enters the calculation is increased 100-fold—i.e., from 109 to 1011—because initially 0.1 ml of the T4 bacteriophage stock was taken and only 0.1 ml of the T4 dilution was added to the soft agar/E. coli mixture (also see Anticipated Results). The T4 stock suspension can only be used for aerosol tagging in the containment experiments on the cell sorter if its titer is >1 × 109 pfu/ml (see Critical Parameters and Anticipated Results). Usually, titers of 1 × 1011 or 1 × 1012 pfu/ml can be achieved using the method described in Support Protocol 1. See Troubleshooting if the titer of the T4 bacteriophage stock solution is lower than required.
Safety Procedures and Quality Control
3.3.9 Current Protocols in Cytometry
Supplement 1
Determine expected throughput of viable T4 bacteriophage/min on the cell sorter 7. Determine the flow rate on the cell sorter and adjust it to a fast flow rate (see Basic Protocol 1, steps 1 to 2). 8. Calculate the expected throughput of T4 bacteriophage in the cell sorter in pfu/min on the basis of the concentration of T4 in the stock suspension as determined by serial dilution and the (fast) flow rate of the instrument. Note that 1 mg of consumed sample equals 1 ìl. For an example of this calculation, see Anticipated Results.
Measure the actual throughput 9. Equip the instrument with a sheath fluid that maintains T4 viability, e.g., HBSS. Place a 12 × 75–mm culture tube containing 1 ml of T4 bacteriophage stock suspension onto the sample-introduction port of the cell sorter. 10. Weigh an empty 12 × 75–mm culture tube. While running the T4 stock suspension through the cell sorter collect the center stream into this tube as it leaves the nozzle for exactly 1 min. Reweigh the tube to determine the total amount of liquid collected. Note that it may take ∼1 min for the T4 stock suspension to reach the nozzle tip after the fluidic switch is first set to the Run position.
11. Immediately transfer 0.1 ml of the collected sample into 0.9 ml of dilution broth. This dilution will prevent loss of T4 activity through interaction with the sheath fluid.
12. Place 0.9 ml dilution broth in each of seven labeled 12 × 75–mm culture tubes. Add 0.1 ml of the diluted T4 suspension from step 11 to the first tube and mix well. Make serial dilutions by transferring 0.1 ml of this dilution to the next tube, mixing well, then transferring 0.1 ml from that tube to the next one, and so on. Preferably, mix by pipetting up and down; do not vortex extensively because T4 bacteriophage will lose infectivity from excessive shearing forces.
13. Combine a broth culture of E. coli with soft agar and place 4 ml of the mixture into each of three 15-ml culture tubes (see step 2 above). 14. Quickly add 0.1 ml each of the last three dilutions prepared in step 12 to the three culture tubes and mix well (see step 3 and attached annotations above). 15. Quickly plate the three tubes onto petri dishes containing bottom agar (see step 4 above and Fig. 3.3.2). 16. Incubate dishes 18 hr for plaque formation (see step 5 above). 17. Count the plaques on each petri dish and select a dish with an intermediate number of plaques to calculate the T4 bacteriophage pfu/min emerging from the cell sorter. Perform the calculation (see Anticipated Results).
Testing Aerosol Containment During Cell Sorting
Dishes with a number of plaques between 10 and 100 are usually selected for the calculation. Note that the final dilution factor that enters the calculation is increased 10-fold—i.e., from 106 to 107—because only 0.1 ml of the T4 dilution was added to the soft agar/E. coli mixture (also see Anticipated Results). The titer of viable T4 bacteriophage contained in the center stream emerging from the cell sorter will be lower than the titer measured in the T4 stock suspension because of an ∼200-fold dilution of the suspension with sheath fluid and loss of activity due to the physical stress of shearing forces. In order to keep the loss of titer acceptable, a sheath fluid that maintains T4 bacteriophage activity—e.g., HBSS—must be used. The T4 suspension leaving the cell sorter must have a titer >1 × 107 pfu/min to be used successfully for assessment of aerosol containment on the instrument (see Critical Parameters and Anticipated Results). See Troubleshooting if the titer of T4 bacteriophage emerging from the cell sorter is lower than required.
3.3.10 Supplement 1
Current Protocols in Cytometry
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Bottom agar 10.0 g Minimal agar Davis (Difco) 13.0 g Bacto-tryptone (Difco) 8.0 g NaCl 2.0 g sodium citrate (dihydrate) 1.3 g glucose H2O to 1 liter Adjust pH to 7.2 to 7.4 with 1 N NaOH Autoclave 15 min at 121°C Store up to 6 months at 4°C Dilution broth 10.0 g Bacto-tryptone (Difco) 5.0 g NaCl H2O to 1 liter Adjust pH to 7.2 to 7.4 with 1 N NaOH Autoclave 15 min at 121°C Store up to 6 months at 4°C Nutrient broth 8.0 g Bacto nutrient broth (Difco) 5.0 g NaCl 1.0 g glucose H2O to 1 liter Adjust pH to 7.2 to 7.4 with 1 N NaOH Autoclave 15 min at 121°C Store up to 6 months at 4°C Soft agar 6.5 g Minimal agar Davis (Difco) 13.0 g Bacto-tryptone (Difco) 8.0 g NaCl 2.0 g sodium citrate (dihydrate) 3.0 g glucose H2O to 1 liter Adjust pH to 7.2 to 7.4 with 1 N NaOH Autoclave 15 min at 121°C Store up to 6 months at 4°C COMMENTARY Background Information Aerosols have been shown to be of importance in the spread of infectious diseases (Sattar and Ijaz, 1987), and exposure to aerosols produced during handling and processing of viable biologic specimens will pose a health risk to laboratory personnel. This risk is generally minimized by performing all necessary sample manipulations within biological safety cabinets. However, cell sorters—which generally create aerosol droplets (with a size range of 40
to 200 µm), microdroplets (with a size range of 3 to 7 µm), and secondary aerosols (containing various droplet sizes) during normal operation (Bakker, 1992)—are too large to fit into a safety cabinet. Routine biosafety features designed to reduce production and escape of aerosols on most modern cell sorters include a vacuum-exhausted catch tube for the central stream and an enclosed sample collection chamber. The installation of a vacuum pump for evacuation of the sort chamber via negative pressure is a
Safety Procedures and Quality Control
3.3.11 Current Protocols in Cytometry
Supplement 1
Testing Aerosol Containment During Cell Sorting
custom modification designed to improve containment of aerosols (Bakker, 1992; Giorgi, 1994). Dispersion of electric charges to prevent droplet spraying can be achieved by attaching a grounded wire to sort-sample collection vials (Merrill, 1981), using silanized glass tubes for collection, and adding serum to the sample-collection buffer. Aerosol-control measures become particularly important during instrument malfunctions—e.g., if the vacuum exhaust system fails—because failure modes on the cell sorter can lead to increased aerosol production. If aerosols escape into the room, they could be potentially hazardous to the sorter operator. To reduce the risk of exposure for personnel, aerosol containment on the flow sorter must be verified before attempting to sort viable samples that may harbor known or unknown infectious agents. However, testing of aerosol escape may be of general importance because many reagents used for cell staining are toxic and could be harmful to the sorter operator if they are inhaled. Furthermore, any jet-in-air flow cytometer can produce aerosols even when not used for cell sorting, particularly when the nozzle tip is partially clogged; therefore it is advisable to also test aerosol containment in instruments that are used for data acquisition on unfixed samples. The movement and velocity of aerosolized particles in air is quite complex; however, it is generally true that large (visible) droplets in the size range >1 µm settle rapidly from the air because of gravitational forces. Thus, diffusion of large droplets is limited to the area close to the aerosol source. Containment efficiency on flow sorters with respect to these aerosol droplets can be measured passively (see Basic Protocol) by using bacteriophage for tagging of droplets and a detection system consisting of settle plates with E. coli bacterial lawns placed near the aerosol source. Several investigators have successfully tested this method (Merrill, 1981; Giorgi, 1994; Ferbas et al., 1995). However, for submicrometer particles that may remain suspended in air by Brownian motion for a prolonged time, the passive settle-plate testing method will not suffice. These particles, for the most part, freely diffuse into the environment until they evaporate. They are then referred to as droplet nuclei and are more difficult to detect than larger droplets because they are less likely to settle from air onto petri dishes by gravitational forces. Active air-sampling methods, therefore, must be utilized for accurate measurement of submicrometer particles. An example of active sampling is presented in this
unit (see Alternate Protocol) using an Andersen N6 air sampler (Andersen, 1958); room air is drawn through a collection device and directed onto an E. coli lawn. The Andersen sampler does not require any additional expertise other than that required to perform the Basic Protocol. Ferbas et al. (1995) have successfully used this single-stage air sampler for testing room air for aerosol containment on a Coulter EPICS Elite flow sorter. For further reference, Bakker (1992), in her thesis on testing aerosol-control measures, describes the use of bacteriophage (and alternatively bacteria) in combination with various air samplers for detection of aerosol escape from a FACStar (Becton Dickinson) cell sorter. From the standpoint of biosafety in the cellsorting laboratory, the aerosol-droplet size and mobility are significant, because these parameters determine the location of particle deposition that occurs during inhalation (Andersen, 1958; Bakker, 1992). It is important to realize that the majority of the aerosol mass is confined to large droplets that rapidly settle from room air. Small (submicrometer) droplets, however, are troublesome because their biohazardous potential is directly related to their ability to withstand dehydration, and when inhaled they are deposited deep within the lung of the exposed individual. Microdroplets are not large enough to carry eukaryotic cells, but they may contain cell-free viruses, bacterial cells, and other infectious or toxic agents. It may be argued that the risk of hazard exposure through droplet nuclei is negligible in the typical cellsorting laboratory, as few pathogens that are transmitted by this aerosol route are likely to be encountered. Thus, the biohazard potential of submicrometer-sized droplets should be considered on a case-by-case basis, and will determine method(s) of assessment of aerosol containment used for the instrument. In addition to testing the efficiency of aerosol containment on flow sorters, it is recommended that operators wear protective gear when performing sorts of viable and known biohazardous specimens. Training of sorter operators in strict adherence to safety practices as outlined (Giorgi, 1994; Ferbas et al., 1995; Schmid et al. 1997) is considered essential for sorting of fixed samples.
Critical Parameters The titer of the T4 bacteriophage stock suspension used for the aerosol-containment test of the cell sorter should be >1 × 109 pfu/ml to provide optimal sensitivity. Furthermore, it is
3.3.12 Supplement 1
Current Protocols in Cytometry
Table 3.3.1
Troubleshooting Guide for Aerosol-Containment Testing
Problem
Possible cause
Solution
Titer of T4 bacteriophage stock suspension is <1 × 109 pfu/ml
Mistake in medium preparation E. coli used for propagation of T4 may not have been in log-phase growth
Check pH of medium, redo preparation Check E. coli density with spectrophotometer (get advice from a microbiologist) Redo titration and plate lower dilutions (e.g., 10−4, 10−5, 10−6) of the T4 stock to assess the titer
Inaccurate titration of bacteriophage suspension Titer of T4 bacteriophage emerging from cell sorter is <1 × 107 pfu/min
Lack of E. coli lawns on petri dishes
Incompatible sheath fluid (e.g., antibiotics or bacteriostatic agents may be present in the sheath fluid)
Change sheath fluid to a T4-compatible solution (e.g., HBSS)
Inaccurate titration of bacteriophage suspension
Redo titration and plate lower dilutions of T4 to assess titer (see above)
E. coli may have been killed by being added to soft agar warmer than 50°C
Monitor temperature of soft agar with a thermometer
Soft agar solidifies too fast to prepare all of the petri dishes Petri dishes inside the sort chamber show large and uneven areas of lysis
Use a 40° to 50°C water bath to keep agar in liquid state Soft agar was damaged (e.g., petri dishes were placed too near to the sort streams)
essential that the loss of T4 bacteriophage activity resulting from dilution with sheath fluid and physical stress that occurs during sorting as the bacteriophage suspension is run through the instrument be within acceptable limits. At least 1 × 107 pfu/min of viable T4 bacteriophage should emerge from the sort nozzle to permit sensitive testing of aerosol containment. Finally, standardization of the cell-sorter setup— e.g., adjustment to a standard flow rate, a constant sort-decision rate, and a constant dropdrive frequency—is important in order to be able to compare test results that were obtained at different times. Figure 3.3.2 depicts a standard form recording the instrument setup used to obtain a given set of results. Table 3.3.1 lists problems that can arise in aerosol-containment testing along with their possible causes and solutions.
Anticipated Results Figure 3.3.3 shows a report sheet with results generated from a successful aerosol containment test performed as outlined in the Basic Protocol on a FACStar Plus (Becton Dickinson) cell sorter. In section II of the report sheet, results of a titration of the T4 bacteriophage stock suspension, performed as described in Support Protocol 3, are calculated as follows.
Handle petri dishes containing soft agar very carefully and insure that the streams are not hitting the dishes
A 10−7 dilution resulted in a number of plaques too numerous to count (TNTC); a 10−8 dilution resulted in 872 plaques; and a 10−9 dilution resulted in 85 plaques. The pfu/ml is then arrived at by multiplying 85 by the dilution factor, 1011, which is in turn calculated by multiplying the inverse of the dilution (109) by a factor of 100 (i.e., only 0.1 ml was originally taken from the T4 stock suspension for the serial dilution and only 0.1 ml of that serial dilution was plated, resulting in a further dilution of 0.1 × 0.1 = 0.01. Taking the inverse of this results in a factor of 100 by which the dilution factor of 109 must be increased—i.e., 109 × 102 = 1011). Hence, the result in pfu/ml is 85 × 1011 = 8.5 × 1012 pfu/ml. In section III of the report sheet shown in Figure 3.3.3, the expected T4 bacteriophage throughput through the cell sorter is calculated (on the basis of the pfu/ml calculated above) as follows: sample throughput = 40 µl/min 40 µl/min × 8.5 × 1012 pfu/1000 µl = 3.4 × 1011 pfu/min. In section IV of Figure 3.3.3, results for the titration of T4 bacteriophage in pfu/min emerging from the cell sorter, performed as outlined in Support Protocol 3, are calculated as follows.
Safety Procedures and Quality Control
3.3.13 Current Protocols in Cytometry
Supplement 1
plated, to give a dilution factor of 107. The pfu/min is then arrived at as follows:
The volume per min collected from the center stream on the cell sorter is 2.37 ml. A dilution of 10−6 resulted in 14 plaques, a dilution of 10−7 resulted in 1 plaque, and a dilution of 10−8 resulted in 0 plaques. Choose the dilution resulting in 14 plaques to calculate the dilution factor, which is calculated as the inverse of the dilution (1/10−6 = 106) multiplied by a factor of 10 because only 0.1 ml of the dilution was
14 plaques × (107/ 0.1 ml) × 2.37 ml/min = 3.3 × 109 pfu/min. In the report shown in Figure 3.3.3, all test parameters for the cell sorter and for the biological materials used for the instrument test were within the established criteria. Aerosol
T4 BACTERIOPHAGE SORT FOR TESTING OF AEROSOL CONTAINMENT I.
CALCULATION OF SAMPLE RATE (µl/min) Volume fed to machine (µl): Sample rate (µl/min):
409 µl
Feeding time (min):
10 min
40 µl
NO
YES
Does flow rate meet established criteria (30-40 µl/min)? II. TITRATION OF T4 BACTERIOPHAGE STOCK (MINIMUM 109 pfu/ml)
Plate ID A1 A2 A3
Dilution 10–7 10–8 10–9
No. of plaques TNTC 872 85
Dilution factor
Vol. on plate
pfu/ml
1011
0.1 ml
8.5 x 1012
Does the T4 phage stock meet minimal concentration
NO
YES
requirement (109 pfu/ml)? III. CALCULATION OF EXPECTED T4 FLOWING PER MIN THROUGH CELL SORTER Sample rate: 40 µl/min
T4 stock conc: 8.5 x 1012 pfu/ml
Expected T4 throughput: 3.4 x 1011 pfu/min
IV. T4 FLOWING/MIN THROUGH CELL SORTER (MINIMUM 107 pfu/min) Collect sheath stream for 1 min Volume collected (ml):
Plate ID B1 B2 B3
2.37 ml
Dilution 10–5 10–6 10–7
No. of plaques 14 1 0
Dilution factor 106
Vol. on plate 0.1 ml
YES
Does the T4 phage stock meet optimal established
Sheath rate 2.37 ml/min
pfu/min 3.3 x 109
NO
criteria for flow rate through flow cytometer (> 107 pfu/min)? V. AEROSOL CONTAINMENT IN THE REGULAR SORTING MODE
Plate ID C1 C2 C3 C4 C5 C6
Plate location inside right inside left door right door left x adjust top of machine
No. of plaques 12 0 0 0 0 0
INSTRUMENT SETUP Sheath pressure 11 psi Drop frequency ∼27 khz ∼2000 decision/sec Sort rate Droplets sorted 3.0 Sample flow rate 30-40 µl/min Door closed yes Vacuum on yes Nozzle tip 70 µm Auxilliary vacuum on Drop drive amplitude ∼5 V
Time 2 hr " " " " "
Is aerosol contained under regular sorting conditions?
YES
NO
YES
NO
VI. AEROSOL CONTAINMENT DURING FAILURE MODE
Plate ID D1 D2 D3 D4 D5 D6
Plate location inside right inside left door right door left x adjust top of machine
No. of plaques 22 250 0 0 0 0
Is aerosol contained under failure-mode sorting conditions?
Testing Aerosol Containment During Cell Sorting
REMARKS: Successful aerosol containment experiment
Time 15 min " " " " "
Signed:
Figure 3.3.3 Sample report sheet for testing aerosol containment on a cell sorter. For discussion of calculations, see Anticipated Results.
3.3.14 Supplement 1
Current Protocols in Cytometry
containment was found to be efficient because no plaques were counted on the petri dishes with E. coli lawns that were placed outside the sort chamber during the regular sorting mode and during the failure mode. The positive control petri dishes placed inside the sort chamber showed plaques, confirming that the detection system was valid.
Time Considerations Because preparation of some of the biologicals for the aerosol containment test will take several days, it is advisable to follow the sequence outlined here. 1. Generate the T4 bacteriophage stock suspension at least 1 week before the actual aerosol containment test on the cell sorter. This will take little time: ∼15 min on 3 consecutive days followed by ∼30 min for the final steps of the procedure on the fourth day. 2. Prepare the petri dishes with bottom agar several days prior to the instrument test. This will take ∼45 min. 3. Initiate the E. coli broth culture on the day before the aerosol containment test; this will take ∼10 min. This culture is used for the generation of the E. coli lawns (which will take 20 to 30 min) on the day of the test. An additional 2 to 4 hr are needed for incubation. Simultaneously, the E. coli culture can be used for the preparation of petri dishes for the titration of the T4 bacteriophage stock suspension and the titration of T4 emerging from the cell sorter, when these titrations are done on the same day as the aerosol test. Allow ∼45 min for cell-sorter setup, including adjustment of the flow rate for the aerosol-containment test and titration of viable T4 bacteriophage emerging from the instrument, with an additional 10 to 15 min for setup of the air sampler if the Alternate Protocol is to be performed. The actual containment test will take 2.5 to 3 hr. The next day, after overnight
incubation, an additional ∼30 min are needed for plaque counting, calculation, and reporting of results. Allow ∼1 hr for performing the titration of the T4 bacteriophage stock suspension and T4 emerging from the cell sorter if both are done at the same time, ∼40 min for each if done separately. If the T4 bacteriophage titrations are done separately from the instrument test (possibly 1 or 2 days before), the next day after overnight incubation ∼30 min are needed for plaque counting, calculation, and reporting of results; otherwise, all result evaluations can be performed at the same time.
Literature Cited Andersen, A.A. 1958. New sampler for the collection, sizing and enumeration of viable airborne particles. J. Bacteriol. 76:471-484. Bakker, A.M. 1992. Evaluation of a biological containment system for a fluorescence-activated cell sorter. M.S. thesis. University of San Francisco. Ferbas, J., Chadwick, K.R., Logar, A., Patterson, A.E., Gilpin, R.W., and Margolick, J.B. 1995. Assessment of aerosol containment on the ELITE flow cytometer. Cytometry 22:45-47. Giorgi, J.V. 1994. Cell sorting of biohazardous specimens for assay of immune function. Methods Cell Biol. 42:359-369. Merrill, J.T. 1981. Evaluation of selected aerosolcontrol measures on flow sorters. Cytometry 1:342-345. Sattar, S.A. and Ijaz, M.K. 1987. Spread of viral infections by aerosols. CRC Crit. Rev. Environ. Control 17:89-131. Schmid, I., Nicholson, J.K.A., Giorgi, J.V., Janossy, G., Kunkl, A., Lopez, P.A., Perfetto, S., Seamer, C.L., and Dean, P.N. 1997. Biosafety guidelines for sorting of unfixed cells. Cytometry 28:99117.
Contributed by Ingrid Schmid, Lance E. Hultin, and John Ferbas UCLA School of Medicine Los Angeles, California
Safety Procedures and Quality Control
3.3.15 Current Protocols in Cytometry
Supplement 1
Safe Use of Hazardous Chemicals
UNIT 3.4
Carrying out the protocols in this manual may result in exposure to toxic chemicals or carcinogenic, mutagenic, or teratogenic reagents (see Table 3.4.1). Cautionary notes and some specific guidelines are included in many instances throughout the manual; however, users must proceed with the prudence and precautions associated with good laboratory practice, under the supervision of those responsible for implementing lab safety programs at their institutions. It is not possible in the space available to list all the precautions required for handling hazardous chemicals. Many texts have been written about laboratory safety (see Literature Cited and Key References). Obviously, all national and local laws should be obeyed, as well as all institutional regulations. Controlled substances are regulated by the Drug Enforcement Administration (http://www.doj.gov/dea). By law Material Safety Data Sheets (MSDSs) must be readily available. All laboratories should have a Chemical Hygiene Plan (29CFR Part 1910.1450); institutional safety officers should be consulted as to its implementation. Help is (or should be) available from your institutional Safety Office; use it. Chemicals must be stored properly for safety. Certain chemicals cannot be easily or safely mixed with and should not be stored near certain other chemicals, because their reaction is violently exothermic or yields a toxic product. Some examples of incompatibility are listed in Table 3.4.2. When in doubt, always consult a current MSDS for information on reactivity, handling, and storage. Chemicals should be separated into general hazard classes and stored appropriately. For example, flammable chemicals such as alcohols, ketones, aliphatic and aromatic hydrocarbons, and other materials labeled flammable should be stored in approved flammable storage cabinets, with those also requiring refrigeration being kept in explosion-proof refrigerators. Strong oxidizers must be segregated. Strong acids (e.g., sulfuric, hydrochloric, nitric, perchloric, and hydrofluoric) should be stored in a separate cabinet well removed from strong bases and from flammable organics. An exception is glacial acetic acid, which is both corrosive and flammable, and which must be stored with the flammables. Facilities should be appropriate for working with hazardous chemicals. In particular, hazardous chemicals should be handled only in chemical fume hoods, not in laminar flow cabinets. The functioning of the fume hoods should be checked periodically. Laboratories should also be equipped with safety showers and eye-wash facilities. Again, this equipment should be tested periodically to ensure that it functions correctly. Other safety equipment may be required depending on the nature of the materials being handled. In addition, researchers should be trained in the proper procedures for handling hazardous chemicals as well as other laboratory operations—e.g., handling of compressed gases, use of cryogenic liquids, operation of high-voltage power supplies, and operation of lasers of all types. Before starting work, know the physical and chemical hazards of the reagents used. Wear appropriate protective clothing and have a plan for dealing with spills or accidents; coming up with a good plan on the spur of the moment is very difficult. For example, have the appropriate decontaminating or neutralizing agents prepared and close at hand. Small spills can probably be cleaned up by the researcher. In the case of larger spills, the area should be evacuated and help should be sought from those experienced in and equipped for dealing with spills—e.g., the institutional Safety Office. Protective equipment should include, at a minimum, eye protection, a lab coat, and gloves. In certain circumstances other items of protective equipment may be necessary (e.g., a face shield). Different types of gloves exhibit different resistance properties (Table 3.4.3).
Safety Procedures and Quality Control
Contributed by George Lunn and Gretchen Lawler
3.4.1
Current Protocols in Cytometry (2002) 3.4.1-3.4.33 Copyright © 2002 by John Wiley & Sons, Inc.
Supplement 20
Table 3.4.1
Commonly Used Hazardous Chemicalsa
Chemical
Hazards
Acetic acid, glacial Acetonitrile
Corrosive, flammable liquid Flammable liquid, teratogenic, toxic Carcinogenic, mutagenic Carcinogenic, toxic
Acridine orange Acrylamide
Alcian blue 8GX Alizarin red S (monohydrate) p-Amidinophenylmethanesulfonyl fluoride (APMSF) 7-Aminoactinomycin D (7-AAD) 4-(2-Aminoethyl)benzenesulfonyl fluoride (AEBSF) Ammonium hydroxide, concentrated Azure A Azure B Benzidine (BDB) Bisacrylamide Boron dipyrromethane derivatives (BODIPY dyes) Brilliant blue R 5-Bromodeoxyuridine (BrdU) Cetylpyridinium chloride (CPC) Cetyltrimethylammonium bromide (CTAB) Chloroform Chlorotrimethylsilane Chromic/sulfuric acid cleaning solution
Remarksb
See Basic Protocol 2 Use dust mask; polyacrylamide gels contain residual acrylamide monomer and should be handled with gloves; acrylamide may polymerize with violence on melting at 86°C See Basic Protocol 2
Enzyme inhibitor
See Basic Protocol 11
Carcinogenic Mutagenic, enzyme inhibitor
See Basic Protocol 11
Corrosive, lachrymatory, toxic Mutagenic Mutagenic Carcinogenic, toxic Toxic Toxic Carcinogenic, mutagenic Mutagenic, teratogenic, photosensitizing Toxic Corrosive, teratogenic, toxic Carcinogenic, teratogenic, toxic Carcinogenic, corrosive, flammable liquid, toxic Carcinogenic, corrosive, oxidizer, toxic
Chromomycin A3 (CA3) Congo red Coomassie brilliant blue G Crystal violet Cresyl violet acetate Cyanides (e.g., KCN, NaCN)
Teratogenic, toxic Mutagenic, teratogenic Mutagenic
Cyanines (e.g., Cy3, Cy5) Cyanogen bromide (CNBr)
Toxic Toxic
Mutagenic Toxic
See Basic Protocol 2 See Basic Protocol 2 See Basic Protocol 1
See Basic Protocol 2
Reacts violently with water; see Basic Protocol 3 Replace with suitable commercially available cleanser See Basic Protocol 2 See Basic Protocol 2 See Basic Protocol 2 See Basic Protocol 2 Contact with acid will liberate HCN gas; see Basic Protocol 4 See Basic Protocol 4 continued
3.4.2 Supplement 20
Current Protocols in Cytometry
Table 3.4.1
Commonly Used Hazardous Chemicalsa, continued
Chemical
Hazards
2′-Deoxycoformycin (dCF, pentostatin) 4′,6-Diamidino-2-phenylindole (DAPI) Diaminobenzidine (DAB) 1,4-Diazabicyclo[2,2,2]-octane (DABCO)
Teratogenic, toxic Mutagenic Carcinogenic Toxic
Dichloroacetic acid (DCA) Dichloromethane (methylene chloride)
Carcinogenic, corrosive, toxic Carcinogenic, mutagenic, teratogenic, toxic Corrosive, flammable liquid, toxic Carcinogenic, toxic Carcinogenic, teratogenic, toxic Highly toxic, cholinesterase inhibitor, neurotoxin Corrosive, flammable liquid, toxic Carcinogenic, toxic Flammable liquid, toxic
Diethylamine (DEA) Diethylpyrocarbonate (DEPC) Diethyl sulfate Diisopropyl fluorophosphate (DFP) Dimethyldichlorosilane Dimethyl sulfate (DMS) Dimethyl sulfoxide (DMSO) Diphenylamine (DPA) 2,5-Diphenyloxazole (PPO) Dithiothreitol (DTT) Eosin B Erythrosin B Ether
Ethidium bromide (EB) Ethyl methanesulfonate (EMS) Fluorescein and derivatives 5-Fluoro-2′-deoxyuridine (FUdR) Fluoroorotic acid (FOA) Formaldehyde
Remarksb
See Basic Protocol 1 Forms an explosive complex with hydrogen peroxide
See Basic Protocol 5 See Basic Protocol 11 See Basic Protocol 3 See Basic Protocol 5 Enhances absorption through skin
Teratogenic, toxic Toxic Toxic Carcinogenic, mutagenic Flammable liquid, toxic
Formamide Formic acid
Mutagenic, toxic Carcinogenic, toxic Carcinogenic, toxic Teratogenic, toxic Toxic Carcinogenic, flammable liquid, teratogenic, toxic Teratogenic, toxic Corrosive, toxic
Glutaraldehyde Guanidinium thiocyanate Hoechst 33258 dye Hydrochloric acid, concentrated
Corrosive, teratogenic, toxic Toxic Mutagenic, toxic Corrosive, teratogenic, toxic
See Basic Protocol 2 See Basic Protocol 2 May form explosive peroxides on standing; do not dry with NaOH or KOH See Basic Protocol 2 or 6 See Basic Protocol 5
May explode when heated >180°C in a sealed tube
continued
3.4.3 Current Protocols in Cytometry
Supplement 20
Table 3.4.1
Commonly Used Hazardous Chemicalsa, continued
Chemical
Hazards
Remarksb
Hydrogen peroxide (30%)
Carcinogenic, corrosive, mutagenic, oxidizer
Hydroxylamine
Corrosive, flammable, mutagenic, toxic Carcinogenic Corrosive, toxic Carcinogenic, mutagenic, toxic Carcinogenic, mutagenic Carcinogenic, toxic Stench, toxic Teratogenic, toxic Teratogenic, toxic Carcinogenic, mutagenic, teratogenic, toxic Mutagenic, toxic Carcinogenic, toxic Teratogenic, toxic Mutagenic
Avoid bringing into contact with organic materials, which may form explosive peroxides; may decompose violently in contact with metals, salts, or oxidizable materials; see Basic Protocol 7 Explodes in air at >70°C
3-β-Indoleacrylic acid (IAA) Iodine Iodoacetamide Janus green B Lead compounds 2-Mercaptoethanol (2-ME) Mercury compounds Methionine sulfoximine (MSX) Methotrexate (amethopterin) Methylene blue Methyl methanesulfonate (MMS) Mycophenolic acid (MPA) Neutral red Nigrosin, water soluble Nitric acid, concentrated Nitroblue tetrazolium (NBT) Orcein, synthetic Oxonols Paraformaldehyde Phenol
See Basic Protocol 8 See Basic Protocol 2
See Basic Protocol 9
See Basic Protocol 2 See Basic Protocol 5 See Basic Protocol 2 See Basic Protocol 2
Corrosive, oxidizer, teratogenic, toxic Toxic See Basic Protocol 2
Phenylmethylsulfonyl fluoride (PMSF) Phorbol 12-myristate 13-acetate (PMA) Phycoerythrins (PE) Piperidine Potassium hydroxide, concentrated
Toxic Toxic Carcinogenic, corrosive, teratogenic, toxic Enzyme inhibitor Carcinogenic, toxic Toxic Flammable liquid, teratogenic, toxic Corrosive, toxic
Propane sultone Propidium iodide (PI) Pyridine Rhodamine and derivatives Rose Bengal Safranine O
Carcinogenic, toxic Mutagenic Flammable liquid, toxic Toxic Carcinogenic, teratogenic Mutagenic
Readily absorbed through the skin See Basic Protocol 11
Produces a highly exothermic reaction when solid is added to water See Basic Protocol 5 See Basic Protocol 2 or 6
See Basic Protocol 2 See Basic Protocol 2
continued
3.4.4 Supplement 20
Current Protocols in Cytometry
Table 3.4.1
Commonly Used Hazardous Chemicalsa, continued
Chemical
Hazards
Remarksb
Sodium azide
Carcinogenic, toxic
Adding acid liberates explosive volatile, toxic hydrazoic acid; can form explosive heavy metal azides, e.g., with plumbing fixtures—do not discharge down drain; see Basic Protocol 10
Sodium deoxycholate (Na-DOC) Sodium dodecyl sulfate (sodium lauryl sulfate, SDS) Sodium hydroxide, concentrated
Carcinogenic, teratogenic, toxic Sensitizing, toxic
Sodium nitrite Sulfuric acid, concentrated
SYTO dyes Tetramethylammonium chloride (TMAC) N,N,N′,N′-Tetramethyl-ethylenediamine (TEMED) Texas Red (sulforhodamine 101, acid chloride) Toluene Toluidine blue O Nα-p-Tosyl-L-lysine chloromethyl ketone (TLCK) N-p-Tosyl-L-phenylalanine chloromethyl ketone (TPCK) Trichloroacetic acid (TCA) Triethanolamine acetate (TEA) Trifluoroacetic acid (TFA) Trimethyl phosphate (TMP) Trypan blue Xylenes
Corrosive, toxic
A highly exothermic reaction ensues when the solid is added to water
Carcinogenic Corrosive, oxidizer, teratogenic, toxic Reaction with water is very exothermic; always add concentrated sulfuric acid to water, never water to acid Toxic Toxic Corrosive, flammable liquid, toxic Toxic Flammable liquid, teratogenic, toxic Mutagenic, toxic Toxic, enzyme inhibitor
See Basic Protocol 2 See Basic Protocol 11
Toxic, mutagenic, enzyme inhibitor
See Basic Protocol 11
Carcinogenic, corrosive, teratogenic, toxic Carcinogenic, toxic Corrosive, toxic Carcinogenic, mutagenic, teratogenic May explode on distillation Carcinogenic, mutagenic, teratogenic See Basic Protocol 2 Flammable liquid, teratogenic, toxic
aFor extensive information on the hazards of these and other chemicals as well as cautionary details, see Bretherick (1986), O’Neil (2001), Furr (2000),
Lewis (1999), Lunn and Sansone (1994a), and Bretherick et al. (1999). bCAUTION: These chemicals should be handled only in a chemical fume hood by knowledgeable workers equipped with eye protection, lab coat, and
gloves. The laboratory should be equipped with a safety shower and eye wash. Additional protective equipment may be required.
No gloves resist all chemicals, and no gloves resist any chemicals indefinitely. Disposable gloves labeled “exam” or “examination” are primarily for protection from biological materials (e.g., viruses, bacteria, feces, blood). They are not designed for and usually have not been tested for resistance to chemicals. Disposable gloves generally offer extremely marginal protection from chemical hazards in most cases and should be removed immediately upon contamination before the chemical can pass through. If possible, design handling procedures to eliminate or reduce potential for contamination. Never assume that disposable gloves will offer the same protection or even have the same properties as
Safety Procedures and Quality Control
3.4.5 Current Protocols in Cytometry
Supplement 20
Table 3.4.2
Examples of Chemical Incompatibility
Chemical
Incompatible with
Acetic acid
Aldehydes, bases, carbonates, chromic acid, ethylene glycol, hydroxides, hydroxyl compounds, metals, nitric acid, oxidizers, perchloric acid, peroxides, phosphates, permanganates, xylene Acids, amines, concentrated nitric and sulfuric acid mixtures, oxidizers, plastics Copper, halogens, mercury, oxidizers, potassium, silver Acids, aldehydes, carbon dioxide, carbon tetrachloride or other chlorinated hydrocarbons, halogens, ketones, plastics, sulfur, water Acids, aldehydes, amides, calcium hypochlorite, hydrofluoric acid, halogens, heavy metals, mercury, oxidizers, plastics, sulfur Acids, alkalis, chlorates, chloride salts, flammable and combustible materials, metals, organic materials, phosphorus, reducing agents, sulfur, urea Acids, aluminum, dibenzoyl peroxide, oxidizers, plastics Any reducing agent Acids, heavy metals, oxidizers Acetaldehyde, alcohols, alkalis, amines, ammonia, combustible materials, ethylene, fluorine, hydrogen, ketones (e.g., acetone, carbonyls), metals, petroleum gases, sodium carbide, sulfur Acids, ethanol, fluorine, organic materials, water Alkali metals, calcium hypochlorite, halogens, oxidizers Sodium Acids, ammonium salts, finely divided organic or combustible materials, powdered metals, sulfur Acetylene or other hydrocarbons, alcohols, ammonia, benzene, butadiene, butane, combustible materials, ethylene, flammable compounds (e.g., hydrazine), hydrogen, hydrogen peroxide, iodine, metals, methane, nitrogen, oxygen, propane (or other petroleum gases), sodium carbide, sodium hydroxide Ammonia, hydrogen, hydrogen sulfide, mercury, methane, organic materials, phosphine, phosphorus, potassium hydroxide, sulfur Acetic acid, acetone, alcohols, alkalis, ammonia, bases, benzene, camphor, flammable liquids, glycerin (glycerol), hydrocarbons, metals, naphthalene, organic materials, phosphorus, plastics Acetylene, calcium, hydrocarbons, hydrogen peroxide, oxidizers Acids (organic or inorganic) Acids, alkaloids, aluminum, iodine, oxidizers, strong bases Ammonium nitrate, chromic acid, halogens, hydrogen peroxide, nitric acid, oxidizing agents in general, oxygen, sodium peroxide All other chemicals
Acetone Acetylene Alkali metals, alkaline earth metals Ammonia (anhydrous)
Ammonium nitrate
Aniline Arsenical materials Azides Bromine
Calcium oxide Carbon (activated) Carbon tetrachloride Chlorates Chlorine
Chlorine dioxide
Chromic acid, chromic oxide
Copper Cumene hydroperoxide Cyanides Flammable liquids
Fluorine Safe Use of Hazardous Chemicals
continued
3.4.6 Supplement 20
Current Protocols in Cytometry
Table 3.4.2
Examples of Chemical Incompatibility, continued
Chemical
Incompatible with
Hydrocarbons (liquid or gas) Hydrocyanic acid Hydrofluoric acid
See flammable liquids
Hydrogen peroxide Hydrogen sulfide Hydroperoxide Hypochlorites Iodine Mercury Nitric acid Nitrites Nitroparaffins Oxalic acid Oxygen
Perchloric acid Peroxides, organic Phosphorus (white) Potassium chlorate
Potassium perchlorate Potassium permanganate Selenides and tellurides Silver Sodium Sodium nitrate Sodium peroxide
Sulfides Sulfuric acid
Alkali, nitric acid Ammonia, metals, organic materials, plastics, silica (glass, including fiberglass), sodium All organics, most metals or their salts, nitric acid, phosphorus, sodium, sulfuric acid Acetylaldehyde, fuming nitric acid, metals, oxidizers, sodium, strong bases Reducing agents Acids, activated carbon Acetylaldehyde, acetylene, ammonia, hydrogen, metals, sodium Acetylene, aluminum, amines, ammonia, calcium, fulminic acid, lithium, oxidizers, sodium Acids, nitrites, metals, most organics, plastics, sodium, sulfur, sulfuric acid Acids Amines, inorganic bases Mercury, oxidizers, silver, sodium chlorite All flammable and combustible materials, ammonia, carbon monoxide, grease, metals, oil, phosphorus, polymers All organics, bismuth and alloys, dehydrating agents, grease, hydrogen halides, iodides, paper, wood Acids (organic or mineral), avoid friction, store cold Air, alkalis, oxygen, reducing agents Acids, ammonia, combustible materials, fluorine, hydrocarbons, metals, organic materials, reducing agents, sugars Alcohols, combustible materials, fluorine, hydrazine, metals, organic matter, reducing agents, sulfuric acid Benzaldehyde, ethylene glycol, glycerin, sulfuric acid Reducing agents Acetylene, ammonium compounds, fulminic acid, oxalic acid, ozonides, peroxyformic acid, tartaric acid Acids, carbon dioxide, carbon tetrachloride, hydrazine, metals, oxidizers, water Acetic anhydride, acids, metals, organic matter, peroxyformic acid, reducing agents Acetic anhydride, benzaldehyde, benzene, carbon disulfide, ethyl acetate, ethyl or methyl alcohol, ethylene glycol, furfural, glacial acetic acid, glycerin, hydrogen sulfide, metals, methyl acetate, oxidizers, peroxyformic acid, phosphorus, reducing agents, sugars, water Acids Alcohols, bases, chlorates, perchlorates, permanganates of potassium, lithium, sodium, magnesium, calcium Safety Procedures and Quality Control
3.4.7 Current Protocols in Cytometry
Supplement 20
Table 3.4.3
Safe Use of Hazardous Chemicals
Chemical Resistance of Commonly Used Glovesa,b
Chemical
Neoprene Latex gloves gloves
Butyl gloves
Nitrile gloves
*Acetaldehyde Acetic acid *Acetone Ammonium hydroxide *Amyl acetate Aniline *Benzaldehyde *Benzene Butyl acetate Butyl alcohol Carbon disulfide *Carbon tetrachloride *Chlorobenzene *Chloroform Chloronaphthalene Chromic acid (50%) Cyclohexanol *Dibutyl phthalate Diisobutyl ketone Dimethylformamide Dioctyl phthalate Epoxy resins, dry *Ethyl acetate Ethyl alcohol *Ethyl ether *Ethylene dichloride Ethylene glycol Formaldehyde Formic acid Freon 11, 12, 21, 22 *Furfural Glycerin Hexane Hydrochloric acid Hydrofluoric acid (48%) Hydrogen peroxide (30%) Ketones Lactic acid (85%) Linseed oil Methyl alcohol Methylamine Methyl bromide *Methyl ethyl ketone *Methyl isobutylketone Methyl methacrylate Monoethanolamine
VG VG G VG F G F P G VG F F F G F F G G P F G VG G VG VG F VG VG VG G G VG F VG VG G G VG VG VG F G G F G VG
VG VG VG VG F F G P F VG F P F P F F G G G G F VG G VG VG F VG VG VG F G VG P G G G VG VG F VG G G VG VG VG VG
G VG P VG P P G F P VG F G P E F F VG G P G VG VG F VG G P VG VG VG G G VG G G G G P VG VG VG G F P P F VG
G VG VG VG P F F P F VG F P P P P P F P F F P VG F VG G P VG VG VG P G VG P G G G VG VG P VG F F G F G G
continued
3.4.8 Supplement 20
Current Protocols in Cytometry
Table 3.4.3
Chemical Resistance of Commonly Used Glovesa,b, continued
Chemical
Neoprene Latex gloves gloves
Butyl gloves
Nitrile gloves
Morpholine Naphthalene Naphthas, aliphatic Naphthas, aromatic *Nitric acid Nitric acid, red and white fuming Nitropropane (95.5%) Oleic acid Oxalic acid Palmitic acid Perchloric acid (60%) Perchloroethylene Phenol Phosphoric acid Potassium hydroxide Propyl acetate i-Propyl alcohol n-Propyl alcohol Sodium hydroxide Styrene (100%) Sulfuric acid Tetrahydrofuran *Toluene Toluene diisocyanate *Trichloroethylene Triethanolamine Tung oil Turpentine *Xylene
VG G VG G G P F VG VG VG VG F VG VG VG G VG VG VG P G P F F F VG VG G P
VG F F P F P F G VG VG G P G VG VG G VG VG VG P G F P G P G F F P
G G VG G F P F VG VG VG G G F VG VG F VG VG VG F G F F F G VG VG VG F
VG F F P F P P F VG VG F P F G VG F VG VG VG P G F P G F G P F P
aPerformance varies with glove thickness and duration of contact. An asterisk indicates limited use. Abbreviations: VG,
very good; G, good; F, fair; P, poor (do not use). bAdapted from the July 8, 1998, version of the DOE OSH Technical Reference Chapter 5 (APPENDIX C at
http://tis.eh.doe.gov/docs/osh_tr/ch5c.html). For more information also see Forsberg and Keith (1999).
nondisposables. Select gloves carefully and always look for some evidence that they will protect against the materials being used. Inspect all gloves before every use for possible holes, tears, or weak areas. Never reuse disposable gloves. Clean reusable gloves after each use and dry carefully inside and out. Observe all common-sense precautions—e.g., do not pipet by mouth, keep unauthorized persons away from hazardous chemicals, do not eat or drink in the lab, wear proper clothing in the lab (sandals, open-toed shoes, and shorts are not appropriate). Order hazardous chemicals only in quantities that are likely to be used in a reasonable time. Buying large quantities at a lower unit cost is no bargain if someone (perhaps you) has to pay to dispose of surplus quantities. Substitute alcohol-filled thermometers for mercury-filled thermometers, which are a hazardous chemical spill waiting to happen.
Safety Procedures and Quality Control
3.4.9 Current Protocols in Cytometry
Supplement 20
Although any number of chemicals commonly used in laboratories are toxic if used improperly, the toxic properties of a number of reagents require special mention. Chemicals that exhibit carcinogenic, corrosive, flammable, lachrymatory, mutagenic, oxidizing, teratogenic, toxic, or other hazardous properties are listed in Table 3.4.1. Chemicals listed as carcinogenic range from those accepted by expert review groups as causing cancer in humans to those for which only minimal evidence of carcinogenicity exists. No effort has been made to differentiate the carcinogenic potential of the compounds in Table 3.4.1. Oxidizers may react violently with oxidizable material (e.g., hydrocarbons, wood, and cellulose). Before using any of these chemicals, thoroughly investigate all its characteristics. Material Safety Data Sheets are readily available; they list some hazards but vary widely in quality. A number of texts describing hazardous properties are listed at the end of this unit (see Literature Cited). In particular, Sax’s Dangerous Properties of Industrial Materials, 10th ed. (Lewis, 1999) and the Handbook of Reactive Chemical Hazards, 6th ed. (Bretherick et al., 1999) give comprehensive listings of known hazardous properties; however, these texts list only the known properties. Many chemicals, especially fluorochromes, have been tested only partially or not at all. Prudence dictates that, unless there is good reason for believing otherwise, all chemicals should be regarded as volatile, highly toxic, flammable human carcinogens and should be handled with great care. Waste should be segregated according to institutional requirements, for example, into solid, aqueous, nonchlorinated organic, and chlorinated organic material, and should always be disposed of in accordance with all applicable federal, state, and local regulations. Extensive information and cautionary details along with techniques for the disposal of chemicals in laboratories have been published (Bretherick, 1986; Lunn and Sansone, 1994a; O’Neil, 2001; Furr, 2000). Some commonly used disposal procedures are outlined in Basic Protocols 1 to 11. Incorporation of these procedures into laboratory protocols can help to minimize waste disposal problems. Alternate Protocols 1 to 7 describe decontamination methods for some of the chemicals. Support Protocols 1 to 9 describe analytical techniques that are used to verify that reagents have been decontaminated; with modification, these assays may also be used to determine the concentration of a particular chemical. DISPOSAL METHODS A number of procedures for the disposal of hazardous chemicals are available; protocols for the disposal and decontamination of some hazardous chemicals commonly encountered in cytometry laboratories are listed in Table 3.4.4. These procedures are necessarily brief; for full details consult the original references or a collection of these procedures (see Lunn and Sansone, 1994a). CAUTION: These disposal methods should be carried out only in a chemical fume hood by workers equipped with eye protection, a lab coat, and gloves. Additional protective equipment may be necessary. BASIC PROTOCOL 1
Safe Use of Hazardous Chemicals
DISPOSAL OF BENZIDINE AND DIAMINOBENZIDINE Benzidine and diaminobenzidine can be degraded by oxidation with potassium permanganate (Castegnaro et al., 1985; Lunn and Sansone, 1991a). This protocol presents a method for decontamination of benzidine and diaminobenzidine in bulk. This method can also be adapted to the decontamination of benzidine and diaminobenzidine spills (see Alternate Protocol 1). These compounds can also be removed from solution using horseradish peroxidase in the presence of hydrogen peroxide (see Alternate Protocol 2). Destruction and decontamination are >99%. Support Protocol 1 is used to test for the presence of benzidine and diaminobenzidine.
3.4.10 Supplement 20
Current Protocols in Cytometry
Table 3.4.4
Protocols for Disposal of Some Hazardous Chemicals
Protocol
Disposal method for
Basic Protocol 1 Alternate Protocol 1 Alternate Protocol 2 Support Protocol 1
Benzidine and diaminobenzidine Spills of benzidine and diaminobenzidine Aqueous solutions of benzidine and diaminobenzidine Analysis for benzidine and diaminobenzidine
Basic Protocol 2 Alternate Protocol 3 Support Protocol 2
Biological stains Large volumes of dilute biological stains Analysis for biological stains
Basic Protocol 3
Silanes
Basic Protocol 4 Support Protocol 3
Cyanide and cyanogen bromide Analysis for cyanide
Basic Protocol 5
Dimethyl sulfate, diethyl sulfate, methyl methanesulfonate, ethyl methanesulfonate, diepoxybutane, 1,3-propane sultone Analysis for dimethyl sulfate, diethyl sulfate, methyl methanesulfonate, ethyl methanesulfonate, diepoxybutane, 1,3-propane sultone
Support Protocol 4 Basic Protocol 6 Alternate Protocol 4 Alternate Protocol 5 Alternate Protocol 6 Support Protocol 5
Ethidium bromide and propidium iodide Equipment contaminated with ethidium bromide Ethidium bromide in isopropanol containing cesium chloride Ethidium bromide in alcohols Analysis for ethidium bromide and propidium iodide
Basic Protocol 7
Hydrogen peroxide
Basic Protocol 8
Iodine
Basic Protocol 9 Alternate Protocol 7 Support Protocol 6
Mercury compounds Waste water containing mercury Analysis for mercury
Basic Protocol 10 Support Protocol 7 Support Protocol 8
Sodium azide Analysis for sodium azide Analysis for nitrite
Basic Protocol 11 Support Protocol 9
Enzyme inhibitors Analysis for enzyme inhibitors
Materials Benzidine or diaminobenzidine tetrahydrochloride dihydrate 0.1 M HCl (for benzidine) 0.2 M potassium permanganate: prepare immediately before use 2 M sulfuric acid Sodium metabisulfite 10 M potassium hydroxide (KOH) Additional reagents and equipment for testing for the presence of aromatic amines (see Support Protocol 1) 1. For each 9 mg benzidine, add 10 ml of 0.1 M HCl or for each 9 mg diaminobenzidine tetrahydrochloride dihydrate, add 10 ml water. Stir the solution until the aromatic amine has completely dissolved.
Safety Procedures and Quality Control
3.4.11 Current Protocols in Cytometry
Supplement 20
2. For each 10 ml of solution, add 5 ml freshly prepared 0.2 M potassium permanganate and 5 ml of 2 M sulfuric acid. Allow the mixture to stand for ≥10 hr. 3. Add sodium metabisulfite until the solution is decolorized. 4. Add 10 M KOH to make the solution strongly basic, pH >12. CAUTION: This reaction is exothermic.
5. Dilute with 5 vol water and pass through filter paper to remove manganese compounds. 6. Test the filtrate for the presence of aromatic amines (i.e., benzidine or diaminobenzidine; see Support Protocol 1). 7. Neutralize the filtrate with acid and discard. ALTERNATE PROTOCOL 1
DECONTAMINATION OF SPILLS INVOLVING BENZIDINE AND DIAMINOBENZIDINE Additional Materials (also see Basic Protocol 1) Glacial acetic acid 1:1 (v/v) 0.2 M potassium permanganate/2 M sulfuric acid: prepare immediately before use Absorbent material (e.g., paper towels, Kimwipes) High-efficiency particulate air (HEPA) vacuum (Fisher) Additional reagents and equipment for testing for the presence of aromatic amines (see Support Protocol 1) CAUTION: This procedure may damage painted surfaces and Formica. 1. Remove as much of the spill as possible using absorbent material and high-efficiency particulate air (HEPA) vacuuming. 2. Wet the surface with glacial acetic acid until all the amines are dissolved, then add an excess of freshly prepared 1:1 (v/v) 0.2 M potassium permanganate/2 M sulfuric acid to the spill area. Allow the mixture to stand ≥10 hr. 3. Ventilate the area and decolorize with sodium metabisulfite. 4. Mop up the liquid with paper towels. Squeeze the solution out of the towels and collect in a suitable container. Discard towels as hazardous solid waste. 5. Add 10 M KOH to make the solution strongly basic, pH ≥12. CAUTION: This reaction is exothermic.
6. Dilute with 5 vol water and filter through filter paper to remove manganese compounds. 7. Test the filtrate for the presence of aromatic amines (i.e., benzidine or diaminobenzidine; see Support Protocol 1). 8. Neutralize the filtrate with acid and discard it. 9. Verify complete decontamination by wiping the surface with a paper towel moistened with water and squeezing the liquid out of the towel. Test the liquid for the presence of benzidine or diaminobenzidine (see Support Protocol 1). Repeat steps 1 to 9 as necessary.
Safe Use of Hazardous Chemicals
3.4.12 Supplement 20
Current Protocols in Cytometry
DECONTAMINATION OF AQUEOUS SOLUTIONS OF BENZIDINE AND DIAMINOBENZIDINE The enzyme horseradish peroxidase catalyzes the oxidation of the amine to a radical which diffuses into solution and polymerizes. The polymers are insoluble and fall out of solution.
ALTERNATE PROTOCOL 2
Additional Materials (also see Basic Protocol 1) Aqueous solution of benzidine or diaminobenzidine 1 N HCl or NaOH 3% (v/v) hydrogen peroxide Horseradish peroxidase (see recipe) 1:1 (v/v) 0.2 M potassium permanganate/2 M sulfuric acid 5% (w/v) ascorbic acid Porous glass filter or Sorvall GLC-1 centrifuge or equivalent Additional reagents and equipment for testing for the presence of aromatic amines (see Support Protocol 1) 1. Adjust the pH of the aqueous benzidine or diaminobenzidine solution to 5 to 7 with 1 N HCl or NaOH as required and dilute so the concentration of aromatic amines is ≤100 mg/liter. 2. For each liter of solution, add 3 ml of 3% hydrogen peroxide and 300 U horseradish peroxidase. Let the mixture stand 3 hr. 3. Remove the precipitate by filtering the solution through a porous glass filter or by centrifuging 5 min at room temperature in a benchtop centrifuge to pellet the precipitate. The precipitate is mutagenic and should be treated as hazardous waste.
4. Immerse the porous glass filter in 1:1 (v/v) 0.2 M potassium permanganate/2 M sulfuric acid. Clean the filter in a conventional fashion and discard potassium permanganate/sulfuric acid solution as described for benzidine and diaminobenzidine (see Basic Protocol 1). 5. For each liter of filtrate, add 100 ml of 5% ascorbic acid. 6. Test the filtrate for the presence of aromatic amines (see Support Protocol 1). 7. Discard the decontaminated filtrate. ANALYTICAL PROCEDURES TO DETECT BENZIDINE AND DIAMINOBENZIDINE Reversed-phase HPLC (Snyder et al., 1997) is used to test for the presence of aromatic amines. The limit of detection is 1 µg/ml for benzidine and 0.25 µg/ml for diaminobenzidine.
SUPPORT PROTOCOL 1
Materials Decontaminated aromatic amine solution 10:30:20 (v/v/v) acetonitrile/methanol/1.5 mM potassium phosphate buffer (1.5 mM K2HPO4/1.5 mM KH2PO4) (benzidine) or 75:25 (v/v) methanol/1.5 mM potassium phosphate buffer (diaminobenzidine) 250-mm × 4.6-mm-i.d. Microsorb C-8 reversed-phase HPLC column (Varian) or equivalent Additional reagents and equipment for reversed-phase liquid chromatography (Snyder et al., 1997)
Safety Procedures and Quality Control
3.4.13 Current Protocols in Cytometry
Supplement 20
Analyze the decontaminated aromatic amine solution by reversed-phase HPLC using a 250-mm × 4.6-mm-i.d. Microsorb C-8 column or equivalent. To detect benzidine, elute with 10:30:20 (v/v/v) acetonitrile/methanol/1.5 mM potassium phosphate buffer at a flow rate of 1.5 ml/min and UV detection at 285 nm. To detect diaminobenzidine, elute with 75:25 (v/v) methanol/1.5 mM potassium phosphate buffer at a flow rate of 1 ml/min and UV detection at 300 nm. BASIC PROTOCOL 2
DISPOSAL OF BIOLOGICAL STAINS Biological stains (Table 3.4.5), as well as ethidium bromide and propidium iodide, can be removed from solution using the polymeric resin Amberlite XAD-16. The decontaminated solution may be disposed of as nonhazardous aqueous waste and the resin as hazardous solid waste. The volume of contaminated resin generated is much smaller than the original volume of the solution of biological stain, so the waste disposal problem is greatly reduced. The final concentration of any remaining stain should be less than the limit of detection (see Support Protocol 2 and Table 3.4.5). In each case decontamination should be >99%. This protocol describes a method for batch decontamination in which the resin is stirred in the solution to be decontaminated and removed by filtration at the end of the reaction time. Large volumes of biological stain can be decontaminated using a column (see Alternate Protocol 3). For full details refer to the original literature (Lunn and Sansone, 1991b) or a compilation (Lunn and Sansone, 1994a).
Table 3.4.5
Safe Use of Hazardous Chemicals
Decontamination of Biological Stains
Compound
Time required for complete decontamination
Volume of solution (ml) decontaminated per gram resin
Acridine orange Alcian blue 8GX Alizarin red S Azure A Azure B Brilliant blue R Congo red Coomassie brilliant blue G Cresyl violet acetate Crystal violet Eosin B Erythrosin B Ethidium bromide Janus green B Methylene blue Neutral red Nigrosin Orcein Propidium iodide Rose Bengal Safranine O Toluidine blue O Trypan blue
18 hr 10 min 18 hr 10 min 10 min 2 hr 2 hr 2 hr 2 hr 30 min 30 min 18 hr 4 hr 30 min 30 min 10 min 2 hr 2 hr 2 hr 3 hr 1 hr 30 min 2 hr
20 500 5 80 80 80 40 80 40 200 40 10 20 80 80 500 80 200 20 20 20 80 40
3.4.14 Supplement 20
Current Protocols in Cytometry
Materials Amberlite XAD-16 resin (Supelco) 100 µg/ml biological stain in water Additional reagents and equipment for testing for the presence of biological stain (see Support Protocol 2) For batch decontamination of 20 ml stain 1a. Add 1 g Amberlite XAD-16 to 20 ml of 100 µg/ml biological stain in water. For aqueous solutions having stain concentrations other than 100 ìg/ml, use proportionately greater or lesser amounts of resin to achieve complete decontamination. For solutions of erythrosin B, use 2 g Amberlite XAD-16 for 20 ml stain.
2a. Stir the mixture for at least the time indicated in Table 3.4.5. For batch decontamination of larger volumes of stain 1b. Add 1 g Amberlite XAD-16 to the volume of a 100 µg/ml biological stain in water indicated in Table 3.4.5. 2b. Stir the mixture for at least 18 hr. 3. Remove the resin by filtration through filter paper. 4. Test the filtrate for the presence of the biological stain (see Support Protocol 2). 5. Discard the resin as hazardous solid waste and the decontaminated filtrate as liquid waste. CONTINUOUS-FLOW DECONTAMINATION OF AQUEOUS SOLUTIONS OF BIOLOGICAL STAINS
ALTERNATE PROTOCOL 3
For treating large volumes of dilute aqueous solutions of biological stains (Table 3.4.5), it is possible to put the resin in a column and run the contaminated solution through the column in a continuous-flow system (Lunn et al., 1994). Limited grinding of the Amberlite XAD-16 resin increases its efficiency. Additional Materials (also see Basic Protocol 2) 25 µg/ml biological stain in water Methanol (optional) 300-mm × 11-mm-i.d. glass chromatography column fitted with threaded adapters and flow-regulating valves at top and bottom nut and insert connectors, and insertion tool (Ace Glass) or 300-mm × 15-mm-i.d. glass chromatography column (Spectrum 124010, Fisher) Glass wool 1.5-mm-i.d. × 0.3-mm-wall Teflon tubing Waring blender (optional) Rubber stopper fitted over a pencil QG 20 lab pump (Fluid Metering) Additional reagents and equipment for testing for the presence of biological stain (see Support Protocol 2)
Safety Procedures and Quality Control
3.4.15 Current Protocols in Cytometry
Supplement 20
Table 3.4.6 Breakthrough Volumes for Continuous-Flow Decontamination of Biological Stains
Compound Acridine orange Alizarin red S Azure A Azure B Cresyl violet acetate Crystal violet Ethidium bromide Janus green B Methylene blue Neutral red Safranine O Toluidine blue O
Breakthrough volume (ml) Limit of detection
1 ppm
5 ppm
465 120 615 630 706 1020 260 170 420 >2480 365 353
>990 150 810 882 >1396 >1630 312 650 645 >2480 438 494
>990 240 >975 >1209 >1396 >1630 416 >870 1050 >2480 584 606
Using a slurry of Amberlite XAD-16 1a. Prepare a 300-mm × 11-mm-i.d. glass chromatography column. To prevent clogging of the column outlet, place a small plug of glass wool at the bottom of the chromatography column. Connect 1.5-mm-i.d. × 0.3-mm wall Teflon tubing to the adapters using nut and insert connectors. Attach the tubing using an insertion tool. 2a. Mix 10 g Amberlite XAD-16 and 25 ml water in a beaker and stir 5 min to wet the resin. Using a finely ground Amberlite XAD-16 slurry 1b. Prepare a 300-mm × 15-mm-i.d. glass chromatography column. To prevent clogging of the column outlet, place a small plug of glass wool at the bottom of the chromatography column. 2b. Grind 20 g Amberlite XAD-16 with 200 ml water for exactly 10 sec in a Waring blender. 3. Pour the resin slurry into the column through a funnel. As the resin settles, tap the column with a rubber stopper fitted over a pencil to encourage even packing. Attach a QG 20 lab pump. 4. Pump the 25-µg/ml biological stain solution through the column at 2 ml/min. Alternatively, gravity flow coupled with periodic adjustment of the flow-regulating valve can be used.
5. Check the effluent from the column for the presence of biological stain (see Support Protocol 2). Stop the pump when stain is detected. Table 3.4.6 lists breakthrough volumes at different detection levels for a number of biological stains.
6. Discard the decontaminated effluent and the contaminated resin appropriately. 7. Many biological stains can be washed off the resin with methanol so the resin can be reused. Discard the methanol solution of stain as hazardous organic liquid waste. Safe Use of Hazardous Chemicals
3.4.16 Supplement 20
Current Protocols in Cytometry
Table 3.4.7
Methods for Detecting Biological Stainsa
Compound
Reagentb
Procedure Wavelength(s)(nm)
Limit of detection (ppm)
Acridine orange Alcian blue 8GX Alizarin red S Azure A Azure B Brilliant blue R Congo red Coomassie brilliant blue G Cresyl violet acetate Crystal violet Eosin B Erythrosin B Ethidium bromide Janus green B Methylene blue Neutral red Nigrosin Orcein Propidium iodide Rose Bengal Safranine O Toluidine blue O Trypan blue
DNA solution
F A A A A A A A
ex 492, em 528 615 556 633 648 585 497 610
0.0032 0.9 0.46 0.15 0.13 1.0 0.25 1.7
F A A F F A A A A A F F F A A
ex 588, em 618 588 514 ex 488, em 556 ex 540, em 590 660 661 540 570 579 ex 350, em 600 ex 520, em 576 ex 460, em 582 626 607
0.021 0.1 0.21 0.025 0.05 0.6 0.13 0.6 0.8 1.15 0.1 0.04 0.03 0.2 0.22
1 M KOH
pH 5 buffer
pH 5 buffer 1 M KOH DNA solution
aAbbreviations: A, absorbance; em, emission; ex, excitation; F, fluorescence bSee Support Protocol 2
ANALYTICAL PROCEDURES TO DETECT BIOLOGICAL STAIN Depending on the biological stain, the filtrate or eluate from the decontamination procedure can be analyzed using either UV absorption (A) or fluorescence detection (F).
SUPPORT PROTOCOL 2
Materials Filtrate or eluate from biological stain decontamination (see Basic Protocol 2 or Alternate Protocol 3) pH 5 buffer (see recipe) 1 M KOH solution 20 µg/ml calf thymus DNA in TBE electrophoresis buffer, pH 8.1 (APPENDIX 2A) Test the filtrate or eluate from the biological stain decontamination procedure using the appropriate method as indicated in Table 3.4.7. Traces of acid or base on the resin may induce color changes in the stain. Accordingly, with cresyl violet acetate or neutral red, mix aliquots of the filtrate with 1 vol pH 5 buffer before analyzing. With alizarin red S and orcein, mix aliquots of the filtrate with 1 vol of 1 M KOH solution before analyzing. Increase the fluorescence of solutions of acridine orange, ethidium bromide, and propidium iodide by mixing an aliquot of the filtrate with an equal volume of 20 ìg/ml calf thymus DNA in TBE electrophoresis buffer, pH 8.1. Let the solution stand 15 min before measuring the fluorescence.
Safety Procedures and Quality Control
3.4.17 Current Protocols in Cytometry
Supplement 20
BASIC PROTOCOL 3
DISPOSAL OF CHLOROTRIMETHYLSILANE AND DICHLORODIMETHYLSILANE Silane-containing compounds are hydrolyzed to hydrochloric acid and polymeric siliconcontaining material (Patnode and Wilcock, 1946). 1. Hydrolyze silane-containing compounds by cautiously adding 5 ml silane to 100 ml vigorously stirred water in a flask. Allow the resulting suspension to settle. 2. Remove any insoluble material by filtration and discard it with the solid or liquid hazardous waste. 3. Neutralize the aqueous layer with base and discard it.
BASIC PROTOCOL 4
DISPOSAL OF CYANIDES AND CYANOGEN BROMIDE Inorganic cyanides (e.g., NaCN) and cyanogen bromide (CNBr) are oxidized by sodium hypochlorite (NaOCl; e.g., Clorox) in basic solution to the much less toxic cyanate ion (Lunn and Sansone, 1985a). Destruction is >99.7%. Materials Cyanide (e.g., NaCN) or cyanogen bromide (CNBr) 1 M NaOH 5.25% (v/v) sodium hypochlorite (NaOCl; i.e., standard household bleach) Additional reagents and equipment for testing for the presence of cyanide (see Support Protocol 3) 1. Dissolve cyanide in water to give a concentration ≤25 mg/ml or dissolve CNBr in water to give a concentration ≤60 mg/ml. If necessary, dilute aqueous solutions so the concentration of NaCN or CNBr does not exceed the limit.
2. Mix 1 vol NaCN or CNBr solution with 1 vol 1 M NaOH and 2 vol fresh 5.25% NaOCl. Stir the mixture 3 hr. IMPORTANT NOTE: With age bleach may become ineffective; use of fresh bleach is strongly recommended.
3. Test the reaction mixture for the presence of cyanide (see Support Protocol 3). 4. Neutralize the reaction mixture and discard it. SUPPORT PROTOCOL 3
ANALYTICAL PROCEDURE TO DETECT CYANIDE This protocol is used to detect cyanide or cyanogen bromide at ≥3 µg/ml. Materials Cyanide or cyanogen bromide decontamination reaction mixture (see Basic Protocol 4) Phosphate buffer (see recipe) 10 mg/ml sodium ascorbate in water: prepare fresh daily 100 mg/liter NaCN in water: prepare fresh weekly 10 mg/ml chloramine-T in water: prepare fresh daily Cyanide detection reagent (see recipe) Sorvall GLC-1 centrifuge or equivalent
Safe Use of Hazardous Chemicals
1. If necessary to remove suspended solids, centrifuge two 1-ml aliquots of the cyanide or cyanogen bromide decontamination reaction mixture 5 min in a benchtop centrifuge, room temperature. Add each supernatant to 4 ml phosphate buffer in separate tubes.
3.4.18 Supplement 20
Current Protocols in Cytometry
2. If an orange or yellow color appears, add 10 mg/ml freshly prepared sodium ascorbate dropwise until the mixture is colorless, but do not add more than 2 ml. 3. Add 200 µl of 100 mg/liter NaCN to one reaction mixture (control solution). 4. Add 1 ml freshly prepared 10 mg/ml chloramine-T to each tube. Shake the tubes and let them stand 1 to 2 min. 5. Add 1 ml cyanide detection reagent, shake, and let stand 5 min. A blue color indicates the presence of cyanide. If destruction has been complete and the analytical procedure has been carried out correctly, the treated reaction mixture should be colorless and the control solution, which contains NaCN, should be blue.
6. Centrifuge tubes 5 min, room temperature, if necessary to remove suspended solids. Measure the absorbance at 605 nm with appropriate standards and blanks. DISPOSAL OF DIMETHYL SULFATE, DIETHYL SULFATE, METHYL METHANESULFONATE, ETHYL METHANESULFONATE, DIEPOXYBUTANE, AND 1,3-PROPANE SULTONE
BASIC PROTOCOL 5
Dimethyl sulfate is hydrolyzed by base to methanol and methyl hydrogen sulfate (Lunn and Sansone, 1985b). Subsequent hydrolysis of methyl hydrogen sulfate to methanol and sulfuric acid is slow. Methyl hydrogen sulfate is nonmutagenic and a very poor alkylating agent. The other compounds can be hydrolyzed with base in a similar fashion (Lunn and Sansone, 1990a). Destruction is >99%. A method to verify that decontamination is complete is also provided (see Support Protocol 4). NOTE: The reaction times given in the protocol should give good results; however, reaction time may be affected by such factors as the size and shape of the flask and the rate of stirring. The presence of two phases indicates that the reaction is not complete, and stirring should be continued until the reaction mixture is homogeneous. Materials Dimethyl sulfate, diethyl sulfate, methyl methanesulfonate, ethyl methanesulfonate, diepoxybutane, or 1,3-propane sultone 5 M NaOH Additional reagents and equipment for testing for the presence of dimethyl sulfate, diethyl sulfate, methyl methanesulfonate, ethyl methanesulfonate, diepoxybutane, or 1,3-propane sultone (see Support Protocol 4) For bulk quantities of dimethyl sulfate 1a. Add 100 ml dimethyl sulfate to 1 liter of 5 M NaOH. Stir the reaction mixture. 2a. Fifteen minutes after all the dimethyl sulfate has gone into solution, neutralize the reaction mixture with acid. For bulk quantities of diethyl sulfate 1b. Add 100 ml diethyl sulfate to 1 liter of 5 M NaOH. Stir the reaction mixture 24 hr. 2b. Neutralize the reaction mixture with acid.
Safety Procedures and Quality Control
3.4.19 Current Protocols in Cytometry
Supplement 20
For bulk quantities of methyl methanesulfonate, ethyl methanesulfonate, diepoxybutane, and 1,3-propane sultone 1c. Add 1 ml methyl methanesulfonate, ethyl methanesulfonate, or diepoxybutane, or 1 g of 1,3-propane sultone to 10 ml of 5 M NaOH. Stir the reaction mixture 1 hr for 1,3-propane sultone, 2 hr for methyl methanesulfonate, 22 hr for diepoxybutane, or 24 hr for ethyl methanesulfonate. 2c. Neutralize the reaction mixture with acid. 3. Test the reaction mixture for the presence of the original compound (see Support Protocol 4). 4. Discard the decontaminated reaction mix. SUPPORT PROTOCOL 4
ANALYTICAL PROCEDURE TO DETECT THE PRESENCE OF DIMETHYL SULFATE, DIETHYL SULFATE, METHYL METHANESULFONATE, ETHYL METHANESULFONATE, DIEPOXYBUTANE, AND 1,3-PROPANE SULTONE This protocol is used to verify decontamination of solutions containing dimethyl sulfate, diethyl sulfate, methyl methanesulfonate, ethyl methanesulfonate, diepoxybutane, or 1,3-propane sultone. The detection limit for this assay is 40 µg/ml for dimethyl sulfate, 108 µg/ml for diethyl sulfate, 84 µg/ml for methyl methanesulfonate, 1.1 µg/ml for ethyl methanesulfonate, 360 µg/ml for diepoxybutane, and 264 µg/ml for 1,3-propane sultone. Materials Reaction mixture containing dimethyl sulfate, diethyl sulfate, methyl methanesulfonate, ethyl methanesulfonate, diepoxybutane, or 1,3-propane sultone 98:2 (v/v) 2-methoxyethanol/acetic acid 5% (w/v) 4-(4-nitrobenzyl)pyridine in 2-methoxyethanol Piperidine 2-Methoxyethanol 1. Dilute an aliquot of the reaction mixture with 4 vol water. 2. Add 100 µl diluted reaction mixture to 1 ml of 98:2 (v/v) 2-methoxyethanol/acetic acid. Swirl to mix. 3. Add 1 ml of 5% (w/v) 4-(4-nitrobenzyl)pyridine in 2-methoxyethanol. Heat 10 min at 100°C, then cool 5 min in ice. 4. Add 0.5 ml piperidine and 2 ml of 2-methoxyethanol. 5. Measure the absorbance of the violet reaction mixture at 560 nm against an appropriate blank. The absorbance of a decontaminated solution should be 0.000.
BASIC PROTOCOL 6
Safe Use of Hazardous Chemicals
DISPOSAL OF ETHIDIUM BROMIDE AND PROPIDIUM IODIDE Ethidium bromide and propidium iodide in water and buffer solutions may be degraded by reaction with sodium nitrite and hypophosphorous acid in aqueous solution (Lunn and Sansone, 1987); destruction is >99.87%. This reaction may also be used to decontaminate equipment contaminated with ethidium bromide (see Alternate Protocol 4; Lunn and Sansone, 1989) and to degrade ethidium bromide in organic solvents (see Alternate Protocol 5 and Alternate Protocol 6; Lunn and Sansone, 1990b). Ethidium bromide and propidium iodide may also be removed from solution by adsorption onto Amberlite XAD-16 resin (see Basic Protocol 2).
3.4.20 Supplement 20
Current Protocols in Cytometry
Materials Ethidium bromide– or propidium iodide–containing solution in water, buffer, or 1 g/ml cesium chloride 5% (v/v) hypophosphorous acid: prepare fresh daily by diluting commercial 50% reagent 1/10 0.5 M sodium nitrite: prepare fresh daily Sodium bicarbonate Additional reagents and equipment for testing for the presence of ethidium bromide or propidium iodide (see Support Protocol 5) 1. If necessary, dilute the ethidium bromide– or propidium iodide–containing solution so the concentration of ethidium bromide or propidium iodide is ≤0.5 mg/ml. 2. For each 100 ml solution, add 20 ml of 5% hypophosphorous acid solution and 12 ml of 0.5 M sodium nitrite. Stir briefly and let stand 20 hr. 3. Neutralize the reaction mixture by adding sodium bicarbonate until the evolution of gas ceases. 4. Test the reaction mixture for the presence of ethidium bromide or propidium iodide (see Support Protocol 5). 5. Discard the decontaminated reaction mixture. DECONTAMINATION OF EQUIPMENT CONTAMINATED WITH ETHIDIUM BROMIDE
ALTERNATE PROTOCOL 4
Glass, stainless steel, Formica, floor tile, and the filters of transilluminators have been successfully decontaminated using this protocol. No change in the optical properties of the transilluminator filter could be detected, even after a number of decontamination cycles. Materials Equipment contaminated with ethidium bromide Decontamination solution (see recipe) Sodium bicarbonate Additional reagents and equipment for testing for the presence of ethidium bromide (see Support Protocol 5) 1. Wash the equipment contaminated with ethidium bromide once with a paper towel soaked in decontamination solution. The pH of the decontamination solution is 1.8. If this would be too corrosive for the surface to be decontaminated, wash with a paper towel soaked in water instead.
2. Wash the surface five times with paper towels soaked in water using a fresh towel each time. 3. Soak all the towels 1 hr in decontamination solution. 4. Neutralize the decontamination solution by adding sodium bicarbonate until the evolution of gas ceases. 5. Test the decontamination solution for the presence of ethidium bromide (see Support Protocol 5). 6. Discard the decontamination solution and the paper towels as nonhazardous liquid and solid wastes.
Safety Procedures and Quality Control
3.4.21 Current Protocols in Cytometry
Supplement 20
ALTERNATE PROTOCOL 5
DECONTAMINATION OF ETHIDIUM BROMIDE IN ISOPROPANOL SATURATED WITH CESIUM CHLORIDE Materials Ethidium bromide in isopropanol saturated with cesium chloride Decontamination solution (see recipe) Sodium bicarbonate Additional reagents and equipment for testing for the presence of ethidium bromide (see Support Protocol 5) 1. If necessary, dilute the ethidium bromide in isopropanol saturated with cesium chloride so the concentration of ethidium bromide is ≤1 mg/ml. 2. For each volume of ethidium bromide solution, add 4 vol decontamination solution. Stir the reaction mixture 20 hr. 3. Neutralize the reaction mixture by adding sodium bicarbonate until the evolution of gas ceases. 4. Test the reaction mixture for the presence of ethidium bromide (see Support Protocol 5). 5. Discard the decontaminated solution.
ALTERNATE PROTOCOL 6
DECONTAMINATION OF ETHIDIUM BROMIDE IN ISOAMYL ALCOHOL AND 1-BUTANOL Materials Ethidium bromide in isoamyl alcohol or 1-butanol Decontamination solution (see recipe) Activated charcoal Sodium bicarbonate Separatory funnel Additional reagents and equipment for testing for the presence of ethidium bromide 1. If necessary, dilute the ethidium bromide in isoamyl alcohol or 1-butanol so the concentration is ≤1 mg/ml final. 2. For each volume of ethidium bromide solution, add 4 vol decontamination solution. Stir the two-phase reaction mixture rapidly for 72 hr. 3. For each 100 ml of reaction mixture, add 2 g activated charcoal. Stir another 30 min. 4. Filter the reaction mixture. 5. Neutralize the filtrate by adding sodium bicarbonate until the evolution of gas ceases. Separate the layers using a separatory funnel. More alcohol may tend to separate from the aqueous layer on standing.
6. Test the alcohol and aqueous layers for the presence of ethidium bromide. 7. Discard the alcohol and aqueous layers appropriately. Discard the activated charcoal as solid waste. The aqueous layer contains 4.6% 1-butanol or 2.3% isoamyl alcohol.
Safe Use of Hazardous Chemicals
3.4.22 Supplement 20
Current Protocols in Cytometry
ANALYTICAL PROCEDURE TO DETECT ETHIDIUM BROMIDE OR PROPIDIUM IODIDE This protocol is used to verify that solutions no longer contain ethidium bromide or propidium iodide. The limits of detection are 0.05 parts per million (ppm) for ethidium bromide and 0.1 ppm for propidium iodide.
SUPPORT PROTOCOL 5
Materials Reaction mixture containing ethidium bromide or propidium iodide TBE buffer, pH 8.1 (APPENDIX 2A) 20 µg/ml calf thymus DNA in TBE buffer, pH 8.1 1. Mix 100 µl reaction mixture containing ethidium bromide or propidium iodide with 900 µl TBE buffer, pH 8.1. 2. Add 1 ml of 20 µg/ml calf thymus DNA in TBE, pH 8.1. Prepare a blank solution (100 µl water + 900 µl TBE + 1 ml of 20 µg/ml calf thymus DNA) and control solutions containing known quantities of ethidium bromide or propidium iodide. Let the mixtures stand 15 min. 3. To detect ethidium bromide, measure the fluorescence with an excitation wavelength of 540 nm and an emission wavelength of 590 nm. To detect propidium iodide, measure the fluorescence with an excitation wavelength of 350 nm and an emission wavelength of 600 nm. If a spectrophotofluorometer is not available, fluorescence of ethidium bromide can be qualitatively determined using a hand-held UV lamp on the long-wavelength setting (Lunn and Sansone, 1991c).
DISPOSAL OF HYDROGEN PEROXIDE Hydrogen peroxide can be reduced with sodium metabisulfite (Lunn and Sansone, 1994b).
BASIC PROTOCOL 7
Materials 30% hydrogen peroxide 10% (w/v) sodium metabisulfite 10% (w/v) potassium iodide 1 M HCl 1% (w/v) starch indicator solution 1. Add 5 ml of 30% hydrogen peroxide to 100 ml of 10% sodium metabisulfite. Stir the mixture at room temperature until the temperature starts to drop, indicating that the reaction is over. 2. Test for the presence of hydrogen peroxide by adding a few drops of the reaction mixture to an equal volume of 10% potassium iodide. Add a few drops of 1 M HCl to acidify the reaction mixture, then add a drop of 1% starch indicator solution. A deep blue color indicates the presence of excess oxidant. If necessary, add more 10% sodium metabisulfite until the starch test is negative.
3. Discard the decontaminated mixture.
Safety Procedures and Quality Control
3.4.23 Current Protocols in Cytometry
Supplement 20
BASIC PROTOCOL 8
REDUCTION OF IODINE Iodine is reduced with sodium metabisulfite to iodide (Lunn and Sansone, 1994b). Materials Iodine 10% (w/v) sodium metabisulfite 1 M HCl 1% (w/v) starch indicator solution 1. Add 5 g iodine to 100 ml of 10% sodium metabisulfite. Stir the mixture until the iodine has completely dissolved. 2. Acidify a few drops of the reaction mixture by adding a few drops of 1 M HCl. Add 1 drop of 1% starch indicator solution. A deep blue color indicates the presence of iodine. If reduction is not complete, add more sodium metabisulfite solution.
3. Dispose of the decontaminated solution. BASIC PROTOCOL 9
DISPOSAL OF MERCURY COMPOUNDS Solutions of mercuric acetate can be decontaminated using Dowex 50X8-100, a strongly acidic gel-type ion-exchange resin with a sulfonic acid functionality. Solutions of mercuric chloride can be decontaminated using Amberlite IRA-400(Cl), a strongly basic gel-type ion-exchange resin with a quaternary ammonium functionality. The final concentration of mercury is <3.8 ppm (Lunn and Sansone, 1994a). On a small scale it is most convenient to stir the resin in the solution to be decontaminated, but on a larger scale, or for routine use, it may be more convenient to pass the solution through a column packed with the resin. Although the volume of waste that must be disposed of is greatly reduced using this technique, a small amount of waste (i.e., the resin contaminated with mercury) remains and must be discarded appropriately. Resin can be regenerated by washing with acid, but the concentrated metal-containing solution generated by this must also be disposed of appropriately. Mercury may also be removed from laboratory waste water using a column of iron powder (see Alternate Protocol 7). Support Protocol 6 is used to detect the presence of mercury. Materials Solution containing ≤1600 ppm mercuric acetate or ≤1350 ppm mercuric chloride Dowex 50X8-100 ion-exchange resin or Amberlite IRA-400(Cl) ion-exchange resin Additional reagents and equipment to test for the presence of mercury (see Support Protocol 6) 1a. For mercuric acetate: For each 200 ml of solution containing ≤1600 ppm mercuric acetate, add 1 g Dowex 50X8-100 ion-exchange resin. Stir the mixture 1 hr, then filter through filter paper. 1b. For mercuric chloride: For each 200 ml of solution containing ≤1350 ppm mercuric chloride, add 1 g Amberlite IRA-400(Cl) ion-exchange resin. Stir the mixture 6 hr, then filter through filter paper. The speed and efficiency of decontamination will depend on factors such as the size and shape of the flask and the rate of stirring.
Safe Use of Hazardous Chemicals
3. Test the filtrate for the presence of mercury (see Support Protocol 6). 4. Discard the decontaminated filtrate and the mercury-containing resin appropriately.
3.4.24 Supplement 20
Current Protocols in Cytometry
DECONTAMINATION OF WASTE WATER CONTAINING MERCURY Laboratory waste water that contains mercury is decontaminated by passing it through a column of iron powder. The mercury forms mercury amalgam and stays on the column. Some metallic mercury remains in solution but this can be removed by aeration. The final concentration of mercury is <5 ppb (Shirakashi et al., 1986).
ALTERNATE PROTOCOL 7
Materials Iron powder, 60 mesh Waste water containing ≤2.5 ppm mercury 6-mm-i.d. column 1. Pack a 6-mm-i.d. column with 1 g of 60-mesh iron powder. Use a fresh column for each treatment.
2. Pass ≤2 liters of water containing ≤2.5 ppm of mercury through the column at a flow rate of 250 ml/hr. Solutions containing a higher concentration of mercury may also be treated, but this will result in a higher final concentration of mercury (e.g., treating a 100-ppm solution in this fashion yielded 33 ppb final). Some iron ends up in solution and can be removed by adjusting the pH to 8. The resulting precipitated Fe(OH)3 can then be removed by filtration.
3. Aerate the resulting effluent to remove traces of metallic mercury and continue aeration 30 min after the last of the effluent has emerged from the column. Vent the metallic mercury removed from the solution by aeration into the fume hood or capture it in a mercury trap. The effluent contains <5 ppb mercury. The presence of iodide or polypeptone may necessitate several treatments to reduce the mercury to an acceptable level.
ANALYTICAL PROCEDURE TO DETECT MERCURY Atomic absorption spectroscopy with detection at 253.7 nm, a slit width of 0.7 nm, and a limit of detection of 3.8 ppm can be used to determine the concentration of mercury in solution for experiments involving ion-exchange resins. A Hiranuma mercury meter model HG-1 can be used for experiments involving iron powder.
SUPPORT PROTOCOL 6
DISPOSAL OF SODIUM AZIDE Sodium azide can be oxidized by ceric ammonium nitrate (Manufacturing Chemists Association, 1973) to nitrogen (Mason, 1967) or by nitrous acid (National Research Council, 1983) to nitrous oxide (Mason, 1967); destruction is >99.996%. Sodium azide in buffer solution may also be degraded by the addition of sodium nitrite (Lunn and Sansone, 1994a). The reaction proceeds much more readily at low pH, but if sufficient sodium nitrite is added, it will proceed to completion even at high pH. At low pH, it may be possible to completely degrade the azide present in the buffer with less than the amount of sodium nitrite indicated. However, the reaction mixture must be carefully checked to ensure that no azide remains (see Support Protocol 7). At high pH it is possible for unreacted azide to remain in the presence of excess nitrite. Residual nitrite can be detected using Support Protocol 8.
BASIC PROTOCOL 10
CAUTION: Some toxic nitrogen dioxide may be produced as a by-product of these reactions, so they should always be carried out in a chemical fume hood. Safety Procedures and Quality Control
3.4.25 Current Protocols in Cytometry
Supplement 20
Materials Sodium azide or solution containing sodium azide Ceric ammonium nitrate 10% (w/v) potassium iodide 1 M HCl 1% (w/v) starch indicator solution Sodium nitrite 4 M sulfuric acid Additional reagents and equipment to test for the presence of sodium azide (see Support Protocol 7) or nitrite (see Support Protocol 8) Decontamination using ceric ammonium nitrate 1a. For each gram of sodium azide, add 9 g ceric ammonium nitrate to 30 ml of water, and stir until it has dissolved. 2a. Dissolve each gram of sodium azide in 5 ml water. Add this solution to the ceric ammonium nitrate solution at the rate of 1 ml each min. Stir 1 hr. If the reaction is carried out on a larger scale, an ice bath may be required for cooling.
3a. Check that the reaction is still oxidizing by adding a few drops of the reaction mixture to an equal volume of 10% potassium iodide. Acidify the mixture with 1 drop 1 M HCl and add 1 drop 1% starch indicator solution. The deep blue color of the starch-iodine complex indicates that excess oxidant is present. If excess oxidant is not present, add more ceric ammonium nitrate.
4a. Test for the presence of sodium azide (see Support Protocol 8). 5a. Discard the decontaminated reaction mixture. Decontamination using sodium nitrite 1b. For each 5 g sodium azide, dissolve 7.5 g sodium nitrite in 30 ml water. 2b. Dissolve each 5 g sodium azide in 100 ml water. Add the sodium nitrite solution with stirring. Slowly add 4 M sulfuric acid until the reaction mixture is acidic to litmus. Stir 1 hr. CAUTION: It is important to add the sodium nitrite, then the sulfuric acid. Adding these reagents in reverse order will generate explosive, volatile, toxic hydrazoic acid. If the reaction is carried out on a large scale, an ice bath may be required for cooling.
3b. Check that there is excess nitrous acid in the reaction. Add a few drops of the reaction mixture to an equal volume of 10% potassium iodide. Acidify the mixture with 1 drop 1 M HCl. Add 1 drop starch indicator solution. The deep blue color of the starch-iodine complex indicates that excess nitrous acid is present. If excess nitrous acid is not present, add more sodium nitrite.
4b. If excess nitrous acid is present, test for the presence of sodium azide (see Support Protocol 7). 5b. Discard the decontaminated reaction mixture.
Safe Use of Hazardous Chemicals
Decontamination of sodium azide in buffer 1c. If necessary, dilute the buffer solution with water so the concentration of sodium azide is ≤1 mg/ml. 2c. For each 50 ml buffer solution add 5 g sodium nitrite. Stir the reaction 18 hr.
3.4.26 Supplement 20
Current Protocols in Cytometry
3c. Test for the presence of sodium azide (see Support Protocol 7). 4c. Discard the decontaminated reaction solution. ANALYTICAL PROCEDURES TO DETECT SODIUM AZIDE Sodium azide is analyzed by reacting azide ion with 3,5-dinitrobenzoyl chloride to form 3,5-dinitrobenzoyl azide, which can be detected by reversed-phase HPLC. The limit of detection of this assay is 0.2 µg/ml sodium azide. This protocol works only in the absence of nitrite; verify that all of the nitrite has been destroyed by sulfamic acid by using the method detailed later in this unit (see Support Protocol 8).
SUPPORT PROTOCOL 7
Materials Reaction mixture from sodium azide treated with ceric ammonium nitrate or sodium nitrite 1 M KOH Acetonitrile Sodium azide indicator solution (see recipe) 0.2 M HCl 20% (w/v) sulfamic acid 3,5-dinitrobenzoyl chloride 50:50 (v/v) acetonitrile/water Sorvall GLC-1 centrifuge or equivalent 25-cm × 4.6-mm-i.d. Microsorb C-8 reversed-phase HPLC column (Varian) or equivalent Additional reagents and equipment for reversed-phase liquid chromatography (Snyder et al., 1997) To analyze for azide in the presence of ceric salts 1a. To a 10-ml aliquot of the reaction mixture from sodium azide treated with ceric ammonium nitrate add 40 ml water. Add 5 ml of this diluted solution to 3 ml of 1 M KOH and mix by shaking. If <3 ml of 1 M KOH is used, precipitation of ceric salts will not be complete.
2a. Centrifuge the mixture 5 min, room temperature. 3a. Remove 2 ml supernatant and add to 1 ml acetonitrile. Add 1 drop sodium azide indicator solution, add 0.2 M HCl dropwise until the mixture turns yellow, then add 1 drop more. To analyze for azide in the presence of nitrite 1b. To 5 ml of the reaction mixture from sodium azide treated with sodium nitrite add ≥1 ml sulfamic acid to remove excess nitrite. Let stand ≥3 min. More sulfamic acid solution may be required for strongly basic reaction mixtures or those containing high concentrations of nitrite. Complete removal of nitrite can be checked by using a modified Griess reagent (see Support Protocol 8). At high pH the reaction between azide and nitrite is quite slow, so the presence of excess nitrite does not mean that all the azide has been degraded.
2b. Add 1 drop sodium azide indicator solution, then basify the mixture by adding 1 M KOH until it turns purple (typically, 3 to 10 ml are required). 3b. Add 2 ml acetonitrile. Add 0.2 M HCl dropwise until the mixture turns yellow, then add 1 drop more. If >1 ml sulfamic acid is used, add 4 ml acetonitrile.
Safety Procedures and Quality Control
3.4.27 Current Protocols in Cytometry
Supplement 20
4. Prepare a 10% (w/v) solution of 3,5-dinitrobenzoyl chloride in acetonitrile. 5. Add 50 µl of 10% dinitrobenzoyl chloride/acetonitrile to the reaction mix (step 3a or 3b). Shake the mixture and let it stand ≥3 min. Longer standing times have no effect on the HPLC analysis. However, it is crucial to use freshly prepared 3,5-dinitrobenzoyl chloride solution within minutes of its preparation. It is generally most convenient to prepare all the analytical samples with the fresh solution at the beginning of the day and analyze them over the course of the day.
6. Analyze 20 µl of each reaction mixture by reversed-phase HPLC (Snyder et al., 1997) using a mobile phase of 50:50 (v/v) acetonitrile/water with a flow rate of 1 ml/min and UV detection at 254 nm. The peak for 3,5-dinitrobenzoyl azide elutes at ∼9 min. SUPPORT PROTOCOL 8
ANALYTICAL PROCEDURE TO DETECT NITRITE This protocol uses a modified Griess reagent to test for the presence of nitrite. The limit of detection of this assay is 0.06 µg/ml nitrite. A similar procedure uses N-(1-naphthyl)ethylenediamine (Cunniff, 1995). Materials α-Naphthylamine 15% (v/v) aqueous acetic acid Sulfanilic acid solution (see recipe) Reaction mixture treated to remove excess nitrite (see Support Protocol 7, step 1b)
Table 3.4.8
Compound
Safe Use of Hazardous Chemicals
AEBSF AEBSF AEBSF APMSF APMSF APMSF APMSF DFP DFP DFP DFP PMSF PMSF PMSF TLCK TLCK TLCK TPCK TPCK TPCK
Conditions for the Destruction of Enzyme Inhibitors
Concentration 1 mM 20 mM 20 mM 2.5 mM 25 mM 25 mM 100 mM 10 mM 200 mM pure 10 mM 10 mM 100 mM 100 mM 1 mM 5 mM 5 mM 1 mM 1 mM 1 mM
Solvent
Solution: 1 M NaOH
Time
Buffer(pH 5.0-8.0) DMSO Isopropanol Buffer(pH 5.0-8.0) DMSO 50:50 isopropanol:pH 3 buffer Water Buffer (pH 6.4-7.2) DMF — Water Buffer (pH 5.0-8.0) DMSO Isopropanol Buffer (pH 5.0-8.0) DMSO Water Buffer (pH 5.0-8.0) DMSO Isopropanol
1:0.1 1:10 1:10 1:0.1 1:5 1:5 1:5 1:0.2 1:2 1:25 1:0.2 1:0.1 1:5 1:5 1:0.1 1:5 1:0.1 1:0.1 1:0.1 1:0.1
1 hr 24 hr 24 hr 1 hr 24 hr 24 hr 24 hr 18 hr 18 hr 1 hr 18 hr 1 hr 24 hr 24 hr 18 hr 18 hr 18 hr 18 hr 18 hr 18 hr
3.4.28 Supplement 20
Current Protocols in Cytometry
1. Prepare the modified Griess reagent by boiling 0.1 g α-naphthylamine in 20 ml water until it dissolves. While the solution is still hot, pour it into 150 ml of 15% aqueous acetic acid. Add 150 ml sulfanilic acid solution. This reagent should be stored at room temperature in a brown bottle. CAUTION: α-Naphthylamine is a carcinogen.
2. Add 3 ml of the reaction mixture treated to remove excess nitrite to 1 ml modified Griess reagent. Let stand 6 min at room temperature. 3. Measure the absorbance at 520 nm against a suitable blank. DISPOSAL OF ENZYME INHIBITORS The enzyme inhibitors p-amidinophenylmethanesulfonyl fluoride (APMSF), 4-(2-aminoethyl)benzenesulfonyl fluoride (AEBSF), phenylmethylsulfonyl fluoride (PMSF; Lunn and Sansone, 1994c), diisopropyl fluorophosphate (DFP; Lunn and Sansone, 1994d), Nα-p-tosyl-L-lysine chloromethyl ketone (TLCK), and N-p-tosyl-L-phenylalanine chloromethyl ketone (TPCK; Lunn and Sansone, 1994c) may be degraded by reaction with 1 M NaOH. Destruction is >99.8% (except TPCK >98.3%). The exact reaction conditions depend on the solvent (see Table 3.4.8). The solutions that were decontaminated are representative of those described in the literature.
BASIC PROTOCOL 11
Materials Solutions of APMSF, AEBSF, PMSF, DFP, TLCK, or TPCK in buffer, DMSO, isopropanol, or water 1 M NaOH Glacial acetic acid Additional reagents and equipment for testing for the presence of the enzyme inhibitors (see Support Protocol 9) 1. If necessary, dilute the solutions with the same solvent so that the concentrations given in Table 3.4.8 are not exceeded. Bulk quantities of AEBSF, PMSF, and TPCK may be dissolved in isopropanol and bulk quantities of APMSF and TLCK may be dissolved in water at the concentrations shown in Table 3.4.8. Bulk quantities of DFP (a liquid) may be mixed directly with 1 M NaOH, taking care to make sure that all the DFP has mixed thoroughly, in the ratio shown in Table 3.4.8 (e.g., 40 ìl DFP with 1 ml of 1 M NaOH).
2. Add 1 M NaOH so that the ratio of solution to 1 M NaOH is that listed in Table 3.4.8. 3. Shake to ensure complete mixing, check that the solution is strongly basic (pH ≥12), and allow to stand for the time given in Table 3.4.8. 4. Neutralize the reaction mixture with acetic acid and test for the presence of residual enzyme inhibitor (see Support Protocol 9). 5. Discard the decontaminated reaction mixture.
Safety Procedures and Quality Control
3.4.29 Current Protocols in Cytometry
Supplement 20
Table 3.4.9 HPLC Conditions for Enzyme Inhibitors
Compound
Mobile phase
Detector
Retention time
Limit of detection
AEBSF
40:60 (v/v) acetonitrile:0.1% trifluoroacetic acid 40:60 (v/v) acetonitrile:0.1% trifluoroacetic acid 50:50 (v/v) acetonitrile:water 40:60 (v/v) acetonitrile:0.1% trifluoroacetic acid 48:52 (v/v) acetonitrile:10 mM pH 7 phosphate buffer
UV 225 nm
9.5 min
0.1 µg/ml
UV 232 nm
7.7 min
0.5 µg/ml
UV 220 nm
8 min
0.9 µg/ml
UV 228 nm
9.5 min
0.37 µg/ml
UV 228 nm
10.5 min
2 µg/ml
APMSF
PMSF TLCK
TPCK
SUPPORT PROTOCOL 9
ANALYTICAL PROCEDURES TO DETECT ENZYME INHIBITORS DFP can be determined using a complex procedure involving the inhibition of chymotrypsin activity. For more information, refer to Lunn and Sansone (1994d). A gas chromatographic method has also been described by Degenhardt-Langelaan and Kientz (1996). AEBSF, APMSF, PMSF, TLCK, and TPCK may be determined by reversed-phase HPLC (Snyder et al., 1997). The chromatographic conditions and limits of detection are shown in Table 3.4.9 (Lunn and Sansone, 1994c). Materials Decontaminated enzyme inhibitor solutions Acetonitrile (HPLC grade) Water (HPLC grade) 0.1% (v/v) trifluoroacetic acid in water 10 mM phosphate buffer, pH 7 250-mm × 4.6 mm-i.d. Microsorb C-8 reversed-phase HPLC column (Varian) or equivalent Additional reagents and equipment for reversed-phase liquid chromatography (Snyder et al., 1997) Analyze the decontaminated enzyme inhibitor solutions by reversed-phase HPLC using a 250-mm × 4.6-mm-i.d. Microsorb C-8 reversed-phase column, or equivalent, using the conditions shown in Table 3.4.9. In each case, the injection volume was 20 µl, the separation occurred at ambient temperature, and the flow rate was 1 ml/min. Check the analytical procedures by spiking an aliquot of the acidified reaction mixture with a small quantity of a dilute solution of the compound of interest.
Safe Use of Hazardous Chemicals
3.4.30 Supplement 20
Current Protocols in Cytometry
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Cyanide detection reagent Stir 3.0 g barbituric acid in 10 ml water. Add 15 ml of 4-methylpyridine and 3 ml concentrated HCl while continuing to stir. Cool and dilute to 50 ml with water. Store at room temperature. CAUTION: This reaction is exothermic.
Decontamination solution Dissolve 4.2 g sodium nitrite (0.2 M final) and 20 ml hypophosphorous acid (3.3% w/v final) in 300 ml water. Prepare fresh. Horseradish peroxidase Dissolve hydrogen-peroxide oxidoreductase (EC 1.11.1.7 [Type II]; specific activity 150 to 200 purpurogallin U/mg, Sigma) in 1 g/liter sodium acetate to give 30 U/ml. Prepare fresh daily. For small-scale reactions, a more dilute solution can be used to avoid working with inconveniently small volumes.
pH 5 buffer 2.04 g potassium hydrogen phthalate (0.05 M final) 38 ml 0.1 M potassium hydroxide (15 mM) H2O to 200 ml Store at room temperature Phosphate buffer 13.6 g monobasic potassium phosphate (KH2PO4; 0.1 M final) 0.28 g dibasic sodium phosphate (Na2HPO4; 2 mM final) 3.0 g potassium bromide (KBr; 25 mM final) 1 liter H2O Store at room temperature Potassium bromide is necessary to make the assay for cyanide work correctly.
Sodium azide indicator solution 0.1 g bromocresol purple (0.4% final) 18.5 ml 0.01 M potassium hydroxide (KOH; 7.4 mM final) H2O to 25 ml Store at room temperature Sulfanilic acid solution Dissolve 0.5 g sulfanilic acid in 150 ml of 15% (v/v) aqueous acetic acid. Use immediately. LITERATURE CITED Bretherick, L. (ed.) 1986. Hazards in the Chemical Laboratory, 4th ed. Royal Society of Chemistry, London. Bretherick, L., Urben, P.G., and Pitt, M.J. 1999. Bretherick’s Handbook of Reactive Chemical Hazards, 6th ed. Butterworth-Heinemann, London. Castegnaro, M., Barek, J., Dennis, J., Ellen, G., Klibanov, M., Lafontaine, M., Mitchum, R., van Roosmalen, P., Sansone, E.B., Sternson, L.A., and Vahl, M. (eds.) 1985. Laboratory Decontamination and Destruction of Carcinogens in Laboratory Wastes: Some Aromatic Amines and 4-Nitrobiphenyl. IARC Scientific Publications No. 64. International Agency for Research on Cancer, Lyon, France. Cunniff, P. (ed.) 1995. Official Methods of Analysis of the Association of Official Analytical Chemists, 16th ed., Ch. 4, p. 14. Association of Official Analytical Chemists, Arlington, Va.
Safety Procedures and Quality Control
3.4.31 Current Protocols in Cytometry
Supplement 20
Degenhardt-Langelaan, C.E.A.M. and Kientz, C.E. 1996. Capillary gas chromatographic analysis of nerve agents using large volume injections. J. Chromatogr. A.723:210-214. Forsberg, K. and Keith, L.H. 1999. Chemical Protective Clothing Performance Index Book, 2nd ed. John Wiley & Sons, New York. Furr, A.K. (ed.) 2000. CRC Handbook of Laboratory Safety, 5th ed. CRC Press, Boca Raton, Fla. Lewis, R.J. Sr. 1999. Sax’s Dangerous Properties of Industrial Materials, 10th ed. John Wiley & Sons, New York. Lunn, G. and Sansone, E.B. 1985a. Destruction of cyanogen bromide and inorganic cyanides. Anal. Biochem. 147:245-250. Lunn, G. and Sansone, E.B. 1985b. Validation of techniques for the destruction of dimethyl sulfate. Am. Ind. Hyg. Assoc. J. 46:111-114. Lunn, G. and Sansone, E.B. 1987. Ethidium bromide: Destruction and decontamination of solutions. Anal. Biochem. 162:453-458. Lunn, G. and Sansone, E.B. 1989. Decontamination of ethidium bromide spills. Appl. Ind. Hyg. 4:234-237. Lunn, G. and Sansone, E.B. 1990a. Validated methods for degrading hazardous chemicals: Some alkylating agents and other compounds. J. Chem. Educ. 67:A249-A251. Lunn, G. and Sansone, E.B. 1990b. Degradation of ethidium bromide in alcohols. BioTechniques 8:372-373. Lunn, G. and Sansone, E.B. 1991a. The safe disposal of diaminobenzidine. Appl. Occup. Environ. Hyg. 6:49-53. Lunn, G. and Sansone, E.B. 1991b. Decontamination of aqueous solutions of biological stains. Biotech. Histochem. 66:307-315. Lunn, G. and Sansone, E.B. 1991c. Decontamination of ethidium bromide spills-author’s response. Appl. Occup. Environ. Hyg. 6:644-645. Lunn, G. and Sansone, E.B. 1994a. Destruction of Hazardous Chemicals in the Laboratory, 2nd ed. John Wiley & Sons, New York. Lunn, G. and Sansone, E.B. 1994b. Safe disposal of highly reactive chemicals. J. Chem. Educ. 71:972-976. Lunn, G. and Sansone, E.B. 1994c. Degradation and disposal of some enzyme inhibitors. Scientific note. Appl. Biochem. Biotechnol. 48:57-59. Lunn, G. and Sansone, E.B. 1994d. Safe disposal of diisopropyl fluorophosphate (DFP). Appl. Biochem. Biotechnol. 49:165-171. Lunn, G., Klausmeyer, P.K., and Sansone, E.B. 1994. Removal of biological stains from aqueous solution using a flow-through decontamination procedure. Biotech. Histochem. 69:45-54. Manufacturing Chemists Association. 1973. Laboratory Waste Disposal Manual. p. 136. Manufacturing Chemists Association, Washington, D.C. Mason, K.G. 1967. Hydrogen azide. In Mellor’s Comprehensive Treatise on Inorganic and Theoretical Chemistry, Vol. VIII (Suppl. II) pp. l-15. John Wiley & Sons, New York. National Research Council. 1983. Prudent Practices for Disposal of Chemicals from Laboratories, p. 88. National Academy Press, Washington, D.C. O’Neil, M.J. (ed.) 2001. The Merck Index, 13th ed. Merck & Co., Whitehouse Station, N.J. Patnode, W. and Wilcock, D.F. 1946. Methylpolysiloxanes. J. Am. Chem. Soc. 68:358-363. Shirakashi, T., Nakayama, K., Kakii, K., and Kuriyama, M. 1986. Removal of mercury from laboratory waste water with iron powder. Chem. Abstr. 105:213690y. Snyder, L.R., Kirkland, J.J., and Glajch, J.L. 1997. Practical HPLC Method Development, 2nd ed. John Wiley & Sons, New York.
KEY REFERENCES The following are good general references for laboratory safety. American Chemical Society, Committee on Chemical Safety. 1995. Safety in Academic Chemistry Laboratories, 6th ed. American Chemical Society, Washington, D.C. Castegnaro, M. and Sansone, E.B. 1986. Chemical Carcinogens. Springer-Verlag, New York. DiBerardinis, L.J., First, M.W., Gatwood, G.T., and Seth, A.K. 2001. Guidelines for Laboratory Design, Health and Safety Considerations, 3rd ed. John Wiley & Sons, New York. Safe Use of Hazardous Chemicals
Fleming, D.D., Richardson, J.H., Tulis, J.J., and Vesley, D. 1995. Laboratory Safety, Principles and Practices, 2nd ed. American Society for Microbiology, Washington, D.C. Freeman, N.T. and Whitehead, J. 1982. Introduction to Safety in the Chemical Laboratory. Academic Press, San Diego.
3.4.32 Supplement 20
Current Protocols in Cytometry
Fuscaldo, A.A., Erlick, B.J., and Hindman, B. (eds.) 1980. Laboratory Safety, Theory and Practice. Academic Press, San Diego. Lees, R. and Smith, A.F. (eds.) 1984. Design, Construction, and Refurbishment of Laboratories. Ellis Horwood, Chichester, United Kingdom. Montesano, R., Bartsch, H., Boyland, E., Della Porta, G., Fishbein, L., Griesemer, R.A., Swan, A.B., and Tomatis, L. (eds.) 1979. Handling Chemical Carcinogens in the Laboratory, Problems of Safety. IARC Scientific Publications No. 33. International Agency for Research on Cancer, Lyon, France. National Research Council. 1995. Prudent Practices in the Laboratory: Handling and Disposal of Chemicals. National Academy Press, Washington, D.C. Occupational Health and Safety. 1993. National Safety Council, Chicago. Pal, S.B. (ed.) 1991. Handbook of Laboratory Health and Safety Measures, 2nd ed. Kluwer Academic Publishers, Hingham, Mass. Rosenlund, S.J. 1987. The Chemical Laboratory: Its Design and Operation: A Practical Guide for Planners of Industrial, Medical, or Educational Facilities. Noyes Publishers, Park Ridge, N.J. Young, J.A. (ed.) 1991. Improving Safety in the Chemical Laboratory: A Practical Guide, 2nd ed. John Wiley & Sons, New York.
INTERNET RESOURCES http://www.ilpi.com/msds/index.html Where to find MSDSs on the internet. Contains links to general sites, government and nonprofit sites, chemical manufacturers and suppliers, pesticides, and miscellaneous sites. http://www.OSHA.gov OSHA web site. http://www.osha-slc.gov/OshStd_data/1910_1450.html Text of OSHA Standard 29 CFR 1910.1450: Occupational Exposure to Hazardous Chemicals in Laboratories. http://www.osha-slc.gov/OshStd_data/1910_1000_TABLE_Z-1.html Tables of permissible exposure limits (PELs) for air contaminants. http://www.osha-slc.gov/OshStd_data/1910_1000_TABLE_Z-2.html Tables of PELs for toxic and hazardous substances. http://hazard.com/msds/index.php Main site for Vermont SIRI. One of the best general sites to start a search. Browse manufacturers alphabetically (for sheets not in the SIRI collection) or do a keyword search in the SIRI MSDS database. Lots of additional safety links and information. http://siri.uvm.edu/msds Alternate site for Vermont SIRI. http://tis.eh.doe.gov/docs/osh_tr/ch5.html DOE OSH technical reference chapter on personal protective equipment.
Contributed by George Lunn Baltimore, Maryland Gretchen Lawler Purdue University West Lafayette, Indiana
Safety Procedures and Quality Control
3.4.33 Current Protocols in Cytometry
Supplement 20
Method for Visualizing Aerosol Contamination in Flow Sorters Deflection-droplet cell sorters are an important part of scientific research because of their capability to separate different populations of cells based on size and granularity as well as on extrinsic fluorescent properties. In the operation of a cell sorter, a cell sample is hydrodynamically focused within a pressurized flowing stream of sheath fluid. Under normal circumstances, i.e., without external forces, the stream emerging from an orifice is hydrodynamically unstable, meaning that the droplet breakoff point is unpredictable and, therefore, not stable.
UNIT 3.5
BASIC PROTOCOL
Stabilization of the droplet breakoff in time and space is achieved by applying vibration to the stream with a steady frequency (by means of a piezoelectric crystal transducer), thereby fixing the droplet breakoff point. When the desired criteria for sorting are met, droplets are charged positively or negatively, and then passed through an electrostatic field. The highly positive or negative deflection plates attract negative or positive droplets, respectively, causing a separation in the stream. Aerosols are fine mists or sprays that can contain minute particles. During routine operation of a flow cytometer, microdroplets and droplets of varying sizes (3 to 7 µm, 40 to 200 µm, respectively) are formed when the pressurized sheath fluid leaves the flow-cell orifice, resulting in the generation of aerosols. In addition, splashing of side streams into receptacles, as well as center waste-stream splashing, can also form aerosols. Since the normal operation of cell sorters generates aerosols, the potential for aerosol creation is even greater in failure mode due to clogs or air in the fluidics. Usually, aerosols are contained within the collection chamber (Ferbas et al., 1995). However, this potential for aerosol generation should be a concern for sorter operators as well as for other individuals present in the facility who can potentially be exposed to contamination during sorting of biohazardous samples. Most commercially available sorters have some means of reducing the potential for exposure to biohazardous aerosols; however, none is designed to be 100% effective in eliminating aerosols. By using established precautionary guidelines and regularly testing the efficiency of an instrument’s capability to reduce aerosols, sorter operators can protect themselves as well as others around them. This unit provides a visual method to examine aerosol containment using a modified, commercially available product called Glo Germ. Glo Germ was developed for teaching aseptic techniques in hospitals, industry, restaurants, and schools and is visualized with UV or black light. Glo Germ is available in three forms: a white powder, an orange oil-based suspension of a melamine copolymer resin, or the orange melamine copolymer resin in dry form. The white powder is not used in this protocol because it fluoresces blue under black-light illumination, making it difficult to differentiate from ever-present lint. The orange oil-based suspension was successfully used prior to the availability of the dry form (Oberyszyn and Robertson, 2001). The newly available dry orange resin is described in this unit. The reported size of the Glo Germ particles (≤5 µm) is similar to that of yeast and bacteria, making it a comparable indicator. The Glo Germ suspension is run as a normal sample and sorted, thereby mimicking a biological sample sort. Placing slides in and around the sort collection area and examining the slides for orange fluorescent particles either by eye or under a fluorescent microscope allows for the detection of aerosols. Safety Procedures and Quality Control Contributed by Andrew S. Oberyszyn Current Protocols in Cytometry (2002) 3.5.1-3.5.7 Copyright © 2002 by John Wiley & Sons, Inc.
3.5.1 Supplement 19
The preparation of the suspension of resin is extremely straightforward, inexpensive, and quick. The visualization of orange fluorescent particles under a microscope is very easy due to their very intense fluorescence. Materials Glo Germ melamine copolymer resin (Glo Germ) 95% and 100% ethanol PBS without calcium or magnesium (PBS (−)) containing 10% filtered fetal bovine serum (FBS; APPENDIX 2A or Life Technologies) 100% acetone 15-ml conical tubes (Falcon) Deflected-droplet cell sorter to be tested Black light source (e.g., GE Bright Stik) Microscope slides Kimwipes (or any low-lint/lint-free paper) Fluorescent microscope (with DAPI or FITC exciter filter) 12 × 75–mm culture tubes (Falcon) Tubing (∼1.3-mm i.d.) Compressed air with thin extension tube Prepare Glo Germ sample suspension 1. Weigh out 0.5 g dry orange Glo Germ and place in a 15-ml conical tube containing 10 ml 95% or 100% ethanol. It is best to prepare the Glo Germ suspension as far away from the cell sorter as possible to prevent potential spreading of resin during preparation, preferably in another room. Without the ethanol suspension step, the Glo Germ will not suspend in the PBS (−)/10% FBS.
2. Vortex tube well to completely suspend resin. 3. Centrifuge 15-ml conical tube 5 min at 2000 × g, 25°C. 4. Remove supernatant and resuspend pellet in 10 ml PBS (−)/10% FBS. Vortex sample periodically to keep resin in suspension. The Glo Germ resin concentration is ∼5%. Higher resin concentration can cause clumping and a decline in sample flow rate during sorting. If clumping does occur, a final concentration of 1% can also be used successfully. The FBS is included to mimic a real sort sample.
Prepare cell sorter for sorting 5. Use standard manufacturer-suggested method for setting up the cell sorter (e.g., determining frequency, drive, and delay). Make sure the cell sorter is working optimally to be able to examine normal sort conditions.
6. With as many lights off as possible, examine the cell sorter area with a black light source. Wearing gloves, clean off any lint or fluorescent particles with 100% ethanol to make sure the area is “spotless.” Lint will glow light blue when excited by a black light source. It is important to eliminate as much nonspecific fluorescence as possible prior to starting the aerosolization evaluation. Methods for Visualizing Aerosol Contamination in Flow Sorters
7. Wearing gloves, meticulously clean several microscope slides by rinsing with 100% acetone. Air dry the slides and wipe clean with a Kimwipe. Rinse the slides with 100% ethanol and wipe dry with a Kimwipe.
3.5.2 Supplement 19
Current Protocols in Cytometry
E A
B D C
Figure 3.5.1 Examples of positions of slides for examination of aerosol extent for a Beckman Coulter Elite ESP; (A,B,C,D) Slides located outside sort chamber on workstation; (E,F) Slides located inside sort chamber.
Acetone essentially “melts” any Glo Germ resin that may be on the slide. Ethanol cleans the slides further to make sure nothing fluorescent is present on the slides.
8. Examine slides using a fluorescent microscope to make sure that no fluorescence is present. It is imperative that the slides do not contain any fluorescence. Lint and dust will fluoresce and can give false positive results.
9. Label the microscope slides and place in and around the sort collection area in various locations (Fig. 3.5.1). Locations should include: directly inside the collection chamber, directly outside the collection chamber door, directly below sample chamber, near the computer keyboard, and so on. Make a note of slide location to be able to identify region of aerosol deposition.
10. Place collection tubes into cell sorter sample holders inside the collection chamber. Collection tubes should mimic a normal sorter collection tube (i.e., medium or PBS, with FBS).
Perform sorting in regular mode 11. Using forward scatter (FS) versus side scatter (SS), set sort regions. Since Glo Germ particles do not show distinct fluorescent peaks, FS versus SS is sufficient for gate setting. The regions selected should include large and small particle sizes.
12. Sort the Glo Germ resin for 1 to 2 hr at rates ranging from the lowest to the highest regularly used on the instrument. Keep collection chamber closed during sorting except for replacement of collection tubes. If the sorter is equipped with a biohazard filter, make sure it is on. This protocol was developed on a Beckman-Coulter Elite fitted with the high-speed ESP upgrade. Rates examined were 1000 to 10,000 events/sec. For regular-mode sorting, it is best to follow normal sorting procedure. This will allow one to determine if precautions regularly used are sufficient.
Safety Procedures and Quality Control
3.5.3 Current Protocols in Cytometry
Supplement 19
13. Turn off as many lights as possible, visualize the area in and around the sort collection chamber with a black light source, and look for any orange fluorescence. If aerosolization is extensive, fluorescent regions or particles may be seen with the naked eye.
14. Carefully remove slides and proceed with microscopic visualization. Perform sorting in mock-failure mode 15. Repeat steps 6 to 10. If normal-mode sorting was performed prior to mock-failure mode, cleaning is imperative due to the fact that orange fluorescence will be seen near the waste drain as well as the location where the center stream falls inside the collection chamber.
16. Connect tubing to the thin extension tube from the compressed air and attach behind and below the deflection plate, making sure to point the end of the tubing toward the center stream (Fig. 3.5.2). 17. Sort the Glo Germ resin as described in step 12. Keep collection chamber open during sorting. If the sorter is equipped with a biohazard filter, make sure it is off. Since this is a failure-mode run, the principle is to try to achieve a “worst-case” scenario. To examine actual containment of aerosols, a “best-case” scenario should also be tested where all safety measures available are in place (i.e., biohazard filter on, collection chamber door closed, etc.) with the addition of the failure mode. What would be expected in a situation like this is that aerosols should be contained within the sorting collection chamber. This will determine if the employed safety measures are sufficient.
18. During sorting, spray a 10- to 30-sec burst of compressed air, deflecting the center stream to mimic a clogged flow cell.
center stream
compressed air
deflection plate
sort streams tubing
Methods for Visualizing Aerosol Contamination in Flow Sorters
“mock” clog
Figure 3.5.2 Tubing attached to compressed air was positioned below the deflection plates, behind the center waste stream. A 10-sec burst was used to simulate a failure mode.
3.5.4 Supplement 19
Current Protocols in Cytometry
Alternatively, with the deflection voltage on, turn off the drive, which gives extensive fanning of the sorting streams.
19. Turn off as many lights as possible, visualize the area in and around the sort collection chamber with a black light source, and look for any orange fluorescence. 20. Carefully remove slides and proceed with microscopic visualization. Visualize aerosolization extent by fluorescent microscopy 21. Place slides on the stage of a fluorescent microscope. Examine areas of the slides looking for any orange fluorescence. A DAPI-exciter filter (405 nm) gives the brightest orange fluorescence; however a FITCexciter filter (490 nm) gives a very similar intensity. A Texas Red–exciter filter (570 nm) has also been used, although the orange fluorescence is noticeably less intense. In many instances, particles can be visualized with bright-field illumination; however, very small particles are more easily visualized by their intense orange fluorescence.
COMMENTARY Background Information A major advantage of sorting flow cytometers is the ability to separate cell populations based on cell size, cell density, or fluorescence. With increased analysis and separation of viable biological specimens comes an increase in the hazard of sorter operator exposure to unknown as well as known pathogens, such as hepatitis and HIV. Additionally, hazards can include commonly used staining reagents that can be toxic if inhaled. The normal working of the cell sorter forms droplets generated by pressurized fluid coming out of a vibrating nozzle. If desired sorting criteria are met, droplets containing the desired cells are electrostatically charged positively and/or negatively and are deflected into receptacles. Due to sample pressurization and electrostatic charges, aerosols of small droplets and microdroplets (size range: 40 to 200 µm and 3 to 7 µm, respectively) are formed, causing potential hazard to the operator. Secondary aerosols can also be formed when the sorting streams splash into receptacles (Merril, 1981; Bakker, 1992). Failure modes of the cell sorter such as a clogged tip, air in the fluidics system, or other instrumental malfunction can cause increased production of aerosols. The potential safety of the sorter operator, as well as of individuals in the sorter area, is further compromised due to the fact that aerosols have been shown to be of importance in the spread of infectious diseases (Almeida et al., 1971; Sattar and Ijaz, 1987; Schoenbaum et al., 1990; Ijaz et al., 1994). As a result, the International Society for Analytical Cytology (ISAC) Biohazard Working Group (Schmid et al., 1997) has developed guidelines as precautions for cell sorter operators. It is important to
remember that sorters used for analysis and not sorting of biohazardous samples still generate aerosols and should also be tested. The generally large size of cell sorters makes it impossible for them to be placed inside a biological hood. It is also not practical to create a “room-sized” biological hood to house them. Most current models of commercially available flow cytometers have designs that reduce the formation of aerosols and their subsequent release outside the sorting collection chamber (Ferbas et al., 1995). Custom-designed modifications to contain aerosols have been developed (Bakker, 1992; Giorgi, 1994). Methods for assessing the efficiency of aerosol containment during the process of sorting biohazardous samples have been developed as well (Anderson, 1958; Merril, 1981; UNIT 3.3). The manufacturer of the Glo Germ resin reports that the particle sizes are ≤5 µm. However, examining the mixture of Glo Germ resin and Flow-Check beads (10-µm nominal diameter; Beckman Coulter) in Figure 3.5.3 reveals that this is not completely correct. Regardless of this discrepancy, particles as small as ∼1 µm can be easily visualized under 20× magnification. Current methods for testing of aerosol containment during cell sorting include monitoring gravitational deposition of droplets (Merril, 1981; UNIT 3.3), and the use of air samplers (Anderson, 1958; UNIT 3.3). There are several advantages to using Glo Germ. Since there is no need for knowledge of microbiology, there is less potential for error compared to titration of T4 bacteriophage and agar plate handling (UNIT 3.3). Preparation takes <30 min, does not need to be done in advance, and requires mini-
Safety Procedures and Quality Control
3.5.5 Current Protocols in Cytometry
Supplement 19
small Glo Germ particles
Flow-Check beads 10 µm
Figure 3.5.3 Comparison of Glo Germ size to Flow-Check beads under 20× magnification with DAPI-exciter filter illumination.
mal reagents. Glo Germ is inert, therefore not biohazardous, and will not cause potential contamination of future sorts. One of the major advantages is that results are immediate as opposed to the need for overnight incubation of E. coli plates. In conjunction with an image analysis system, it can potentially be quantitative as well.
resin will last for many tests of sorter aerosolization. The resin suspension can be prepared in <30 min, allowing sufficient time to perform test of varying times and parameters. Analysis of slides microscopically can potentially take the longest amount of time, depending on how many slides are used around the sorter area.
Literature Cited Critical Parameters It is of the utmost importance that the microscope slides and sort area be immaculately clean to prevent any false positive results. It is critical to examine the sample stage area since removing a sample tube under pressure causes a considerable amount of aerosol formation.
Anticipated Results Under ideal conditions, no fluorescent orange particle should be detected outside the sorting collection chamber. As would be expected, orange fluorescence will be visible around the waste drain as well as the location where the center waste stream hits the bottom of the collection chamber. In failure mode, aerosols may be seen more extensively within the sort collection area but should not be present outside the area.
Time Considerations Methods for Visualizing Aerosol Contamination in Flow Sorters
Because the preparation of the resin suspension is very simple and quick, this test can be performed at any time with little advance planning. A 4-oz. (∼113 g) sample of Glo Germ
Almeida, J.D., Kulatilake, A.E., Makay, D.H., Shackman, R., Chisholm, G.D., MacGregor, A.B., O’Donoghue, E.P., and Waterson, A.P. 1971. Possible airborne spread of serum-hepatitis virus within a hemodialysis unit. Lancet 2:849-850. Anderson, A.A. 1958. New sampler for the collection, sizing and enumeration of viable airborne particles. J. Bacteriol. 76:471-484. Bakker, A.M. 1992. Evaluation of a biological containment system for a fluorescent activated cell sorter. M.S. Thesis. University of San Francisco. Ferbas, J., Chadwick, K.R., Logar, A., Patterson, A.E., Gilpin, R.W., and Margolick, J.B. 1995. Assessment of aerosol containment on the ELITE Flow Cytometer. Cytometry 22:45-47. Giorgi, J.V. 1994. Cell sorting of biohazardous specimens for assay of immune function. Methods Cell Biol. 42:359-369. Ijaz, M.K., Sattar, S.A., Alkarmi, T., Dar, F.K., Bhatti, A.R., and Elhag, K.M. 1994. Studies on the survival of aerosolized bovine rotavirus (UK) and a murine rotavirus. Comp. Immunol. Microbiol. Inf. Dis. 17:91-98. Merril, J.T. 1981. Evaluation of selected aerosolcontrol measures on flow cytometers. Cytometry 1:342-345.
3.5.6 Supplement 19
Current Protocols in Cytometry
Oberyszyn, A.S. and Robertson, F.M. 2001. Novel rapid method for visualization of extent and location of aerosol contamination during highspeed sorting of potentially biohazardous samples. Cytometry 43:217-222. Sattar, S.A. and Ijaz, M.K. 1987. Spread of viral infections by aerosols. CRC Crit. Rev. Environ. Control 17:89-131. Schmid, I., Nicholson, J.K., Giorgi, J.V., Janossy, G., Kunkl, A., Lopez, P.A., Perfetto, S., Seamer, L.C., and Dean, P.N. 1997. Biosafety guidelines for sorting of unfixed cells. Cytometry 28:99117.
Schoenbaum, M.A., Zimmermann, J.J., Beran, G.W., and Murphy, D.P. 1990. Survival of pseudorabies virus in aerosol. Am. J. Vet. Res. 51:331-333.
Contributed by Andrew S. Oberyszyn The Ohio State University Comprehensive Cancer Center Analytical Cytometry Shared Resource Laboratory Ohio State University Columbus, Ohio
Safety Procedures and Quality Control
3.5.7 Current Protocols in Cytometry
Supplement 19
CHAPTER 4 Molecular and Cellular Probes INTRODUCTION
T
he concept of cellular identification first arose in the time of Ehrlich and Metchnikoff, who pioneered diagnostic process in infectious disease (as discussed in Weissman, 1992). Over a century later, histological diagnosis of disease by pathology is the gold standard to which all other methods are compared. The development of absorptive dyes to bring out the special features of different cells is the accepted tool of pathologists for this purpose. It was only natural to use the light microscope to distinguish between healthy and unhealthy cells. In the early 1940s antibodies were introduced as a means for more precise identification (Coons et al., 1941; Moller, 1961). Cytochemical reagents (initially lightabsorbing dyes) were attached to antibodies; when the antibodies bound to cells, a color reaction was produced at the site of antibody deposition. Later other probes, such as lectins to surface carbohydrates, were developed as alternatives to antibody probes. In the early 1960s, fluorescent dyes began to be substituted for light-absorbing dyes, although primarily only for research specimens. Even today, immunofluorescence is hardly used by anatomical pathologists. With the association of computers to flow cytometers and improvements in immunofluorescence detection and instrumentation, however, flow cytometry using fluorochrome-conjugated antibodies has emerged as a major approach to automated cellular identification. The success of this methodology has been driven by its ability to examine large numbers of cells, measure fluorescence quantitatively, and combine several cellular probes in a single specimen. New applications are being developed that will not only reveal a cell’s identity, but define its function at the same time. The probes now available (e.g., from Molecular Probes) consist not only of antibodies, but also of substrates within the cell itself: molecular probes, lectins, and ion-sensitive substrates.
Information about the nature of conjugated probes, with particular reference to antibodies, is presented in UNIT 4.1. The unit discusses techniques for ascertaining the optimal titer for individual, dual, and multiple antibodies used for simultaneous phenotyping, and also stresses the importance of quality control in making batches of antibody for routine use. Methods used to conjugate antibodies to ensure optimal fluorochrome/protein ratios are discussed in detail in UNIT 4.2. UNIT 4.3 describes the physicochemical structures of a number of commonly used nucleic acid probes and provides an excellent background for other protocols in this manual that employ these probes. This unit should be considered preliminary reading for any studies involving nucleic acid probes. UNIT 4.4 discusses molecular probes in general, providing considerable background detail for a number of probes found in the protocols of Chapter 9. For example, probes for hydrogen peroxide determination (UNIT 9.7), calcium indicators (UNIT 9.8), pH indicators (UNIT 9.3), membrane potential probes (UNIT 9.6), and green fluorescence protein (UNIT 9.5) are all covered by this unit. The material in UNIT 4.4 provides an excellent basis for understanding the chemical and physiological interactions of these fluorescent indicators.
Molecular and Cellular Probes Contributed by J. Paul Robinson Current Protocols in Cytometry (2000) 4.0.1-4.0.2 Copyright © 2000 by John Wiley & Sons, Inc.
4.0.1 Supplement 14
UNIT 4.5 takes the nature of fluorescent probes one step further and provides a quantitative approach to analysis of DNA and RNA probes using spectroscopic techniques. The study of cellular fluorescence associated with many fluorescent probes is complemented by a better quantitative understanding of the relationship between the probe and the relevant organelle, in this case nuclear material.
LITERATURE CITED Coons, A.H., Creech, H.J., and Jones, R.N. 1941. Immunological properties of an antibody containing a fluorescent group. Proc. Soc. Exp. Biol. Med. 47:200-210. Moller, G. 1961. Demonstration of mouse isoantigens at the cellular level by the fluorescent antibody technique. J. Exp. Med. 114:415-432. Weissmann, G. 1992. Inflammation: Historical perspective. In Inflammation: Basic Principle and Clinical Correlates (J.I. Gallin, I.M. Goldstein, and R. Snyderman, eds.) pp. 5-9. Raven Press, New York.
J. Paul Robinson
Introduction
4.0.2 Supplement 14
Current Protocols in Cytometry
Titering Antibodies
UNIT 4.1
The antibodies used for immunophenotyping are primarily monoclonal, derived from hybridoma fusions of mouse cells immunized against the appropriate antigen. They are almost always of the IgG class, most frequently IgG1, IgG2a, and IgG2b. IgM is a less commonly used class of monoclonal antibody; because it is approximately five times the molecular weight of IgG, it can be more troublesome. Polyclonal antibodies derived from sera of immunized animals are less frequently used as a primary antibody directed against the desired antigen, but are almost always used as the second fluorochrome-conjugated antibody. The F(ab′)2 fragment should always be used, to minimize Fc receptor binding of polyclonal antibodies. F(ab′)2 fragments of monoclonal antibodies are not readily available because rodent-derived F(ab′)2 fragments are difficult to prepare in high yield. This unit will define the strategy for titering antibodies to give the highest discrimination of positive cells from negative cells. Once titered, antibodies may be combined for providing simultaneous evaluation of multiple antigens expressed by cells. TITERING DIRECTLY CONJUGATED ANTIBODIES TO EXTRACELLULAR ANTIGENS
BASIC PROTOCOL 1
This procedure is designed to determine the appropriate concentration of antibody for use in immunophenotyping. It is based on the concept that the best concentration is that which produces the best discrimination between the positive and negative cells. Accordingly, the most important measurement is the fluorescence intensity of staining (signal) vs. nonspecific binding (noise), the signal-to-noise ratio. Materials Stock solution of specific, fluorochrome-conjugated antibody to be titered Target cells for antibody to be titered PBS without calcium or magnesium (e.g., APPENDIX 2A or Life Technologies) Normal mouse IgG (e.g., Caltag) 2% ultrapure formaldehyde in PBS (prepared from 10% EM-grade solution; Polysciences) Ammonium chloride lysing solution (APPENDIX 2A), prepared fresh Isotype control myeloma proteins (fluorochrome conjugated) 12 × 75–mm polypropylene test tubes (Falcon) Sorvall centrifuge and H1000B rotor, or equivalent Flow cytometer Titer antibodies 1. Determine the concentration of specific antibody conjugate in the stock solution and centrifuge 10 min at 15,000 × g, 4°C. Leave aggregated antibody in pellet. 2. Prepare 30 µl containing 9 µg antibody (300 µg/ml) in PBS. Prepare six 1/3 serial dilutions (10 µl to 30 µl) in PBS. 3. Prepare a target cell suspension containing 5–10 × 106 cells/ml and 200 µg normal mouse IgG/ml. 4. Add 10 µl of each antibody dilution to 50-µl aliquots of cell suspension in separate 12 × 75–mm polyporpylene test tubes. Also prepare two control tubes: an isotope control containing only normal mouse IgG and an autofluorescence control tube containing only cell suspension. Incubate 15 min at 4°C. Contributed by Carleton C. Stewart and Sigrid J. Stewart Current Protocols in Cytometry (1997) 4.1.1-4.1.13 Copyright © 1997 by John Wiley & Sons, Inc.
Molecular and Cellular Probes
4.1.1
5. Add 3 ml freshly prepared ammonium chloride lysing solution. Centrifuge 3 min at 1500 × g, 4°C. 6. Remove supernatant and resuspend cells in residual solution. 7. Add 200 µl of 2% ultrapure formaldehyde in PBS. 8. Acquire 2000 target cells using a flow cytometer. Perform data analysis 9. Display a histogram for each dilution, as shown in Figure 4.1.1. 10. Adjust markers using auto sample or isotype control sample, so that <1% of events are above the marker. 11. Determine the mean channel linear fluorescence intensity (MCF) of both positive (signal) and negative (noise) cells for the six tubes. 12. Compute the signal-to-noise ratio by dividing the MCF value for positive cells by that for the negative cells. Plot these values as a function of antibody dilution as shown in Figure 4.1.2.
Cell number
The optimal titer is the one that generates the highest ratio, because this provides the greatest discrimination between positive and negative cells, regardless of the absolute value of fluorescence intensity.
100
Titering Antibodies
101
3 µg S/N = 2.5
1µg S/N = 2.1
0.3µg S/N = 2.4
0.1µg S/N = 4.1
0.03 µg S/N = 4.8
0.01µg S/N = 4.6
0.003 µg S/N = 3.5
0.001µg S/N = 3.2
auto
102
103 104 100 101 102 103 104 100 Fluorescence intensity
101
102
103
104
Figure 4.1.1 Histograms showing determination of antibody titer through serial dilutions. S/N, signal-to-noise ratio; auto, autofluorescence control. Microgram values shown in the figure are the amounts used to stain 106 cells in a total volume of 100 µl.
4.1.2 Current Protocols in Cytometry
5
Signal–to–noise ratio
titer
4
3
2 0
1
2
3
4
5
6
7
8
Serial dilution number
Figure 4.1.2 Choice of working titer at maximum signal-to-noise ratio.
TITERING BIOTINYLATED OR HAPTEN-CONJUGATED ANTIBODIES In contrast to direct labeling of epitopes by conjugated antibodies, a two-step procedure may be used to bind and label an antigen. The following protocol can be used when biotinor hapten-conjugated antibody is used in a first step, and a secondary fluorescent antibody is directed against it.
ALTERNATE PROTOCOL 1
Additional Materials (also see Basic Protocol 1) Biotinylated or hapten-conjugated primary antibody Labeled strepavidin, or anti-hapten, secondary antibody with conjugated fluorochrome Proceed as for titering directly conjugated antibodies (see Basic Protocol 1) with the following modifications at step 6: 6a. Remove the supernatant and resuspend cells in residual solution. Add 10 µl of appropriately titered streptavidin or anti-hapten antibody with conjugated fluorochrome and incubate 15 min. Add 3 ml PBS and centrifuge tubes 3 min at 1500 × g, 4°C. Repeat supernatant removal and resuspension. TITERING OF INDIRECT ANTIBODY TO EXTRACELLULAR ANTIGEN Often antibodies are not conjugated, such as those derived from hybridoma culture supernatants, ascitic fluid, or antisera from immunized animals. This requires the use of a second fluorochrome-conjugated antibody directed against the primary antibody. The titer of the secondary antibody must be well established (see Basic Protocol 1). Normal IgG is used to block sites that may give rise to nonspecific binding. Additional Materials (also see Basic Protocol 1) Antibodies or antisera against extracellular antigen Normal goat IgG against primary antibody species, fluorochrome conjugated (Caltag)
ALTERNATE PROTOCOL 2
Molecular and Cellular Probes
4.1.3 Current Protocols in Cytometry
Proceed as for titering directly conjugated antibodies (see Basic Protocol 1) with the following modifications at steps 3 to 6: 3b. Prepare a target cell suspension containing 5–10 × 106 cells/ml and 200 µg normal IgG of the second antibody species (e.g., goat). IMPORTANT NOTE: Do not use normal mouse IgG.
4b. Add 10 µl of each antibody dilution to 50 µl of cell suspension in separate test tubes. Also prepare an isotype control tube and a tube containing only cell suspension. Incubate 15 min at 4°C. 5b. Add 3 ml lysing solution. Centrifuge tubes 3 min at 1500 × g, 4°C. Remove supernatant and resuspend cells in residual solution. Add 10 µl of appropriately titered fluorochrome-conjugated second antibody and incubate 15 min at 4°C. 6b. Add 3 ml PBS and centrifuge tubes 3 min at 1500 × g, 4°C. Remove supernatant and resuspend cells in residual solution. BASIC PROTOCOL 2
TITERING DIRECTLY CONJUGATED ANTIBODIES TO INTRACELLULAR ANTIGENS When antigens are intracellular, they must be made available for binding by permeabilizing the cell membrane so that antibodies can penetrate. There is no universal method for fixing and permeabilizing all types of cells, so some adjustments may be required to the recommended starting procedure. There are several excellent commercially available fixatives and permeabilizing solutions for this purpose. Another serious consideration is that enzymes, such as trypsin, that are often used to detach cultured cells may also strip the desired antigen epitopes. To further complicate matters, fixation may damage or destroy the epitope. Alternatively, a freeze-thaw can be performed by rapidly freezing cells in an ethanol/dry ice bath and thawing them at 37°C for three to five cycles. This procedure has the advantage that epitopes are not distorted, but has the drawback of significant cell loss. Additional Materials (also see Basic Protocol 1) Target cells that do and do not express the desired epitope Prepare titered antibody and control 1. Determine the concentration of antibody in the stock solution and centrifuge 10 min at 15,000 × g, 4°C. Leave aggregated antibody in pellet. 2. Determine the concentration of an isotype control immunoglobulin and centrifuge at 15,000 × g, 4°C. Leave aggregated immunoglobulin in pellet. 3. Prepare 30 µl of antibody and 30 µl of isotype control each containing 9 µg (300 µg/ml) in PBS. Prepare six 1/3 serial dilutions (10 µl to 30 µl). Incubate fixed permeabilized target cells with antibody 4. Prepare fixed permeabilized target cells that express the desired epitope and fixed permeabilized target cells that do not express the desired epitope. Both cell suspensions should be at 5–10 × 106 cells/ml. Cells can be permeabilized with commercially available reagents or the freeze-thaw method. If erythrocytes are present they should be lysed with ammonium chloride lysing solution (APPENDIX 2A) before fixation.
Titering Antibodies
4.1.4 Current Protocols in Cytometry
5. Add 10 µl of each antibody dilution and 10 µl of each isotype control dilution to 50 µl of each target cell in the appropriate tubes. Also prepare tubes containing each cell type alone. 6. Incubate cells 45 min at 4°C and then add 3 ml PBS. Incubate an additional 45 min, centrifuge 3 min at 1500 × g, 4°C, and discard the supernatant. Resuspend cells in residual solution. 7. Acquire 2000 target cells using a flow cytometer.
isotype control
antibody
Cell number
ligand, 1 µg MCF Ab 278 IC 5.8
isotype control
antibody
Cell number
ligand, 1 µg MCF Ab 100 IC 3.6
isotype control
antibody
Cell number
ligand, 1 µg MCF Ab 25.7 IC 2.6
100
101
102
103
104
Cytokeratin
Figure 4.1.3 Determining titer of antibody to intracellular antigens by serial dilution. For this antibody against cytokeratin, histograms of the mean channel fluorescence (MCF) of the third (top) through fifth (bottom) serial dilution of specific antibody are shown in comparison with isotype controls.
Molecular and Cellular Probes
4.1.5 Current Protocols in Cytometry
60 titer
Signal–to – noise ratio
50
40
30
20
10
0 0
1
2
3
4
5
6
Serial dilution number
Figure 4.1.4 Titer at maximum signal-to-noise ratio for an antibody against intracellular antigens. Dilutions 1 and 2 are not shown in Fig. 4.1.3. S/N is calculated by the MCF values determined by comparing antibody binding (signal) against epitope-positive and -negative cells, in proportion to the binding of nonspecific isotype control MCF (noise). The antibody concentration at which S/N is maximal is the appropriate working titer.
Perform data analysis 8. Display histogram for each dilution as shown in Figure 4.1.3 for antibody and isotype control for both positive and negative cells. 9. Determine the mean channel linear fluorescence intensity (MCF) of each. 10. Determine the S/N ratio of the specific antibody to isotype control and plot these values as a function of dilution, as shown in Figure 4.1.4. ALTERNATE PROTOCOL 3
TITERING BIOTINYLATED, HAPTEN-CONJUGATED, OR SECONDARY ANTIBODIES TO INTRACELLULAR ANTIGENS Additional Materials (also see Basic Protocol 2 and Alternate Protocol 1) Avidin or anti-hapten antibody, or second antibody with conjugated fluorochrome Proceed as for titering directly conjugated antibodies to intracellular antigens (see Basic Protocol 2) with the following modifications at step 6: 6a. Add 10 µl of appropriately titered avidin or anti-hapten antibody, or second antibody with conjugated fluorochrome, and incubate cells 45 min. Add 3 ml PBS and incubate an additional 45 min. Centrifuge cells 3 min at 1500 × g, 4°C, discard the supernatant, and resuspend cells in residual solution.
Titering Antibodies
4.1.6 Current Protocols in Cytometry
EPITOPE TITERING The previous procedures (see Basic Protocols 1 and 2 and Alternate Protocols 1 and 2) determine the antibody titer at a constant epitope concentration. The working titer of any antibody is also dependent upon the concentration of epitopes. The optimal antibody concentration required to saturate cells having 105 epitopes is different from that required to saturate cells having 104 epitopes, and both cell types can be present in the same suspension. To determine this titer, the antibody concentration is held constant and the epitope concentration is varied by increasing the cell concentration.
BASIC PROTOCOL 3
1. Adjust target cell concentration to 128 × 109/ml and prepare four serial 1/4 dilutions to produce 64 (neat), 16, 4, and 1 × 106 cells in 50 µl of PBS. If blood or bone marrow leukocytes are used as target cells, add 5 ml of cell solution to a 50-ml centrifuge tube containing 45 ml ammonium chloride lysing solution (APPENDIX 2A). Centrifuge 5 min at 1500 × g, 4°C. Remove the supernatant and resuspend cells in residual solution, adjusting the concentration of the suspension before making serial dilutions.
2. Mix 50 µl of cell suspension with 10 µl of antibody whose titer was determined using 0.5–1 × 106 cells. Incubate 15 min at 4°C. 3. Add 3 ml PBS. Centrifuge tubes 3 min at 1500 × g, 4°C. 4. Remove supernatant and resuspend cells in residual solution. 5. Add the amount of 2% ultrapure formaldehyde in PBS needed to give a final concentration of 106 cells/ml. 6. Acquire 2000 target cells using a flow cytometer. 7. Display histogram for each dilution. 8. Adjust markers using auto sample (or isotype control sample) so that <1% of events are above the marker. 9. Determine the mean channel linear fluorescence intensity (MCF) of both positive (signal) and negative (noise) cells for the four tubes. VERIFYING PERFORMANCE OF ANTIBODY COMBINATIONS When antibodies are purchased as separate reagents with the intent to combine them for clinical immunophenotyping, it is necessary for the clinical laboratory to verify their performance. Each combination must prove to stain properly and each subsequent batch must provide the same performance. The laboratory must also show the intended use of the combination is efficacious. The authors recommend using only antibodies with directly conjugated fluorochromes, as they perform more reliably. A simple test will determine if antibodies perform properly when combined.
BASIC PROTOCOL 4
Additional Materials (also see Basic Protocol 1) Specific antibodies with appropriate fluorochromes Isotype control myeloma proteins with appropriate fluorochromes 1. Using each antibody at its proper titer, add each one to separate test tubes. To the last tube add all antibodies in combination. 2. To each tube add 50 µl of target cell suspension at 5–10 × 106/ml, containing 200 µg IgG/ml. 3. Incubate cells 15 min (45 min for intracellular antigens) at 4°C. Lyse erythrocytes and fix cells (see Basic Protocol 1, steps 5 to 7).
Molecular and Cellular Probes
4.1.7 Current Protocols in Cytometry
4. Acquire 2000 target cells using a flow cytometer. 5. Display univariate histograms of each file of cells stained with only one antibody, as shown in Figure 4.1.5. 6. Overlay the univariate histograms for each antibody from the file derived for the antibody combination. Each antibody must yield the same histogram whether used singly or in combination. If they are not identical, there is a problem with the combination or with the instrument setup. When producing an antibody combination, ensure that the histograms from cells stained with each antibody separately overlap the histogram from cells stained with the combination, as shown in Figure 4.1.5. BASIC PROTOCOL 5
BATCH PRODUCTION OF ANTIBODY COMBINATIONS Once the antibody combination has been validated, batches of the desired number of tests can be created—e.g., 100-test batches. Batches provide consistency, reduce errors, and promote staining efficiency, avoiding the variability created by adding antibodies individually to cells at the time of staining. Because the volume of each antibody solution dilutes the others, the concentration of each antibody before mixing should be three to four times its appropriate titer, so they can be mixed together to give a 1× titer. Additional Materials (also see Basic Protocols 1 and 4) Previous antibody combination batch Antibody combination batch Isotype control combination batch 1. To one test tube add the appropriate amount of the antibody combination batch to 50 µl of target cells at 5–10 × 106/ml, containing 200 µg IgG/ml. 2. To a second test tube add the appropriate amount of a previous antibody combination batch to 50 µl of target cell suspension at 5–10 × 106/ml, containing 200 µg IgG/ml. 3. Incubate cells 15 min (45 min for intracellular antigens) at 4°C. Lyse erythrocytes and fix cells (see Basic Protocol 1, steps 5 to 7). 4. Acquire 2000 target cells using a flow cytometer. Data analysis 5. Display bivariate histograms of each appropriate fluorochrome in the combination and load file as shown in Figure 4.1.6, top row. 6. Create a template consisting of regions around discrete target cell populations. Compute the mean channel linear fluorescence intensity (MCF) for the two antibodies and record. This will be the batch standard for future comparisons. 7. Load newly prepared batch and compute the MCF values for each antibody, as shown in Figure 4.1.6, bottom row. The newly prepared batch should provide exactly the same pattern as the old batch when the bivariate views of each fluorochrome is displayed. Similarity of the patterns is quantified by comparing the MCF for selected target cell populations. For a resolution of 128, the authors accept a ±5-channel variation in the MCF for each antibody. However, if a newly acquired antibody from a supplier exhibits a difference greater than ±5 channels, use the new MCF value for the antibody. A variation in MCF greater than ±5 channels for a different lot of antibody from the same supplier is not uncommon.
Titering Antibodies
4.1.8 Current Protocols in Cytometry
Cell number
101
102 103 FITC-CD3
104
100
101
102 PE-CD4
103
104
100
101
102 TC-CD8
103
104
Cell number
Cell number
100
Figure 4.1.5 Verification of antibody combinations. The dark histograms represents the first batch produced, while the light histogram is the second. A perfect overlap indicates an acceptable second batch. If overlap does not occur, the reason must be determined. Molecular and Cellular Probes
4.1.9 Current Protocols in Cytometry
104 R1 PE-CD4
103
x = 69 y = 94
x = 98 R2
102 101 100 104 R1
PE-CD4
103
x = 97 pass R2
102
x = 62 fail y= 96 pass
101 100 100
101 102 103 FITC-CD3
104 100
101
102 103 TC-CD8
104
Figure 4.1.6 Batch verification of antibody combinations. Top row, control batch or historic control file; bottom row, new batch. While the cluster for CD3+ CD4+ is in the R1 region for the new batch, the mean channel fluorescence for CD3 is only 62. This exceeds a difference of 69 ±5 channels, so the batch fails.
COMMENTARY Background Information
Titering Antibodies
One of the most important issues in immunophenotyping is using the correct amount of antibody. Most users depend on the manufacturer’s recommendation of appropriate amount. Unfortunately, this provides a false sense of security because standards for titration vary. Some suppliers offer antibodies at higher concentration than stated, whereas others may sell their antibodies below titer. One company offers an antibody at two separate concentrations. Another problem arises when an inappropriate target cell (cell lines, fixed, or frozen cells) is used to titer the antibody, because the target cell may not be representative of those that will be tested. It is important to distinguish between specific and nonspecific binding. Some monoclonal antibodies exhibit more nonspecific binding than others. This can easily be determined using the signal-to-noise (S/N) method of titering. When a “clean” antibody is added at high concentration, there is little change in the position of negative cells, and positive cells can be explicitly resolved. Unfortunately, most monoclonal antibodies bind nonspecifically to a lesser or greater extent; the ideal condition is rarely achieved, and the amount of nonspecific
binding increases with antibody concentration. Clean monoclonal antibodies exhibit only epitope-positive cells above the marker over a wide range of concentrations, whereas others may only exhibit a single optimal concentration. Most monoclonal antibodies fall between these two extremes. Because nonspecific binding is of much lower affinity than epitope binding, it decreases more rapidly than specific binding as antibody concentration is decreased. To determine the optimal amount of antibody to use it is necessary to determine the amount needed to maximize epitope binding, but not such an excess that nonspecific binding becomes significant. This is achieved by determining the signal-to-noise ratio for several antibody dilutions in PBS. A major reason for the heterogeneity in nonspecific binding among monoclonal antibodies is their structural diversity. Monoclonal antibodies are the product of hybridomas that are created by fusing normal with malignant B cells. During the screening process, the primary selection criterion is the quantity of antibody capable of binding to its epitope, and not antibody quality. Indeed, monoclonal antibodies of the same isotype and subclass are not only
4.1.10 Current Protocols in Cytometry
structurally different from one another, but they are also different from normal Ig. In order to properly titer antibodies, it is necessary to have both the highest signal (greatest number of true positive cells) and lowest noise (least number of false positive cells). Fc binding can be blocked using normal mouse IgG (200 µg/ml) when biotinylated or directly conjugated antibodies are used. If a second fluorochrome-conjugated antibody is used, the blocking IgG must be from the same animal species as that of the conjugated antibody. Antibodies useful for immunohistochemistry may prove inadequate for flow cytometry. Specific staining represented by bright reaction product can be distinguished microscopically from a dimmer general background. Because the flow cytometer measures total fluorescence, the cumulative difference, rather than the spatial difference, may be so small that no resolution of positive cells occurs. Sometimes it is not possible to find an antibody concentration for which the signal-tonoise ratio is >3. The fluorochrome may be inappropriate, because the epitope density on the target cell is too low for good resolution of positives. The problem can be corrected by using a brighter fluorochrome, such as phycoerythrin (also see UNIT 4.2). Alternatively, a second reagent might be used to improve fluorescence, but this approach can also increase noise. If these approaches fail to improve the S/N, the antibody itself may be of poor quality and an alternative source may be needed. The expression of any particular epitope on a population of cells can vary by an order of magnitude. For example, the epitope number for CD19 molecules on B cells is an order of magnitude less than the epitope frequency for CD20. Similarly, the frequency of CD4 epitopes on T cells is about five times greater than on monocytes. The epitope concentration, therefore, is equal to the average number of molecules per cell times the number of cells divided by the volume. It is important to know the epitope titer when scaling up the staining of target cells. For example, if the antibody titer is known for 106 cells/ml, 100 times this amount may not be required to stain 108 cells/ml. Because of the cost of antibodies, it is important not to use more than necessary.
Critical Parameters and Troubleshooting Erythrocyte lysing is critical because unlysed red cells in the gated region produce an erroneous denominator. Although nearly every
supplier of antibodies to human antigens offers a lysing solution of some kind, the authors recommend the ammonium chloride solution described here (for which a recipe is provided in APPENDIX 2A) because of its universal applicability and the lack of resulting debris in the suspension. Testing of most of the other available lysing solutions has shown that they produce variable lysis, depending on the source of the specimen, and variable amounts of debris. Furthermore, all of them are considerably more expensive than ammonium chloride. Aggregated immunoglobulin may increase nonspecific binding. Spontaneous aggregation of immunoglobulins may be aggravated by freezing and thawing or by lyophilizing. To remove aggregates, all newly acquired antibodies should be centrifuged at high speed in a microcentrifuge. The aggregates can be left in the bottom, or the supernatant antibody solution transferred to a new tube. IgG at high concentration is used to block Fc receptor binding and to some extent nonspecific binding. Purified mouse IgG can be used in all situations except when it is also the primary antibody to be titered. In this instance the blocking IgG must be from the same species as the source of second conjugated antibody. This will prevent the second antibody from binding to the nonspecific blocking IgG. For intracellular antigens it is important to use both epitope-positive and -negative cell lines. Doing so confirms the antibody has bound to its specific epitope. The protein concentration inside cells is orders of magnitude higher than membrane protein, thereby exacerbating nonspecific binding (NSB). Because it is desirable to resolve specific binding from NSB, it is necessary to systematically evaluate both. The isotype control Ig provides an estimate of the NSB component, whereas specific antibody binding is composed of both specific and NSB components. As both ligands are diluted, for acceptable antibodies the mean channel fluorescence of NSB falls faster than that of the specific component. The ratio between the MCF of the antibody to isotype must be >3 to be acceptable. When concentrations of antibody >3.0 µg/ml are required, the antibody should not be used, because this indicates that either the affinity or the specific activity is too low. The best antibodies usually stain at concentrations of 0.01 to 1.0 µg/ml. This is equivalent to 10 to 1000 ng per test. The concentration of antibody at titer should be calculated and expressed in ng/µl or µg/ml. An historical record of titers for
Molecular and Cellular Probes
4.1.11 Current Protocols in Cytometry
Titering Antibodies
antibodies can provide information on important formulation changes that may not be announced by the supplier. When cells are stained for intracellular antigens, an Ig block is usually ineffective against nonspecific binding because the internal pool of proteins is so high. When a neat antibody and second fluorochrome-conjugated step is combined with staining by other biotinylated, hapten-conjugated, or directly conjugated antibodies, however, it is necessary to block with IgG. After the initial incubation to allow the antibody or isotype control to enter permeabilized cells, the cells are diluted in PBS and incubated an additional time to allow the unbound antibody or isotype control to leave the cells. This extended wash reduces NSB effectively while not affecting specific binding. It is important to prepare tubes containing the second antibody alone to distinguish between NSB due to Ig, revealed by the isotype control, and that produced by the second antibody itself. If NSB of the second reagent is high, it may be necessary to retiter or find a new supplier. The F(ab′)2 fragment should be used for second antibodies. The unstained cells provide a reference upon which all other binding can be evaluated. When antibodies are combined for multicolor analysis, it must be determined that they behave independently and do not interact with one another. Once the combination is shown to give reliable results, it can be used for immunophenotyping. The combination should be made in batches whose size depends on the number of specimens to be evaluated before the earliest antibody expires. If an additional batch is required, it should be prepared in time to have its performance verified against the current batch. When a new combination is designed, it must be tested to verify its intended use. A minimum of 25 controls and samples from 25 patients approximating the ethnic origin and age distribution of the target population needs to be tested. Reference ranges for each resolved population are determined and become a part of the combination’s verification record. As the combination is used, newly acquired values can be added to the reference ranges to further enhance the combination’s validity. When antibodies are combined, each one’s volume dilutes the others so that each one is at a different concentration when mixed. This can have a profound effect on the combination’s
staining quality. For example, if each of three antibodies has a titer of 100 ng in 20 µl, the concentration of one antibody alone after addition to 50 µl of cells is 100 ng/(20 µl + 50 µl) = 1.4 ng/µl, the antibody’s 1× titer. If each of the three antibodies has the same titer, when they are mixed together their combined volume is 60 µl; when the combination is added to 50 µl of cells, the concentration of each antibody is now only 100 ng/ (20 µl + 20 µl + 20 µl + 50 µl) = 0.9 ng/µl, a 36% reduction in the titer of each type of antibody in the combination. Each antibody in the combination is no longer at its optimal concentration required to stain cells. Thus, when mixing antibodies, stock concentrations should be high enough so the volume required is low, to minimize mutual dilution. For any number of antibodies in combination, the amount of each to mix is determined by simultaneous equations as follows: Ab1V1 = C1 (Vc + V1 + V2 + + Vn ) Ab 2 V2 = C2 (Vc + V1 + V2 + + Vn ) Ab n Vn = Cn (Vc + V1 + V2 + + Vn )
where Abn is the concentration of stock antibody, Vn is the required volume of stock antibody to be combined, Cn is the concentration of Abn required to optimally stain cells, and Vc is the final volume of the cell suspension after adding all antibodies. As an example, suppose one wishes to mix together two antibodies to stain cells in 50 µl, and has the following data: Ab1: Stock concentration 10 ng/µl, titer 1 ng/µl. Ab2: stock concentration 30 ng/µl, titer 6 ng/µl. Substituting these values and solving for V1 and V2 yields the following: Ab 1V1 = C1 (Vc + V1 + V2 ) V1 = 5.55 + 0.11V2 Ab 2 V2 = C2 (Vc + V1 + V2 ) V2 = 11.7 µl V1 = 5.55 + 0.11V2 = 6.84 µl
Thus, for one test it is necessary to mix together 6.84 µl of Ab1 and 11.7 µl of Ab2 and add this to 50 µl of cells, for a total volume of 68.54 µl. To make a 100-test batch, it is neces-
4.1.12 Current Protocols in Cytometry
sary to combine 684 µl of Ab1 with 1170 µl of Ab2, and use 18.5 µl per test for staining cells in 50 µl.
Anticipated Results High-quality antibodies produce signal-tonoise ration (S/N) >3 when <0.5 µg is used for staining 106 cells in a 100-µl volume. If a concentration >1 µg/100 µl is required, the antibody is unsuitable for use. For some cell types, there is a continuum from positive to negative cells without a clearcut distinction. Their epitopes are especially difficult to titer, resulting in a plateau effect or no peak in the signal-to-noise ratio. Identifying antibodies with the greatest possible S/N assists in distinguishing positive and negative cells. Antibodies that never show a plateau in MCF or a peak S/N should not be used, and an alternate supplier found.
Time Considerations The titering procedures for epitopes or antibodies should take <1 hr, once cells are prepared. After collecting the data, an analysis program can be designed permitting data reduction in 5 min. Cell preparation time for permeabilization will depend upon the choice of reagents from different manufacturers. If freeze/thaw is used, <1 hr is sufficient. The procedures using prepared permeabilized cells, requiring two 45min incubations, may take 2 to 3 hr.
Key Reference Stewart, C.C. and Stewart, S.J. 1994. Cell preparation for the identification of leukocytes. Methods Cell Biol. 41:39-60. General reference on immunophenotyping.
Contributed by Carleton C. Stewart and Sigrid J. Stewart Roswell Park Cancer Institute Buffalo, New York
Molecular and Cellular Probes
4.1.13 Current Protocols in Cytometry
Conjugation of Fluorochromes to Monoclonal Antibodies
UNIT 4.2
The most widely used application of flow cytometry is the detection of cell surface molecules labeled by monoclonal or polyclonal antibodies conjugated with a fluorochrome. The specificity provided by monoclonal antibodies makes them ideal for use as diagnostic reagents, and therefore the ability to conjugate these proteins with a variety of fluorochromes adds to their flexibility and utility in flow cytometric applications. This unit consists of protocols for tagging monoclonal antibodies with fluorescein, biotin, Texas Red, and phycobiliproteins. In addition, a procedure is included for preparing a PE−Texas Red tandem conjugate dye that can then be used to conjugate antibodies. Using these protocols, investigators can label antibodies of their choice with multiple fluorochromes; not only is this cost-effective, but it permits more combinations of antibodies to be used in multicolor flow cytometric applications. The protocols for the conjugation of fluorescein (see Basic Protocol 1), biotin (see Basic Protocol 2), and Texas Red (see Basic Protocol 3) are similar—consisting of simple chemistry and dialysis steps—as these are all small inorganic molecules. Differences in the labeling procedures depend upon the type of reactive group attached to the fluorochrome: labeling buffers are optimized for the reactive groups, not the fluorochrome itself. Thus, for example, Basic Protocol 1 is for fluorescein with an isothiocyanate reactive group (FITC); other forms of fluorescein are available with succidimyl ester reactive groups, and for these the biotin labeling protocol should be substituted. The protocols for conjugation of phycobiliproteins (see Basic Protocol 4) and preparation of the PE−Texas Red tandem conjugate dye (see Basic Protocol 5) are more involved, requiring size separation of products on large gel filtration columns, and more complex chemistries. CAUTION: DMSO, DMF, and THF are hazardous; follow appropriate precautions for handling and disposal when performing these procedures. LABELING ANTIBODY WITH FLUORESCEIN ISOTHIOCYANATE (FITC) Conjugation of fluorescein isothiocyanate (FITC) to purified antibody is an extremely valuable technique for identifying surface molecules using either fluorescence microscopy or flow cytometry. In the procedure that follows, the amino groups of the antibody molecule are coupled with fluorescein derivatives. Following removal of unbound FITC, the fluorochrome/antibody ratio is determined and the labeled antibody is used in the basic and alternate protocols.
BASIC PROTOCOL 1
Materials 1 to 2 mg/ml purified monoclonal antibody FITC labeling buffer (prepare ≤2 weeks before use; see recipe) 5 mg/ml FITC, isomer I, in anhydrous dimethyl sulfoxide (DMSO) Final dialysis buffer (see recipe) Sephadex G-25 column (Pharmacia Biotech PD-10; optional) Dialysis tubing
Molecular and Cellular Probes Contributed by Kevin L. Holmes, Larry M. Lantz, and Wesley Russ Current Protocols in Cytometry (1997) 4.2.1-4.2.12 Copyright © 1997 by John Wiley & Sons, Inc.
4.2.1
Conjugate FITC and antibody 1. Dialyze purified monoclonal antibody against 500 ml FITC labeling buffer at 4°C with two or three changes over 2 days. Allow ≥4 hr between buffer changes. This step removes free NH4+ ions and raises the pH to 9.2. Generally, up to 5 ml antibody can be dialyzed against 500 ml buffer. For discussion of dialysis and a detailed procedure, see Andrew and Titus (1991).
2. Determine antibody concentration based upon A280. Concentration of antibody (mg/ml) = A280 × 0.74 × (dilution factor).
3. Add 20 µl of 5 mg/ml FITC in DMSO for each milligram of antibody. Incubate 2 hr at room temperature. Both the dye and organic solvent must be anhydrous; prepare FITC/DMSO solution immediately before use.
4. Remove unbound FITC by dialysis against 500 ml final dialysis buffer at 4°C with two or three changes over 2 days. Alternatively, filter on a Sephadex G-25 column. Determine FITC/antibody ratio 5. Dilute a small volume (∼100 µl) FITC-IgG complex with dialysis buffer so that A280 = <2.0. 6. Determine and record A280 and A492. 7. Calculate protein concentration as follows: mg/ml protein =
A280 − ( A492 × 0.35) 1.4
where 1.4 is the reciprocal of the FITC-conjugated antibody molar coefficient. 8. Calculate moles of protein: mg/ml protein 1.5 × 10 5 A492 moles FITC = 0.69 × 10 5
moles protein =
where 1.5 × 105 = mol. wt. Ig and 0.69 × 105 = mol. wt. FITC. 9. Determine fluorochrome/protein (F/P) ratio: F/ P =
moles of FITC moles of protein
An F/P of 5 to 6:1 is usually optimal for flow cytometry.
Stabilize antibody-dye conjugate 10. Dilute FITC-IgG complex 1:1 with stabilizing buffer.
Conjugation of Fluorochromes to Monoclonal Antibodies
4.2.2 Current Protocols in Cytometry
LABELING ANTIBODY WITH LONG-ARMED BIOTIN Biotin is a naturally occurring vitamin with a molecular weight of 244 Da and an extremely strong affinity for avidin (Kd = 10 to 15 M−1). Thus, biotin-labeled antibodies can be detected using commercially available avidin coupled to fluorochromes. Labeling antibodies with biotin provides flexibility by offering a choice of different fluorochromes to be used depending on the needs of the experiment. Moreover, because avidin has four binding sites for biotin and multiple biotin molecules can be conjugated to a single antibody, the fluorescent signal is considerably amplified when biotin/avidin is used, compared to that obtained by direct conjugation of the antibody with the fluorochrome.
BASIC PROTOCOL 2
Because the binding of biotin or the subsequent binding of avidin may induce changes in protein structure, many companies now supply biotin containing a spacer between the protein-binding site and the avidin-binding site (sometimes known as long-armed or spacer biotin). Biotin can also be easily coupled to antibodies via a hydroxysuccinimide ester, usually without disturbing the biological properties of the antibody. The following protocol is for conjugating either IgG or IgM antibodies; alternative information appropriate for the two types of antibodies is indicated in certain steps. Conjugation of IgM antibodies using dialysis buffer at pH 7.5, rather than pH 8.4, provides consistently better labeling, perhaps due to overlabeling og the IgM at higher pH. Materials 1 to 2 mg/ml purified monoclonal antibody Succinimide ester labeling buffer or IgM labeling buffer (see recipes) 10 mg/ml long-armed biotin (Zymed) in anhydrous N, N-dimethylformamide (DMF) Dialysis tubing 1. Dialyze 1 to 2 mg/ml purified antibody against 500 ml succinimide ester labeling buffer (for IgG) or IgM labeling buffer (for IgM) at 4°C with two to three changes over 2 days. Allow ≥4 hr between buffer changes. For discussion of dialysis and a detailed procedure, see Andrew and Titus (1991).
2. Determine protein concentration by measuring A280. Concentration of antibody (mg/ml) = A280 × dilution factor × 0.7 (for IgG) or 0.8 (for IgM).
3. Add 10 µl of 10 mg/ml biotin in DMF for each milligram of antibody. Incubate 1 hr at room temperature. Both the dye and organic solvents must be anhydrous; prepare biotin/DMF solution immediately before use.
4. Remove unbound biotin by dialysis against final dialysis buffer at 4°C as in step 1. Biotin/protein ratio cannot be determined spectrophotometrically, but titration comparison of the same antibody labeled with FITC can indicate whether relabeling is necessary.
5. Dilute biotin-antibody complex solution 1:1 with stabilizing buffer.
Molecular and Cellular Probes
4.2.3 Current Protocols in Cytometry
BASIC PROTOCOL 3
LABELING WITH TEXAS RED−X Texas Red, the sulfonylchloride derivative of sulforhodamine 101, has been used for many years in dual-laser multiparameter flow cytometry. However, directly labeling antibodies with this dye can be difficult, depending upon the class of the antibody and host species (Titus et al., 1982). Concentrations required to achieve adequate dye/protein ratios often precipitate the antibody-dye conjugates. The recent development of the modified Texas Red−X succinimidyl ester has greatly improved Texas Red labeling, allowing a greater range of antibodies to be labeled with substantially less precipitation of antibody-dye conjugates. The procedure is similar to the protocol for biotin labeling, with the modifications detailed below. Materials 1 to 2 mg/ml purified monoclonal antibody Succinimide ester labeling buffer (see recipe) 5 mg/ml Texas Red−X succinimidyl ester (Molecular Probes) in N,N-dimethylformamide (DMF) Final dialysis buffer (see recipe) Stabilizing buffer (see recipe) Dialysis tubing Sephadex G-25 column (Pharmacia Biotech; optional) 1. Dialyze purified monoclonal antibody against 500 ml succinimide ester labeling buffer at 4°C with two or three changes over 2 days. Allow ≥4 hr between buffer changes. For discussion of dialysis and a detailed procedure, see Andrew and Titus (1991).
2. Determine antibody concentration based upon A280 and adjust to 1 to 2 mg/ml. Concentration of antibody (mg/ml) = A280 × 0.7 × dilution factor.
3. Add 5 µl of 5 mg/ml Texas Red−X in DMF for each milligram of antibody. Incubate 1 hr at room temperature. Both the dye and organic solvents must be anhydrous; prepare Texas Red−X/DMF solution immediately before use.
4. Remove unbound Texas Red−X by dialysis at 4°C as in step 1, but using final dialysis buffer. Alternatively, filter on a Sephadex G-25 column. 5. Remove any precipitated antibody by centrifuging 3 min at 10,000 × g. 6. Determine Texas Red/antibody ratio by measuring A596/A280. A ratio of 0.5 to 0.7 usually gives the best results and probably represents two to three Texas Red molecules bound per antibody, based upon a molar extinction coefficient for antibody bound to Texas Red of 8.4 × 104 M−1 at 596 nm (Titus et al., 1982).
7. Dilute Texas Red–Ig complex solution 1:1 with stabilizing buffer. BASIC PROTOCOL 4
Conjugation of Fluorochromes to Monoclonal Antibodies
LABELING ANTIBODY WITH PHYCOBILIPROTEINS Coupling phycobiliproteins such as phycoerythrin (PE) and allophycocyanin (APC) to antibodies is more difficult than labeling with FITC or biotin. A sulfhydryl-maleimide linkage is used to couple the antibody to the phycobiliprotein. The unbound antibody and phycobiliprotein are then separated by size on a gel filtration column. The procedure described here is for PE-antibody coupling. The step for APC coupling is identical except where noted.
4.2.4 Current Protocols in Cytometry
Materials 10 to 25 mg/ml phycoerythrin (PE; purchased as suspension in buffered ammonium sulfate solution) Coupling buffer, pH 5.5 and 7.5 (see recipe) Sulfhydryl addition reagent: N-succinimidyl-S-acetylthioacetate (SATA; Calbiochem; store under nitrogen after opening) Dimethylformamide (DMF) Nitrogen Deacetylation buffer (see recipe) Heterobifunctional cross-linker: γ-maleimidobutyric acid N-hydroxysuccinimide ester (GMBS; Calbiochem; store under nitrogen after opening) Tetrahydrofuran (THF) 0.1 mg/ml cysteine Running buffer, degassed (see recipe) Dialysis tubing AcA 34 column (IBF Biotechnics) Sephacryl S-200 column (Pharmacia Biotech; optional) Prepare the PE-SATA conjugate 1. Dialyze PE against 500 ml coupling buffer, pH 7.5, at 4°C with two or three changes over 2 days. Use sufficient PE to give a PE/IgG (w:w) ratio of 3:1 and allow ≥4 hr between buffer changes. For discussion of dialysis and a detailed procedure, see Andrew and Titus (1991). The precise concentration of PE must be determined by spectrophotometric measurements at A596 and the concentration adjusted with coupling buffer to fall within the indicated range.
2. Dilute SATA to 1 mg/ml in DMF. 3. Add 10 µl diluted SATA solution for each milligram of PE to be labeled. Incubate 30 min at room temperature. 4. Dialyze PE-SATA conjugate in 500 ml coupling buffer, pH 7.5, at 4°C with two or three changes to remove unreacted SATA. Store at 4°C for later use. Label the antibody and isolate the conjugate 5. Dialyze purified antibody in 500 ml coupling buffer, pH 7.5, as for FITC labeling (see Basic Protocol 1, step 1) to a final IgG concentration of ≥1 mg/ml. 6. Dilute GMBS to 2 mg/ml (7.14 mM) in THF. 7. Deacetylate PE-SATA conjugate from step 4 by adding 100 µl deacetylation buffer for each milliliter of PE-SATA. Incubate 1 hr at room temperature. 8. Add 10 µl diluted GMBS solution for each milligram of antibody to be labeled. Incubate 30 min at room temperature. 9. Wash one Sephadex G-25 column for each 2.0 ml IgG-GMBS conjugate solution to be loaded by adding 10 ml coupling buffer (pH 5.5) per column. Load 2.0 ml IgG-GMBS solution onto washed column. Monitor eluate spectrophotometrically using a 280-nm filter and collect the portion represented by the first peak. Proceed immediately to step 10. The first peak is the GMBS-labeled antibody. The second peak is free GMBS and should be discarded. Molecular and Cellular Probes
4.2.5 Current Protocols in Cytometry
unbound PE and antibody
OD280
1:1 > 1:1
Fraction Figure 4.2.1 Elution profile of phycoerythrin-labeled IgG from a gel filtration column. The initial peaks have greater than a 1:1 ratio of dye bound to antibody, and are not used. Optimal material is found in the center fractions designated 1:1. The trailing peaks contain unbound dye and unconjugated antibodies, and are discarded.
Couple PE-SATA and IgG-GMBS conjugates 10. Mix deacetylated PE-SATA conjugate from step 7 with IgG-GMBS conjugate from step 9 immediately after isolating the latter. Incubate 2 hr at room temperature. Use a 3:1 ratio of PE/IgG for optimum yield, but use a 2:1 ratio of allophycocyanin/IgG.
11. Quench residual maleimide groups by adding 0.1 mg/ml cysteine to twice the antibody concentration. For example, add 25 ìl of 0.1 mg/ml cysteine (570 ìM) per milligram of IgG.
12. Separate PE-IgG conjugate from unconjugated PE and free IgG using an AcA 34 column. Sample volume loaded onto the column should be between 0.5% and 4% of total column bed volume. Pour an appropriately sized column, using degassed running buffer, according to manufacturer’s directions. Due to slow packing and running rates, it generally requires one night to pack a column and an additional night to isolate the sample. Therefore it is advisable to pack the column before labeling.
13. Load sample onto column and run column at manufacturer’s suggested rates. Collect fractions 1⁄20 the column volume. Two well-separated red bands will appear on the column and several peaks will appear on the column A280 printouts. The first peaks are PE-IgG conjugates with more than one PE per IgG. The peak with one PE per IgG will appear immediately before the largest peak consisting of unconjugated PE and IgG. See Figure 4.2.1 for sample results. Confirm number of PE per IgG using flow cytometry techniques or spectrophotometrically using A596 /A280 ratios. Best results have come from using the one-PE-per-Ig conjugate. BASIC PROTOCOL 5
Conjugation of Fluorochromes to Monoclonal Antibodies
CONJUGATION OF TEXAS RED TO R-PHYCOERYTHRIN TO PRODUCE AN ENERGY TRANSFER FLUOROCHROME An advantage of flow cytometric analysis is the ability to distinguish functionally significant populations of cells. The need to further subdivide these populations using antibody probes against known cell surface antigens requires increasingly complex multicolor analyses. Fluorochromes excitable with a single excitation source and possessing emission wavelengths distinct enough to be detected separately are needed to distinguish the different antigens. A highly effective way to achieve this purpose is the use of energy-transfer fluorochromes, which allow three- and four-color single-excitation
4.2.6 Current Protocols in Cytometry
flow cytometry. One of the more useful energy-transfer fluorochromes is the conjugate of R-PE and Texas Red, whose preparation and use is detailed in this protocol. The emission from R-PE overlaps with the absorption of Texas Red, allowing energy transfer if the two molecules are placed within a limiting distance from each other (Glazer and Stryer, 1983). Materials 10 to 50 mg R-phycoerythrin (R-PE; purchased as suspension in buffered ammonium sulfate solution) R-PE dialysis buffer (prepared within 2 days of use; see recipe) Conjugation buffers A and B (see recipes) Texas Red−sulfonyl chloride (Molecular Probes) N,N-Dimethylformamide (DMF) Glycine (ultrapure or ACS grade) 0.5 M hydroxylamine⋅HCl, pH 7.2 (prepared as for deacetylation buffer, but without EDTA; see recipe) Equilibration buffer (see recipe for HIC column buffers) First-wash buffer (see recipe for HIC column buffers) Elution buffers A and B (see recipe for HIC column buffers) Dialysis tubing Sephadex G-50 fine columns, one with ~5 ml capacity and one larger (e.g., ∼50 ml capacity for labeling 10 mg of R-PE) Fraction collector HIC (hydrophobic interaction column) TSK-Gel Toyopearl Butyl 650M (e.g., 25 ml capacity for labeling 10 mg of R-PE) Gradient maker with 200-ml capacity UV monitor and chart recorder Magnetic stir plate 10- to 15-ml glass test tube Flea (small stir-bar) Peristaltic pump Centrifuge and appropriate tubes (e.g., Beckman tabletop with 50-ml conical tubes) Spectrophotometer with quartz cuvettes 1. Dialyze 10 to 50 mg R-phycoerythrin (R-PE) against 500 ml R-PE dialysis buffer with two or three changes over 2 days at 4°C. Allow ≥4 hr between buffer changes. Protect R-PE from light by covering containers with foil during dialysis and in all subsequent steps when practical. For discussion of dialysis and a detailed procedure, see Andrew and Titus (1991). R-PE concentration should be in the range of 20 to 30 mg/ml.
2. Determine the total amount of R-PE: determine the concentration of R-PE, and then measure the total volume of R-PE solution. Concentration of R-PE (mg/ml) = A565 × 0.122 × dilution factor. mg R − PE = R − PE × volume (ml).
3. Decant the R-PE into a 10- to 20-ml glass test tube suspended in an ice bath on a magnetic stir plate. Mix gently with a flea. 4. While stirring, add dropwise sufficient conjugation buffer A to equal 20% the volume of R-PE. 5. While stirring, add dropwise sufficient conjugation buffer B to equal 25% the volume of R-PE.
Molecular and Cellular Probes
4.2.7 Current Protocols in Cytometry
6. Calculate quantity of Texas Red–sulfonyl chloride to use to give a 37:1 Texas Red/PE conjugation ratio. mg Texas Red required = (mg R-PE/240,000 mg/mmol) × 37 × (625 mg/mmol).
7. Increase the mixing rate of the R-PE solution to a very rapid rate. 8. Dissolve twice the amount of Texas Red required (as determined in step 6) in DMF. Mix in an open glass tube or vial on a vortex until dissolved. 9. Immediately add the dissolved Texas Red to the R-PE solution mixing in the ice bath. 10. Increase the mixing to the most rapid rate possible and maintain for 2 to 3 min, then reduce to the normal mixing rate. 11. After 10 min, remove ∼20 µl of the solution and load it on a Sephadex G-50 column containing ∼5 ml resin. Collect the first peak and measure A565 and A596. If the A565/A596 ratio is between 2.3 and 3.3, proceed to the next step. If the ratio is <2.3, the R-PE is overlabeled with Texas Red. Discard overlabeled conjugate and start again at step 1 using a lower Texas Red/PE ratio (see step 6). If the ratio is >3.3, more Texas Red needs to be reacted with the R-PE and the process repeated.
12. While continuing to mix, add 5 mg solid glycine per mg Texas Red (determined in step 6) to quench the reaction. From step 2, it should take ~3 hr for the conjugation and 2 hr for chromatography. After glycine is added, the product is stable ≤5 days at 4°C.
13. Once the glycine is dissolved, add a volume of 0.5 M hydroxylamine⋅HCl, pH 7.2, equal to 10% of the mixing R-PE. Measure the volume. 14. Pass the resulting solution over a Sephadex G-50 fine column equilibrated with R-PE dialysis buffer, using 25 ml resin per ml solution volume. At this point the product may be stored ≤5 days at 4°C.
15. Collect the first peak and dialyze overnight against equilibration buffer at 4°C. 16. Centrifuge 20 min at 900 × g, 4°C. Remove the supernatant using a pipet, being careful not to disturb the pellet. Calculate the concentration of R-PE in the R-PE− Texas Red complex as in step 2. The concentration prior to starting the next step should be ≤4 mg/ml; adjust if necessary using equilibration buffer. 17. Equilibrate a HIC Toyopearl butyl column in equilibration buffer, using 1.5 ml of resin per mg of R-PE−Texas Red complex. Flow rate should be 1 to 3 ml/min. 18. Load the dye solution onto the column and wash with 2 column volumes of equilibration buffer. 19. Wash the column with 10 column volumes first-wash buffer, at a flow rate of 1 to 3 ml/min. 20. Connect a UV monitor to the column. Run a 1:2 gradient of elution buffer A to elution buffer B, using 15 ml elution buffer A and 30 ml of elution buffer B for each 1.5 ml of column resin. Collect fractions equal to one-tenth the column volume. A broad peak of weakly colored fractions will precede a higher region with intense color trailing off to fractions with less color. The fractions in the middle are optimal for conjugation to antibodies. Conjugation of Fluorochromes to Monoclonal Antibodies
From step 16 to this point takes ~4 hours. The product may be stored up to 15 days at 4°C. To store longer, add sodium azide to 1% final concentration. The sodium azide will require removal by dialysis prior to conjugation.
4.2.8 Current Protocols in Cytometry
21. For conjugation to antibodies, determine the concentration of the R-PE−Texas Red dye as for R-PE. On the same day as conjugation, centrifuge the solution 20 min at 600 × g, 4°C; then conjugate using the procedure described for labeling antibody with phycobiliproteins (see Basic Protocol 4). REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Conjugation buffers Conjugation buffer A (1.4 M sodium sulfate): Dissolve 20 g sodium sulfate in 100 ml of R-PE dialysis buffer (see recipe). Adjust pH to 7.2 with 1 M KOH. Conjugation buffer B (1 M potassium borate, pH 9.8): Prepare a 1 M solution of boric acid (H3BO3) in water. Adjust pH to 9.8 with 8 M KOH. The solid borate will slowly go into solution and lower the pH; slight adjustments with addition of KOH will allow the borate to dissolve completely.
Coupling buffer 0.1 M Na2HPO4⋅7H2O 0.1 M NaCl 1 mM EDTA Adjust pH to 7.5 or 5.5 with concentrated HCl Deacetylation buffer Dissolve 3.47 g hydroxylamine (mono HCl; 0.5 M final) and 0.73 g EDTA (anhydrous free acid; 0.025 M final) in ∼50 ml water and adjust pH to 7.5 with solid anhydrous disodium hydrogen phosphate. Add water to 100 ml final volume. Final dialysis buffer 0.1 M Tris⋅Cl, pH 7.4 0.1% (w/v) NaN3 0.2 M NaCl Adjust pH to 7.4 with 5 M NaOH Store at 4°C FITC labeling buffer 0.05 M boric acid (H3BO3) 0.2 M NaCl Adjust pH to 9.2 with 5 M NaOH Store at 4°C HIC column buffers Equilibration buffer 100 mM K2HPO4 2 mM EDTA 200 mM sodium sulfate Adjust pH to 7.2 with 1 M KOH First-wash buffer 100 mM K2HPO4 2 mM EDTA 135 mM sodium sulfate Adjust pH to 7.2 with 1 M KOH
Elution buffer A 100 mM K2HPO4 2 mM EDTA 70 mM sodium sulfate Adjust pH to 7.2 with 1 M KOH Elution buffer B 100 mM K2HPO4 2 mM EDTA Adjust pH to 7.2 with 1 M KOH Molecular and Cellular Probes
4.2.9 Current Protocols in Cytometry
Supplement 1
IgM labeling buffer 0.1 M Na2HPO4⋅7H2O 0.15 M NaCl Adjust pH to 7.5 with concentrated HCl Store at room temperature R-PE dialysis buffer 100 mM K2HPO4 2 mM EDTA Adjust pH to 7.2 with 1 M KOH Store at room temperature This buffer may be used for up to one week.
Running buffer 81.82 g NaCl 4 ml glycerol Dissolve in 3.8 liters phosphate-buffered saline (PBS; APPENDIX 2A) Adjust pH to 7.5 with concentrated HCl Add PBS to 4 liters To degas buffer, place room temperature buffer in an Erlenmeyer flask equipped with a one-hole stopper and tubing (alternatively, a side-arm vacuum flask and stopper may be used). Apply vacuum through the tubing (or side-arm flask) while stirring buffer vigorously. Sample is degassed when no more bubbles rise out of solution.
Stabilizing buffer Hanks’ balanced salt solution (HBSS) without phenol red (APPENDIX 2A) 0.1% (w/v) NaN3 5.0% (w/v) bovine serum albumin (BSA; fraction V) Store at 4°C Succinimide ester labeling buffer 0.1 M NaHCO3 0.1 M NaCl Adjust pH to 8.4 with concentrated HCl Store at room temperature COMMENTARY Background Information
Conjugation of Fluorochromes to Monoclonal Antibodies
The choice of the dye to conjugate with an antibody is dependent upon several factors, including the time the investigator is willing to invest in conjugation, the available excitation source(s), whether the conjugate will be used in combination with other dyes, and the density of the antigen that is being detected. Labeling procedures with fluorescein isothiocyanate (FITC), biotin, and Texas Red are easy to perform, whereas labeling procedures with phycobiliproteins such as phycoerythrin (PE) and allophycocyanin (APC) and with phycoerythrin–Texas Red (PE−Texas Red) are more difficult and time consuming, taking several days to complete (Brinkley, 1992). In addition, FITC, biotin, and Texas Red can be efficiently conjugated to small amounts (e.g., 1 mg) of
purified antibody, but phycobiliprotein and PE−Texas Red conjugation require larger amounts of purified antibody and give a lower final yield. FITC, PE and PE−Texas Red are all excitable by 488 nm argon lasers, but Texas Red is excitable by argon-dye (rhodamine 6G) or krypton (operating at 568 nm). For detection of surface antigens with low density, phycobiliproteins or tandem conjugate dyes offer better signal-to-noise ratios than FITC or Texas Red, because of their large quantum yields and extinction coefficients. Methods for conjugating with phycobiliproteins using sulfhydryl-maleimide linkages are presented (Duncan et al., 1981; Tanimori et al., 1983), although other linkages can be used (Kitagawa et al., 1981; Hashida et al., 1984; Blattler et al., 1985; Kronick, 1988).
4.2.10 Supplement 1
Current Protocols in Cytometry
Critical Parameters and Troubleshooting Preparation and storage of reagents Although the protocols for labeling antibodies with fluorochromes and biotin are simple, results are highly dependent upon the quality of reagents used. The organic solvents (DMF and DMSO) and the dye powders must be anhydrous. For this reason it is recommended that dyes be purchased in small amounts and stored in a desiccator. Organic solvents can be purchased packed under nitrogen in syringe vials. Solutions of FITC, Texas Red−X, and biotin should be prepared just prior to use. It is recommended that FITC labeling buffer and coupling buffer be made <2 weeks before use. In immunophenotypic analysis, IgG antibodies can bind to Fc receptors regardless of their antigen specificity. This problem can be minimized by ultracentrifugation (e.g., using a Beckman Airfuge). IgG-FITC, -biotin, or Texas Red conjugates may be airfuged 15 min at 100,000 × g to remove aggregates, and retitered for optimal dilution. For IgG-PE, -PE–Texas Red, or all IgM conjugates, centrifuge 5 min at 12,000 × g to remove aggregates prior to retitration. It is essential that the Texas Red sulfonylchloride ester be added to the rapidly mixing R-PE quickly, as the ester is reactive for only a few minutes. There will be some R-PE−Texas Red complexes remaining on the hydrophobic interaction column. This material is not suitable for conjugation to antibodies. When attempting to label small amounts of protein (≤0.5 mg), it is advisable to use an apparatus such as the Pierce Microdialyzer to avoid loss or dilution of antibody. It is not advisable to use a Sephadex G-25 column for separation of labeled antibody from unbound fluorochrome, as this will cause considerable dilution of small-volume samples. Because Texas Red is hydrophobic, the optimal method for separation of Texas Red-X–labeled antibody from the fluorochrome is a Sephadex G-25 column. Antibody-fluorochrome conjugates should be stored at 4°C, protected from light. Dilution of antibody-fluorochrome (or -biotin) conjugates with stabilizing buffer greatly increases shelf life by preventing aggregation of conjugates. Conjugates may be stored 1 year or longer at 4°C, although individual antibodies may vary. For longer storage, most antibodies (diluted in stablilizing buffer) may be dispensed in small aliquots and frozen at −70°C. Small volumes should be pretested for stability
after freeze-thaw. Antibodies should be frozen only once. Antibody-fluorochrome (-biotin) conjugates can be filter-sterilized as necessary, using a sterile syringe filter equipped with a 0.22-µm low-protein-binding membrane. Deterioration of antibody-fluorochrome (-biotin) conjugates may be indicated by visible precipitation, or by loss of fluorescent signal in standardized flow cytometric analyses. Deterioration of PE antibody–Texas Red conjugates may be indicated by the inability to perform sufficient electronic compensation of PE– Texas Red signal from the PE detector on the flow cytometer. This may arise from uncoupling of Texas Red from PE, resulting in decreased efficiency of energy transfer, and therefore more PE emmission. Final dialysis buffer contains sodium azide, which when dried at high concentrations may spontaneously combust. Sodium azide–containing solutions are highly toxic, and should be disposed of by dilution with large quantities of water. Optimization of fluorescence A high fluorochrome (or biotin)/protein ratio improves fluorescent signal-to-noise ratio in flow cytometric analysis. The amounts of fluorochrome (or biotin) to be used per milligram of antibody cited in the protocols are guidelines only. Because of inherent differences in monoclonal antibodies, it may be necessary to label several batches of antibody, varying the amount of fluorochrome (or biotin) used above and below those suggested amounts in order to achieve optimal labeling. If the efficiency of the energy transfer fluorochrome is poor, then the wash step prior to the elution gradient should be lengthened. A fluorescent spectrophotometer can determine the point when the dye complex is eluting off the column very accurately. The ratio of emission at 575 nm to emission at 613 nm when the eluant was excited at 488 nm is representative of the efficiency of the transfer process. Anticipated Results Labeling of antibodies with either FITC or biotin generally results in excellent yield and very little loss of protein (<5%). Texas Red labeling, on the other hand, generally results in a considerable loss of protein (10% to 20%)— presumably due to overlabeling—that can be visualized as a precipitate after labeling. It is advisable to isolate and discard this precipitate by centrifugation as described.
Molecular and Cellular Probes
4.2.11 Current Protocols in Cytometry
For the preparation of the energy transfer fluorochrome, the final amount of usable product is ~25% of the amount of phycoerythrin initially used. For the conjugation of phycoerythrin to antibody, the final amount of usable product is ∼25% of the amount of antibody used.
Time Considerations The labeling of purified antibodies with FITC, biotin, and Texas Red-X will take ∼3 days. These procedures require relatively small amounts of hands-on time (2 to 4 hr). The majority of time is required for dialysis steps. Sephadex G-25 columns can be used in place of dialysis, but fractionation with a column will usually result in more protein loss and dilution than dialysis. A good stopping point in any of these procedures is after the start of any dialysis step. Labeling of antibodies with phycobiliproteins is more labor-intensive than the FITC, biotin, and Texas Red-X protocols. This procedure will take ~5 to 7 days. Most of this time is required for dialysis, column preparation, and fractionation. Good stopping points are after the start of any dialysis step, after the PE-SATA conjugation (Basic Protocol 4, step 4), and after residual maleimide groups are quenched with cysteine (Basic Protocol 4, step 11). For the preparation of the energy transfer fluorochrome, time to complete the various steps are listed within the protocol. The handson time required is ~10 hr; 3 to 4 days are necessary to complete the dialysis steps.
Duncan, R.J.S., Weston, P.D., and Wrigglesworth, R. 1981. A new reagent which may be used to introduce sulfydryl groups into proteins, and its use in the preparation of conjugates for immunoassay. Anal. Biochem. 132:68-73. Glazer, A.N. and Stryer, L. 1983. Fluorescent tandem phycobiliprotein conjugates. Biophys. J. 43:383-386. Hashida, S., Imagawa, M., Inoue, S., Ruan, K-H., and Ishikawa, E. 1984. More useful maleimide compounds for the conjugation of Fab′ to horseradish peroxidase through thiol groups in the hinge. J. Appl. Biochem. 6:56-63. Kitagawa, T., Shimozono, T., Aikawa, T., Yoshida, T., and Nishimura, H. 1981. Preparation and characterization of hetero-bifunctional crosslinking reagents for protein modification. Chem. Pharm. Bull. (Tokyo) 29:1130-1135. Kronick, M.N. 1988. Phycobiliproteins as labels in immunoassay. In Nonisotopic Immunoassay. (T.T. Ngo, ed.) pp. 163-185. Plenum Press, NY. Tanimori, H., Ishikawa, F., and Kitagawa, T. 1983. A sandwich enzyme immunoassay of rabbit immunoglobulin G with an enzyme labeling method and a new solid support. J. Immunol. Methods 62:123-131. Titus, J.A., Haugland, R., Sharrow, S.O., and Segal, D.M. 1982. Texas Red, a hydrophilic, red-emitting fluorophore for use with fluorescein in dual parameter flow microfluorometric and fluorescence microscopic studies. J. Immunol. Methods 50:193-204.
Key Reference Brinkley, 1992. See above. Wong, S.S. 1991. Chemistry of Protein Conjugation and Cross-Linking. CRC Press, Inc. A good reference book for both protein-protein coupling and reactive group chemistry
Literature Cited Andrew, S.M. and Titus, J.A. 1991. Dialysis and concentration of protein solutions. In Current Protocols in Immunology (J.E. Coligan, A.M. Kruisbeck, D.H. Margulies, E.M. Shevach, and W. Strober, eds.) pp. A.3H.1-A.3H.2. John Wiley and Sons, New York.
Contributed by Kevin L. Holmes and Larry M. Lantz National Institute of Allergy and Infectious Diseases Bethesda, Maryland
Blattler, W.A., Kuenzi, B.S., Lambert, J.M., and Senter, P.D. 1985. New heterobifunctional protein cross-linking reagent that forms an acidlabile link. Biochemistry 24:1517-1524.
Wesley Russ Kirkegaard & Perry Laboratories Gaithersburg, Maryland
Brinkley, M. 1992. A brief survey of methods for preparing protein conjugates with dyes, haptens, and cross-linking reagents. Bioconjugate Chem. 3:2-13.
Conjugation of Fluorochromes to Monoclonal Antibodies
4.2.12 Current Protocols in Cytometry
Nucleic Acid Probes CHARACTERISTICS OF NUCLEIC ACID STAINS A major application of flow and image cytometry is the automated classification of populations of cells on the basis of differences in DNA or RNA content. The DNA content of a cell varies according to its cycle stage of the cell cycle. Tumor cells may or may not exhibit a nuclear DNA content that differs from their euploid (“normal”) counterparts. The RNA content of a cell, on the other hand, gives information on whether this particular cell is progressing along a pathway toward a differentiated stage. To obtain information on the DNA and RNA content of each individual cell, it is necessary to use dyes and staining procedures that meet specific, well-defined criteria. Chapter 7 of this volume details methods for analyses of cell cycle distribution, nuclear ploidy, and cell differentiation. The purpose of this unit is to discuss the key characteristics of fluorescent dyes that can be used in these methods; their most important features are displayed in Table 4.3.1. Nucleic acid dyes described to date fluoresce after cell staining because of a noncovalent association with their nucleic acid target. This unique mode of staining imposes four specific requirements for a dye to be useful for the quantitation of DNA and RNA on a per-cell basis. 1. The dye should be specific for nucleic acids. Dyes that give fluorescence while binding to a non–nucleic acid target will cause nonspecific background staining. 2. The dye should exhibit a reasonable degree of DNA or RNA selectivity. Dyes that bind to DNA and RNA equally well are of little use, because they do not allow quantitation of either nucleic acid. 3. After cell staining, fluorescence emission from the dye should be stoichiometric with either the cellular DNA or RNA content. If there is no stoichiometric relationship between DNA or RNA content on one hand and cell fluorescence on the other, the amount of fluorescence per cell does not reflect a quantitative feature of the cell under study. 4. Ideally, a nucleic acid stain should show a strong degree of fluorescence enhancement upon binding to its nucleic acid target. Dyes that meet only the first three criteria may have some use, but a dye showing strong fluores-
Contributed by Martin Poot Current Protocols in Cytometry (2003) 4.3.1–4.3.10 Copyright © 2003 by John Wiley & Sons, Inc.
UNIT 4.3 cence enhancement is preferred over a dye exhibiting less fluorescence enhancement upon nucleic acid binding. Dyes of widely varying chemistry have met the criteria outlined above. The chemistry and range of uses of fluorescent nucleic stains are extensively reviewed in the sixth edition of the Handbook of Fluorescent Probes and Research Chemicals by Haugland (1996), to which the reader is referred for a more in-depth general discussion of fluorescent dyes and a compilation of their chemical structures. A convenient means of classifying nucleic acid stains is by their mode of nucleic acid binding. One can distinguish DNA minor groove–binding dyes; dyes that intercalate between the bases of nucleic acids; dyes that develop fluorescence after a dual mode of nucleic acid binding; and dyes that exhibit different modes of DNA versus RNA binding (nucleic acid–precipitating and metachromatic dyes). The bisbenzimidazoles (also known as the Hoechst dyes) bind exclusively to the minor groove of double-stranded DNA. This feature endows Hoechst dyes with unsurpassed selectivity for DNA. Dyes that bind to both DNA and RNA by intercalation between the bases are the phenanthridines (e.g., propidium iodide and ethidium bromide), 7-aminoactinomycin D, and the dimeric cyanine dyes. The third group of dyes is comprised of those that fluoresce only after intercalating between the bases and simultaneously binding to the minor or major groove of a nucleic acid: the monomeric cyanine dyes fall into this category. The fourth group of dyes show different fluorescence emission spectra depending upon which polynucleotide (DNA or RNA) is bound; dyes of this type (e.g., the acridines and pyronin Y) can be used to detect RNA independently of the nuclear DNA content. Each group is considered separately in the sections that follow.
MINOR GROOVE–BINDING DYES Two bisbenzimidazoles that bind exclusively and specifically to the minor groove of DNA are currently in use in cytometry: Hoechst 33342 and Hoechst 33258. Hoechst 33258 penetrates cell membranes poorly, whereas Hoechst 33342 is taken up reasonably well by intact cells. After binding to double-stranded DNA, both dyes show an absorption maximum
Molecular and Cellular Probes
4.3.1 Supplement 26
Supplement 26
AT
dsDNA dsDNA
Hoechst 33258
DNA (RNA)
Dual-binding SYTOX Green
DNA (RNA)
Propidium iodide (PI) DNA (RNA)
dsDNA (RNA)
Ethidium homodimer
DRAQ5
No
dsDNA (RNA)
No
No
No
No
GC (weak)
DNA (RNA)
Interbase-intercalating 7-Aminoactinomycin D (7-AAD) Ethidium bromide
AT
AT
DNA
Base pair specificity
4′,6-Diamidino-2phenylindole (DAPI) Hoechst 33342
Minor groove–binding
Target moleculeb
504
646
536
528
510
545
346
346
359
Excitation maximum (nm)
Properties of Nucleic Acid Stains Used in Flow Cytometrya
Dye name and binding mode
Table 4.3.1
Nucleic Acid Probes
4.3.2
Current Protocols in Cytometry
523
681
617
617
595
647
460
460
461
Emission maximum (nm)
Argon (457) 488 nm
HeNe 633 nm; Kr 647 nm
(Mercury arc lamp) argon 488 nm Argon 488 nm
Mercury arc lamp; HeNe 543 nm (Mercury arc lamp) argon 488 nm
Mercury arc lamp; argon 360 nm Mercury arc lamp; argon 360 nm Mercury arc lamp; argon 360 nm
Excitation sourcesc
continued
Impermeant; used for cell cycle studies; provides live/dead discrimination
Impermeant; provides live/dead discrimination Permeant; used for cell cycle studies
Impermeant
Weakly permeant; provides live/dead discrimination Impermeant
Impermeant; used for cell cycle studies
Used for ploidy analysis and cell cycle studies Permeant; used for cell cycle studies
Comments
Current Protocols in Cytometry
DNA (RNA) DNA (RNA) DNA (RNA) DNA (RNA) DNA (RNA) DNA (RNA)
TO-PRO-1
TO-PRO-3
TOTO-1
TOTO-3
YO-PRO-1
YOYO-1
RNA DNA DNA RNA
460 502 559 560
491
491
642
514
642
515
Excitation maximum (nm)
cMajor excitation sources for each dye; excitation sources that may also work for the dye in parentheses.
No No No No
No
No
No
No
No
No
Base pair specificity
bMajor nucleic acid target of the dye; minor target in parentheses.
aData from Haugland (1996).
Pyronin Y
Acridine orange (AO)
Nucleic acid–precipitating and metachromatic
Target moleculeb
Properties of Nucleic Acid Stains Used in Flow Cytometrya, continued
Dye name and binding mode
Table 4.3.1
Molecular and Cellular Probes
4.3.3
650 526 569 573
509
509
660
533
661
531
Emission maximum (nm)
Mercury arc lamp; argon 457, 488 nm Argon 488 nm HeNe 543 nm
Argon 457, 488 nm (mercury arc lamp) Argon 457, 488 nm (mercury arc lamp)
HeNe 633 nm; diode 635 nm (mercury arc lamp) Argon 488, 514 nm (mercury arc lamp) HeNe 633 nm; diode 635 nm (mercury arc lamp)
Argon 488, 514 nm (mercury arc lamp)
Excitation sourcesc
Impermeant; provides DNA/RNA discrimination with Hoechst 33342
Provides DNA/RNA discrimination
Impermeant; used for cell cycle studies
Impermeant; used for cell cycle studies
Impermeant; used for cell cycle studies
Impermeant; used for cell cycle studies
Impermeant; used for cell cycle studies
Impermeant; used for cell cycle studies
Comments
Nucleic Acid Probes
at 346 nm, so they are well excited by the light from a mercury arc lamp or an argon laser (Table 4.3.1). By X-ray crystallography (Pjura et al., 1987), MPE-Fe(II) footprinting (Harshman and Dervan, 1985), and NMR spectroscopy (Searle and Embrey, 1990), Hoechst 33258 was shown to bind to the minor groove of double-stranded DNA at stretches of at least three AT base pairs flanked by one GC base pair. DNA binding studies over a wide range of Hoechst 33258 dye to DNA base pair ratios revealed multiple binding stoichiometries (Loontiens et al., 1990). The affinity of the dye for DNA or a synthetic polynucleotide appeared to be sensitive to ionic strength and to the presence of ethanol; the latter solvent eliminated one high-affinity DNA binding mode (Loontiens et al., 1990). Studies of fluorescence emission as a function of dye-to-DNA base (D/B) ratio showed that Hoechst 33258 follows at least two binding patterns: at a low dye-to-DNA ratio (D/P < 0.05) fluorescence emission with a maximum at 460 nm was recorded, whereas at higher D/B ratios fluorescence at 460 nm was quenched, and a new peak of fluorescence emission emerged (Stokke and Steen, 1985). This emission was maximal at 490 nm and exhibited a much lower quantum yield than the emission at 460 nm. At very high D/B ratios (> 0.4) the dye−DNA complex precipitated, irrespective of DNA strandedness and base composition (Stokke and Steen, 1985). In the presence of 25% ethanol the binding leading to fluorescence around 490 nm did not occur, and the affinity constant for the binding mode that gave fluorescence at 460 nm was strongly decreased. Ethanol prevented the precipitation of dyeDNA complex at very high D/B ratios (Stokke and Steen, 1985). Flow cytometric analysis of stained intact nuclei confirmed the results obtained with isolated DNA (Stokke and Steen, 1986). Because of the specificity of the Hoechst dyes for AT stretches of DNA, their fluorescence emission does not reflect total DNA content of cells. Therefore, fluorescence emission of the Hoechst dyes cannot be used to compare genome sizes of different species. However, these dyes can be successfully used to analyze by flow cytometry cell cycle distributions of cultures of a single cell type either directly after staining (with Hoechst 33342), after treatment with a detergent, or after cell fixation (Hoechst 33258). The complexity of the binding modes of the Hoechst dyes, which at first sight may
suggest a relatively simple relationship to their targets, calls for careful control of staining conditions. Inadvertent variation of conditions may jeopardize obtaining meaningful quantitative data. An interesting feature of fluorescence generation after binding of Hoechst dye to DNA is that fluorescence after DNA binding is decreased if halogenated uridines are incorporated into the DNA (Latt and Stetten, 1976). For instance, incorporation of 5-bromodeoxyuridine during the S phase of the cell cycle distinguishes cycling cells from those that did not traverse the S phase during the labeling interval. On the basis of this observation, a flow cytometric method to determine cell cycle kinetic parameters of cultures of cells has been devised (Rabinovitch, 1983; Rabinovitch et al., 1986). The physicochemical basis of this phenomenon has not been resolved, but it is inferred that the method whereby the Hoechst dye binds with its target may be drastically altered if base analogs are incorporated in its binding site (Loontiens et al., 1991). Indoles, such as 4′,6-diamidino-2-phenylind ole ( DAPI) and 4′,6-dicarboxyamide-2phenylindole (DCI), also bind to the minor groove of DNA, but in addition they show other modes of polynucleotide binding (Kapuscinski and Skoczylas, 1978; Tanious et al., 1992). The latter mode may lead to the formation of a small amount of nonspecific (RNA-dependent) fluorescence in the cytoplasm.
DYES THAT INTERCALATE BETWEEN BASES Historically, the first group of dyes found to exhibit fluorescence enhancement upon binding to nucleic acids were phenanthridinium derivatives such as ethidium bromide and propidium iodide. Waring (1965) showed that ethidium bromide binds to DNA and RNA with high affinity, and its binding saturates at one dye molecule for every four to five nucleotides; no dependency on base composition or strandedness was observed. At high ratios of dye molecules to DNA base pairs additional binding to the level of one dye molecule per DNA base pair may occur, but this leads to precipitation of the complex. Neutralization of the negative charges of the phosphate groups on the DNA by the ethidium cations may account for this. This inference is supported by Waring’s observation that high concentrations of monovalent and low concentrations of divalent cations compete with ethidium bromide in binding DNA.
4.3.4 Supplement 26
Current Protocols in Cytometry
Although ethidium bromide may bind to single-stranded polynucleotides, it experiences fluorescence enhancement (25- to 30-fold over free dye) only if the polynucleotide is doublestranded DNA, a DNA-RNA hybrid, or doublestranded RNA. LePecq and Paoletti (1967) reported that in this configuration ethidium bromide can be excited by light of wavelengths between 270 and 370 nm and between 440 and 560 nm. Binding of ethidium bromide to DNA increased its thermal stability and viscosity, and on the basis of flow dichroism data, they believed that the dye intercalated between the bases of its polynucleotide target. Unlike the mode of polynucleotide binding that leads to fluorescence, at very low salt concentrations ethidium bromide may bind electrostatically to DNA or RNA, but this type of binding does not lead to fluorescence enhancement. These results led LePecq and Paoletti to propose that fluorescence emission results from forcing the exocyclic phenyl ring into the plane formed by the phenanthridinium moiety of the ethidium bromide molecule. This concept has been used by numerous workers seeking to synthesize dyes with higher selectivity and greater fluorescence enhancement (see below). Reviewing the characteristics of many drugs that intercalate between the bases of DNA, Waring (1981) pointed out that these follow a “nearest neighbour exclusion rule.” This rule dictates that once the space between a base pair is occupied by a drug molecule, a second molecule cannot bind in the space created by the base pair immediately neighboring the space that is already occupied. The physical basis for this rule may be that the intercalating drug partially unwinds the polynucleotide. Waring proposed that the site immediately following the occupied site may be distorted to the extent that it is no longer accessible to a second drug molecule. Whereas the phenanthridines ethidium bromide and propidium iodide exhibit no base specificity, the third base-intercalating dye, 7aminoactinomycin D (7-AAD), shows weak GC selectivity. In addition, its excitation and emission maxima are red-shifted from those of ethidium bromide and propidium iodide (see Table 4.3.1). This allows its use for simultaneous analysis of cellular DNA and RNA content in combination with pyronin Y (Toba et al., 1995). Neither the phenanthridines nor 7-AAD penetrates intact plasma membranes readily; thus, these dyes can be conveniently used to distinguish viable cells from those with compromised plasma membranes. Ethidium bro-
mide and propidium iodide show some, and 7-AAD strong, selectivity for the differentiation state of stained cells. After treatment of nuclei with hydrochloric acid, which removes histones and other chromatin proteins, the difference in fluorescence intensity between differentiated and undifferentiated cells is lost, but acid-treated samples show a considerable increase in staining (Darzynkiewicz et al., 1984). These results indicate that base-intercalating dyes may be useful for monitoring alterations in chromatin structure during cell differentiation. Because none of the base-intercalating dyes is absolutely specific for DNA, enzymatic digestion of RNA is required to assess nuclear ploidy by measuring cellular fluorescence using these dyes. Recently a novel anthraquinone dye (DRAQ5) has become available for cytometry (Smith et al., 2000). In addition to excitation maxima at 622 and 676 nm, this dye also shows significant excitation at 240 and 314 nm, while it fluoresces maximally at 681 with half-maximal fluorescence at 660.5 and 762.5 nm. Since DRAQ5 fluorescence is red shifted 58 nm relative to propidium iodide, DRAQ5 can be used in combination with a large number of green-, yellow-, and red-fluorescing dyes, including MitoTracker Orange (CMTMRos). Ethanolfixed cells retain DRAQ5 stainability, such that multiparameter cell cycle analysis after cyclin B1 and DRAQ5 staining can be performed (Smith et al., 2000). Currently available data indicate that DRAQ5 carries great promise as a deep-red-fluorescing DNA dye for studies by flow and image cytometry of viable and ethanol-fixed cells. A novel use of the intercalation concept for the development of highly fluorescent nucleic acids stains has been the synthesis of bis-intercalating dyes. A first attempt was the synthesis of ethidium homodimer, which showed an incremental improvement over the parent ethidium bromide dye in fluorescence enhancement after DNA binding (Table 4.3.2). A breakthrough was achieved when a series of bis-intercalating dyes based on thiazole orange congeners was synthesized. This family of dyes, known as dimeric cyanine dyes, consists of dimers of thiazoles and oxazoles in which the monomeric chromophores are linked by a long bis-cationic bridge (Fig. 4.3.1). Their bis-intercalative mode of nucleic acid binding and the bis-cationic linker enable the dimeric cyanine dyes to bind with high affinity to both single-stranded and double-stranded nucleic acids (Rye and Glazer, 1995). Polarized
Molecular and Cellular Probes
4.3.5 Current Protocols in Cytometry
Supplement 26
Table 4.3.2
Optical Properties of Intercalating and Bis-Intercalating Dyesa,b
ε × 103
Fbound/Ffree
QY
6.0 DNA
25-30
ND
Ethidium bromide Ethidium homodimer
5.6 dsDNA 8.9 dsDNA
25-30 35
ND
TO-PRO-1 TOTO-1
62.8 DNA 117.0 DNA
18,900 1,400
0.25 0.34
TO-PRO-3 TOTO-3
102.0 DNA 154.1 DNA
ND ND
0.11 0.06
YO-PRO-1 YOYO-1
52.0 DNA 98.9 DNA
700 3,200
0.44 0.52
Dye name Propidium iodide
ND
aData from Haugland (1996) and Rye et al. (1992). bds, double-stranded; ε, molecular extinction coefficient; F, fluorescence; ND, not determined; QY, quantum yield.
Nucleic Acid Probes
light spectroscopy has revealed that the long axis of the oxazole or thiazole moiety is situated perpendicular to the nucleic acid axis, which is consistent with dye bis-intercalation (Larsson et al., 1994). At high ratios of dye to nucleic acid base, a second, external mode of binding may occur (Larsson et al., 1994). Dimers of thiazole orange are named TOTO; monomers are called TO-PRO. Binding of dimeric cyanine dyes to DNA leads to unwinding of the helix by 60° and consequent increase of the DNA helical repeat to 12 base pairs (Spielman et al., 1995). With the dimeric dye TOTO1, for example, the bis-cationic linker chain is situated in the minor groove and the benzothiazole ring of the TOTO-1 molecule is twisted relative to the quinoline ring after DNA binding (Spielman et al., 1995). Locking of the two chromophore moieties of the TOTO-1 system between the DNA bases, and the consequent fixation of their relative orientation, forces the TOTO-1 molecule to release absorbed energy by fluorescence emission (Spielman et al., 1995). This restriction of chromophore torsion around the methenyl bridge between the benzothiazole and quinoline moieties accounts for the large fluorescence enhancement found with the TOTO-1 dye and other members of the dimeric cyanine dye family (Netzel et al., 1995; Spielman et al., 1995). Inspection of Table 4.3.2, however, reveals that the dimer of a yellow oxazole (YOYO-1) exhibits increased fluorescence enhancement relative to the monomeric YO-PRO-1. In the case of the TOPRO-1 and TOTO-1 pair (monomer and dimer respectively), the opposite relationship holds. A confounding factor that may account for this unexpected behavior of the TO-PRO-1 and
TOTO-1 pair of dyes could be the fact that Förster energy transfer may occur between the bis-intercalated chromophores of the TOTO-1 dye (Larsson et al., 1994); the energy thus lost will not be detected as fluorescence. Therefore the possible contribution of Förster energy transfer between chromophores within a dye molecule should be considered in the design of novel fluorochromes. The TOTO and YOYO dyes have been used successfully for cell cycle and cytogenetic analyses in a conventional flow cytometry system (Hirons et al., 1994). TOTO-3 was found to suit an experimental flow cytometry system equipped with a laser diode as excitation source (Doornbos et al., 1994). In both systems fluorescence emission was stoichiometric with DNA content. Because dimeric cyanine dyes do not pass through intact cell membranes, the staining buffer must contain a detergent. It is likely that shear forces upon the detergenttreated cells may have removed most of the cytoplasm, reducing the possible contribution of cytoplasmic RNA to the total cellular fluorescence emission. To date no studies have been published that assess this possible contribution of RNA. Nevertheless the procedures described by Hirons and coworkers (1994) and by Doornbos and coworkers (1994) indicate the usefulness of dimeric cyanine dyes for cell cycle studies. In contrast to the Hoechst 33342 and 33258 dyes TO-PRO-3 was found to show enhanced fluorescence after staining of 5-bromodeoxyuridine-containing DNA (Beisker et al., 1999). Although the physicochemical basis of this fluorescence enhancement is not known, this characteristic dye has been used success-
4.3.6 Supplement 26
Current Protocols in Cytometry
methenyl bridge
S CH
N
CH3
N+ CH3
benzothiazole moiety
quinoline moiety
Figure 4.3.1 Thiazole orange.
fully to distinguish cells that have incorporated 5-bromodeoxyuridine from cells that did not proliferate during a labeling period. Other cyanine-dimer dyes (such as TO-PRO-5 and YOYO-3) show little fluorescence enhancement and some (e.g., TO-PRO-1, YO-PRO-3) show no fluorescence enhancement after staining of 5-bromodeoxyuridine-containing DNA. In combination with Hoechst 33258, TO-PRO3 show excellent resolution between cells that have undergone zero, one, or two rounds of DNA replication (M. Poot, unpub. observ.). TO-PRO-3 fluorescence is 44-nm red shifted relative to propidium iodide, which allows its use in experiments that involve two-color immunostaining in conjunction with Hoechst fluorescence–based resolution of proliferating and non-proliferating cells, similar to the method developed by Rosato et al. (2001).
DYES WITH A DUAL-BINDING MODE The discovery that an asymmetric dye based upon the fusion of a benzothiazole moiety to a quinoline system via a methenyl bridge leads to a brightly fluorescent nucleic stain (Lee et al., 1986) inspired the synthesis of a series of monomeric cyanine dyes. The original dye of this type, thiazole orange (Fig. 4.3.1), has an excitation maximum at 509 nm and an emission maximum at 533 nm. The rather short Stokes shift of thiazole orange is of little concern, because the dye has a broad range of excitation, which tails into the blue part of the visible spectrum, and a high molar absorption coefficient after binding to RNA (Lee et al., 1986). These characteristics allow excitation of the dye bound to nucleic acids with the 488 and 514 nm lines of an argon laser. By systematic variation of the components that make up thiazole orange, a series of dyes
with even higher molar absorption coefficients but retaining favorable excitation and emission ranges has been developed. Although there has been no detailed investigation of the mode of polynucleotide binding of thiazole orange or any of the monomeric cyanine dyes, one can surmise that these dyes may exhibit features found with the dimeric cyanine dyes. Thus, the benzoxazole or benzothiazole moiety may bind to the minor groove of the polynucleotide, whereas the other moiety of the dye molecule may intercalate between the bases. Binding to the nucleic acid would force the methenyl bridge between the two chromophores into a fixed orientation, resulting in strongly enhanced fluorescence, whereas the relative flexibility of the dye allows comparatively nonselective binding of the dye with high affinity. In addition, the integration of the two chromophores and the methenyl bridge into a single fluorophore would red-shift the excitation and emission maxima of the resulting dyes relative to those of the component chromophores. This hypothetical scheme for nucleic acid binding of the monomeric cyanine dyes is in agreement with existing physical data, but needs experimental support. For now, the model outlined indicates a dual mode of nucleic-acid binding for this group of dyes. Neither thiazole orange nor the novel monomeric cyanine dyes have been systematically investigated for the relative contribution of DNA and RNA toward their fluorescence emission; little is known about their selectivity toward one or the other polynucleotide. Monomeric cyanine dyes do not penetrate intact plasma membranes, yet results published by several investigators (Doornbos et al., 1994; Hirons et al., 1994; Van Hooidonk et al., 1994) indicate that some monomeric cyanine dyes
Molecular and Cellular Probes
4.3.7 Current Protocols in Cytometry
Supplement 26
may be useful for assessing cell cycle distributions of nucleic acids.
NUCLEIC ACID–PRECIPITATING AND METACHROMATIC DYES
Nucleic Acid Probes
In the discussion of the bis-benzimidazoles and ethidium bromide, precipitation of the dye−nucleic acid complex was noted as an untoward side effect. Precipitation of such complexes can be of practical use, however. The tendency of a dye to precipitate after binding gives information regarding the denaturability of the nucleic acid target. In addition, spectral properties of the dye–nucleic acid complex may change if the nucleic acid denatures or the complex precipitates. Two dyes, acridine orange and pyronin Y, derive their usefulness from properties of this kind. Specific methods involving these dyes are presented in Chapter 7 of this volume; the discussion here is limited to their physicochemical characteristics. After staining with acridine orange, the spectrum of fluorescence emission depends upon the strandedness of the nucleic acid: double-stranded nucleic acids give rise to fluorescence in the green region of the spectrum (maximum at 530 nm), and single-stranded nucleic acids show red fluorescence emission (maximum at 640 nm). Treating samples such that all RNA is single stranded, while retaining DNA in a double-stranded conformation, followed by staining with acridine orange indicates simultaneously the cell cycle stage (green fluorescence) and total cellular RNA content (red fluorescence) of each cell in the sample. Flow cytometric analysis of acridine orange–stained samples revealed that during lymphocyte activation the amount of RNA increases before cells enter the S phase of the cell cycle (Darzynkiewicz et al., 1976). Likewise, the small keratinocytes in the basal layer of the skin have to enter a narrow window of increased RNA content and cell size to be competent for cell proliferation and to initiate DNA synthesis (Poot et al., 1990). The physicochemical basis of the spectral difference in fluorescence emission after acridine orange staining has been elucidated (Darzynkiewicz et al., 1983). Ribosomal RNA stained with acridine orange at concentrations ranging from 4 to 20 µM shows red emission and a strong increase in light scatter, whereas with DNA this phenomenon is observed with dye concentrations ranging from 10 to 50 µM (Darzynkiewicz et al., 1983). To obtain the same effect in cells, higher dye concentrations have to be used. In the presence of EDTA the
phenomenon occurs at lower dye concentrations, whereas extraction of basic proteins with 80 mM hydrochloric acid reduces the dye concentration at which DNA fluoresces red. Therefore, differences in denaturability of the nucleic acid target determine the emission spectrum after acridine orange staining. Complexes of dye with single-stranded nucleic acids concomitantly precipitate (Darzynkiewicz et al., 1983). Additional support for the contention that red emission after acridine orange staining corresponds to denaturation of the nucleic acid target has been obtained with DNA conformation probes such as formaldehyde and diethyl pyrocarbonate (Kapuscinski and Darzynkiewicz, 1983); the ability to denature doublestranded nucleic acids is apparently common to all molecules that intercalate between the bases of the nucleic acid target. These results of are in agreement with the model proposed by Waring (1981). Pyronin Y shows a different pattern of interaction with nucleic acids. Depending on whether the polynucleotide is DNA or RNA and on its base composition, pyronin Y can cause it to denature and precipitate; precipitates of pyronin Y with single-stranded nucleic acids are nonfluorescent, but can be detected by their visible color (Kapuscinski and Darzynkiewicz, 1987). The range of pyronin Y concentrations at which nucleic acid−dye complex precipitation occurs varies by almost two orders of magnitude, with poly(rA) being the most sensitive and calf thymus DNA the least (Darzynkiewicz at al., 1987). At low concentrations of pyronin Y, precipitation of dye−nucleic acid complexes does not occur and fluorescence emission is obtained. The quantum yield of fluorescent emission of pyronin Y–nucleic acid complexes depends upon the base composition of the nucleic acid (Kapuscinski and Darzynkiewicz, 1987). As a result, the amount of fluorescence after pyronin Y staining is not stoichiometric with the amount of nucleic acid present. This characteristic violates the third requirement for a dye to be useful in cytometry. Notwithstanding the fact that pyronin Y cannot be used to quantify any nucleic acid, the dye has been applied successfully in flow cytometry. Brachet observed that the fluorescence of pyronin Y in complexes with DNA is quenched in the presence of methyl green (Brachet, 1940). Later, Hoechst 33258 was found to exhibit a similar characteristic, and a method that uses this pair of dyes has been devised for simultaneous detection of DNA and RNA by flow cytometry
4.3.8 Supplement 26
Current Protocols in Cytometry
(Crissman et al., 1985). This method, although it does not allow the quantitative analysis of cellular RNA content, gives some insight into the cell cycle. Mitotic cells exhibit less fluorescence with pyronin Y than do G2-phase cells (Darzynkiewicz et al., 1987). This result was explained by taking into account that polyribosomes dissociate during mitosis, and therefore much of the cellular RNA is in a singlestranded (nonfluorescent) conformation (Darzynkiewicz et al., 1987). Thus, the G2 and mitotic phases of the cell cycle can be resolved by simultaneous staining of cells with Hoechst 33258 and pyronin Y. Because pyronin Y does not penetrate intact plasma membranes, this method requires prior cell fixation. Another complicating factor is that the method requires dual-laser excitation (UV and 488 nm); the fluorescence quantum yield of pyronin Y is so low that mercury arc lamp excitation is not sufficient (Darzynkiewicz et al., 1987). Advantages of pyronin Y/Hoechst over the acridine orange method for simultaneous analysis of cellular DNA and RNA are high sensitivity for RNA detection, specificity of each dye for each nucleic acid, and high accuracy for the quantitation of DNA (Darzynkiewicz et al., 1987). In addition, pyronin Y provides important information regarding the conformation of RNA (Traganos et al., 1988).
CONCLUDING REMARKS The preceding sets the stage for selection of nucleic acid stains to suit the particular purpose(s) of the reader. New dyes are being developed continuously. Because of their novelty, their physicochemical properties and usefulness for cytometry are only starting to be characterized. Novel dyes carry great promise for the future.
Darzynkiewicz, Z., Evenson, D., Kapuscincki, J., and Melamed, M. 1983. Denaturation of RNA and DNA in situ induced by acridine orange. Exp. Cell Res. 148:31-46. Darzynkiewicz, Z., Traganos, F., Kapuscincki, J., Staiano-Coico, L., and Melamed, M. 1984. Accessibility of DNA in situ to various fluorochromes: Relationship to chromatin changes during erythroid differentiation of Friend leukemia cells. Cytometry 5:355-363. Darzynkiewicz, Z., Kapuscinski, J., Traganos, F., and Crissman, H. 1987. Application of pyronin Y(G) in cytochemistry of nucleic acids. Cytometry 8:138-145. Doornbos, R., De Grooth, B.G., Kraan, Y.M., Van Der Poel, C.J., and Greve, J. 1994. Visible diode lasers can be used for flow cytometric immunofluorescence and DNA analysis. Cytometry 15:267-271. Harshman, K.D. and Dervan, P.B. 1985. Molecular recognition of B-DNA by Hoechst 33258. Nucl. Acids Res. 13:4825-4835. Haugland, R.P. 1996. Handbook of Fluorescent Probes and Research Chemicals, 6th ed. Molecular Probes, Eugene, Ore. Hirons, G.T., Fawcett, J.J., and Crissman, H.A. 1994. TOTO and YOYO: New very bright fluorochromes for DNA content analyses by flow cytometry. Cytometry 15:129-140. Kapuscinski, J. and Darzynkiewicz, Z. 1983. Increased accessibility of bases in DNA upon binding of acridine orange. Nucl. Acids Res. 11:75557568. Kapuscinski, J. and Darzynkiewicz, Z. 1987. Interactions of pyronin Y(G) with nucleic acids. Cytometry 8:129-137. Kapuscinski, J. and Skoczylas, B. 1978. Fluorescent complexes of DNA with DAPI 4′,6-diamidino2-phenylindole⋅2 H Cl o r DCI 4′,6-dicarboxyamide-2-phenylindole. Nucl. Acids Res. 5:3775-3799.
LITERATURE CITED
Larsson, A., Carlsson, C., Johnsson, M., and Albinsson, B. 1994. Characterization of the binding of the fluorescent dyes YO and YOYO to DNA by polarized light spectroscopy. J. Am. Chem. Soc. 116:8459-8465.
Beisker, W., Weller-Mewe, E.M., and Nusse, M. 1999. Fluorescence enhancement of DNAbound TO-PRO-3 by incorporation of bromodeoxyuridine to monitor cell cycle kinetics. Cytometry 37:221-229.
Latt, S.A. and Stetten, G. 1976. Spectral studies of 33258 Hoechst and related bisbenzimidazole dyes useful for fluorescent detection of deoxyribonucleic acid synthesis. J. Histochem. Cytochem. 24:24-33.
Brachet, J. 1940. La détection histochimique des acides pentose nucléiques. Comptes Rend. Soc. Biol. 133:88-90.
Lee, L.G., Chen, C.H., and Chu, L.A. 1986. Thiazole orange: A new dye for reticulocyte analysis. Cytometry 7:508-517.
Crissman, H., Darzynkiewicz, Z., Tobey, R.A., and Steinkamp, J.A. 1985. Correlated measurements of DNA, RNA and protein in individual cells by flow cytometry. Science 228:1321-1324.
LePecq, J.B. and Paoletti, C. 1967. A fluorescent complex between ethidium bromide and nucleic acids. Physicochemical characterization. J. Mol. Biol. 27:87-106.
Darzynkiewicz, Z., Traganos, F., Sharpless, T.K., and Melamed, M. 1976. Lymphocyte stimulation. A rapid multiparameter analysis. Proc. Natl. Acad. Sci. U.S.A. 73:2881-2884.
Molecular and Cellular Probes
4.3.9 Current Protocols in Cytometry
Supplement 26
Loontiens, F.G., Regenfuss, P., Zechel, A., Dumortier, L., and Clegg, R.M. 1990. Binding characteristics of Hoechst 33258 with calf thymus DNA, poly[d(A-T)], and d(CCGGAATTCCGG): Multiple stoichiometries and determination of tight binding with a wide spectrum of site affinities. Biochemistry 29:9029-9039. Loontiens, F.G., McLaughlin, L.W., Diekman, S., and Clegg, R.M. 1991. Binding of Hoechst 33258 and 4′,6-diamidino-2-phenylindole to self-complementary decadeoxynucleotides with modified exocyclic base substituents. Biochemistry 30:182-189. Netzel, T.M., Nafisi, K., Zhao, M., Lenhard, J.R., and Johnson, I. 1995. Base-content dependence of emission enhancements, quantum yields, and lifetimes for cyanine dyes bound to doublestranded DNA: Photophysical properties of monomeric and bischromophoric DNA stains. J. Phys. Chem. 195:17936-17947. Pjura, P.E., Grzeskowiak, K., and Dickerson, R.E. 1987. Binding of Hoechst 33258 to the minor groove of B-DNA. J. Mol. Biol. 197:257-271. Poot, M., Rizk-Rabin, M., Hoehn, H., and Pavlovitch, J.H. 1990. Cell size and RNA content correlate with cell differentiation and proliferative capacity of rat keratinocytes. J. Cell Physiol. 143:279-286. Rabinovitch, P.S. 1983. Regulation of human fibroblast growth rate by both noncycling cell fraction and transition probability is shown by growth in 5-bromodeoxyuridine followed by Hoechst 33258 flow cytometry. Proc. Natl. Acad. Sci. U.S.A. 80:2951-2955. Rabinovitch, P.S., Kubbies, M., Chen, Y.C., Schindler, D., and Hoehn, H. 1986. BrdU-Hoechst flow cytometry: A unique tool for quantitative cell cycle analysis. Exp. Cell Res. 174:309-318. Rosato, M.T., Jabbour, A.J., Ponce, R.A., Kavanagh, T.J., Takaro, T.K., Hill, J.P., Poot, M., Rabinovitch, P.S., and Faustman, E.M. 2001. Simultaneous analysis of surface marker expression and cell cycle progression in human peripheral blood mononuclear cells. J. Immunol. Methods 256:35-46.
Nucleic Acid Probes
Searle, M.S. and Embrey, K.J. 1990. Sequence-specific interaction of Hoechst 33258 with the minor groove of an adenine-tract DNA duplex studied in solution by 1H NMR spectroscopy. Nucl. Acids Res. 18:3753-3767. Smith, P.J., Blunt, N., Wiltshire, M., Hoy, T., Teesdale-Spittle, P., Craven, M.R., Watson, J.V., Amos, W.B., Errington, R.J., and Patterson, L.H. 2000. Characteristics of a novel deep red/infrared fluorescent cell-permeant DNA probe, DRAQ5, in intact human cells analyzed by flow cytometry, confocal and multiphoton microscopy. Cytometry 40:280-291. Spielman, H.P., Wemmer, D.E., and Jacobsen, J.P. 1995. Solution structure of a DNA complex with the fluorescent bis-intercalator TOTO determined by NMR spectroscopy. Biochemistry 34:8542-8553. Stokke, T. and Steen, H.B. 1985. Multiple binding modes for Hoechst 33258 to DNA. J. Histochem. Cytochem. 33:333-338. Stokke, T. and Steen, H.B. 1986. Binding of Hoechst 33258 to chromatin in situ. Cytometry 7:227234. Tanious, F.A., Veal, J.M., Buczak, H., Ratmeyer, L.S., and Wilson, W.D. 1992. DAPI (4′,6diamidino-2-phenylindole) binds differently to DNA and RNA: Minor-groove binding at AT sites and intercalation at AU sites. Biochemistry 31:3103-3112. Toba, K., Winton, E.F., Koike, T., and Shibata, A. 1995. Simultaneous three-color analysis of the surface phenotype and DNA-RNA quantitation using 7-aminoactinomycin D and pyronin Y. J. Immunol. Methods 182:193-207. Traganos, F., Crissman, H., and Darzynkiewicz, Z. 1988. Staining with pyronin Y detects changes in conformation of RNA during mitosis and hyperthermia of CHO cells. Exp. Cell Res. 179:535544. Van Hooidonk, C.A.E.M., Glade, C.P., and Van Erp, P.E.J. 1994. TO-PRO-3 iodide: A novel HeNe laser-excitable DNA stain as an alternative for propidium iodide in multiparameter flow cytometry. Cytometry 17:185-189.
Rye, H.S. and Glazer, A.N. 1995. Interaction of dimeric intercalating dyes with single-stranded DNA. Nucl. Acids Res. 23:1215-1222.
Waring, M.J. 1965. Complex formation between ethidium bromide and nucleic acids. J. Mol. Biol. 13:269-282.
Rye, H.S., Yue, S., Wemmer, D.E., Quesada, M.A., Haugland, R.P., Mathies, R.A., and Glazer, A.N. 1992. Stable fluorescent complexes of doublestranded DNA with bis-intercalating asymmetric cyanine dyes: Properties and applications. Nucl. Acids Res. 20:2803-2812.
Waring, M.J. 1981. DNA modification and cancer. Annu. Rev. Biochem. 50:159-192.
Contributed by Martin Poot University of Washington Seattle, Washington
Dr. Poot wishes to acknowledge the support of Molecular Probes, Inc., in Eugene, Oregon, where he developed this material.
4.3.10 Supplement 26
Current Protocols in Cytometry
Cellular Function Probes Flow cytometry provides statistical information on cellular function parameters measured at the single-cell level, allowing analysis of uniformity, in addition to magnitude, across an entire cell population. A fundamental requirement in this process, the selective detection of a single function amidst a multitude of others, is fulfilled by the use of fluorescent probes. How probes currently in use perform this role is the central theme of this article. Of particular concern are two properties that primarily determine the utility of a probe: the dependence of its fluorescence emission spectrum and/or quantum yield on environmental factors, and the level of probe uptake and distribution within the cell. The wide range of accessible functional parameters adds a powerful extra dimension to flow cytometric analysis, enabling simultaneous measurement and correlation of multiple parameters, and ultimately affording a more comprehensive view of complex biological processes. For example, the process of programmed cell death (apoptosis) has many functional expressions that can be detected using currently available probes (Table 4.4.1). Furthermore, multiparameter flow cytometry is ideally suited for exploration of the temporal and mechanistic connections between signal transduction and functional expression processes (Simons, 1993).
PROBES FOR INTRACELLULAR IONS Ca2+, Mg2+, and K+ Indicators An extensive family of fluorescent indicators for intracellular Ca2+ based on a set of design concepts originated by Tsien (1980) has Table 4.4.1
UNIT 4.4
been developed over the last 15 years (Haugland, 1996a). The common elements of these indicators are: (1) the BAPTA chelator, with Ca2+ binding affinity matching typical intracellular concentrations, high selectivity for Ca2+ over Mg2+, and low pH-sensitivity, and (2) intracellular delivery in the form of membrane-permeant acetoxymethyl (AM) ester derivatives (Fig. 4.4.1). Existing indicators are differentiated primarily by their spectral characteristics, which determine compatibility with various instrument configurations, and their Ca2+ dissociation constants (Kd), which determine the accessible Ca2+ concentration measurement range. For flow cytometry, the optimum spectral characteristics are strong absorption at 488 nm and a Ca2+-dependent emission wavelength shift. Unfortunately, development of a single indicator possessing these characteristics has proved elusive (Table 4.4.2). The dual-emissio n respo nses exhibited by indo-1 (Grynkiewicz et al., 1985) and the combination of fluo-3 and Fura Red (Fig. 4.4.2) enable a Ca2+-dependent ratio of fluorescence intensities measured at two different wavelengths to be calculated, resulting in cancellation of artifactual fluorescence signal variations that might otherwise be misinterpreted as changes in Ca2+ concentration. Sources of artifact include cell size variations and differences in the intracellular indicator concentration in the experimental cell population, as well as instrumental factors such as excitation source instability. Disadvantages of indo-1 include the requirement for UV excitation, photodegradation (Scheenen et al., 1996), and shifts in the Kd value and emission spectra caused by interac-
Functional Indicators of Apoptosis
Functiona
Probe
Phosphatidylserine exposure Loss of mitochondrial membrane potential
Fluorescein-labeled annexin V DiO6(3), CMXRos
Increased O2− production Intracellular acidification
Dihydroethidium, DHR123, DCFH-DA SNARF-1
Glutathione depletion Protease activation
mBBr, mBCl Rhodamine 110 bisamide
aThis listing excludes chromatin condensation and DNA fragmentation, both of which are detected using nucleic acid probes (UNIT 4.3).
Contributed by Iain D. Johnson Current Protocols in Cytometry (1997) 4.4.1-4.4.17 Copyright © 1997 by John Wiley & Sons, Inc.
Molecular and Cellular Probes
4.4.1 Supplement 2
O
O
O
(CH3COCH2 OCCH2)2N
O
N(CH2COCH2 OCCH3)2 OCH2 CH2O
NH
CH3 nonpolar, Ca2+-insensitive
COCH2 OCCH3 O
O
cell membrane esterase O
O
(−OCCH2)2N
OCH2 CH2O
NH
CH3
CO− O
N(CH2CO−)2
O
+
5HCH
polar, Ca2+-sensitive O
+
5CH3 COH
Figure 4.4.1 Schematic diagram of the processes involved in loading cells using membrane-permeant acetoxymethyl (AM) ester derivatives of polar fluorescent indicators, exemplified by indo-1. Note the generation of potentially toxic byproducts (formaldehyde and acetic acid). The BAPTA chelator responsible for Ca2+ binding is indicated by shading.
Cellular Function Probes
tions with proteins and other environmental factors (Owen et al., 1991; Hove-Madsen and Bers, 1992). Fluo-3 (Minta et al., 1989) is an indicator with close to optimal excitation at 488 nm, facilitating combined analysis of intracellular Ca2+ and immunofluorescence detection of cell surface antigens using R-phycoerythrin (Greimers et al., 1996). Fluo-3 lacks the emission ratio detection capability of indo-1 (Fig. 4.4.2), however, and separate Ca2+ response calibrations are needed for each experimental sample to control for variations in intracellular indicator concentration (Vandenberghe and Ceuppens, 1990). In addition, unambiguous identification of heterogeneous Ca2+ responses within a cell population is more difficult with fluo-3 than with indo-1. Like fluo-3, the Calcium Green and Oregon Green 488 BAPTA indicator families developed by Molecular Probes (Haugland, 1996a) exhibit fluorescence increases but no wavelength shift upon binding Ca2+. These indicators differ from fluo-3 and from each other primarily in terms of Kd (Table 4.4.2) and the dynamic range of the Ca2+-dependent fluores-
cence increase. The absorption maxima of the new Oregon Green 488 BAPTA indicators are closer to 488 nm than those of fluo-3 and the Calcium Green series. Co-loading two indicators allows one to obtain a Ca2+-dependent dual emission response with excitation at 488 nm. Suitable indicator pairs consist of: (1) fluo-3, or a Calcium Green or Oregon Green 488 BAPTA indicator, and (2) a Ca2+-sensitive (Fura Red; June and Rabinovitch, 1994; Novak and Rabinovitch, 1994) or Ca2+-insensitive (SNARF-1; Rijkers et al., 1990; Spencer and Berlin, 1995) indicator with emission at wavelengths >570 nm (Fig. 4.4.2). The validity of these methods depends on consistency of the relative intracellular concentrations of the two indicators from cell to cell, a requirement that is not always fulfilled when loading with fluo-3 and Fura Red AM esters (Floto et al., 1996). Note that it is usual to load fluo-3 and Fura Red at a nominal concentration ratio of at least 1:2 to compensate for the much lower fluorescence quantum yield of the latter. Other preconditions, such as equivalent intracellular distributions and Ca2+
4.4.2 Supplement 2
Current Protocols in Cytometry
binding kinetics for the two indicators, appear to be more generally valid (Lipp and Niggli, 1993; Floto et al., 1996). Replacement of the tetracarboxylate BAPTA chelator by the tricarboxylate APTRA yields indicators for detection of intracellular Mg2+ (typical levels are ∼1 mM) as well as for elevated Ca2+ concentrations up to ∼500 µM Table 4.4.2
(Tse and Hille, 1994). The spectroscopic properties of these indicators are essentially similar to those of their BAPTA counterparts. Thus Mag-indo-1 [with a Kd (Mg2+) of 2.7 mM] has a dual-emission response to ion binding, and can be used for flow cytometric detection of Mg2+ in the same way as indo-1 for Ca2+ (Rijkers and Griffioen, 1993). For excitation at
Characteristics of Intracellular Ca2+ Indicators Used for Flow Cytometrya
Indicator
EX (nm)
EM− (nm)
EM+ (nm)
Modeb
Kd (nM)c
Indo-1 Fluo-3d
351-364 488
475 None
401 526
EM+/EM− EM+
230 390
Calcium Green-1d Calcium Green-2
488 488
531 536
531 536
EM+ EM+
190 550
Calcium Green-5N Oregon Green 488 BAPTA-1e
488 488
532 523
532 523
EM+ EM+
14,000 170
Oregon Green 488 BAPTA-2e Oregon Green 488 BAPTA-5Ne
488 488
523 521
523 521
EM+ EM+
580 20,000
Fura Red Fluo-3 + Fura Redf
488 488
657 657
637 526
EM− EM+/EM−
140 380
a Abbreviations: EX, fluorescence excitation wavelength; EM−, fluorescence emission maximum in the absence of Ca2+; EM+, fluorescence emission maximum in presence of Ca2+ (all in nm). bCa2+ detection modes: EM+/EM−, ratio of fluorescence intensities at wavelengths EM+ and EM−; EM+, fluorescence increase at wavelength EM+; EM−, fluorescence decrease at wavelength EM−. cCa2+ dissociation constant, generally measured in vitro at 22°C in 100 mM KCl/10 mM MOPS, pH 7.2. Values depend
on temperature, ionic strength, pH, and other factors, and are often significantly higher in vivo. dCalcium Green 1 is more fluorescent than fluo-3 in both Ca2+-bound and Ca2+-free forms. The magnitude of the Ca2+-dependent fluorescence increase is greater for fluo-3. eThe molar absorptivities of the Oregon Green BAPTA 488 indicators at 488 nm are approximately twice those of the corresponding Calcium Green indicators. fK value determined in vitro, as reported by Lipp and Niggli (1993). See Floto et al. (1996) for discussion of technical d complexities involved in calibrating the Ca2+ response of this indicator pair.
Fluorescence emission
A
B 39.8 µM free Ca2+ 1.35 0.60 0.35 0.23 0.15 0.10 0.065 0.038 0.017
500
550 600 Wavelength (nm)
0 µM free Ca2+ 0.017 0.038 Fluo-3 0.065 0.10 0.15 0.23 0.35 0.60 1.35
Fura Red
39.8 650 500
550 600 650 Wavelength (nm)
700
Figure 4.4.2 (A) The Ca2+-dependent fluorescence quantum yield increase of fluo-3 is not accompanied by a spectral shift. (B) Emission spectra of a mixture (1:10 mol/mol) of fluo-3 and Fura Red excited at 488 nm showing the counterbalanced responses of the two probes to addition of Ca2+. Excitation wavelength is 488 nm in both panels.
Molecular and Cellular Probes
4.4.3 Current Protocols in Cytometry
Supplement 2
488 nm, Magnesium Green [Kd (Mg2+) of 1.0 mM] is analogous to the Ca2+ indicator Calcium Green-1. The crown ether chelators used in indicators for Na+ and K+ are somewhat less selective than the BAPTA chelator is for Ca2+. Nevertheless, SBFI (Minta and Tsien, 1989) has proved to be a valuable indicator for intracellular Na+, although its spectroscopic response to Na+ is analogous to the Ca2+ indicator fura-2 (Grynkiewicz et al., 1985) rather than indo-1; it is therefore not generally suitable for flow cytometric applications. Sodium Green, a Na+ indicator suitable for 488 nm excitation, has recently been developed by Molecular Probes. Flow cytometric measurements of Na+ in CHO cells using Sodium Green have been validated by correlation with spectrofluorometric measurements using SBFI (Amorino and Fox, 1995). In some cell types, however, it appears that the Na+-dependent fluorescence increase of Sodium Green is severely attenuated, probably as a result of binding to proteins.
Indicators for Intracellular pH
Cellular Function Probes
The major intracellular pH indicators for flow cytometric applications are BCECF and SNARF-1 (Boyer and Hedley, 1994), both of which exhibit spectral shifts upon deprotonation of a phenolic hydroxyl group with pKa values within the normal cytosolic pH range (7.0 for BCECF, 7.5 for SNARF-1). For both indicators, cell loading can be accomplished using membrane-permeant ester derivatives in the same way as for Ca2+ indicators (see Fig. 4.4.1). The pH-dependent ratio measurement of BCECF fluorescence intensities excited at 440 and 490 nm that is often used for microscopy and spectrofluorometry is difficult to implement in flow cytometry. Instead, ratios of fluorescence emission intensities at 525 and 640 nm (excited at 488 nm) can be used (Wang et al., 1990; Boyer and Hedley, 1994; Chow et al., 1996), even though the emission spectrum of BCECF exhibits much smaller pH-dependent wavelength shifts than the excitation spectrum. In contrast, the emission spectra of SNARF1 are strongly pH dependent, with a peak shift from 585 nm to 635 nm on deprotonation (Whitaker et al., 1991), resulting in much larger ratio changes than with BCECF (Van Erp et al., 1991; Chow et al., 1996). As in the case of indo-1 (see above), the extent of the emission shift and the dissociation constant (pKa) value for SNARF-1 may be significantly modified by environmental factors (Owen, 1992; Owen et
al., 1992). Disadvantages of SNARF-1 relative to BCECF are primarily its lower fluorescence quantum yield and higher pKa. 1,4-dihydroxyphthalonitrile (DHPN) is a useful flow cytometric pH indicator in experiments requiring emission at shorter wavelengths than BCECF or SNARF-1. The pH-dependent emission spectral shift of DHPN permits ratio detection at 512/455 nm (Kurtz and Balaban, 1985); disadvantages of DHPN include the requirement for UV excitation, rapid extracellular leakage, and a relatively high pKa value (∼8.0).
MEMBRANE POTENTIAL PROBES Probes used for flow cytometric detection of membrane potential can be divided into three primary types: carbocyanines, oxonols, and rhodamine/rosamine dyes (Fig. 4.4.3). All three types operate via potential-dependent changes in their transmembrane distribution and are suitable for detecting changes in average membrane potentials of nonexcitable cells caused by respiratory activity, ion channel permeability, receptor activation, drug binding, and other factors. A detailed review of the response mechanism of these dyes has recently been published (Plášek and Sigler, 1996). In most cases, it is difficult to discriminate between multiple sources of potential-dependent uptake (e.g., plasma membrane versus mitochondrial membrane potentials) and other staining processes that are not potential dependent (Shapiro, 1994).
Carbocyanines Membrane hyperpolarization results in increased cellular uptake of moderately lipophilic, cationic carbocyanine dyes, such as DiOC6(3) (Fig. 4.4.3) and DiOC5(3). Flow cytometric protocols utilize low extracellular staining concentrations (<0.1 µM), thereby minimizing the usual toxic effects associated with carbocyanine dyes (Shapiro, 1994; Petit et al., 1996). Staining at low extracellular dye concentrations predominantly reflects the mitochondrial membrane potential rather than that of the plasma membrane (Wilson et al., 1985; Petit et al., 1996). DiOC6(3) has almost ideal spectral characteristics for 488 nm excitation (εmax = 154,000 cm−1 M−1 at 484 nm). For other excitation sources, such as the 633nm helium-neon laser, cyanine dyes with longer wavelength absorption characteristics [e.g., DiIC1(5); εmax = 238,000 cm−1 M−1 at 638 nm] can be substituted. Potential-dependent accumulation of the
4.4.4 Supplement 2
Current Protocols in Cytometry
A
B
O
Cl
O CH
CH
CH 3
CH 3
CH 2
CH 2 N
Cl
N+
N
Cl
N
CH
CH
CH
CH
N+
N
(CH 2)5
(CH 2)5
CH 2
CH 2
CH 3
CH 3
CH 3
CH 3
Cl
C
D
CH 3(CH 2)3
O−
O
N
(CH 2)3CH 3
H 2N
+
NH2
O
N
O
CH
CH
CH
O
N O
CH 3(CH 2)3
O
E
N
C
(CH 2)3CH 3
O
OCH3
F N
O
N+ +
(CH3)2N
N (CH2)8
N(CH3)2
CH 3
CH2Cl
Figure 4.4.3 Membrane potential probes and mitochondrial stains. (A) DiOC6(3); (B) JC-1; (C) DiBAC4(3); (D) rhodamine 123; (E) CMXRos; (F) nonyl acridine orange.
carbocyanine dye JC-1 (Fig. 4.4.3) by mitochondria results in a pronounced emission wavelength shift (∼527 nm to 590 nm) due to formation of J-aggregates (Chen and Smiley, 1993). The ratio of green to red JC-1 fluorescence excited at 488 nm can be used, therefore, to detect mitochondrial membrane potential changes (Cossarizza et al., 1993). MitoTracker Green FM (Molecular Probes) is a newly developed carbocyanine dye designed primarily as an alternative to rhodamine 123 (see below) for mitochondrial staining. The absorption maximum of MitoTracker Green FM at 490 nm is ideally suited for 488 nm excitation; its emission spectrum is blueshifted relative to that of rhodamine 123, resulting in less signal spillover in applications involving simultaneous detection of immunofluorescence labels such as R-phycoerythrin. Furthermore, as with the rosamine derivatives CMTMRos and CMXRos (Fig. 4.4.3), MitoTracker Green staining persists through fixation and permeablization treatments.
Oxonols Oxonols, of which DiBAC4(3) is the most widely used in flow cytometric applications, are lipophilic anions (Fig. 4.4.3). Membrane hyperpolarization results in decreased cellular staining by oxonols, in contrast to the increases shown by cationic carbocyanines. Oxonols are more specific indicators of plasma membrane potential than carbocyanines, because their anionic character results in exclusion from mitochondria. Oxonols are superior to carbocyanines and rhodamines for discrimination of bacterial viability on the basis of membrane potential (Mason et al., 1995). DiBAC4(3) absorption (εmax = 123,000 cm−1 M−1 at 493 nm) is well matched to the 488-nm argon ion laser line; as in the case of carbocyanines, longerwavelength analogs (e.g., DiBAC4(5), εmax = 160,000 cm−1 M−1 at 590 nm) are available for use with longer-wavelength excitation sources (Mandler et al., 1988). Molecular and Cellular Probes
4.4.5 Current Protocols in Cytometry
Supplement 2
Rhodamines and Rosamines Rhodamine 123, with spectral properties similar to fluorescein, is a lipophilic cation (Fig. 4.4.3) with long-established utility for staining mitochondria in living cells (Chen, 1989). Because staining is potential dependent, it directly reflects cell viability (Kaprelyants and Kell, 1992), as well as enabling discrimination of cell populations based on respiratory activity levels (Skowronek et al., 1990; Spangrude and Johnson, 1990). Rhodamine 123 staining of carcinoma cells is particularly intense (Summerhayes et al., 1982), probably as the result of abnormal plasma and mitochondrial membrane potentials (Davis et al., 1985). Enhanced efflux of rhodamine 123 via the P-glycoprotein transporter has led to its widespread use in functional assays for multidrug resistance (Ludescher et al., 1991; Broxterman et al., 1996). Differences in the influx rates of the dye between resistant and sensitive cells have been observed in some cases (Dordal et al., 1995) but not in others (Altenberg et al., 1994). The rosamine derivatives CMTMRos and CMXRos (Fig. 4.4.3) combine the potentialdependent mitochondrial staining properties of rhodamine 123 and the capacity for coupling to intracellular peptides via a thiol-reactive chloromethyl substituent (Poot et al., 1996b). Mitochondrial staining is retained, therefore, following disruption of the membrane potential by fixation, permeabilization, or other treatments. CMXRos is not optimally excited at 488 nm, but its fluorescence emission is at sufficiently long wavelengths (∼600 nm) to facilitate combined immunofluorescence detection of intracellular antigens using fluorescein-labeled antibodies (Macho et al., 1996a.).
Nonyl Acridine Orange
Cellular Function Probes
Staining of mitochondria by nonyl acridine orange (NAO, Fig. 4.4.3) is not potential driven; it results instead from interaction of the cationic dye with the acidic phospholipid cardiolipin, which is particularly abundant in the inner mitochondrial membrane. NAO therefore serves as a useful contrast to the potential-sensitive dyes discussed above, allowing simultaneous analysis of mitochondrial activity and mass (Ferlini et al., 1996; Petit et al., 1996). The use of NAO in combination with green fluorescent dyes such as rhodamine 123 and DiOC6(3) is facilitated by the absorption and fluorescence emission maxima of the NAO:cardiolipin complex at ∼450 and 640 nm, respectively (Gallet et al., 1995). These spectral characteristics are very different from those of the monomeric dye
and are similar to those of the nucleic acid stain acridine orange complexed with RNA (see UNIT 4.3).
PROBES FOR PROTEINS AND ENZYMES Green Fluorescent Protein The green fluorescent protein (GFP) from the jellyfish Aequorea victoria incorporates an intrinsic fluorophore formed by spontaneous post-translational cyclization of a Ser-Tyr-Gly sequence (residues 65 to 67; Heim et al., 1994). The crystal structure of the 238-amino-acid protein (Ormö et al., 1996) shows the fluorophore located in the center of an 11-stranded β barrel that completely shields it from the surrounding solvent. Heterologous expression of the cDNA for GFP provides the basis for its utilization as a powerful fluorescent reporter of gene expression that is not constrained by requirements for exogenous substrates or co-factors (Chalfie et al., 1994; Galbraith et al., 1995; see also UNIT 9.5). GFP has been expressed in a wide variety of cell types, including mammalian cells, plants, yeast, and bacteria, by the use of standard transfection techniques or viral vectors. The fluorescence excitation spectrum of native GFP exhibits a maximum at 395 nm with a weaker long-wavelength band at 476 nm. These two peaks are assigned respectively to the neutral and ionized forms of the Tyr-66 sidechain. Replacement of Ser-65 by Thr shifts the equilibrium towards the ionized form, shifting the excitation maximum to 489 nm (Heim et al., 1995; Brejc et al., 1997). The resulting mutant GFP (S65T) yields substantially improved detection sensitivity in flow cytometry using 488-nm excitation (Ropp et al., 1995; Lybarger et al., 1996). Combined use of spectrally distinct GFP mutants allows simultaneous flow cytometric detection of expression from two different genes within the same cell (Anderson et al., 1996). Furthermore, S65T fluorescence is compatible with simultaneous detection of cell surface antigens using R-phycoerythrin or allophycocyanin (Lybarger et al., 1996). Stable expression of a plasmid encoding S65T GFP under the control of the HIV-1 long terminal repeat promotor in a T cell line is the basis of a new assay system for HIV infection, in which infected cells are identified by a 100- to 1000fold increase in fluorescence (Gervaix et al., 1997).
4.4.6 Supplement 2
Current Protocols in Cytometry
A HO HO
OO OH
O OH
OH O
O
HO HO
OH OH
O
O
OO
O
O
OH
OH β-galactosidase
C
OH
O
β-galactosidase −O
O
O C
B CBZ Pro Arg NH
O
O−
O
NH Arg Pro CBZ O
O−
O
peptidase +
CBZ Pro Arg NH
NH2
O
C O− O
H2N peptidase
+
NH2
O
C O− O
Figure 4.4.4 (A) Conversion of nonfluorescent fluorescein di-β-D-galactopyranoside (FDG) to highly fluorescent fluorescein by sequential β-galactosidase cleavages. (B) Conversion of a nonfluorescent rhodamine 110 bisamide substrate to highly fluorescent rhodamine 110 by sequential peptidase cleavages. The substrate shown is specific for serine proteases (e.g., trypsin, thrombin). For both substrates, the monofunctional intermediate is weakly fluorescent relative to the final product. CBZ, benzyloxycarbonyl.
Fluorogenic Substrates for Glycosidases Fluorogenic substrates, probes that are converted from a nonfluorescent to a fluorescent form by the action of an enzyme, have been developed primarily for hydrolytic enzymes, such as glycosidases, peptidases, and phosphatases, and for oxidative enzymes, such as peroxidases and microsomal dealkylases (Haugland, 1996b). The β-galactosidase substrate fluorescein di-β-D-galactopyranoside (FDG, Fig. 4.4.4A) typifies the characteristics of fluorogenic substrates suitable for flow cytometry. Methods that use FDG to detect the E. coli lacZ reporter gene in transformed cells illustrate some practical problems with the application of fluorogenic substrates in living cells (Nolan et al., 1988; Roederer et al., 1991; see also UNIT 9.5). These problems include:
1. The substrate is not cell permeant and must be loaded by transient hypotonic shock permeabilization. 2. The amount of substrate loaded is finite and cannot be readily replenished. In cells with high enzyme activity levels, measurements must be made either very soon after loading, or in the presence of a competitive inhibitor, to prevent the fluorescence intensity from being attenuated due to substrate depletion. 3. Analysis must be carried out at 4°C to minimize efflux of the fluorescent product (fluorescein). 4. Endogenous enzyme activity may compromise detection of gene expression. FDGlcU, a β-glucuronidase substrate analogous to FDG, can be used for detection of the gus reporter gene in cell types that are unable to express lacZ (Lorincz et al., 1996; see also UNIT 9.5). Researchers at Molecular Probes have
Molecular and Cellular Probes
4.4.7 Current Protocols in Cytometry
Supplement 2
O
O
O
O
(CH3COCH2 OCCH2)2 NCH2 CH3CO O
CH2N(CH2COCH2 OCCH3)2 OCCH3 calcein AM (nonfluorescent)
O
O O
O
esterase O
O +
−
+
( OCCH2)2 NHCH2 −
O
O
CH2NH(CH2CO−)2 O calcein (fluorescent)
C
O−
O
Figure 4.4.5 Conversion of calcein AM to calcein by intracellular esterases. Note the nonpolar structural characteristics of calcein AM, which result in membrane permeability, in contrast to the polar character of calcein.
made structural modifications to FDG and other analogous glycosidase substrates, adding thiol-reactive chloromethyl or pentafluorobenzoyl groups (Lorincz et al., 1997) or lipophilic alkyl tails (Sommerfelt and Sorscher, 1994) to provide improved loading and product localization.
Peptidase Substrates
Cellular Function Probes
Rhodamine 110–based peptidase substrates are analogous to fluorescein-based glycosidase substrates in that they require two successive cleavages to release fluorescent rhodamine 110 (R110), which can be excited at 488 nm for analysis by flow cytometry (Fig. 4.4.4B). These substrates are intrinsically cell permeant and exhibit little toxicity in several mammalian cell types. Enzyme specificity is determined by the peptide sequences coupled to R110 via C-terminal amide linkages. R110 derivatives with a pair of single unblocked amino acid substituents are substrates for aminopeptidases. Endopeptidase substrates have N-terminal blocked amino acid or dipeptide substituents (Fig. 4.4.4B), the latter resulting in greater enzyme specificity (Leytus et al., 1983). Cleavage of the internal peptide bond of a dipeptide substituent can be detected via coupling to a subsequent and more rapid aminopetidase cleavage of the amino acid residue attached
directly to R110 (Ganesh et al., 1995).
Esterase Substrates Conversion of nonfluorescent, nonpolar substrates into fluorescent, polar products by intracellular esterases (Fig. 4.4.5) is a versatile strategy with three primary applications: 1. Detection of esterase activity per se. For example, esterase content is an indicator of granulocytic versus monocytic differentiation in peripheral blood leukocytes (Robinson et al., 1994a). 2. Cell viability determination. Fluorogenic esterase substrates provide the most stringent indication of cell viability available without resorting to multiple probe labeling, since fluorescent staining of cells requires plasma membrane integrity for product retention as well as enzymatic activity. This technique is not always effective in some types of cells, e.g., bacteria (Kaneshiro et al., 1993; Diaper and Edwards, 1994), however, because of substrate permeability limitations. Active extrusion of carboxyfluorescein (generated enzymatically from its diacetate derivative) has been proposed as a method for flow cytometric determination of yeast viability (Breeuwer et al., 1994). 3. Intracellular loading of a wide range of fluorescent indicators (Figs. 4.4.1, 4.4.6A, and 4.4.7B). Cytoplasmic labeling using fluoro-
4.4.8 Supplement 2
Current Protocols in Cytometry
A O
O
CH3C O
O CCH3
O
Cl
HO
Cl C OH
H
O
Cl esterase
nonfluorescent
O
O H
HO
Cl C OH O
Cl
Cl C OH O
H2O2
nonfluorescent
O
O
fluorescent
B H2N
NH2 H
N CH2CH3
weakly fluorescent (blue)
H2N
NH2 N+ CH2CH3
O2−
weakly fluorescent (red) DNA
H2N
NH2 N+ CH2CH3
DNA
intensely fluorescent (red)
Figure 4.4.6 Mechanisms of intracellular oxidant detection by (A) 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) and (B) dihydroethidium. Note the dependence of the fluorescence signal on ethidium binding to DNA for the latter.
genic esterase substrates also allows analysis of processes such as gap junction communication (Tomasetto et al., 1993), cell adhesion (Hauss et al., 1996), and multidrug resistance (Holló et al., 1994). Most fluorogenic esterase substrates suitable for flow cytometry are derivatives of fluorescein diacetate. Of these, calcein AM (Fig. 4.4.5) is the generally the first choice, for several reasons: It is less sensitive to environmental factors such as pH; it exhibits minimal interference with cellular function (De Clerck et al., 1994); and it has a large net negative charge resulting in long-term intracellular retention (Fig. 4.4.5).
Probes for Oxidative Activity 2′,7′-dichlorofluorescin (also known as 2′,7′-dichlorodihydrofluorescein diacetate; DCFH-DA) is a freely membrane-permeant probe that is converted to dichlorodihydrofluo-
rescein (DCFH) by cytoplasmic esterases (Fig. 4.4.6A). The fluorescence increase resulting from oxidation of DCFH to 2′,7′-dichlorofluorescein (DCF) can be used to detect enzymatic generation of reactive oxygen species in processes such as the NADPH oxidase–linked oxidative burst of phagocytic cells (Bass et al., 1983). DCF is primarily oxidized by H2O2, although the rate of conversion is slow unless peroxidase or other cofactors (Fe2+, cytochrome c) are also present (Henderson and Chappell, 1993; Royall and Ischiropoulos, 1993). Dihydrorhodamine 123 (DHR 123) is more sensitive to H2O2 than is DCFH (Vowells et al., 1995) and exhibits better intracellular retention (Royall and Ischiropoulos, 1993) due to mitochondrial sequestration of the oxidation product (rhodamine 123; see above). DHR 123 does not require activation by esterase cleavage, providing a further advantage in cells such as
Molecular and Cellular Probes
4.4.9 Current Protocols in Cytometry
Supplement 2
A
SH O
H3C CH3
O N N
O
γ-Glu Cys Gly CH3
H3C
GST
CH2Cl
O N N
CH3
CH3 CH2S
γ-Glu Cys nonfluorescent
Gly
fluorescent
B O CH3C O
O O CCH3
O O
O
–O
O
O
esterase
C O− O
CH2Cl nonfluorescent
CH2Cl fluorescent
SH γ-Glu Cys Gly GST –O
O
O C
O−
O CH2S γ-Glu Cys Gly fluorescent
Figure 4.4.7 (A) Coupling of monochlorobimane (mBCl) to glutathione mediated by glutathioneS-transferase (GST). Note that reaction of monobromobimane (mBBr) with glutathione yields an identical product but does not require GST. (B) Conversion of 5-chloromethylfluorescein diacetate (CMFDA) to 5-chloromethylfluorescein followed by GST-mediated coupling to glutathione.
Cellular Function Probes
monocytes, in which DCFH-DA hydrolysis is incomplete (Robinson et al., 1988). The sensitivity and response time of oxidative burst detection by DHR 123 and DCFH-DA are impaired by their location in the cytoplasm rather than at the site of oxidative activity in the phagocytic vacuole. In addition, false positive responses can be generated in unstimulated cells because of intercellular diffusion of H2O2 (Henderson and Chappell, 1993). Coupling DCFH to an immune complex overcomes these problems, producing a probe targeted to the phagocytic vacuole via Fc receptor–mediated internalization (Ryan et al., 1990). The response mechanism of dihydroethidium (also frequently referred to as hydroethidine) to intracellular oxidants is considerably different from that of DCFH-DA (Fig. 4.4.6B). Unlike DHR123 or DCFH-DA, dihydroethidium can be oxidized directly by superoxide anion (O2−), which is generated earlier in the oxidative burst reaction sequence than H2O2 (Rothe and Valet, 1990). The spectral characteristics of DNA-bound ethidium (absorption
and emission maxima at 518 nm and 605 nm respectively) are suitable for simultaneous excitation and detection in combination with DCFH-DA. This procedure provides better discrimination between functionally distinct neutrophil populations than does single-probe analysis (Rothe and Valet, 1990; Robinson et al., 1994b). Dihydroethidium and dihydrorhodamine 6G (a close analog of DHR123) exhibit increased accumulation in epithelial cells expressing the cystic fibrosis transmembrane conductance regulator, and can therefore be used to detect functional activation of this cAMPdependent chloride channel (Wersto et al., 1996).
Probes for Intracellular Thiols
The tripeptide glutathione (GSH; γ-GluCys-Gly) is the most abundant nonprotein thiol in eukaryotic cells. Its primary role is to provide the reducing capacity required to prevent damage to DNA by free radicals and other oxidants. GSH also participates in the removal of xenobiotics, thereby contributing to the develop-
4.4.10 Supplement 2
Current Protocols in Cytometry
ment of multidrug resistance. The primary problems encountered in flow cytometric determination of GSH are obtaining a stoichiometric fluorescence response to the prevailing high intracellular GSH concentrations (typically >2 mM) and specificity for GSH versus other thiols (Hedley and Chow, 1994a,b). Monochlorobimane (mBCl) and monobromobimane (mBBr) are currently the most widely used probes for intracellular GSH. Both are freely membrane permeant and initially nonfluorescent; conjugation to GSH yields a fluorescent thioether with fluorescence excitation and emission maxima at ∼390 and 480 nm, respectively (Fig. 4.4.7A). Because the thiol reactivity of mBCl is low, coupling to GSH only occurs to a significant extent via the action of glutathione-S-transferase (GST). Coupling of the more reactive mBBr occurs spontaneously. Although enzymatic mediation of mBCl coupling provides high GSH specificity, the probe is unfortunately a poor substrate for many GST isoenzymes, leading to underestimation of GSH levels (Ublacker et al., 1991; Hedley and Chow, 1994a). For this reason, mBBr is generally preferred over mBCl (Hedley and Chow, 1994a,b). Extrusion of the fluorescent conjugate by organic ion transporters may also contribute to underestimation of GSH (Poot et al., 1996a). Cell-permeant 5-chloromethylfluorescein diacetate (CMFDA) is converted to a fluorescent product by the action of intracellular esterases. The cleavage product (5-chloromethylfluorescein) can then couple to GSH via the same enzymatic process as mBCl. The significant difference in the case of CMFDA is that GSH coupling produces no change in fluorescence (Fig. 4.4.7B). The principal advantage of CMFDA over mBBr and mBCl is that its absorption maximum is well matched to the 488nm argon-ion laser line. CMFDA is less GSH specific than the mBBr and mBCl according to published comparisons (Poot et al., 1991; Barhoumi et al., 1993; Hedley and Chow, 1994a). Discrepancies appear to arise from the fluorescence of unconjugated 5-chloromethylfluorescein rather than from conjugation to thiols other than GSH (Hedley and Chow, 1994a). Nevertheless, GSH levels measured using CMFDA have correlated well with those determined using other methods in several instances (Barhoumi et al., 1993; Coates and Tripp, 1995). Coupling of CMFDA to GSH and other intracellular thiols provides a basis for longterm retention of the probe in the cells, except
in cases where retention is compromised by active extrusion of the fluorescent conjugate (see above). Labeling of cell populations with well-retained fluorescent tracers allows functional properties expressed at the multicellular level, such as adhesion, fusion, and gap junction communication, to be detected and quantitated. A general experimental strategy for these applications is to label two cell populations with probes that have contrasting emission spectra; cell-cell interactions result in probe transfer and the subsequent detection of a third, dual-labeled cell population. For example, CMFDA has been used in combination with an analogous rhodamine derivative (5-[and-6][{(4-chloromethyl)benzoyl}amino] tetraethylrhodamine) to quantitatively detect hybrid cells produced by electrofusion (Jaroszeski et al., 1994). Other probes resistant to spontaneous intercellular transfer for these applications include amine-reactive carboxyfluorescein diacetate succinimidyl ester (CFDA SE), thiol-reactive 4,5,6,7-tetrafluorofluorescein diacetate (TFFDA), calcein AM, and lipophilic carbocyanines (see below). Unlike CMFDA (Fig. 4.4.7B), TFFDA is only thiol reactive prior to esterase cleavage; the fluorescent product 4,5,6,7-tetrafluorofluorescein is not reactive (Gee et al., 1996). CFDA SE is widely used to label cells for in vivo migration studies (Weston and Parish, 1990). Because the label distributes equally among cell progeny, successive generations of dividing cells are marked by halving of cellular fluorescence (Lyons and Parish, 1994).
PROBES FOR LIPIDS AND MEMBRANES Most lipid probes exhibit marked fluorescence enhancement when partitioned from aqueous solutions into nonpolar environments such as membrane and lipoprotein interiors. It is more difficult to obtain sufficient fluorescence response specificity to detect particular molecular species or metabolic processes. Nevertheless, several valuable approaches have been devised.
Annexin V Annexin V is a phospholipid-binding protein with high affinity for negatively charged phosphatidylserine. Binding of fluorescent annexin V conjugates provides sensitive identification of cells in the early stages of apoptosis, which are characterized by translocation of phosphatidylserine from the inner monolayer
Molecular and Cellular Probes
4.4.11 Current Protocols in Cytometry
Supplement 2
A
O C OCH2 C OCH
NH N N
O
O
O +
CH2O P OCH2 CH2N(CH3)3 O−
NO2
B H3C
F N
CHOH
F B
N
C NHCH O
CH2OH
H3C
C
D H3C CH3
H3C CH3
O− C O
CH CH CH N+
N
F BN F N H3C
CH3
Figure 4.4.8 Fluorescent lipid probes: (A) C6-NBD-PC; (B) BODIPY FL C5-ceramide; (C) DiIC18(3); (D) BODIPY FL C12.
to the outer monolayer of the plasma membrane (Vermes et al., 1995). Although fluorescein-labeled annexin V has been used exclusively in applications reported to date, other dyes with longer wavelength emission characteristics could presumably be substituted for applications involving simultaneous detection of probes such as DiOC6(3) or fluo-3. Simultaneous detection of phosphatidylserine exposure and other expressions of apoptosis can be accomplished using fluorescein-labeled annexin V and CMXRos (mitochondrial membrane potential disruption; Castedo et al., 1996) or SNARF-1 (intracellular acidification; Meisenholder et al., 1996 ).
BODIPY and NBD Probes Cellular Function Probes
The BODIPY FL fluorophore is intrinsically lipophilic, and can be incorporated into
phospholipids, sphingolipids, and fatty acids without major perturbation of the structural and functional properties of the corresponding native lipids (Fig. 4.4.8). Flow cytometric applications of these lipid analogs include quantitative analysis of cell type–dependent fatty acid uptake (Macho et al., 1996b) and detection of intracellular phospholipase A activity (Meshulam et al., 1992). Neutral BODIPY dyes are probes for intracellular accumulation of neutral lipids and are reported to be superior to the conventionally used dye Nile Red for this purpose (Gocze and Freeman, 1994). BODIPY FL is efficiently excited at 488 nm and has narrow spectral bandwidths, resulting in minimal detection spillover when it is used in combination with probes emitting at longer wavelengths (Furlong et al., 1997). Furthermore, BODIPY FL exhibits a fluorescence emission shift from
4.4.12 Supplement 2
Current Protocols in Cytometry
∼520 nm to ∼620 nm when it becomes highly concentrated. Consequently, cells labeled with BODIPY FL C5–ceramide exhibit bright red staining of the Golgi apparatus, the primary site of metabolic accumulation (Pagano et al., 1991). The same technique has also recently been used to identify populations of endosomes characterized by different levels of BODIPY FL C5–sphingomyelin incorporation (Chen et al., 1997). NBD fluorescence is generally weaker and more environment dependent than that of BODIPY FL. A useful property of NBD is its reduction to a nonfluorescent product by dithionite, which can be exploited for analysis of phospholipid movement across the plasma membrane (Nolan et al., 1995).
Lipophilic Carbocyanines Carbocyanine dyes with C12 or longer alkyl substituents are highly lipophilic, localizing predominantly in the plasma membrane, in contrast to the potential-dependent transmembrane distribution adopted by less lipophilic analogs such as DiOC6(3) (see above). This type of probe is exemplified by DiOC18(3) and PKH2 (Horan et al., 1990), which are both green fluorescent, and orange-fluorescent DiIC18(3) and PKH26. Typical flow cytometric applications of lipophilic carbocyanines, such as detection of cell-cell clustering and fusion (Gant et al., 1992; Spötl et al., 1995), are based on their capacity to produce stable labeling of cell populations with minimal cytotoxicity and physiological perturbation. In two-color methods for analysis of viability (Chang et al., 1993) and gap junction communication (Tomasetto et al., 1993), lipophilic carbocyanine probes function as inert reference markers for cells that are not stained by a second “active” probe (propidium iodide for viability, calcein for gapjunction transfer). A new derivative of DiIC18(3) incorporating a thiol-reactive choromethyl substituent (CM-DiI, Molecular Probes) is resistant to fixation and permeabilization treatments that degrade labeling by other lipophilic carbocyanines. Flow cytometric applications of CM-DiI have included tracking the circulatory migration of lymphocytes (Andrade et al., 1996) and detecting endocytosis of live bacteria by amoebae (Harf et al., 1997).
CONCLUDING REMARKS This article presents a sampling of the many applications of cellular function probes. The scope for future experimentation offered by existing probes, used singly or in combination with others (including the array of nucleic acid
stains and immunofluorescent cell surface markers discussed elsewhere in this volume; see UNITS 4.3, 4.2 & 6.2), is enormous. Further advances will be facilitated by improving existing probes and designing and synthesizing new ones. Existing technology can provide fluorescent analogs of almost any biologically active molecule for investigation of transport, metabolism, and receptor binding. Development of specific indicators for ions and small molecules is technically challenging, but potentially rewarding, in view of what has already been achieved with currently available probes.
LITERATURE CITED Altenberg, G.A., Vanoye, C.G., Horton, J.K., and Reuss, L. 1994. Unidirectional fluxes of rhodamine 123 in multidrug-resistant cells: Evidence against direct drug extrusion from the plasma membrane. Proc. Natl. Acad. Sci. U.S.A. 91:4654-4657. Amorino, G.P. and Fox, M.H. 1995. Intracellular Na+ measurements using Sodium Green tetraacetate with flow cytometry. Cytometry 21:248-256. Anderson, M.T., Tjioe, I.M., Lorincz, M.C., Parks, D.R., Herzenberg, L.A., Nolan, G.P., and Herzenberg, L.A. 1996. Simultaneous fluorescenceactivated cell sorter analysis of two distinct transcriptional elements within a single cell using engineered green fluorescent proteins. Proc. Natl. Acad. Sci. U.S.A. 93:8508-8511. Andrade, W., Seabrook, T.J., Johnston, M.G., and Hay, J.B. 1996. The use of the lipophilic fluorochrome CM-DiI for tracking the migration of lymphocytes. J. Immunol. Methods 194:181189. Barhoumi, R., Bowen, J.A., Stein, L.S., Echols, J., and Burghardt, R.C. 1993. Concurrent analysis of intracellular glutathione content and gap junctional intercellular communication. Cytometry 14:747-756. Bass, D., Parce, J.W., Dechatelet, L.R., Szejda, P., Seeds, M.C., and Thomas, M.J. 1983. Flow cytometric studies of oxidative product formation by neutrophils: A graded response to membrane stimulation. J. Immunol. 130:1910-1917. Boyer, M.J. and Hedley, D.W. 1994. Measurement of intracellular pH. Methods Cell Biol. 41:135148. Breeuwer, P., Drocourt, J.-L., Rombouts, F.M., and Abee, T. 1994. Energy-dependent, carrier-mediated extrusion of carboxyfluorescein from Saccharomyces cerevisiae allows rapid assessment of cell viability by flow cytometry. Appl. Environ. Microbiol. 60:1467-1472. Brejc, K., Sixma, T.K., Kitts, P.A., Kain, S.R., Tsien, R.Y., Ormö, M., and Remington, S.J. 1997. Structural basis for dual excitation and photoisomerization of the Aequorea victoria green fluorescent protein. Proc. Natl. Acad. Sci. U.S.A. 94:2306-2311.
Molecular and Cellular Probes
4.4.13 Current Protocols in Cytometry
Supplement 2
Broxterman, H.J., Sonneveld, P., Feller, N., Ossenkoppele, G.J., Währer, D.C.R., Eekman, C.A., Schoester, M., Lankelma, J., Pinedo, H.M., Löwenberg, B., and Schuurhuis, G.J. 1996. Quality control of multidrug resistance assays in adult acute leukemia: Correlation between assays for P-glycoprotein expression and activity. Blood 87:4809-4816.
Dordal, M.S., Ho, A.C., Jackson-Stone, M., Fu, Y.F., Goolsby, C.L., and Winter, J.N. 1995. Flow cytometric assessment of the cellular pharmacokinetics of fluorescent drugs. Cytometry 20:307314.
Castedo, M., Hirsch, T., Susin, S.A., Zamzami, N., Marchetti, P., Macho, A., and Kroemer, G. 1996. Sequential acquisition of mitochondrial membrane alterations during early lymphocyte apoptosis. J. Immunol. 157:512-521.
Ferlini, C., Di Cesare, S., Rainaldi, G., Malorni, W., Samoggia, P., Biselli, R., and Fattorossi, A. 1996. Flow cytometric analysis of the early phases of apoptosis by cellular and nuclear techniques. Cytometry 24:106-115.
Chalfie, M., Tu, Y., Euskirchen, G., Ward, W.W., and Prasher, D.C. 1994. Green fluorescent protein as a marker for gene expression. Science 263:802805.
Floto, R.A., Mahaut-Smith, M.P., Somasundaram, B., and Allen, J.M. 1996. IgG-induced Ca2+ oscillations in differentiated U937 cells: A study using laser scanning confocal microscopy and co-loaded fluo-3 and Fura Red fluorescent probes. Cell Calcium 18:377-389.
Chang, L., Gusewitch, G.A., Chritton, D.B.W., Folz, J.C., Lebeck, L.K., and Nehlsen-Cannarella, S.L. 1993. Rapid flow cytometric assay for the assessment of natural killer cell activity. J. Immunol. Methods 166:45-54. Chen, C.-S., Martin, O.C., and Pagano, R.E. 1997. Changes in the spectral properties of a plasma membrane lipid analog in the first seconds of endocytosis in living cells. Biophys. J. 72:37-50. Chen, L.B. 1989. Fluorescent labeling of mitochondria. Methods Cell Biol. 29:103-123. Chen, L.B. and Smiley, S.T. 1993. Probing mitochondrial membrane potential in living cells by a J-aggregate forming dye. In Fluorescent and Luminescent Probes for Biological Activity (W.T. Mason, ed.) pp.124-132. Academic Press, San Diego.
Furlong, S.T., Koziel, H., Bartlett, M.S., McLaughlin, G.L., Shaw, M.M., and Jack, R.M. 1997. Lipid transfer from human epithelial cells to Pneumocystis carinii in vitro. J. Infect. Dis. 175:661-668. Galbraith, D.W., Lambert, G.M., Grebenok, R.J., and Sheen, J. 1995. Flow cytometric analysis of transgene expression in higher plants: Green fluorescent protein. Methods Cell Biol. 50:3-13. Gallet, P.F., Maftah, A., Petit, J.M., Denis-Gay, M., and Julien, R. 1995. Direct cardiolipin assay in yeast using the red fluorescence emission of 10-N-nonyl acridine orange. Eur. J. Biochem. 228:113-119.
Chow, S., Hedley, D., and Tannock, I. 1996. Flow cytometric calibration of intracellular pH measurements in viable cells using mixtures of weak acids and bases. Cytometry 24:360-367.
Ganesh, S., Klingel, S., Kahle, H., and Valet, G. 1995. Flow cytometric determination of aminopeptidase activities in viable cells using fluorogenic rhodamine 110 substrates. Cytometry 20:334-340.
Coates, A. and Tripp, E. 1995. Comparison of two fluorochromes for flow cytometric assay of cellular glutathione content in human malignant melanoma. Melanoma Res. 5:107-111.
Gant, V.A., Shakoor, Z., and Hamblin, A.S. 1992. A new method for measuring clustering in suspension between accessory cells and T lymphocytes. J. Immunol. Methods 156:179-189.
Cossarizza, A., Baccarani-Contri, B., Kalashnikova, G., and Franceschi, C. 1993. A new method for the cytofluorimetric analysis of mitochondrial membrane potential using the J-aggregate forming lipophilic cation 5,5′,6,6′-tetrachloro1,1′,3,3′-tetraethylbenzimidazolylcarbocyanine iodide (JC-1). Biochem. Biophys. Res. Commun. 197:40-45.
Gee, K.R., Sun, W.-C., Klaubert, D.H., Haugland, R.P., Upson, R.H., Steinberg, T.H., and Poot, M. 1996. Novel derivatization of protein thiols with fluorinated fluoresceins. Tetrahedron Lett. 37:7905-7908.
Davis, S., Weiss, M.J., Wong, J.R., Lampidis, T.J., and Chen, L.B. 1985. Mitochondrial and plasma membrane potentials cause unusual accumulation and retention of rhodamine 123 by human breast adenocarcinoma–derived MCF-7 cells. J. Biol. Chem. 260:13844-13850. De Clerck, L.S., Bridts, C.H., Mertens, A.M., Moens, M.M., and Stevens, W.J. 1994. Use of fluorescent dyes in the determination of adherence of human leukocytes to endothelial cells and the effect of fluorochromes on cellular function. J. Immunol. Methods 172:115-124. Cellular Function Probes
Diaper, J.P. and Edwards, C. 1994. The use of fluorogenic esters to detect viable bacteria by flow cytometry. J. Appl. Bacteriol. 77:221-228.
Gervaix, A., West, D., Leoni, L.M., Richman, D.D., Wong-Staal, F., and Corbeil, J. 1997. A new reporter cell line to monitor HIV infection and drug susceptibility. Proc. Natl. Acad. Sci. U.S.A. 94:4653-4658. Gocze, P.M. and Freeman, D.A. 1994. Factors underlying the variability of lipid droplet fluorescence in MA-10 Leydig tumor cells. Cytometry 17:151-158. Greimers, R., Trebak, M., Moutschen, M., Jacobs, N., and Boniver, J. 1996. Improved four-color flow cytometry method using fluo-3 and triple immunofluorescence for analysis of intracellular calcium ion ([Ca2+]i) fluxes among mouse lymph node B- and T-lymphocyte subsets. Cytometry 23:205-217.
4.4.14 Supplement 2
Current Protocols in Cytometry
Grynkiewicz, G., Poenie, M., and Tsien, R.Y. 1985. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J. Biol. Chem. 260:3440-3450. Harf, C., Goffinet, S., Meunier, O., Monteil, H., and Colin, D.A. 1997. Flow cytometric determination of endocytosis of viable labeled Legionella pneumophila by Acanthamoeba palestinensis. Cytometry 27:269-274. Haugland, R.P. 1996a. Indicators for Ca2+, Mg2+, Zn2+ and other metals. In Handbook of Fluorescent Probes and Research Chemicals, 6th Ed. (M.T.Z. Spence, ed.) pp. 503-550. Molecular Probes, Eugene, OR. Haugland, R.P. 1996b. Enzymes, enzyme substrates and enzyme inhibitors. In Handbook of Fluorescent Probes and Research Chemicals, 6th Ed. (M.T.Z. Spence, ed.) pp. 201-250. Molecular Probes, Eugene, OR. Hauss, P., Selz, F., and Fischer, A. 1996. Comparative analysis of CD4-mediated down-regulation of T cell adhesion to B cells by flow cytometry and fluorescence microscopy. Cytometry 23:3947. Hedley, D.W. and Chow, S. 1994a. Evaluation of methods for measuring cellular glutathione content using flow cytometry. Cytometry 15:349358. Hedley, D.W. and Chow, S. 1994b. Glutathione and cellular resistance to anti-cancer drugs. Methods Cell Biol. 42:31-44. Heim, R., Cubitt, A.B., and Tsien, R.Y. 1995. Improved green fluorescence. Nature 373:663-664. Heim, R., Prasher, D.C., and Tsien, R.Y. 1994. Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. U.S.A. 91:12501-12504. Henderson, L.M. and Chappell, J.B. 1993. Dihydrorhodamine 123: A fluorescent probe for superoxide generation? Eur. J. Biochem. 217:973980. Holló, Z., Homolya, L., Davis, W.C., and Sarkadi, B. 1994. Calcein accumulation as a fluorometric functional assay of the multidrug transporter. Biochim. Biophys. Acta 1191:384-388. Horan, P.K., Melnicoff, M.J., Jensen, B.D., and Slezak, S.E. 1990. Fluorescent cell labeling for in vivo and in vitro cell tracking. Methods Cell Biol. 33:469-490. Hove-Madsen, L. and Bers, D.M. 1992. Indo-1 binding to protein in permeabilized ventricular myocytes alters its spectral and calcium binding properties. Biophys. J. 63:89-97. Jaroszeski, M.J., Gilbert, R., and Heller, R. 1994. Detection and quantitation of cell-cell electrofusion products by flow cytometry. Anal. Biochem. 216:271-275. June, C.H. and Rabinovitch, P.S. 1994. Intracellular ionized calcium. Methods Cell Biol. 41:149-174. Kaneshiro, E.S., Wyder, M.A., Wu, Y.-P., and Cushion, M.T. 1993. Reliability of calcein acetoxymethyl ester and ethidium homodimer or
propidium iodide for viability assessment of microbes. J. Microbiol. Methods 17:1-16. Kaprelyants, A.S. and Kell, D.B. 1992. Rapid assessment of bacterial viability and vitality by rhodamine 123 and flow cytometry. J. Appl. Bacteriol. 72:410-422. Kurtz, I. and Balaban, R.S. 1985. Fluorescence emission spectroscopy of 1,4-dihydroxyphthalonitrile: A method for determining intracellular pH in cultured cells. Biophys. J. 48:499-508. Leytus, S.P., Patterson, W.L., and Mangel, W.F. 1983. New class of sensitive and selective fluorogenic substrates for serine proteases. Biochem. J. 215:253-260. Lipp, P. and Niggli, E. 1993. Ratiometric confocal Ca2+ measurements with visible wavelength indicators in isolated cardiac myocytes. Cell Calcium 14:359-372. Lorincz, M., Herzenberg, L.A., Diwu, Z., Barranger, J.A., and Kerr, W.G. 1997. Detection and isolation of gene-corrected cells in Gaucher disease via a fluorescence activated cell sorter assay for lysosomal glucocerebrosidase activity. Blood 89:3412-3420. Lorincz, M., Roederer, M., Diwu, Z., Herzenberg, L.A., and Nolan, G.P. 1996. Enzyme-generated intracellular fluorescence for single-cell reporter gene analysis utilizing Escherichia coli β-glucuronidase. Cytometry 24:321-329. Ludescher, C., Gattringer, C., Drach, J., Hofmann, J., and Grunicke, H. 1991. Rapid functional assay for the detection of multidrug-resistant cells using the fluorescent dye rhodamine 123. Blood 78:1385-1390. Lybarger, L., Dempsey, D., Franek, K.J., and Chervenak, R. 1996. Rapid generation and flow cytometric analysis of stable GFP-expressing cells. Cytometry 25:211-220. Lyons, A.B. and Parish, C.R. 1994. Determination of lymphocyte division by flow cytometry. J. Immunol. Methods 171:131-137. Macho, A., Decaudin, D., Castedo, M., Hirsch, T., Susin, S.A., Zamzami, N., and Kroemer, G. 1996a. Chloromethyl-X-rosamine is an aldehyde-fixable potential-sensitive fluorochrome for the detection of early apoptosis. Cytometry 25:333-340. Macho, A, Mishal, Z., and Uriel, J. 1996b. Molar quantification by flow cytometry of fatty acid binding to cells using dipyrrometheneboron difluoride derivatives. Cytometry 23:166-173. Mandler, R.N., Schaffner, A.E., Novotny, E.A., Lange, G.D., and Barker, J.L. 1988. Flow cytometric analysis of membrane potential in embryonic rat spinal cord cells. J. Neurosci. Methods 22:203-213. Mason, D.J., Lopéz-Amorós, Allman, R., Stark, J.M., and Lloyd, D. 1995. The ability of membrane potential dyes and calcofluor white to distinguish between viable and non-viable bacteria. J. Appl. Bacteriol. 78:309-315. Molecular and Cellular Probes
4.4.15 Current Protocols in Cytometry
Supplement 2
Meisenholder, G.W., Martin, S.J., Green, D.R., Nordberg, J., Babior, B.M., and Gottlieb, R.A. 1996. Events in apoptosis. J. Biol. Chem. 271:16260-16262.
Plášek, J. and Sigler, K. 1996. Slow fluorescent indicators of membrane potential: A survey of different approaches to probe response analysis. J. Photochem. Photobiol. Biol. 33:101-124.
Meshulam, T., Herscovitz, H., Casavant, D., Bernardo, J., Roman, R., Haug land, R.P., Strohmeier, G.S., Diamond, R.D., and Simons, E.R. 1992. Flow cytometric kinetic measurements of neutrophil phospholipase A activation. J. Biol. Chem. 267:21465-21470.
Poot, M., Hudson, F.N., Grossmann, A., Rabinovitch, P.S., and Kavanagh, T.J. 1996a. Probenicid inhibition of fluorescence extrusion after MCB-staining of Rat-1 fibroblasts. Cytometry 23:78-81.
Minta, A. and Tsien, R.Y. 1989. Fluorescent indicators for cytosolic sodium. J. Biol. Chem. 264:19449-19457. Minta, A., Kao, J.P.Y., and Tsien, R.Y. 1989. Fluorescent indicators for cytosolic calcium based on rhodamine and fluorescein chromophores. J. Biol. Chem. 264:8171-8178. Nolan, J.P., Magargee, S.F., Posner, R.G., and Hammerstedt, R.H. 1995. Flow cytometric analysis of transmembrane phospholipid movement in bull sperm. Biochemistry 34:3907-3915. Nolan, G.P., Fiering, S., Nicolas, J.-F., and Herzenberg, L.A. 1988. Fluorescence-activated cell analysis and sorting of viable mammalian cells b ased on β-D-galactosidase activity after transduction of Escherichia coli lacZ. Proc. Natl. Acad. Sci. U.S.A. 85:2603-2607. Novak, E.J. and Rabinovitch, P.S. 1994. Improved sensitivity in flow cytometric intracellular ionized calcium measurements using fluo-3/Fura Red fluorescence ratios. Cytometry 17:135-141. Ormö, M., Cubitt, A.B., Kallio, K., Gross, L.A., Tsien, R.Y., and Remington, S.J. 1996. Crystal structure of the Aequorea victoria green fluorescent protein. Science 273:1392-1395. Owen, C.S. 1992. Comparison of the spectrumshifting intracellular pH probes 5′(and 6′)-carboxy-10-dimethylamino-3-hydroxyspiro[7Hbenzo[c]xanthene-7,1′(3′H)-isobenzofuran]-3′o n e and 2′,7′-biscarboxyethyl-5(and-6)-carboxyfluorescein. Anal. Biochem. 204:65-71. Owen, C.S., Carango, P., Grammer, S., Bobyock, S., and Leeper, D.B. 1992. pH-dependent intracellular quenching of the indicator carboxy SNARF-1. J. Fluorescence 2:75-80. Owen, C.S., Sykes, N.L., Shuler, R.L., and Ost, D. 1991. Non-calcium environmental sensitivity of intracellular indo-1. Anal. Biochem. 192:142148. Pagano, R.E., Martin, O.C., Kang, H.C., and Haugland, R.P. 1991. A novel fluorescent ceramide analogue for studying membrane traffic in animal cells: Accumulation at the Golgi apparatus results in altered spectral properties of the sphingolipid precursor. J. Cell Biol. 113:1267-1279. Petit, P.X, Glab, N., Marie, D., Kieffer, H., and Métézeau, P. 1996. Discrimination of respiratory dysfunction in yeast mutants by confocal microscopy, image, and flow cytometry. Cytometry 23:28-38. Cellular Function Probes
Poot, M., Zhang, Y.-Z., Krämer, J.A., Wells, K.S., Jones, L.J., Hanzel, D.K., Lugade, A.G., Singer, V.L., and Haugland, R.P. 1996b. Analysis of mitochondrial morphology and function with novel fixable fluorescent stains. J. Histochem. Cytochem. 44:1363-1372. Poot, M., Kavanagh, T.J., Kang, H.C., Haugland, R.P., and Rabinovitch, P.S. 1991. Flow cytometric analysis of cell cycle–dependent changes in cell thiol level by combining a new laser dye with Hoechst 33342. Cytometry 12:184-187. Rijkers, G.T. and Griffioen, A.W. 1993. Changes in free cytoplasmic magnesium following activation of human lymphocytes. Biochem. J. 289:373-377. Rijkers, G.T., Justement, L.B., Griffioen, A.W., and Cambier, J.C., 1990. Improved method for measuring intracellular Ca2+ with fluo-3. Cytometry 11:923-927. Robinson, J.P., Narayanan, P.K., and Carter, W.O. 1994a. Functional measurements using HL-60 cells. Methods Cell Biol. 42:423-436. Robinson, J.P., Carter, W.O., and Narayanan, P.K. 1994b. Oxidative product formation analysis by flow cytometry. Methods Cell Biol. 41:437-447. Robinson, J.P., Bruner, L.H., Bassoe, C.F., Hudson, J.L., Ward, P.A., and Phan, S.H. 1988. Measurement of intracellular fluorescence of human monocytes relative to oxidative metabolism. J. Leukocyte Biol. 43:304-310. Roederer, M., Fiering, S., and Herzenberg, L.A. 1991. FACS-Gal: Flow cytometric analysis and sorting of cells expressing reporter gene constructs. Methods 2:248-260. Ropp, J.D., Donahue, C.J., Wolfgang-Kimball, D., Hooley, J.J., Chin, J.Y.W., Hoffman, R.A., Cuthbertson, R.A., and Bauer, K.D. 1995. Aequorea green fluorescent protein analysis by flow cytometry. Cytometry 21:309-317. Rothe, G. and Valet, G. 1990. Flow cytometric analysis of respiratory burst activity in phagocytes with hydroethidine and 2′,7′-dichlorofluorescein. J. Leukocyte Biol.47:440-448. Royall, J.A. and Ischiropoulos, H. 1993. Evaluation o f 2′7′-dichlorofluorescein and dihydrorhodamine 123 as fluorescent probes for intracellular H2O2 in cultured endothelial cells. Arch. Biochem. Biophys. 302:348-355. Ryan, T.C., Weil, G.J., Newburger, P.E., Haugland, R., and Simons, E.R. 1990. Measurement of superoxide release in the phagovacuoles of immune complex–stimulated human neutrophils. J. Immunol. Methods 130:223-233.
4.4.16 Supplement 2
Current Protocols in Cytometry
Scheenen, W.J.J.M., Makings, L.R., Gross, L.R., Pozzan, T., and Tsien, R.Y. 1996. Photodegradation of indo-1 and its effect on apparent Ca2+ concentrations. Chem. Biol. 3:765-774. Shapiro, H.M. 1994. Cell membrane potential analysis. Methods Cell Biol. 41:121-133.
Vandenberghe, P.A. and Ceuppens, J.L. 1990. Flow cytometric measurement of cytoplasmic free calcium in human peripheral blood T lymphocytes with fluo-3, a new fluorescent calcium indicator. J. Immunol. Methods 127:197-205.
Skowronek, P., Krummeck, G., Haferkamp, O., and Rödel, G. 1990. Flow cytometry as a tool to discriminate respiratory-competent and respiratory-deficient yeast cells. Curr. Genet. 18:265267.
Van Erp, P.E.J., Jansen, M.J.J.M., De Jongh, G.J., Boezeman, J.B.M., and Schalkwijk, J. 1991. Ratiometric measurement of intracellular pH in cultured human keratinocytes using carboxy SNARF-1 and flow cytometry. Cytometry 12, 127-132.
Simons, E.R. 1993. Flow cytometry: Use of multiparameter kinetics to evaluate several activation parameters simultaneously in individual living cells. In Fluorescent and Luminescent Probes for Biological Activity (W.T. Mason, ed.) pp. 298309. Academic Press, San Diego.
Vermes, I., Haanen, C., Steffens-Nakken, H., and Reutelingsperger, C. 1995. A novel assay for apoptosis: Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein-labeled annexin V. J. Immunol. Methods 184:39-51.
Sommerfelt, M.A. and Sorscher, E.J. 1994. Use of fluorescence-activated cell sorting for rapid isolation of insect cells harboring recombinant baculovirus. Methods Cell Biol. 42:563-574.
Vowells, S.J., Sehhsaria, S., Malech, H.L., Shalit, M., and Fleisher, T.A. 1995. Flow cytometric analysis of the granulocyte respiratory burst: A comparison study of fluorescent probes. J. Immunol. Methods 178:89-97.
Spangrude, G.J. and Johnson, G.R. 1990. Resting and activated subsets of mouse multipotent hematopoietic stem cells. Proc. Natl. Acad. Sci. U.S.A. 87:7433-7437. Spencer, C.I. and Berlin, J.R. 1995. A method for recording intracellular [Ca2+] transients in cardiac myocytes using Calcium Green-2. Pfluegers Arch. Eur. J. Physiol. 430:579-583. Spötl, L., Sarti, A., Dierich, M.P., and Möst, J. 1995. Cell membrane labeling with fluorescent dyes for the demonstration of cytokine-induced fusion between monocytes and tumor cells. Cytometry 21:160-169. Summerhayes, I.C., Lampidis, T.J., Bernal, S.D., Nadakavukaren, J.J., Nadakavukaren, K.K., Shepherd, E.L., and Chen, L.B. 1982. Unusual retention of rhodamine 123 by mitochondria in muscle and carcinoma cells. Proc. Natl. Acad. Sci. U.S.A. 79:5292-5296. Tomasetto, C., Neveu, M.J., Daley, J., Horan, P.K., and Sager, R. 1993 Specificity of gap junction communication among mammary cells and connexin transfectants in culture. J. Cell Biol. 122:157-167. Tse, F.W., Tse, A., and Hille, B. 1994. Cyclic Ca2+ changes in intracellular stores of gonadotropes during gonadotropin-releasing hormone–stimulated Ca2+ oscillations. Proc. Natl. Acad. Sci. U.S.A. 91:9750-9754. Tsien, R.Y. 1980. New calcium indicators and buffers with high selectivity against magnesium and protons: Design, synthesis, and properties of prototype structures. Biochemistry 19: 23962404. Ublacker, G.A., Johnson, J.A., Siegel, F.L., and Mulcahy, R.T. 1991. Influence of glutathione-Stransferases on cellular glutathione determination by flow cytometry using monochlorobimane. Cancer Res. 51:1783-1788.
Wang, Z., Chu, G.L., Hyun, W.C., Pershadsingh, H.A., Fulwyler, M.A., and Dewey, W.C. 1990. Comparison of DMO and flow cytometric methods for measuring intracellular pH and the effect of hyperthermia on the transmembrane pH gradient. Cytometry 11:617-623. Wersto, R.P., Rosenthal, E.R., Crystal, R.G., and Spring, K.R. 1996. Uptake of fluorescent dyes associated with the functional expression of the cystic fibrosis transmembrane conductance regulator in epithelial cells. Proc. Natl. Acad. Sci. U.S.A. 93:1167-1172. Weston, S.A. and Parish, C.R. 1990. New fluorescent dyes for lymphocyte migration studies. Analysis by flow cytometry and fluorescence microscopy. J. Immunol. Methods 133:87-97. Whitaker, J.E., Haugland, R.P., and Prendergast, F.G. 1991. Spectral and photophysical studies of Benzo[c]xanthene dyes: Dual emission pH sensors. Anal. Biochem. 194:330-344. Wilson, H.A., Seligmann, B.E., and Chused, T.M. 1985. Voltage-sensitive cyanine dye fluorescence signals in lymphocytes: Plasma membrane and mitochondrial components. J. Cell. Physiol. 125:61-71.
INTERNET RESOURCES http://www.probes.com The Molecular Probes’ World Wide Web site contains the Molecular Probes bibliography database, created by Dr. Richard Haugland, which is the source for many of the papers cited in this unit.
Contributed by Iain D. Johnson Molecular Probes, Inc. Eugene, Oregon
Molecular and Cellular Probes
4.4.17 Current Protocols in Cytometry
Supplement 9
Spectroscopic Analysis Using DNA and RNA Fluorescent Probes
UNIT 4.5
Reliable quantitation of nanogram and microgram amounts of DNA and RNA in solution is essential to researchers in cell and molecular biology. In addition to the traditional absorbance measurements at 260 nm (see Basic Protocol), two more sensitive fluorescence techniques (see Alternate Protocols 1 and 2) are presented below. These three procedures cover a range from 5 to 10 ng/ml DNA to 50 µg/ml DNA (see Commentary and Table 4.5.3 therein). Absorbance measurements are straightforward as long as any contribution from contaminants and the buffer components are taken into account. Fluorescence assays are less prone to interference than A260 measurements and are also simple to perform. As with absorbance measurement, a reading from the reagent blank is taken prior to adding the DNA. In instruments where the readout can be set to indicate concentration, a known concentration is used for calibration and subsequent readings are taken in ng/ml or µg/ml DNA. DETECTION OF NUCLEIC ACIDS USING ABSORPTION SPECTROSCOPY Absorption of the sample is measured at several different wavelengths to assess purity and concentration of nucleic acids. A260 measurements are quantitative for relatively pure nucleic acid preparations in microgram quantities. Absorbance readings cannot discriminate between DNA and RNA; however, the ratio of A at 260 and 280 nm can be used as an indicator of nucleic acid purity. Proteins, for example, have a peak absorption at 280 nm that will reduce the A260/A280 ratio. Absorbance at 325 nm indicates particulates in the solution or dirty cuvettes; contaminants containing peptide bonds or aromatic moieties, such as protein and phenol, absorb at 230 nm.
BASIC PROTOCOL
This protocol is designed for a single-beam ultraviolet to visible range (UV-VIS) spectrophotometer. If available, a double-beam spectrophotometer will simplify the measurements, as it will automatically compare the cuvette holding the sample solution to a reference cuvette containing the blank. In addition, more sophisticated double-beam instruments will scan various wavelengths and report the results automatically. Materials 1× TNE buffer (see recipe) DNA sample to be quantitated Calf thymus DNA standard solutions (see recipe) Matched quartz semi-micro spectrophotometer cuvettes (1-cm pathlength) Single- or dual-beam spectrophotometer (ultraviolet to visible) 1. Pipet 1.0 ml of 1× TNE buffer into a quartz cuvette. Place the cuvette in a single- or dual-beam spectrophotometer, read at 325 nm (note contribution of the blank relative to distilled water if necessary), and zero the instrument. Use this blank solution as the reference in double-beam instruments. For single-beam spectrophotometers, remove blank cuvette and insert cuvette containing DNA sample or standard suspended in the same solution as the blank. Take reading. Repeat this process at 280, 260, and 230 nm. It is important that the DNA be suspended in the same solution as the blank. Molecular and Cellular Probes Contributed by Sean R. Gallagher Current Protocols in Cytometry (2000) 4.5.1-4.5.8 Copyright © 2000 by John Wiley & Sons, Inc.
4.5.1 Supplement 14
2. To determine the concentration (C) of DNA present, use the A260 reading in conjunction with one of the following equations: Single-stranded DNA :
C (pmol/µl) = C (µg/ml) =
A
260
10 × S A 260
0.027 A 260 Double-stranded DNA : C (pmol/µl) = 13.2 × S A C (µg/ml) = 260 0.020 A Single-stranded RNA : C (µg/ml) = 260 0.025 Oligonucleotide :
C (pmol/µl) = A
260
×
100 1.5 N
A
+ 0.71 N
C
+ 1.20 N
G
+ 0.84 N
T
where S represents the size of the DNA in kilobases and N is the number or residues of base A, G, C, or T. For double- or single-stranded DNA and single-stranded RNA: These equations assume a 1-cm-pathlength spectrophotometer cuvette and neutral pH. The calculations are based on the Lambert-Beer law, A = εCl, where A is the absorbance at a particular wavelength, C is the concentration of DNA, l is the pathlength of the spectrophotometer cuvette (typically 1 cm), and ε is the extinction coefficient. For solution concentrations given in mol/liter and a cuvette of 1-cm pathlength, ε is the molar extinction coefficient and has units of M−1cm−1. If concentration units of ìg/ml are used, then ε is the specific absorption coefficient and has units of (ìg/ml)−1cm−1. The values of ε used here are as follows: ssDNA, 0.027 (ìg/ml)−1cm−1; dsDNA, 0.020 (ìg/ml)−1cm−1; ssRNA, 0.025 (ìg/ml)−1cm−1. Using these calculations, an A260 of 1.0 indicates 50 ìg/ml double-stranded DNA, ∼37 ìg/ml singlestranded DNA, or ∼40 ìg/ml single-stranded RNA (adapted from Applied Biosystems, 1989). For oligonucleotides: Concentrations are calculated in the more convenient units of pmol/ìl. The base composition of the oligonucleotide has significant effects on absorbance, because the total absorbance is the sum of the individual contributions of each base (Table 4.5.1).
3. Use the A260/A280 ratio and readings at A230 and A325 to estimate the purity of the nucleic acid sample. Ratios of 1.8 to 1.9 and 1.9 to 2.0 indicate highly purified preparations of DNA and RNA, respectively. Contaminants that absorb at 280 nm (e.g., protein) will lower this ratio.
Table 4.5.1 Molar Extinction Coefficients of DNA Basesa
Spectroscopic Analysis Using DNA and RNA Fluorescent Probes
Base
ε1M 260 nm
Adenine Cytosine Guanosine Thymine
15,200 7,050 12,010 8,400
aMeasured at 260 nm; see Wallace and Miyada, 1987.
4.5.2 Supplement 14
Current Protocols in Cytometry
Table 4.5.2
Spectrophotometric Measurements of Purified DNAa
Wavelength (nm)
Absorbance
A260/A280
Conc. (µg/ml)
325 280 260 230
0.01 0.28 0.56 0.30
— — 2.0 —
— — 28 —
aTypical absorbancy readings of highly purified calf thymus DNA suspended in 1× TNE buffer. The concentration of DNA was nominally 25 µg/ml.
Absorbance at 230 nm reflects contamination of the sample by phenol or urea, whereas absorbance at 325 nm suggests contamination by particulates and dirty cuvettes. Light scatter at 325 nm can be magnified 5-fold at 260 nm (K. Hardy, pers. comm.). Typical values at the four wavelengths for a highly purified preparation are shown in Table 4.5.2.
DNA DETECTION USING THE DNA-BINDING FLUOROCHROME HOECHST 33258
ALTERNATE PROTOCOL 1
Use of fluorometry to measure DNA concentration has gained popularity because it is simple and much more sensitive than spectrophotometric measurements. Specific for nanogram amounts of DNA, the Hoechst 33258 fluorochrome has little affinity for RNA and works equally well with either whole-cell homogenates or purified preparations of DNA. The fluorochrome is, however, sensitive to changes in DNA composition, with preferential binding to AT-rich regions. A fluorometer capable of an excitation wavelength of 365 nm and an emission wavelength of 460 nm is required for this assay. Additional Materials (also see Basic Protocol) Hoechst 33258 assay solution (working solution; see recipe) Dedicated filter fluorometer (e.g., Amersham Pharmacia Biotech DQLOO) or scanning fluorescence spectrophotometer (e.g., Shimadzu model RF-5000 or Perkin-Elmer model LS-5B or LS-3B) Fluorometric square glass cuvettes or disposable acrylic cuvettes (Sarstedt) Teflon stir rod 1. Prepare the scanning fluorescence spectrophotometer by setting the excitation wavelength to 365 nm and the emission wavelength to 460 nm. The dedicated filter fluorometer has fixed wavelengths at 365 and 460 nm and does not need adjustment.
2. Pipet 2.0 ml Hoechst 33258 assay solution into cuvette and place in sample chamber. Take a reading without DNA and use as background. If the fluorometer has a concentration readout mode or is capable of creating a standard curve, set instrument to read 0 with the blank solution. Otherwise note the readings in relative fluorescence units. Be sure to take a blank reading for each cuvette used, as slight variations can cause changes in the background reading.
3. With the cuvette still in the sample chamber, add 2 µl DNA standard to the blank Hoechst 33258 assay solution. Mix in the cuvette with a Teflon stir rod or by capping and inverting the cuvette. Read emission in relative fluorescence units or set the concentration readout equal to the final DNA concentration. Repeat measurements with remaining DNA standards using fresh assay solution (take background zero reading and zero instrument if needed).
Molecular and Cellular Probes
4.5.3 Current Protocols in Cytometry
Supplement 14
If necessary, the DNA standards should be quantitated by A260 measurement (Basic Protocol) before being used here. Small-bore tips designed for loading sequencing gels minimize errors of pipetting small volumes. Prerinse tips with sample and make sure that no liquid remains outside the tip after drawing up the sample. Read samples in duplicate or triplicate, with a blank reading taken each time. Unusual or unstable blank readings indicate a dirty cuvette or particulate material in the solution, respectively.
4. Repeat step 3 with unknown samples. A dye concentration of 0.1 ìg/ml is adequate for final DNA concentrations up to ∼500 ng/ml. Increasing the working dye concentration to 1 ìg/ml Hoechst 33258 will extend the assay’s range to 15 ìg/ml DNA, but will limit sensitivity at low concentrations (5 to 10 ng/ml). Sample volumes of ≤10 ìl can be added to the 2.0-ml aliquot of Hoechst 33258 assay solution. ALTERNATE PROTOCOL 2
DNA AND RNA DETECTION WITH ETHIDIUM BROMIDE FLUORESCENCE In contrast to the fluorochrome Hoechst 33258, ethidium bromide is relatively unaffected by differences in the base composition of DNA. Although capable of detecting nanogram levels of DNA, ethidium bromide is not as sensitive as Hoechst 33258 and will also bind to RNA. In preparations of DNA with minimal RNA contamination or with DNA samples having an unusually high guanine and cytosine (GC) content where the Hoechst 33258 signal can be quite low, ethidium bromide offers a relatively sensitive alternative to the more popular Hoechst 33258 DNA assay. A fluorometer capable of an excitation wavelength of 302 or 546 nm and an emission wavelength of 590 nm is required for this assay. Additional Materials (also see Basic Protocol) Ethidium bromide assay solution (see recipe) 1. Pipet 2.0 ml ethidium bromide assay solution into cuvette and place in sample chamber. Set excitation wavelength to 302 nm or 546 nm and emission wavelength to 590 nm. Take an emission reading without DNA and use as background. If the instrument has a concentration readout mode or is capable of creating a standard curve, set instrument to read 0 with the blank solution. Otherwise note the readings in relative fluorescence units. The excitation wavelength of this assay can be either in the UV range (∼302 nm) using a quartz cuvette or in the visible range (546 nm) using a glass cuvette. In both cases the emission wavelength is 590 nm.
2. Read and calibrate these samples as described in step 3 of the Hoechst 33258 assay. 3. Read emissions of the unknown samples as in step 4 of the Hoechst 33258 assay. A dye concentration of 5 ìg/ml in the ethidium bromide assay solution is appropriate for final DNA concentrations up to 1000 ng/ml. 10 ìg/ml ethidium bromide in the ethidium bromide assay solution will extend the assay’s range to 10 ìg/ml DNA, but is used only for DNA concentrations >1 ìg/ml. Sample volumes of up to 10 ìl can be added to the 2.0-ml aliquot of ethidium bromide assay solution.
Spectroscopic Analysis Using DNA and RNA Fluorescent Probes
4.5.4 Supplement 14
Current Protocols in Cytometry
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Calf thymus DNA standard solutions Kits containing calf thymus DNA standard for fluorometry are available (Fluorometry Reference Standard Kits, Hoefer). Premeasured, CsCl-gradient-purified DNA of defined GC content, for use in absorption and fluorometric spectroscopy, is available from Sigma (e.g., calf thymus DNA, 42% GC; Clostridium perfringens DNA, 26.5% GC). Ethidium bromide assay solution Add 10 ml of 10× TNE buffer (see recipe) to 89.5 ml H2O. Filter through a 0.45-µm filter, then add 0.5 ml of 1 mg/ml ethidium bromide. Add the dye after filtering, as ethidium bromide will bind to most filtration membranes. CAUTION: Ethidium bromide is hazardous; wear gloves and use appropriate care in handling, storage, and disposal.
Hoechst 33258 assay solutions Stock solution: Dissolve in H2O at 1 mg/ml. Stable for ∼6 months at 4°C. Working solution: Add 10 ml of 10× TNE buffer (see recipe) to 90 ml H2O. Filter through a 0.45-µm filter, then add 10 µl of 1 mg/ml Hoechst 33258. Hoechst 33258 is a fluorochrome dye with a molecular weight of 624 and a molar extinction coefficient of 4.2 × 104 M−1cm−1 at 338 nm. The dye is added after filtering because it will bind to most filtration membranes. CAUTION: Hoechst 33258 is hazardous; use appropriate care in handling, storage, and disposal.
TNE buffer, 10× 100 mM Tris base 10 mM EDTA 2.0 M NaCl Adjust pH to 7.4 with concentrated HCl As needed, dilute with H2O to desired concentration COMMENTARY Background Information In deciding what method of nucleic acid measurement is appropriate, three issues are critical: specificity, sensitivity, and interfering substances. Properties of the three assays described in this section are listed in Table 4.5.3. The traditional method for determining the amount of DNA in solution is by measuring absorbance at 260 nm. Because many potential contaminants of DNA and RNA preparations also absorb in the UV range, absorption spectroscopy is a reliable method to assess both the purity of a preparation and the quantity of DNA or RNA present. Absorption spectroscopy does have serious limitations. Relatively large amounts of DNA are required to get accurate readings—for example, 500 ng/ml DNA is equivalent to only 0.01 A260 units. Furthermore,
the method cannot discriminate between RNA and DNA, and UV-absorbing contaminants such as protein will cause discrepancies. The assay using Hoechst 33258 dye (Alternate Protocol 1) is the only procedure in common use that is specific for DNA (i.e., it does not measure RNA). This assay is the method of choice for rapid measurement of low quantities of DNA, with a detection limit of ∼1 ng DNA. Concentrations of DNA in both crude cell lysates and purified preparations can be quantified (Labarca and Paigen, 1980). Because the assay quantifies a broad range of DNA concentrations—from 10 ng/ml to 15 µg/ml—it is useful for the measurement of both small and large amounts of DNA (e.g., in verifying DNA concentrations prior to performing electrophoretic separations and Southern blots). The
Molecular and Cellular Probes
4.5.5 Current Protocols in Cytometry
Supplement 14
Table 4.5.3 Properties of Absorbance and Fluorescence Spectrophotometric Assays for DNA and RNA
Property Sensitivity (µg/ml) DNA RNA Ratio of signal (DNA/RNA)
Absorbance (A260) 1-50 1-40
0.01-15 n.a.
0.1-10 0.2-10
0.8
400
2.2
Hoechst 33258 assay is also useful for measuring products of the polymerase chain reaction (PCR) synthesis. Upon binding to DNA, the fluorescence characteristics of Hoechst 33258 change dramatically, showing a large increase in emission at ∼458 nm. Hoechst 33258 is nonintercalating and apparently binds to the minor groove of the DNA, with a marked preference for AT sequences (Portugal and Waring, 1988). The fluorochrome 4′,6-diamidino-2-phenylindole (DAPI; Daxhelet et al., 1989) is also appropriate for DNA quantitation, although it is not as commonly used as Hoechst 33258. DAPI is excited with a peak at 344 nm. Emission is detected at ∼466 nm, similar to Hoechst 33258. Ethidium bromide is best known for routine staining of electrophoretically separated DNA and RNA, but it can also be used to quantify both DNA and RNA in solution (Le Pecq, 1971). Unlike Hoechst 33258, ethidium bromide fluorescence is not significantly impaired by high GC content. The ethidium bromide assay (with excitation at 546 nm) is ∼20-fold less sensitive than the Hoechst 33258 assay.
Critical Parameters
Spectroscopic Analysis Using DNA and RNA Fluorescent Probes
Fluorescence H33258 EB
Care should be taken when handling sample cuvettes in all spectrophotometric procedures. Fluorometers use cuvettes with four optically clear faces, because excitation and emitting light enter and leave the cuvette through directly adjacent sides. Thus, fluorometric cuvettes should be held by the upper edges only. In contrast, transmission spectrophotometers use cuvettes with two opposite optical windows, with the sides frosted for easy handling. It is important to check that the optical faces of cuvettes are free of fingerprints and scratches. In addition, for accurate absorbance readings, spectrophotometer cuvettes must be perfectly matched and scrupulously clean. Proteins in general have A280 readings considerably lower than nucleic acids on an equivalent weight basis. Thus, even a small increase
in the A280 relative to A260 (i.e., a lowering of the A260/A280 ratio) can indicate severe protein contamination. Other commonly used buffer components absorb strongly at 260 nm and can cause interference if present in high enough concentrations. EDTA, for example, should not be present at ≥10 mM. Sensitivity of the Hoechst 33258 fluorescence assay decreases with nuclease degradation, increasing GC content, or denaturation of DNA (Labarca and Paigen, 1980; Stout and Becker, 1982). Increased temperature of the assay solution and ethidium bromide contamination also decrease the Hoechst 33258 signal. Sodium dodecyl sulfate (>0.01% final concentration) also interferes with accurate readings (Cesarone et al., 1979). The pH of the assay solution is critical to sensitivity and should be ∼7.4 (Stout and Becker, 1982; Labarca and Paigen, 1980). At a pH <6.0 or >8.0 the background becomes much higher and there is a concomitant loss of fluorescence enhancement. High-quality double-stranded DNA is recommended. With very small fragments of DNA, the Hoechst 33258 dye binds to doublestranded DNA only. Thus, the assay will not work with single-stranded oligomers. Linear and circular DNA give approximately the same levels of fluorescence (Daxhelet et al., 1989). When preparing DNA standards, an attempt should be made to equalize the GC content of the standard DNA and that of the sample DNA. In most situations, salmon sperm or calf thymus DNA is suitable. An extensive list of estimated GC content for various organisms is available (Marmur and Doty, 1962). Eukaryotic cells vary somewhat in GC content but are generally in the range of 39% to 46%. Within this range, the fluorescence per microgram of DNA does not vary substantially. In contrast, the GC content of prokaryotes can vary from 26% to 77%, causing considerable variation in the fluorescence signal. In these situations, the sample DNA should be first quantitated via transmission spectroscopy and compared to a readily
4.5.6 Supplement 14
Current Protocols in Cytometry
Relative flourescence units
1000 900 800 700 600 500 400 300 200 100 0
A
H33258 Ex: 365 nm r 2: = 0.9999 Ex: 455 nm
60 50 40 30 20 10 0
H33258
0
10
20
ethidium bromide Ex: 546 nm r 2: = 0.9993 Ex: 590 nm
50
30
B
40 30 20 10 0
0
100
200
300
400
DNA (ng/ml)
500
Figure 4.5.1 Fluorochrome Hoechst 33258 (H33258) (A) and ethidium bromide (B) DNA concentration standard curves. Assays were performed as described in alternate protocols, at indicated excitation and emission wavelengths. The concentrations of H33258 and ethidium bromide were 0.1 and 5 µg/ml, respectively. Assays contained the indicated concentrations of calf thymus DNA standards suspended in a final volume of 2.0 ml. Inset shows low DNA concentration curve for the H33258 assay. Note that, under these conditions, H33258 produces ∼20 times more relative fluorescence units than ethidium bromide. A Shimadzu RF-5000 scanning fluorescence spectrometer was used for both assays.
available standard (e.g., calf thymus DNA). Future measurements would then use calf thymus as a standard, but with a correction factor for difference in fluorescence yield between the two DNA types. For further troubleshooting information, see Van Lancker and Gheyssens (1986), in which the effects of interfering substances on the Hoechst 33258 assay (and several other assays) are compared. In the ethidium bromide assay, singlestranded DNA gives approximately half the signal as double-stranded calf thymus DNA. Ribosomal RNA also gives about half the fluorescent signal as double-stranded DNA, and RNase and DNase both severely decrease the signal. Closed circular DNA also binds less ethidium bromide than nicked or linear DNA. Further critical parameters of the ethidium bromide assay are described by Le Pecq (1971).
(r2) of 0.98 to 0.99 (Fig. 4.5.1). Table 4.5.3 provides a comparison of the sensitivities and specificities of the three assays.
Anticipated Results
Literature Cited
The detection limit of absorption spectroscopy will depend on the sensitivity of the spectrophotometer and any UV-absorbing contaminants that might be present. The lower limit is generally ∼0.5 to 1 µg nucleic acid. Typical values for a highly purified sample of DNA are shown in Table 4.5.2. For the Hoechst 33258 and ethidium bromide assays, a plot of relative fluorescence units or estimated concentration (y axis) versus actual concentration (x axis) typically produces a linear regression with a correlation coefficient
Applied Biosystems. 1989. User Bulletin Issue 11, Model No. 370. Applied Biosystems, Foster City, Calif.
Time Considerations The three assays described can be performed in a short period of time. In a well-planned series of assays, 50 samples can be prepared and read comfortably in 1 hr. Although some error might be introduced, DNA samples can be sequentially added to the same cuvette containing working dye solution. The increase in fluorescence with each sample is noted and subtracted from the previous reading to give relative fluorescence or concentration of the new sample, eliminating the need to change solutions for each sample. Be certain that the final amount of DNA does not exceed the linear portion of the assay.
Cesarone, C.F., Bolognesi, C., and Santi, L. 1979. Improved microfluorometric DNA determination in biological material using 33258 Hoechst. Anal. Biochem. 100:188-197. Daxhelet, G.A., Coene, M.M., Hoet, P.P., and Cocito, C.G. 1989. Spectrofluorometry of dyes with DNAs of different base composition and conformation. Anal. Biochem. 179:401-403. Molecular and Cellular Probes
4.5.7 Current Protocols in Cytometry
Supplement 14
Labarca, C. and Paigen, K. 1980. A simple, rapid, and sensitive DNA assay procedure. Anal. Biochem. 102:344-352. Le Pecq, J.-B. 1971. Use of ethidium bromide for separation and determination of nucleic acids of various conformational forms and measurement of their associated enzymes. In Methods of Biochemical Analysis, Vol. 20 (D. Glick, ed.) pp. 41-86. John Wiley & Sons, New York. Marmur, J. and Doty, P. 1962. Determination of the base composition of deoxyribonucleic acid from its thermal denaturation temperature. J. Mol. Biol. 5:109-118. Portugal, J. and Waring, M.J. 1988. Assignment of DNA binding sites for 4′,6-diamidine-2- phenylindole and bisbenzimide (Hoechst 33258): A comparative footprinting study. Biochem. Biophys. Acta 949:158-168. Stout, D.L. and Becker, F.F. 1982. Fluorometric quantitation of single-stranded DNA: A method applicable to the technique of alkaline elution. Anal. Biochem. 127:302-307.
Van Lancker, M. and Gheyssens, L.C. 1986. A comparison of four frequently used assays for quantitative determination of DNA. Anal. Lett. 19:615-623. Wallace, R.B. and Miyada C.G. 1987. Oligonucleotide probes for the screening of recombinant DNA libraries. In Methods of Enzymology, Vol. 152: Guide to Molecular Cloning Techniques (S.L. Berger and A.R. Kimmel, eds.) pp. 432442. Academic Press, San Diego.
Key References Labarca and Paigen, 1980. See above. Contains a detailed description of the Hoechst 33258 fluorometric DNA assay.
Contributed by Sean R. Gallagher Motorola Corporation Tempe, Arizona
Spectroscopic Analysis Using DNA and RNA Fluorescent Probes
4.5.8 Supplement 14
Current Protocols in Cytometry
CHAPTER 5 Specimen Handling, Storage, and Preparation INTRODUCTION
T
his chapter describes well-established, quality-controlled methods and procedures for the handling, storage, and preparation of subcellular components (such as chromosomes), and tissue for flow cytometric assay. Numerous assays can be performed with samples prepared following these fundamental techniques.
Common to all flow cytometric procedures is the requirement for a single-cell, singlenucleus, or single-chromosome preparation. Although human blood is obtained conveniently in cell suspension, specific procedures must still be followed, as discussed in UNIT 5.1, to ensure the quality of the resulting single-cell preparation, which can be affected by a variety of parameters. Conditions for obtaining, transporting, and storing material prior to assay should prevent or minimize artefactual debris, degradation of protein, RNA, or DNA, and loss of functional activity. In many cases it is also desirable to obtain specific blood cell fractions desired for analysis in enriched or purified form; procedures for obtaining such fractions are also detailed in UNIT 5.1. The same critical parameters that apply to blood samples are equally important in the preparation of other human tissues. Unlike blood cells, such tissues are not initially obtained initially as single-cell suspensions but must be prepared as such. The cell parameters or components to be measured (e.g., protein, DNA, or RNA) and the localization of the markers (e.g., cell surface, cytoplasmic, or nuclear) will influence the choice of dissociation technique as well as the optimal methods for collection, transport, and storage of specimens. Several methods to accommodate flow cytometric analysis of both whole cells and nuclei are described in UNIT 5.2. Flow cytometric analysis and sorting of either animal or plant chromosomes is an invaluable method of obtaining DNA to be used for physical gene mapping, isolation of molecular markers, or construction of chromosome-specific DNA libraries, among other applications. Varying sample handling, storage, and preparation conditions are needed depending on the cell type used as the chromosome source. UNIT 5.3 details procedures for accumulation of plant cells in metaphase, for preparation of plant chromosome suspensions, and for flow cytometric analysis and sorting; methods for further processing of sorted plant chromosomes are also provided. Specific applications, such as those described in other chapters of this manual, may require a unique method that works optimally with that application and may not be included in this chapter. These applications may require a different fixative, medium, or time frame for preparation. Therefore, users are advised to review their specific flow cytometry applications prior to the preparation of cells. Alberto Orfao Specimen Handling, Storage, and Preparation Contributed by Alberto Orfao Current Protocols in Cytometry (1999) 5.0.1 Copyright © 1999 by John Wiley & Sons, Inc.
5.0.1 Supplement 9
Handling, Storage, and Preparation of Human Blood Cells
UNIT 5.1
Human peripheral blood samples are probably the most common specimens submitted to the flow cytometry laboratory for analysis. How specimens are obtained, handled, and stored is intimately related to the reason for obtaining them. Although the majority of specimens are for lymphocyte immunophenotyping, peripheral blood is also commonly obtained for the analysis of platelets, neutrophil function, lymphocyte mitogenesis or blastogenesis, and intracellular antigen expression, and for reticulocyte enumeration. Many of these assays impose specific requirements for specimen acquisition, transport, and storage. Moreover, although flow cytometry can be used to analyze heterogeneous cell populations, it is often desirable to enrich, purify, or separate cell populations prior to performing flow cytometry. Preparation of cells for flow cytometry will vary depending on which cell lineage is to be examined and what assay is to be performed on the cells. Therefore, it is not possible to present a single, generic protocol for acquisition and transportation of all specimens. However, as a general principle, cells should be examined as soon as possible after the specimen is obtained and in as close to the native state as possible (i.e., with the least amount of processing possible), and certain general procedures for obtaining blood samples apply. 1. Before taking the sample, read thoroughly the procedure for the specific assay to be performed. Contact the venipuncturist with any information concerning the volume of blood required for the assay and the anticoagulant to be used, as well as any other pertinent information. 2. Procure blood samples from the venipuncture team as quickly as possible. Maintain blood sample at room temperature unless the specific protocol dictates otherwise. In general, avoid subjecting specimens to extremes of temperature or holding them for prolonged periods of time prior to processing. 3. Add tissue culture medium, anticoagulant, preservatives, or other additives as directed by the specific protocol. 4. Be sure to label all specimens properly with type of specimen, time and date of collection, patient name (if appropriate), and test to be performed. Also note the type of anticoagulant used if not evident from the color of the stopper or another indicator. 5. Store the specimen as appropriate for the assay to be performed. If no specific additives are indicated, maintain blood sample aseptically at room temperature until needed; this temperature is usually acceptable for short-term storage, although this may not be universally true. Gentle rocking may help in preserving cellular aggregation. (Table 5.1.1 provides general recommendations for anticoagulants and storage times for blood samples obtained for a variety of common assays.) This unit presents protocols for separating white cells by lysing erythrocytes (see Basic Protocol 1), isolating mononuclear cells using trilineage separation (see Basic Protocol 2), enrichment of lymphocytes by adherence (see Support Protocol), and assorted methods of isolating or enriching specific cell populations, including peripheral blood monocytes (see Basic Protocol 3), neutrophils (see Basic Protocol 4), T lymphocytes (see Basic Protocols 5 and 7), B lymphocytes (see Basic Protocols 6, 7, and 8), NK cells (see Basic Protocol 8), stem cells (see Basic Protocol 9), and platelets (see Basic Protocol 10). Specimen Handling, Storage, and Preparation Contributed by J. Philip McCoy, Jr. Current Protocols in Cytometry (1998) 5.1.1–5.1.13 Copyright © 1998 by John Wiley & Sons, Inc.
5.1.1 Supplement 5
Table 5.1.1 Recommended Anticoagulants and Storage Times for Commonly Performed Assays
Assay
Anticoagulant
Time limitation
Lymphocyte immunophenotyping Myeloid immunophenotyping Neutrophil function Platelet activation Platelet markers Reticulocyte enumeration DNA analysis
Sodium heparin or EDTA
Store ≤72 hr
EDTA
Use immediately
Sodium heparin or EDTA EDTA EDTA EDTA Sodium heparin or EDTA
Use immediately Use immediately Use immediately Store ≤72 hr at 4°C Use immediate for cell-cycle analysis; store ≤72 hr for ploidy analysis
CAUTION: When working with human blood, cells, or infectious agents, biosafety practices should be followed; see Critical Parameters for further information. NOTE: All solutions and equipment coming in contact with cells must be sterile, and proper sterile techniques should be used accordingly. BASIC PROTOCOL 1
PREPARATION OF WHITE CELL SUSPENSION BY LYSIS OF ERYTHROCYTES Although erythrocytes can be separated from mononuclear cells by density-gradient separation (see Basic Protocol 2), many laboratories prefer lysis methods to eliminate erythrocytes from various specimens. Lysis is much quicker than gradient separation and in general leaves the remaining white cell populations relatively unperturbed. Blood samples may be treated with any anticoagulant. This procedure, in which erythrocytes are lysed with ammonium chloride, may be used for unstained blood or blood that has already been incubated with monoclonal antibodies. In general, this method will not affect the pattern of staining observed for most lymphoid markers. The viability of white blood cells subjected to this treatment is good. Commercial reagents such as FACSLyse, ImmunoLyse, and Optilyse can be used in place of ammonium chloride to lyse erythrocytes. However, as FACSLyse contains a fixative, staining for cell surface markers on leukocytes should be performed prior to lysis of the erythrocytes with this reagent. Stain whole blood with antibodies, then lyse erythrocytes according to manufacturer’s instructions. Optilyse may be used without subsequent washing of the leukocytes. Materials Whole blood sample 1× ammonium chloride lysing solution (prepare fresh from 10× stock; see APPENDIX 2A) or FACSLyse (Becton Dickinson Immunocytometry), ImmunoLyse (Coulter), or Optilyse (Immunotech) Phosphate-buffered saline (PBS; APPENDIX 2A) 1. Place 200 µl whole blood sample in a centrifuge tube and add 3 ml fresh 1× ammonium chloride lysing solution. Incubate 10 min at room temperature.
Handling, Storage, and Preparation of Human Blood Cells
2. Centrifuge 5 min at 300 × g, room temperature (22° to 25°C). 3. Discard supernatant. Resuspend cells in 2 ml PBS. Centrifuge 5 min at 300 × g, room temperature.
5.1.2 Supplement 5
Current Protocols in Cytometry
4. Discard supernatant and repeat wash with 2 ml PBS. 5. Resuspend cells as necessary for specified assay. Use immediately or store as indicated for assay. ISOLATION OF MONONUCLEAR CELLS BY DENSITY GRADIENT SEPARATION
BASIC PROTOCOL 2
Trilineage separation methods (Boyum, 1968, 1977) are used when purification of cell populations is required rather than simple removal of erythroid contaminants. Densitygradient separation techniques may not yield as many cells as the simple lysis methods, but they have other distinct advantages. Separating cells by Ficoll-Hypaque centrifugation often decreases the cytometry time for acquisition and removal of nonviable cells. Materials 1.077 g/ml Ficoll-Hypaque (Pharmacia Biotech) or Histopaque-1077 (Sigma) Anticoagulated blood in heparin or EDTA Phosphate-buffered saline (PBS; APPENDIX 2A) Tissue culture medium (optional; APPENDIX 3B) 15- and 50-ml conical centrifuge tubes Centrifuge 1. With a sterile pipet, place the Ficoll-Hypaque solution into a 50-ml conical centrifuge tube, using 2 ml Ficoll-Hypaque per ml blood. Volume of Ficoll-Hypaque may vary with brand used. Consult manufacturer’s recommendations.
2. Mix anticoagulated blood with an equal volume of PBS. 3. Slowly layer the diluted blood over the Ficoll-Hypaque solution by gently pipetting the diluted blood down the side of the tube containing the Ficoll-Hypaque. 4. Centrifuge 40 min at 400 × g, 22°C, with no brake. 5. Using a sterile Pasteur pipet, carefully remove the mononuclear cells, located at the interface between the plasma (upper layer) and the Ficoll-Hypaque (bottom). 6. Transfer the aspirated mononuclear cells to a 15-ml conical tube. Add 10 ml PBS or tissue culture medium and mix thoroughly. Centrifuge 10 min at 400 × g, 4°C. 7. Discard the supernatant and repeat wash with PBS or tissue culture medium as needed. ENRICHMENT OF LYMPHOCYTES FROM MONONUCLEAR CELL PREPARATIONS BY DEPLETION OF MONOCYTOID CELLS THROUGH ADHERENCE TO PLASTIC
SUPPORT PROTOCOL
The density-gradient procedure described in Basic Protocol 2 yields a preparation of mononuclear cells that may include monocytoid cells. For most individuals, this protocol provides a specimen with more lymphoid than monocytoid cells. However, in some instances, even further purity of the lymphocytes is desired. This protocol enriches the lymphocyte population by depleting monocytoid cells from the mononuclear cell preparation using adherence to a plastic tissue culture flask. Additional Materials (also see Basic Protocol 2) Mononuclear cell preparation (see Basic Protocol 2) Tissue culture medium (APPENDIX 3B) containing ≥10% FBS (APPENDIX 2A)
Specimen Handling, Storage, and Preparation
5.1.3 Current Protocols in Cytometry
75-cm2 or 150-cm2 plastic tissue culture flask Humidified 37°C, 5% CO2 incubator Additional reagents and equipment for assessing cell viability (UNIT 9.2) 1. Suspend mononuclear cells in tissue culture medium to a final concentration of 1–2 × 106 cells/ml. 2. Transfer cell suspension to tissue culture flask. An appropriate volume for a 150-cm2 flask is 50 ml.
3. Incubate cells 1 hr in a humidified 37°C, 5% CO2 incubator. 4. Gently pour supernatant containing nonadherent lymphoid cells into a 50-ml centrifuge tube. Rinse flask gently with 10 ml tissue culture medium and add to centrifuge tube. 5. Check cell suspension manually or with a hematology analyzer to determine percentage of monocytoid cells remaining in the specimen. If additional depletion of monocytoid cells is desired, repeat steps 1 to 4. 6. Assess viability of lymphocytes using trypan blue exclusion (APPENDIX 3B) or flow cytometry (UNIT 9.2). BASIC PROTOCOL 3
ISOLATION OF PERIPHERAL BLOOD MONOCYTES BY GRADIENT CENTRIFUGATION WITH FICOLL-HYPAQUE FOLLOWED BY PERCOLL Peripheral blood monocytes are isolated from blood by centrifugation through a FicollHypaque or Histopaque gradient (Denholm and Wolber, 1991) followed by Percoll. Materials Venous blood in sodium citrate Phosphate-buffered saline (APPENDIX 2A) 1.077 g/ml Ficoll-Hypaque (Pharmacia Biotech) or Histopaque-1077 (Sigma) 10× Hanks balanced salt solution (HBSS; APPENDIX 2A) Percoll, specific gravity 1.130 g/ml (Sigma) Tissue culture medium (optional; APPENDIX 3B) 50-ml conical centrifuge tubes 10 × 15–cm round bottom polypropylene tubes silanized with Surfasil 1. Mix 30 to 40 ml of blood with an equal volume of PBS. Layer diluted blood over Ficoll-Hypaque solution, using 3 parts diluted blood to 2 parts Ficoll-Hypaque. Centrifuge 30 min at 500 × g, 25°C. 2. Using a sterile Pasteur pipet, carefully collect mononuclear cells, located at the interface between the plasma (upper layer) and Ficoll-Hypaque solution (bottom of tube). 3. Transfer cells to a 50-ml centrifuge tube. Add 10 ml PBS. Centrifuge 10 min at 400 × g, 4°C. 4. Discard supernatant and resuspend cells in 4 ml PBS.
Handling, Storage, and Preparation of Human Blood Cells
5. Mix 1.65 ml of 10× Hanks balanced salt solution with 10 ml Percoll. Adjust pH to 7.0 with 0.1 N HCl (~30 µl should be sufficient). 6. Mix 8 ml Percoll solution with 4 ml mononuclear cells in a silanized 10 × 1.5–cm round-bottom polypropylene tube. Mix thoroughly by inverting tube three or four times. Centrifuge 25 min at 370 × g, 25°C.
5.1.4 Current Protocols in Cytometry
7. Aspirate monocytes by gentle pipetting into a clean test tube. Add PBS or tissue culture medium as needed. After centrifugation, monocytes appear as a cloudy layer in the top 5 mm of the gradient.
8. Wash the cells in PBS or tissue culture medium. ISOLATION OF NEUTROPHILS BY PERCOLL GRADIENT CENTRIFUGATION
BASIC PROTOCOL 4
In general, neutrophils require more prompt and delicate handling than lymphoid cells. Specimens more than a few hours old are not optimal for use in neutrophil isolation procedures. Materials Blood sample in acid citrate dextrose, formula A 0.6% (w/v) dextran in 0.9% (w/v) NaCl 0.9% NaCl 1.10 g/ml, 1.095 g/ml, and 1.085 g/ml Percoll in phosphate-buffered saline (PBS; APPENDIX 2A) 15-ml polypropylene tubes 1. Mix 45 ml of blood with 10 ml of 6% dextran in 0.9% NaCl. Allow to stand for 1 hr at 25°C. 2. Remove leukocyte-rich supernatant and centrifuge 10 min at 400 × g, 4°C. 3. Discard supernatant and resuspend cells in 0.9% NaCl to a final concentration of 5 × 107 cells/ml. 4. Prepare a discontinuous Percoll gradient: transfer 4 ml of 1.10 g/ml Percoll in PBS to a 15-ml polypropylene tube; then overlay this with 3 ml of 1.095 g/ml Percoll and finally with 3 ml of 1.085 g/ml Percoll. 5. Layer 5 ml of cell suspension over the Percoll gradient. Centrifuge 30 min at 700 × g, 25°C, in a swinging-bucket rotor with no braking. 6. Sample will contain three bands and a cell pellet. Using sterile polypropylene pipets, collect the two lower bands, which contain the enriched neutrophil fraction. Wash in 0.9% NaCl. ISOLATION OF T LYMPHOCYTES BY ERYTHROCYTE ROSETTING Erythrocyte rosetting (Madsen et al., 1979; Indiveri et al., 1980) is one of several methods used to isolate T lymphocytes. Other methods include panning (see Basic Protocol 7) and separation using magnetic beads coated with appropriate antibodies (see Basic Protocol 6). E-rosetting, a classical method for isolating T cells, involves separating T cells from other mononuclear cells by exploiting the unique ability of the lymphocytes to bind to and form rosettes with sheep red blood cells. The method remains very useful and inexpensive today. Materials Sheep red blood cells in Alsevers solution Phosphate-buffered saline, isotonic, pH 7.0 (PBS; APPENDIX 2A) 2-aminoethylisothiouronium hydrobromide (AET) solution: 0.143 M AET in H2O, pH 9.0 (adjust pH with 4 N NaOH), prepared fresh for each use Tissue culture medium (APPENDIX 3B)
BASIC PROTOCOL 5
Specimen Handling, Storage, and Preparation
5.1.5 Current Protocols in Cytometry
FBS (heat inactivated; APPENDIX 2A) 1–2 × 107 cell/ml mononuclear cell preparation (see Basic Protocol 2) prepared fresh in tissue culture medium 1.077 g/ml Ficoll-Hypaque (Pharmacia Biotech) or Histopaque-1077 (Sigma) 10× ammonium chloride lysing solution (APPENDIX 2A; prepare fresh) 15-ml polystyrene round-bottom centrifuge tubes 15- and 50-ml conical centrifuge tubes 1. Place 6 ml sheep red blood cells in a 50-ml conical centrifuge tube. Add 30 ml isotonic PBS at room temperature. Centrifuge 10 min at 400 × g, 4°C. Discard supernatant and repeat wash two more times. 2. Transfer 2 ml of washed sheep erythrocytes to a 50-ml conical centrifuge tube. Add 10 ml AET solution. Vortex the mixture gently and incubate 15 min at 37°C, mixing gently every 5 min. 3. Wash the AET-treated cells three times with 30 ml of 4°C PBS as in step 1. Resuspend cells in 50 ml tissue culture medium. 4. Place 1 ml mononuclear cells in a 15-ml round bottom centrifuge tube. Add 2 ml AET-treated sheep erythrocytes and 2 ml heat-inactivated FBS. Mix well and incubate 10 min at 37°C. 5. Centrifuge 10 min at 200 × g, 4°C. Remove the supernatant and incubate pellet 60 min at 4°C. 6. Gently resuspend pelleted mixture with a pipet and slowly layer over an equal volume of Ficoll-Hypaque solution in a 15-ml conical centrifuge tube. Centrifuge 40 min at 400 × g, room temperature. T cells will pellet at the bottom of the centrifuge tube.
7. Remove T cells with a pipet. To lyse erythrocytes, add 15 ml ammonium chloride lysing solution to pellet and let stand 10 min (until solution becomes transparent), then transfer to 50-ml conical tube containing tissue culture medium. Centrifuge 10 min at 400 × g, 4°C. 8. Discard supernatant and resuspend T cells in tissue culture medium at a concentration appropriate for their desired use. BASIC PROTOCOL 6
Handling, Storage, and Preparation of Human Blood Cells
ISOLATION OF B LYMPHOCYTES BY MAGNETIC BEAD–ANTIBODY SEPARATION Lymphoid subsets can be isolated by either positive or negative selection. Positive selection is more direct, but in some instances the cells of interest may suffer from being manipulated. In those instances, negative selection can be applied by using antibodies that will bind all lymphoid subsets other than the one of interest. The positive selection procedure that follows (based on procedures developed by Rasmussen et al., 1992, and Molday et al., 1977) employs magnetic beads coated with anti-CD19 antibodies to bind B cells. Materials 1% FBS/PBS: phosphate-buffered saline (PBS; APPENDIX 2A) with 1% (v/v) FBS (heat inactivated; APPENDIX 2A), 4°C Heparinized blood Tissue culture medium with 10% FBS (APPENDIX 2A) Detach-a-bead (Dynal) or other anti-mouse Fab antibody (optional)
5.1.6 Current Protocols in Cytometry
Anti-CD19-coated magnetic beads (Dynal) Magnetic separation device (e.g., Dynal) 75-cm2 tissue culture flask 15-ml polypropylene test tube Humidified 37°C, 5% CO2 incubator Bind B cells with anti-CD19-coated magnetic beads 1. Thoroughly mix 100 µl of resuspended beads with 10 ml of 4°C 1% FBS/PBS in a 15-ml polypropylene test tube. Place tube in magnetic separation device for 5 to 10 min to draw beads to the side of the tube. Gently remove the PBS by pipetting or aspiration. 2. Repeat suspension and separation as in step 1. Resuspend cells in 2 ml of 4°C 1% FBS/PBS. 3. Mix 10 ml heparinized blood with 30 ml of 4°C 1% FBS/PBS. Place diluted blood in tissue culture flask or polypropylene centrifuge tube. Add washed magnetic beads to the flask and incubate 30 min at 4°C with gentle rocking. 4. Separate the bead-cell complexes using a magnet or by placing tube in magnetic separation device for 5 to 10 min. For this separation, cells may be transferred to a different vessel if necessary.
5. Aspirate and discard fluid from the tube or separation vessel. Remove vessel from magnetic separation device. 6. Resuspend bead-cell complexes in 10 ml of 4°C 1% FBS/PBS. Incubate 30 min at 4°C with gentle rocking. 7. Repeat separation and washing with 1% FBS/PBS as many times as desired. Dissociate B cells from beads by any of several methods, including the following: Capping: 8a. Suspend bead-cell complexes in 10 ml complete tissue culture medium with 10% FBS in a tissue culture flask and incubate 12 to 24 hr at 37°C in a humidified 37°C, 5% CO2 incubator. 9a. Expose flask to a magnet or place in magnetic separation device for 5 to 10 min. Exposure to anti-mouse antibodies (Detach-a-bead): 8b. Suspend bead-cell complexes in 100 to 200 µl complete tissue culture medium with 10% FBS in a polypropylene test tube. Add 10 to 20 µl Detach-a-bead or an appropriate volume of other anti-mouse Fab antibody. Incubate 1 hr at 25°C with gentle rocking. 9b. Add 5 to 10 ml tissue culture medium with 10% FBS, mix thoroughly, and expose tube to a magnet or place in magnetic separation device for 5 to 10 min. 10. Remove and save suspension, which contains unbound B cells. Repeat capping or exposure to anti-mouse antibodies as desired to remove more beads. 11. Centrifuge 10 min at 200 × g, 4°C. Resuspend cells in tissue culture medium. Specimen Handling, Storage, and Preparation
5.1.7 Current Protocols in Cytometry
BASIC PROTOCOL 7
ISOLATION OF T OR B LYMPHOCYTES BY PANNING T or B lymphocyte subsets may be purified through the use of panning techniques (Wysocki and Sato, 1978). These are very similar to the magnetic bead separation technique described in Basic Protocol 6, except the antibodies are used to adhere the cells of interest to a plastic vessel while the unwanted cells are washed away. Yield will vary with cell type to be isolated and the antibody used. Materials Highly specific anti-mouse IgG (Dako) 0.05 M Tris⋅Cl, pH 9.5 Specific mouse monoclonal antibodies (e.g., CD19 for B lymphocytes and CD3 for mature T lymphocytes) 10% sodium azide (optional) Phosphate-buffered saline, pH 7.2 (PBS; APPENDIX 2A) Mononuclear cell preparation (see Basic Protocol 2) 5% and 1% FBS/PBS: PBS with 5% or 1% (v/v) FBS (heat inactivated; APPENDIX 2A), 4°C Complete tissue culture medium (APPENDIX 3B) 100 × 15 mm–petri dish, bacteriological grade 15-ml polystyrene centrifuge tubes Additional reagents and equipment for counting cells and assessing cell viability by trypan blue exclusion (APPENDIX 3B) or using propidium iodide (UNIT 9.2) 1. Dilute anti-mouse IgG to ∼10 µg/ml in 0.05 M Tris⋅Cl, pH 9.5. Place 10 ml in a petri dish and incubate 1 hr at room temperature. 2. Add 2 × 107 mononuclear cells to a 15-ml polystyrene tube. Add specific monoclonal antibody and, optionally (depending on what the cells will be used for), 0.01% sodium azide. Incubate 15 min to 1 hr, at room temperature if sodium azide is used or at 4°C without azide. The amount of antibody to be used will be variable. Roughly the same concentration which yields the brightest staining by immunofluorescence or immunoperoxidase methods is generally appropriate.
3. While cells are incubating, remove unbound IgG from petri dish using a sterile pipet. Add 10 ml of 5% FBS/PBS to plate, swirl, and remove. Repeat wash with a second 10 ml of 5% FBS/PBS. 4. Wash the antibody-treated cells from step 2 by adding 10 ml of 5% FBS/PBS, centrifuging 10 min at 400 × g, and removing supernatant. Repeat wash, then resuspend the cells in 3 ml of 5% FBS/PBS. 5. Remove FBS/PBS from coated petri dish and pour cells onto dish. Incubate 1 hr at 4°C on a flat surface. Swirl cells once midway through the incubation to ensure even coating of dish. 6. Remove nonadherent cells by gently pipetting supernatant from the dish. Wash cells gently as in step 3 with 5 to 10 ml of 1% FBS/PBS. Repeat wash. 7. Remove adherent cells by adding 5 to 10 ml of 1% FBS/PBS to the dish and gently scraping with a pipet bulb or tissue culture scraper. Alternatively, harvest cells by aggressive pipetting. Handling, Storage, and Preparation of Human Blood Cells
8. Wash cells (as in step 4) in 1% FBS/PBS, count (APPENDIX 3A), and resuspend in complete tissue culture medium. 9. Check viability using trypan blue exclusion (APPENDIX (UNIT 9.2).
3B)
or propidium iodide
5.1.8 Current Protocols in Cytometry
ENRICHMENT OF B OR NK CELLS BY REMOVAL OF T LYMPHOCYTES THROUGH COMPLEMENT-MEDIATED LYSIS
BASIC PROTOCOL 8
This protocol presents a relatively simple procedure for using negative selection to deplete T cells from mononuclear cell preparations when B cells or NK cells are the population of interest. T cells are lysed with antibody and complement. The subtleties of this procedure revolve around each lot of reagent used. The complement must not be cytotoxic to any of the cells in the absence of specific antibody and the amount of complement and antibody to be used must be determined by titration. Materials Rabbit complement (Harlan Bioproducts for Science) in tissue culture medium Mononuclear cell preparation (see Basic Protocol 2) Anti–T cell antibodies: e.g., anti-CD2, CD3, CD5, or CD7 15-ml conical centrifuge tube 1% FBS/PBS: phosphate-buffered saline (PBS; APPENDIX 2A) with 1% (v/v) FBS (heat inactivated; APPENDIX 2A) 1. Place 107 mononuclear cells in a 15-ml conical centrifuge tube. Centrifuge 10 min at 400 × g, 10°C. 2. Remove supernatant and resuspend cells in an appropriate dilution of antibody. Incubate 30 min at 25°C. 3. Centrifuge 10 min at 400 × g, 10°C. 4. Remove the supernatant and resuspend cells in the appropriate concentration of complement in tissue culture medium. Incubate 45 to 60 min at 25°C. The concentration of complement that gives maximum antibody-specific cytotoxicity is generally 20% to 50% (depending on the source). This can be determined by titration: simultaneously perform the procedure with different dilutions of complement, and then use the concentration that gives minimal nonspecific cytotoxicity and maximum antibody-specific cytotoxicity.
5. Wash the enriched cells from step 4 with 10 ml of 1% FBS/PBS, centrifuging 10 min at 400 × g, 4°. ISOLATION OF PERIPHERAL BLOOD STEM CELLS USING AN IMMUNOSORBENT COLUMN
BASIC PROTOCOL 9
The method for isolating peripheral blood stem cells may vary depending on how one defines stem cells. In this protocol, CD34+ cells are considered to be synonymous with stem cells. If stem cells are defined by additional markers or by antigen density, this method may be used as an enrichment technique. Materials CellPro Ceperate LC Kit containing: Avidin column PBS P30 gel 1% and 5% BSA in PBS Precolumn gel Primary antibody (CD34) Secondary biotinylated antibody Sample/wash chamber Sample stand
Specimen Handling, Storage, and Preparation
5.1.9 Current Protocols in Cytometry
Cell preparation (see Basic Protocol 2) Peristaltic pump 15-ml centrifuge tubes 1. Mount sample/wash chamber assembly onto the sample stand. Remove assembly from its pouch and mount on the sample stand as directed in manufacturer’s instructions. Remove cap at the bottom of the three-way valve. 2. Remove the avidin column from its foil pouch, remove luer lock cap from top of column, and attach top of the column to bottom of the three-way valve on sample assembly. Tighten the luer nut at the connection. 3. Insert tubing from the bottom of avidin column into the peristaltic pump. Cut end of tubing and place tubing into a 15-ml centrifuge. 4. Prime the sample/wash chamber and column with PBS to remove any bubbles that may interfere with cell separation. Open sample chamber and turn three-way valve to permit flow between the sample chamber and column. Squeeze the chamber to remove air, and, while squeezing, add 5 ml PBS to sample chamber. Release chamber and turn the three-way valve to allow PBS to flow from the sample chamber to the wash chamber. Turn the three-way valve to permit PBS to flow from the sample chamber into the column. Add 10 ml PBS to wash chamber and 5 ml PBS to sample chamber. Add 0.5 ml P30 gel to sample chamber and allow it to settle onto filter near the bottom of the chamber. 5. Wash P30 gel and column with 5 ml PBS, then wash with 5 ml of 5% BSA/PBS using the peristaltic pump. Stop when the fluid reaches the top of the P30 gel in the sample chamber. 6. Wash cells in PBS with 1% BSA and suspend at a density of 108 cells/ml in 1% BSA/PBS. Add the appropriate amount of CD34 antibody to cells, mix, and incubate 25 min at 25°C. 7. Wash cells twice with 1% BSA/PBS. Resuspend cells to a density of 108 cells/ml in 1% BSA/PBS and add the secondary biotinylated antibody. Mix and incubate 25 min at 25°C. 8. Wash cells in 5% BSA/PBS and resuspend at a density of 108 cells/ml in 5% BSA/PBS. 9. Load labeled cells onto the column by layering cells onto the P30 gel in the sample chamber. Pump cells through the column at a flow rate of 0.75 ml/min. When the last of the cell suspension reaches the top of the P30 gel, add 2 ml of 5% BSA/PBS to the sample chamber and pump the remaining cells into the column. When sample chamber wash is nearing the top of the P30 gel, turn the three-way valve to allow 4 ml of PBS from the wash chamber to be pumped through the column. Close the clamp below the wash chamber. 10. Collect the adsorbed CD34+ stem cells. Cut the tubing 10 to 15 cm below the column, but before the pump. Place the end of the tubing into a new 15-ml centrifuge tube. Open the clamp below the wash chamber to permit PBS to start flowing through the column and squeeze the column ten times. Allow column to drain into the collection tube. Allow all of the PBS in the wash chamber to drain into the column. Handling, Storage, and Preparation of Human Blood Cells
The centrifuge tube contains the enriched CD34+ stem cells. These may be assayed as desired. If the cells are not used immediately, they should be transferred to another medium for optimal storage.
5.1.10 Current Protocols in Cytometry
PREPARATION OF PLATELET-ENRICHED PLASMA For many platelet assays, the platelets do not need to be purified by density-gradient separation. Platelet-enriched plasma, prepared by enrichment of platelets from peripheral blood (Ault, 1988), is often an acceptable specimen.
BASIC PROTOCOL 10
Materials Peripheral blood in EDTA or appropriate anticoagulant Tyrode’s buffer (see recipe) 15-ml conical centrifuge tube 1. Centrifuge 7 ml blood (in collection tube) 10 min at 200 × g, 25°C. 2. With a sterile pipet, transfer the plasma layer to a 15-ml conical centrifuge tube. Centrifuge 10 min at 1600 × g, 25°C. 3. Remove and discard supernatant. Resuspend pellet containing platelets in Tyrode’s buffer or a buffer containing EDTA. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Tyrode’s buffer 137 mM NaCl 12 mM NaHCO3 5.5 mM glucose 2 mM KCl 1 mM MgCl2 0.3 mM Na2HPO4 Adjust pH to 7.4 with HCl COMMENTARY Background Information Flow cytometry is a relatively unique technology in that cells need not be purified or separated for the study of a particular subpopulation or clone within a more heterogenous population of cells. Multiparameter analyses and electronic gating permit flow cytometry studies to be performed without purification of the subpopulations. It is even theoretically possible to analyze peripheral blood leukocytes without prior lysis of erythrocytes using multiparameter methods (Terstappen et al., 1991). Indeed, “live” gating techniques where data are gathered during acquisition and prior to analysis my be viewed as “electronic separation” of cell populations, permitting analysis of small populations of cells without their physical separation. Additionally, flow cytometry provides an excellent means of isolating specific cell populations with high purity, if high yield is not an overriding factor. Although flow cytometry may be used to analyze cells in heterogeneous populations, it
is quite often desirable to enrich, purify, or in some way separate populations of cells prior to flow cytometric analysis. The most common example of this is the removal of erythrocytes from peripheral blood prior to analysis of leukocytes. Though not absolutely required, the ease with which erythrocyte removal can be performed, the minimal impact this has on most assays, and the extent of the cytometric problems this avoids makes this removal desirable. Separation or enrichment of cell populations prior to flow cytometric sorting studies may also be well worth the effort. The throughput and, ultimately, the yield of a desired population of cells may be greatly enhanced by enriching for the target cells in the starting population. The techniques described in this unit can be used in many different ways. Protocols using antibodies for separation can be applied to a variety of cell types if the appropriate antibodies and cell surface antigens are available. Several of the methods, including the immunosor-
Specimen Handling, Storage, and Preparation
5.1.11 Current Protocols in Cytometry
bent columns and the magnetic bead separation techniques, can be used in either a positive- or a negative-selection mode. Again, the manner in which these techniques are used depends on the cell surface antigens and their corresponding antibodies, as well as any effect these may have on cell function.
Critical Parameters
Handling, Storage, and Preparation of Human Blood Cells
Numerous points must be considered in using preenriching or separation techniques prior to flow cytometric studies. In addition to yield and purity, possible alterations in cell function must be considered when cells are enriched for subsequent functional studies. Viability and processing time are other issues that may affect the separation or enrichment methods used, or whether any precytometric method should be used at all. Precytometric enrichment or separation techniques are not without ancillary effects, both beneficial and detrimental. For example, erythrocyte lysis techniques are quite effective in removing mature erythroid cells from blood, but may leave numerous immature, nucleated erythroid cells in the specimen if they were present in the original blood sample. These may be detected in the lymphoid gate by light scatter and may confuse the determination of how lymphoid subsets are distributed (Slade et al., 1988). On the other hand, a separation technique such as Ficoll-Hypaque not only removes granulocytes from the mononuclear cells, it has the added advantage of removing many dead lymphoid cells from the specimen as well. If cell selection methods are used prior to flow cytometric immunophenotyping, care should be taken that these methods do not affect the staining or light-scatter attributes of the cells of interest. For example, use of antibodies in these methods may interfere with subsequent immunofluorescence staining. Alternatively, granulocytes, for example, may have significantly different scatter properties after cell separation procedures. After any cell separation or preparation technique, particularly if cell surface antigens are to be examined, it is highly recommended that cellular viability be determined. Cells should also be thoroughly washed in isotonic buffer to assure the cells are as free of contaminants as possible and in a native condition. The latter may be of particular concern when staining cells with lectins. Lectins bind carbohydrate structures on the cells, and this may be hindered by residual sugars from the medium or from density gradients.
One last note concerning the handling and preparation of human specimens for flow cytometric analyses: All human-derived material should be considered potentially infectious and should be handled with appropriate caution! The precise precautions taken—e.g., whether specimens are prepared under a laminar flow hood—may vary somewhat from laboratory to laboratory. Certain universal precautions, such as wearing of latex gloves, disinfecting of countertops and equipment with bleach solutions, and the use of face shields should always be followed. No specimen from a human, even a “normal control,” should ever be considered “safe.” Fixation of cells with formalin or paraformaldehyde after staining reduces the infectious potential of human specimens considerably, yet prudent precautions are still advised even when working with fixed specimens.
Troubleshooting In attempting to troubleshoot the procedures presented in this unit, it is critical that an initial assessment be made of the specimen (cells) to be purified or enriched. The percentage of cells of interest in the initial specimen, and the absolute number of cells of interest should be determined prior to any of the procedures. Similarly, the types and numbers of unwanted cells contaminating the initial specimen should be determined. In general, this may best be accomplished by flow cytometric immunopyhenotyping together with manual or automatic cell counts. There is usually a trade-off between obtaining maximum purity and maximum yield in isolating cells of interest. This should be borne in mind when separating cells and whether purification (maximum purity) or enrichment (maximum yield) is the ultimate goal. This goal is obviously determined by the final use of the cells and the influence of the remaining, contaminating cells in preparation in the final application. Unwanted contaminants in the final cell preparation can be the result of any of a number of factors. Assuming protocols are followed precisely and the beginning specimen was typical, a common reason for failure in purification or enrichment of cells is reagent failure. This may be due to substandard lots of reagents, improper storage, or the use of inappropriate or expired reagents. If possible, all reagents should be checked (at least grossly) prior to use. Such checks might include examining the reagent bottles for obvious contamination, re-
5.1.12 Current Protocols in Cytometry
agent expiration date, and any unusual discoloration or precipitate. It is also crucial in most instances to assess the viability of separated cells. This may be easily done by any number of methods (e.g., see APPENDIX 3B & UNIT 5.2). Additionally, viability should be determined prior to separation procedures. Low initial viability may greatly alter purification, even if viable cells are not needed in the final specimen. Low viability in the final specimen may be due to the presence of cytotoxic agents, such as preservatives, in one of the reagents. A reagent to be considered is the fetal bovine serum, which may not have been properly heat inactivated. Alternatively, poor viability may be due to overly harsh treatment of the cells (e.g., centrifugation).
Anticipated Results The yield and purity from each of these methods is different. As stated above, before using any of these separation procedures is used, a decision must be made whether this is to be a purification or enrichment process. The resulting yields must be anticipated and evaluated in this light.
Time Considerations The length of time required to separate cells by these procedures ranges from several minutes to many hours. With most procedures, optimal results will be obtained if cells are processed as soon as possible after collection from the donor. This will minimize problems with viability and any possible alteration of the cell surface antigens.
Literature Cited Ault, K.S. 1988. Flow cytometric measurement of platelet-associated immunoglobulin. Pathol. Immunopathol. Res. 7:395-408. Boyum, A. 1968. Isolation of mononuclear cells and granulocytes from human blood. Scand. J. Clin. Lab. Invest. Suppl. 21:77-89. Boyum, A. 1977. Separation of lymphocytes, lymphocyte subgroups and monocytes: A review. Lymphology 10:71-76. Denholm, E.M. and Wolber, F.M. 1991. A simple method for the purification of human peripheral blood monocytes. A substitute for SepracellMN. J. Immunol. Methods 144:247-251. Indiveri, F., Huddlestone, J., Pellegrino, M., and Ferrone, S. 1980. Isolation of human T lymphocytes: Comparison between wool filtration and rosetting with neuraminidase (VCN) and 2-aminoethylisothiouronium bromide (AET)–treated sheep red blood cells. J. Immunol. Methods 34:107-115.
Madsen, M., Johnsen, H.E., and Kissmeyer-Nielsen, F. 1979. Separation of human T and B lymphocytes using ET-treated sheep red blood cells. Transplant. Proc. 11:1381-1382. Molday, R.S., Yen, S.P., and Rembaum, A. 1977. Application of magnetic microspheres in labelling and separation of cells. Nature 268:437-438. Rasmussen, A.M., Smeland, E.B., Erikstein, B.K., Caignault, L., and Funderud, S. 1992. A new method for detachment of Dynabeads from positively selected B lymphocytes. J. Immunol. Methods 146:195-202. Slade, H.B., Greenwood, J.H., Hudson, J.L., Beekman, R.H., Riedy, M.C., and Schwartz, S.A. 1988. Lymphocyte phenotyping of infants with congenital heart disease: Comparison of cell preparation techniques. Diagn. Clin. Immunol. 5:249-255. Terstappen, L.W., Johnson, D., Mickaels, R.A., Chen, J., Olds, G., Hawkins, J.T., Loken, M.R., and Levin, J. 1991. Multidimensional flow cytometric blood cells differentiation without erythrocyte lysis. Blood Cells 17:585-602. Wysocki, L.J. and Sato, V.L. 1978. Panning for lymphocytes: A method for cell separation. Proc. Natl. Acad. Sci. U.S.A. 75:2844-2848.
Key References NCCLS. 1989. Clinical Applications of Flow Cytometry: Quality Assurance and Immunophenotyping of Peripheral Blood Lymphocytes. 1989. NCCLS Doc. H42-P, Vol. 9, No.13. Provides general guidelines for performance of flow cytometric immunophenotyping. NCCLS. 1993. Clinical Applications of Flow Cytometry: Quality Assurance and Immunophenotyping of Leukemic Cells; Proposed Guidelines. 1993. NCCLS Doc. H43-P, Vol. 13, No. 23. Provides more detailed guidelines for flow cytometric immunophenotyping of leukemias and lymphomas. McCoy, J.P., Carey, J.L., and Krause, J.R. 1990. Quality control in flow cytometry for diagnostic pathology: I. Cell surface phenotyping and general laboratory procedures. Am. J. Pathol. 93(Suppl. 1):S27-S37. Discusses quality control issues in the performance of low cytometric immunophenotyping. Stewart, C.C. 1990. Cell preparation for the identification of leukocytes. Methods Cell Biol. 33:411-426. Details sample preparation methods for immunophenotyping studies.
Contributed by J. Philip McCoy, Jr. Cooper Hospital/University Medical Center Camden, New Jersey
Specimen Handling, Storage, and Preparation
5.1.13 Current Protocols in Cytometry
Handling, Storage, and Preparation of Human Tissues
UNIT 5.2
In contrast to blood cells, human tissue for use in flow cytometry must first be prepared as an adequate single-cell suspension. Selection of the appropriate methods for specimen collection, transport, storage, and tissue dissociation depend on the cell parameters being measured (e.g., protein, DNA, or RNA) and the localization of the markers (e.g., cell surface, cytoplasm, or nucleus). This unit includes a general method for collecting and transporting tissue samples to the flow cytometry laboratory (see Basic Protocol 1) and preparing a tissue imprint to assess the sample (see Support Protocol 1). Single-cell suspensions of whole cells can be prepared from fresh tissue using mechanical disaggregation (see Basic Protocol 2) or enzymatic disaggregation (see Alternate Protocol 1). Single-cell suspensions of whole cells can also be prepared from fine-needle aspitates (see Basic Protocol 3) or pleural effusions, abdominal fluids, or other fluids (see Alternate Protocol 2). Once a single-cell suspension is obtained, the cells can be fixed (see Support Protocol 2), used for cytospin preparations (see Support Protocol 3), or frozen for future use (see Support Protocol 4). Intact nuclei can be prepared for flow cytometry from fresh or frozen tissue (see Basic Protocol 4) or from paraffin-embedded tissue (see Alternate Protocol 3). Debris can be removed from nuclear preparations, from frozen or paraffinembedded samples, by a sucrose step gradient (see Support Protocol 5). CAUTION: Handle all standards, controls, and patient samples as if they are capable of transmitting hepatitis and AIDS. Refer to Universal Precaution Procedures for correct handling procedures. COLLECTION AND TRANSPORT OF SAMPLES TO THE FLOW CYTOMETRY LABORATORY
BASIC PROTOCOL 1
Most often human tissues are received from a hospital pathology laboratory or surgical suite. Before using tissue for any assay, the investigator should ensure that appropriate tissue has already been obtained for diagnosis by the pathologist and that, if needed, appropriate consent forms have been signed. The collaborating pathologist or tissue procurement person should be aware of the needs of the study so that representative tissue is appropriately collected and transported, especially if the use for the tissue requires special handling (see Table 5.2.1). For example, if RNA must remain intact, the tissue should be frozen immediately, but freezing is not a requirement for DNA ploidy and S-phase determinations. For most protein and DNA flow cytometry assays, tissue can be collected at room temperature and placed in a clean (sterile, if possible) vessel containing growth medium. The sample is placed on ice and transported to the laboratory. For Table 5.2.1
Sample Collection and Transport Conditions
To study
Collect tissue
Transport tissue
DNA (e.g., ploidy, % S) Nuclear proteins
Fresh Fresha
At 4°C At 4°Ca
Cell-surface proteins RNA
Fresha Freeze immediately
At 4°Ca In liquid nitrogen or isopentene
aDepending on the nature of the protein—e.g., enzymes—the conditions for transport may vary and may require
freezing; the investigator will need to identify optimal conditions depending on the specific protein being assayed.
Contributed by Lynn G. Dressler and Dan Visscher Current Protocols in Cytometry (1997) 5.2.1-5.2.15 Copyright © 1997 by John Wiley & Sons, Inc.
Specimen Handling, Storage, and Preparation
5.2.1 Supplement 1
efficiency, refrigerated prefilled containers of growth medium can be kept at the sample collection site. Cells in suspension as aspirates or fluid samples can simply be obtained in a sterile container, bag, or syringe and transported on ice to the laboratory. Once the tissue is received, it is essential to include a quality assurance check to ensure that appropriate tissue (normal or malignant) is obtained; this can be accomplished in a variety of methods including analyzing tissue imprints (see Support Protocol 1), frozen sections, or cytospin preparations (see Support Protocol 3). Materials Growth medium (e.g., MEM or RPMI 1640) containing 5% (v/v) heat-inactivated FBS (APPENDIX 2A) or Hanks’ balanced salt solution (HBSS; APPENDIX 2A) Liquid nitrogen or isopentane bath Sample containers (e.g., 50-ml conical centrifuge tubes or cryovials), sterile Curved and straight forceps, sterile Cryovials Additional reagents and equipment for preparing a tissue imprint (see Support Protocol 1) To collect samples for DNA or protein assays 1a. Add 25 ml growth medium containing 5% FBS to a 50-ml conical centrifuge tube. Keep tube on ice. Alternatively, use tubes that have been filled with growth medium and refrigerated. Sterility is preferable but not essential.
2a. Have the pathologist or appropriate person procure a representative piece of tissue and trim away fat and necrotic areas. If possible, prepare a tissue imprint (see Support Protocol 1). 3a. Using sterile forceps, place sample in conical tube on ice. Label tube with appropriate identification information, date, and time of collection. Transport tube to lab on ice. 4a. Keep sample at 4°C and dissociate as soon as possible, preferably within 2 hr. Depending on the assay, samples may be stored overnight in the refrigerator, if necessary, before dissociation (see analysis protocols for details).
To collect samples for RNA assays or those which are susceptible to protein or nucleic acid degradation 1b. Have the pathologist or appropriate person procure a representative piece of tissue, and trim away fat and necrotic areas. Prepare a tissue imprint (see Support Protocol 1). Sample may need to be cut into multiple pieces depending on size of tissue and investigator need.
2b. Place tissue in a dry cryovial, marked with appropriate identification information and date of collection. Using long forceps, place vial in liquid nitrogen or isopentane bath for 30 to 60 seconds. 3b. Transfer vial to bucket containing dry ice or keep in liquid nitrogen or isopentane bath and transport to lab. Keep sample frozen during transport. Handling, Storage, and Preparation of Human Tissues
4b. Store sample at −70°C or lower.
5.2.2 Supplement 1
Current Protocols in Cytometry
PREPARING A TISSUE IMPRINT Tissue imprints are used to verify that the correct tissue has been obtained and to provide a preliminary evaluation of cellularity. The tissue sample is touched to a slide, and the slide is fixed and stained.
SUPPORT PROTOCOL 1
Materials Tissue specimen, freshly collected 95% ethanol Forceps Glass slides Coplin jar Papanicolaou staining set or hematoxylin-eosin staining series 1. Gently hold the tissue specimen with forceps. Touch the freshly cut surface of the tissue with a glass slide. 2. Fix the slide immediately by immersing it in a Coplin jar containing 95% ethanol ≥30 sec. These preparations will air dry almost immediately because of their relatively low cellularity, so it is essential that slides be fixed immediately to prevent drying artifacts.
3. Stain slide with Papanicolaou or hematoxlyin-eosin, add a coverslip, and examine for presence of appropriate cells and cellularity (e.g., for cancer specimens, ensure that ≥15% of cells are malignant). MECHANICAL DISAGGREGATION OF WHOLE CELLS FROM FRESH HUMAN TISSUE
BASIC PROTOCOL 2
Flow cytometric analysis of fresh solid tumors requires the recovery of sufficient numbers of intact, viable tumor cells, as a single-cell suspension, either by mechanical or enzymatic disaggregation (see Alternate Protocol 1). Mechanical disaggregation involves the dispersion of cells by scraping and mincing the tumor with a scalpel. The resulting cell suspension is filtered, counted, and fixed (see Support Protocol 2) prior to flow cytometry (Hitchcock, et al., 1996). Materials Tumor tissue, freshly obtained (see Basic Protocol 1) Growth medium (e.g., RPMI 1640 medium, Life Technologies) FBS, heat-inactivated (Life Technologies or APPENDIX 2A) 0.4% (w/v) trypan blue (Sigma) Gauze, sterile No. 22 scalpel blades and handle 100-mm petri dishes 80-µm metal-mesh sieve (EC Apparatus) 15- and 50-ml conical plastic centrifuge tubes IEC CL centrifuge and swinging bucket rotor (or equivalent) Additional reagents and equipment for counting cells (APPENDIX 3A) and assessing viability (APPENDIX 3B), and fixing cell suspension (see Support Protocol 2) NOTE: This protocol should be carried out in biological safety cabinet.
Specimen Handling, Storage, and Preparation
5.2.3 Current Protocols in Cytometry
1. Remove fresh tumor tissue from medium and blot on sterile gauze. Although it is optimal to dissociate cells as soon as possible, for most DNA flow cytometry protocols, samples may be stored at 4°C up to 24 hr without deleterious effects.
2. Using a no. 22 scalpel blade, trim fat and areas of necrosis. Weigh remaining tissue. 3. Place tissue in a petri dish containing 5 to 10 ml growth medium. Using forceps and a no. 22 scalpel blade, gently scrape the tissue until the solution becomes turbid. Then bisect the specimen to provide more surface area and repeat. Continue to scrape and mince the tissue until no more cells are released. 4. Pour the solution through an 80-µm metal-mesh sieve into a 15-ml conical centrifuge tube. Centrifuge 10 min at 250 × g (4000 rpm in swinging bucket rotor), room temperature. 5. Remove supernatant and resuspend pellet in 1 ml growth medium plus 1 ml heat-inactivated FBS. 6. Combine 10 µl of cell suspension with 10 µl of 0.4% trypan blue. Using a hemacytometer, count the cells (APPENDIX 3A) and assess viability (APPENDIX 3B). CAUTION: Trypan blue is a potential carcinogen and should be handled carefully. Dead cells take up trypan blue.
7. Fix cell suspension (see Support Protocol 2). If the antigen to be assayed is destroyed by fixation, the cells do not have to be fixed, except to prepare cytospin slides. ALTERNATE PROTOCOL 1
ENZYMATIC DISAGGREGATION OF WHOLE CELLS FROM FRESH HUMAN TISSUE Enzymatic disaggregation of solid tumors uses trypsin to disrupt intercellular adhesions and collagenase Type II to digest stromal components. DNase I is also added to ensure that ploidy is determined only from populations of intact cells. Enzymatic disaggregation can be used alone or in combination with mechanical methods and it is recommended for complete dissociation of tumors of squamous origin. Additional Materials (also see Basic Protocol 2) Trypsin Collagenase type II DNAse I Additional reagents and equipment for counting cells (APPENDIX 3A) and assessing viability (APPENDIX 3B), and fixing cell suspension (see Support Protocol 2) 1. In a 50-ml centrifuge tube, prepare a 10-ml enzyme cocktail containing 2.5 mg/ml trypsin, 0.5 mg/ml collagenase Type II, and 20 µg/ml DNase I in growth medium. Warm 15 min at 37°C. 2. Place fresh tumor tissue or mechanically disaggregated tissue (Basic Protocol 2, step 5) into the enzyme cocktail. Incubate 1 hr at 37°C with gentle agitation.
Handling, Storage, and Preparation of Human Tissues
3. Pour the solution through an 80-µm metal-mesh sieve into a 15-ml centrifuge tube. Add 2 ml heat-inactivated FBS to stop the reaction. Centrifuge 10 min at 250 × g (4000 rpm in swinging bucket rotor), room temperature. 4. Discard the supernatant and resuspend the pellet in growth medium.
5.2.4 Current Protocols in Cytometry
5. Combine 10 µl of cell suspension with 10 µl of 0.4% trypan blue. Using a hemacytometer, count the cells (APPENDIX 3A) and assess viability (APPENDIX 3B). CAUTION: Trypan blue is a potential carcinogen and should be handled carefully. Dead cells take up trypan blue.
6. Fix cell suspension (see Support Protocol 2). If the antigen to be assayed is destroyed by fixation, the cells do not have to be fixed, except to prepare cytospin slides.
PREPARATION OF CELLS FROM FINE-NEEDLE ASPIRATES Frequently human cells are obtained as a single-cell suspension in fine-needle aspirates, bone marrow aspirates, or pleural, ascitic, or other fluid samples (see Alternate Protocol 2). No dissociation is required for these samples. If the sample contains red blood cells, hemolyzing reagent should be added to lyse them.
BASIC PROTOCOL 3
Materials Needle or bone marrow aspirate in a syringe or sterile container Growth medium (e.g., MEM, HBSS, or RPMI 1640) containing 5% (v/v) heat-inactivated FBS (Life Technologies or APPENDIX 2A) Hemolyzing reagent: 3.6% (w/v) NaCl in H2O, cold Heparin 15-ml or 50-ml conical centrifuge tubes Beckman RJ-6 centrifuge and TH-4 rotor (or equivalent) Forceps, sterile Additional reagents and equipment for counting cells (APPENDIX 3A) 1. Collect needle or bone marrow aspirate in sterile syringe or container. Transport to lab on ice and store at 4°C. Cells should be processed (washed and centrifuged) within 2 hr for DNA flow cytometry.
2. Add contents of syringe and needle or container to a 15- or 50-ml conical tube containing 5 to 10 ml growth medium with 5% FBS. If possible, add needle as well to obtain additional cells. Keep tube on ice. 3. Remove needle from tube with sterile forceps. Cap tube and invert several times to mix contents. 4. Optional: If the suspension is bloody, centrifuge 5 min at 125 × g (800 rpm in TH-4 rotor), 4°C. Resuspend pellet in 12 ml cold water and add 4 ml of cold hemolyzing reagent. Invert tube several times to mix contents and incubate a total of ∼20 sec. Centrifuge 5 min at 125 × g, 4°C. If the pellet is still bloody, repeat. A 20-sec incubation should be sufficient to lyse red blood cells.
5. Centrifuge 5 min at 125 × g (800 rpm in TH-4 rotor), 4°C, to pellet cells. 6. Resuspend cells in 1 to 2 ml fresh growth medium. Count a sample of the suspension in a hemacytometer (see APPENDIX 3A). Adjust cell concentration to 1 to 3 × 106 cells/ml. Sample can be either used for flow cytometry or frozen and stored for later use (see Support Protocol).
Specimen Handling, Storage, and Preparation
5.2.5 Current Protocols in Cytometry
ALTERNATE PROTOCOL 2
PREPARATION OF CELLS FROM PLEURAL EFFUSIONS, ABDOMINAL FLUIDS, AND OTHER FLUIDS Large volumes of fluid (pleural, abdominal, or other fluid) need to be centrifuged and the cell pellets washed and pooled for use. It is important to add hemolyzing agent if the sample contains red blood cells. Additional Materials (also see Basic Protocol 3) Pleural effusion, abdominal fluid, or other fluid 25-ml pipet and bulb, sterile 15- or 50-ml centrifuge tubes or 250-ml centrifuge bottles, sterile Additional reagents and equipment for counting cells (APPENDIX 3B) 1. Using a 25-ml pipet and bulb or a Pasteur pipet, transfer pleural effusion, abdominal fluid, or other fluid sample to 15-ml or 50-ml sterile centrifuge tubes or 250-ml sterile centrifuge containers (depending on total volume). Centrifuge 5 min at 150 × g (1200 rpm in TH-4 rotor), 4°C. 2. Discard supernatant and pool sample pellets. 3. Optional: If the suspension is bloody, centrifuge 5 min at 125 × g (800 rpm in TH-4 rotor), 4°C. Resuspend pellet in 12 ml cold water and add 4 ml of cold hemolyzing reagent. Invert tube several times to mix contents and incubate a total of ∼20 sec. Centrifuge 5 min at 125 × g, 4°C. If the pellet is still bloody, repeat. A 20-sec incubation should be sufficient to lyse red blood cells.
4. Discard supernatant and resuspend pellets in an appropriate volume (for cell counts) of growth medium containing 5% FBS. Count cells in hemacytometer (APPENDIX 3A). Adjust cell concentration to 1 to 2 × 106 cells/ml for immediate assay or 3 to 5 × 106 cells/ml for freezing (see Support Protocol 4) and future use. SUPPORT PROTOCOL 2
ETHANOL FIXATION OF CELL SUSPENSIONS Fixation of disaggregated solid tumor single-cell suspensions is accomplished by adding 70% ethanol drop-wise while vortexing. Fixed specimens can be stored at 4°C for up to 5 days without compromising nuclear integrity. This method is sufficient for most flow cytometry DNA ploidy analysis and can be used in conjunction with simultaneous immunofluorescence for a variety of intracellular antigens including intermediate filaments and/or leukocyte common antigen. Materials Whole cell suspension (see Basic Protocol 2 or 3 or Alternate Protocol 1 or 2) 70% ethanol, 4°C 15-ml centrifuge tube 1. Transfer 2 ml resuspended single-cell suspension into a 15-ml centrifuge tube. While vortexing, add 6 ml cold 70% ethanol drop-wise. This cell/70% ethanol ratio results in a final ethanol concentration of 50%. If the sample is >2 ml, use multiple tubes or larger tubes with the same cell/ethanol ratio.
2. Cap centrifuge tube and store at 4°C for a minimum of 30 min to a maximum of 5 days. Handling, Storage, and Preparation of Human Tissues
3. Use fixed cells for flow cytometry or to prepare cytospin slides (see Support Protocol 3).
5.2.6 Current Protocols in Cytometry
PREPARATION OF CYTOSPIN SLIDES FROM CELL SUSPENSIONS Cytospin slides prepared from dissociated solid tumor cell suspensions can be used as an important quality control for flow cytometry (Dressler and Bartow, 1989). In addition to reflecting tumor cell versus non-malignant cell representation, cytospin slides also provide an assessment of specimen quality in terms of relative amounts of debris and cellular damage. Prepared cytospin slides are stained using either Papanicolaou or Leishman stains.
SUPPORT PROTOCOL 3
Materials PBS (see APPENDIX 2) Ethanol-fixed single-cell suspension (see Support Protocol 2) 0.1% (w/v) bovine serum albumin (BSA, Baxter) in PBS Leishman or Papanicolaou stain (Sigma) 12 × 75-mm tube Glass slides Cytocentrifuge (Shandon), slide chambers, filter cards Dilute cells 1. Pipet 500 µl PBS into a 12 × 75–mm tube. 2. Add 50 µl ethanol-fixed single-cell suspension at an adjusted concentration of 1 × 106 cells/ml. Mix gently. Assemble cytospin apparatus 3. Label glass slide with specimen number, date, and other identification information. 4. Assemble the slide, filter card, and cytospin chamber, carefully aligning the hole in the filter card with the opening of the chamber. 5. Insert the slide/chamber unit into a slot in the centrifuge head. Balance the cytocentrifuge. Prepare slides 6. Add 300 µl diluted cell suspension to the chamber. 7. Add 50 µl 0.1% mg/ml BSA to maintain cellular integrity during cytocentrifugation. 8. Close and lock the cover of the cytocentrifuge. Centrifuge slide/chamber units 5 min at 1000 rpm. 9. When the cytocentrifuge cover unlocks, carefully remove the slide/chamber unit. 10. Discard the filter card and allow the preparation to air dry. Wash the chamber before reusing. 11. Stain slides with Leishman or Papanicolaou stain according to manufacturer’s instructions.
Specimen Handling, Storage, and Preparation
5.2.7 Current Protocols in Cytometry
Supplement 6
SUPPORT PROTOCOL 4
FREEZING DISSOCIATED SINGLE CELLS Single whole cell suspensions prepared from tumor tissue can be frozen and stored for future use. Pelleted whole cells are resuspended in cryopreservation solution and frozen at −70°C. Materials Whole cell suspension (see Basic Protocol 2 or 3 or Alternate Protocol 1 or 2), unfixed Growth medium (e.g., RPMI, MEM, Medium 199) containing 5% (v/v) heat-inactivated FBS (APPENDIX 2A) Cryopreservation solution (see recipe) 1.25-ml cryovials 12 × 75–mm polystyrene tubes Beckman RJ-6 centrifuge and TH-4 rotor 1. Transfer 1.0 ml whole-cell suspension at a concentration of 3 to 5 × 106 cells/ml in growth medium containing 5% FBS into 12 × 75–mm polypropylene tube. Centrifuge 5 min at 150 × g (1200 rpm in TH-4 rotor), 4°C. 2. Decant supernatant and quickly resuspend cells in 1.0 ml cryopreservation solution. 3. Immediately transfer contents of tube into a labeled 1.25-ml cryovial and place in −70°C freezer. Samples can be stored for 4 to 6 months. The resuspension and transfer should be done carefully but quickly to minimize contact of cells with DMSO in warm cryopreservation solution.
BASIC PROTOCOL 4
ISOLATION OF NUCLEI FROM FRESH OR FROZEN TISSUE Nuclei can be prepared from fresh or frozen tissue by mechanical or enzymatic procedures. Some investigators use a combination of both. The goal is to obtain nuclei that are relatively free of any contaminating cytoplasmic debris and yet retain an intact nuclear membrane. The simple technique can be applied to fresh or frozen tissue. Nuclei are isolated from single-cell suspensions of whole cells prepared from tumor tissue. Alternatively, nuclei can be prepared from paraffin-embedded tissues (see Alternate Protocol 3). For assays that require intact whole cells see Basic Protocol 2. Materials Tumor tissue, fresh or frozen Growth medium (e.g., MEM or RPMI 1640) containing 5% (v/v) heat-inactivated FBS (APPENDIX 2A), 4°C Sample container, sterile and dry Petri dish, disposable Curved and straight forceps, sterile Scalpel and no. 20 blades 50- to 70-µm nylon mesh (Tetko) 15-ml conical centrifuge tube Beckman RJ-6 centrifuge and TH-4 rotor (or equivalent) Additional reagents and equipment for counting cells (APPENDIX 3A)
Handling, Storage, and Preparation of Human Tissues
1a. For fresh tumor tissue: Obtain tumor tissue in a dry sterile container and put on ice or at 4°C. 1b. For frozen tumor tissue: Remove vial or container from −70°C freezer and place on ice. Incubate on ice ∼8 to 10 min until the tissue reaches ∼4°C.
5.2.8 Supplement 6
Current Protocols in Cytometry
2. Using sterile forceps, transfer the tissue to a disposable petri dish on ice. Add 5 to 10 ml cold growth medium containing 5% FBS. Keep the tissue on ice or at 4°C. 3. Cut the tissue into small pieces using a scalpel with a no. 20 blade. 4. Hold the tissue in the medium using a pair of forceps. Using moderate pressure, scrape the cut surface of the tissue with a glass slide held at a 45° angle until only connective tissue remains (1 to 3 min). The medium will turn cloudy as cells are released from the tissue.
5. Using a Pasteur pipet, collect the cells and medium and filter through a 50- to 70-µm nylon mesh into a 15-ml conical centrifuge tube. 6. Centrifuge 5 min at 150 × g (1200 rpm in TH-4 rotor), 4°C, to gently pellet the cells. 7. Resuspend the pellet in 2 ml growth medium containing 5% FBS. Centrifuge 5 min at 150 × g, 4°C. If the sample is small or there is minimal cell yield, this step can be eliminated to reduce cell loss.
8. Resuspend the pellet in 1 ml growth medium containing 5% FBS. Count the cells (APPENDIX 3A) and adjust the concentration to 1 to 2 × 106 cells/ml for flow cytometry or 3 to 5 × 106 cells/ml for freezing and storage. These cells will lyse and release nuclei when they are added to the staining solution for flow cytometry.
ISOLATION OF NUCLEI FROM PARAFFIN-EMBEDDED TISSUE Nuclei can also be isolated from tumor tissue in formalin-fixed paraffin blocks for flow cytometric determination of DNA. Sections cut from the block are first dewaxed with xylene and rehydrated with ethanol and water. The sample is then digested with pepsin to disaggregate the tissue and isolate the nuclei. Isolated nuclei can be stained for flow cytometric analysis of DNA content (Dressler and Bartow, 1989).
ALTERNATE PROTOCOL 3
Materials Formalin-fixed paraffin-embedded tissue Hematoxylin-eosin or Diff-Qwik stain Xylene or Histoclear Ethanol series: 100%, 95%, 70% and 50% Pepsin solution: 0.5% (w/v) bovine pancreatic pepsin in 0.9% (w/v) NaCl Growth medium (e.g., MEM, RPMI, HBSS) containing 5% (v/v) heat-inactivated FBS (APPENDIX 2A) Propidium iodide buffer (see recipe) Microtome (adjustable, to cut 50-µm and 4-µm sections) Glass slides 13 × 100–mm glass screw top tubes 50-µm nylon mesh (Tetko4) Additional reagents and equipment for counting cells (APPENDIX 3A) Cut sections 1. Deface block of formalin-fixed paraffin-embedded tissue. Mount block on a microtome and cut one 4-µm section. Prepare a slide with this section, label it “top,” and include appropriate identifying information. Set aside for routine staining (step 4). The block should be examined by a pathologist who should mark the block for areas of malignant and nonmalignant tissue. If distinct areas are marked, first make a razor cut
Specimen Handling, Storage, and Preparation
5.2.9 Current Protocols in Cytometry
Supplement 6
along the demarcation line so malignant and nonmalignant sections for each area can be processed separately.
2. Cut three 50-µm sections from the same block. Let each section curl up. Using forceps, carefully transfer each thick section to a glass 13 × 100–mm screw top tube. Label tube appropriately with solvent pen. 3. Cut one 4-µm section from the same block. Prepare a slide, label it “bottom,” and include other identifying information. Set aside for routine staining (step 4). 4. Dewax, rehydrate, and stain the “top” and “bottom” slides using hematoxylin-eosin or Diff-Qwik stain according to manufacturer’s instructions. 5. Evaluate top and bottom sections to ensure that representative tissue is present on both sections before processing and analyzing material from thick sections. Keep stained sections on file for future reference. Dewax and rehydrate thick sections 6. Dewax thick sections using two changes of xylene (or Histoclear): add 10 ml xylene to each tube and let sit 10 min at room temperature (without vortexing). Carefully remove xylene with Pasteur pipet, replace with 10 ml fresh xylene, and incubate 10 min. CAUTION: Perform steps 6 and 7 in a fume hood. CAUTION: Collect used xylene in a discard beaker and transfer to a xylene waste container. Never pour xylene down the sink!
7. Remove xylene and rehydrate by incubating 10 min in 10 ml of each of the following (do not vortex): 100% ethanol, two 10-min changes 95% ethanol 70% ethanol 50% ethanol With each successive ethanol step, the tissue becomes more fragile and will break up into pieces. Be careful to remove only the alcohol solution without losing the tissue pieces.
8. After the final ethanol step, remove ethanol and add 10 ml distilled water to each tube, cap tube, and let stand overnight at room temperature. This step should allow any remaining alcohol trapped in the tissue to be released. Sections can stay in distilled water up to 10 days without adverse effects, but must stay ≥12 to 24 hr for optimal results.
Disaggregate tissue and isolate nuclei 9. After ≥12 to 24 hr, remove distilled water from each tube and add 10 ml fresh distilled water. Let stand 10 min at room temperature. During this 10-min incubation, take pepsin out of freezer and warm to room temperature. Add 10 ml fresh distilled water to each tissue tube. 10. Prepare 1.5 ml pepsin solution per tube. Warm solution 5 min at 37°C before adding to tubes. 11. Remove distilled water from tube and place tube in a rack in a 37°C water bath. 12. Add 1.5 ml pepsin solution to each tube and vortex 10 seconds. Handling, Storage, and Preparation of Human Tissues
13. Incubate tube 30 min in 37°C water bath, vortexing each tube at 5-min intervals. If cells stick to side of tube, wash them down with a Pasteur pipet.
5.2.10 Supplement 6
Current Protocols in Cytometry
The solution in the tube should turn cloudy as cells are released from tissue.
14. After the incubation, filter digested suspension through 50-µm nylon mesh. Wear gloves. If the filtered suspension contains considerable debris, remove the debris on a sucrose step gradient (see Support Protocol 5).
15. Centrifuge 5 min at 150 × g (1200 rpm in TH-4 rotor), 4°C, to obtain a cell pellet. Resuspend in 1.0 to 2.0 ml growth medium containing 5% FBS. Count cells and nuclei in a hemacytometer (APPENDIX 3A) and adjust to a concentration of 1 to 2 × 106 cells and nuclei/ml in propidium iodide buffer. Long-term storage of unstained nuclei is not recommended. Once isolated, nuclei should remain in growth medium at 4°C and, for optimal results, should be processed within 2 hr. Isolated, unstained nuclei can be held overnight at 4°C and stained within 24 hr; however, the risk of losing select populations (e.g., aneuploid nuclei) increases the longer nuclei are held. Time course experiments have shown that propidium iodide–stained nuclei can be held for longer periods of time, up to 24 to 72 hr; however, the risk of selective nuclear loss remains, usually of deteriorating aneuploid nuclei, which appear to have more fragile nuclear membranes. Experimental studies have shown a decrease in the aneuploid population as a function of storage time; this is somewhat variable from sample to sample. Therefore, for optimal results, use nuclei for flow cytometry within 2 hr of preparation.
REMOVING DEBRIS FROM SUSPENSIONS OF NUCLEI Sometimes filtered nuclear suspensions prepared from fresh, frozen, or paraffin-embedded tissue may contain considerable debris or increased amounts of fibrous connective tissue. In these cases, subcellular debris can be removed by centrifuging the suspension on a sucrose step gradient.
SUPPORT PROTOCOL 5
Materials 1.75 M sucrose in PBS (APPENDIX 2A) 1.5 M sucrose in PBS (APPENDIX 2A) Nuclear suspension from fresh, frozen, or paraffin-embedded tissue, filtered (see Basic Protocol 4 or Alternate Protocol 3) Growth medium containing 5% (v/v) heat-inactivated FBS (APPENDIX 2A) 12 × 75–mm centrifuge tube Beckman RJ-6 centrifuge and TH-4 rotor (or equivalent) 1. Add 800 µl of 1.75 M sucrose to a 12 × 75–mm centrifuge tube. 2. Carefully layer 800 µl of 1.5 M sucrose on top of the 1.75 M sucrose. 3. Carefully add 1.5 ml filtered nuclear suspension on top of the sucrose cushions. 4. Centrifuge 40 min at 250 × g (4000 rpm in TH-4 rotor), 4°C, to remove subcellular debris. 5. Discard supernatant. Resuspend the nuclear pellet in 1 ml growth medium containing 5% FBS.
Specimen Handling, Storage, and Preparation
5.2.11 Current Protocols in Cytometry
Supplement 1
REAGENTS AND SOLUTIONS Use sterile deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2; for suppliers, see SUPPLIERS APPENDIX.
Cryopreservation solution 1.765 g NaCl (0.3 M final) 15 ml dimethyl sulfoxide (DMSO; 15% final) H2O to 100 ml Store at 4°C Propidium iodide buffer 250 ml H2O 1.211 g Tris base (0.04 M final) 0.2541 g MgCl2⋅6H2O (5 mM final) 75 µl 0.3% (v/v) Nonidet P-40 (NP-40) 0.0125 g propidium iodide (0.05 mg/ml final) Adjust pH to near 7.0 with concentrated HCl Adjust pH to exactly 7.0 with 1 M HCl Store up to 3 weeks at 4°C in a foil-wrapped container COMMENTARY Background Information
Handling, Storage, and Preparation of Human Tissues
A fundamental requirement of flow cytometry is a sample prepared as a single-cell or single-nucleus suspension. It is imperative that the user understand the nature of the parameter being measured before choosing a procedure for sample collection, transport, and preparation (see Table 5.2.1). For example, paraffin block material can be used to measure nuclear parameters (e.g., DNA ploidy, S phase, and nuclear proteins); this fixed material cannot be used for measuring membrane proteins because the fixation and dissociation process destroys cell membrane integrity. Analysis of membrane proteins requires fresh tissue or dissociated cells from optimally fixed fresh tissue. Therefore, the user needs to understand the specific flow cytometric application and identify optimal conditions for assay, including transport and storage. Dissociation of intact cells can be accomplished by both mechanical (see Basic Protocol 2) and enzymatic (see Alternate Protocol 1) disaggregation techniques. The success of any disaggregation procedure can be evaluated by several parameters including cell yield, cell viability, maintenance and integrity of aneuploid populations, and histogram quality (coefficient of variation and baseline debris; Visscher and Crissman, 1984). The ability to successfully address these issues varies with disaggregation method as well as with tumor type. Mechanical disaggregation involves scraping and mincing to disperse cells from solid
tumors. Although cell yield and viability can vary, mechanical disaggregation has been demonstrated to be superior to enzymatic digestion for aneuploid cell recovery in breast (McDivitt et al., 1994), colon (Crissman et al., 1988), prostate (König et al., 1993), and ovarian tumors (Costa et al., 1987). Neoplasms of squamous origin or differentiation, however, require enzymatic digestion for their complete dissociation. Cell yield and viability and aneuploid cell recovery are all increased when compared to mechanical methods for these tumors (Bijman et al., 1985; Ensley et al., 1988). There is no universal disaggregation method, so any technique should be optimized to obtain maximum recovery of target cell populations comprising a solid tumor and individualized for tumor type. The most frequent use of nuclei prepared from formalin-fixed paraffin-embedded tissue is for measurement of DNA ploidy and S phase by DNA flow cytometry. David Hedley (Hedley et al., 1983; Hedley, 1994) revolutionized DNA flow cytometry with a procedure that can be applied to tissue fixed in paraffin blocks. Since then, many variations of this technique have been developed to improve the quality of the DNA histogram obtained. Basic to all procedures is the isolation of nuclei from fixed tissue by proteolytic dissolution of the disulfide linkages formed during formalin fixation. Regardless of the technique used, the user must first select an adequately fixed, representative block of tissue that contains minimal necrosis and, if possible, minimal fibrosis. This first step
5.2.12 Supplement 1
Current Protocols in Cytometry
of quality assurance is essential to any DNA flow cytometric assay (Dressler, 1993). Optimally, for flow cytometry, material should be fixed in buffered formalin for ∼24 hr prior to paraffin processing and embedding. However, the investigator usually has little control over the initial tissue processing because this is usually done by the hospital or clinical laboratory following a routine histologic procedure. Finished blocks should never be exposed to extremes of temperature; they should be transported neither frozen, nor under conditions that cause paraffin to melt (>80°F). To ensure that tissue processed for flow cytometry is representative, thin 4- to 5-µm sections immediately preceding and after the thick sections cut for assay should be obtained and stained for histopathologic evaluation. The number of thick sections required for assay depends on the cellularity of the lesion. For most solid tumors, three 50-µm sections should provide a sufficient number of nuclei for flow cytometry. For lymph nodes (either as controls or from lymphoma cases), one to two 50-µm sections are usually sufficient. Although some investigators have used 30-µm sections, most have found that a thicker cut, at 50-µm, reduces debris caused by cutting through nuclei and yields high-quality histograms. It is imperative that a pathologist review the corresponding hematoxylin-eosin stained slides, not only to ensure representative tissue on the section, but also to determine if selective areas of the block should be prepared for flow cytometry. For example, if tumor tissue is localized in one portion of the block and nonmalignant tissue in another, each portion should be processed separately. The pathologist can mark the stained slide (and the block, if possible) with a line or semicircle demarcating the malignant versus nonmalignant areas for cutting by the histotechnician. This is an excellent way to obtain malignant and nonmalignant areas and to provide a nonmalignant diploid control (Dressler and Bartow, 1989; Dressler and Seamer, 1994).
Critical Parameters and Troubleshooting Three main critical parameters need to be addressed: tissue representation, disaggregation, and cell yield (Dressler and Seamer, 1994; Shankey, et al., 1993). To a great extent, representation of appropriate cells (neoplastic or nonneoplastic) in a dissociation protocol reflects adequacy of sampling, although a variety of biologic factors may also interfere with tumor representation. These include admixture of
neoplastic and benign tissue in some lesions as well as admixture of preinvasive and invasive tissue in other lesions. In general, sampling is optimized if tissue is obtained from the areas of the “invasion front” (i.e., the advancing edge) of the tumor. For this reason, consistent sampling of fresh, frozen, or paraffin-embedded tissue requires an experienced technologist or pathologist. For some tumor systems, such as prostate carcinoma, it may be difficult to distinguish benign from malignant tissue. Further, for fresh or frozen tissue, the decision to sample tissue requires diagnostic evaluation in order to assess proximity to margins and neoplasm size. Experienced diagnostic evaluation is essential in any experimental protocol that employs unfixed human tissue samples. Fine needle aspiration will often selectively withdraw aneuploid cells, as the intercelluar bridges of aneuploid cells are often less intact than those of diploid cells. Disaggregation protocols sometimes result in excessive debris, usually observed as baseline elevation of the DNA histogram or as increased coefficient of variation. In general, fresh samples have less debris than fixed specimens, but this is not always the case. Fine needle aspirates may also contain excess debris. Debris may reflect inadequate filtering of the cellular suspension, presence of abundant mucin within the neoplasm, tissue autolysis prior to dissociation, or sampling from necrotic or hemorraghic areas with low cellularity. In paraffin-embedded material, debris may also result from slicing through nuclei and cells, especially in thin sections. Debris can be minimized by using a double-layer sucrose cushion (see Support Protocol 5); this technique can be applied to whole cells or nuclei, fresh or frozen, in addition to paraffin-embedded tissue. Refrigeration of tissue prior to dissociation is suggested to avoid tissue autolysis. Use of an isotonic growth medium also optimizes cellular preservation. Addition of 5% heat-inactivated fetal bovine serum (FBS) to the medium appears to enhance this preservation. For fresh or frozen samples, tissue scraping to release cells may be insufficiently vigorous to yield cells or so vigorous that cells are damaged. It is useful to examine the scalpel blade after scraping the tissue to look for the presence of adherent, slightly turbid fluid on the blade surface. Minimal observable fluid means the dissociation was probably not sufficiently vigorous, and a cytospin preparation will reveal large clusters or aggregates of cells. If there are fragments of tissue, then dissociation has likely been too
Specimen Handling, Storage, and Preparation
5.2.13 Current Protocols in Cytometry
vigorous and cell viability will decline noticeably. In either case, overall cell yield decreases. The third critical parameter is cell yield. Optimal specimens have high cellularity, so after processing, cell yield is sufficient and also representative of the original specimen. An inherent limitation to dissociation of fresh tissue specimens (as compared to dissociation of formalin-fixed paraffin-embedded tissue), is that histologic features of the tissue are not necessarily known at the time of sampling. Although a tissue imprint (see Support Protocol 1) is suggested for quality assurance of representative tissue, the cell yield of dissociation will still be a function of the volume of tissue dissociated, the inherent cellularity of the tissue or neoplasm, the ratio of neoplastic cells to benign stromal or inflammatory cells, and factors intrinsic to the dissociation protocol. It should be noted that there is considerable intercase variability in these parameters, even for a neoplasm of a given type. For example, some breast carcinomas are characterized by extensive areas of “desmoplastic” (scar-like) stromal reaction which contains relatively few neoplastic cells. Another type of breast carcinoma (medullary carcinoma) is characterized by a high percentage of tumor-infiltrating lymphocytes.
Anticipated Results
Typically, cell yield is ∼107 cells per milligram of tissue. Optimally, 58% of the dissociated cells are intact, as estimated by viability studies for fresh tissue dissociation. Approximately 20% to 80% of the cells from a tumor slice will be neoplastic, depending on the degree of host cell contribution. An evaluation of the degree of benign and/or nonneoplastic cell admixture as well as assessment of neoplastic cell morphology and representation is critical. As noted, tissue imprints, cytospin preparations, and paraffin block sections stained with Papanicolau or hemotoxylin-eosin are necessary for evaluation. The quality of the results obtained from these different dissociation techniques will largely depend on the quality and cellularity of the initial sample obtained for assay.
Time Considerations
Handling, Storage, and Preparation of Human Tissues
Disaggregation of solid tumors from fresh or frozen specimens, fixation, and cytospin preparations can be performed in a few hours. Formalin-fixed, paraffin-embedded tissue samples require 2 days for preparation: 2 to 3 hr are required on the first day for processing thin
sections and rehydration, and ∼3 hr on the second day for isolation of nuclei from the hydrated tissue sections (depending on the total number of samples that are processed). Twelve to eighteen samples is a reasonable number for manual preparation by one technician. For procuring fresh samples, the investigator should establish a coordinated protocol with the cooperating pathologist and/or tissue procurement person. The time required for tissue procurement is usually <1 hr, but delays may occur in the surgical suite, the pathology suite, and transport to the laboratory.
Literature Cited Bijman, J., Wagener, D.J., van Rennes, H., Wessles, J.M.C., and van den Brock, P. 1985. Flow cytometric evaluation of cell dispersion from human head and neck tumors. Cytometry 6:334341. Costa, A, Silvestrini, R., Del Bino, G., and Motta, R. 1987. Implications of disaggregation procedures on biological presentation of human solid tumors. Cell Tissue Kinet. 20:171-180. Crissman, J.D., Zarbo, R.J., Niebylski C.D., Corbett, T., and Weaver, D. 1988. Flow cytometric DNA analysis of colon adenocarcinomas: A comparative study of preparatory techniques. Mod. Pathol. 1:198-204. Dressler, L.G. 1993. DNA flow cytometry measurements as surrogate endpoints in chemoprevention trials: Clinical, biological, and quality control considerations. J. Cell. Biochem. 17G:212218. Dressler, L.G. and Bartow, S.A. 1989. DNA flow cytometry in solid tumors: Practical aspects and clinical applications. Semin. Diagn. Pathol. 6:55-82. Dressler, L.G. and Seamer, L.C. 1994. Controls, standards, and histogram interpretation in DNA flow cytometry. In Methods in Cell Biology, Vol. 41, 2nd ed. (Z. Darzynkiewicz, J.P. Robinson, and H.A. Crissman, eds.) pp. 241-262. Academic Press, San Diego. Ensley, J.F., Aciorowski, Z., Hassan, M., Peitraszkiewicz, H., Kish, J., Tapazoglu, E., Jacobs, J., Mathog, R., Weaver, A., Atkinson, D., Binns, P., Ahmed, K., Al-Sarraf, M., Sakr, W., Zarbo, R., and Crissman, J. 1988. The potential and pitfalls of solid tumor flow cytometry with respect to squamous cell cancers of the head and neck. In Head and Neck Oncology Research— Proceedings of the Second International Research Conference on Head and Neck Cancer (G.T. Wold and T. Carey, eds.) pp. 213-224. Kugler, Amsterdam. Hedley, D.W. 1994. DNA analysis from paraffinembedded blocks. In Methods in Cell Biology, Vol. 41, 2nd ed. (Z. Darzynkiewicz, J.P. Robinson, and H.A. Crissman, eds.) pp. 231-240. Academic Press, San Diego.
5.2.14 Current Protocols in Cytometry
Hedley, D.W., Friedlander, M.L., and Taylor, I.W. 1983. Method for analysis of cellular DNA content of paraffin-embedded pathological material using flow cytometry. J. Histochem. Cytochem. 31:1333-1335.
Shankey, T.V., Rabinovitch, P.S., Bagwell, B., Bauer, K.D., Duque, R.E., Hedley, D.W., Mayall, B.H., and Wheeless L. 1993. Guidelines for implementation of clinical DNA cytometry. Cytometry 14:472-477.
Hitchcock, C.L., Ensley, J.F, and Zalupski, M. 1996. Processing of solid tumors for DNA analysis by flow cytometry. In Basic and Clinical Applications of Flow Cytometry—Proceedings of the 24th Annual Detroit Cancer Symposium (F.A. Valeriate, A. Nakeff, and M. Valdivieso, eds.) pp. 159-178. Kluwer Academic Publishers, Boston.
Visscher, D.W. and Crissman, J.D. 1994. Dissociation of intact cells from tumors and normal tissues. In Methods in Cell Biology, Vol 41: Flow Cytometry, 2nd ed. Part A. (Z. Darzynkiewicz, J.P. Robinson, and H.A. Crissman, eds.) pp 1-13. Academic Press, San Diego.
König, J.J., van Dongen, J.W., and Schröder, F.H. 1993. Preferential loss of abnormal prostate carcinoma cells by collagenase treatment. Cytometry 14: 805-810. McDivitt, R.W., Stone, K.R., and Meyer, J.S. 1984. A method for dissociation of viable human breast cancer cells that produces flow cytometric kinetic information similar to that obtained by thymidine labelling. Cancer Res. 44:2628-2633.
Contributed by Lynn G. Dressler University of North Carolina Chapel Hill, North Carolina Dan Visscher Harper Hospital Detroit, Michigan
Specimen Handling, Storage, and Preparation
5.2.15 Current Protocols in Cytometry
Flow Analysis and Sorting of Plant Chromosomes
UNIT 5.3
Methods of chromosome analysis and sorting by flow cytometry (flow cytogenetics) are increasingly used in plant cytogenetics. Flow cytometry allows rapid classification of isolated chromosomes according to their relative DNA content. The resulting distributions, generally referred to as flow karyotypes, may be used for detection of numerical and structural chromosome aberrations. Flow-sorted chromosomes are an invaluable source of DNA, and their availability may greatly simplify the analysis and mapping of complex plant genomes. For instance, sorted chromosomes may be used for physical gene mapping, isolation of molecular markers, and construction of chromosome-specific DNA libraries. A general outline of the procedure for flow cytometric analysis and sorting of plant chromosomes consists of the following steps: (1) accumulation of cells in metaphase, (2) preparation of chromosome suspensions, (3) flow analysis and sorting, and (4) processing of sorted chromosomes. Basic Protocol 1 provides a procedure for accumulation of cells in metaphase. Originally developed for the field bean (Vicia faba), this procedure is also suitable for other large-seeded legumes such as the garden pea (Pisum sativum). A modification of the procedure that is suitable for cereals such as barley (Hordeum vulgare) and rye (Secale cereale) is given in Alternate Protocol 1. Both protocols were developed for use with roots of young seedlings. The modification of Basic Protocol 1 for the use with immortal root (hairy root) cultures is described in Alternate Protocol 2. The extent of cell-cycle synchrony may be monitored according to Support Protocol 1. Preparation of chromosome suspensions from synchronized root tips is described in Basic Protocol 2. Isolated chromosomes can be subjected either to univariate flow analysis (Basic Protocol 3) or to bivariate analysis (Alternate Protocol 3). Both protocols describe chromosome sorting as well. Alternate Protocol 4 presents a procedure for two-step sorting that is useful for sorting chromosomes whose frequency in the original suspension is too low, or for chromosomes that are difficult to sort with acceptable purity. The alignment of the flow cytometer is crucial to achieve the highest purity of sorted chromosome fractions and is described in Support Protocol 2. The usefulness of the sorted chromosome fractions depends on their purity. Support Protocol 3 gives a method for detecting contamination of the sorted fractions and for estimating the degree of purity. Sorted chromosomes are suitable for a variety of uses. Basic Protocol 4 provides a procedure for physical gene mapping. ACCUMULATION OF ROOT-TIP CELLS AT METAPHASE IN LARGE-SEEDED LEGUMES
BASIC PROTOCOL 1
This protocol induces a high degree of metaphase synchrony in meristem root-tip cells. The procedure uses a combination of hydroxyurea (a DNA synthesis inhibitor) and the antimicrotubular drug amiprophos-methyl. It has been developed for use with largeseeded legumes such as field bean and garden pea. Optimal concentrations and treatment times are given for field bean and garden pea, and should be determined experimentally for other legumes as they may vary from species to species (see Critical Parameters). For hydroxyrurea, the concentration as well as incubation and wash times should be optimized to achieve the highest degree of mitotic synchrony. For amiprophos-methyl, the optimal concentration is the lowest concentration
Contributed by Jaroslav DoleÓel, Ji¯í Macas, and Sergio Lucretti Current Protocols in Cytometry (1999) 5.3.1-5.3.33 Copyright © 1999 by John Wiley & Sons, Inc.
Specimen Handling, Storage, and Preparation
5.3.1 Supplement 9
that effectively blocks cells at metaphase. In both cases, Support Protocol 1 may be used to assess metaphase synchrony. NOTE: Adjust the temperature of all solutions to 25° ± 0.5°C prior to use. Perform all incubations in the dark in a biological incubator at 25° ± 0.5°C. Aerate all solutions. Keep aeration stones and tubing clean to avoid extensive contamination by bacteria and fungi. Materials Seeds Inert substrate for seed germination (e.g., perlite) 1× Hoagland’s nutrient solution (see recipe) Hydroxyurea treatment solution (see recipe) Amiprophos-methyl treatment solution (see recipe) Aquarium bubbler with tubing and aeration stones 4-liter plastic tray (e.g., 25 cm long × 15 cm wide × 11 cm high) 750-ml plastic tray (e.g., 14 cm long × 8 cm wide × 10 cm high) including an open-mesh basket to hold germinated seeds Biological incubator (heating/cooling) with internal temperature adjusted to 25° ± 0.5°C Germinate seeds 1. Imbibe seeds for 24 hr in deionized H2O with aeration. Approximately 30 seedlings are needed to prepare one sample (1 ml of chromosome suspension prepared in Basic Protocol 2).
2. Wet an inert substrate with 1× Hoagland’s nutrient solution and put it into a 4-liter plastic tray. 3. Wash seeds in deionized H2O, spread them over the surface of the wet substrate, and cover them with a 1-cm layer of wet substrate. 4. Cover the tray with aluminum foil and germinate the seeds at 25° ± 0.5°C in a biological incubator in the dark. Optimal root length is ∼4 cm and should be achieved in 2 to 3 days.
5. Remove seedlings from the substrate and wash them in deionized H2O. Accumulate root-tip cells in metaphase 6. Select ∼30 seedlings with primary roots of similar length. 7. Thread seedling roots through the holes of an open-mesh basket placed in a 750-ml plastic tray filled with deionized H2O. 8. Transfer the basket with seedlings to a second plastic tray containing hydroxyurea treatment solution. 9. Incubate for 18.5 hr (field bean) or 18 hr (garden pea) at 25°C. 10. Wash the roots vigorously in several changes of deionized H2O. 11. Incubate in hydroxyurea-free 1× Hoagland’s nutrient solution for 4.5 hr (field bean) or 3 hr (garden pea) at 25°C. 12. Transfer the basket with seedlings to a tray filled with amiprophos-methyl treatment solution and incubate for 2 hr at 25°C. Flow Analysis and Sorting of Plant Chromosomes
To determine the degree of synchrony, see Support Protocol 1. Seedlings should not be stored before chromosome suspensions are prepared (see Basic Protocol 2).
5.3.2 Supplement 9
Current Protocols in Cytometry
ACCUMULATION OF ROOT-TIP CELLS AT METAPHASE IN CEREAL SPECIES
ALTERNATE PROTOCOL 1
Basic Protocol 1 has been modified for cereals and optimized for preparation of chromosome suspensions in barley and rye. Additional Materials (also see Basic Protocol 1) Cereal seeds 18-cm-diameter glass petri dish Paper towels cut to 18-cm diameter Filter paper cut to 18-cm diameter 1. Place several layers of paper towels into an 18-cm glass petri dish and top them with a single sheet of filter paper. 2. Moisten the paper layers with deionized H2O. 3. Spread seeds on the filter paper surface. Approximately 50 seedlings are needed to prepare one sample (1 ml of chromosome suspension prepared in Basic Protocol 2).
4. Cover the petri dish and germinate the seeds at 25° ± 0.5°C in a biological incubator in the dark. Optimal root length is 2 to 3 cm and should be achieved in 2 to 3 days.
5. Select ∼50 seedlings with primary roots of similar length and process through hydroxyurea and amiprophos-methyl treatment as described (see Basic Protocol 1, steps 6 to 12). For hydroxyurea treatment, incubate 18 hr. For hydroxyurea-free Hoagland’s nutrient solution, incubate for 6.5 hr (barley) or 5.5 hr (rye). 6. Transfer the basket with seedlings to a plastic tray filled with an ice-water bath (1° to 2°C). 7. Place the container in a refrigerator and leave overnight. To determine the degree of synchrony, see Support Protocol 1. Seedlings should not be stored before chromosome suspensions are prepared (see Basic Protocol 2).
ACCUMULATION OF ROOT-TIP CELLS AT METAPHASE FROM ROOTS CULTURED IN VITRO
ALTERNATE PROTOCOL 2
This protocol has been developed for use with in vitro cultured hairy roots of some legumes (field bean and garden pea). Such immortal cultures can be obtained after transformation with a suitable strain of Agrobacterium rhizogenes (Tempé and CasseDelbart, 1989). NOTE: Sterilize all solutions by filtration through a 0.22-µm filter. Perform all manipulations under aseptic conditions in a laminar flow hood. Perform all incubations on an orbital shaker in the dark at 25° ± 0.5°C. Materials Hairy root cultures (Tempé and Casse-Delbart, 1989) B5 nutrient medium (see recipe) with and without 2 mM hydroxyurea, sterile Amiprophos-methyl Sterile pipets, forceps, scalpel Orbital shaker Biological incubator (heating/cooling) at 25°C ± 0.5°C
Specimen Handling, Storage, and Preparation
5.3.3 Current Protocols in Cytometry
Supplement 9
1. Select rapidly growing hairy root cultures. 2. Cut root tips 1 to 2 cm from the tip and transfer them to a 250-ml Erlenmeyer flask containing 25 ml sterile B5 nutrient medium supplemented with 2 mM hydroxyurea. Approximately 300 root tips are needed to prepare one sample (1 ml of chromosome suspension prepared in Basic Protocol 2). Root tips should be taken only from actively growing cultures.
3. Incubate for 18 hr on an orbital shaker at 25° ± 0.5°C. 4. Wash roots twice in sterile hydroxyurea-free B5 nutrient medium. 5. Incubate in sterile hydroxyurea-free B5 medium for 6 hr on an orbital shaker at 25°C. 6. Transfer roots to sterile B5 medium containing 10 µM amiprophos-methyl and incubate for 2 hr at 25°C. To determine the degree of synchrony, see Support Protocol 1. Seedlings should not be stored before chromosome suspensions are prepared (see Basic Protocol 2). SUPPORT PROTOCOL 1
ANALYSIS OF THE DEGREE OF METAPHASE SYNCHRONY This support protocol is used to estimate the frequency of metaphase cells following the procedures for metaphase accumulation (Basic Protocol 1, and Alternate Protocols 1 and 2). Materials Root tips synchronized in metaphase (see Basic Protocol 1, Alternate Protocol 1, or Alternate Protocol 2) 3:1 (v/v) ethanol/glacial acetic acid, freshly prepared 70% and 96% (v/v) ethanol 5 N HCl Schiff’s reagent (see recipe) 45% (v/v) acetic acid Fructose syrup (see recipe) Xylene DePeX (Serva) Microscope slides 18 × 18–mm coverslips Coplin jars Microscope Fix and stain cells 1. Harvest 1-cm root tips in deionized H2O. About ten root tips should be used to allow for accidental losses during the procedure and for bad squashes. At least five good squashes should be used for microscopic evaluation.
2. Fix in 3:1 ethanol/glacial acetic acid overnight at 4°C. CAUTION: Glacial acetic acid is volatile. Concentrated acids must be handled with great care. Wear gloves and safety glasses and work in a chemical fume hood.
3. Remove fixative with several washes in 70% ethanol. Fixed tips may be stored in 70% ethanol at 4°C for up to one year.
4. Wash tips in several changes of deionized H2O. 5. Hydrolyze tips in 5 N HCl at room temperature for 25 min. Flow Analysis and Sorting of Plant Chromosomes
CAUTION: Concentrated HCl is volatile. Concentrated acids should be handled with great care. Wear gloves and safety glasses and work in a chemical fume hood.
5.3.4 Supplement 9
Current Protocols in Cytometry
6. Wash in deionized H2O and incubate in Schiff’s reagent for 1 hr at room temperature. Prepare and analyze squashes 7. Wash tips in deionized H2O and macerate (soften) them for ∼1 min in 45% acetic acid at room temperature. To make slides for immediate use: 8a. Cut off the darkly stained meristem tip and squash it in a drop of fructose syrup between a microscope slide and an 18 × 18–mm coverslip. Repeat to prepare at least five different slides. 9a. On each of five slides, analyze ≥1000 cells and determine the proportion of cells in metaphase. Squash preparations in fructose syrup can be maintained for a few days in a refrigerator, but are not permanent.
To make permanent slides: 8b. Squash the darkly stained meristem tip in a drop of 45% acetic acid and immediately place the slide on a block of dry ice. Allow the slide to freeze. Repeat to prepare at least five different slides. 9b. Peel off coverslips. Dehydrate tips in two changes of 96% ethanol in Coplin jars and leave to air dry overnight. 10b. Dip slides in xylene and mount each in a drop of DePeX. CAUTION: Xylene is flammable and may cause a narcotic effect, lung irritation, chest pain, and edema. Wear gloves and safety glasses and work in a chemical fume hood.
11b. On each of five slides, analyze ≥1000 cells and determine the proportion of cells in metaphase. PREPARATION OF SUSPENSIONS OF INTACT PLANT CHROMOSOMES After a mild fixation with formaldehyde, chromosomes are mechanically released from synchronized root tips into a polyamine buffer (LB01), which stabilizes their structure. Chromosome suspensions prepared according to this procedure are suitable for flow cytometric analysis and sorting (see Basic Protocol 3, Alternate Protocol 3, and Alternate Protocol 4).
BASIC PROTOCOL 2
Materials Root tips synchronized in metaphase (see Basic Protocol 1, Alternate Protocol 1, or Alternate Protocol 2) Formaldehyde fixative (see recipe) Tris buffer (see recipe) LB01 lysis buffer (see recipe) 0.1 mg/ml DAPI stock solution (see recipe) 5°C water bath 5-ml polystyrene tubes (e.g., Falcon 2054; Becton Dickinson) Mechanical homogenizer (e.g., Polytron PT1200 with a PT-DA 1205/5 probe; Kinematica) 50-µm (pore size) nylon mesh in 4 × 4–cm squares 0.5-ml tubes for polymerase chain reaction (PCR) Microscope slides Fluorescence microscope with 10× to 20× objective and DAPI filter set
Specimen Handling, Storage, and Preparation
5.3.5 Current Protocols in Cytometry
Supplement 9
Prepare chromosome suspension 1. Harvest 1-cm root tips and transfer into deionized H2O. Root tips must be harvested immediately after treatment with amiprophos-methyl (see Basic Protocol 1, step 12, or see Alternate Protocol 2, step 6) or ice water (see Alternate Protocol 1, step 7).
2. Immediately transfer root tips to 25 ml formaldehyde fixative and fix at 5°C for 30 min (field bean) or 20 min (garden pea, barley, and rye). 3. Wash roots in 25 ml Tris buffer three times for 5 min each at 5°C. 4. Excise root meristems and transfer them to a 5-ml polystyrene tube containing 1 ml LB01 lysis buffer. 5. Isolate chromosomes by homogenizing at 9500 rpm using a Polytron PT1200 for 15 sec (field bean and garden pea) or 10 sec (barley and rye). 6. Filter the suspension through 50-µm nylon mesh into a 5-ml polystyrene tube. 7. Store the suspension on ice. Although the chromosome suspension can be stored overnight, it is recommended to analyze the chromosomes on the same day. Alternatively, root tips may be homogenized using a razor blade. Transfer fixed root tips into 1.25 ml LB01 lysis buffer in a 6-cm glass petri dish. Chop meristem root tips individually using a sharp razor blade, avoiding dispersion or drying. Filter the suspension through a 50-ìm nylon mesh into a polystyrene tube. Pass the suspension once through a 22-G needle to disperse intact metaphase chromosomes, and store the suspension on ice. This method is more laborious and inconvenient in species with small root tips. However, it results in higher yield of longer chromosomes in species with large chromosomes, such as field bean.
Examine quality of chromosomes 8. Transfer 50 µl chromosome suspension into a 0.5-ml PCR tube. 9. Add 1 µl of 0.1 mg/ml DAPI stock solution. 10. Place a small drop (∼10 µl) of DAPI-stained suspension on a microscope slide. 11. Using a fluorescence microscope, observe the suspension under low magnification (10× to 20×). Do not cover with a coverslip. The suspension should contain intact nuclei and chromosomes. The concentration of chromosomes in the sample should be ≥5 × 105/ml. If the chromosomes are damaged (broken and/or appear as long extended fibers), the formaldehyde fixation was too weak and should be prolonged. If the chromosomes are aggregated and/or the cells remain intact, the fixation was too strong and should be shortened. BASIC PROTOCOL 3
Flow Analysis and Sorting of Plant Chromosomes
UNIVARIATE FLOW KARYOTYPING AND CHROMOSOME SORTING OF PLANT CHROMOSOMES This protocol describes the analysis and sorting of plant chromosomes stained with DAPI. The flow cytometer must be equipped with a UV light source to excite this dye. Materials Chromosome suspension (see Basic Protocol 2) 0.1 mg/ml DAPI stock solution (see recipe) LB01 lysis buffer (see recipe) Collection liquid Sheath fluid SF50 for flow cytometric analysis: 40 mM KCl/10 mM NaCl (sterilize by autoclaving)
5.3.6 Supplement 9
Current Protocols in Cytometry
Computer with spreadsheet or other software for theoretical flow karyotypes (available from DoleÓel) 20-µm-pore-size nylon mesh in 4 × 4–cm squares Flow cytometer and sorter (e.g., Becton Dickinson FACSVantage) with a UV argon laser (e.g., Coherent Innova 305) and a 424 ± 44–nm band-pass filter Microscope slides Fluorescence microscope with DAPI filter set Additional reagents and equipment for aligning the flow cytometer and adjusting the sorting device (see Support Protocol 2), and for determining purity of sorted chromosomes (see Support Protocol 3) Prepare theoretical flow karyotypes 1. Prepare theoretical flow karyotypes using either a spreadsheet or dedicated computer software (for details see Conia et al., 1989; DoleÓel, 1991). 2. Predict the assignment of chromosomes to chromosome peaks on a flow karyotype. 3. Determine the resolution (coefficient of variation) of chromosome peaks needed to discriminate individual chromosome types. Theoretical flow karyotypes can be modeled based on relative length or DNA content of individual chromosomes, and are very useful in planning experiments with chromosome analysis. The model predicts the complexity of the analysis and the limitations of univariate flow karyotyping. It may be used to predict positions of peaks representing specific chromosomes on a flow karyotype, and to study the effect of resolution (coefficient of variation) of chromosome peaks on discriminiation of individual chromosome types.
Perform flow cytometry 4. Stain a chromosome suspension (∼1 ml/sample) by adding 0.1 mg/ml DAPI stock solution to a final concentration of 2 µg/ml. Analysis can be performed immediately after addition of DAPI, without incubation. If necessary, the stained suspension can be kept on ice.
5. Filter the suspension through a 20-µm nylon mesh. 6. Make sure that the flow cytometer is properly aligned for univariate analysis (see Support Protocol 2) and that a 424 ± 44–nm band-pass filter is placed in front of the DAPI fluorescence detector. 7. Run a dummy sample (LB01 lysis buffer containing 2 µl/ml DAPI) to equilibrate the sample line. This ensures stable peak positions during analysis and sorting.
8. Introduce the sample and let it stabilize at the appropriate flow rate (e.g., 200 particles/sec). If possible, do not change the flow rate during the analysis. Significant changes in the flow rate during the analysis may result in peak shifts.
9. Set a gating region on a dot plot of forward scatter (FS) and DAPI peak/pulse height to exclude debris, nuclei, and large clumps. 10. Adjust photomultiplier voltage and amplification gains so that chromosome peaks are evenly distributed on a histogram of DAPI signal pulse area/integral. 11. Collect 20,000 to 50,000 chromosomes and save the results on a computer disk. Sort chromosomes 12. Make sure that the sorting device is properly adjusted (see Support Protocol 2).
Specimen Handling, Storage, and Preparation
5.3.7 Current Protocols in Cytometry
Supplement 9
13. Run the sample and display the signals on a dot plot of DAPI signal pulse width versus area/integral. 14. Adjust the DAPI pulse width amplifier gain and width offset as needed to achieve optimal resolution of the width signal. 15. Check for stability of the break-off point and of the side streams. 16. Define sorting region for the largest chromosome on the dot plot of DAPI pulse width versus DAPI pulse area/integral. 17. Select one-droplet sort envelope (number of deflected droplets) and sort mode giving the highest purity and count precision (e.g., counter mode in FACSVantage instrument). 18. Sort an exact number of chromosomes (e.g., 50) onto a microscope slide. 19. Check the number of chromosomes using a fluorescence microscope (do not cover the drop with a coverslip). 20. If the number is not correct, repeat adjustment of the sorting device using fluorescent beads (see Support Protocol 2). 21. Define a sorting region for the chromosome to be sorted on the dot plot of DAPI pulse width versus DAPI pulse area/integral. 22. Select sort mode and sort envelope according to required purity, number of chromosomes to be sorted, and desired volume for the sorted fraction. Consult manufacturer’s instructions for explanation of sort modes and sort envelopes. 23. Sort the required number of chromosomes into a polystyrene tube containing the appropriate amount of collection liquid. The amount and composition of the collection liquid depends on the number of sorted chromosomes and on their subsequent use. For PCR, use a small volume (20 to 60 ìl) of sterile, deionized H2O in a 0.5-ml PCR tube.
24. Microcentrifuge the tube for 5 to 10 sec at room temperature. 25. Sort chromosomes onto a microscope slide for determination of purity (see Support Protocol 3). ALTERNATE PROTOCOL 3
BIVARIATE FLOW KARYOTYPING AND CHROMOSOME SORTING This protocol gives a procedure for chromosome isolation and bivariate analysis after dual staining with DAPI, and mithramycin, which preferentially bind AT-rich and GC-rich regions of DNA, respectively. Additional Materials (also see Basic Protocol 3) 100 mM MgSO4 solution (filter through a 0.22-µm filter; store at 4°C) 1 mg/ml mithramycin stock solution (see recipe) Flow cytometer and sorter (e.g., FACSVantage, Becton Dickinson) with two argon ion lasers (e.g., Coherent Innova 305), a 424 ± 44–nm band-pass filter, and a 490-nm long-pass filter
Flow Analysis and Sorting of Plant Chromosomes
Perform flow cytometry 1. To a chromosome suspension (∼1 ml/sample), add 100 mM MgSO4 solution to a final concentration of 10 mM.
5.3.8 Supplement 9
Current Protocols in Cytometry
2. Stain chromosomes by adding 0.1 mg/ml DAPI stock solution to a final concentration of 1.5 µg/ml and 1 mg/ml mithramycin stock solution to a final concentration of 20 µg/ml. 3. Allow to equilibrate for 30 min on ice. 4. Make sure that the flow cytometer is properly aligned for bivariate analysis (see Support Protocol 2). Use a half mirror to split the DAPI fluorescence through a 424 ± 44–nm band-pass filter and the mithramycin fluorescence through a 490-nm long-pass filter. Because of the optical design of the dual-laser FACSVantage instrument (which employs spatially separated beam geometry), a half mirror is used to reflect all light from the second laser at 90 degrees toward the mithramycin detector. The DAPI fluorescence (excited by the first laser) is not reflected and enters the DAPI detector directly.
5. Run a dummy sample (LB01 lysis buffer containing 1.5 µg/ml DAPI and 20 µg/ml mithramycin) to equilibrate the sample line. This ensures stable peak positions during analysis and sorting.
6. Filter the sample through a 20-µm nylon mesh. 7. Run the sample and let it stabilize at the appropriate flow rate (e.g., 200 particles/sec). If possible, do not change the flow rate during the analysis. Significant change in the flow rate during the analysis may result in peak shifts.
8. Set a gating region on a dot plot of FS versus DAPI peak/pulse height. Gating is used to exclude small debris as well as large clumps and nuclei. Use this gate to display other parameters (DAPI pulse area/integral and mithramycin pulse area/integral).
9. Adjust photomultiplier voltages and amplification gains so that chromosome peaks are evenly distributed on histograms of DAPI pulse area/integral and mithramycin pulse area/integral. 10. Display the data on a dot plot of DAPI pulse area/integral versus mithramycin pulse area/integral. 11. Collect 20,000 to 50,000 chromosomes and save the results on a computer disk. Sort chromosomes 12. Make sure that the sorting device is properly adjusted (see Support Protocol 2). 13. Run the sample and display the signals on a dot plot of DAPI pulse area/integral versus mithramycin pulse area/integral. 14. Check for stability of the break-off point and of the side streams. 15. Define sorting region for the largest chromosome on the dot plot of DAPI pulse area/integral versus mithramycin pulse area/integral. 16. Proceed with cell sorting as described (see Basic Protocol 3, steps 17 to 25), but define a sorting region on the dot plot of DAPI pulse area/integral versus mithramycin pulse area/integral (step 21).
Specimen Handling, Storage, and Preparation
5.3.9 Current Protocols in Cytometry
Supplement 9
ALTERNATE PROTOCOL 4
TWO-STEP SORTING This protocol is used to sort chromosomes when their frequency in the original suspension is too low. This is frequently the case for large chromosomes, which break more easily than smaller chromosomes during chromosome isolation. During the first sort, the sample is enriched for the required chromosome. During the second sort, the chromosomes are sorted with a high purity. In some cases, it may be practical to enrich the sample for more than one chromosome. Individual chromosomes are sorted during the second sort. Additional Materials (also see Basic Protocol 3 and Alternate Protocol 3) 1.5-ml polystyrene cup (e.g., Deltalab) 1.5-ml polystyrene PCR tube Enrich sample for desired chromosome(s) 1. Make sure that the sorting device is properly adjusted (see Support Protocol 2). 2. Run the sample and display the signals on a suitable distribution (see Basic Protocol 3, steps 4 to 10 and 13 to 14, or see Alternate Protocol 3, steps 1 to 10 and 13). 3. Perform trial sorting onto a microscipe slide and check the number of chromosomes (see Basic Protocol 3, steps 15 to 20), using a dot plot of DAPI pulse width versus DAPI area/integral to define the sorting region (step 16) for univariate analysis, or using DAPI area/integral versus mithramycin area/integral for bivariate analysis. 4. Select the sort mode and sort envelope that allow for the highest recovery (e.g., enrich mode and three deflected droplets in FACSVantage instrument). 5. On a suitable distribution (see step 3), define a sorting region for the chromosome(s) to be sorted. 6. Sort ≥100,000 chromosomes into 400 µl LB01 lysis buffer in a 1.5-ml polystyrene cup. The actual number of chromosomes that should be sorted depends on the number of chromosomes that will be sorted during the second sort. It is recommended to sort at least five times more chromosomes than the final number required.
Perform second sort 7. Add fluorescent dye(s) to reach recommended final concentrations (see Basic Protocol 3, step 4, or see Alternate Protocol 3, steps 1 to 3). 8. Run the sample and define a sorting region for the chromosome to be sorted. If only one chromosome has been presorted, this step should use the same sorting regions defined in step 2. However, it is also possible to presort two or more different chromosome types. In this case, the sort window is modified prior to the second sort when only one chromosome is sorted at a time.
9. Select sort mode and sort envelope according to required purity, number of chromosomes to be sorted, and desired volume for the sorted fraction. Consult manufacturer’s instructions for explanation of sort modes and sort envelopes. 10. Sort the required number of chromosomes into a 1.5-ml polystyrene PCR tube containing the appropriate amount of collection liquid. The amount and composition of the collection liquid depends on the number of sorted chromosomes and on their subsequent use. For PCR, use a small volume (20 to 60 ìl) of sterile, deionized H2O in 0.5-ml PCR tube.
11. Microcentrifuge the tube for 5 to 10 sec at room temperature. Flow Analysis and Sorting of Plant Chromosomes
12. Sort chromosomes onto a microscope slide for determination of purity (see Support Protocol 3).
5.3.10 Supplement 9
Current Protocols in Cytometry
ALIGNMENT OF FLOW CYTOMETER FOR CHROMOSOME ANALYSIS AND SORTING
SUPPORT PROTOCOL 2
The alignment of the flow cytometer is crucial to achieve the highest purity in the sorted chromosome fraction. This protocol describes suitable setups and fine tuning of the instrument for chromosome analysis and sorting. Refer to manufacturer’s instructions for operation and basic alignment of the instrument. Additional Materials (also see Basic Protocol 3 and Alternate Protocol 3) Calibration beads (Polysciences): BB beads (univariate analysis) or YG beads (bivariate analysis) 530 ± 30–nm and 585 ± 42–nm band-pass filters (bivariate analysis) Align flow cytometer 1. Switch on the laser(s): a. Univariate analysis: Operate the argon ion laser in multi-UV mode (351.1 to 363.8 nm) with 300 mW output power. b. Bivariate analysis: Operate the first argon ion laser in multi-UV mode (351.1 to 363.8 nm) with 300 mW output power, and the second argon ion laser at 457.9 nm with 300 mW output power. 2. Allow the laser(s) to stabilize for 30 min. Peak the laser optics for maximum light output. 3. Empty the waste container and fill the sheath container with sterile sheath fluid SF50. 4. Adjust sheath fluid pressure (e.g., 10 psi for FACSVantage instrument) and leave the fluid running to fill all plastic lines and filters in the instrument. 5. Install a nozzle (70-µm orifice) and check for air bubbles. 6. Install appropriate optical filters for alignment: a. Univariate analysis: Use a 424 ± 44–nm band-pass filter in front of the DAPI detector. b. Bivariate analysis: Use a 530 ± 30–nm band-pass filter in front of the DAPI detector and a 585 ± 42–nm band-pass filter in front of the mithramycin detector. Use a half mirror to split the fluorescence from the first and the second lasers. 7. Trigger on forward scatter (FS) and select a threshold level. 8. Run fluorescent beads at a flow rate of 200 particles/sec, using BB beads for univariate analysis and YG beads for bivariate analysis. 9. Display the data on a dot plot of FS versus DAPI, and on one-parameter histograms of FS and DAPI fluorescence. 10. Align the instrument to achieve maximum signal intensity and minimum coefficient of variation of FS and DAPI signals. 11. For bivariate analysis, use a one-parameter histogram of mithramycin fluorescence and align the second laser to achieve maximum intensity and the lowest coefficient of variation of the mithramycin signal. Change only settings specific for the second laser; do not adjust other controls. Adjust dual-laser delay and dead-time parameters as needed. Adjust sorting device 12. Switch on the sorting device and warm up the deflection plates for 30 min.
Specimen Handling, Storage, and Preparation
5.3.11 Current Protocols in Cytometry
Supplement 9
13. Run calibration beads at a flow rate of 200 particles/sec. It is important that adjustment of the sorting device be done with the sample running.
14. Switch on the test sort mode. 15. Adjust the drop drive frequency and drop drive amplitude to break the stream at a suitable distance from the laser intercept point (check for satellite drops). 16. Adjust the drop drive phase to obtain single side streams. 17. Adjust the position of side streams so that they enter the collection tubes. 18. Switch off the test sort mode. 19. Calculate drop delay and perform its optimization: a. Define sorting region for single beads (avoiding doublets and clumps). b. Select one-droplet sort envelope (number of deflected droplets) and sort mode giving the highest purity and count precision (e.g., counter mode in FACSVantage instrument). c. Sort 20 beads onto a microscope slide. d. Check the number of beads using a fluorescence microscope (do not cover the drop with a coverslip). e. If the number of sorted beads is not correct, change the drop delay by a factor of 0.25 and repeat steps c to e until the drop delay setting results in the highest number of sorted beads. To determine optimal drop delay, it may be convenient to sort 20 beads (at 0.25-step settings) on the same slide. SUPPORT PROTOCOL 3
ESTIMATION OF PURITY OF SORTED FRACTIONS USING PRINS During the initial flow karyotyping experiment for a given plant species or line, it is very useful to sort chromosomes from a stained suspension onto a slide to determine the chromosome content of individual peaks, and to examine the purity that may be achieved when sorting specific chromosomes. Additionally, when chromosomes are sorted for molecular analyses (e.g., physical mapping; see Basic Protocol 4), chromosomes should be sorted from the stained suspension onto a slide after they are sorted from a tube. This permits examination of the purity of a sorted chromosome fraction achieved under the same sorting conditions. This protocol allows precise estimation of the purity of a sorted fraction. The procedure is based on specific fluorescence labeling of repetitive DNA sequences that show a characteristic pattern of distribution among the chromosomes. The sequences are labeled with FITC using primed in situ (PRINS) labeling and the chromosomes are evaluated using a fluorescence microscope. NOTE: Experiments involving PCR require extremely careful technique to prevent contamination (see Kramer and Coen, 1999).
Flow Analysis and Sorting of Plant Chromosomes
Materials LB01 lysis buffer (see recipe)/10% (w/v) sucrose or PRINS buffer (see recipe)/10% (w/v) sucrose PRINS reaction mix (see recipe) containing fluorescein-labeled nucleotides Stop buffer (see recipe) Wash buffer (see recipe)
5.3.12 Supplement 9
Current Protocols in Cytometry
0.1 mg/ml DAPI stock solution (see recipe) Vectashield antifade solution (Vector Laboratories) Rubber cement Microscope slides Aseptic box (Steril Helios, Steril S.p.A.) 25-µl Frame-Seal chambers (MJ Research) with 26 × 40–mm polyester coverslips Thermal cycler equipped with a flat plate for microscope slides (e.g., PTC-100, MJ Research) 10-cm glass petri dish 18 × 18–mm glass coverslips Fluorescence microscope with epiillumination and filter sets for DAPI, fluorescein isothiocyanate (FITC), and dual DAPI/FITC fluorescence Additional reagents and equipment for sorting chromosomes (see Basic Protocol 3, Alternate Protocol 3, or Alternate Protocol 4) Sort chromosomes 1. Pipet 15 µl LB01 lysis buffer/10% sucrose (for cereal chromosomes) or PRINS buffer/10% sucrose (for legume chromosomes) onto a clean microscope slide. 2. Immediately sort 1000 chromosomes into the drop (see Basic Protocol 3, steps 4 to 10 and 12 to 22; see Alternate Protocol 3, steps 1 to 10 and 12 to 16; or see Alternate Protocol 4, steps 1 to 9). 3. Air dry in an aseptic box for ∼1 hr and leave overnight at room temperature. Although the drops dry within 1 hr, the PRINS reaction should not be performed until the next day.
Perform PRINS reaction 4. Stick a 25-µl Frame-Seal chamber to the slide over the specimen area. 5. Pipet 25 µl PRINS reaction mix into the frame and place the polyester coverslip over the frame. 6. Place slides in a thermal cycler and run the PRINS reaction using the following PCR cycles: 1 cycle:
8 cycles:
1 cycle:
5 min 5 min 10 min 1 min 1 min 3 min 1 min 5 min 10 min
94°C 55°C 72°C 94°C 55°C 72°C 94°C 55°C 72°C
(denaturation) (annealing) (extension) (denaturation) (annealing) (extension) (denaturation) (annealing) (extension).
The actual conditions for the PRINS reaction (i.e., denaturation and annealing temperatures) must be optimized for a given species and primer pair.
7. Remove the cover, add 100 µl stop buffer, and incubate for 2 min at 70°C. 8. Remove the stop buffer and transfer the slide to a 10-cm glass petri dish. 9. Add 70 µl wash buffer and incubate at room temperature for 5 min. Repeat twice. Specimen Handling, Storage, and Preparation
5.3.13 Current Protocols in Cytometry
Supplement 9
Examine slides 10. Add 70 µl wash buffer containing 0.2 µg/ml DAPI to counterstain the chromosomes and incubate for 15 min at room temperature. 11. Drain excess fluid, but do not dry. 12. Add 8 µl Vectashield antifade solution and cover with an 18 × 18–mm glass coverslip. 13. Gently squeeze out excess solution and seal with rubber cement. 14. Examine with a fluorescence microscope using a DAPI filter first to localize sorted chromosomes. Avoid prolonged exposure to excitation light, which rapidly bleaches both DAPI and FITC. BASIC PROTOCOL 4
PHYSICAL MAPPING OF DNA SEQUENCES USING PCR This protocol describes the use of PCR on flow-sorted chromosomes for physical mapping of DNA sequences to individual chromosomes or their regions. NOTE: Experiments involving PCR require extremely careful technique to prevent contamination (see Kramer and Coen, 1999). Materials PCR premix (see recipe) Loading buffer (see recipe) 1.5% (w/v) agarose gel (see recipe) 1× TAE electrophoresis buffer (see recipe) DNA molecular weight markers 0.5 µg/ml ethidium bromide solution (see recipe) 0.5-ml PCR tubes Thermal cycler Horizontal gel electrophoresis apparatus and power supply UV transilluminator and gel documentation system Additional reagents and equipment for sorting chromosomes (see Basic Protocol 3, Alternate Protocol 3, or Alternate Protocol 4) Sort chromosomes 1. Prepare 0.5-ml PCR tubes containing 19 µl sterile, deionized H2O. The final volume after sorting will be ∼20 ìl.
2. Sort 500 chromosomes into each tube (see Basic Protocol 3, steps 4 to 10 and 12 to 22; see Alternate Protocol 3, steps 1 to 10 and 12 to 16; or see Alternate Protocol 4, steps 1 to 9). 3. Freeze the tube and store at −20°C for up to 6 months. It is important to freeze the tubes even if the reaction is to be performed on the same day of sorting.
Perform PCR 4. Thaw a chromosome fraction and add 30 µl PCR premix. Vortex and microcentrifuge briefly. Flow Analysis and Sorting of Plant Chromosomes
5.3.14 Supplement 9
Current Protocols in Cytometry
5. Place tube in a thermal cycler and perform PCR amplification using the following cycles: Initial step: 35 cycles:
2 min 1 min 1 min 2 min 10 min indefinitely
1 cycle: Final step:
94°C 94°C 58°C 72°C 72°C 4°C
(denaturation) (denaturation) (annealing) (extension) (extension) (hold)
Annealing temperature must be optimized for a given primer pair and template.
Analyze PCR products 6. Take equal amounts of PCR products (5 to 10 µl) from each tube and add 1 to 2 µl loading buffer. 7. Load samples onto a 1.5% agarose gel bathed in 1× TAE electrophoresis buffer. Also load DNA molecular weight markers. 8. Run electrophoresis at a constant voltage of 4 to 5 V/cm until the bromphenol blue reaches a point 3 cm from the edge of the gel. 9. Stain the gel with 0.5 µg/ml ethidium bromide solution. 10. Photograph the gel and analyze the presence of products in individual lanes. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Agarose gel, 1.5% (w/v) 3 g electrophoresis-grade agarose 200 ml 1× TAE electrophoresis buffer (see recipe) Soak for 20 min at room temperature. Boil in a microwave oven to dissolve completely. Adjust volume to 200 ml with H2O. Pour into a casting form and let solidify. Amiprophos-methyl treatment solution Prepare a 20 mM stock solution by dissolving 60.86 mg amiprophos-methyl in 10 ml cold acetone. Store up to 1 year at −20°C in 1-ml aliquots. Prepare the treatment solution immediately before use by combining amiprophos-methyl stock solution and 0.1× or 1× Hoagland’s nutrient solution (see recipe) as specified in Table 5.3.1.
Table 5.3.1 Preparation of Amiprophos-Methyl Treatment Solution
Hoagland’s nutrient solution
20 mM amiprophos-methyl
Species Concentration Volume (ml) Vicia faba Pisum sativum Hordeum vulgare Secale cereale
1× 1× 0.1× 1×
750 750 750 750
Amount (µl)
Final concentration (µM)
94.8 380.0 94.8 19.0
2.50 10.0 2.50 0.50
Specimen Handling, Storage, and Preparation
5.3.15 Current Protocols in Cytometry
Supplement 9
Flow Analysis and Sorting of Plant Chromosomes
B5 nutrient medium Solution A: 300 mg H3BO3 758 mg MnSO4⋅H2O 200 mg ZnSO4⋅7H2O Adjust volume to 1 liter with H2O Store up to 6 months at 4°C Solution B: 25 mg Na2MoO4⋅2H2O 2.5 mg CoCl2⋅6H2O 75 mg KI 2.5 mg CuSO4⋅5H2O Adjust volume to 100 ml with H2O Store up to 6 months at 4°C Solution C: 100 mg thiamine 10 mg pyridoxine 10 mg nicotinic acid Adjust volume to 100 ml with H2O Store up to 1 month at 4°C Solution D: 100 mg m-inositol Adjust volume to 100 ml with H2O Store up to 1 month at 4°C Solution E: 3.36 g Na2EDTA 2.79 g FeSO4 Adjust volume to ∼400 ml with H2O Heat to 70°C while stirring until the color turns yellow-brown Cool and then adjust volume to 500 ml Store up to 6 months at 4°C B5 stock solution, 10×: 30 g KNO3 5 g MgSO4⋅7H2O 1.34 g (NH4)2SO4 1.5 g CaCl2⋅2H2O 1.5 g NaH2PO4⋅H2O 100 ml solution A (above) 10 ml solution B (above) Adjust volume to 500 ml with H2O Store up to 6 months at 4°C B5 nutrient medium, 1×: 100 ml 10× B5 stock solution (above) 10 ml solution C (above) 10 ml solution D (above) 5 ml solution E (above) Adjust volume to 1 liter with H2O Adjust final pH to 5.5 using 1 N NaOH Sterilize by autoclaving Prepare just before use This recipe is from Gamborg et al. (1968).
5.3.16 Supplement 9
Current Protocols in Cytometry
4′,6-Diamidino-2-phenylindole (DAPI) stock solution, 0.1 mg/ml Dissolve 5 mg DAPI in 50 ml H2O by stirring for 60 min. Pass through a 0.22-µm filter to remove small particles. Store up to 1 year at −20°C in 0.5-ml aliquots. CAUTION: DAPI is a possible carcinogen. It may be harmful if inhaled, swallowed, or absorbed through the skin, and may also cause irritation. Use gloves when handling. Be careful of particulate dust when weighing out the dye. Consult local institutional safety officer for specific handling and disposal procedures.
Ethidium bromide solution, 0.5 ìg/ml Prepare a 0.5 mg/ml aqueous stock solution and stir for 60 min. Store up to 1 year at 4°C in the dark. For the working solution, dilute 1:1000 in H2O (final 0.5 µg/ml). The working solution may be used several times. CAUTION: Ethidium bromide is a powerful mutagen and is moderately toxic. It may be harmful if inhaled, swallowed, or absorbed through the skin. Use gloves when handling. Be careful of particulate dust when weighing out the dye. Consult local institutional safety officer for specific handling and disposal procedures.
Formaldehyde fixative 0.303 g Tris base (10 mM final) 0.931 g Na2EDTA (10 mM final) 1.461 g NaCl (100 mM final) 250 µl Triton X-100 (0.1% [v/v] final) Adjust volume to 200 ml with H2O, and adjust pH to 7.5 using 1 N NaOH. Add 37% (v/v) formaldehyde stock solution (e.g., Merck) as indicated in Table 5.3.2. Adjust final volume to 250 ml with H2O. Prepare just before use. CAUTION: Formaldehyde is toxic and is also a carcinogen. It is readily absorbed through the skin and is irritating or destructive to the skin, eyes, mucous membranes, and upper respiratory tract. Wear gloves and safety glasses. Always work in a chemical fume hood. Consult local institutional safety officer for specific handling and disposal procedures.
Fructose syrup 30 g fructose 20 ml H2O Incubate at 37°C overnight Add one crystal of thymol Store up to 1 year at 4°C
Table 5.3.2 Preparation of Formaldehyde Fixative
Formaldehyde Species
Vicia faba Pisum sativum Hordeum vulgare Secale cereale
Volume of stock (ml)
Final % concentration (v/v)
27 20 13.5 13.5
4 3 2 2
Specimen Handling, Storage, and Preparation
5.3.17 Current Protocols in Cytometry
Supplement 9
Hoagland’s nutrient solution, 1× and 0.1× Solution A: 280 mg H3BO3 340 mg MnSO4⋅H2O 10 mg CuSO4⋅5H2O 22 mg ZnSO4⋅7H2O 10 mg (NH4)6Mo7O24⋅4H2O Adjust volume to 100 ml with H2O Store up to 1 year at 4°C Solution B: 0.5 ml concentrated H2SO4 Adjust volume to 100 ml with H2O Store up to 1 year at 4°C Solution C: 3.36 g Na2EDTA 2.79 g FeSO4 Adjust volume to ∼400 ml with H2O Heat to 70°C while stirring until the color turns yellow-brown Cool and then adjust volume to 500 ml Store up to 1 year at 4°C Hoagland’s stock solution, 10×: 4.7 g Ca(NO3)2⋅4H2O 2.6 g MgSO4⋅7H2O 3.3 g KNO3 0.6 g (NH4)H2PO4 5 ml solution A (above) 0.5 ml solution B (above) Adjust volume to 500 ml with H2O Store up to 1 year at 4°C Hoagland’s nutrient solution, 1×: 100 ml 10× stock solution 5 ml solution C (above) Adjust volume to 1 liter with H2O Prepare just before use Hoagland’s nutrient solution, 0.1×: 10 ml 10× stock solution 0.5 ml solution C (above) Adjust volume to 1 liter with H2O Prepare just before use This recipe is from Gamborg and Wetter (1975).
Hydroxyurea treatment solution Prepare treatment solution immediately before use by combining hydroxyurea and 0.1× or 1× Hoagland’s nutrient solution (see recipe) as indicated in Table 5.3.3.
Flow Analysis and Sorting of Plant Chromosomes
LB01 lysis buffer 0.363 g Tris base (15 mM) 0.149 g Na2EDTA (2 mM) 0.348 g spermine⋅4HCl (0.5 mM) 1.193 g KCl (80 mM) 0.234 g NaCl (20 mm)
continued
5.3.18 Supplement 9
Current Protocols in Cytometry
Table 5.3.3 Preparation of Hydroxyurea Treatment Solution
Hoagland’s nutrient solution Species
Concentration Volume (ml)
Vicia faba Pisum sativum Hordeum vulgare Secale cereale
1× 1× 0.1× 1×
750 750 750 750
Hydroxyurea Amount (mg)
Final concentration (mM)
71.3 71.3 114.0 142.6
1.25 1.25 2.00 2.50
200 µl Triton X-100 (0.1%) Adjust volume to 200 ml with H2O, and final pH to 7.5 using 1 N HCl. Filter through a 0.22-µm filter to remove small particles. Add 220 µl 2-mercaptoethanol and mix well. Store up to 1 year at −20°C in 10-ml aliquots. This recipe is from DoleÓel et al. (1989). CAUTION: 2-Mercaptoethanol may be fatal if inhaled or absorbed through the skin and is harmful if swallowed. High concentrations are extremely destructive to the skin, eyes, mucous membranes, and upper respiratory tract. Wear gloves and safety glasses and work in a chemical fume hood.
Loading buffer 2 ml 0.5 M EDTA, pH 8.0 (APPENDIX 2A; 100 mM final) 0.1 g SDS (1% final) 5 mg bromphenol blue (0.05% final) 5 mg xylene cyanol (0.05% final) 5 ml glycerol (50% final) Adjust volume to 10 ml with H2O Store up to 1 year at room temperature Mithramycin stock solution, 1 mg/ml Dissolve 50 mg mithramycin A in 50 ml deionized H2O by stirring for 60 min. Filter through a 0.22-µm filter to remove small particles. Store up to 1 year at −20°C in 0.5-ml aliquots. CAUTION: Mithramycin is a possible carcinogen. It may be harmful if inhaled, swallowed, or absorbed through the skin. Use gloves when handling. Be careful of particulate dust when weighing out the dye. Consult the local institutional safety officer for specific handling and disposal procedures.
PCR premix 5 µl 10× Taq DNA polymerase buffer (1× final; buffer does not contain MgCl2) 3 µl 25 mM MgCl2 (1.5 mM final) 1 µl 10 mM 4dNTP mix (final 0.2 mM each dATP, dCTP, dGTP, dTTP) 1 µl 50 µM forward primer (1 µM final) 1 µl 50 µM reverse primer (1 µM final) 0.5 µl 5 U/µl Taq DNA polymerase (2.5 U/50 µl final) 18.5 µl sterile, deionized H2O Mix well and microcentrifuge briefly (5 to 10 sec at 2000 × g at room temperature) Prepare on ice shortly before use
Specimen Handling, Storage, and Preparation
5.3.19 Current Protocols in Cytometry
Supplement 9
PRINS buffer 0.605 g Tris base (10 mM final) 1.864 g KCl (50 mM final) 0.203 g MgCl2⋅6H2O (2 mM final) 2.5 ml 10% (v/v) Tween 20 Adjust volume to 500 ml with deionized H2O Adjust pH to 8.0 using 1 N HCl Sterilize by autoclaving Store up to 6 months at 4°C PRINS reaction mix 5 µl 10× Taq DNA polymerase buffer (1× final, containing 1.5 mM MgCl2) 5 µl 25 mM MgCl2 (4 mM final) 2.5 µl 2 mM dCTP/dGTP (0.1 mM each final) 2 µl 0.2 mM fluorescein-12-dUTP (8 µM final) 2 µl 0.2 mM fluorescein-15-dATP (8 µM final) 4.25 µl 0.2 mM dTTP (17 µM final) 4.25 µl 0.2 mM dATP (17 µM final) 5 µl 20 µM forward primer (2 µM final) 5 µl 20 µM reverse primer (2 µM final) 0.6 µl 5 U/µl Taq DNA polymerase (3 U/50 µl final) Adjust to 55 µl with sterile, deionized H2O (includes 5 µl for evaporation) Prepare shortly before use. Actual composition of the mix (e.g., MgCl2, concentration, ratio of labeled and unlabeled nucleotides, primer concentration) should be optimized for a given primer pair and species.
Schiff’s reagent 30 ml 1 N HCl 2 g parafuchsin (Serva; color index 42500) 3.8 g K2S2O5 170 ml H2O Stir for 2 hr in a tightly closed bottle and leave overnight. Add 2 g active charcoal, mix 1 min, and filter through a paper filter moistened with 1 N HCl. Repeat filtration if solution is not colorless. Store up to 6 months in a tightly closed bottle at 4°C. Stop buffer 2.923 g NaCl (0.5 M final) 1.861 g Na2EDTA (0.05 M final) Adjust volume to 100 ml with H2O Adjust pH to 8.0 using 1 N NaOH Sterilize by autoclaving Store up to 6 months at 4°C TAE electrophoresis buffer, 50× and 1× 242 g Tris base (2 M) 57.1 ml glacial acetic acid (1 M acetate final) 200 ml 0.5 M EDTA, pH 8.0 (APPENDIX 2A; 100 mM final) Adjust volume to 1 liter with H2O Store up to 1 year at room temperature Dilute 1:50 in H2O before use Final 1× concentrations are 40 mM Tris, 20 mM acetate, and 2 mM EDTA. Flow Analysis and Sorting of Plant Chromosomes
CAUTION: Glacial acetic acid is volatile. Concentrated acids must be handled with great care. Wear gloves and safety glasses and work in a chemical fume hood.
5.3.20 Supplement 9
Current Protocols in Cytometry
Tris buffer 0.606 g Tris base (10 mM final) 1.861 g Na2EDTA (10 mM final) 2.922 g NaCl (100 mM final) Adjust volume to 500 ml with H2O Adjust pH to 7.5 using 1 N NaOH Store up to 6 months at 4°C Wash buffer 1.161 g maleic acid (0.1 M final) 0.876 g NaCl (0.15 M final) 0.5 ml Tween 20 (0.05% final) Adjust volume to 100 ml with H2O Adjust pH to 7.5 using 1 N NaOH Sterilize by autoclaving Store up to 6 months at 4°C COMMENTARY Background Information A majority of agriculturally important plant species have a complex nuclear genome that makes gene mapping and isolation very difficult. This daunting task may be simplified by dividing the genome into well-defined parts, e.g., chromosomes. Generally, individual chromosomes can be isolated either by chromosome microdissection or by flow sorting. The advantage of flow cytometry is that chromosomes can be sorted in large numbers. Flow analysis and sorting of mitotic chromosomes (flow cytogenetics) greatly stimulated progress in human genome mapping. Chromosomes stained with DNA fluorochromes were individually classified according to their dye content, and the resulting distributions (flow karyotypes) were shown to be useful for detection of numerical and structural chromosome aberrations (Otto 1988; Cooke et al., 1989; Boschman et al., 1992). Flow-sorted chromosomes were used for gene mapping (Lebo, 1982), for construction of chromosomespecific DNA libraries (Van Dilla and Deaven, 1990), and for generation of chromosome painting probes (Carter, 1994). In addition, flow cytogenetics has been instrumental for analysis of genomes from several animal species (Dixon et al., 1992; Langford et al., 1996; Burkin et al., 1997). Attempts to develop flow cytogenetics in plants were hampered by difficulties in preparing suspensions of intact chromosomes and by the inability to discriminate single chromosome types (DoleÓel et al., 1994). These problems were overcome by the development of new procedures and experimental approaches,
and flow cytometric analysis and sorting has been reported for twelve plant species, including some agriculturally important legumes and cereals (DoleÓel et al., 1999). Considering these results, it may be expected that flow cytogenetics will play an increasingly important role in plant genomics. Materials used for preparation of plant chromosome suspensions The first plant chromosome suspension for flow cytometric analysis was prepared using suspension-cultured cells of Haplopappus gracilis (De Laat and Blaas, 1984). Subsequently, chromosomes were isolated from cultured cells of Lycopersicon esculentum, L. pennellii, and Triticum aestivum (Arumuganathan et al., 1991, 1994; Wang et al., 1992). Cell suspensions appear to be an attractive system for chromosome isolation mainly because large numbers of cells can be grown and manipulated under defined conditions. However, they have certain disadvantages, the most serious being their karyological instability. Some authors isolated chromosomes from protoplasts derived from leaf cells (Conia et al., 1987, 1988, 1989). In the young leaf, the majority of mesophyll cells are reversibly arrested at the G0/G1 phase of the cell cycle. Upon transfer to the appropriate culture medium, protoplasts derived from these cells resume the cell cycle, traverse the S and G2 phases, and enter mitosis with a certain degree of synchrony. Although relatively simple, this approach has not been used frequently, mainly because the extent of synchrony is relatively low.
Specimen Handling, Storage, and Preparation
5.3.21 Current Protocols in Cytometry
Supplement 9
The third system, developed by DoleÓel et al. (1992) and described in this unit, involves the use of root-tip meristems and has been by far the most frequently employed. Examples of successful applications include agriculturally important legumes (Lucretti et al., 1993; Gualberti et al., 1996) and cereals (Lee et al., 1996, 1997; Lysák et al., 1999). The advantage of roots is that they can be easily obtained from seeds for the majority of plant species, they are cheap to produce, and they are easy to grow and manipulate. A further advantage of root-tip meristems is their karyological stability. Although generally applicable, this experimental system may not be accessible for genotypes that are difficult to propagate via seeds. Here the use of immortal (hairy root) cultures produced after infection with the bacterium Agrobacterium rhizogenes represents an elegant solution (Veuskens et al., 1992, 1995; Neumann et al., 1998). However, due to their smaller size, more root tips must be used than is the case for seedlings.
Flow Analysis and Sorting of Plant Chromosomes
Cell cycle synchronization and accumulation of metaphase chromosomes While “spontaneous” cell cycle synchrony is observed in cultured mesophyll protoplasts, it must be artificially induced in suspensioncultured cells and root tips. Most frequently, this has been achieved by treatment with DNA synthesis inhibitors like hydroxyurea or aphidicolin. The treatment leads to the accumulation of cycling cells mainly at the G1/S interface. After a suitable period, cells are released from the block and traverse the cycle synchronously. Although the mechanisms of action of hydroxyurea and aphidicolin are different (Young and Hodas, 1964; Sala et al., 1980), the degree of induced cell-cycle synchrony is similar. However, hydroxyurea is considerably cheaper and thus has been used more frequently. In order to accumulate a sufficient number of synchronized cells in metaphase, chromosome movement in mitosis must be inhibited. This can be conveniently achieved using mitotic spindle poisons. One of them, the alkaloid colchicine, has been frequently employed in procedures for chromosome isolation (De Laat and Blaas, 1984; Conia et al., 1987; Wang et al., 1992; Arumuganathan et al., 1994). Although widely used, colchicine has serious disadvantages. Its affinity to plant tubulins is low, and thus it has been applied at relatively high (millimolar) concentrations. Another serious drawback of colchicine is its tendency to induce extensive chromosome stickiness and clump-
ing. In contrast to colchicine, synthetic herbicides like amiprophos-methyl, oryzalin, and trifluralin show visible antimicrotubular effects at micromolar concentrations, and thus have been used frequently in chromosome isolation protocols (DoleÓel et al., 1992; Veuskens et al., 1995; Lee et al., 1996). Preparation of chromosome suspensions Chromosome isolation is complicated by the rigid cell walls typical of plant cells. Two approaches have been used to release chromosomes from synchronized cells. The first involves removal of the cell wall by the action of hydrolytic enzymes such as pectinases and cellulases. Protoplasts thus obtained are lysed in a hypotonic buffer, where they burst and liberate chromosomes (De Laat and Blaas, 1984). The second approach, which is described in this unit, involves direct mechanical release of chromosomes from synchronized root meristems after formaldehyde fixation (DoleÓel et al., 1992). Compared to the protoplast lysis method, the latter has several advantages. Mechanical isolation is rapid and avoids extended enzyme incubations, during which chromosome decondensation may occur. Moreover, enzyme mixtures used to digest cell walls may be contaminated by DNases. Lee et al. (1996) introduced a modified version of DoleÓel’s procedure where the fixation step is omitted. However, the original method based on formaldehyde fixation has several advantages. Fixed chromosomes are resistant to mechanical shearing, and thus can be isolated using a mechanical homogenizer, which takes only a few seconds. Mechanical stability of isolated chromosomes permits two-step sorting to improve the purity of sorted fractions (Lucretti et al., 1993). The fixation step also significantly increases the yield of chromosomes. The morphology of isolated chromosomes is well preserved and their DNA is suitable for PCR (Macas et al., 1993). In addition, chromosomes prepared according to this protocol were shown to be suitable for scanning electron microscopy (Schubert et al., 1993), in situ hybridization (Fuchs et al., 1994), PRINS (Kubaláková et al., 1997), and immunolocalization of chromosomal proteins (Binarová et al., 1998). Flow karyotyping The majority of plant species have morphologically similar chromosomes with similar DNA content. Due to this, the number of chromosomes that can be discriminated and sorted in various species is usually low (DoleÓel et al.,
5.3.22 Supplement 9
Current Protocols in Cytometry
1994, 1999). Various strategies have been employed to overcome this problem. Bivariate flow karyotyping is a standard method of human flow cytogenetics that allows discrimination of almost all chromosome types (see UNIT 8.6). With rare exceptions (Arumuganathan et al., 1994), this approach does not help to discriminate higher numbers of chromosomes in plants when compared to univariate analysis (Lucretti and DoleÓel, 1997; Schwarzacher et al., 1997). The reason for this is most probably a high proportion of repetitive DNA sequences dispersed more or less evenly in plant genomes. Lucretti et al. (1993) suggested the use of chromosome translocation lines to aid in discrimination and sorting of individual chromosome types. Although developed for the field bean (DoleÓel and Lucretti, 1995), this approach has also been found useful for the garden pea (Neumannn et al., 1998) and barley (Lysák et al., 1999), thus confirming its overall usefulness. Furthermore, chromosomes sorted from translocation lines were found useful for localization of DNA sequences at the subchromosomal level (Macas et al., 1993). In species where translocation lines or other chromosomal stocks are not available, primed in situ DNA labeling en suspension (PRINSES) appears to be a solution. This method was used to discriminate chromosomes of similar size in V. faba (Macas et al., 1995; Pich et al., 1995). However, because the method is rather complicated, its use is warranted only in those cases where other approaches are not feasible. Chromosome sorting The problems associated with discrimination of single chromosome types underline the importance of procedures for identification of chromosomal content of individual peaks on a flow karyotype and for determining the purity of sorted chromosome fractions. While dot blotting (Arumuganathan et al., 1994) or PCR with primers for chromosome-specific markers (Lysák et al., 1999) may be sufficient for chromosome assignment, microscopic evaluation of a sorted fraction is needed to determine the frequency and the nature of contaminating particles. Usually, morphological analysis of sorted chromosomes is not sufficient, as the length and even the arm ratio may differ significantly among sorted chromosomes of the same type. An elegant approach is to localize DNA sequences that are chromosome specific or show chromosome-specific distribution. Thus, Lucretti et al. (1993) used fluorescence in situ
hybridization (FISH) with a probe for rDNA to evaluate the purity of a sorted chromosome fraction from field bean. Compared to FISH, PRINS has several advantages. It does not require labeled probes and is faster, thus permitting rapid evaluation of sorted chromosomes. Its usefulness to identify sorted chromosomes and evaluate their purity was demonstrated by Gualberti et al. (1996) and Lysák et al. (1999). Utility of sorted chromosome fractions Flow-sorted plant chromosomes were found invaluable in a number of studies, including the generation of chromosome-specific DNA libraries. Such libraries can significantly reduce the efforts required for isolation of DNA sequences including molecular markers from defined regions of the genome. Thus, Wang et al. (1992) constructed a chromosome-enriched DNA library from sorted Triticum aestivum chromosome 4A. The library showed 20-fold enrichment of clones as compared to a random genomic library. Arumuganathan et al. (1994) reported generation of a chromosome-2-specific DNA library in L. pennellii. Using degenerate oligonucleotide–primed PCR (DOPPCR), the library was constructed from only one thousand sorted chromosomes. Furthermore, the authors isolated eleven chromosome2-specific probes that detected restriction fragment length polymorphisms (RFLPs) and placed them on the genetic linkage map of tomato. Macas et al. (1996) constructed a set of short-insert chromosome-specific DNA libraries covering the whole field bean genome. This is the first case in plants where a whole genome is available in the form of chromosome-specific DNA libraries. Physical mapping of DNA sequences is another spectacular application of flow-sorted chromosomes. Macas et al. (1993) used PCR with sequence-specific primers to physically map five seed-specific protein-coding genes and pseudogenes in field bean. In this work, the authors sorted chromosomes from a set of translocation lines that enabled mapping at the subchromosomal level. This approach is generally attractive for localization of those genes that, due to a lack of allelic variants, cannot be mapped genetically. More recently, Lysák et al. (1999) demonstrated the utility of chromosomes sorted from barley translocation lines for physical mapping of microsatellite markers. In addition to the above mentioned examples, sorted plant chromosomes have the potential to be used for isolation of chromosome-specific cDNAs, for generation of chromosome-
Specimen Handling, Storage, and Preparation
5.3.23 Current Protocols in Cytometry
Supplement 9
specific painting probes, for construction of plant artificial chromosomes, and for chromosome-mediated gene transfer. Compared to chromosome microdissection (see UNIT 8.6), flow sorting offers the possibility of sorting large numbers of chromosomes in a relatively short time. If chromosomal DNA is amplified by PCR (e.g., DOP-PCR), sorted chromosome preparations allow a higher number of template chromosomes to be used in a smaller number of PCR cycles compared to chromosomes isolated by microdissection, which in turn reduces the probability of preferential amplification of specific sequences from total DNA. Furthermore, only flow sorting may provide enough plant chromosomes for direct cloning and construction of chromosome-specific YAC and BAC libraries.
Critical Parameters and Troubleshooting
Flow Analysis and Sorting of Plant Chromosomes
Cell cycle synchronization and metaphase accumulation The quality of a chromosome suspension depends not only on the procedure for chromosome isolation but also on the proportion of metaphase cells in the source tissue. Although the mitotic activity in meristem root tips may be relatively high, it is usually not sufficient for chromosome isolation. Thus, the protocols described here rely on artificial synchronization of the cell cycle based on the inhibition of cell cycle progression at the G1/S interface by the action of hydroxyurea. The effect of the treatment depends critically on the hydroxyurea concentration. Because cell cycle kinetics are sensitive to external factors including temperature, care should be taken to adjust the temperature of all solutions prior to use. Also, the length of all treatments should be monitored, and residual treatment solutions should be carefully washed away before transferring seedlings to other solutions. Seedlings with healthy and actively growing roots should always be used. To achieve reproducible results, seedlings with a similar root length should be selected. In the protocols presented in this unit, synchronized cells are arrested at metaphase by the action of a mitotic spindle poison, amiprophosmethyl. Although higher frequencies of metaphase cells may be obtained after longer treatments, a short (e.g., 2-hr) treatment decreases the proportion of single chromatids and avoids chromosome decondensation. Compared to other mitosis blockers, amiprophos-methyl has a high affinity for plant tubulins and can be used
in very low concentrations. If needed, other mitotic spindle inhibitors may be used instead. This will require determination of optimal concentrations. In some species, including cereals such as barley, wheat, and rye, overnight incubation of synchronized seedlings in ice water improves the spreading of chromosomes within cells. This, in turn, results in improved chromosome yield and lower frequency of chromosome clumps. It is important examine the frequency of metaphase cells in the root tips prior to chromosome isolation (see Support Protocol 1). If the frequency is <40%, or if the cells are not arrested at metaphase, check the procedure for metaphase accumulation. First, analyze the degree of mitotic synchrony induced by the hydroxyurea treatment. Take samples of root tips at 1-hr intervals after removal from the hydroxyurea block and determine the mitotic activity on Feulgen-stained preparations. The mitotic activity should gradually increase and peak at ∼50%. If this is not the case, check the quality of the plant material and of all reagents and solutions. Based on the authors’ experience, treatment with a mitotic spindle inhibitor should be started 30 to 90 min before the peak of mitotic activity. If the synchronized cells are not arrested at metaphase, test the mitosis blocker using nonsynchronized cells (root tips). Try to determine the optimal concentration that will arrest cells at metaphase (no anaphase or telophase cells should be observed) and that will not induce chromosome clumping. The above-described troubleshooting strategy may be also used to modify the procedure for other species not mentioned in this unit. Preparation of chromosome suspensions Start with the formaldehyde fixation of root tips immediately after finishing the incubation in the mitosis blocker. The extent of formaldehyde fixation is a critical determinant of chromosome morphology and yield. Thus, the concentration of formaldehyde, the temperature of the fixative, and the length of fixation must be strictly followed. Also note that the formaldehyde fixative used in this protocol is not neutralized. Check the quality of the chromosome suspension under a fluorescence microscope. Weak fixation results in poor preservation of chromosome morphology. Most of the isolated chromosomes are damaged (extended fibers) and the suspensions contain a large amount of chromosome debris. However, if the fixation is too strong, the homogenized suspensions will
5.3.24 Supplement 9
Current Protocols in Cytometry
contain large numbers of chromosome clumps and even intact cells. When starting with the chromosome preparation protocol, it is recommended to test several variants of fixation. For instance, change the length of the fixation in 5-min steps, or change the concentration of formaldehyde in 1% steps, and observe the effect under the microscope. Always compare chromosome suspensions prepared on the same day from the same batch of synchronized root tips. Select the optimum fixation giving the highest yield of morphologically intact chromosomes. However, the final decision on the optimum extent of fixation should be based on flow-cytometric analysis of isolated chromosomes. The fluorescence distribution should contain well-resolved chromosome peaks with the least amount of background debris, chromatids, and chromosome clumps. Storage of fixed root tips in Tris buffer prior to chromosome isolation may decrease chromosome yield, so it is advisable to perform the isolation immediately after fixation. Although suspensions of isolated chromosomes may be stored at 4°C, the best resolution of flow karyotypes is usually obtained with freshly prepared samples. Flow karyotyping The resolution of flow karyotypes depends critically on the performance of the cytometer. Always make sure that the instrument is properly aligned prior to starting chromosome analysis. This is usually done using suitable fluorescent microspheres to achieve the lowest coefficient of variation (CV) for fluorescent peaks. Naturally, the lowest CV that can be achieved depends also on the quality of the calibration beads. Use the best particles available with CV ≤1.5%. If the resolution of fluorescent peaks remains poor, check the fluidic system and the nozzle. All tubing must be clean and free of air bubbles. A dirty nozzle is probably the most frequent reason for poor resolution. When in doubt, always clean the nozzle. This can be done in a small sonicating water bath, but note that only the nozzle itself may be sonicated, as the holder may disintegrate. A convenient way to check the state of the nozzle and fluidics is by analysis of forward light scatter. The resulting distribution should be tight and have a CV similar to that observed in fluorescence channels. Compared to human or animal chromosome suspensions, plant chromosome suspensions are usually less concentrated and, therefore, should be run at lower speeds (e.g., 200 to 500
chromosomes/sec) to achieve the best resolution. The peak position can show instability due to slow equilibration of fluorescent dyes in the sample line. To minimize this problem, always run a dummy sample prior to analysis of the suspension. Even then, it may take a few minutes before the chromosome peak positions are stable and data may be acquired. Because isolated plant chromosomes vary greatly in the degree of condensation and, hence, also in length, better resolution is obtained on histograms of fluorescence pulse area rather than fluorescence pulse height. The procedures for preparation of plant chromosome suspensions described here are based on formaldehyde fixation of root tips. The fixation results in lower resolution of flow karyotypes after staining with DNA intercalators such as ethidium bromide or propidium iodide. Always use DAPI or similar dyes (e.g., Hoechst 33258) to stain chromosomes isolated according to this protocol for high-resolution univariate flow karyotyping. Chromosome sorting Sorting of pure chromosome fractions requires that the sorter be well aligned. In addition to the resolution of fluorescence and forward scatter (see above), this means adjustment of the sort module. Always check its performance prior to actual chromosome sorting. Adjust the drop drive phase to obtain single side streams without fanning, and make sure that the sorted stream is positioned so that the deflected drops arrive at the collecting tube or on a slide. Check the adjustment of the drop delay so that the number of sorted particles indicated by the machine is equal to that actually sorted. Carefully follow the stability of the break-off point and immediately stop the sorting if a change is observed. If this happens, the sort module must be readjusted. Note that a common reason for break-off point instability is a dirty nozzle. Although the purity of sorted chromosome fractions is primarily determined by the resolution of flow karyotypes, it is also affected by the presence of debris particles, chromatids, and chromosome clumps in the chromosome suspension. For instance, chromatids and fragments of larger chromosomes may be indiscernible from small chromosomes. It may be difficult to discriminate doublets of short chromosomes or other chromosome aggregates from large chromosomes. Although the analysis of fluorescence pulse width may be helpful (see Basic Protocol 3, Alternate Protocol 3, and
Specimen Handling, Storage, and Preparation
5.3.25 Current Protocols in Cytometry
Supplement 9
Alternate Protocol 4), it is preferable to minimize the frequency of these particles. Always check the purity of sorted chromosome fractions. Because the degree of condensation varies greatly among sorted chromosomes, simple fluorescence staining is not sufficient to determine the purity of a sorted chromosome fraction. This is best done after fluorescence labeling of sorted chromosomes using either chromosome-specific DNA sequences or sequences that show a chromosomespecific labeling pattern (see Support Protocol 3). If the purity of a sorted fraction is not satisfactory, use two-step sorting (see Alternate Protocol 4). During the first sort, many aggregates split and, at the same time, the suspension is enriched for the required chromosome, which can then be sorted at high purity during the second sort. When sorting, the instrument should be triggered on the fluorescence pulse height signal and the threshold set to a minimum so that all DNA-containing particles are detected. This is important for recognition of particles that should not be sorted, and for avoiding contamination of the sorted fraction. Even if the sample line is equilibrated with dye, a slight change in chromosome peak position may be observed, especially during long-term sorting. Monitor peak positions during sorting and make adjustments to the sort window as needed.
Flow Analysis and Sorting of Plant Chromosomes
Determining the purity of sorted chromosome fractions using PRINS Compared to FISH, PRINS is much faster and thus more useful for analyzing the purity of sorted fractions. As with FISH, discrimination of individual chromosome types is possible only when primers are available for repetitive sequences that can be visualized and that show a characteristic pattern of chromosome distribution. PRINS is sensitive to chromatin structure, and several parameters—including the age of preparations, reaction mix composition, and conditions for PRINS reaction—should be optimized when the procedure is modified for other species (Kubaláková et al., 1997; Kubaláková and DoleÓel, 1998). Reaction concentrations must be consistent during thermal cycling to ensure proper amplification. It is thus critical that the reaction chamber be tightly sealed to prevent vapor leakage. Reaction chambers of different volumes may be conveniently created using frames that are stuck to the slides and tightly closed by a polyester coverslip. These are available from MJ Research or Hybaid. It is critical that heat
be transferred to and from the slide rapidly and uniformly, and that heat loss from the top surface of the slide due to radiation and convection be minimized. Although it is possible to use aluminum foil to adapt a standard block-type thermal cycler to hold the slides, it is advised to use either a flat plate equipped with a temperature sensor (e.g., Slide Griddle from MJ Research) or, preferably, a thermal cycler dedicated to microscope slides. To improve heat transfer, the bottom of the slides may be covered with mineral oil. The resulting layer must be thin and free of bubbles. Physical mapping of DNA sequences PCR with primers derived from chromosome-specific DNA sequences is a rapid method for physical mapping of DNA sequences to individual chromosomes or their parts. However, reliable results may be obtained only under optimum conditions. Reagent purity is critical as well as avoiding contamination. It is important that the conditions of the PCR reaction (e.g., MgCl2 concentration, annealing temperature) be optimized for a given primer pair. This is most conveniently done using genomic DNA as a template. Negative and positive controls should always be included to check whether the PCR ran properly. A convenient positive control may be obtained after sorting all different chromosome types from a given species. This ensures that a target DNA sequence (whose chromosome localization is not known) will be present in the PCR tube and will be amplified. A chromosome fraction where the sequence is known to be absent provides a convenient negative control. The use of PCR for physical mapping is possible only if the purity of the sorted fraction approaches 100%; otherwise, false positive signals will be obtained. Even with sorted fractions of high purity, care should be taken not to run too many PCR cycles, to avoid amplification of the target sequence from minor contaminants. Freezing and thawing of sorted chromosomes before PCR is essential. Most probably, this step increases the accessibility of the template for DNA polymerase.
Anticipated Results Preparation of chromosome suspensions Depending on the species, the procedure for metaphase accumulation should result in metaphase frequencies ranging between 40% and 60%. This level of cell cycle synchrony is essential for preparation of good chromosome
5.3.26 Supplement 9
Current Protocols in Cytometry
suspensions. Besides the degree of cell synchrony, the chromosome yield depends on the strength of formaldehyde fixation and the procedure for chromosome release itself. Under optimal conditions, >106 chromosomes can be isolated from 30 root tips of field bean and garden pea. Approximately 5 × 105 chromosomes can be isolated after homogenization of 50 root tips of rye and barley. In the case of hairy roots, on average five times more root tips are needed to achieve the same chromosome concentration. Flow karyotyping Figure 5.3.1 presents typical univariate flow karyotypes obtained after the analysis of DAPIstained chromosomes isolated from garden pea. Root-tip meristems were synchronized according to Basic Protocol 1, isolated according to Basic Protocol 2, and sorted according to Basic Protocol 3. Panel A shows a flow karyotype obtained in a line with a standard karyotype. Note that only chromosome 5 can be fully resolved. Chromosome 7 is resolved only partially and the remaining chromosomes form a composite peak. Panel B demonstrates the flow karyotype of the chromosome translocation line JI 148. In addition to a wild-type chromosome 5, three additional chromosomes (1, 2, and 7) can be resolved. Note that the flow karyotypes also contain minor peaks repre-
senting chromosomal debris, chromatids, and chromosome clumps. The results of bivariate flow karyotyping in field bean are demonstrated in Figure 5.3.2. Chromosome suspensions were prepared according to Basic Protocols 1 and 2, and then stained simultaneously with DAPI and mithramycin and analyzed according to Alternate Protocol 3. Two translocation lines were analyzed: BKH (panel A) and EF (panel B). Note that in both translocation lines, the chromosome peaks lie on straight diagonal lines, indicating negligible differences in the AT/GC ratio among the chromosomes. Chromosome sorting Figure 5.3.3 demonstrates definition of a gating region (R3) to sort the largest chromosome (1) in field bean. The gate is placed on a dot plot of fluorescence pulse area versus fluorescence pulse width. The analysis of pulse width is used here to discriminate doublets of short acrocentric chromosomes (2 to 6, region R2) that have similar relative DNA content as the large metacentric chromosome 1. Region R1 represents single acrocentric chromosomes (2 to 6). Two-step chromosome sorting is demonstrated in Figure 5.3.4. Root-tip meristems of barley translocation line T2-6y were synchronized according to Alternate Protocol 1. Chromosomes were isolated according to Basic Pro-
A
B
Number of events
125 100
3, 4, 6
3, 4, 6
2 75 1
1, 2 7
50
5
5 7
25 0 0
200
400
600
800
1000
0
200
400
600
800
1000
Relative fluorescence intensity (channel number)
Figure 5.3.1 Univariate flow karyotypes obtained after analysis of DAPI-stained chromosomes isolated from two lines of garden pea (Pisum sativum L., 2n = 14). (A) Cultivar “Ctirad” with a standard karyotype. Only chromosome 5 can be clearly discriminated. (B) Chromosome translocation line JI 148. In addition to wild-type chromosome 5, three additional chromosomes (1, 2, and 7) can be resolved in the translocation line.
Specimen Handling, Storage, and Preparation
5.3.27 Current Protocols in Cytometry
Supplement 9
A
B
DAPI fluorescence
1000 1
1
800 3
600
4 2
400 4
5
2, 5, 6
200
6 3
0 0
200
400
600
800
1000
0
200
400
600
800
1000
Mithramycin fluorescence
Figure 5.3.2 Bivariate flow karyotypes of two field bean chromosome translocation lines (Vicia faba L., 2n = 12): (A) BKH and (B) EF. The chromosome peaks lie on straight diagonal lines, indicating negligible differences in the AT/GC ratio among the chromosomes in both lines. Data shown as relative fluorescence intensity (channel number).
A
B Fluorescence pulse width
1000
1
2
3
4
5
6
800 600
R3
400 R1
R2
200 0 0
200
400
600
800
1000
Fluorescence pulse area
Figure 5.3.3 Idiogram (A) and dot plot of fluorescence pulse area versus fluorescence pulse width (B) obtained after the analysis of chromosomes isolated from a field bean line with a standard karyotype (Vicia faba L., 2n = 12). This type of analysis permitted discrimination of the largest chromosome (1, region R3) from the doublets of short acrocentric chromosomes (2 to 6, region R2), which have similar relative DNA content. Region R1 represents single acrocentric chromosomes (2 to 6).
Flow Analysis and Sorting of Plant Chromosomes
tocol 2, and were sorted according to Alternate Protocol 4. Panel A shows a flow karyotype obtained during the first sort run when the chromosome 26 was sorted. Panel B shows a histogram of relative fluorescence intensity ob-
tained after the analysis of the sorted fraction. Note that the fraction contains almost exclusively chromosome 26. From this fraction, the chromosome can be sorted at high speed and with high purity.
5.3.28 Supplement 9
Current Protocols in Cytometry
A
B
Number of events
125 26
1, 3, 4, 7
100 75 62
50
5 26
25 0 0
200
400
600
800
1000
0
200
400
600
800
1000
Relative fluorescence intensity (channel number)
Figure 5.3.4 Two-step sorting in barley translocation line T2-6y (Hordeum vulgare L., 2n = 14). (A) Flow karyotype obtained during the first sort run when chromosome 26 was sorted. (B) Histogram of relative fluorescence intensity obtained after analysis of the sorted fraction. Note that the fraction contains almost exclusively chromosome 26.
Determining the purity of sorted chromosome fractions using PRINS Examples of six chromosome types sorted from the field bean translocation line EF are shown in Figure 5.3.5. Sorted chromosomes were dried on a microscope slide and subjected to a PRINS reaction with primers specific for a repetitive sequence (FokI). The labeling pat-
tern thus achieved may be used for unambiguous discrimination of individual chromosome types and to determine the presence of contaminating chromosomes. Images of sorted chromosomes were captured by a charge-coupled device (CCD) camera and assembled using Adobe Photoshop software.
1
4
2
5
3
6
Figure 5.3.5 Six chromosome types sorted from the field bean translocation line EF (Vicia faba L., 2n = 12). Sorted chromosomes were subjected to a PRINS reaction with primers specific for a repetitive sequence (FokI). The labeling pattern thus obtained was used for unambiguous discrimination of individual chromosome types. Note the significant variation in length among chromosomes of the same type.
Specimen Handling, Storage, and Preparation
5.3.29 Current Protocols in Cytometry
Supplement 9
A
B M
26
62
C
26
62
C
M
bp
bp
1000
1000
500
500
MWG557
MWG878
MWG557
C
62
MWG878
6 2
26
Figure 5.3.6 An example of physical mapping of DNA sequences to specific chromosome regions using PCR on flow-sorted chromosomes. Agarose gel electrophoresis was performed with products of PCR using primers specific for two RFLP markers, MWG557 (A) and MWG878 (B), previously mapped to barley chromosome 2. PCR was performed on 500 chromosomes 26 (lane 26) and 500 chromosomes 62 (lane 62) sorted from translocation line T2-6y. Additionally, 3500 chromosomes (∼500 chromosomes of each type) were used as a positive control (lane C). (C) Scheme of wild-type chromosomes 2 and 6 and their translocation products 26 and 62. Black bands indicate the positions of Giemsa N bands; arrowheads point to breakpoint positions.
Flow Analysis and Sorting of Plant Chromosomes
5.3.30 Supplement 9
Current Protocols in Cytometry
Physical mapping of DNA sequences PCR on sorted chromosomes enables chromosomal localization of known DNA sequences. Provided chromosome translocation lines are available, subchromosomal localization is possible. Figure 5.3.6 gives an example of subchromosomal localization of two RFLP markers in barley using translocation line T26y (Marthe and Künzel, 1994). Presence of PCR products indicates the position of the markers with respect to the translocation breakpoint.
Time Considerations The whole procedure, from the preparation of chromosome suspensions to their analysis and chromosome sorting, may take up to one week. Thus, the experiment needs careful planning. Although the protocols described here are not complicated, they require expertise in several different areas. Preparation of chromosome suspensions The quality of chromosome suspensions is critical for successful chromosome sorting, and it may take several weeks to master the preparation method. The chromosome yield depends critically on the frequency of metaphase cells in the root tip. In some cases, it may be necessary to optimize the protocol for cell-cycle synchronization and metaphase accumulation. Generally, the preparation of plant material from seed germination to accumulation of root tips in metaphase takes 5 to 6 days. It is convenient to germinate the seeds over a weekend. For hairy root cultures, the time needed to prepare the material depends on the growth phase of the culture. Provided the culture is rapidly growing, the procedure for metaphase accumulation takes 2 days. The processing of root tips for chromosome isolation is not time consuming and takes ∼1 hr including formaldehyde fixation. However, it must be performed immediately after metaphase accumulation. Flow analysis and sorting The operation of the flow cytometer for flow karyotyping and chromosome sorting requires a trained specialist. The analysis must be performed at highest possible resolution. This often requires time-consuming alignment of the instrument optics and fluidics. If the flow cytometer is well aligned and its fluidics are stable, chromosome sorting can be routinely performed at 5 to 20 chromosomes/sec. Naturally, the actual sort rate depends on the quality of the suspension, the frequency of the chro-
mosome in the suspension, and the sort mode and envelope. Nevertheless, sorting of small amounts of chromosomes for PCR or microscopic evaluation is convenient as the sorting itself takes only several minutes. On the other hand, sorting of large numbers of chromosomes needed for direct cloning requires many hours. Here problems with flow stability, including occasional clogging of the nozzle, may result in additional delays. Estimating the purity of sorted chromosomes Contamination of the sorted fraction by other chromosomes can be conveniently detected using PRINS. The reaction is fast and takes only 1 to 3 hr. However, before the reaction, the sorted chromosomes must be air dried on a slide. Although this may take only ∼1 hr, it may be convenient to dry the slides overnight. In this case, the results of the analysis are available the following day after sorting. Microscopic evaluation of one slide (analyzing ∼100 chromosomes) takes ∼30 min. Physical mapping of DNA sequences Provided that primers for the sequence to be mapped are available and reaction conditions for the given primer pair have been optimized, the protocol can be completed in a single day. The PCR itself takes ∼4 hr, and electrophoresis and staining take another 3 to 5 hr.
Literature Cited Arumuganathan, K., Slattery, J.P., Tanksley, S.D., and Earle, E.D. 1991. Preparation and flow cytometric analysis of metaphase chromosomes of tomato. Theor. Appl. Genet. 82:101-111. Arumuganathan, K., Martin, G.B., Telenius, H., Tanksley, S.D., and Earle, E.D. 1994. Chromosome 2-specific DNA clones from flow-sorted chromosomes of tomato. Mol. Gen. Genet. 242:551-558. Binarová, P., Hause, B., DoleÓel, J., and Dráber, P. 1998. Association of γ-tubulin with kinetochore/centromeric region of plant chromosomes. Plant J. 14:751-757. Boschman, G.A., Manders, E.M.M., Rens, W., Slater, R., and Aten, J.A. 1992. Semi-automated detection of aberrant chromosomes in bivariate flow karyotypes. Cytometry 13:469-477. Burkin, D.J., O’Brien, P.C.M., Broad, T.E., Hill, D.F., Jones, C.A., Wienberg, J., and FergusonSmith, M.A. 1997. Isolation of chromosomespecific paints from high-resolution flow karyotypes of the sheep (Ovis aries). Chromosome Res. 5:102-108. Carter, N.P. 1994. Cytogenetic analysis by chromosome painting. Cytometry 18:2-10. Conia, J., Bergounioux, C., Perennes, C., Muller, P., Brown, S., and Gadal, P. 1987. Flow cytometric
Specimen Handling, Storage, and Preparation
5.3.31 Current Protocols in Cytometry
Supplement 9
analysis and sorting of plant chromosomes from Petunia hybrida protoplasts. Cytometry 8:500508. Conia, J., Bergounioux, C., Brown, S., Perennes, C., and Gadal, P. 1988. Caryotype en flux biparametrique de Petunia hybrida. Tri du chromosome numero I. C.R. Acad. Sci. (Paris) 307:609615.
Flow Analysis and Sorting of Plant Chromosomes
Conia, J., Muller, P., Brown, S., Bergounioux, C., and Gadal, P. 1989. Monoparametric models of flow cytometric karyotypes with spreadsheet software. Theor. Appl. Genet. 77:295-303. Cooke, A., Tolmie, J.L., Colgan, J.M., Greig, C.M., and Connor, J.M. 1989. Detection of an unbalanced translocation (4;14) in a mildly retarded father and son by flow cytometry. Hum. Genet. 83:83-87. De Laat, A.M.M. and Blaas, J. 1984. Flow-cytometric characterization and sorting of plant chromosomes. Theor. Appl. Genet. 67:463-467. Dixon, S.C., Miller, N.G.A., Carter, N.P., and Tucker, E.M. 1992. Bivariate flow cytometry of farm animal chromosomes: A potential tool for gene mapping. Anim. Genet. 23:203-210. DoleÓel, J. 1991. KARYOSTAR: Microcomputer program for modeling of monoparametric flow karyotypes. Biológia 46:1059-1064. DoleÓel, J. and Lucretti, S. 1995. High-resolution flow karyotyping and chromosome sorting in Vicia faba lines with standard and reconstructed karyotypes. Theor. Appl. Genet. 90:797-802. DoleÓel, J., Binarová, P., and Lucretti, S. 1989. Analysis of nuclear DNA content in plant cells by flow cytometry. Biol. Plant. 31:113-120. DoleÓel, J., Cíhalíková, J., and Lucretti, S. 1992. A high-yield procedure for isolation of metaphase chromosomes from root tips of Vicia faba L. Planta 188:93-98. DoleÓel, J., Lucretti, S., and Schubert, I. 1994. Plant chromosome analysis and sorting by flow cytometry. Crit. Rev. Plant Sci. 13:275-309. DoleÓel, J., Lysák, M.A., Kubaláková, M., ´imková, H., Macas, J., and Lucretti, S. 1999. Sorting of plant chromosomes. In Flow Cytometry, 3rd ed. (Z. Darzynkiewicz, H.A. Crissman, and J.P. Robinson, eds.) Academic Press, New York. In press. Fuchs, J., Joos, S., Lichter, P., and Schubert, I. 1994. Localization of vicilin genes on field bean chromosome II by fluorescent in situ hybridization. J. Hered. 85:487-488. Gamborg, O.L. and Wetter, L.R. 1975. Plant Tissue Culture Methods. Natl. Res. Council of Canada, Saskatoon, Saskatchewan, Canada. Gamborg, O.L., Miller, R.A., and Ojima, K. 1968. Nutrient requirement of suspension cultures of soybean root cells. Exp. Cell Res. 50:151-158. Gualberti, G., DoleÓel, J., Macas, J., and Lucretti, S. 1996. Preparation of pea (Pisum sativum L.) chromosome and nucleus suspensions from single root tips. Theor. Appl. Genet. 92:744-751. Kramer, M.F. and Coen, D.M. 1999. Enzymatic amplification of DNA by PCR: Standard procedures and optimization. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 15.1.1-15.1.15. John Wiley & Sons, New York.
Kubaláková, M. and DoleÓel, J. 1998. Optimization of PRINS and C-PRINS for detection of telomeric sequences in Vicia faba. Biol. Plant. 41:177-184. Kubaláková, M., Macas, J., and DoleÓel, J. 1997. Mapping of repeated DNA sequences in plant chromosomes by PRINS and C-PRINS. Theor. Appl. Genet. 94:758-763. Langford, C.F, Fischer, P.E., Binns, M.M., Holmes, N.G., and Carter, N.P. 1996. Chromosome-specific paints from a high-resolution flow karyotype of the dog. Chrom. Res. 4:115-123. Lebo, R.V. 1982. Chromosome sorting and DNA sequence localization: A review. Cytometry 3:145-154. Lee, J.-H., Arumuganathan, K., Kaeppler, S.M., Kaeppler, H.F., and Papa, C.M. 1996. Cell synchronization and isolation of metaphase chromosomes from maize (Zea mays L.) root tips for flow cytometric analysis and sorting. Genome 39:697-703. Lee, J.-H., Arumuganathan, K., Yen, Y., Kaeppler, S. Kaeppler, H., and Baezinger, P.S. 1997. Root tip cell cycle synchronization and metaphase-chromosome isolation suitable for flow sorting in common wheat (Triticum aestivum L.). Genome 40:633-638. Lucretti, S. and DoleÓel, J. 1997. Bivariate flow karyotyping in broad bean (Vicia faba). Cytometry 28:236-242. Lucretti, S., DoleÓel, J., Schubert, I., and Fuchs, J. 1993. Flow karyotyping and sorting of Vicia faba chromosomes. Theor. Appl. Genet. 85:665-672. Lysák, M.A., Cíhalíková, J., Kubaláková, M., ´imková, H., Künzel, G., and DoleÓel, J. 1999. Flow karyotyping and sorting of mitotic chromosomes of barley (Hordeum vulgare L.). Chrom. Res. In press. Macas, J., DoleÓel, J., Lucretti, S., Pich, U., Meister, A., Fuchs, J., and Schubert, I. 1993. Localization of seed storage protein genes on flow-sorted field bean chromosomes. Chrom. Res. 1:107-115. Macas, J., DoleÓel, J., Gualberti, G., Pich, U., Schubert, I., and Lucretti, S. 1995. Primer-induced labeling of pea and field bean chromosomes in situ and in suspension. BioTechniques 19:402-408. Macas, J., Gualberti, G., Nouzová, M., Samec, P., Lucretti, S., and DoleÓel, J. 1996. Construction of chromosome-specific DNA libraries covering the whole genome of field bean (Vicia faba L.). Chrom. Res. 4:531-539. Marthe, F. and Künzel, G. 1994. Localization of translocation breakpoints in somatic metaphase chromosomes of barley. Theor. Appl. Genet. 89:240-248. Neumann, P., Lysák, M., DoleÓel, J., and Macas, J. 1998. Isolation of chromosomes from Pisum sativum L. hairy root cultures and their analysis by flow cytometry. Plant Sci. 137:205-215. Otto, F.J. 1988. Assessment of persisting chromosome aberrations by flow karyotyping of cloned Chinese hamster cells. Z. Naturforsch. 43c:948954. Pich, U., Meister, A., Macas, J., DoleÓel, J., Lucretti, S., and Schubert, I. 1995. Primed in situ labelling facilitates flow sorting of similar sized chromosomes. Plant J. 7:1039-1044.
5.3.32 Supplement 9
Current Protocols in Cytometry
Sala, F., Parisi, B., Burroni, D., Amileni, A.R., Pedrali-Noy, G., and Spadari, S. 1980. Specific and reversible inhibition by aphidicolin of the α-like DNA polymerase of plant cells. FEBS Lett. 117:93-98. Schubert, I., DoleÓel, J., Houben, A., Schertan, H., and Wanner, G. 1993. Refined examination of plant metaphase chromosome structure at different levels made feasible by new isolation methods. Chromosoma 102:96-101.
Key References
Schwarzacher, T., Wang, M.L., Leitch, A.R., Miller, N., Moore, G., and Heslop-Harrison, J.S. 1997. Flow cytometric analysis of the chromosomes and stability of a wheat cell-culture line. Theor. Appl. Genet. 94:91-97. Tempé, J. and Casse-Delbart, F. 1989. Plant gene vectors and genetic transformation: Agrobacterium Ri plasmids. In Cell Culture and Somatic Cell Genetics of Plants, Vol. 6 (J. Schell, ed.) pp. 25-49. Academic Press, New York. Van Dilla, M.A. and Deaven, L.L. 1990. Construction of gene libraries for each human chromosome. Cytometry 11:208-218. Veuskens, J., Marie, D., Hinnisdaels, S., and Brown, S.C. 1992. Flow cytometry and sorting of plant chromosomes. In Flow Cytometry and Cell Sorting (A. Radbruch, ed.) pp. 177-188. SpringerVerlag, Berlin-Heidelberg. Veuskens, J., Marie, D., Brown, S.C., Jacobs, M., and Negrutiu, I. 1995. Flow sorting of the Y sex-chromosome in the dioecious plant Melandrium album. Cytometry 21:363-373. Wang, M.L., Leitch, A.R., Schwarzacher, T., Heslop-Harrison, J.S., and Moore, G. 1992. Construction of a chromosome-enriched HpaII library from flow-sorted wheat chromosomes. Nucl. Acids Res. 20:1897-1901. Young, C.W. and Hodas, S. 1964. Hydroxyurea: Inhibitory effect on DNA metabolism. Science 146:1172.
Lucretti et al., 1993. See above. The first paper on flow analysis and sorting in plant chromosome translocation lines.
DoleÓel et al., 1992. See above. The original paper describing the procedure for preparation of suspensions of intact chromosomes from synchronized root tips after formaldehyde fixation. DoleÓel et al., 1994, 1999. See above. Review papers on plant flow cytogenetics.
Macas et al., 1993. See above. Plant gene mapping using flow-sorted chromosomes and PCR with specific primer pairs.
Internet Resources http://www.ueb.cas.cz/olomouc1 Web site that contains useful information and protocols on plant flow cytogenetics.
Contributed by Jaroslav DoleÓel Institute of Experimental Botany Olomouc, Czech Republic Ji¯í Macas Institute of Plant Molecular Biology Ceské Budejovice, Czech Republic Sergio Lucretti ENEA, Casaccia Research Center Rome, Italy
The authors wish to acknowledge the support of research grant no. 521/96/K117 from the Grant Agency of the Czech Republic and of grant no. 6046 from the National Agency for Agricultural Research, Czech Republic.
Specimen Handling, Storage, and Preparation
5.3.33 Current Protocols in Cytometry
Supplement 9
Quality Control in Phenotypic Analysis by Flow Cytometry INTRODUCTION As with any technology transfer from the research to the clinical laboratory, development and implementation of quality control and assurance procedures for flow cytometric analysis are essential for the successful application of this technology in the “routine” clinical laboratory. Quality control practices are performed in clinical laboratories to determine the precision and accuracy of procedures. The ultimate goal of quality control practices in clinical laboratories is to determine and report correct results for patient care. During the last few years, numerous papers have appeared in the literature discussing various aspects of flow cytometry quality control and standardization (Duckworth, 1987; Bachner, 1996; also see Table 3.2.1). Yet, currently there are no formal guidelines for flow cytometric operations in clinical laboratories, and determining quality control procedures remains the responsibility of individual laboratory directors. However, in 1987, the National Committee for Clinical Laboratory Standards (NCCLS) established a Subcommittee on Flow Cytometry to develop quality assurance procedures for clinical laboratory flow cytometry. The primary objective of the subcommittee was to develop a set of guidelines to allow clinical laboratories using different commercially available instruments to obtain comparable results on commonly analyzed types of samples. The first task before the subcommittee was to formulate a set of guidelines for peripheral blood lymphocyte subset immunophenotyping, one of the most frequently performed flow cytometric assays in clinical laboratories. An additional set of guidelines is being formulated under NCCLS auspices for immunophenotyping of leukemia/lymphoma. Finally, an entire quality control section has been written as part of the Leukemia/Lymphoma Consensus Conference and is the most aggressive and scientifically sound effort to date. This work was sponsored by the National Institutes of Health (NIH), the Clinical Cytometry Society (CCS), and the International Society of Clinical Cytometry; it is due to be published in 1997. There are a number of ways of considering quality control. In a strictly clinical interpretation, quality control is used to determine the Contributed by Anne A. Hurley Current Protocols in Cytometry (1997) 6.1.1-6.1.4 Copyright © 1997 by John Wiley & Sons, Inc.
precision and accuracy of a test or procedure. Precision and accuracy are two very different parameters: precision is the reproducibility of the same value; accuracy is how close that value is to the “true” value. Quality control for precision has not been difficult to address in flow cytometry. A number of methods currently exist to assess instrument precision, or reproducibility. But the issue of accuracy remains—i.e., a value can be extremely precise without approaching the true value. However, flow cytometry in the clinical setting is at a disadvantage in the area of accuracy. There are no set and true standards as there are in clinical chemistry, where, for example, the correct value of a glucose or potassium can be quantitatively determined. Even in other long-standing medical laboratory disciplines—e.g., clinical hematology—there is difficulty with controls and standards for any test other than hemoglobin determination. Therefore, given the strictly clinical interpretation of quality control, precision and accuracy must be determined for correct assessment of instrument function and test results. A high degree of both will, in turn, indicate a high degree of confidence in the test results. This is the ultimate purpose of quality control—to report correct results that will allow proper patient care. Quality control may also be approached from a different angle—directed more toward the interpretation than the final result. There has always been the need in clinical laboratory science to distinguish between a sample aberration and an instrument or test malfunction. Properly used, quality control helps in this area. The problem is less critical in a laboratory that is involved in more quantitative procedures such as clinical chemistry, but becomes increasingly important in those areas that rely more on subjectivity—e.g., blood differentials in hematology or interpretation of dot plots in flow cytometry. To address these issues, a specific daily quality control program is proposed in this unit to ensure instrument reliability and accurate results within the present constraints. This quality control program is intended to assess instrument performance and makes no guarantee of the overall test results, because interpretation depends on many variables—e.g., sample preparation. Flow cytometry has the advan-
UNIT 6.1
Phenotypic Analysis
6.1.1
tages, for example, of being able to look at large numbers of cells and distinguish variations in fluorescence in a more quantitative manner than has previously been possible. The ideal arrangement is to combine these positive features of flow cytometry with rigorous quality control to obtain accurate and reliable results.
STEPS FOR QUALITY CONTROL IN IMMUNOPHENOTYPING The following is an outline of the procedures involved as a laboratory prepares to run immunophenotyping samples. In general, these samples could be from a patient with any condition for which immunophenotyping is the preferred source of information (e.g., HIV or leukemia/lymphoma). The flow cytometer as an instrument is set up in much the same way for all tests; it is the antibody panels that change to reflect the different diagnostic questions being asked.
Initial Instrument Setup
Quality Control in Phenotypic Analysis by Flow Cytometry
The following are the first steps in setting up the flow cytometer. 1. Perform optical alignment (if necessary). 2. Establish desired values for particles to be used in regular monitoring. 3. Create a quality control log to record and follow values. 4. Establish test-specific and assay-specific instrument settings and channels—e.g., PMT voltages, compensation values, and expected coefficient of variation (CV). 5. Determine mean and standard deviation of monitored variables. The laboratory should perform multiple runs (at least 20) over multiple days (at least 5) to establish initial values. However, the procedure and numbers may vary according to individual laboratory practice and instrument/reagent manufacturer. Currently, most instrument manufacturers include training on instrument setup and verification as part of the instrument purchase package. Often, the flow cytometer manufacturer may provide either special software and/or suggested protocols for instrument setup and monitoring. However, laboratory personnel need to understand the principles behind these “automatic” processes. For their own protection, laboratories should have backup manual procedures and materials for instrument setup and quality control if problems arise— e.g., software failure or the unavailability of routine materials.
Daily Instrument Monitoring After the initial instrument setup has been completed, the laboratory must document and monitor results for change from expected variability (see UNIT 3.2 for discussion on quality control charts). The Center for Disease Control and Prevention (CDC) guidelines for daily verification are summarized below (also see the original document for these guidelines; information on this publication is listed in Table 3.2.1). 1. Align optics daily (or as required by instrument type). This ensures that the brightest and tightest peaks are produced in all parameters. Note that some flow cytometers can be aligned only by qualified service personnel. Alignment is usually performed using a stable calibration material (e.g., microbeads labeled with fluorochromes). 2. Standardize daily. This ensures that the flow cytometer is performing optimally each day and that its performance is the same from day to day. Procedures include selecting optimal instrument settings for the particular fluorochrome-labeled specimens, then using microbeads or other stable standardization material to place the scatter and fluorescence peaks in the same scatter and fluorescence channels each day. 3. Determine fluorescence resolution daily. The flow cytometer must differentiate between possible dim peak staining and auto fluorescence in each fluorescence channel. Laboratories can evaluate standardization or calibration material, or cells that have low-level fluorescence that can be separated from autofluorescence (e.g., using microbeads with low-level and negative fluorescence or CD56-labeled lymphocyte preparations). 4. Compensate for spectral overlap. This step corrects the spectral overlap of one fluorochrome into the fluorescence spectrum of another. Procedures include using either a microbead or cellular compensation material containing three populations (no fluorescence, PE fluorescence only, and FITC fluorescence only) for two-color immunofluorescence, four populations for three-color immunofluorescence, and five populations for four-color immunofluorescence. If calibration particles (microbeads) have been used to set compensation, the settings must be confirmed using lymphocytes labeled with appropriately conjugated monoclonal antibodies that recognize the separate cell populations, but do not overlap. These populations should have the brightest expected signals.
6.1.2 Current Protocols in Cytometry
5. Repeat the above four steps whenever instrument problems occur, and after service. 6. Maintain and monitor instrument quality control logs. The next step of a quality control program is to run a process control. This step is critical in assessing the total system: optics, fluidics, and technique. Virtually all specimens submitted for analysis contain a significant complement of normal hematopoietic cells. Analysis of the reactivity of panel antibody cocktails on these normal cells provides invaluable data regarding the technical performance of the reagents. If care is taken in the antibody-panel configuration, and the normal cells present in the sample provide distinct positive and negative controls for each antibody in use, then there is little or no need for a normal-donor process control to be run with each patient analysis run.
ANTIBODY SELECTION In the area of antibody selection for flow cytometric immunophenotyping for HIV, the CDC guidelines are specific in recommending a six-tube panel that includes CD45, CD14, CD3, CD4, CD8, CD19, CD16, and/or CD56, as well as isotype controls. The isotype reagent control is used to determine nonspecific binding of the mouse monoclonal antibody to the cells and to set markers for distinguishing fluorescence-negative and fluorescence-positive cell populations. The isotype control is recommended for the complete panel as described above. For monoclonal antibody panels containing antibodies to CD3, CD4, and CD8 only, this control may not be needed—because labeling with these anti bodies results in fluorescence patterns in which the unlabeled cells are clearly separated from the labeled cells. In these cases, the negative cells in the histogram are the appropriate “isotype” control. The preferred antibody panel for leukemia/lymphoma is an area of relative controversy. According to the most recent revision of the Committee on Antibody Panel Selection, Assembly and Utilization (a section of the Leukemia/Lymphoma Consensus Conference), there is a strong consensus in the community that specific isotype controls provide no more useful information than unstained cells alone, or than negative cells in selected antibody combinations (Poon et al., 1995). Therefore, it is very important to select antibody panels for lymphoproliferative diseases that allow combinations of markers clearly demonstrating negative populations for cells, for use in setting
quadrant statistics to quantify percentages of cells positive for expression.
QUANTITATIVE FLOW CYTOMETRY There is currently a movement toward more quantitative quality control procedures, called “quantitative flow cytometry” (QFCM). In this approach, molecules of equivalent soluble fluorochrome (MESF) and relative antibody binding capacity (ABC) have both been used and reported. MESF units have become internationally recognized as useful calibration units—as shown by their use as a measure of fluorescence intensity in the blind panel of the Fifth International Workshop on Human Leukocyte Differentiation Antigens. The determination of a correlation between a fluorescence signal and a flow cytometer’s response via calibrated standards is performed in accordance with three main requirements: 1. The standards used for calibrating the flow cytometer must have the same excitation and emission spectra as the fluorochromes used on the samples being analyzed; 2. A calibration standard must demonstrate a uniform enough signal that its population is described by a representative peak channel; 3. The calibration standard must be calibrated in some meaningful practical fluorescent unit. In a classification system proposed by Abe Schwartz in 1994, there are three types of quality control materials that are used to accomplish adequate quantitative performance: Type I (alignment), Type II (reference), and Type III (calibration). The technique is explained more fully in Zagursky et al. (1995) and Schwartz et al. (1996). In addition, a quality assurance overview using this technique will be provided in Stelzer et al. (submitted).
CONCLUSION It has been the purpose of this discussion to examine the need for quality control programs in flow cytometry and to offer a realistic, dayto-day approach that allows for excellent quality control as well as time and cost savings. The necessity of quality control as a tool in providing accurate results and as an aid in troubleshooting has been described. Future progress will entail the continued development of materials to provide more accurate assessments of the appropriate parameters. This will allow for more confident interpretation of the most complex of data.
Phenotypic Analysis
6.1.3 Current Protocols in Cytometry
LITERATURE CITED Bachner, P. 1996. Quality assurance of the analytic process: Pre- and postanalytic variation. Clin. Lab. Med. Vol. 6. Duckworth, J.K. 1987. Laboratory licensure and accreditation. In Laboratory Quality Assurance (P.J. Howanitz and J.H. Howanitz, eds.). McGraw-Hill, New York. Poon, R., Johnson, R., Stanton, T., Fernandez-Repollet, E., and Schwartz, A. 1995. Calibration of fluorescence intensity microbead standards and assignment of effective F/P ratios to the FITCconjugated secondary antibody used in the Vth International Leukocyte Typing Workshop. In Leukocyte Typing V pp. 13-16. Oxford University Press, Oxford.
Schwartz, A., Fernandez-Repollett, E., Vogt, R., and Gratama, J. 1996. Standardizing flow cytometry: Construction of a standardized fluorescence calibration plot using matching spectral calibrators. Cytometry 26:22-31. Stelzer, G. T., Marti, G., Hurley, A. A., McCoy, P., Lovett, E. J., and Schwartz, A. Standardization of technical components of leukemia/lymphoma analysis by flow cytometric immunophenotyping. Submitted for publication. Zagursky, R. J., Sharp, D., Solomon, K., and Schwartz, A. 1995. Quantitation of cellular receptors by a new immunocytochemical flow cytometry technique. Biotechniques. 18:3.
Contributed by Anne A. Hurley Comprehensive Cytometric Consulting Ballwin, Missouri
Quality Control in Phenotypic Analysis by Flow Cytometry
6.1.4 Current Protocols in Cytometry
Immunophenotyping
UNIT 6.2
There are four basic methods for staining cells with antibodies for immunophenotyping by flow cytometry. The first method (see Basic Protocol 1) is an indirect one that employs a primary unconjugated monoclonal antibody followed by a secondary, fluorochromeconjugated polyclonal antibody. Because the target antibody is not conjugated with a fluorochrome in this method, a second fluorochrome-conjugated polyclonal antibody— derived from a different animal species and directed against the IgG from the species that generated the first antibody—is used. An intact polyclonal antibody preparation should never be used for immunophenotyping; the fluorochrome-conjugated F(ab′)2 fragment should always be used (see Critical Parameters). The second method (see Alternate Protocol 1) is also an indirect method, in this case employing a hapten-conjugated primary antibody followed by a fluorochrome-conjugated polyclonal antibody against the hapten. Examples of haptens are digoxigenin, dior trinitrophenol, and sometimes biotin. A variation on this method included in Alternate Protocol 1 employs a biotinylated antibody followed by a fluorochrome-conjugated streptavidin. The fourth method (see Alternate Protocol 2), which is the method of choice, employs a directly conjugated monoclonal antibody against the desired antigen. Combinations of the above methods are also described for two-color staining (see Basic Protocol 2 and Alternate Protocols 3 and 4) as well as three- and four-color staining (see Alternate Protocols 5 to 9). The ethidium monoazide (EMA) procedure for detecting nonviable cells in a cell population is also included (see Support Protocol 1). Finally, analysis of the data acquired from flow cytometry using cells stained by the above procedures is detailed (see Support Protocols 2, 3, and 4). BASIC INDIRECT STAINING This is the most traditional method of staining. It uses unconjugated primary antibody followed by fluorochrome-conjugated secondary antibody. The most commonly used fluorochromes are fluorocein (FITC), phycoerythrin (PE), PE-Cy5 tandem complex, and PerCP. The PE—Texas red tandem complex is utilized to a lesser extent, mainly because of the lack of availability of directly conjugated reagents. Cy5 and allophycocyanin are likely to become important fluorochromes with the availability of dual laser excitation. Antibodies are available from several suppliers—e.g., Becton Dickinson Immunocytometry, Caltag Labs, Coulter, Gen Trak, Pharmingen, Sigma, and R & D Systems (see SUPPLIERS APPENDIX).
BASIC PROTOCOL 1
Materials 3 mg/ml normal goat IgG 5–10 × 106 cell/ml target cell suspension in PBS (APPENDIX 2A) Unconjugated monoclonal antibody against cell-surface antigen of interest, appropriately titered (UNIT 4.1) Isotype control: myeloma protein of appropriate isotype Erythrocyte-lysing solution (see recipe), 4°C Fluorochrome-conjugated polyclonal goat anti–mouse IgG F(ab′)2, appropriately titered Phosphate-buffered saline (PBS; APPENDIX 2A) 2% formaldehyde (see recipe) Centrifuge and rotor capable of 2000 × g, refrigerated Phenotypic Analysis Contributed by C.C. Stewart and S.J. Stewart Current Protocols in Cytometry (1997) 6.2.1-6.2.18 Copyright © 1997 by John Wiley & Sons, Inc.
6.2.1
1. Add 67 µl of 3 mg/ml goat IgG per ml of target cell suspension (200 µg/ml IgG final). Incubate 10 min in an ice bath. See Critical Parameters for additional discussion of incubation temperature. The IgG is added to block all Fc receptors and prevent nonspecific binding; see Critical Parameters.
2. Add the appropriate amount of unconjugated monoclonal antibody to a fresh tube, then add 50 µl of blocked cell suspension from step 1. Prepare an isotype control tube and a blank tube with no added reagents (just cells) in the same manner. 3. Incubate 15 min in an ice bath, then add 3 ml of 4°C erythrocyte-lysing solution. 4. Centrifuge 3 min at 1500 × g, 4°C. Decant the supernatant and resuspend cells in the residual solution. 5. Add the appropriate amount of fluorochrome-conjugated goat anti–mouse IgG F(ab′)2. Incubate cells 15 min in an ice bath, then add 3 ml PBS. 6. Centrifuge 3 min at 1500 × g, 4°C. Decant the supernatant and resuspend cells in the residual solution 7. Add 200 µl of 2% formaldehyde and incubate ≥1 hr at room temperature before acquiring data. See Critical Parameters for additional discussion of fixation time. ALTERNATE PROTOCOL 1
INDIRECT STAINING USING BIOTINYLATED OR HAPTEN-CONJUGATED ANTIBODY This technique should be used when only biotinylated or hapten-conjugated antibodies are available. Proceed as in the basic indirect staining procedure (see Basic Protocol 1) with the following modifications at steps 1, 2, and 5. Additional Materials (also see Basic Protocol 1) 3 mg/ml normal mouse IgG Biotinylated or hapten-conjugated monoclonal antibody against cell-surface antigen of interest, appropriately titered (UNIT 4.1) Isotype control: biotinylated or hapten-conjugated myeloma protein Fluorochrome-conjugated goat anti-hapten F(ab′)2 or fluorochrome-conjugated streptavidin, appropriately titered 1a. Use 3 mg/ml normal mouse IgG in place of the normal goat IgG. 2a. Use biotinylated or hapten-conjugated monoclonal antibody in place of the unconjugated monoclonal antibody. Use biotinylated or hapten-conjugated monoclonal antibody against irrelevant antigen for isotype control. 5a. Use fluorochrome-conjugated goat anti-hapten F(ab′)2 or fluorochrome-conjugated streptavidin in place of the fluorochrome-conjugated goat anti–mouse IgG F(ab′)2.
ALTERNATE PROTOCOL 2
Immunophenotyping
DIRECT STAINING PROCEDURE Directly conjugated antibodies offer faster staining and are the most convenient tools in clinical immunology, as the mixtures of antibodies can be subjected to a high level of quality control. These directly conjugated antibodies are available with all the fluorochromes from the suppliers named in Basic Protocol 1. Proceed as in the basic indirect staining procedure (see Basic Protocol 1) with the following modifications at steps 1, 2, and 5.
6.2.2 Current Protocols in Cytometry
Additional Materials (also see Basic Protocol 1) 3 mg/ml normal mouse IgG Fluorochrome-conjugated monoclonal antibody against cell-surface antigen of interest, appropriately titered (UNIT 4.1) Isotype control: fluorochrome-conjugated myeloma protein 1a. Use 3 mg/ml normal mouse IgG in place of the normal goat IgG. 2a. Use fluorochrome-conjugated monoclonal antibody in place of the unconjugated monoclonal antibody. Use fluorochrome-conjugated myeloma protein for isotype control. 5a. Omit step 5 of Basic Protocol 1 (addition of secondary antibody); also omit subsequent centrifugation (step 6). Proceed directly with formaldehyde fixation (step 7) using cells suspended in the residual erythrocyte-lysing solution (see step 4). As a monoclonal antibody directly conjugated to fluorochrome is used, no labeled secondary antibody is needed.
TWO-COLOR IMMUNOPHENOTYPING USING UNCONJUGATED PRIMARY ANTIBODY/FLUOROCHROME-CONJUGATED SECONDARY ANTIBODY IN COMBINATION WITH BIOTINYLATED ANTIBODY
BASIC PROTOCOL 2
The four basic procedures for staining cells with antibody (i.e., indirect using unconjugated primary antibody and fluorochrome-conjugated secondary antibody, indirect using biotinylated antibody, indirect using hapten-conjugated antibody, and direct using fluorochrome-conjugated monoclonal antibody) can be combined in various ways to produce simultaneous evaluation of multiple antigens. This protocol combines the first two of these indirect staining procedures. There are two general rules that must be adhered to for obtaining valid results with this protocol. First, the unconjugated primary antibody/fluorochrome-conjugated secondary antibody-labeling step must be performed separately and before labeling with the next procedure. Second, when the first labeling reaction is complete but before addition of subsequent antibodies, blocking (see Critical Parameters) must be performed with IgG from the species from which the unconjugated antibody was derived (e.g., mouse IgG for mouse antibodies). Materials 3 mg/ml normal goat IgG 5–10 × 106 cell/ml target cell suspension Unconjugated monoclonal antibody against cell-surface antigen of interest, appropriately titered (UNIT 4.1) Isotype control: unconjugated myeloma protein Erythrocyte-lysing solution (see recipe) Fluorochrome-conjugated polyclonal goat anti–mouse IgG F(ab′)2, appropriately titered Phosphate-buffered saline (PBS; APPENDIX 2A) 3 mg/ml normal mouse IgG Biotinylated monoclonal antibody against a second cell-surface antigen of interest, appropriately titered (UNIT 4.1) Isotype control: biotinylated myeloma protein Fluorochrome-conjugated streptavidin, of concentration such that appropriate quantity is contained in 10 µl 2% formaldehyde (see recipe) Centrifuge and rotor capable of 2000 × g, refrigerated
Phenotypic Analysis
6.2.3 Current Protocols in Cytometry
Label with unconjugated primary antibody and fluorochrome-conjugated secondary antibody 1. Add 67 µl of 3 mg/ml goat IgG per ml of target cell suspension (200 µg/ml IgG final). Incubate 10 min in an ice bath. See Critical Parameters for additional discussion of incubation temperature. The IgG is added to block all Fc receptors and prevent nonspecific binding; see Critical Parameters.
2. Add the appropriate amount of unconjugated monoclonal antibody to a fresh tube, then add 50 µl of blocked cell suspension (from step 1). Prepare an isotype control tube and a blank tube with no added reagents (just cells) in the same manner. 3. Incubate 15 min in an ice bath, then add 3 ml erythrocyte-lysing solution. 4. Centrifuge 3 min at 1500 × g, 4°C. Decant the supernatant and resuspend cells in the residual solution. 5. Add the appropriate amount of fluorochrome-conjugated goat anti–mouse IgG F(ab′)2, Incubate 15 min in an ice bath, then add 3 ml PBS. 6. Centrifuge 3 min at 1500 × g, 4°C. Decant the supernatant and resuspend cells in the residual solution. Label with biotinylated antibody and fluorochrome-conjugated streptavidin 7. Add 10 µl of 3 mg/ml mouse IgG to each tube. Incubate 10 min in an ice bath. 8. Add the appropriate amount of biotinylated monoclonal antibody (or biotinylated isotype control). Incubate 15 min in an ice bath, then add 3 ml erythrocyte-lysing solution. 9. Centrifuge 3 min at 1500 × g, 4°C. Decant the supernatant and resuspend cells in the residual solution. 10. Add 10 µl fluorochrome-conjugated streptavidin. Incubate 15 min in an ice bath, then add 3 ml PBS. 11. Centrifuge 3 min at 1500 × g, 4°C. Decant the supernatant and resuspend the cells in the residual solution. 12. Add 200 µl of 2% formaldehyde and incubate at least ≥1 hr at room temperature before acquiring data. See Critical Parameters for additional discussion of fixation time. ALTERNATE PROTOCOL 3
TWO-COLOR IMMUNOPHENOTYPING USING BIOTINYLATED ANTIBODY IN COMBINATION WITH HAPTEN-CONJUGATED ANTIBODY This modification of Basic Protocol 2 provides a method for staining cells with a biotinylated and hapten-conjugated antibodies. One fluorochrome is conjugated to the biotin while the second antibody to the hapten(s) is conjugated with the other fluorochrome(s). While hapten-conjugated antibodies have been used in research, they have not had general use in clinical immunophenotyping. Because many different haptens could be used, each with their specific fluorochrome-conjugated second antibody, they present a potential strategy for creating complex multicolor combinations of primary antibodies, such as FITC, PE, PE–Texas red, and PE-Cy5 or PerCP. For two-laser excitation, the PE–Texas red is replaced by using either APC or Cy5.
Immunophenotyping
6.2.4 Current Protocols in Cytometry
Additional Materials (also see Basic Protocol 2) 3 mg/ml normal mouse IgG Biotinylated monoclonal antibody against one cell-surface antigen of interest and hapten-conjugated monoclonal antibody against a second antigen, appropriately titered (UNIT 4.1) Isotype control: biotinylated and hapten-conjugated myeloma protein Fluorochrome-conjugated streptavidin, of concentration such that appropriate quantity is contained in 10 µl Fluorochrome-conjugated monoclonal antibody, of concentration such that appropriate quantity is contained in 10 µl Proceed as in the basic protocol for two-color immunophenotyping (see Basic Protocol 2) with the following modifications at steps 1, 2, and 5. 1a. Use 3 mg/ml normal mouse IgG in place of the normal goat IgG. 2a. In place of the unconjugated monoclonal antibody, use a combination of appropriate quantities of a biotinylated monoclonal antibody against one antigen of interest and a hapten-conjugated antibody against a second antigen. Prepare an isotype control tube and a blank tube with no reagents (just cells) in the same manner. 5a. Use 20 µl of a combination of fluorochrome-conjugated streptavidin and fluorochrome-conjugated anti-hapten monoclonal antibody in place of the fluorochromeconjugated goat anti–mouse IgG F(ab′)2. Continue with subsequent centrifugation and resuspension (step 6), but omit additional treatment with biotinylated antibody and streptavidin (steps 7 to 11). Proceed directly with formaldehyde fixation (step 12) using cells suspended in residual PBS from step 6. TWO-COLOR IMMUNOPHENOTYPING USING BIOTINYLATED ANTIBODY IN COMBINATION WITH DIRECTLY CONJUGATED ANTIBODY
ALTERNATE PROTOCOL 4
Sometimes the desired fluorochrome-conjugated antibody is not available but a biotinylated one is. This situation is especially apparent when a PE–Texas red tandem is desired. By combining the biotinylated antibody with the directly conjugated ones, complex multicolor cocktails can be produced. Additional Materials (also see Basic Protocol 2) 3 mg/ml normal mouse IgG Biotinylated monoclonal antibody against one cell-surface antigen of interest and fluorochrome-conjugated monoclonal antibody against a second antigen, appropriately titered (UNIT 4.1) Isotype control: biotinylated and directly conjugated myeloma proteins Fluorochrome-conjugated streptavidin, of concentration such that appropriate quantity is contained in 10 µl Proceed as in the basic protocol for two-color immunophenotyping (see Basic Protocol 2) with the following modifications at steps 1, 2, and 5. 1a. Use 3 mg/ml normal mouse IgG in place of the normal goat IgG. 2a. In place of the unconjugated monoclonal antibody, use a combination of appropriate quantities of a biotinylated monoclonal antibody against one antigen of interest and a fluorochrome-conjugated monoclonal antibody against a second antigen. Prepare an isotype control tube and a blank tube with no reagents (just cells) in the same manner. 5a. Use 10 µl of appropriately titered fluorochrome-conjugated streptavidin in place of the fluorochrome-conjugated goat anti–mouse IgG F(ab′)2. Continue with sub-
Phenotypic Analysis
6.2.5 Current Protocols in Cytometry
sequent centrifugation and resuspension (step 6), but omit additional treatment with biotinylated antibody and streptavidin (steps 7 to 11). Proceed directly with formaldehyde fixation (step 12) using cells suspended in residual PBS from step 6. ALTERNATE PROTOCOL 5
THREE-OR FOUR-COLOR IMMUNOPHENOTYPING USING UNCONJUGATED ANTIBODY, BIOTINYLATED ANTIBODY, AND DIRECTLY CONJUGATED ANTIBODY Staining with three, four, or more colors can be accomplished by putting the antibody and second reagent steps together appropriately. To do this, one must always stain cells first with an unconjugated primary antibody and second fluorochrome-conjugated antibody. This is the only step in which a primary antibody cannot be combined with the others. The two most extensively used three-color combinations are FITC, PE, and PE-Cy5 tandem complexes or PerCP. The fourth color can be a PE–Texas red complex for single-laser excitation. The APC or Cy5 fluorochrome is used for the fourth color when a second laser operating at 635 nm is used. For example, PE-Cy5 second antibody could be used with the unconjugated primary antibody, a PE–Texas red–streptavidin could be used with the biotinylated antibody, and primary antibodies directly conjugated with FITC and PE could be used to complete the four-color staining paradigm. Additional Materials (also see Basic Protocol 2) One or two fluorochrome-conjugated antibodies against cell-surface antigens of interest, appropriately titered (UNIT 4.1) Proceed as in the basic protocol for two-color immunophenotyping (see Basic Protocol 2) with the following modification at step 8. 8a. Add appropriate amount of one or two fluorochrome-conjugated monoclonal antibodies along with the biotinylated monoclonal antibody.
ALTERNATE PROTOCOL 6
THREE- OR FOUR-COLOR IMMUNOPHENOTYPING USING BIOTINYLATED, HAPTEN-CONJUGATED, AND DIRECTLY CONJUGATED ANTIBODIES Whenever possible, it is better not to use an unconjugated primary antibody with a second fluorochrome-conjugated antibody. This is because two extra steps are required, and if blocking is not properly done, the conjugated antibodies added subsequently can bind to the second fluorochrome-conjugated antibody leading to artifactual data. Biotinylated, hapten-conjugated, and directly conjugated primary antibodies can be combined into cocktails as described in UNIT 4.1. The fluorochrome-conjugated streptavidin and anti-hapten second antibody can also be combined into a cocktail to produce a more efficient and flexible staining procedure.
Immunophenotyping
Additional Materials (also see Basic Protocol 2) 3 mg/ml normal mouse IgG Hapten-conjugated antibody against cell-surface antigen of interest, appropriately titered (UNIT 4.1) One or two fluorochrome-conjugated monoclonal antibodies against cell-surface antigens of interest, appropriately titered (UNIT 4.1) Isotype control: biotinylated, hapten-conjugated, and directly conjugated myeloma proteins Fluorochrome-conjugated streptavidin, of concentration such that appropriate quantity is contained in 10 µl Fluorochrome-conjugated anti-hapten monoclonal antibody, of concentration such that appropriate quantity is contained in 10 µl
6.2.6 Current Protocols in Cytometry
Proceed as in the basic protocol for two-color immunophenotyping (see Basic Protocol 2) with the following modifications at steps 1, 2 and 5. 1a. Use 3 mg/ml normal mouse IgG in place of the normal goat IgG. 2a. In place of the unconjugated monoclonal antibody, use a combination of appropriate quantities of biotinylated, hapten-conjugated, and one or two directly conjugated monoclonal antibodies. Prepare an isotype control tube and a blank tube with no reagents (just cells) in the same manner. 5a. Use 20 µl of a combination of fluorochrome-conjugated streptavidin and fluorochrome-conjugated anti-hapten monoclonal antibody in place of the fluorochromeconjugated goat anti-mouse IgG F(ab′)2. Continue with subsequent centrifugation and resuspension (step 6), but omit additional treatment with biotinylated antibody and streptavidin (steps 7 to 11). Proceed directly with formaldehyde fixation (step 12) using cells suspended in residual PBS from step 6. THREE- OR FOUR-COLOR IMMUNOPHENOTYPING USING BIOTINYLATED ANTIBODY AND DIRECTLY CONJUGATED ANTIBODIES
ALTERNATE PROTOCOL 7
Another strategy for multicolor immunophenotyping is to use a biotinylated primary antibody in combination with two or three directly conjugated antibodies. This procedure is most relevant because of greater availability for biotinylated reagents. All primary antibodies can be combined into a cocktail (see UNIT 4.1). Additional Materials (also see Basic Protocol 2) 3 mg/ml normal mouse IgG Biotinylated monoclonal antibody against cell-surface antigen of interest, appropriately titered (UNIT 4.1) Isotype control: biotinylated and fluorochrome-conjugated myeloma proteins Two or three fluorochrome-conjugated monoclonal antibodies against cell-surface antigens of interest, appropriately titered (UNIT 4.1) Proceed as in the basic protocol for two-color immunophenotyping (see Basic Protocol 2) with the following modifications at steps 1, 2 and 5. 1a. Use 3 mg/ml normal mouse IgG in place of the normal goat IgG. 2a. In place of the unconjugated monoclonal antibody, use a combination of appropriate quantities of biotinylated antibodies and two or three directly conjugated monoclonal antibodies. Prepare an isotype control tube and a blank tube with no reagents (just cells) in the same manner. 5a. Use 10 µl of fluorochrome-conjugated streptavidin in place of the fluorochrome-conjugated goat anti-mouse IgG F(ab′)2. Continue with subsequent centrifugation and resuspension (step 6), but omit additional treatment with biotinylated antibody and streptavidin (steps 7 to 11). Proceed directly with formaldehyde fixation (step 12) using cells suspended in residual PBS from step 6.
Phenotypic Analysis
6.2.7 Current Protocols in Cytometry
ALTERNATE PROTOCOL 8
THREE- OR FOUR-COLOR IMMUNOPHENOTYPING USING UNCONJUGATED PRIMARY ANTIBODY AND BIOTINYLATED SECONDARY ANTIBODY IN COMBINATION WITH DIRECTLY CONJUGATED ANTIBODIES Another strategy is to use a primary antibody followed by a biotinylated second antibody that is then colored with a fluorochrome-conjugated streptavidin. When applying this strategy, stain cells with the unconjugated primary antibody first, and follow this with the biotinylated second antibody. After blocking with normal mouse IgG, a cocktail containing all the other directly conjugated antibodies and the fluorochrome-conjugated streptavidin can then be added to finish the staining. Additional Materials (also see Basic Protocol 2) Biotinylated polyclonal goat anti–mouse IgG F(ab′)2 (Caltag Labs) Isotype control: unconjugated and fluorochrome-conjugated myeloma proteins Fluorochrome-conjugated streptavidin, of concentration such that appropriate quantity is contained in 10 µl Two or three fluorochrome-conjugated monoclonal antibodies against cell-surface antigens of interest, appropriately titered (UNIT 4.1) Proceed as in the basic protocol for two-color immunophenotyping (see Basic Protocol 2) with the following modifications at steps 5, 8, and 10. 5a. Use biotinylated goat anti–mouse IgG F(ab′)2 in place of the fluorochrome-conjugated goat anti–mouse IgG F(ab′)2. Prepare an isotype control tube and a blank tube with no reagents (just cells) in the same manner. 8a. Add 10 µl fluorochrome-conjugated streptavidin along with a combination of two or three fluorochrome-conjugated monoclonal antibodies in place of the biotinylated monoclonal antibody. 9a. Omit step 10 of Basic Protocol 2 and subsequent centrifugation and resuspension (step 11). Proceed directly with formaldehyde fixation (step 12) using cells suspended in residual PBS (step 9).
ALTERNATE PROTOCOL 9
THREE- OR FOUR-COLOR IMMUNOPHENOTYPING USING A COMBINATION OF DIRECTLY CONJUGATED ANTIBODIES The easiest, most efficient, and least troublesome procedure is to use all directly conjugated antibodies. Except for the PE–Texas red reagents, just about everything is available from one supplier or another. In addition, if a reagent is not available with the desired fluorochrome, it can be obtained as a custom conjugate from Becton Dickinson Immunocytometry, Caltag Labs, or Molecular Probes. In some instances, it may be necessary to supply them with the purified antibody. Additional Materials (also see Basic Protocol 1) 3 mg/ml normal mouse IgG Fluorochrome-conjugated monoclonal antibodies against cell-surface antigens of interest, appropriately titered (UNIT 4.1) Isotype control: fluorochrome-conjugated myeloma proteins Proceed as in the basic indirect staining procedure (see Basic Protocol 1) with the following modifications to steps 1, 2, and 5. 1a. Use 3 mg/ml normal mouse IgG in place of the normal goat IgG.
Immunophenotyping
2a. Use a combination of fluorochrome-conjugated monoclonal antibodies in place of the unconjugated monoclonal antibody. Prepare an isotype control tube and a blank tube with no reagents (just cells) in the same manner.
6.2.8 Current Protocols in Cytometry
5a. Omit step 5 of Basic Protocol 1 (addition of secondary antibody); also omit subsequent centrifugation (step 6). Proceed directly with formaldehyde fixation (step 7) using cells suspended in the residual erythrocyte-lysing solution (see step 4). As monoclonal antibodies directly conjugated to fluorochromes are used, no labeled secondary reagents are needed.
EMA PROCEDURE FOR DETECTING NONVIABLE CELLS IN A CELL POPULATION TO BE FIXED
SUPPORT PROTOCOL 1
Dead cells bind antibodies nonspecifically. Each antibody, however, binds dead cells to a different extent—which could lead to misinterpretation. It is important, therefore, to minimize dead cells in the analysis window. Ethidium monoazide (EMA) stains the dead cells, making them much brighter than the viable cells. EMA staining is done on a separate aliquot of the same cell suspension used in any of the staining procedures in this unit. The EMA tubes are analyzed on the flow cytometer along with the other immunophenotyping panels as described in Support Protocol 4 (also see Critical Parameters). For additional discussion of the purpose of EMA staining and the difference between EMA and other stains for nonviable cells (e.g., propidium iodide), see Background Information. Materials Cell suspension for analysis Erythrocyte-lysing solution (optional; see recipe) EMA working solution (see recipe) Phosphate-buffered saline (PBS; APPENDIX 2A) 2% formaldehyde (see recipe) 12 × 75–mm test tubes Centrifuge and rotor capable of 2000 × g Fluorescent desk lamp White pan 1. Add 50 µl of cell suspension to a 12 × 75 mm test tube. If necessary, add 3.5 ml of erythrocyte-lysing solution to lyse erythrocytes. 2. Centrifuge 3 min at 1500 × g, 4° to 25°C. Decant the supernatant and resuspend cells in the residual solution. 3. Add 5 µl EMA working solution. Determine optimal concentration for working solution for each batch of EMA.
4. Lay tube(s) flat in a white pan 20 cm beneath a fluorescent desk lamp at room temperature and irradiate for 10 min with fluorescent light. At end of irradiation, add 3.5 ml PBS. 5. Centrifuge 3 min at 1500 × g, 4° to 25°C. Decant the supernatant and resuspend cells in the residual solution. 6. Add 0.5 ml of 2% formaldehyde and incubate at least 1 hr before acquiring data.
Phenotypic Analysis
6.2.9 Current Protocols in Cytometry
DATA ANALYSIS FOR FLOW CYTOMETRIC IMMUNOPHENOTYPING Two basic ways to analyze flow cytometry data are the marker approach (described in Support Protocol 2), in which a marker is placed on a histogram to designate positive and negative cells, and the template approach (described in the Commentary), in which a region is drawn circumscribing the geometric pattern created by the cell populations. The marker approach (see Support Protocol 2) is well suited for the measurement of one or two colors of fluorescence, but for three and four colors, the template approach (see Background Information) may be more meaningful. SUPPORT PROTOCOL 2
Marker Approach Using Population Gate This approach is illustrated using five-parameter data. There are two ways to address analysis with the marker approach. The first, described in this protocol, is to establish a “population” gate using FS versus SS. The second (see Support Protocol 3) is to establish a “cell” gate using SS versus cell-specific antibody. For both these approaches, the region R9 is used for this purpose. The procedure here is illustrated with human leukocytes. This protocol requires listmode files with the acquired data. 1. Display bivariate plots of all parameters: 3 parameters = FS vs. SS and green fluorescence univariate histogram 4 parameters = FS vs. SS and green fluorescence vs. orange fluorescence (see Fig. 6.2.1) 5 parameters = FS vs. SS, green fluorescence vs. orange fluorescence, red fluorescence vs. orange fluorescence, and red fluorescence vs. green fluorescence (see Fig. 6.2.2 and Fig. 6.2.3) 2. To establish the desired population—e.g., lymphocytes—that express CD45 brightly, load the listmode file of cells stained with CD45 and CD14. 3. Refer to Figure 6.2.1A and draw a region (R9) to circumscribe the lymphocyte cluster and another region (R2) to circumscribe the granulocytes completely in the FS versus SS plot. This latter region is drawn in the FS versus SC display so that some granulocytes that overlap B cells that are less CD45-bright than T cells and NK cells can be excluded.
4. Draw a region R3 (Fig. 6.2.1B) around the CD45-bright–CD14-negative cells in that bivariate plot. Create the Boolean logical gate “R3 and NOT R2” and gate the FS versus SS display (Fig. 6.2.1C). The CD45+ cells (i.e., all lymphocytes) and the CD14− cells (i.e., NOT monocytes) then exclude the granulocytes (R2) by scatter.
5. Gate the CD45 versus CD14 display using R9 (Fig. 6.2.1D). Create results and determine the percentage of lymphocytes (CD45-bright) inside the region R9 and their purity from the cells inside R3. The purpose of this step is to determine the purity and the percentage (or yield) of lymphocytes that are in R9.
6. Load the file containing the unstained cells (or an isotype control file; Fig. 6.2.2), gated on region R9 and adjust quadrant markers in the green fluorescence versus orange fluorescence and red fluorescence versus green fluorescence display so that <2% of the events are beyond the marker. Immunophenotyping
The purpose of this step is to determine markers for positive and negative cells.
6.2.10 Current Protocols in Cytometry
A
B
SSC
PE-CD14
R2
R9
R3 FITC-CD45
FSC
D
C
purity = 94%
SSC
PE-CD14
R2
lymphocytes = 97%
R9
R3 FITC-CD45
FSC
PE-IgG2b
PE-IgG2b
Figure 6.2.1 Bivariate plots for establishing a lymphocyte gate.
3 104
101
104
101
102 103 TC-IgG2a
104
SSC
FITC-IgG1
101 102 103 FITC-IgG1
7 102 103 TC-IgG2a
10 30 50 70 90 110 130 FSC
Figure 6.2.2 Isotype control bivariate displays. See color figure.
Phenotypic Analysis
6.2.11 Current Protocols in Cytometry
PE-CD4
2
PE-CD4
1
4
3 101
102
103
104
101
FITC-CD3 6
7
8
104
SSC
PE-CD4
5
102 103 TC-CD8
101
102 103 TC-CD8
104
10 30 50 70 90 110 130 FSC
Figure 6.2.3 Bivariate displays of stained cells. See color figure.
7. Create the Boolean logical gates for all the combinations as shown in Table 6.2.1 and assign a color to each. The purpose of this step is to create the four (two-color) or eight (three-color) possible binary populations of positive and negative cells and to identify them by their color.
8. Load each file containing stained cells and record the results as shown in Figure 6.2.3 and Table 6.2.1. The purpose of this step is to analyze the acquired data. SUPPORT PROTOCOL 3
Marker Approach Using Cell Gate Instead of a population gate in FS versus SS, a cell gate can be used around cells stained with a specific antibody. This requires that the gating antibody be present in every tube. The gating antibody could be a lineage-specific one like CD3 for T cells, CD19 for B cells, or CD34 for hematopoietic progenitor cells. It might also be a combination like CD56/NOT CD3 for NK cells, or CD45 for all leukocytes. 1. Display bivariate plots of SS versus the gating antibody, FS versus SS, and antibody 1 versus antibody 2. For four-color data, the SS versus gating antibody, antibody 1 versus antibody 2, antibody 3 versus antibody 2, and antibody 1 versus antibody 3 are displayed as shown in Figure 6.2.4.
2. In the SS versus gating antibody view, draw region R9 and gate all other histograms on R9. Immunophenotyping
This step provides the gate for the cells of interest.
6.2.12 Current Protocols in Cytometry
PE-CD69
PE-CD69
102 103 TC-CD4
104
101
102 103 ECD-CD8
104
101
102 103 ECD-CD8
104
TC-CD4
FITC-CD3
101
10 30 50 70 90 110 130 SSC
Figure 6.2.4 Four-color bivariate displays. See color figure.
3. Set quadrant markers using the file from cells stained with only the gating antibody, with or without an isotype control, so that <2% of events are beyond the marker. This step provides the markers to distinguish positive and negative cells,
4. Create the Boolean logical gates for all the combinations and assign a color to each, as shown in Table 6.2.1. 5. Load each file containing stained cells and record the results. Table 6.2.1
Analysis of Data Using Quadrant Markers
Boolean expression
Cluster color
One color ––– +––
R3 and not R8 R4 and R5
Black Yellow
1 3
Two colors –+– ++– ––+
R1 and not R6 R2 and R5 R8 and R3
Cyan Green Brown
2 49 6
Three colors +–+ –++ +++
R6 and R4 R1 and R8 R2 and R6
Blue Violet Red
17 1 4
No. of colors
% positive cells in Figure 6.2.3
Phenotypic Analysis
6.2.13 Current Protocols in Cytometry
R2
% dead in gate = 1%
SSC
SSC
FL3-EMA
% dead = 5%
R9 FSC
R9 FSC
FSC
Figure 6.2.5 Determination of dead cells using EMA. See color figure.
SUPPORT PROTOCOL 4
Determining Viable Cells Because EMA staining (Support Protocol 1) is done on a separate aliquot of cells (see Critical Parameters), the location of dead cells is determined in the FS versus SS plot so they can be gated out of the analysis. Refer to Figure 6.2.5. 1. Display FS versus SS and FS versus EMA fluorescence. Draw a region R2 around the dead cells. 2. Gate FS versus SS on R2. Draw a gating region R9 so that >95% of cells are viable. By gating the FS versus red fluorescence display on R9, the percent of dead cells in R9 can be determined.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
EMA solution Prepare a stock solution of 5 mg/ml ethidium monoazide (EMA; Molecular Probes) in PBS (APPENDIX 2A) and store up to 2 years at −20°C; wrap aluminum foil around the container to keep in total darkness. Prepare working dilution of 5 µg/ml biannually, divide into 50-µl aliquots in 0.5-ml microcentrifuge tubes, and store at 4°C (wrapped in aluminum foil). Open tubes one at a time as necessary. CAUTION: EMA is a DNA dye and should be handled using protective clothing and gloves. Never pipet by mouth.
Erythrocyte-lysing solution 4.13 g ammonium chloride (154 mM final) 0.5 g potassium bicarbonate (10 mM final) 0.0185 g tetrasodium EDTA (0.082 mM final) 500 ml double-distilled H2O Formaldehyde, 2% 200 ml 10% formaldehyde (ultrapure, EM Grade; Polysciences) 800 ml Dulbecco’s PBS (Life Technologies)
Immunophenotyping
6.2.14 Current Protocols in Cytometry
COMMENTARY Background Information Antibodies The antibodies used as second antibodies or antibodies to haptens in these protocols are almost always polyclonal antibodies. These are derived from the serum of mammals that have been immunized with the immunoglobulin fraction of serum obtained from the animal species from which the first antibody was derived or with a hapten that has been conjugated to a large protein such as poly-L-lysine. Serumderived polyclonal antibodies consist of IgM, all subclasses of IgG, and IgA. All of these antibodies will have specificities for the immunogen. Because pentameric IgM and quadrameric IgA are present, the F(ab′)2 fragments should always be used, as all antibodies will then be dimeric. This produces a uniform preparation with the lowest amount of undesired cell binding. Polyclonal antibodies are usually purified by passing the F(ab′)2 preparations over a column composed of protein A or protein G or anti-IgG directed against the animal’s purified IgG light chains or heavy chains. The latter method is best because only those immunoglobulin molecules specific for the immunogen are recovered. Detection of dead cells Propidium iodide (PI) stains DNA and is commonly used to detect dead cells in a fixed preparation because it is excluded by viable cells with intact membranes. In flow cytometry, use of viable cells presents both a health hazard and the inconvenience of requiring immediate evaluation; therefore cells are frequently fixed. Unfortunately, when cells are fixed all cells stain with PI because they are no longer viable. PI stains reversibly, and will leak out of cells during fixation and spread through the whole preparation. This process also occurs with other dyes, including 7-aminoactinomycin D (7AAD). Ethidium monoazide (EMA) provides a solution to this problem (Riedy et al., 1991). Like PI and 7-AAD, it stains the DNA of dead cells, but not viable ones. When stained cells are exposed to light, an azide group on the dye is photoactivated and covalently bonds to DNA and histones. After exposure, dye that has not been bound can be washed away and the dead cells are permanently stained. Fixation will not change the situation.
Data analysis Software for the analysis of immunophenotyping data has evolved over the years as the number of antibodies combined together has increased. The fundamental approach is to set a marker to resolve antibody-positive cells from negative ones. The most primitive analysis strategy is to create a gate in the FS versus SS view and apply it to all files. Next, the antibodyfluorescence histogram is displayed. For a single antibody a univariate histogram is used. For antibodies in combination, bivariate histograms are displayed and quadrant markers are inserted. The position of markers is set using unstained cells or an isotype control. There is no rule for the percentage of events allowed above the marker, but usually it is <1% to 2%. The number of bivariate histograms (BH) increases with the number of antibodies (n) according to the formula: BH = (n − 1) + (n − 2) + ... + (n − n). The number of bivariate histograms that require quadrant markers is (n − 1). The number of binary populations that can be resolved by the quadrant markers expressed in terms of Boolean algebra is 2n. Thus, 3 antibodies provide 8 populations, 4 provide 16, 5 provide 32, etc. Clearly the ability to visualize all of these populations becomes increasingly difficult (see UNIT 10.4). As interest in measuring more parameters increases, new approaches to data analysis are required. One such approach is cell gating, where one antibody is used to resolve cells of interest, such as CD45 for lymphocytes, CD3 for T cells, CD19 for B cells, CD14 for monocytes, and CD34 for hematopoietic progenitor cells. Other antibody combinations—e.g., CD4/CD8 for T cells or anti-kappa/anti-lambda for B cells, are used to resolve the population subsets. Another approach is to use a template composed of regions that define all the populations resolved by the antibody combination. This approach assumes that distinct clusters occur as well-resolved geometric shapes in the various bivariate views of multidimensional space. Thus, in one view a cluster may appear homogeneous, but a second view reveals two or more clusters. Regions linked by Boolean algebra are used to define all distinct clusters until all are homogeneous. Homogeneity does not imply any particular geometric shape, only that no further separation occurs and all kinds can be found. The template approach can be automated using clustering algorithms. Although several
Phenotypic Analysis
6.2.15 Current Protocols in Cytometry
attempts have been made to apply these mathematical approaches, the results have not been promising, mainly because each file represents an entirely new experience for the algorithm. Applying neural networks or classification and regression trees provides the added dimension of experience for analysis of high-dimensional data.
Critical Parameters Washing Recently, several suppliers have introduced procedures with fluorochrome-conjugated antibodies that require only lysing erythrocytes but not washing the specimen. This may seem to be a good idea, even though it is really a bad one for several reasons. First, the supposition is fewer cells are lost because they are not washed away. In reality, positive cells are lost to detection because the nonspecific binding increases markedly for most antibodies. Although the stained cells do not change in brightness, the increased nonspecific binding and fluorescence in the cell stream causes a decreased signal-to-noise discrimination. When cells are washed, some may be discarded along with the supernatant fraction, but the solution to this problem is faster centrifugation (1500 × g, rather than 300 × g) as recommended in this protocol. The higher speed does not increase cellular aggregation and no intact cells are found in the discarded wash supernatant. Proponents of the lyse/no wash system also believe it saves time by eliminating washes. This is a false belief as well. Using the bare minimum of 1 ml lysing reagent results in a final volume >3× that of the washed cells (after resuspension in 2% formaldehyde)—data acquisition using the flow cytometer will also take >3× as long for the same number of cells. In addition, erythrocyte lysis is likely to be incomplete, leaving intact erythrocytes whose presence in gating regions will affect the calculations for percent positive cells. Washing all samples is strongly recommended. There are several important reasons for this recommendation. First, it is desirable to remove unreacted antibody. Second, higher backgrounds will be evident without washing, which may result in dim antigen expression being missed.
Immunophenotyping
Blocking Intact antibodies, especially monoclonal antibodies, bind to any cell that has unoccupied Fc receptors. Even when the cells are derived
from blood that contains all immunoglobulin isotypes, these receptors are not fully saturated because the immunoglobulin concentration in blood is nonsaturating. For this reason, serum should never be used for blocking in these protocols. It is necessary to add a higher concentration of IgG to block all receptors so that the antibody binds only to its epitope and not to Fc receptors as well. All hematopoietic cells have Fc receptors except T cells and erythrocytes. Sequence of reactions When a second conjugated polyclonal antibody is used, that staining reaction must always be performed first and the IgG block must be from the same animal species as the polyclonal antibody. It cannot be assumed that all Fab sites on the second antibody have bound epitopes on the primary. To the extent these are free binding sites, addition of another primary antibody can bind to these free sites rather than to its epitope. Thus, any cell that is positive for the unconjugated primary antibody can also be positive for all others, unless these free binding sites are blocked. This is done by adding purified IgG from the same species as the one from which the primary unconjugated antibody was derived. Failure to perform this procedure in the correct order will result in flawed data. Temperature The temperature of staining has not been indicated. Many protocols recommend room temperature (∼22°C). The authors recommend placing the rack of tubes in an ice bath (4°C). Many cell types—e.g., myeloid cells—still function at room temperature, albeit more slowly, and can internalize bound antibodies. Fixation A minimum fixation time of 1 hr has been indicated in the various protocols. This is because known hazardous virus will be inactivated, making the samples safer to handle while acquiring data. The FS versus SS distribution of aldehyde-fixed cells changes markedly during the first 8 hr because the continuous crosslinking of proteins results in changes of cell shape and granularity. Because of this, the FS versus SS profile for granulocytes (and to a lesser extent for monocytes) that is obtained after 1 or 2 hr will look different from that obtained after longer fixation times. Lymphocytes are not affected because of their low granularity. A 12-hr fixation time is recommended for complete stabilization. This time
6.2.16 Current Protocols in Cytometry
period is most convenient because it corresponds to an overnight fixation. Fixed suspensions should be evaluated within 5 days because autofluorescence will increase with longer storage times. Dead cells Dead cells bind antibodies nonspecifically, and their presence can lead to misinterpretation. In most instances the use of EMA enables the establishment of a gate to exclude nearly all dead cells based on an FS versus SS gate. The authors use the approach of defining a region on the desired population to optimize exclusion of dead cells so that the detector otherwise allocated to EMA can be used for another fluorochrome. If the gated region on the desired cell population contains >5% dead cells (i.e., <95% viable cells in the region), one can accept that the results of analysis may be flawed to the extent the result is caused by nonspecific antibody binding to dead cells. One can decrease the gate size to reduce the percentage of dead cells, even if it means decreasing the percentage of desired cells evaluated, or one can terminate evaluation of the data—bad data is worse than no data. EMA (like PI or 7-AAD) can be added as a third color after staining cells with FITCand PE-conjugated antibodies. Data analysis For single-color analysis, a simple univariate histogram suffices because cells are either negative or positive for the antibody. When two antibodies are combined, four populations (see Table 6.2.1 and Fig. 6.2.1) can be resolved. For three-antibody combinations there are eight populations; but because two bivariate views are needed to resolve their combination, Boolean expressions are required and a color must be assigned to them so each cell population can be visualized in all the bivariate views. The two regions R9 and R2 are drawn in the FS versus SS plot for determining lymphocyte purity because region R3 must be extended somewhat into the space occupied by granulocytes (CD45-dim) if all B cells are to be included. Some CD45 monoclonal antibodies do not resolve B cells, which are dimmer than T cells and NK cells, from the brighter CD45-dim granulocytes. Because these granulocytes are excluded by forward scatter gating on R9, they also need to be excluded when backgating on R3. This is accomplished using the Boolean combination “R3 and NOT R2.” In this way the percentage of lymphocytes in the gate R9 can be determined. The cells in R3, when gated on
R9, contain the purity of the lymphocyte gate. In the example shown in Figure 6.2.1, 97% of lymphocytes—the cells in R3 that are inside R9—are counted, with a purity of 94% for cells in R9 that are also in R3. These percentages should be >90% for lymphocytes. Large lymphocytes, which are most important in disease, are frequently excluded from the analysis. Note that the large cells with low SS to the right of the dense cluster have been included in R9 because the lymphocyte gate (R9) circumscribes only the dense lymphocyte cluster. These cells are very important and should always be included even though they often reside under the monocyte cluster. R9, rather than R1, is used as the gate region because most analysis software applications assign colors in a hierarchical fashion. If R1 were the gate, it would not be possible to assign colors to the various cell populations resolved by the Boolean logical gates shown in Table 6.2.1. It is desirable to use multiparameter software that allows at least nine regions, as shown in Figures 6.2.1 and 6.2.2 and Table 6.2.1. Regions 1 to 8 are used for the quadrant markers and regions 9 and higher are used for gating. Some software applications attach a color based on a Venn spectrum to the gate logic, in a hierarchical order instead of allowing userselected colors for each population. In viewing multiparameter data, this software restriction is often undesirable because the color mixtures can duplicate the background color and the population will completely disappear from the view. The hierarchical nature associated with gating means the gate must always be last. Because of hierarchical color assignment, if the gate region is first, it will not be possible to assign colors to any other region. Thus, it must be last so that its color does not dominate. Color assignments for quadrant markers combined by Boolean equations must be assigned by the user and not the software.
Anticipated Results The most important single potential problem with results stems from the quality of the antibody. Refer to UNIT 4.1 for a more complete discussion of these problems. The objective of these protocols is to resolve epitope-positive cells from negative ones. To do so, there must be enough epitopes, and the antibody’s fluorochrome must be bright enough for the investigator’s instrument to measure. One may not be able to do anything about the number of epitopes but the fluorochrome and the instrument
Phenotypic Analysis
6.2.17 Current Protocols in Cytometry
are within the investigator’s power. Often, stream-in-air detection is less sensitive than stream-in-liquid (flow-cell cuvette) detection. Thus, the instrument plays a role in sensitivity. Because PE-containing fluorochromes are five to seven times brighter than the others, a strategy to use them for conjugation to antibodies that will be used to detect the cells with the least number of epitopes can markedly improve results. Finally, Fc-receptor binding, dead cells, and nonspecific binding affect both the result and its interpretation. Fc-receptor binding can be eliminated by appropriate blocking; dead cells can often, but not always be removed from analysis by appropriate gating. Nonspecific binding problems can be reduced by washing. When a cell population is poorly resolved by one antibody, it can often be well resolved by combining it with a second to pull the population away from other cells. Even if both antibodies are directed toward antigens in low abundance, their combined effect can be rewarding. This is one of the best reasons for performing multicolor flow cytometry. Investigators have been taught to expect that immunophenotyping by flow cytometry will resolve positive and negative cells. Clearly, if one desires to have a CD4 count or a CD34 count, this is what to expect. But the pathologist does not expect this result when using immunohistochemistry to identify cells in a tissue. When cell populations are to be identified, it is possible to combine antibodies that resolve different cells and their pattern in bivariate plots can be used to identify them. In the case of pathology, their pattern is the expected result, and deviations from the known pattern may indicate abnormal processes such as malignancy.
Time Considerations One should always set up the necessary tubes and reagents prior to getting cells. If 96 tubes (one full standard rack) are used as an example, this preparation should take ∼30 min. The time to prepare cells depends on where they come from; nevertheless, they should be obtained just before they are to be stained, when-
ever possible. Counting, washing, and blocking takes at most 15 min. Adding cells to the antibody cocktails takes another 5 min. For each block of steps, 20 min is required. Thus—in a procedure where staining with an unconjugated antibody (20 min) and subsequent staining with secondary antibody (20 min) is combined with a block (10 min) prior to adding a biotinylated antibody combined with directly conjugated antibodies (20 min) and finally adding fluorochrome-conjugated antibody or antibodies (20 min)—up to 90 min may be necessary for staining. For acquiring data on the instrument, the authors usually require 1 min per tube. This will vary depending on cell concentration and the number of events acquired. It is difficult to evaluate analysis time, because this depends on how much prior knowledge of the samples is available. If it is a routine evaluation where the analysis protocol is known, it may require only 1 to 2 min per sample. Thus, 96 samples can take as little as 4 hr to process from start to finish.
Literature Cited Riedy, M.C., Muirhead, K.A., Jensen, C.P., and Stewart, C.C. 1991. The use of a photolabeling technique to identify nonviable cells in fixed homogenous or heterologous cell populations. Cytometry 12:133-139.
Key References Stewart, C.C. and Stewart, S.J. 1994. Cell preparation for the identification of leukocytes. In Methods In Cell Biology (Z. Darzynkiewicz, J. Robinson, and H. Crissman, eds.) pp. 39-60. Academic Press, New York. Provides additional information on preparing cells. Stewart, C.C. and Stewart, S.J. 1994. Multiparameter analysis of leukocytes by flow cytometry. In Methods in Cell Biology (Z. Darzynkiewicz, J. Robinson, and H. Crissman, eds.) pp. 61-79. Academic Press, New York. Provides additional information on instrument setup, data acquisition, and data analysis.
Contributed by C.C. Stewart and S.J. Stewart Roswell Park Cancer Institute Buffalo, New York
Immunophenotyping
6.2.18 Current Protocols in Cytometry
High-Sensitivity Immunofluorescence/ Flow Cytometry: Detection of Cytokine Receptors and Other Low-Abundance Membrane Molecules
UNIT 6.3
Immunofluorescence-based flow cytometry, in common with other analytical techniques, has limited sensitivity. Thus, it is important to understand that when it is said that T cells do not express CD32, it is only known with certainty that CD32 is undetectable on T cells with the methods used. The detection limit, or sensitivity, depends on the reagents, staining, and instrument parameters. For the most commonly used procedures and reagents, the sensitivity is ∼2000 molecules of target antigen per cell (this parameter is different from the widely used “molecules of equivalent soluble fluorescein,” MESF). This level of sensitivity is adequate for most purposes. Molecules such as CD3, CD4, and CD8 are expressed at ∼20,000 to 100,000 molecules/cell on the appropriate T cell subsets. Most of the widely used leukocyte markers are expressed at similar levels on the cells of interest (Table 6.3.1). Molecules expressed at much lower concentrations are, however, biologically significant. Growth factors can activate cells through receptors expressed at very low concentrations (e.g., 1000 molecules or less; Table 6.3.1). The same may be true for molecules involved in intercellular interactions, such as B7 (CD80 and CD86) and its ligand CTLA-4 (CD152), receptors for Ig, or adhesion and homing molecules. For the example above, CD32 Table 6.3.1 Typical Concentration of a Selection of Widely Used Immunological Markers, Compared with Cytokine Receptors
Marker
Cell typea
Concentration (molecules/cell)
Reference
Commonly used markers CD2
T cells (blood)
40,000
Martin et al. (1983)
CD3
T cells (blood)
57,000
Bikoue et al. (1996)
CD4
T cell subset (blood)
47,000
Bikoue et al. (1996)
CD8
T cell subset (blood)
145,000
Bikoue et al. (1996)
CD5
T cells (blood)
50,000
Bikoue et al. (1996)
CD19
B cells (blood)
27,000
Bikoue et al. (1996)
CD45
Lymphocytes
sIg
Chronic lymphocytic leukemia
217,000 6,500-22,500
Bikoue et al. (1996) Dighiero et al. (1980)
Cytokine receptors CD121a
Lymphocytes
<100
Dower et al. (1985)
CD25
T cells (blood)
<500
Le Mauff et al. (1987)
CD25
In vitro-activated T cells
>30,000
Le Mauff et al. (1987)
CD122
T cells (blood)
700
Ben Aribia et al. (1989)
CD124
Resting B lymphocytes (mouse)
400
Lowenthal et al. (1988)
CD126
Activated B cells
300
Kishimoto (1989)
a Human, unless indicated otherwise.
Phenotypic Analysis Contributed by Heddy Zola Current Protocols in Cytometry (2004) 6.3.1-6.3.13 C 2004 by John Wiley & Sons, Inc. Copyright
6.3.1 Supplement 30
expression has been demonstrated on some T cells using high-sensitivity immunofluorescence or PCR (Mantzioris et al., 1993). Although the function of this low-affinity IgGreceptor on a subpopulation of T cells is not known, the demonstration of its expression raises the question and permits experimental approaches to determining its function. Similarly, the demonstration of cytokine receptors on both normal and malignant cells that were previously classified as negative opens up a number of interesting functional questions. Flow cytometric immunofluorescence is capable of detecting a single molecule in specialized situations, and 100 molecules/cell or better in practical flow cytometric applications, provided that every step of the staining and analysis procedure is optimized for sensitivity. This unit discusses the underlying principles of high-sensitivity immunofluorescence and provides specific experimental procedures to achieve high levels of sensitivity. Staining and analysis procedures are provided for single-color fluorescence (see Basic Protocol) and multicolor fluorescence (see Alternate Protocol). Support protocols describe the titration and quality control of reagents (see Support Protocol 1) and procedures for obtaining high sensitivity in preparative sorting (see Support Protocol 2).
STRATEGIC PLANNING The major factors that determine sensitivity are the fluorochrome, the staining protocol, the reagent quality, and the instrument parameters. Phycoerythrin (PE) and other phycobiliprotein have a higher extinction coefficient (amount of incident light energy absorbed) and give a better quantum yield (energy of emitted signal for a given incident energy) than most other dyes that can be excited at readily available wavelengths (Table 6.3.2). The recently-described Alexa dyes also have properties that make them suitable for highsensitivity applications (Table 6.3.2). Quantum dots (Lidke et al., 2003) appear very promising, but have not been fully characterized in this respect at the time of writing. The staining protocol should maximize specific binding and minimize nonspecific binding. This may require close attention to parameters such as incubation time, temperature, and number of washes. Two- or three-layer protocols give a greater signal amplification than direct fluorochrome-labeled antibodies; however, if the nonspecific binding increases proportionately, there is no real gain in sensitivity. Fluorochrome-conjugated anti-Ig antibodies frequently stain a subset of cells nonspecifically, even in the absence of a primary antibody. On the other hand, a 5-layer procedure using unlabeled mouse monoclonal antibody, biotinylated anti-mouse Ig, PE-streptavidin, biotinylated anti-PE antibody (Becton Dickinson), and a further layer of PE-streptavidin can give very high sensitivity without significant background (Mavrangelos et al., 2004). Most PE reagents are intended for use Table 6.3.2 Fluorochrome Properties
Fluorochrome Fluorescein R-Phycoerythrin R-PE/Cy5 tandem Cy3
High-Sensitivity Detection of Low-Abundance Membrane Molecules
Absorption maximum wavelength (nm)
Emission maximum wavelength (nm)
Extinction coefficient (mol−1 cm−1 )
Quantum yield
495
520
8.2 × 104
0.3
580
2 × 10
0.8
667
2 × 10
565
1.3 × 10
>0.15
5
546 546 552
6
<0.8
6 5
PerCP
478
677
3.2 × 10
NAa
Alexa 488
495
519
6.5 × 104
NAa
Alexa 647
650
668
NAa
NAa
a NA, not available. Extinction coefficient and quantum yield vary with conjugation of the dye and physical conditions
such as pH.
6.3.2 Supplement 30
Current Protocols in Cytometry
Figure 6.3.1 Staining of human peripheral blood cells with CD25 and CD3 monoclonal antibodies detected with fluorescein-conjugated anti–mouse Ig (left) compared with high-sensitivity procedure (right). Photomultiplier voltages were set to restrict nonspecific staining (IgG1 control) to the first decade. The high-sensitivity procedure showed less nonspecific staining in this case, with significantly brighter staining of CD3 and CD25. This is seen particularly in the greater % positive and in the separation of positive and negative populations with CD25 (results from Ms. Natasha Wuttke). MFI, mean fluorescence intensity.
in two-color experiments where sensitivity is not a requirement, so they may be inadequate for high-sensitivity work. The investigator must compare reagents and select a set that maximizes the separation between the signal from positive cells and the signal from negative cells (Figs. 6.3.1 and 6.3.2). Careful attention to minimizing nonspecific fluorescence allows the potential benefits of improved staining or instrument amplification to be realized. By examining controls lacking individual staining reagents, it is possible to identify the major contributors to nonspecific background (e.g., autofluorescence, uptake of antibodies through Fc receptors, or cross-reactivity of antibodies). Once the major contributors are identified, strategies can be devised to reduce nonspecific staining. Standard analytical flow cytometers or sorters that use an air-cooled, 15-mW, 488-nm laser are capable of adequate sensitivity. A quartz flow cell provides higher sensitivity than does a stream-in-air system. Use of a dye-pumped solid-state (DPSS) laser emitting at 532 nm can improve sensitivity, because the excitation matches the absorption spectrum of PE better than the 488-nm argon-ion laser does (Table 6.3.2). To analyze PE as a single fluorochrome, the narrow-band filters used to separate the emissions of FITC, PE, and other dyes are not needed, and use of a broader-band filter may increase sensitivity. As in the case of increased amplification achieved with multilayer reagents, increases in signal strength that do not improve the separation between specific and nonspecific staining are of no value.
Phenotypic Analysis
6.3.3 Current Protocols in Cytometry
Supplement 30
Figure 6.3.2 Improved sensitivity using samples similar to those in Figure 6.3.1. Samples were stained by the high-sensitivity procedure and were read using a 488-nm argon ion laser and a 532-nm DPSS laser (see text). DPSS excitation yields a significantly greater sensitivity (results from Ms. Natasha Wuttke). MFI, mean fluorescence intensity.
BASIC PROTOCOL
SINGLE-COLOR STAINING AND ANALYSIS Single-color staining and analysis permit the highest sensitivity and should be used unless additional parameters are required to identify the cell population of interest.
Materials Cell suspension: heparinized whole blood or Ficoll-Hypaque-purified cells (mononuclear cell fraction or suspension of tissue cells; UNITS 5.1 & 9.7) DPBS/azide, ice cold: Dulbecco’s PBS (UNIT 9.8) containing 0.02 M sodium azide (or other isotonic buffer with same concentration of azide) Pretitrated monoclonal antibody (MAb) specific for antigen of interest (see Support Protocol 1) in DPBS/azide, ice cold Isotype control in DPBS/azide at same concentration as MAb of interest, ice cold 2:1 (v/v) normal horse serum/normal human serum (or normal serum from appropriate species mixed with normal human serum; see step 6 below) Pretitrated biotinylated anti–mouse Ig (e.g., Vector Labs; see Support Protocol 1 for titration) in DPBS/azide, ice cold Pretitrated phycoerythrin-streptavidin (PE-SA; Caltag Labs; Sigma; see Support Protocol 1 for titration) in DPBS/azide, ice cold Lysing solution (Becton Dickinson Immunocytochemistry) High-Sensitivity Detection of Low-Abundance Membrane Molecules
12 × 75–mm (3-ml) flow cytometry tubes Refrigerated centrifuge (ideally with a 64-tube-capacity rotor) Additional reagents and equipment for counting cells (APPENDIX 3A) and flow cytometry (see Chapter 1)
6.3.4 Supplement 30
Current Protocols in Cytometry
NOTE: Sodium azide reacts with copper in plumbing to form an explosive compound. Check institutional safety regulations concerning disposal of azide-containing liquids. NOTE: Maintain cells cold throughout by keeping reaction tubes and reagents (except normal horse serum/normal human serum) in melting ice and setting the centrifuge at 4◦ C.
Perform primary antibody labeling 1. Count cells (APPENDIX 3A) in whole blood or Ficoll-Hypaque-purified cells. Suspend at 107 cells/ml in ice-cold DPBS/azide. 2. Add the appropriate amount of ice-cold pretitrated MAb and isotype control to separate 12 × 75–mm 3-ml flow cytometry tubes. A final MAb concentration of 5 µg/ml (or 50 µl of undiluted culture supernatant or 1/100 dilution of ascitic fluid added to 50 µl of cell suspension) should be suitable. However, the optimal concentration must be titrated prior to the assay (see Support Protocol 1). The use of an isotype-matched negative control is essential, but does not guarantee that staining is specific, because nonspecific binding is affected by factors such as aggregation or denaturation during purification or storage.
3. Mix cell suspension and add 50 µl to each tube. Mix by gentle vortexing and incubate 30 min on ice, resuspending settled cells once at 15 min. 4. Add 3 ml ice-cold DPBS/azide and centrifuge 5 min at 200 × g, 4◦ C. 5. Remove supernatant by gentle aspiration, being careful to remove as much liquid as possible without disturbing the cell pellet. Resuspend cells in 100 µl ice-cold DPBS/azide, add 3 ml ice-cold DPBS/azide, and vortex gently. Repeat washing step, removing the final supernatant.
Perform indirect fluorescent labeling 6. Resuspend cells thoroughly in 50 µl ice-cold DPBS/azide and add 15 µl of 2:1 (v/v) normal horse serum/normal human serum. Incubate 10 min at room temperature, then return tube to ice/water. The normal serum is added to block Fc receptors and reduce nonspecific binding of the reagents that follow. If the anti-mouse reagent is made in a species other than horse, use blocking serum for the appropriate species. Fc-mediated uptake can also be avoided by using F(ab)’2 preparations, but these reagents tend to give lower sensitivity.
7. Add 50 µl ice-cold pretitrated biotinylated anti–mouse Ig to the cells and mix gently. Incubate and wash twice as above (see steps 3 to 5). Horse anti–mouse Ig can be used at a 1/100 dilution, but a preliminary titration (see Support Protocol 1) is necessary for each new batch or after prolonged storage (>1 month).
8. Add 50 µl ice-cold pretitrated PE-SA to the cells and mix gently. Incubate and wash twice as above (see steps 3 to 5). Sigma PE-SA can be used at a 1/20 dilution, but a preliminary titration (see Support Protocol 1) is necessary for each new batch or after prolonged storage (>1 month).
9. Optional (for whole-blood samples): To lyse erythrocytes, add 2 ml lysing solution, vortex briefly, and allow to stand 10 min at room temperature in the dark. Centrifuge 5 min at 200 × g, and remove supernatant. Wash once with 1 ml DPBS/azide. 10. Resuspend cells thoroughly in 100 to 200 µl ice-cold DPBS/azide. If analysis is to be delayed until another day, resuspend in 200 µl flow cytometry fixative (Lanier and Warner, 1981) and store up to 7 days at 4◦ C.
Phenotypic Analysis
6.3.5 Current Protocols in Cytometry
Supplement 30
Analyze cells 11. Analyze by flow cytometry (see Chapter 1). Select cell population of interest by gating on forward and side scatter, and print out fluorescence-intensity histograms (see Fig. 6.3.1 for examples). ALTERNATE PROTOCOL
MULTICOLOR STAINING AND ANALYSIS In many situations, the expression of markers in a mixed cell population must be analyzed. While the cell subset of interest can first be purified and then examined by single-color analysis, a more elegant solution is to use multicolor analysis. One or more colors are used on cell-subset markers to identify the cell populations, and a different color is used to measure the expression of the marker of interest. The major factor in planning such a study is to reserve the highest-sensitivity label (phycoerythrin) for the low-abundance marker. A number of directly labeled cell-specific reagents are commercially available. Fluorescein- or PE/cyanine 5 (Cy5)–conjugated monoclonal antibodies with specificities for lymphocyte subsets (e.g., anti-CD3, anti-CD4, anti-CD8, anti-CD19) and other cell populations are available in good quality from various suppliers (e.g., Caltag Labs, BD/PharMingen, Beckman Coulter, Sigma). Becton Dickinson Immunocytometry supplies subset markers labeled with peridin chlorophyll protein (PerCP). PerCP and PE/Cy5 energy transfer reagents (e.g., Tricolor from Caltag Labs, CyChrome from BD/PharMingen, PE/Cy5 from Beckman Coulter, or Quantum Red from Sigma) are better than PE/Texas Red combinations, because there is less spectral overlap with PE. Nevertheless, electronic compensation for fluorescence spectral overlap must be set carefully to avoid reducing sensitivity (UNIT 6.14). Recently, allophycocyanin (APC)–MAb reagents have also become available from several manufacturers. These can be used once the flow cytometer is equipped with a red laser. The range of antibodies conjugated to the Alexa dyes is at present limited, but increasing.
Additional Materials (also see Basic Protocol) Normal mouse serum or 1 mg/ml normal mouse Ig (Sigma) in DPBS/azide (store refrigerated), ice cold Pretitrated, direct fluorochrome-conjugated, cell-specific MAb(s) (see introduction; see Support Protocol 1 for titration) 1. Follow steps 1 to 7 of Basic Protocol. 2. Add 50 µl ice-cold normal mouse serum or 50 µl of 1 mg/ml normal mouse Ig and incubate 10 min on melting ice. The purpose of this step is to block sites on the biotinylated anti–mouse Ig that may bind subsequent monoclonal antibodies.
3. Add 5 µl ice-cold pretitrated, direct fluorochrome-conjugated, cell-specific MAb(s) and 50 µl PE-SA to the cells and mix gently. Incubate and wash twice (see Basic Protocol, steps 3 to 5). Cell-specific MAbs and PE-SA can be added together. Sigma PE-SA can be used at a 1/20 dilution, but a preliminary titration (see Support Protocol 1) is necessary for each new batch or after prolonged storage (>1 month). Streptavidin labeled with PE/Cy5 can be used in place of PE-SA (see Strategic Planning; see Critical Parameters). High-Sensitivity Detection of Low-Abundance Membrane Molecules
4. Perform steps 9 (optional) and 10 (see Basic Protocol). 5. Analyze by flow cytometry (see Chapter 1). Select cell population of interest by gating on forward and side scatter, and print out two-color dot diagrams. Alternatively, gate
6.3.6 Supplement 30
Current Protocols in Cytometry
Figure 6.3.3 Two-color staining and analysis. High sensitivity staining of Fas (CD95) and analysis of expression on B cells (CD19), T cells (CD3), and T cell subpopulations (CD4 and CD8). Neg, negative control.
on dual scatter and one or two fluorescence parameters, and print out fluorescence intensity histograms (see Fig. 6.3.3 for examples).
TITRATION AND QUALITY CONTROL OF MONOCLONAL ANTIBODIES AND DETECTION REAGENTS
SUPPORT PROTOCOL 1
The primary monoclonal antibody (MAb) and the detection reagents must be titrated to determine optimal concentrations. This is done by performing staining and analysis with varied concentrations of MAb and detection reagents. Thus, reagents and specific methods are as described (see Basic Protocol). Modifications for titration are described below. Note that the use of an isotype-matched negative control is an essential but not necessarily adequate control, because one antibody can give higher nonspecific binding than another for reasons that are difficult to control, such as aggregation or partial denaturation during purification or storage. Phenotypic Analysis
6.3.7 Current Protocols in Cytometry
Supplement 30
MAb Titration Prepare cells (see Basic Protocol, step 1). Make dilutions of primary monoclonal antibody in DPBS/azide. As a guide, start with ∼5 µg/ml if the antibody concentration is known. For unpurified ascitic fluid, a dilution of 1/100 is a reasonable starting point. When using culture supernatant as antibody, undiluted supernatant is likely to be adequate. Use three to five doubling dilutions to cover the likely range of antibody concentrations. Add cell suspension to tubes (see Basic Protocol, step 3) before adding MAb, then add 50 µl MAb/tube directly to the cell suspension, mix gently, and proceed with the incubation and remaining procedure (steps 3 to 11). Select the optimal concentration on the basis of the best separation between positive and negative populations when the antibody gives a bimodal distribution (Figs. 6.3.1 and 6.3.2). When the antibody does not yield a bimodal distribution, select on the basis of the best discrimination between the antibody and an equivalent concentration of an isotype-matched negative control antibody.
Titration of Detection Reagents To test and titrate detection reagents, use three sets of tubes: a known strong antibody that will resolve positive and negative populations (e.g., CD3), a known antibody that produces low-level staining with a bimodal distribution (e.g., CD25), and a known negative control. While the test can be done as a checkerboard titration to determine optimal concentrations of both the biotinylated anti–mouse Ig and the PE-SA, it is more efficient to vary one reagent at a time. Prepare cells and perform MAb labeling (see Basic Protocol, steps 1 to 6). Make dilutions of biotinylated anti–mouse Ig and PE-SA in DPBS/azide, using the manufacturer’s recommendation as a starting point and adding 2-fold higher and lower concentrations. Use these dilutions to perform indirect labeling and analysis (see Basic Protocol, steps 7 to 11). A typical set of results is shown for human peripheral blood lymphocytes (PBLs) in Figure 6.3.1. This type of analysis provides an important quality control tool, to compare reagent batches, to check reagent activity after storage, and to check instrument performance. The CD3 pattern gives a measurement of overall sensitivity (the separation between the positive and negative peaks), while comparison of CD25 with the negative control provides confirmation that the assay is operating at the necessary level of sensitivity. The CD25 pattern should give a clear shoulder of positive cells. SUPPORT PROTOCOL 2
High-Sensitivity Detection of Low-Abundance Membrane Molecules
CELL SORTING WITH HIGH SENSITIVITY Staining protocols for high-sensitivity analysis can be translated directly to sorting applications, and modern sorters are capable of sensitivity equivalent to that of analytical cytometers. In principle, therefore, there are no difficulties associated with translating a high-sensitivity analytical application to preparative use. In practice, there are a number of variables that can cause difficulty. The first is the concentration of reagents used in staining. Because sorting applications normally involve staining a relatively large number of cells (typically 107 or more, compared to 0.5 × 106 generally used for analysis) there is a temptation to avoid using the same ratio of reagent to cells, to economize on antibodies. However, because this can lead to lower sensitivity, it is important to titrate the reagents to determine the optimum concentrations (see Support Protocol 1). A second potential source of reduced sensitivity derives from the relatively long sorting times, compared with analysis. Antibody may dissociate with time, particularly at higher temperatures, and it may be essential to cool the sample tube during sorting. Stream-in-air optical systems sacrifice some signal due to the air interfaces, and quartz cuvettes are better for sensitivity. Quartz cuvettes are used almost universally for analytical
6.3.8 Supplement 30
Current Protocols in Cytometry
instruments but not on all sorters, and stream-in-air optics are universally used for highspeed sorting. The best approach is to optimize sensitivity on an analyzer, then translate the application as far as possible to the sorter. If sensitivity is not adequate, contact the manufacturer of the sorter for advice—the major companies are aware of the sensitivity issue and can provide help with instrument optimization.
COMMENTARY Background Information Cytokine receptors are complex structures composed of two or more protein chains. One protein confers cytokine-binding specificity while a second protein, which may be shared by several other cytokines, increases binding affinity and mediates signal transduction (Taga and Kishimoto, 1992; Noguchi et al., 1993). Cytokine receptors can be expressed on activated cells at very high levels (>30,000 molecules/cell for CD25), but they are usually expressed at much lower concentrations (100 to 500 molecules) in cells taken directly from blood (Zola et al., 1990, 1995; Zola, 1995). Monoclonal antibodies are available against many cytokine receptor proteins (Schlossman et al., 1995; Zola, 1995; Kishimoto et al., 1997; Mason et al., 2002). Receptor expression may be analyzed using the labeled cytokine as ligand. This approach may give a different result from analysis using monoclonal antibody, since the cytokine will usually bind to the multichain receptor complex, while the antibody detects individual proteins. Other molecules that may be expressed and functional at low concentrations include Fc receptors (Mantzioris et al., 1993) and costimulatory molecules such as B7 and CTLA-4. The technique described in this protocol is also useful for detection of transfectants expressing low levels of transgene, and for detection of low-affinity binding.
Critical Parameters Maintain cells and reagents cold throughout the procedure to reduce loss of antigen and dissociation of the antibody-antigen complex. All reagents should be stored refrigerated, and the reaction tubes should be in contact with melting ice. Centrifugation should be carried out at 4◦ C. Blocking of Fc receptors (see Basic Protocol step 6), however, is performed at room temperature to increase the rate of binding. After centrifugation remove supernatant thoroughly, carefully, and reproducibly. If the amount of supernatant left behind after washing varies from tube to tube, the next
reagent will be diluted to a variable degree, introducing variability in staining. Resuspend cells after each centrifugation, to ensure a single-cell suspension. Careful use of a vortex mixer is recommended. Excessive mixing may denature protein reagents. Although the recommended protocols specify two washes at each stage, it may be worthwhile experimenting with fewer washes, particularly with low-affinity antibodies. Sensitivity As has been emphasized (see Strategic Planning), phycoerythrin (PE) is the fluorochrome that gives the highest sensitivity. The use of a three-step staining protocol provides amplification by increasing the number of fluorochrome molecules per antigen epitope. Different staining reagents that are equivalent in principle may differ greatly in sensitivity, and reagents may lose activity on storage. Titration and quality control are therefore essential. Instrument parameters can also have a major influence on sensitivity (see Strategic Planning). Sensitivity is not improved simply by increasing the signal strength, as can be demonstrated very simply by turning up the photomultiplier voltage. Nonspecific (background) staining levels must be minimized to achieve sensitivity. The major sources of background staining are autofluorescence, cross-reactivity of anti-Ig reagents with cell membrane Ig, and binding of antibodies to Fc receptors or other membrane components. Strategies for reducing Fc-mediated staining have been described in the protocols. Cross-reactivity of anti–mouse Ig with human Ig may be removed by absorption on immobilized human Ig. Include controls to determine the major contributors in a particular situation, and then take steps to minimize the major contributors. Effect of fluorescence compensation on sensitivity Electronic compensation must be carried out carefully (UNITS 1.3 & 1.14). Some loss of sensitivity is generally experienced in multicolor compared with single-color analysis. The
Phenotypic Analysis
6.3.9 Current Protocols in Cytometry
Supplement 30
emission spectrum of fluorescein has a substantial “tail” that overlaps the spectral window used to collect PE signals. Electronic correction for this involves some reduction in the PE signal; if the compensation is perfect the reduction will exactly match the signal from fluorescein and will not reduce the genuine PE signal, but in practice it is difficult to achieve this precisely. In principle, digital fluorescence compensation, available on some instruments, should be more precise. The spectral overlap between PE and PE/Texas Red is also considerable, but that between PE and PE/Cy5 is small. This combination should therefore be chosen where possible, but PE/Cy5 conjugates need to be checked to see if they emit in the PE spectral window, which can happen if there is
High-Sensitivity Detection of Low-Abundance Membrane Molecules
any free PE. To avoid spectral overlap, use the 488-nm laser to excite FITC while using a 532-nm dye-pumped solid-state (DPSS) laser to excite PE. Although each dye will be excited by both lasers, the excitation of FITC at 532 nm will be very low. The signals have to be measured separately and then combined for correlated dual-color analysis, which can be achieved by separating the laser/stream intersection points and using gated amplifiers. Simultaneous analysis of two low-abundance molecules If two low-abundance molecules are to be measured together, use PE and PE/Cy5 as the fluorochromes, because they give the strongest signals and require minimal compensation (but
Figure 6.3.4 Three-color staining and analysis, including two high-sensitivity colors. Cells were stained directly with FITC-conjugated anti-CD4 antibody and PE-conjugated anti-Fas (CD95) antibody (DX-2; BD/PharMingen), and indirectly with anti-CD25 detected with biotinylated horse anti-mouse (Vector Labs) and streptavidin-PE/Cy5 (Sigma). The bottom panel shows CD4 cells with a major population coexpressing CD25 and Fas (results from Ms. Natasha Wuttke).
6.3.10 Supplement 30
Current Protocols in Cytometry
see note about checking PE emission from PE/Cy5, in above discussion on effect of fluorescence compensation). Alternatively, combine PE with one of the Alexa dyes, or use two different Alexa conjugates. The biotin-avidin amplification system can be used for only one of the markers, and is best reserved for the weaker one. A second indirect system can be used, such as that based on digoxigenin, but the availability of reagents usually means that one of the markers will have to be detected with a direct conjugate. In this situation it is worth exploring alternative reagents within the same specificity; for example, Becton Dickinson Immunocytometry’s CD25-PE conjugate and BD/PharMingen’s CD95-PE conjugate give exceptionally good staining.
Troubleshooting Loss of sensitivity is probably due to deterioration of one of the reagents, and they should be retitrated (see Support Protocol 1). A high background signal, usually in the form of a “tail” to the negative population in the nega-
tive control, is probably due to Fc-mediated uptake or cross-reactivity (see Critical Parameters). Identify the cause of the problem and remedy as described in the protocols. Not infrequently, a batch of negative control antibody will start showing nonspecific binding that had not been seen previously. This may result from aggregation or other denaturation and is best remedied by using a new batch.
Anticipated Results Cytokine receptors are still commonly referred to in the literature as absent from cells that have not been activated in vitro. As a result, many immunologists would anticipate negative results with cells taken directly from tissue. However, as may be seen in Figures 6.3.1, 6.3.2, 6.3.3, and 6.3.4 and in a number of publications (Zola and Flego, 1992; Zola et al., 1993a,b), optimization for low-abundance antigens has shown that a number of receptors clearly stain subsets of lymphocytes. Furthermore, the level of staining may have clinical significance (Zola et al.,
Figure 6.3.5 Application of high-sensitivity fluorescence outside the cytokine receptors area: detection of low-avidity, single-chain antibody fragment. Rat thymocyte population was stained with negative control antibody (top panels), scFv (single chain variable fragments) derived from the anti-CD4 antibody (middle panels), and with anti-CD4 antibody (lower panels). While the highsensitivity method did not provide any advantage for staining with whole antibody, it showed clear staining with scFv, which could not be discerned clearly using conventional staining reagents (results from Dr. Michael Thiel). MFI, mean fluorescence intensity.
Phenotypic Analysis
6.3.11 Current Protocols in Cytometry
Supplement 30
1994; Borvak et al., 1995). The value of this method is not restricted to cytokine receptors. It has been used to demonstrate CD32 on T cells (Mantzioris et al., 1993), to detect lowlevel expression of transfected genes, and to detect binding of relatively low-affinity, genetically engineered antibody fragments (Nicholson et al., 1997; Fig. 6.3.5). The measures described above to reduce nonspecific staining usually work well for T cells and less well for B cells, while monocytes tend to produce some nonspecific staining that is difficult to remove entirely, resulting in a small “tail” of positive cells in the negative control.
Time Considerations Allow 1 hr for cell preparation. During cell preparation, label reaction tubes, place in rack in melting ice, and aliquot monoclonal antibodies ready for use. Allow 2.5 hr for the performance of the test, once the cells are ready. The whole procedure, starting from a blood or tissue sample, therefore takes 3.5 hr, excluding flow cytometry (see Chapter 5 for time considerations). The three protocols (single-color, multicolor, and titration) are roughly similar with respect to time. The size of the test (number of samples) will affect the time and may lead to significant variations in incubation periods. The author generally limits a test to 50 to 60 samples. The protocols are all readily performed by one person.
Literature Cited Ben Aribia, M.H., Moire, N., Metivier, D., Vaquero, C., Lantz, O., Olive, D., Charpentier, B., and Senik, A. 1989. IL-2 receptors on circulating natural killer cells and T lymphocytes. Similarity in number and affinity but difference in transmission of the proliferation signal. J. Immunol. 142:490-499. Bikoue, A., George, F., Poncelet, P., Mutin, M., Janossy, G., and Sampol, J. 1996. Quantitative analysis of leukocyte membrane antigen expression: Normal adult values. Cytometry 26:137147. Borvak, J., Chou, C.-S., Bell, K., van Dyke, G., Zola, H., Ramilo, O., and Vitetta, E.S. 1995. Expression of CD25 defines peripheral blood mononuclear cells with productive versus latent HIV infection. J. Immunol. 155:3196-3204.
High-Sensitivity Detection of Low-Abundance Membrane Molecules
Dighiero, G., Bodega, E., Mayzner, R., and Binet, J.L. 1980. Individual cell-by-cell quantitation of lymphocyte surface membrane Ig in normal and CLL lymphocytes and during ontogeny of mouse B lymphocytes by immunoperoxidase assay. Blood 55:93-100. Dower, S.K., Kronheim, S.R., March, C.J., Conlon, P.J., Hopp, T.P., Gillis, S., and Urdal, D.L. 1985.
Detection and characterization of high affinity plasma membrane receptors for human interleukin 1. J. Exp. Med. 162:501-515. Kishimoto, T. 1989. The biology of interleukin-6. Blood 74:1-10. Kishimoto, T., Kikutani, H., von dem Borne, A.E.G., Goyert, S.M., Mason, D.Y., Miyasaka, M., Moretta, L., Okumura, K., Shaw, S., Springer, T.A., Sugamura, K., and Zola, H. 1997. Leucocyte Typing VI: White cell differentiation antigens. Garland Publishing, New York. Lanier, L.L. and Warner, N.L. 1981. Paraformaldehyde fixation of hematopoietic cells for quantitative flow cytometry (FACS) analysis. J. Immunol. Methods 47:25-30. Le Mauff, B., Gascan, H., Olive, D., Moreau, J.F., Mawas, C., Soulillou, J.P., and Jacques, Y. 1987. Parameters of interaction of a novel monoclonal antibody (33B3.1) with the human IL2-receptors: Interrelationship between 33B3.1, anti-Tac, and IL2 binding sites. Hum. Immunol. 19:53-68. Lidke, D.S., Nagy, P., Heintzmann, R., Arndt-Jovin, D., Post, J.N., Grecco, H.E., Jares-Erijman, E.A., and Jovin, T.M. 2004. Quantum dot ligands provide new insights into erbB/HER receptormediated transduction. Nat. Biotech. 22:198203. Lowenthal, J.W., Castle, B.E., Christiansen, J., Schreurs, J., Rennick, D., Arai, N., Hoy, P., Takebe, Y., and Howard, M. 1988. Expression of high affinity receptors for murine interleukin 4 (BSF-1) on hemopoietic and nonhemopoietic cells. J. Immunol. 140:456-464. Mantzioris, B.X., Berger, M.F., Sewell, W., and Zola, H. 1993. Expression of the Fc receptor for IgG (FcgammaRII/CDw32) by human circulating T and B lymphocytes. J. Immunol. 150:51755184. Martin, P.J., Longton, G., Ledbetter, J.A., Newman, W., Braun, M.P., Beatty, P.G., and Hansen, J.A. 1983. Identification and functional characterization of two distinct epitopes on the human T cell surface protein Tp50. J. Immunol. 131:180185. Mason, D.Y., Andr´e, P., Bensussan, A., Buckley, C., Civin, C., Clark, E.A., de Haas, M., Goyert, S., Hadam, M., Hart, D., Horejs´ı, V., Meuer, S., Morrissey, J., Schwartz-Albiez, R., Shaw, S., Simmons, D., Uguccioni, M., van der Schoot, E., Vivier, E., and Zola, H. 2002. CD antigens 2001. Blood. 99:3877-80. Mavrangelos, C., Swart, B., Nobbs, S., Nicholson, I.C., Macardle, P.J., and Zola, H. 2004. Detection of low-abundance membrane markers by immunofluorescence: A comparison of alternative high-sensitivity methods and reagents. J. Immunol. Methods 289:169-78. Nicholson, I.C., Lenton, K.A., Little, D.J., DeCorso, T., Lee, F.T., Scott, A.M., Zola, H., and Hohmann, A.W. 1997. Construction and characterization of a functional single chain Fv fragment for immunotherapy of B lineage leukemia and lymphoma. Mol. Immunol. 34:11571165.
6.3.12 Supplement 30
Current Protocols in Cytometry
Noguchi, M., Nakamura, Y., Russell, S.M., Ziegler, S.F., Tsang, M., Cao, X., and Leonard, W.J. 1993. Interleukin-2 receptor gamma chain: A functional component of the interleukin-7 receptor. Science 262:1877-1880. Schlossman, S.F., Boumsell, L., Gilks, W., Harlan, J.M., Kishimoto, T., Morimoto, C., Ritz, J., Shaw, S., Silverstein, R., Springer, T.A., Tedder, T.F., and Todd, R.F. 1995. Leucocyte Typing V: White cell differentiation antigens. Oxford University Press, Oxford. Taga, T. and Kishimoto, T. 1992. Cytokine receptors and signal transduction. FASEB J. 6:3387-3396. Zola, H. 1995. Use of flow cytometry to detect cytokine receptors. In Current Protocols in Immunology (R. Coico, J.E. Coligan, E.M. Shevach, D.H. Margulies, W. Strober, and A.M. Kruisbeek, eds.) pp. 6.21.1-6.21.24. John Wiley & Sons, New York. Zola, H. and Flego, L. 1992. Expression of interleukin-6 receptor on blood lymphocytes without in vitro activation. Immunology 76:338340. Zola, H., Flego, L., and Weedon, H. 1993a. Expression of membrane receptor for tumor necrosis factor on human blood lymphocytes. Immunol. Cell Biol. 71:281-288.
Zola, H., Neoh, S.H., Mantzioris, B.X., Webster, J., and Loughnan, M.S. 1990. Detection by immunofluorescence of surface molecules present in low copy numbers. High sensitivity staining and calibration of flow cytometer. J. Immunol. Methods 135:247-255. Zola, H., Siderius, N., Flego, L., Beckman Coulter, I.G.R., and Seshadri, R. 1994. Cytokine receptor expression in leukemic cells. Leuk. Res. 18:6573. Zola, H., Fusco, M., Macardle, P.J., Flego, L., and Roberton, D. 1995. Expression of cytokine receptors by human cord blood lymphocytes— comparison with adult lymphocytes. Pediatr. Res. 38:397-403.
Key Reference Zola et al., 1990. See above. This paper describes the high-sensitivity immunofluorescence procedure in detail and shows typical results.
Contributed by Heddy Zola Child Health Research Institute North Adelaide, Australia
Zola, H., Flego, L., and Weedon, H. 1993b. Expression of IL-4 receptor on human T and B lymphocytes. Cell Immunol. 150:149-158.
Phenotypic Analysis
6.3.13 Current Protocols in Cytometry
Supplement 30
Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells
UNIT 6.4
Hematopoietic progenitor cells (HPCs) can be mobilized from the bone marrow into the peripheral blood by cytotoxic drugs, cytokines, or combinations of the two. In the majority of cases, this will allow these primitive cells to be collected by apheresis in sufficient quantities for transplantation procedures. CD34 is the first documented cell-surface antigen whose expression within the hematopoietic system is restricted to stem and progenitor cells of all lineages. Thus, CD34+ HPCs can restore multilineage hematopoiesis in myelo-ablated patients. Flow cytometric enumeration of CD34+ cells per kilogram of recipient body weight has been shown to be the most useful indicator of the hematopoietic reconstitutive capacity of peripheral blood stem cell (PBSC) transplants. As flow cytometry can produce results within an hour, serial assessments of CD34+ cell concentrations in peripheral blood also enable the optimal timing of the apheresis sessions, ensuring the harvest of sufficient numbers of CD34+ cells. As CD34+ cells occur only in low frequencies in peripheral blood and apheresis products following their mobilization (i.e., typically between 0.1% and 2%), accurate enumeration of CD34+ cells represents rare event analysis and it is not uncommon for the number of specifically stained CD34+ cells to be outnumbered by nonspecifically stained events. The Basic Protocol is a modified version of the protocol developed by Sutherland et al. (1994) that was subsequently incorporated into a set of clinical guidelines for the International Society for Hematotherapy and Graft Engineering (ISHAGE; Sutherland et al., 1996a). It features counterstaining of CD34 by the CD45 monoclonal antibody (MAb), allowing the identification of leukocytes (CD45+) and the verification of “true” CD34+ cells as being dim for CD45 fluorescence and having low side scatter (CD45dim, SSlow). This flexible approach to CD34+ cell enumeration can be applied to the widest range of specimens, including those most commonly used for transplantation (e.g., apheresis products, bone marrow, and cord blood). The addition of predefined numbers of counting beads to the cell suspension yields the concentration of CD34+ cells per unit of sample volume (i.e., the absolute CD34+ cell count) from a single flow-cytometric assessment (single-platform technique). Dead cells can be excluded from analysis using the DNA stain 7-aminoactinomycin D (7-AAD). Support Protocol 1 allows the simultaneous enumeration of both CD34+ cells and CD3+ T cells. An ability to accurately enumerate the latter can be particularly useful in the allogeneic transplant setting, where reducing the number of CD3+ cells in the graft has been shown to result in decreased severity of graft-versus-host disease (GVHD). In this modification, CD3 is added (in place of 7-AAD) as a third antibody conjugate. The modified assay can also be performed in the presence of counting beads, allowing the generation of absolute CD3+ and CD34+ cell counts from a single assay tube. Support Protocol 2 addresses the further immunophenotypic characterization of CD34+ cells. This assay is of use in studies to identify more precisely which CD34+ cell subsets may have the highest relevance for prediction of engraftment. ENUMERATION OF ABSOLUTE NUMBERS OF CD34+ CELLS Counterstaining CD34 with CD45 MAb allows the elimination of debris and nonspecifically stained events from the analysis, as well as the generation of a reliable denominator against which to measure CD34+ cells (i.e., leukocytes). Importantly, this approach also allows the discrimination of HPCs (which express relatively low levels of CD45 on their surface) from lymphocytes and monocytes (which express high levels). Just as lymphoContributed by D. Robert Sutherland, Michael Keeney, and Jan W. Gratama Current Protocols in Cytometry (2003) 6.4.1-6.4.23 Copyright © 2003 by John Wiley & Sons, Inc.
BASIC PROTOCOL
Phenotypic Analysis
6.4.1 Supplement 25
cytes, monocytes, and granulocytes form discrete clusters on bivariate plots of CD45 versus side scatter (SS), so do nonmalignant CD34+ hematopoietic stem and precursor cells. Thus, in the Basic Protocol presented here, a multiparameter flow methodology is applied that utilizes the maximum information available of four parameters—i.e., forward scatter (FS), side scatter (SS), CD34 expression, and intensity of CD45 expression. These four parameters are combined in a sequential Boolean gating strategy (UNIT 10.4) that can be used on a variety of clinical samples such as peripheral blood, apheresis products, purified CD34+ cell suspensions, cord blood, and bone marrow specimens. To enumerate absolute numbers of CD34+ cells (i.e., the concentration per unit of volume) by a single flow cytometric measurement (single-platform assay), an accurately measured volume of sample is pipetted into a tube and the same accurately measured volume of counting beads is added. This procedure establishes a ratio of counting beads to volume of sample. The known number of added counting beads allows direct calculation of the absolute CD34+ cell count. To avoid the loss of cell, and/or counting beads, lyse/no-wash sample processing is employed. These modifications convert the ISHAGE Protocol into a single-platform assay (Keeney et al., 1998a). Optional exclusion of dead cells from CD34+ cell enumeration can be achieved by inclusion of the DNA G-C intercalating dye 7-AAD, which allows discrimination between viable and nonviable (CD34+) cells. This modification is simple to apply and may be of value in all sample types. It is highly recommended for apheresis samples over 4 hr old, all “fresh” cord blood and marrow samples, or any samples that have been manipulated in any manner, and is essential for the accurate analysis of post-thawed samples, regardless of source (see below). 7-AAD is excited at 488 nm and has maximum emission at 660 nm. Consequently, the dye cannot be used in single-laser systems with other fluorochromes that emit at >600 nm, such as PerCP or PE-Cy5. See UNIT 9.2 for a discussion on compensation for spectral overlap between PE and 7-AAD. This procedure requires reverse pipetting of the sample and counting beads (see Critical Parameters and Troubleshooting). Additionally, this protocol uses terminology and gating parameters that are unique to specific instruments, i.e., Beckman Coulter XL and BD Biosciences FACS series instruments (see Fig. 6.4.1 and Fig. 6.4.2, respectively). However, the procedure can certainly be adapted for other flow cytometers. Materials Sample of interest: peripheral blood, apheresis, cord blood, bone marrow, selected CD34+ cells, or cryopreserved and thawed samples Phycoerythrin-conjugated CD34 monoclonal antibody (CD34-PE MAb), appropriately titered (UNIT 4.1) Fluorescein isothiocyanate–conjugated CD45 monoclonal antibody (CD45-FITC MAb), appropriately titered (UNIT 4.1) 1× ammonium chloride lysing solution (APPENDIX 2A) Phosphate-buffered saline (PBS; APPENDIX 2A) supplemented with 1% human serum albumen (or similar; PBS/HSA) Flow-Count counting beads (Beckman Coulter) 12 × 75–mm polypropylene tubes Flow cytometer with at least three fluorescence detectors and appropriate filter sets for detection of FITC, PE, and if required, 7-AAD Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells
Additional reagents and equipment for assessing leukocyte count (APPENDIX 3A)
6.4.2 Supplement 25
Current Protocols in Cytometry
1
2
J
1023
1023
AJ
SS
SS
A
B
E
0
0 100
3
102 101 CD45-FITC
103
100
102 101 CD34-PE
103
4 ABCJ
ABJ
SS
1023
SS
1023
C
D
0
0 100
102 101 CD45-FITC
0
103
5 103
G1
G2
G3
G4
1023
H
1023
FS
6 EJ
101
SS
CD34-PE
102
F
100
0 100
103
101 CD45-FITC
102
103
7H
0
1023
8 CAL
1023
J
101
SS
Flowcount
102
100
0
1023 Time
0 100
101 7-AAD
102
103
Figure 6.4.1 Example of the Basic Protocol combined with optional dead-cell exclusion for a Beckman Coulter XL flow cytometer equipped with four fluorescence detectors. Data acquisition was done using System 2 Software and analysis using Expo software. Data are shown from a prediluted apheresis sample containing 16% dead cells (i.e., 7-AAD+) and 300 viable CD34+ cells/µl. In this example, 180 viable CD34+ cells were gated in region D and 6754 singlet beads were gated in region H. The assayed Flow-Count bead concentration was 1046/µl and the cell dilution factor was 1/10.
Phenotypic Analysis
6.4.3 Current Protocols in Cytometry
Supplement 25
1 1000
2 1000
not R8 (G8)
800
600
600
SS
SS
800
not R8 and R1(G1)
R1
400
400 R2 R5
200
200
0
0 100
101
102 CD45-FITC
103
104
100
3
101
102 CD34-PE
104 File: PBSC080499.008 Gate: G1 Total Events: 50253
4
1000
1000 R2 and G1
R3 and G2
800
Gate G1 G2 G3 G4 G5 G6 G7 G8 G9
800 R1
SS
SS
600 400
600 R4
400
R3
200
200
0
Events % Gated 41643 100.00 491 1.18 472 1.13 456 1.10 10782 25.89 4584 11.01 4079 9.80 41643 100.00 37059 88.99
0 100
104
101
102 103 CD45-FITC
104
0
200
400
600
800
1000
FS
5
6 1000
ungated R6
not R8 and R5 (G5)
800
103
102
Stored PBSC (24 hr old) Abs viable CD34+ cells
600 SS
CD34-PE
103
400
× bead conc = G4 G7
R4
× 1046 = 456 4079
101 200
= 117/µl
100
0 100
101
102 103 CD45-FITC
104
0
200
400
600
800
1000
FS
7
8
1000
R6 (G6)
1000 800
600
600
dead cells
R8
FS
SS
800
400
400
R7
200
200 0
0 0
200
400 600 800 Time(1024.00 sec.)
1000
100
101
102 7AAD
103
104
Figure 6.4.2 Example of the Basic Protocol combined with optional dead-cell exclusion for a BD Biosciences FACScan flow cytometer. Data were acquired and analyzed using CellQuest version 3.3 software. Data are shown from apheresis collection, stored overnight at 4°C after dilution with autologous plasma. Sample (diluted 1/10 before staining) contained ∼11% dead cells (i.e., 7-AAD+ gated in R8) and 117 viable CD34+ cells/µl. In this example, 456 viable CD34+ cells were identified in gate 4 and 4079 singlet beads were enumerated in gate 7. The assayed bead concentration was 1046/µl.
Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells
Prepare sample 1. Assess the leukocyte count (APPENDIX 3A) of blood and apheresis samples to determine whether or not dilution of the sample is necessary. Dilute the sample with PBS/HSA so that the leukocyte count is ≤20 × 109/liter in a volume of 0.5 ml. 2. For each sample, label three 12 × 75–mm polypropylene tubes: two for CD34 and one for PBS.
6.4.4 Supplement 25
Current Protocols in Cytometry
3. Mix diluted sample well and add 100 µl to each of the tubes labeled CD34. Add 2 ml PBS to the PBS tube. 4. Add 10 µl CD45-FITC MAb and 10 µl CD34-PE MAb to each of the two CD34 tubes. Gently mix. 5. Incubate tubes 15 min at room temperature, protected from light. 6. Add 2 ml of 1× ammonium chloride lysing solution at room temperature to the CD34 tubes. Vortex gently. Incubate 10 to 15 min at room temperature, protected from light. Incubation for 10 min is usually sufficient to lyse erythrocytes in peripheral blood samples and apheresis products. Incubation for 15 min is recommended for cord blood samples because of the higher concentration and greater resistance to lysis of the erythrocytes. If erythrocytes are only a minor population (e.g., apheresis products or enriched CD34+ cells), lysis is not necessary and samples can simply be resuspended in 2 ml PBS/HSA. Loss of counting beads may occur when the analysis is performed on undiluted apheresis samples using polystyrene tubes in the absence of protein (Brando et al., 2001). Optional (for dead cell exclusion): Before adding lysing solution, prepare a fresh 100 ìg/ml working solution of 7-AAD by diluting a 1 mg/ml stock solution (UNIT 9.2) with PBS. Add 20 ìl working solution to 2 ml ammonium chloride lysing solution and then add this 7-AAD/lysing solution to each tube. Vortex gently and incubate 15 min at room temperature protected from light.
CAUTION: 7-AAD is a suspected carcinogen and should be handled with care. In particular, wear gloves when weighing out the dye and do not inhale dust. 7. Immediately prior to use, resuspend Flow-Count beads gently but thoroughly. This can be achieved manually by end-over-end rotation to avoid the generation of air bubbles. Do not vortex. Add 100 µl Flow-Count beads to the CD34 tubes with the same pipet used for sample dispensing and gently mix to distribute the beads evenly. If necessary Flow-Count beads should be properly resuspended before use by vortexing, especially if they have been sitting for 12 hr or more. This is best performed at least 2 hr prior to use so that any air bubbles inadvertently generated during this process will not be present when beads are pipetted into the sample. Samples without 7-AAD can be analyzed immediately or kept on melting ice for a maximum of 1 hr. Samples should be gently mixed immediately prior to analysis. Samples with 7-AAD should be analyzed immediately.
Set up instrument 8. Position FS, SS, and fluorescence (FL) windows of analysis and adjust electronic correction for spectral overlap according to standard procedures for lyse/no-wash immunophenotyping assays (UNIT 1.3). Flow-Count beads are detected on FACScan and FACSCalibur instruments at lower FS levels than lymphocytes; thus, the FS threshold must be lowered so that singlet beads are not excluded from acquisition (see Fig. 6.4.2 histogram 7).
9. Create a total of seven or eight bivariate histograms (dot plots) as follows: a. b. c. d. e.
Histogram 1: CD45-FITC (green fluorescence) versus SS Histogram 2: CD34-PE (orange-red fluorescence) versus SS Histogram 3: CD45-FITC (green fluorescence) versus SS Histogram 4: FS versus SS Histogram 5: CD45-FITC (green fluorescence) versus CD34-PE (orange-red fluorescence)
Phenotypic Analysis
6.4.5 Current Protocols in Cytometry
Supplement 25
f. Histogram 6: FS versus SS g. Histogram 7: Time versus counting-bead fluorescence (red fluorescence; for Beckman Coulter XL) or time versus FS (for BD Biosciences FACS) Optional (for dead cell exclusion): For Beckman Coulter XL instruments, acquire histogram 8 as 7-AAD (far-red fluorescence) versus SS. On Beckman Coulter XL instruments with three fluorescence detectors, replace the 620-nm red fluorescence band-pass filter with a 675-nm band-pass filter for optimal detection of 7-AAD. For BD Biosciences FACS instruments, acquire histogram 8 as 7-AAD (far-red fluorescence) versus SS.
Create gating regions In the following steps, specific terminology for regions and gates are given parenthetically for different instruments. For example, region A is for Beckman Coulter XL instruments, and region R1 and gate G1 (see below) are for BD Biosciences FACS instruments. See Figures 6.4.1 and 6.4.2. 10. Histogram 1 (leukocyte gate): Display all events. Draw a rectangular gate (A; R1) to include all CD45dim to CD45bright events and to exclude debris, platelets, and unlysed erythrocytes, which are all CD45neg. For Beckman Coulter XL instruments: By ensuring that the right edge of region A does not extend further than the brightest CD45+ cells, the Flow-Count beads (in the top right-hand corner of histogram 1) can be excluded from region A. For BD Biosciences FACS instruments: The counting beads are found in the brightest green, orange-red, and far-red channels and also exhibit very high side scatter. Since the gate statistics (Fig. 6.4.2) are obtained from events gated in R1 (displayed on histogram 2), ensure that all the CD45+ events as well as all the counting beads are gated in R1 by including the highest green fluorescence and SS channels.
11. Histogram 2 (total CD34+ gate): Display events from the leukocyte gate (i.e., gated on A; gate G1 = R1). Draw an amorphous polygon (nonrectangular) region (B; R2) to include all CD34+ events. 12. Histogram 3 (CD34+ blast gate): Display events that fulfill the criteria of both above gates (i.e., regions A and B, gated on AB; regions R1 and R2, G2 = R2 and G1). Draw an amorphous polygon region (C; R3) to include only those events that form a cluster with low to intermediate SS and CD45dim expression. 13. Histogram 4 (lymph-blast gate): Display events that fulfill the criteria of all three above gates (i.e., regions A, B, and C, gated on ABC; regions R1, R2, and R3, G3 = R3 and G2). Draw an amorphous polygon region (D; R4) to include only those events that form a cluster with low to intermediate SS and low to high FS. On BD Biosciences FACS instruments, set logical gate G4 = R4 and G3. The lymph-blast gate serves to exclude platelets and debris that may show weak nonspecific binding of CD34 and CD45 MAbs. Its lower boundaries are verified in histogram 6 (see below).
14. Histogram 5: Display ungated data. Draw a quad-stat region in order to establish the lower limit of CD45 expression by the CD34+ events. On histogram 5, draw a small rectangular region H (R6) to include the brightest events that fall in the highest green and orange-red channels. Although not readily visible, all the Flow-Count beads (e.g., singlets and aggregates) are contained in this region. Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells
15. Histogram 6 (duplicate lymph-blast gate): On histogram 1, draw an amorphous polygon region (E; R5) on the lymphocytes (CD45bright, SSlow) to create a lymphocyte
6.4.6 Supplement 25
Current Protocols in Cytometry
gate. Display the events from this region in histogram 6 (i.e., gated on E; G5 = R5). Then: a. For Beckman Coulter XL: Draw an amorphous polygon region F in histogram 6 that is similar to region D in histogram 4, creating the duplicate lymph-blast gate. Adjust region F on histogram 6 so that lymphocytes from region E are just included. Once region F has been optimized, place region D in histogram 4 similarly. b. For BD Biosciences FACS: Copy region R4 from histogram 4 into histogram 6 (creating the duplicate lymph-blast gate), and adjust the position of the duplicate so that lymphocytes from region R5 are just included. The original region R4 on histogram 4 will automatically move to the same position on histogram 4. This step establishes the minimum FS and SS range for the lymph-blast gate (F; R4) in histogram 6.
16. Histogram 7 (bead gate); gated on H (R6): a. For Beckman Coulter XL: Set a rectangular region (G) to include only single-bead events as per the manufacturer’s recommendations. Adjust counting-bead (red fluorescence) high voltage to ensure that both singlet and doublet bead populations are visible. On the Beckman Coulter XL-MCL, define region G as a CAL region to allow automatic calculation of absolute numbers of CD34+ cells. b. For BD Biosciences FACS: Display events from R6 on histogram 7 (time versus FS). If beads are not visible, lower FS threshold until singlet beads are clearly visible. Set region R7 to include only singlet beads. Set logical gate G7 = R6 and R7. Optional (for dead cell exclusion): On histogram 8 for Beckman Coulter XL, draw a rectangular region (J) to include only living (i.e., 7-AAD−) cells, and display listmode data from region J only in histograms 2 to 6. A summary of the resulting logical gates is shown in Table 6.4.1. For FACS, draw a rectangular region (R8) on histogram 8 that includes the dead cells (7-AAD+) but excludes the counting beads (that are present in the highest SS and far-red fluorescence channels). Exclude dead cells from further analysis by setting logical gates as in Table 6.4.1.
Table 6.4.1 Summary of Instrument-Specific Logical Gates Using Dead Cell Exclusion
Histogram 1 2 3 4 5 6 7 8
Beckman-Coulter XL
BD Biosciences FACS
Gated on J Gated on AJ Gated on ABJ Gated on ABCJ Ungated Gated on EJ Gated on H Ungated
G8 = not R8 G1 = (not R8) and R1 G2 = R2 and G1 G3 = R3 and G2 Ungated G5 = (not R8) and R5 G6 = R6 Ungated Phenotypic Analysis
6.4.7 Current Protocols in Cytometry
Supplement 25
Acquire data NOTE: Data analysis is performed using System 2/Expo Software for Beckman Coulter XL instruments or using CellQuest Software for BD Biosciences FACS series instruments. 17. Acquire a minimum of 100 CD34+ events (region D; G4 = R4 and G3) from the first CD34 tube. If a sensitivity level of 0.1% is sufficient, stop at 50,000 CD45+ events (region A; R1). 18. Repeat step 17 for the second CD34 tube. The total number of events to be acquired is dependent on the desired sensitivity (e.g., a level of 0.1% implies the detection of 100 CD34+ cells among a total of 100,000 CD45+ cells) in the two tubes analyzed.
19. Run the PBS tube for 1 min between two patient specimens to flush fluids and prevent specimen carryover. Flushing the system with PBS is not required between duplicates, but should be done before another pair of duplicate samples is run.
Analyze data 20. If the total number of events in region D (G4 = R4 and G3) of the two CD34 tubes is ≥100, calculate the absolute number of CD34+ cells/µl as follows (for instruments that do not calculate the absolute CD34+ cell number automatically): CD34 + cells / µl =
no. CD34 + × B × DF no. beads
where the number of CD34+ cells is determined from region D (G4 = R4 and G3), the number of singlet beads is determined from region G (G7), B is bead concentration (specified per lot), and DF is the sample dilution factor. Average the results of the two tubes. 21. If it is necessary to know the % CD34+ cells in a sample (i.e., as a fraction of leukocytes) or if the total number of events in region D (G4 = R4 and G3) of the two CD34 tubes is <100, report the CD34 result as “x% CD34+ cells,” in which
x=
100 × no. CD34 + cells no. CD45+ events
where the number of CD34+ cells is determined from region D (G4 = R4 and G3) and the number of CD45+ events is determined from region A (R1). Caution should be used in the case of samples with absolute number of CD34+ events in region D <100, as the precision of the assay will be affected by the reduction in cells enumerated.
Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells
NOTE: On BD Biosciences FACS series instruments, the counting beads are present (but not visible) in the top right corner of region R1 and must be removed from the total number of events present in R1 for accurate calculation of % CD34+ cells. This can be accomplished by manually subtracting the number of total beads (i.e., singlets and doublets, etc.) in region R6 (G6) from the number of events in R1 (G1), or by setting an additional logical gate (e.g., G9 = not R6 and R1) as shown in Figure 6.4.2.
6.4.8 Supplement 25
Current Protocols in Cytometry
As described by Brocklebank et al. (2001), it is also possible to perform the Basic Protocol using TruCOUNT absolute counting tubes (BD Biosciences), instead of Flow-Count microspheres. In this setting, due to the small size of the beads, a threshold cannot be set on forward scatter, making it necessary to set it on a fluorescence parameter. An advantage of this approach is that it eliminates the requirement to carefully suspend and pipet the counting beads. However, accurate pipetting is still required if any predilution of the sample is required, as well as for the delivery of the sample into the TruCOUNT tube. Note that some sample types, e.g., peripheral blood or cord blood, prepared by lyse/nowash sample processing, can contain a large amount of debris. By setting a gate to exclude the CD34–/CD45– events on plot 5 prior to data acquisition, a major portion of this debris can be excluded from the listmode file. An example of a data file acquired using this modification is shown in Figure 6.4.2.
QUANTITATION OF RESIDUAL CD3+ T LYMPHOCYTES IN T CELL–DEPLETED ALLOGENEIC STEM/PROGENITOR CELL GRAFTS
SUPPORT PROTOCOL 1
To simultaneously enumerate the absolute CD34+ and CD3+ cell content of a sample, a CD3 conjugate emitting in the far-red channel is added to the two-color Basic Protocol. Additional Materials (also see Basic Protocol) PE-Cy5-conjugated CD3 monoclonal antibody, appropriately titered (UNIT 4.1) NOTE: Several fluorochromes other than PE-Cy5 may be used as conjugate for the CD3 MAb. Here, PE-Cy5 is selected because of its high signal-to-noise ratio and minimal spectral overlap with PE. For data acquisition with a Beckman Coulter XL instrument with three fluorescence detectors, the use of CD3-PE-Cy5 requires replacement of the 620-nm red fluorescence band-pass filter by a 675-nm band-pass filter. On an XL with four fluorescence detectors, CD45-PE-Cy5 is detected in the far-red fluorescence channel. Prepare sample and set up instrument 1. Dilute sample as described (see Basic Protocol, step 1). 2. Label two tubes “CD34/CD3.” 3. Mix diluted sample well and add 100 µl to each tube. 4. Set up “CD34/CD3" tubes by addition of 10 µl each of CD34-PE, CD45-FITC, and CD3-PE-Cy5 MAbs. Process as described for the single-platform procedure (see Basic Protocol, steps 5 to 7). Set up instrument (see Basic Protocol, step 8). 5. Create the first seven histograms (see Basic Protocol, step 9, for parameters). In addition, create histogram 8 to display events from region A (R1) on CD3-PE-Cy5 versus CD34-PE. Create gating regions 7. Establish gates in histograms 1 to 5 and 7 (see Basic Protocol, steps 10 to 14, and step 16) For histogram 6 display events from gating region H (G8 = R1 and R8). 8. Histogram 8: a. For Beckman Coulter XL: Display data from gating region A (R1). Draw a rectangular gating region I (R8) to include all CD3+ events. Adjust logical gating so that plot 6 displays events from regions A and I (Gate 8 = R1 and R8). Phenotypic Analysis
6.4.9 Current Protocols in Cytometry
Supplement 25
1
2A
1023
1023
SS
SS
A
B
E
0
0 100
103
100
4 ABC
101 102 CD34-PE
103
1023
D
SS
SS
1023
3 AB
102 101 CD45-FITC
CD34+
C 0 100
5 G1
G2
G3
G4
0
103
0
6
FS
1023
AI
1023
H
102 SS
CD34-PE
103
102 101 CD45-FITC
101
F
CD3+
100
0 100
103
101 102 CD45-FITC
103
1023
0 FS
7H
103
8A
CAL 102 CD34-PE
Flowcount
102
101
100
100
1023
0 Time
Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells
I
101
100
101 102 CD3-PE-Cy5
103
Figure 6.4.3 Simultaneous absolute counting of CD34+ and CD3+ cells on a Beckman Coulter XL flow cytometer using Support Protocol 1. Analysis was performed on a peripheral blood sample containing 106 CD34+ cells/µl and 2588 CD3+ cells/µl. In this sample, 450 CD34+ cells and 10,950 CD3+ cells were identified in regions D and F respectively. The assayed bead concentration was 1046/µl.
6.4.10 Supplement 25
Current Protocols in Cytometry
2
1 1000
1000
all events
File: RTCELL011100.005 Gate: G1 Total Events: 69438
R1(G1)
600
600
Gate G1 G2 G3 G4 G5 G6 G7 G8 G9
SS
800
SS
800
R1
R2
400
400
200
200 R5
0
0 100
101
102 103 CD45-FITC
100
104
3
101
102 103 CD34-PE
Events % Gated 69307 100.00 3134 4.52 3067 4.43 3029 4.37 14940 21.56 7312 10.55 6910 9.97 13620 19.65 13247 19.11
104
4 1000
1000 R1 + R2 (G2)
R1 + R2 + R3 (G3) 800
800 R1
600
SS
SS
600
R3
400
R4
400
200
200
0
0
Abs CD34+ cells 100
101
102 103 CD45-FITC
104
0
600
800
1000
FS
R1 and R8 (G8)
R6
800
103
× 1012 = G4 G7 = 444/µl Abs CD3+ cells
1000
all events
× 1012 = G9 G7 = 1940/µl
600 102
SS
CD34-PE
400
6
5
104
200
R4
400 101
200
100
0 100
101
102 103 CD45-FITC
104
0
400
600
800
1000
FS
7 1000
200
104
8
R6 (G6)
R1 (G1)
800
CD34-PE
103
FS
600 400
102 R8
R7 singlet beads
200
101
100
0 0
200 400 600 800 Time (1024.00 sec.)
1000
100
101
102 103 CD3-PE-Cy5
104
Figure 6.4.4 Simultaneous absolute counting of CD34+ and CD 3+ cells on a BD Biosciences FACScan flow cytometer using Support Protocol 1. Analysis was performed on a peripheral blood sample containing 444 CD34+ cells/µl and 1940 CD3+ cells/µl. In this sample, 3029 CD34+ cells and 13,247 CD3+ cells were identified in gates G4 and G9, respectively. The assayed bead concentration was 1012/µl.
Phenotypic Analysis
6.4.11 Current Protocols in Cytometry
Supplement 25
b. For BD Biosciences FACS series: Set logical gate G9 (= G8 and R4). In this way, the absolute CD3+ cell count is derived from events fulfilling the CD45+, CD3+, and light-scatter requirements appropriate for viable CD3+ cells. Nonviable CD3+ cells, nonspecifically stained events, and debris are thereby excluded. Figures 6.4.3 and 6.4.4 show examples of this analysis on a Beckman Coulter XL and a BD Biosciences FACScan respectively. Should low viability be an issue, it is recommended that the Basic Protocol (including viability dye 7-AAD) be performed to enumerate viable CD34+ cells. Alternatively, if a four-color instrument is available, it is possible to perform a four-color assay combining CD45/CD34/CD3 and 7-AAD. However, there are instrument- and fluorochrome-specific issues that must be addressed with this cocktail that are beyond the scope of this unit.
9. Calculate the absolute number of CD34+ and CD3+ cells/µl as follows (for instruments that do not calculate the absolute CD34+ cell number automatically):
CD34+ cells / µl =
no. CD34+ (region D[G 4] × B × DF) no. beads(region CAL [G7])
The absolute CD3+ and CD34+ count from tubes 1 and 2 should be within 10% of the mean of each other. Note that for CD34+ selected samples, the numbers of residual CD3+ cells may be below precise detection. SUPPORT PROTOCOL 2
IMMUNOLOGICAL CHARACTERIZATION OF CD34+ CELLS The Basic Protocol is sufficiently flexible to allow further qualitative analyses of CD34+ cells. This protocol addresses the analyses of expression of any marker of interest other than CD45 by the CD34+ cells. Depending on the number of available detectors and the presence of more than one laser, three- or four-color techniques can be applied. Four-color assays are required for analysis of CD34+ cell subsets defined by two additional markers or for a combination of CD34+ subset analysis with viability staining (UNIT 9.2). For simplicity, this protocol describes a three-color assay for a single-laser instrument. In view of the often dim expression of the markers of interest on CD34+ cells, a lyse-and-wash technique is used to minimize background staining due to debris and unbound MAb. Therefore, reverse pipetting of the sample is not necessary. Two examples of CD34+ cell subtyping are shown in Figures 6.4.5 and 6.4.6. Figure 6.4.5 shows assessment of CD133 coexpression on a Beckman Coulter XL flow cytometer and Figure 6.4.6 shows assessment of CD90 coexpression on a BD Biosciences FACScan instrument. Additional Materials (also see Basic Protocol) Phycoerythrin-conjugated monoclonal antibody identifying marker of interest (CDx-PE MAb), appropriately titered (UNIT 4.1) Fluorescein isothiocyanate–conjugated anti-CD34 monoclonal antibody (CD34-FITC MAb), appropriately titered (UNIT 4.1) PE-Cy5-conjugated anti-CD45 monoclonal antibody (CD45-PE-Cy5 MAb), appropriately titered (UNIT 4.1) Prepare sample 1. Dilute sample as described (see Basic Protocol, step 1).
Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells
2. Set up one tube each for autofluorescence control (a) and coexpression analysis (b). 3. Mix diluted sample well and add 100 µl to each tube.
6.4.12 Supplement 25
Current Protocols in Cytometry
1
2A
1023
1023
SS
SS
A
B
E
0
0 100
103
100
3 AB
4
102 101 CD34-FITC
103
ABC
1023
D
SS
SS
1023
102 101 CD45-PE-Cy5
C
0 100
0
103
0
FS
1023
6E
5 J1
J2
J3
J4
1023
102
101
F
SS
CD34-FITC
103
102 101 CD45-PE-Cy5
100
100
102
7ABCD control I1 99.5%
103
I2 0.5%
102
101
100
0
103
CD34-FITC
CD34-FITC
103
101
I4 0.0%
I3 0.0%
100
101 102 Unstained PE
FS
1023
8 ABCD test H1 35.5%
H2 64.5%
H3 0.0%
H4 0.0%
102
101
100
103
0
100
101 102 CD133-PE
103
Figure 6.4.5 Analysis of a cord blood sample containing 1.5% CD34+ cells (100 × gate D/gate A) of which 64.5% express CD133 using Support Protocol 2. Data were acquired on a Beckman Coulter XL flow cytometer equipped with four fluorescence detectors and were analyzed using Expo software. Histogram 7 is taken from the control tube stained with CD34-FITC and CD45-PE-Cy5 and no fluorochrome on the PE channel. The gate statistics from the test (histogram 8, quadrant 2) show that 64.5% of the CD34+ cells are stained by MAb CD133-PE.
Phenotypic Analysis
6.4.13 Current Protocols in Cytometry
Supplement 25
1 1000
2 1000
all events
800
R1(G1)
800
R1
600
SS
SS
600
400
400
200
R2
200 R5
0
0 100
101
103 102 CD45-PE-Cy5
104
100
3
101
102 CD34-FITC
File: PBSC090499.001 Gate: G1 Total Events: 80955
1000 R1 + R2 (G2)
R1 + R2 + R3 (G3)
800
800
600
R1
Gate G1 G2 G3 G4 G5 G6 G7
SS
SS
600 R3
400
R4
400
200
200
Events % Gated 80113 100.00 860 1.07 847 1.06 815 1.02 13613 16.99 3 0.00 14428 18.01
0
0 100
104
101
103 102 CD45-PE-Cy5
104
0
200
400
600
800
1000
FS
6
5 1000
lymphs R5
800
103
600
R4
SS
CD34-FITC
104
4
1000
102
400 101
200 0
100 100
104
101
7
102 103 CD45-PE-Cy5
104
0
200
400
600
800
1000
FS
PBSC090499.001
104
8
PBSC090499.002
G4 or R5
File: PBSC090499.002 Gate: G1 Total Events: 80053
G4 or R5 103
CD34-FITC
103
CD34-FITC
103
R6 102
Gate G1 G2 G3 G4 G5 G6 G7
R6 102
101
101
100
Events % Gated 80007 100.00 866 1.08 839 1.05 797 1.00 12251 15.31 489 0.61 13048 16.31
100 100
101
103 102 Unstained PE
104
100
101
102 CD90-PE
103
104
Figure 6.4.6 Analysis of a peripheral blood apheresis sample with 1% CD34+ cells (fraction of CD45+ cells), of which 61% express CD90 using Support Protocol 2. Data were acquired on a BD Biosciences FACScan flow cytometer equipped with CellQuest version 3.3 software. The gate statistics of the control tube (file “.001") show that gate G6 contains 3 events or 0.0% of the total CD45+ cells. Gate G6 of the CD90-PE-stained tube (file ”.002") shows that 61% of the gated CD34+ cells express CD90.
6.4.14 Supplement 25
Current Protocols in Cytometry
4. Add 10 µl CD34-FITC MAb and 5 µl CD45-PE-Cy5 MAb to both tubes. Add 10 µl CDx-PE MAb to tube b and 10 µl PBS to tube a. Gently mix. Many markers of interest (e.g., CD90 and CD133) do not identify discrete, nonoverlapping subsets of CD34+ cells, but rather exhibit a continuous spectrum of mainly dim fluorescence signals. Therefore, PE is the fluorochrome of choice for these markers where available. NOTE: Several fluorochromes other than PE-Cy5 may be used as conjugate for the CD45 MAb, depending on instrument type and configuration. For data acquisition with a Beckman Coulter XL instrument with three fluorescence detectors, CD45-ECD (PE-Texas Red) can be used and is detected in the red fluorescence channel. Otherwise, the use of CD45-PE-Cy5 requires replacement of the 620-nm red fluorescence band-pass filter by a 675-nm band-pass filter. On an XL with four fluorescence detectors, CD45-PE-Cy5 is detected in the far-red fluorescence channel. Similarly, PerCP conjugates, detected in the far-red fluorescence channel, can be used on BD Biosciences FACS series instruments.
5. Incubate tubes 15 min at room temperature, protected from light. 6. Add 2 ml of 1× ammonium chloride lysing solution at room temperature. Vortex gently. Incubate 10 to 15 min at room temperature, protected from light. Incubation for 10 min is usually sufficient to lyse erythrocytes in peripheral blood samples and apheresis products. Incubation for 15 min is recommended for cord blood samples because of the higher concentration and greater resistance to lysis of the erythrocytes in these samples. If erythrocytes are only a minor population (e.g., apheresis products or enriched CD34+ cells), lysis is not necessary and samples can simply be resuspended in 2 ml PBS.
7. Centrifuge 5 min at 1000 × g, room temperature. Remove supernatant by pipetting and discard. Add 2.0 ml PBS, vortex gently, and repeat centrifugation and removal of supernatant. 8. Vortex gently and add 0.5 ml PBS. Analyze immediately or keep on melting ice for ≤1 hr. Gently mix the sample immediately prior to analysis. Set up instrument 9. Position FS, SS, and fluorescence windows of analysis and adjust electronic corrections for spectral overlap according to standard procedures for lyse-and-wash immunophenotyping assays (UNIT 1.3). Set color compensation between green fluorescence and orange-red fluorescence and between orange-red fluorescence and far-red fluorescence on both Beckman Coulter XL and BD Biosciences FACS series instruments (also see UNIT 1.3). Verification of appropriate compensation using cells singly stained with CD4-FITC, CD8-PE, and CD45-PE-Cy5 is recommended.
10. Create a total of eight bivariate histograms (dot plots): a. b. c. d. e.
Histogram 1: CD45-PE-Cy5 (far-red fluorescence) versus SS Histogram 2: CD34-FITC (green fluorescence) versus SS Histogram 3: CD45-PE-Cy5 (far-red fluorescence) versus SS Histogram 4: FS versus SS Histogram 5: CD45-PE-Cy5 (far-red fluorescence) versus CD34-FITC (green fluorescence) f. Histogram 6: FS versus SS g. Histogram 7: unstained (orange-red fluorescence) versus CD34-FITC (green fluorescence) h. Histogram 8: CDx-PE (orange-red fluorescence) versus CD34-FITC (green fluorescence)
Phenotypic Analysis
6.4.15 Current Protocols in Cytometry
Supplement 25
Gating regions A to F (Beckman Coulter XL) and R1 to R5 (BD Biosciences FACS) are established as for the Basic Protocol. Quadrants are positioned on histograms 7 and 8 as described below (steps 13 and 14). Region R6 (BD Biosciences FACS) is positioned on histogram 8 as described below.
Acquire data 11a. For Beckman Coulter XL: Acquire tube “a” (autofluorescence control) and collect ≥500 CD34+ cells in region D. Display lymphocytes for region E drawn in histogram 1, in histogram 7. Display events fulfilling the criteria of regions A, B, C, and D and in histogram 8. 11b. For BD Biosciences FACS series: Display events from regions R4 (G4 = R4 and G3) and R5 on histogram 7 (set logical gate G7 = G4 or R5). In this manner, both gated CD34+ cells and lymphocytes are displayed from the control tube.
12. Acquire tube b (coexpression analysis). Collect a similar number of CD34+ cells as for tube a. Analyze data Either approach shown in Figure 6.4.5 and Figure 6.4.6 can be used on any flow cytometer. For example shown in Fig. 6.4.5 To save space, histogram 7 displayed in this example is taken from histogram 8 of the control tube. 13a. Establish the vertical cursor of the quadrant shown in histogram 7 using the autofluorescence control. Set the quadrants such that the upper left quadrant (i.e., green+, orange-red-) contains ≥99% of the events from region D. 14a. Display the events that fulfill the criteria of regions A, B, C, and D (i.e., gated on ABCD, the “true” CD34+ cells) in histogram 8. Quantitate the CD133-PE expression of CD34+ cells in the upper right quadrant (green+, orange-red+) of histogram 8. For example shown in Fig. 6.4.6 To save space, histograms 1 through 7 are from the autofluorescence control tube “a”; histogram 8 is a duplicate of histogram 7 from the CD90-PE-stained tube “b.” 13b. On histogram 7, establish the quadrant such that the CD34+ cells fall in the upper left quadrant (green+, orange-red-). Set region R6 using the quadrant as a guide. Duplicate this quadrant and region R6 on histogram 8. 14b. For tube b, display the events from R4 and R5 (G7 = G4 or R5). The number of CD34+ cells expressing CD90-PE can be obtained from the gate statistics (i.e., gate G6).
COMMENTARY Background Information
Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells
Development of flow cytometry protocols for CD34+ cell enumeration Siena et al. (1989) in Milan were the first to describe a flow cytometric method to count CD34+ cells in peripheral blood. This method was initially based on mononuclear cell enrich-
ment by density-gradient centrifugation, followed by staining with the anti–class I CD34 antibody (see below) MY10 using indirect immunofluorescence. The subsequent development of a conjugated anti–class III CD34 MAb allowed the improvement of this protocol by adapting it into a direct immunofluorescence, whole-blood staining, lyse-and-wash tech-
6.4.16 Supplement 25
Current Protocols in Cytometry
nique (Siena et al., 1991). This so-called Milan protocol used simple FS-versus-SS gating to select leukocytes as denominator. An isotypematched control served to control for nonspecific binding and was used to define the positive analysis region for CD34+ cells. In the CD34stained sample, the number of events brighter than the control and exhibiting the low to intermediate SS of blast cells were counted and used as the numerator in the calculation of the percent CD34+ cells. Either 50,000 total events or a minimum of 50 CD34+ events were counted. The Nordic Stem Cell Laboratory Group (Johnsen and Knudsen, 1996) modified the Milan protocol by gating the CD34+ cells as a single bright cluster of events. The same gating region was used to analyze the isotype control sample, and any nonspecifically stained events were subtracted from the CD34 result. Bender et al. (1994) improved the level of precision in the assessment of percent CD34+ cells by counterstaining CD34-PE with CD45FITC; leukocytes were defined as CD45+. The CD45+ events were then analyzed in a similar manner to the Milan protocol using an isotype and CD34-versus-SS gating to enumerate the CD34+ cells. During the 1990s, the above-mentioned protocols proved to be insufficient for the accurate enumeration of CD34+ cells in samples of less than pristine quality. This awareness stimulated the development of alternative protocols for CD34+ enumeration. Owens and Loken (1995) designed a three-color protocol that makes use of 7-AAD to exclude dead cells, CD14-FITC to exclude monocytes, and CD34-PE to identify HPCs. Living cells (i.e., 7-AAD−) were selected and a plot of CD14-FITC versus SS was generated to exclude monocytes (which bind MAb nonspecifically through their Fc receptors). From this histogram, a third plot of CD34-PE versus SS was generated and compared to the isotype control (IgG1-PE versus SS) as in the Milan protocol. Both dim and bright CD34+ events were included in the calculation. The CD34 result was expressed as a percentage of live plus dead nucleated cells based on FS-versus-SS gating. In the Netherlands, the Foundation for Immunophenotyping in Hemato-Oncology (SIHON; Gratama et al., 1997) developed a threecolor protocol based on Laser Dye Solution 751 (LDS-751) and anti-CD14, anti-CD66e, and anti-CD34 MAbs. LDS-751 stains DNA and RNA, and gating on FS versus LDS-751 allows discrimination between nucleated cells and debris, unlysed erythrocytes, and platelets. Mono-
cytes (CD14+) and granulocytes (CD66e+) are then excluded from further analysis, after which the HPCs are identified as dim or bright CD34+ cells in a CD34-versus-SS dot plot. Nonspecific staining is analyzed using identical gate settings on a control staining in which the anti-CD34 MAb has been replaced by an isotype-control MAb, and any nonspecifically stained events are subtracted from the CD34 result. The counterstaining of anti-CD34 by antiCD45 MAb not only allows a more accurate resolution between leukocytes and other (irrelevant) cells than FS characteristics, but also contributes to the identification of the HPCs cluster, which expresses lower levels of CD45 than mature lymphocytes and monocytes (Borowitz et al., 1993). Sutherland et al. (1994) took advantage of the dim CD45 expression by the CD34+, SSlow HPC in a so-called sequential Boolean gating strategy, which formed the core of the CD34+ enumeration guidelines formulated for ISHAGE (Sutherland et al., 1996a). To qualify as true CD34+ cells, events had to stain with CD45 and CD34, and the CD45dim, CD34+ events had to form a cluster with similar FS,SS characteristics as lymphocytes and blast cells. Using fluorescent counting beads, Keeney et al. (1998a) extended this protocol to be capable of directly generating absolute CD34+ cell counts from a single flow cytometric assessment. A slightly modified version of this protocol, adapted to accommodate the recommendations of the European Working Group on Clinical Cell Analysis (Gratama et al., 1998), is presented here as the Basic Protocol. The basic method has been shown to be sufficiently flexible to include a third antigenic marker of interest to assess subsets of CD34+ cells (Stewart et al., 1995; presented in modified form in Support Protocol 2). Single- versus dual-platform assays for absolute CD34+ cell count assessments Assessments of counts per unit of sample volume (i.e., absolute counts) of immunophenotypically defined cell populations (such as CD34+ HPCs) can be obtained on a single flow-cytometric platform if a known number of fluorescent counting beads is incorporated in the assay. Assessment of the ratio between the numbers of beads and CD34+ cells counted allows direct calculation of the absolute CD34+ cell count. Single-platform assays involve lyse/no-wash staining procedures in order to prevent loss of beads and cells due to washing. Conventionally, absolute CD34+ cell counts
Phenotypic Analysis
6.4.17 Current Protocols in Cytometry
Supplement 25
were derived from a flow cytometrically assessed percent CD34+ cells within nucleated cells or leukocytes, combined with the assessment of the absolute leukocyte count from a hematology cell analyzer (i.e., a dual-platform assay). One advantage of single-platform over dualplatform techniques is reduced work load. In addition, the following three factors may contribute to a higher level of standardization between laboratories. First, single-platform techniques bypass the denominator issue—i.e., whether the percent CD34+ cells should be reported as a proportion of leukocytes (CD45+ cells), total nucleated cells (leukocytes plus nucleated red cells), or total events scattering above a simple FS threshold. This issue is relevant for the correct calculation of absolute CD34+ cell counts in dual-platform assays, as nucleated red cells (which may constitute significant populations in peripheral blood stem cell and cord blood cell transplants) will be included in the leukocyte counts to varying degrees dependent on the type of hematology cell analyzer used. Second, single-platform techniques avoid the variability arising from the hematology analyzer to derive absolute cell counts. Third, single-platform techniques avoid the inaccuracies resulting from rounding up or down low percentages of CD34+ cells in dual-platform techniques, which subsequently are used in calculations of absolute CD34+ cell counts.
Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells
Commercial assay kits for CD34+ cell enumeration Two commercial single-platform assay kits for absolute CD34+ cell counting are currently available: Stem-Kit (Beckman Coulter) and ProCOUNT (BD Biosciences). The Stem-Kit assay is based on the singleplatform ISHAGE protocol of Keeney et al. (1998a) with two additions. The first is a socalled isoclonal control for nonspecific staining (see Critical Parameters and Troubleshooting). Second, Stem-Kit also includes positive control cells termed Stem-Trol. These are fixed KG1a cells that have been modified to present the 581/class III CD34 and J33/pan-CD45 epitopes at levels similar to those found on normal CD34+ hematopoietic cells. Stem-Trol cells are presented at a predefined concentration, and in addition to serving as a “process control” can also be used to assess pipetting accuracy as part of the regular quality control regimen (also see Critical Parameters and Troubleshooting). Recently Beckman Coulter has released auto-
mated acquisition and analysis software, termed StemONE, for use on the Beckman Coulter XL instrument. This software mimics decision-making made by an operator using ISHAGE single-platform methodology. The software has been validated for all known CD34+ sample types with the exception of selected CD34+ products. The ProCOUNT assay has been developed by BD Biosciences and utilizes a proprietary nucleic acid dye (NAD) in combination with a class III CD34-PE and a pan-CD45-PerCP. The NAD serves to delineate bona fide nucleated cells from the large amounts of debris that typically remain in lysed/unwashed samples. The combined use of TruCOUNT tubes containing a calibrated number of TruCOUNT fluorescent beads in a lyse/no-wash protocol allows the absolute number of CD34+ cells to be enumerated directly from the flow cytometer. Because the beads are smaller than the smallest leukocytes, a green fluorescence channel threshold is set on events stained by the nuclear dye, rather than on forward scatter. In manual mode, 60,000 (dye+) events are acquired. CD34+ cells have to satisfy the simultaneous criteria of NAD+, CD45 dull/negative, and CD34+. An isotype control tube is used to assess the levels of nonspecific staining but any such events satisfying the above gating criteria for CD34+ cells are not subtracted from the positive analysis tube. The ProCOUNT assay is intended for use on fresh PB and apheresis samples only. BD Biosciences has also developed specialized software for semiautomated data acquisition and analysis of ProCOUNTstained samples. Using this approach, CD34+ cells can be accurately measured in most fresh PB and apheresis samples. However, the presence of platelet aggregates and other debris that stain nonspecifically with CD34-PE conjugates can cause the software to flag some fresh samples, and BD Biosciences has issued a directive to ProCOUNT users to the effect that manual gating of such flagged samples must be performed by an experienced flow cytometrist. Gutensohn et al. (1999) analyzed 90 fresh apheresis samples with the ProCOUNT system and found that 23% of them were “flagged.” Even after manual reanalysis, significant differences remained between ProCOUNT and both the German Reference and ISHAGE protocols. ProCOUNT is not recommended for the enumeration of CD34+ cells in cord blood or bone marrow because these sample types contain more dead cells and debris and have greater heterogeneity in light scatter and staining char-
6.4.18 Supplement 25
Current Protocols in Cytometry
acteristics than blood or apheresis samples. Further, as the lysing agent employed includes a fixative, ProCOUNT cannot assess viability or be used in CD34+ subset analyses on singlelaser instruments. More recently, BD Biosciences has released a series of reagents (CD45-FITC/CD34-PE pool, TruCOUNT tubes, an ammonium chloride–based lysing agent called PharM Lyse, and a ready-to-use 7-AAD preparation called ViaProbe), that can be used in conjunction with the Basic Protocol.
Critical Parameters and Troubleshooting Sample quality When fresh samples of good viability are analyzed, both the simpler techniques (i.e., Milan-Nordic) and the more elaborate techniques (i.e., SIHON, ISHAGE, and ProCOUNT) for CD34+ cell enumeration usually generate similar results (Gratama et al., 1998). However, all too often samples are in suboptimal condition (e.g., samples shipped overnight, cryopreserved/thawed apheresis or cord blood samples, manipulated products for transplantation containing purified CD34+ cells, and/or purged samples). Such products all contain varying numbers of dead or dying CD34− cells, dead or dying CD34+ cells, platelet aggregates, and/or cellular and other debris, all of which constitute significant obstacles to reliable CD34+ cell enumeration. In this context it is relevant that in the study by Allen et al. (2002), enumeration of viable CD34+ cells showed an improved correlation with engraftment kinetics in comparison to total (i.e., living plus dead) CD34+ cells. Choice of CD34 MAb and fluorochromes The heavy glycosylation of the CD34 molecule has important implications for the choice of appropriate CD34 MAb for CD34+ cell enumeration. The differential sensitivity of the CD34 molecule to cleavage with either neuraminidase and/or a mucin-selective protease established three broad classes of epitopes. Class I epitopes are totally or partially sensitive to neuraminidase and also sensitive to glycoprotease, while class II epitopes are sensitive only to glycoprotease, and class III epitopes are insensitive to both enzymes (reviewed by Sutherland and Keating, 1992). For double-marker analysis of CD34+ cells, CD34 MAbs conjugated with the brightly emitting dye PE are recommended in order to facilitate the identification of the rare CD34+
cells. While PE-labeled CD34 antibodies to class II and class III epitopes generate equivalent results in two-color analyses, the choice of FITC-conjugated antibodies for use in three- or four-marker assays should be restricted to anti– class III MAbs. Class III CD34 MAbs generate equivalent results in direct comparative analyses whether conjugated to FITC or PE. However, the use of class II CD34 MAbs labeled with FITC is discouraged because these reagents have greatly reduced specific binding affinity. That situation is caused by the facts that (1) the epitopes detected by class II MAbs are surrounded by the negatively charged sugars and (2) conjugation of class II MAbs with the negatively charged FITC dye also renders them negatively charged (Sutherland et al., 1996a,b). Finally, class I CD34 MAbs generate the most aberrant data on clinical samples because their sugar-dependent epitopes are not universally expressed on all CD34 glycoforms. As most class I CD34 MAbs also lose binding affinity after conjugation with charged fluorochromes such as FITC, they should not be used for CD34+ cell enumeration. In triple-marker analyses in which the antigen of interest on CD34+ cells shows relatively weak expression, cocktails of CD34-FITC, CD45-PE-Cy5, and CDx-PE can be used with confidence (see Support Protocol). For example, the PE- and FITC-labeled class III antibodies 8G12 (BD Biosciences) and 581 (Beckman Coulter) work well from batch to batch. As different suppliers use various fluorescence-toprotein ratios for each MAb, all MAbs should be carefully evaluated before routine use. The same recommendation holds for PE-Cy5-labeled CD34 MAbs: overconjugation of PE with Cy5 moieties can lead to nonspecific binding of such conjugates to Type I Fc receptors (CD64) on monocytes, and to a lesser extent on neutrophils and B lymphocytes. Currently, the 581 PE-Cy5-labeled MAb (Beckman Coulter) shows an acceptable level of background staining (Sutherland and Keeney, unpub. observ.). Choice of CD45 MAb Pan-CD45 MAbs, which detect not just all isoforms but all glycoforms of CD45, are required. Clones HLE-1 (BD Biosciences) and J33 (Beckman Coulter) are useful antibodies. Although the pan-CD45 MAb ALB12 (Beckman Coulter) detects all isoforms of CD45, it should not be used for primary gating purposes since it detects a sialic acid–dependent epitope that is not invariantly expressed and, indeed, is expressed at low to undetectable levels on
Phenotypic Analysis
6.4.19 Current Protocols in Cytometry
Supplement 9
CD34+ cells (Sutherland, unpub. observ.). As CD45 is brightly expressed by lymphocytes and monocytes, FITC conjugates are generally used in two-marker CD34+ cell enumerations (e.g., the Basic Protocol). For three- or fourmarker analyses, J33-PE-Cy5 can be used with confidence. Interestingly, the increased nonspecific binding characteristics of some other CD45-PE-Cy5 conjugates to monocytes actually makes the identification of true CD34+ cells easier for the analysis using the sequential Boolean gating strategy of the Basic Protocol, as such monocytes will be located even further to the right of true CD34+ cells in histogram 3 (Figs. 6.4.5 and 6.4.6).
Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells
Antibody controls for nonspecific staining A major point of concern with the use of isotype- and fluorochrome-matched MAbs to control for nonspecific staining is that they are not a true representation for the anti-CD34 MAb used. Their methods of purification and their specific protein contents are often different. There is no methodology to verify that a given control MAb is truly representative for any MAb of interest. The use of isotype controls may even contribute to the generation of erroneous results. This point is illustrated by the following scenarios, in which the control MAb shows lower or higher nonspecific binding characteristics than the CD34 MAb, respectively. First, placement of the fluorescence marker based on the isotype-control staining (such as in the Milan protocol) has been shown to be problematic on blood and cord blood samples containing platelet aggregates that nonspecifically bind CD34 MAb but not the fluorochrome- and isotype-matched control MAb from the same commercial source (Sutherland et al., 1997). The use of such isotype control MAb to establish the positive analysis region for bone marrow CD34+ cells in Milantype analyses is even less adequate (Sutherland et al., 1994). Furthermore, if placement of the fluorescence marker is based on the cluster of CD34+ events (such as in the Nordic and SIHON protocols), the failure to detect any events stained by such isotype-control MAb contributes no information, as it would under-represent nonspecific staining. Second, if the control MAb features higher nonspecific binding than the CD34 MAb, subtraction of the percent cells staining with the control MAb from the CD34 result would lead to underestimation of the percent CD34+ cells, not to mention a percent CD34+ cells <0 that will be obtained if the
“nonspecifically” stained cells outnumber the CD34+ cells. Beckman Coulter has addressed this problem by designing an isoclonic control, which consists of a 50-fold excess of unlabeled CD34 MAb relative to CD34-PE MAb. Thus, the unlabeled MAb blocks the specific signal of the conjugated MAb via its Fab region. However, the unlabeled CD34 MAb will also block any nonspecific binding of the labeled CD34 MAb. That situation may lead to the setting of a positive analysis region based on misleadingly low levels of background staining obtained with the isoclonic control. In practice, the authors have not found this control to be of value in the enumeration of CD34+ cells using the basic protocol (Keeney et al., 1998b). In conclusion, the authors consider controlling for nonspecific CD34 MAb binding using fluorochrome- and isotype-matched MAbs in the Basic and Support Protocols to be redundant, because the Boolean gating strategy eliminates the cell populations typically involved in nonspecific MAb binding (i.e., myelomonocytic cells, platelets, and dead cells). However, unstained cells should be included as a reference point for the level of autofluorescence when the coexpression of other markers on CD34+ cells is studied (Support Protocol 2; Keeney et al., 1998b). Setting the positive fluorescence region for immunophenotypic characterization of CD34+ cells Virtually all the antibodies used to date for the immunophenotypic characterization of CD34+ cells fail to identify discrete, nonoverlapping populations. Instead, they generate a continuous spectrum of fluorescence signals ranging in many cases from negative to (dimly) positive. The decision on where to draw the line separating positive from negative CD34+ cells appears to be problematic. The same practical and theoretical arguments cited against the use of fluorochrome- and isotype-matched MAb controls apply here (Keeney et al., 1998b). The approach described herein makes use of the autofluorescence in the orange-red fluorescence channel of the gated CD34+ cells to define the lower fluorescence intensity limit of the positive analysis region from the control tube. Typically, lymphocytes gated from the same control tube should exhibit very similar autofluorescence characteristics (Fig. 6.4.6, plot 7). If they do not, the compensation of the instrument may be suboptimal and must be corrected. The experimentally stained sample
6.4.20 Supplement 9
Current Protocols in Cytometry
is then analyzed using the defined positive analysis region. This approach is adequate for the analysis of the coexpression of CD133 and CD90 by the CD34+ cells. CD133 is expressed on the majority of peripheral blood CD34+ cells (Fig. 6.4.5) while CD90 is usually expressed on a minority of these cells (Fig. 6.4.6), and also on a small subset of T lymphocytes in some samples. One significant advantage of the authors’ approach is that it can be used to optimize the titration of the selected PE conjugate. This can be critically important to the correct interpretation of data if the conjugate of choice to “antigen X” on CD34+ cells also stains lymphocytes nonspecifically at the concentration used. At optimal concentration, the lymphocytes (or a clearly delineated subset thereof) do not show increased fluorescence in the experimental tube over that exhibited by those in the control tube. However, for some antigens of potential interest in HSC research, e.g., CD38, this approach to optimizing the MAb concentration is not appropriate because a clearly distinguishable CD38-"negative" subset of lymphocytes cannot be identified. CD34bright versus CD34dim cells Although CD34bright cells constitute the most widely used parameter for the estimation of the hematopoietic potential of an apheresis product (Johnsen and Knudsen, 1996; Gratama et al., 1997), the precise potential of CD34dim cells for either short- or long-term engraftment is still unresolved. Bone marrow CD34dim cells may include precursors committed to the erythroid, myeloid, and lymphoid lineages (Terstappen et al., 1991), but their relevance to even short-term engraftment in peripheral blood stem cell (PBSC) transplantation is unknown. Importantly, the use of a Boolean gating strategy will exclude artifactually stained events while retaining true CD34+ cells regardless of their staining intensity. This is particularly useful when analyzing bone marrow samples, in which HPCs expressing CD34 in a rather wide range of intensities overlap with unstained cell populations and nonspecifically stained events. In such samples, a cluster of CD34bright cells cannot be distinguished reliably. The gating strategy at the heart of the Basic Protocol was initially devised to enumerate CD34+ cells in bone marrow samples (Sutherland et al., 1994), a source of HPCs generally considered to be the most difficult to enumerate accurately.
Sample preparation: When to wash, when to fix The benefits of washing are the removal of free hemoglobin (from lysed red cells), cellular debris, and unbound MAb, and the ability to resuspend the sample in a relatively small volume of buffer. Consequently, flow cytometry will be less complicated as less time will be needed for data acquisition, and there will be less background fluorescence and less contamination of listmode data files with irrelevant events. The risks of washing are loss of cells of interest and loss or clumping of counting beads if the latter are already present in the sample before washing. Each investigator should verify that cell and/or bead loss is prevented in a particular situation before implementing such modifications (Keeney et al., 1998a). The authors do not recommend the inclusion of washing steps in single-platform analyses. The benefits of fixation are discussed in UNIT 6.2. However, several commercially available lysing reagents contain fixatives and may reduce binding of CD34 MAb to varying degrees depending on the CD34 epitopes recognized, or may cause light-scatter changes that are unpredictable and variable over time. Washing of fixed cells results in higher cell losses when cells are washed under adhesion-promoting conditions (e.g., polystyrene tubes, absence of protein in washing buffers). Therefore, it is recommended not to wash samples lysed with fixative-containing buffers. If the samples must be fixed because of time constraints, safety considerations, or other scenarios, fixation should be performed after washing. Remember that fixed samples may have altered light scatter and increased autofluorescence levels as compared to unfixed samples, requiring adaptation of threshold and analysis region settings. Accurate pipetting The accuracy and reproducibility of conventional pipetting of high-viscosity materials, such as peripheral blood or apheresis samples, is less than that of water. In conventional pipetting, precisely the required amount of sample is aspirated and then ejected. This method runs the risk of inclusion of air bubbles in the dispensed volume which, in such cases, will be smaller than required. To improve the accuracy and reproducibility of sample aliquoting, “reverse pipetting”—preferably using a calibrated automated pipettor—is recommended. With reverse pipetting, the plunger of the micropipettor is pressed to the second stop so that a slight excess of sample is gently aspirated. The fluid is then ejected against the lower end of the wall
Phenotypic Analysis
6.4.21 Current Protocols in Cytometry
Supplement 25
of the tube by pressing the plunger to the first stop, allowing ejection of the precisely required volume. A small part of the sample remains in the tip and deviations from the requested volume by viscosity and the formation of air bubbles are prevented. If blood is spilled to the upper part of the tube, it should be removed with a swab. The accuracy of pipetting is of critical importance in single-platform assays, where the result (i.e., absolute CD34+ cell count) is directly affected by the amount of sample pipetted and by the amount of counting beads if these are added by pipetting (as in the Basic Protocol). In this context it is also important to recall that apheresis products require dilution of the specimen prior to staining in order to obtain cell concentrations in the range for which the CD34+ cell enumeration assays and the hematology analyzers have been calibrated and are linear (see above for a discussion of the accuracy and reproducibility of aliquoting of specimens with high viscosity). Given the dilution factor commonly required (i.e., up to 1/40), a minor error at this stage can have a significant effect on the accuracy of the absolute count irrespective of whether a single- or double-platform assay is subsequently used. Number of events to be acquired To assess the quality of the HPC transplants, it is important to obtain statistically reliable CD34+ cell enumerations. As CD34+ cells are rare events, their number will follow a Poisson distribution. According to that distribution, their coefficient of variation (CV, in %) is (100 × √n )/n, where n is the number of events fulfilling the flow cytometric criteria of CD34+ cells. Hence, the CV of CD34+ cells will vary proportionally to the square root of the number of CD34+ cells acquired. The procedures outlined in the Basic Protocol require a minimum of 100 true CD34+ cells, which ensures a CV of 10%.
Anticipated Results
Enumeration of CD34+ Hematopoietic Stem and Progenitor Cells
Samples that come from apparently healthy individuals, that have not been mobilized with cytokines such as G-CSF or GM-CSF, and that have a normal leukocyte count (i.e., between 6 and 10 × 109/liter) typically contain <0.1% CD34+ cells as a fraction of leukocytes, or <5 CD34+ cells/µl. Following mobilization, peripheral blood CD34+ cell counts are a reliable predictor for harvesting adequate amounts of HPCs for transplantation and may range between <5 and 200 CD34+ cells/µl. If the preapheresis blood sample contains <10 CD34+
cells/µl (i.e., poor mobilization), an apheresis run that day may result in an inadequate yield (i.e., <0.5 × 106 CD34+ cells/kg). Generally, peripheral blood counts between 10 and 20 CD34+ cells/µl may predict apheresis yields of 0.5 to 1 × 106/kg. In a study of 59 patients with hematological malignancies or breast cancer undergoing 235 apheresis procedures, Chapple et al. (1998) showed that if the peripheral blood CD34+ cell count exceeded 5 CD34+ cells/µl, at least 0.5 × 106 CD34+ cells/kg could be collected by apheresis on the same day. Counts between 20 and 40 CD34+ cells/µl may correlate with yields ∼1 × 106 CD34+ cells/kg, and in patients with counts >40 CD34+ cells/µl (i.e., good mobilization), one single harvest may frequently be enough to secure a transplant of 2 × 106 CD34+ cells/kg. In cord blood specimens, CD34+ cell counts ranging between 10/µl and 100/µl can be expected.
Time Considerations Preparation of an individual sample or a series of up to 6 samples takes 30 to 40 min, plus 10 to 15 min if a washing step is applied. Assuming that the flow cytometer has been set up and calibrated appropriately and that userdefined software applications for data acquisition and analysis have been prepared, listmode data acquisition may take up to 5 min per tube (depending on the desired lower level of CD34+ cell detection), while the analysis of these data files will take another 5 min per sample. Thus, the entire procedure may be completed within 1 hr for a single sample and within 1.5 hr if up to 6 samples are processed simultaneously.
Literature Cited Allen, D.S., Keeney, M., Howson-Jan, K., Popma, J., Weir, K., Bhatia, M., Sutherland, D.R., and Chin-Yee, I.H. 2002. Number of viable CD34+ cells reinfused predicts engraftment in autologous hematopoietic stem cell transplantation. Bone Marrow Transplant. 29:967-972. Bender, J.G., Unverzagt, K., and Walker, D. 1994. Guidelines for determination of CD34+ cells by flow cytometry: Application to the harvesting and transplantation of peripheral blood stem cells. In Hematopoietic Stem Cells: The Mulhouse Manual (E. Wunder, H. Sovalat, P.R. Henon, and S. Serke, eds.) pp. 31-43. AlphaMed Press, Dayton, Ohio. Borowitz, M.J., Guenther, K.L., Schultz, K.E., and Stelzer, G.T. 1993. Immuno-phenotyping of acute leukemia by flow cytometry: Use of CD45 and right angle light scatter to gate on leukemic blasts in three color analysis. Am. J. Clin. Pathol. 100:534-540.
6.4.22 Supplement 25
Current Protocols in Cytometry
Brando, B., Göhde Jr., W., Scarpati, B., and D’Avanzo G. 2001. The ‘vanishing counting bead’ phenomenon: Effect on absolute CD34+ cell counting in phosphate-buffered saline-diluted leukapheresis samples. Cytometry 43:154160. Brocklebank, A.M. and Sparrow, R.L. 2001. Enumeration of CD34+ cells in cord blood: A variation on a single-platform flow cytometric method based on the ISHAGE gating strategy. Cytometry 46:254-261. Chapple, P., Prince, H.M., Quinn, M., Bertoncello, I., Juneja, S., Wolf, M., Januszewicz, H., Brettell, M., Gardyn, J., Seymour, C., and Venter, D. 1998. Peripheral blood CD34+ cell count reliably predicts autograft yield. Bone Marrow Transplant. 22:125-130. Gratama, J.W., Kraan, J., Levering, W., Van Bockstaele, D.R., Rijkers, G.T., and Van der Schoot, C.E. 1997. Analysis of variation in results of CD34+ hematopoietic progenitor cell enumeration in a multicenter study. Cytometry 30:109117.
Siena, S., Bregni, M., Brando, B., Ravagnani, F., Bonadonna, G., and Gianni, A.M. 1989. Circulation of CD34+ hematopoietic stem cells in the peripheral blood of high-dose cyclophosphamide-treated patients: Enhancement by intravenous recombinant human granulocyte-macrophage colony-stimulating factor. Blood 74:19051914. Siena, S., Bregni, M., Brando, B., Belli, N., Ravagnani, F., Gandola, L., Stern, A.C., Lansdorp, P.M., Bonadonna, G., and Gianni, A.M. 1991. Flow cytometry for clinical estimation of circulating hematopoietic progenitors for autologous transplantation in cancer patients. Blood 77:400409. Stewart, A.K., Imrie, K., Keating, A., Anania, S., Nayar, R., and Sutherland, D.R. 1995. Optimizing the CD34+,Thy-1+ stem cell content of peripheral blood collections. Exp. Hematol. 23:1619-1627. Sutherland, D.R. and Keating, A. 1992. The CD34 antigen: Structure, biology and potential clinical applications. J. Hematother. 1:115-129.
Gratama, J.W., Orfao, A., Barnett, D., Brando, B., Huber, A., Janossy, G., Johnsen, H.E., Keeney, M., Marti, G.E., Preijers, F., Rothe, G., Serke, S., Sutherland, D.R., Van der Schoot, C.E., Schmitz, G., and Papa, S. (for the European Working Group on Clinical Cell Analysis). 1998. Flow cytometric enumeration of CD34+ hematopoietic stem and progenitor cells. Cytometry 34:128142.
Sutherland, D.R., Keating, A., Nayar, R., Anania, S., and Stewart, A.K. 1994. Sensitive detection and enumeration of CD34+ cells in peripheral and cord blood by flow cytometry. Exp. Hematol. 22:1003-1010.
Gratama, J.W., Menéndez, P., Kraan, J., and Orfao, A. 2000. Loss of CD34+ hematopoietic progenitor cells due to washing can be reduced by the use of fixative-free erythrocyte lysing reagents. J. Immunol. Methods 239:13-23.
Sutherland, D.R., Anderson, L., Keeney, M., Nayar, R., and Chin-Yee, I. 1996b. QBEnd10 (CD34) antibody is unsuitable for routine use in the ISHAGE CD34+ cell determination assay. J. Hematother. 5:601-603.
Gutensohn, K., Carrero, I., Krueger, W., Kroeger, N., Schafer, P, and Luedemann, K. 1999. Semiautomated flow cytometric analysis of CD34-expressing hematopoietic cells in peripheral blood progenitor cell apheresis products. Transfusion 39:1220.
Sutherland, D.R., Anderson, L., Keeney, M., Nayar, R., and Chin-Yee, I. 1997. Re: Toward a worldwide standard for CD34+ enumeration (letter). J. Hematother. 6:85-89.
Johnsen, H.E. and Knudsen, L.M. (for the Nordic Stem Cell Laboratory Group) 1996. Nordic flow cytometry standards for CD34+ cell enumeration in blood and leukapheresis products: Report from the second Nordic workshop. J. Hematother. 5:237-245. Keeney, M., Chin-Yee, I., Weir, K., Popma, J., Nayar, R., and Sutherland, D.R. 1998a. Single platform flow cytometric absolute CD34+ cell counts based on the ISHAGE guidelines. Cytometry 34:61-70. Keeney, M., Gratama, J.W., Chin-Yee, I.H., and Sutherland, D.R. 1998b. Isotype controls in the analysis of lymphocytes and CD34+ stem/progenitor cells by flow cytometry—time to let go! Cytometry 34:280-283. Owens, M.A. and Loken, M.R. 1995. Peripheral blood stem cell quantitation. In Flow Cytometric Principles for Clinical Laboratory Practice, pp. 111-128. Wiley-Liss, New York.
Sutherland, D.R., Anderson, L., Keeney, M., Nayar, R., and Chin-Yee, I. 1996a. The ISHAGE guidelines for CD34+ cell determination by flow cytometry. J. Hematother. 5:213-226.
Terstappen, L.W.M.M., Huang, S., Safford, M., Lansdorp, P.M., and Loken, M.R. 1991. Sequential generations of hematopoietic colonies derived from single nonlineage-committed CD34+CD38- progenitor cells. Blood 77:12181227.
Contributed by D. Robert Sutherland The University Health Network Toronto, Ontario, Canada Michael Keeney London Health Sciences Center London, Ontario, Canada Jan W. Gratama Daniel den Hoed Cancer Center Rotterdam, The Netherlands Phenotypic Analysis
6.4.23 Current Protocols in Cytometry
Supplement 25
Immunophenotypic Analysis of Peripheral Blood Lymphocytes
UNIT 6.5
Identification, classification, and enumeration of peripheral blood lymphocytes can be performed by several methodological approaches (UNIT 6.2) according to availability of reagents and instrument capabilities. Currently, most commercially available flow cytometry instruments have the capability to detect three or four different fluorescence colors. In addition, the vast commercial availability of antibodies directed against lymphocyte-associated antigens, and labeled with various fluorochromes, provides the basis for the common use of multicolor stainings for the analysis of peripheral blood lymphocytes in which three or more monoclonal antibody reagents are combined. Measurements of monoclonal antibody stainings are commonly performed on whole peripheral blood samples in which the erythrocytes have been lysed and the remaining cells washed. Results are expressed as a percentage of the total leukocytes and/or lymphocytes present in the sample. For a presentation of this approach, see Basic Protocol 1. More recently, alternate approaches, using lysis of the erythrocytes but no washing, have been specifically developed to avoid either random or selective cell losses (see ALTERNATE PROTOCOL 1). These later methods not only allow an accurate measurement of percentages of cells, but at the same time also derive the absolute numbers of cells of interest per unit of volume of the sample. Methods using microbeads as an internal reference to derive absolute counts are described in Alternate Protocol 2. Frequently, the analysis of peripheral blood lymphocytes is performed on mononuclear cells that have been isolated through densitygradient centrifugation procedures. For indications and guidelines to perform densitygradient isolation of mononuclear cells, see Basic Protocol 2. The enumeration and characterization of lymphocyte subsets are crucial in specific clinical situations; hence, strategies to perform reliable counts are most important. The reliability of flow cytometric quantitation of cell populations depends not only on instrument performance, but also on the reagents and gating strategies used to specifically identify the cells of interest with the highest sensitivity. Support Protocols 1, 2, and 3 discuss the most common combinations of reagents and gating strategies for the global identification of the lymphocytes present in a peripheral blood sample. Its is relevant to stress that whichever protocol is used, all procedures should comply with general quality control practices and guidelines (UNIT 6.1). STRATEGIC PLANNING The selection of antibodies to perform the immunologic classification and enumeration of peripheral blood lymphocytes has been strongly biased by the most popular clinical application of immunophenotyping of peripheral blood lymphocytes—testing of persons infected with the human immunodeficiency virus (HIV). This fact has resulted in several recommendations regarding panels of antibodies whose most critical goal is the reliable enumeration of CD4+ T cells, since CD4 count remains the optimal clinical indicator of the effects of the infection on the immune system in HIV-infected patients. Table 6.5.1, Table 6.5.2, and Table 6.5.3 describe the panels recommended by the United States Centers for Disease Control and Prevention (CDC) for two-, three- and four-color immunophenotyping of peripheral blood lymphocytes in persons infected with HIV (U.S. Department of Health and Human Services, 1997; Nicholson et al., 1996). Other organizations, such as the National Committee for Clinical Laboratory Standards (NCCLS) in the United States, have recommended, not exclusively for HIV-infected Contributed by Alejandro Ruiz-Argüelles and Beatriz Pérez-Romano Current Protocols in Cytometry (2000) 6.5.1-6.5.14 Copyright © 2000 by John Wiley & Sons, Inc.
Phenotypic Analysis
6.5.1 Supplement 11
Table 6.5.1 Two-Color Monoclonal Antibody Panel Recommended by the U.S. Centers for Disease Control
Green fluorescence
Red fluorescence
1 2 3 4 5
CD45 Isotype CD3b CD3 b CD3 b
CD14 Isotype CD4 CD8 CD19
6
CD3 b
CD16/56
Tube
Purpose of admixture Gating on lympohocytesa Determine background fluorescence Count CD3+/CD4+ T cells Count CD3+/CD8+ T cells Count total T (CD3) and B (CD19) cells Count total T (CD3) and NK (CD16/56) cells
aLymphocyte gating on FS and SS should yield >98% CD45++ and <2% CD14+ cells. This approach assumes
that the efficiency of the lysing system will remain constant for the rest of the tubes in the panel. bThe repeated use of CD3 in four tubes serves as a control for tube-to-tube variability; the values of all four
tubes should be within 3% of each other.
Table 6.5.2
Panel Aa
Three-Color Monoclonal Antibody Panels Recommended by the CDC
Antibodies
Purpose of admixture
CD3/CD4/CD45b
Gate on CD45++ and side scatter, count CD3/CD4 cells Gate on CD45++ and side scatter, count CD3/CD8 cells Gate on CD45++ and side scatter, count CD3 and CD19 cells Count T, B, and NK cells Count total T (CD3), CD3/CD4, and CD3/CD8 cells
CD3/CD8/CD45b CD3/CD19/CD45b Bc
CD3/CD19/CD16-56 CD3/CD4/CD8
aPanel A is recommended for instruments incapable of yielding absolute cell numbers directly, and isotype
control is not needed, for CD45 identifies leukocyte subpopulations based on fluorescence intensity. bThe repeated use of CD3 serves as a control for tube-to-tube variability; the values of all tubes should be
within 3% of each other. cPanel B is recommended for systems capable of counting absolute cell numbers directly from the flow
cytometer.
Table 6.5.3
Four-Color Monoclonal Antibody Panel Recommended by the CDC
Antibodies Tube 1a Tube 2b
Purpose of admixture
Gate on CD45++ and side scatter, count total, CD3+/CD4+, and CD3+/CD8+ T cells CD3/CD19/CD56/CD45 Gate on CD45++ and side scatter, count total T, B, and NK cells CD3/CD4/CD8/CD45
aThe repeated use of CD3 serves as a control for tube-to-tube variability; the values obtained from all tubes should
Immunophenotypic Analysis of Peripheral Blood Lymphocytes
be within 3% of each other. bCD56 can be replaced by CD16, or both antibodies might be used simultaneously in a single color.
6.5.2 Supplement 11
Current Protocols in Cytometry
persons, the use of the same antibodies as those recommended by the CDC, in two-color admixtures but in no specific combinations except for the tube used for gating purposes, which contains antibodies directed against CD45 and CD14 (see Support Protocol 1; National Committee for Clinical Laboratory Standards, 1992). Therefore, laboratory workers should be free to choose the most appropriate antibody panels and combinations according to their particular needs and interests. Since CD4+ T-cell determination is surely the most common application of lymphocyte immunphenotyping, Basic Protocol 1, Alternate Protocol 1, and Alternate Protocol 2 are intended to serve this purpose. IMMUNOPHENOTYPIC ANALYSIS OF THE RELATIVE DISTRIBUTION OF LYMPHOCYTE SUBSETS IN ERYTHROCYTE-LYSED-ANDTHEN-WASHED WHOLE PERIPHERAL BLOOD SAMPLES
BASIC PROTOCOL 1
This is an efficient, rapid, and reliable method to enumerate the relative distribution of different subsets of peripheral blood lymphocytes. The monoclonal antibodies employed for this purpose should be conjugated with fluorescein isothiocyanate (FITC), phycoerythrin (PE), either the phycoerythrin-cyanine 5 tandem complex (PE-Cy5) or peridinin chlorophyll protein (PerCP) as a third color, and either the PE–Texas Red energy coupling dye (ECD) or allophycocyanin (APC) as a fourth one. After addition of the dye, the erythrocytes are lysed with ammonium chloride, the cells are washed, and then flow cytometric analysis is carried out. Materials Blood to be tested, with tripotassium EDTA or heparin as anticoagulant Fluorochrome-conjugated monoclonal antibodies directed against the cell surface antigens of interest Ammonium chloride lysing solution (APPENDIX 2A) Phosphate-buffered saline (PBS; APPENDIX 2A), filtered through a 0.45-µm filter PBS (APPENDIX 2A) containing 1% paraformaldehyde (prepare fresh) 12 × 75–mm polypropylene tubes Flow cytometer equipped with four fluorescence detectors Stain cells 1. Label as many 12 × 75–mm polypropylene tubes as necessary. 2. Pipet 100 µl of whole anticoagulated blood into each tube. 3. Add the desired admixtures of FITC-, PE-, PE-Cy5- (or PerCP)-, and ECD- (or APC)-conjugated antibodies to each one of the blood-containing tubes. Gently mix for 5 sec. Some suppliers provide ready-to-use pretitered combinations of antibodies. If the antibodies are purchased separately, the amount of each antibody to be added must be determined (UNIT 4.1). Selection of fluorochrome conjugates should be based on the type of laser and optics available on the instrument that will be used for the measurements. For four-color analysis in conventional XL instruments (Beckman Coulter), combinations of FITC, PE, PE-Cy5, and ECD fluorophores should be used, while for conventional FACScalibur instruments (Becton Dickinson Biosystems), combinations of FITC, PE, PerCP (or PECy5), and APC fluorophores should be selected.
4. Incubate 15 min at room temperature in the dark, or otherwise as recommended by the antibody supplier. Phenotypic Analysis
6.5.3 Current Protocols in Cytometry
Supplement 11
Lyse erythrocytes and wash cells 5. Add 2 ml ammonium chloride lysing solution and vortex for 5 sec. Alternatively, reagents are commercially available for manual erythrocyte lysis and fixation (Coulter Lysing Kit, Immunotech Lyse and Fix Reagent, Immunotech OptiLyse, Becton Dickinson FACS Lysing Solution, Caltag Whole Blood Lysing Solution, PharMingen PharMLyse, or Leinco Easy-Lyse). Instructions for use are supplied with the reagents.
6. Incubate 10 min at room temperature in the dark. 7. Centrifuge 5 min at 540 × g, room temperature. 8. Discard the supernatant, resuspend the cell pellet by vortexing for 5 sec, and add 2 ml filtered PBS to each tube. 9. Centrifuge again for 5 min at 540 × g, room temperature. 10. Discard the supernatant, resuspend the cell pellet by vortexing for 5 sec, and add 1 ml of PBS containing 1% paraformaldehyde to each tube. Perform flow cytometric analysis 11a. For flow cytometers in which a single 488-nm laser line is used: Collect forward scatter (FS), side scatter (SS), log green fluorescence (525 nm), log orange fluorescence (575 nm), log red fluorescence (610 nm), and log red-purple fluorescence (675 nm). Acquire and store listmode data on at least 5000 lymphocytes. 11b. For flow cytometers equipped with both 488- and 635-nm laser lines: Collect forward scatter (FS), side scatter (SS), log green fluorescence (525 nm), log orange fluorescence (575 nm), log red fluorescence (675 nm), and log red-purple fluorescence (661 nm). Acquire and store listmode data on at least 5000 lymphocytes. For gating strategies used during the analysis, refer to Support Protocols 1 to 3. ALTERNATE PROTOCOL 1
IMMUNOPHENOTYPIC ANALYSIS OF THE RELATIVE DISTRIBUTION OF LYMPHOCYTE SUBSETS IN ERYTHROCYTE-LYSED-NON-WASHED WHOLE PERIPHERAL BLOOD SAMPLES Recently, erythrocyte-lyse-non-wash procedures have been developed and applied to the enumeration of peripheral blood lymphocyte subsets. These protocols have two major advantages: They decrease sample manipulation and increase the speed of sample preparation. Additionally, if an automated lyse-non-wash instrument is available, this is a very rapid approach. Additional Materials (also see Basic Protocol 1) Coulter Immunoprep lysing and fixing kit Coulter Q-Prep immunology workstation 1. Stain the anticoagulated whole blood samples (see Basic Protocol 1, steps 1 to 4). 2. Lyse and fix the cells using the Coulter Q-Prep apparatus, set in the 35 sec cycle, according to the manufacturer’s instructions. Alternatively, reagents are commercially available for manual erythrocyte lysis and fixation (Coulter Lysing Kit, Immunotech Lyse and Fix Reagent, Immunotech OptiLyse, Becton Dickinson FACS Lysing Solution, Caltag Whole Blood Lysing Solution, PharMingen PharMLyse, or Leinco Easy-Lyse). Instructions for use are supplied with the reagents.
Immunophenotypic Analysis of Peripheral Blood Lymphocytes
3. Perform flow cytometry (see Basic Protocol 1, step 11a or 11b).
6.5.4 Supplement 11
Current Protocols in Cytometry
FLOW CYTOMETRIC IMMUNOPHENOTYPIC ANALYSIS OF ABSOLUTE NUMBERS OF PERIPHERAL BLOOD LYMPHOCYTE SUBSETS
ALTERNATE PROTOCOL 2
If absolute lymphocyte counts independent of hematology analyzers are preferred, internal volume control beads may be used. Single-platform methods to derive absolute lymphocyte counts have been shown to decrease variability in the results observed. Additional Materials (also see Basic Protocol 1) Coulter Immunoprep lysing and fixing kit Coulter Q-Prep immunology workstation Coulter Flow COUNT beads 1. Stain the anticoagulated whole blood samples (see Basic Protocol 1, steps 1 to 4) 2. Lyse and fix the cells using the Coulter Q-Prep apparatus, set in the 35-sec cycle, according to the manufacturer’s instructions. Alternatively, reagents are commercially available for manual erythrocyte lysis and fixation (Coulter Lysing Kit, Immunotech Lyse and Fix Reagent, Immunotech OptiLyse, Becton Dickinson FACS Lysing Solution, Caltag Whole Blood Lysing Solution, PharMingen PharMLyse, or Leinco Easy-Lyse). Instructions for use are supplied with the reagents.
3. Add 100 µl Flow COUNT beads to each tube by reverse pipetting (see UNIT 6.4) and mix gently. 4. Perform flow cytometry (see Basic Protocol 1, step 11a or 11b). Commercial tubes containing a known amount of bead reagents are available from Becton Dickinson (TRUE-COUNT tubes). If using these tubes, omit step 3. In any event, the integrity of the beads and their fluorescence once placed in a specific erythrocyte lysing solution should be tested prior to analysis.
LYMPHOCYTE IMMUNOPHENOTYPIC ANALYSIS OF PERIPHERAL BLOOD MONONUCLEAR CELL (PBMC)–ENRICHED SUSPENSIONS
BASIC PROTOCOL 2
The presence of a large proportion of dead leukocytes or the use of enrichment procedures prior to lymphocyte culture is not that infrequent; in very rare instances blood samples are heavily contaminated with cells other than leukocytes (e.g., nonhemopoietic tissue, bacteria, or fungi). In any of these cases, it might be useful to prepare a mononuclear cell–enriched suspension by density-gradient centrifugation, instead of by directly staining a whole-blood sample. Materials Blood to be tested, with tripotassium EDTA or heparin as anticoagulant Phosphate-buffered saline (PBS; APPENDIX 2A), 10°C Lymphocyte separation medium (LSM; e.g., Ficoll-Paque PLUS from Pharmacia Biotech, Lymphoprep 1077 or Nycoprep 1077 from Life Technologies, LymphoSep from ICN, Histopaque 1077 from Sigma, or LeucoPREP from Becton Dickinson), 10°C Fluorophore-conjugated monoclonal antibodies directed against the cell surface antigens of interest 15-ml conical polypropylene centrifuge tubes 12 × 75–mm polypropylene tubes Centrifuge and rotor capable of 2000 × g, refrigerated Flow cytometer Additional reagents and equipment for counting cells (APPENDIX 3A)
Phenotypic Analysis
6.5.5 Current Protocols in Cytometry
Supplement 11
Isolate PBMC 1. Dilute blood 1 part blood with 2 parts PBS. 2. Place 3 ml LSM in a 15-ml conical polypropylene centrifuge tube. 3. Carefully overlay the LSM with 9 ml diluted blood. 4. Centrifuge 35 min at 200 × g, 10°C. 5. Carefully remove the peripheral blood mononuclear cells (PBMC) from the interface with a Pasteur pipet. 6. Wash PBMC with PBS twice, each time by centrifuging 5 min at 540 × g, 10°C. 7. Resuspend PBMC pellet in PBS, count cells (APPENDIX 3A), and adjust to 2–5 × 106 cells/ml. Stain and wash cells 8. Place 100 µl of cell suspension (containing 2–5 × 105 total cells) in labeled 12 × 75–mm polypropylene tubes. 9. Add the desired admixtures of conjugated antibodies into each of the tubes containing PBMC suspensions. Some suppliers provide ready-to-use pretitered combinations of antibodies. If the antibodies are purchased separately, the amount of each antibody to be added must be determined (UNIT 4.1).
10. Incubate 15 min at room temperature in the dark, or otherwise as recommended by the antibody supplier. 11. Wash with PBS by centrifuging 10 min at 540 × g, 10°C. 12. Remove supernatant and resuspend pellet in 0.5 ml PBS. Perform flow cytometric analysis 13a. For flow cytometers in which a single 488-nm laser line is used: Collect forward scatter (FS), side scatter (SS), log green fluorescence (525 nm), log orange fluorescence (575 nm), log red fluorescence (610 nm), and log red-purple fluorescence (675 nm). Acquire and store listmode data on at least 5000 lymphocytes. 13b. For flow cytometers equipped with both 488- and 635-nm laser lines: Collect log green fluorescence (525 nm), log orange fluorescence (575 nm), log red fluorescence (675 nm), and log red-purple fluorescence (661 nm). Acquire and store listmode data on at least 5000 lymphocytes. For gating strategies used during the analysis, refer to Support Protocols 1 to 3. Centrifugation on density cushions with specific gravity of 1.077 enriches for mononuclear cells; therefore, a variable proportion of monocytes is expected to be present and should be considered for gating strategies.
Immunophenotypic Analysis of Peripheral Blood Lymphocytes
6.5.6 Supplement 11
Current Protocols in Cytometry
LIGHT-SCATTER GATING STRATEGY FOR THE IMMUNOPHENOTYPIC ANALYSIS OF PERIPHERAL BLOOD LYMPHOCYTES
SUPPORT PROTOCOL 1
1. Display the following bivariate plots (six parameters): forward scatter (FS) versus side scatter (SS) log green fluorescence versus log orange fluorescence log orange fluorescence versus log red fluorescence log green fluorescence versus log red fluorescence log red fluorescence versus log red-purple fluorescence log green fluorescence versus log red-purple fluorescence log orange fluorescence versus log red-purple fluorescence. 2. Gate the lymphocyte population in the low SS versus low FS region according to light scatter properties. Scatter gating of peripheral blood lymphocytes should be checked for its purity based on strong CD45 staining or the T, B, and NK sum using simultaneous stainings for the identification of these three lymphocyte subsets (i.e., CD3, CD19, and CD56). Whenever the percentage of lymphocytes contained in the FS versus SS gate is low (<95%) or the percentage of lymphocytes outside the gate is high (>5%), the bitmap gate should be redrawn to obtain a maximum of purity and specificity. If the abovementioned numbers can not be reached, a different strategy than the FS versus SS gating approach should be used.
3. Using unstained cells as a negative control, determine negative, positive, and doublepositive events in the remaining in-gate bivariate histograms. 4. Display an in-gate prism histogram to enumerate the proportions of cells in the eight theoretically possible combinations of the three antibodies. Prism histograms can be displayed with the Coulter XL System Two, Elite, and Expo 2.0 software programs. Equivalent features are provided by other suppliers.
CD45 IMMUNO-GATING STRATEGY FOR THE PHENOTYPIC ANALYSIS OF PERIPHERAL BLOOD LYMPHOCYTES
SUPPORT PROTOCOL 2
1. Display five bivariate plots of six parameters: CD45 (log green fluorescence) versus side scatter (SS) forward scatter (FS) versus side scatter (SS) log orange fluorescence versus log red fluorescence log orange fluorescence versus log red-purple fluorescence log red fluorescence versus log red-purple fluorescence. 2. Gate the lymphocyte population in the CD45 versus SS histogram according the intensity of CD45. Check that all the gated events have a uniformly low FS versus low SS distribution. 3. Using unstained cells as a negative control, determine negative, positive, and doublepositive events in the remaining in-gate bivariate histograms. 4. Display an in-gate prism histogram to enumerate the proportions of cells in the eight theoretically possible combinations of the three antibodies other than CD45. Prism histograms can be displayed with the Coulter XL System Two, Elite, and Expo 2.0 software programs. Equivalent features are provided by other suppliers.
Phenotypic Analysis
6.5.7 Current Protocols in Cytometry
Supplement 11
SUPPORT PROTOCOL 3
MULTIPARAMETRIC IMMUNO-GATING STRATEGY FOR THE PHENOTYPIC ANALYSIS OF PERIPHERAL BLOOD LYMPHOCYTES In the example below, the strategy of the LYMPHOGRAM reagent (Exalpha Biologicals) for three-color stainings is examplified. 1. Display five bivariate plots of the following pairs of parameters: forward scatter (FS) versus side scatter (SS) side scatter (SS) versus log green fluorescence (CD19-CD8) side scatter (SS) versus log orange fluorescence (CD56-CD3) log green fluorescence (CD8-CD19) versus log orange fluorescence (CD56-CD3) log green fluorescence (CD8-CD19) versus log red fluorescence (CD4). 2. Gate the lymphocyte population in the SS versus CD19-CD8+ and the SS versus CD56-CD3+ histograms as those events being positive for any of the four markers and displaying low SS values. Check that all the gated events have a uniform low FS versus low SS distribution. 3. Draw a region around each cluster of cells identified in the CD19-CD8 versus CD3-CD56 as well as the CD3-CD56 versus CD4 bivariate plots. Calculate the proportions of cells in each region. COMMENTARY Background Information
Immunophenotypic Analysis of Peripheral Blood Lymphocytes
The identification and enumeration of lymphocyte subsets has been of interest in many clinical fields for many years. Ever since the original observations that lymphocytes, although they have homogeneous morphological features, could be subclassified into different functional compartments, attempts have been undertaken to establish direct ways of identifying and enumerating them. Rosetting methods played an important historical role; these techniques assumed the concept of membrane receptor–ligand interactions, which remain fundamental for the detection of cell-membrane molecules using new molecular probes. Monoclonal antibodies have proved to be extremely useful tools for detecting the presence of diverse molecules on the surface membranes of leukocytes; in fact, the CD nomenclature and workshops stemmed from the diversity of monoclonal antibodies that react with such molecules (Kishimoto et al., 1997). The conjugation of monoclonal antibodies with fluorochromes, along with the detection of their reactivity with leukocytes by means of flow cytometry, was the cornerstone that boosted the applications of flow cytometry in the clinical laboratory (Macey, 1999; Bauer et al., 1993). Immunophenotyping of peripheral blood lymphocytes is useful in the diagnosis and classification of primary immunodeficiency
diseases. Examples of diagnostic applications include the investigation of CD19, CD20, and CD21 on peripheral blood lymphocytes from patients with X-linked agammaglobulinemia, which indicates an arrest of B-cell maturation (CD19+, CD20+, CD21−) in this disease, a phenomenon that is not found in transient hypogammaglobulinemia of infancy. Another example is the analysis of the expression of CD43 (sialophorin), which is markedly decreased on lymphocytes from patients with WiskottAldrich syndrome. In the classification of patients affected by severe combined immunodeficiency, allocation of each newly diagnosed case to a given category is based exclusively on the patterns of expression of CD3, CD4, CD8, and CD38 on the lymphocyte surface membrane. Extensive reviews of the clinical utility of lymphocyte immunophenotyping in primary immunodeficiency might be found elsewhere (Keren et al., 1994). Infection with the human immunodeficiency virus (HIV) and acquired immunodeficiency syndrome (AIDS) represent the most common clinical application of flow cytometric immunophenotyping of peripheral blood lymphocytes. CD4+ T cells are the main cellular targets of HIV infection; they are destroyed by several mechanisms and their numbers decrease with progression of disease. According to certain organizations, clinical decisions con-
6.5.8 Supplement 11
Current Protocols in Cytometry
cerning initiation of antiviral therapy or prophylaxis of opportunistic infections are based on the numbers of CD4+ T cells; therefore, accuracy and precision in measuring lymphocyte subsets are crucial during the follow-up of these patients. There is no doubt that the increasing demand for reliable CD4+ T-cell counts on a rapidly growing number of HIV/AIDS patients worldwide has boosted the installation of flow cytometry facilities in clinical laboratories. This explains why corporations have reacted by developing reagents and even instruments for this single purpose, and also why most standardization efforts so far undertaken in clinical flow cytometry have focused on CD4+ T-cell enumeration. Peripheral blood immunophenotyping has also proved to be very useful in the diagnosis and follow-up of patients with certain chronic lymphoproliferative disorders, particularly chronic lymphatic leukemia of B-cell origin, where clonality of the neoplastic cell population is defined on the basis of the abnormal proportions of κ and λ immunoglobulin light chains on the membranes of otherwise mature normal B cells. Subclassification of the major chronic B-lymphoid leukemias is made on the distinct immunophenotypes borne by each subtype (Loken et al., 1990; Ruiz-Argüelles et al., 1998). In transplantation medicine, changes in subpopulations of T cells occur after the grafting itself, and subsets of CD4+ T cells reflect specific changes secondary to immunosuppression. Patients receiving azathioprine or cyclosporine have decreased numbers of CD4+ T cells that coexpress the CD29 antigen, and increased numbers of CD4+ T cells that coexpress CD45RA. Lymphocytes coexpressing CD4 and CD57, a subset whose function remains obscure, are increased in patients receiving azathioprine but not in those receiving cyclosporine. Some reports have suggested that patients with the lowest CD4/CD8 ratios have a higher frequency of rejection episodes and show resistance to antirejection therapy. An increase in the expression of HLA-DR in the CD8+ T-cell population might be used as an early marker of rejection, and a decrease to normal follows successful antirejection therapy (Rich, 1996; Roitt et al., 1998). Imbalances of T-cell subsets are seen in chronic inflammatory diseases such as systemic lupus erythematosus, mixed connective tissue disease, progressive systemic sclerosis, and rheumatoid arthritis, and some reports suggest that the degree of such changes might
reflect the activity of the underlying disease. In the mononucleosic syndromes, dramatic changes in the proportion of CD4 and CD8 cells occur, and normalization of this ratio coincides with remission of clinical symptoms (Lahita, 1992). Apart from the clinical applications mentioned above, the identification, enumeration, and phenotypic characterization of peripheral blood lymphocytes have shown to be of great utility once combined with other phenotypic and functional lymphocyte markers in immunologic research. Examples of these applications include simultaneous detection on specific subsets of lymphocytes—e.g., the major T, B, or NK cells—of intracellular cytokine production, activation-associated antigens, or apoptotic markers, among others.
Critical Parameters Selection of antibodies A critical step in the immunophenotypic analysis of peripheral blood lymphocytes is the selection of the monoclonal antibodies to be used. In spite of the fact that high-quality monoclonal antibody conjugates are currently available, the selection of a specific clone and fluorochrome conjugate, as well as the antigen to be used for specific and sensitive identification of a given cell subset, represents a critical step for the success of the analytical approach. As an example, while it has been claimed for years that CD19 and CD20 are B cell–specific markers, it is now well established that a subset of normal T cells is CD20 positive, which limits the value of this marker for the specific identification of peripheral-blood B lymphocytes. Guidelines and recommendations provided by several institutions and professional organizations should be followed to minimize analytical variability. In this area, special attention should be paid to instrument setup and calibration procedures and how to control such variability. In spite of the fact that the use of direct immunofluorescence combined with an erythrocyte-lysed whole-blood technique has dramatically decreased the problems with nonspecific staining, appropriate controls, such as lyophilized/stabilized cells or cell lines, should be used when testing an antibody for the first time, to rule out this possibility. Estimation of absolute lymphocyte counts In recent years it has become possible to derive absolute counts directly from the flow cytometer. Such a possibility is rapidly expand-
Phenotypic Analysis
6.5.9 Current Protocols in Cytometry
Supplement 11
ing among diagnostic laboratories. It should be stressed that once these approaches are used, a non-wash technique is recommended. In this situation, appropriate control of the volumes of sample and control beads should be made by using calibrated pipets and reverse pipetting procedures (UNIT 6.4). If the leukocyte count and the white blood cell differential are obtained in a hematology cell analyzer, it is very important to confirm that the quality-control practices, standardization, and calibration procedures of the hematology analyzer comply strictly with recommendations. Electronic analyzers that perform automated differential counts usually yield much more reproducible results than manual counts do, and their routine use is recommended. Sample preparation Sample storage, handling, and preparation are very important parameters that might affect the accuracy of lymphocyte enumeration (see UNIT 5.1); as an example, low temperatures affect the viability of CD4+ T cells more than the viability of CD8+ T cells, while density-gradient isolation of lymphocytes enriches for CD4+ T cells, because there is a selective loss of CD8+ T cells in the centrifugation process. Accordingly, guidelines and recommendations provided by several institutions and professional organizations should be followed to minimize analytical variability. Isotype controls The use of fluorochrome-conjugated, isotype-matched monoclonal immunoglobulins as controls for nonspecific staining has been stressed for many years by several organizations. However, it is clear that these “isotypic controls” are far from optimal to subtract the binding of antibodies to the lymphocyte membrane which results from reactions other than the specific CD antigen–antibody reaction. Although such reagents are claimed by manufacturers to serve this purpose, the method of purification, protein content, and fluorescein/protein (F/P) ratio are often different from those of the MAbs of interest. For these reasons, and considering that all antibodies used for immunophenotyping of lymphocytes can clearly discriminate negative from positive cells within a clean lymphocyte gate, the inclusion of these control antibodies is not recommended.
Number of events to be acquired Lymphocytes that have been positively stained with the majority of the recommended antibodies display a Gaussian or Poisson distribution. Accordingly, to obtain an acceptable (<10%) coefficient of variation, the minimum number of positive events that have to be acquired is 100 (see UNIT 6.4). In the healthy individual, the least frequent lymphocyte subpopulation corresponds to NK cells, which are normally found above 3%. Hence, a total of 3000 gated (lymphocytes) cells should suffice for this purpose. However, in AIDS patients it is not that uncommon to find ∼1% CD4+ T cells, whose reliable count would require the acquisition of ∼10,000 gated cells. A practical approach might be to set the instrument’s stop at 10,000 gated (lymphocyte) events, and according to the estimated frequency of CD4+ T cells, the acquisition might be interrupted earlier. Dead cells Cell death due to prolonged or inadequate storage, holding, or transportation of samples is seldom equivalent among lymphocyte subpopulations; hence, the relative proportion of each subpopulation is affected differently by the proportion of dead cells. The best way to deal with the presence of dead cells is to prevent it. Samples should be prepared and analyzed as soon as possible after being drawn; however, if the sample is obtained from a patient at a location distant from a flow cytometry facility, and therefore must be shipped, the anticoagulant, time, and temperature should be taken into consideration. According to CDC recommendations, samples drawn with tripotassium EDTA should be analyzed within 30 hr, while those drawn with acid citrate dextrose (ACD) or heparin might be analyzed within 48 hr without significant loss of viability. In any event, the samples must be kept at room temperature (21° to 25°C) during shipment, and extreme temperatures should be avoided. When a sample does not comply with these specifications, a second sample should be requested. In a special scenario where a second sample cannot be obtained, enrichment of mononuclear cells by density-gradient centrifugation is a method for getting rid of dead cells; however, the drawbacks of such an approach should be taken into consideration.
Immunophenotypic Analysis of Peripheral Blood Lymphocytes
6.5.10 Supplement 11
Current Protocols in Cytometry
1023
FS
CD4+
102 101
+ –
100
0 0
100
1023
101 CD8+
SS –+
–+ 102
CD8+
102
CD4+
102
101 + –
100
100
101 CD3+
101 + –
100
102
100 Prism
101 CD3+
102
977 3.8%
35.5%
29.8%
5.7%
18.8%
5.2%
1.1%
0.1% CD4+ CD8+ CD3+
– – –
+ – –
– + –
+ + –
– – +
+ – +
– + +
+ + +
Figure 6.5.1 Three-color immunophenotyping of lymphocytes. Gating is performed by light-scatter properties. Bivariate histograms depict coexpression of CD3, CD4, and CD8. The prism histogram shows the proportion of cells in each of the eight possible phenotypic combinations.
Phenotypic Analysis
6.5.11 Current Protocols in Cytometry
Supplement 11
103
Lymphocytes
102
102
CD4+
CD45+
103
101
101
100
100 0
100
1023
101
102 CD8+
SS –+
–+
103
103 CD8+
CD4+
103
102
102
+ –
101
+ –
101
100
100 100
101
102
100
102
101
102
102
CD3+
CD3+ 1011 30.4%
5.4%
37.9%
2.3%
21.0%
2.6%
CD4+ CD8+ CD3+
– – –
+ – –
– + –
0.0% + + –
– – +
+ – +
– + +
0.4% + + +
Figure 6.5.2 Four-color immunophenotyping of lymphocytes. Gating is performed according to SS and CD45 intensity. Bivariate histograms depict coexpression of CD3, CD4, and CD8. The prism histogram shows the proportion of cells in each of the eight possible phenotypic combinations.
Anticipated Results
Immunophenotypic Analysis of Peripheral Blood Lymphocytes
When enumerating lymphocyte subpopulations, a sample from a healthy individual must be run simultaneously with patients’ samples, and the obtained values must fall within a reference interval or range. The reference range for each measured cell population must be established by each laboratory in 50 to 100 normal blood donors. Age-related and circadian changes have been described for several
T-cell subsets, including CD4+ T cells, and must be taken into consideration when analyzing normal as well as pathological samples. If peripheral blood from normal individuals is not available on a daily basis, lyophilized or stabilized cells with predefined target values are commercially available. These cells can be also used as an internal control for the staining procedure(s) and the instrument performance. In addition, different populations of lympho-
6.5.12 Supplement 11
Current Protocols in Cytometry
1023
FS
Lymphocytes 30.6%
0 0
1023
SS
A1 6.6%
A2 22.2%
A3 31.7%.
A4 39.6%
CD8+
102
101
100
100
101
102
CD3+
Figure 6.5.3 Two-color immunophenotyping of lymphocytes. The upper panel depicts gating of lymphocytes according to their light-scatter properties. The lower panel shows percentages of the cells according to their staining for combination of CD3 and CD8 monoclonal antibody reagents.
cytes display distinct scatter characteristics. For example, NK cells, TCR-γδ+ T cells, and CD8+/TCR-αβ+ T cells display higher scatter values than do B lymphocytes. CD4+ T lymphocytes usually display a heterogeneous distribution and mix with both B cells and NK cells in FS versus FS bivariate plots. Some expected results are shown in Figure 6.5.1, Figure 6.5.2, and Figure 6.5.3.
Keren, D.F., Hanson, C.A., and Hurtubise, P.E. 1994. Flow Cytometry and Clinical Diagnosis. ASCP Press, Chicago, Ill.
Literature Cited
Loken, M.R., Brosnan, J.M., Bach, B.A., and Ault, K.A. 1990. Establishing optimal lymphocyte gates for immunophenotyping by flow cytometry. Cytometry 11:453-459.
Bauer, K.D., Duque, R.E., and Shankey, T.V. 1993. Clinical Flow Cytometry: Principles and Application. Williams & Wilkins, Baltimore.
Kishimoto, T., Kikutani, H., von dem Borne, A.E.G.K., Goyert, S.M., Mason, D.Y., Miyasaka, M., Moretta, L., Okumura, K., Shaw, S., Springer, T.A., Sugamura, K., and Zola, H. 1997. Leukocyte Typing VI. White Cell Differentiation Antigens. Garland Publishing, New York. Lahita, R.G. 1992. Systemic Lupus Erythematosus. 2nd ed. Churchill Livingston, New York.
Phenotypic Analysis
6.5.13 Current Protocols in Cytometry
Supplement 11
Macey, M.G. 1999. Flow Cytometry. Clinical Applications. Blackwell Scientific, Oxford. National Committee for Clinical Laboratory Standards. 1992. Clinical Applications of Flow Cytometry: Quality Assurance and Immunophenotyping of Peripheral Blood Lymphocytes. NCCLS Publication H42-T. NCCLS, Wayne, Pa. Nicholson, J.K.A., Hubbard, M., and Jones, B.M. 1996. Use of CD45 fluorescence and side scatter characteristics for gating lymphocytes when using the whole blood lysis procedure and flow cytometry. Cytometry 26:16-21. Rich, R.E. 1996. Clinical Immunology. Principles and Practice. Mosby, Baltimore, Md. Roitt, I., Brostoff, J., Male, D. 1998. Immunology. 5th ed. Mosby, London.
http://www.journals.wiley.com/cytometry/ John Wiley & Sons cytometry Web site. http://www.cytometry.org/ Clinical Cytometry Society Web site. http://dpalm2.med.uth.tmc.edu/edprog/00000014. htm Diagnostic flow cytometry tutorial. http://www.club.ch/scs/cytometr.htm Cytometry HotLinks; links to many flow cytometry– related sites. http://flosun.salk.edu Salk Institute flow cytometry laboratory home page.
Ruiz-Argüelles, A., Duque, R.E., and Orfao, A. 1998. Report on the First Latin American Consensus Conference for Flow Cytometric Immunophenotyping of Leukemia. Cytometry 34:3942.
http://www.ncrr.nih.gov/ncrrprog/btceltec.htm#fl ow
U.S. Department of Health and Human Services. Centers for Disease Control and Prevention. 1997. Revised Guidelines for Performing CD4+ T Cell Determinations in Persons Infected With Human Immunodeficiency Virus (HIV). MMWR 46:RR-2;1-29.
Beckman Coulter home page.
Internet Resources
National Flow Cytometry Resource. http://www.coulter.com
http://www.enterprise.net/appcysys/ Applied Cytometry Systems page. http://www.phnxflow.com Phoenix Flow Systems home page.
http://www.cyto.purdue.edu Purdue University cytometry labs. http://www.nih.gov National Institutes of Health home page.
Contributed by Alejandro Ruiz-Argüelles and Beatriz Pérez-Romano Laboratorios Clínicos de Puebla Puebla, Mexico
http://www.cdc.gov U.S. Centers for Disease Control home page. http://www.cyto.purdue.edu/step/stephome.htm ISAC Specialty Training and Education Program home page.
Immunophenotypic Analysis of Peripheral Blood Lymphocytes
6.5.14 Supplement 11
Current Protocols in Cytometry
Immunophenotypic Analysis of Human Mast Cells by Flow Cytometry
UNIT 6.6
The low frequency at which mast cells (MCs) are usually present in human bone marrow (BM), together with the lack of MC-specific antigens, has been an important limitation for the accurate direct identification and enumeration of MCs in different human body fluids and tissues. Furthermore, MCs are found in close relationship with the tissue stroma and stromal cells, which may represent an additional limitation on their accurate immunophenotypical identification, enumeration, and characterization in cell suspensions from BM or other tissue samples. On the other hand, flow cytometry is a technology well suited for the analysis of single-cell suspensions, even when such cells are present at very low frequencies. This is due, at least to a certain extent, to the multiparametric analytical capabilities of flow cytometry. In this sense, previous studies have shown that flow cytometry can be successfully applied for the identification and enumeration of rare events in samples from normal individuals or subjects suffering from different pathological conditions. The enumeration of CD34+ hemopoietic progenitor cells (HPCs), the identification and monitoring of minimal residual disease in BM samples from patients with hematological malignancies who achieved morphologic complete remission, as well as the immunophenotypic identification and enumeration of human (bone marrow mast cells) BMMCs, represent a clear demonstration of the utility of flow cytometry for rare event analysis. The use of a double-step acquisition procedure, in which an appropriate live-gate acquisition is performed, has largely contributed to the detection and study of small populations of cells present not only in human BM, but also in other single-cell suspension samples. This method allows identification of cells even when their frequency is as low as one event in up to a total of 106 unwanted cells. Additionally, this approach has overcome the problem associated with the acquisition and storage of multiparametric information on a large number of events, typically more than 106 or 107. Based on this approach, previous studies have shown that BMMCs are clearly identifiable by flow cytometry on the basis of their light-scatter properties and strong CD117 expression. These CD117+ cells are negative for the CD34 antigen. In addition, they are CD33+ and express significant amounts of the high-affinity IgE receptor (FcεRI). Due to the unique biological characteristics of MCs, special attention should be devoted to specific flow cytometry technical questions related to sample manipulation, staining procedures, and reagents to be used, as well as data acquisition and analysis. The basic protocols described in this unit allow the identification and enumeration of BMMCs using multiparametric flow cytometry on the basis of technical approaches used for the identification of cells present at low frequencies. This approach can be applied to a wide range of specimens, including BM (see Basic Protocol 1), peripheral blood (PB), ascitic fluid (see Basic Protocol 2), and lymphoid tissue, e.g., adenoids (see Basic Protocol 3), among others. In addition, the Support Protocol describes flow cytometric methods for the immunophenotypical characterization of MCs in different tissues, both from normal subjects and from patients suffering from mastocytosis and other pathological conditions. Special emphasis is placed on cell markers that have been suggested to be of great utility to distinguish between normal, reactive, and pathological MCs. From the clinical point of view, it has been suggested that the immunophenotypical analysis of MCs could be of great utility to support the diagnosis of tissue involvement in mastocytosis. Phenotypic Analysis Contributed by Luis Escribano, Raquel Navalón, Rosa Núñez, Beatriz Díaz Agustín, and Pilar Bravo
6.6.1
Current Protocols in Cytometry (2000) 6.6.1-6.6.18 Copyright © 2000 by John Wiley & Sons, Inc.
Supplement 12
BASIC PROTOCOL 1
IDENTIFICATION AND ENUMERATION OF BONE MARROW MAST CELLS (BMMCs) This protocol allows the identification and enumeration of BMMCs using multi-parametric flow cytometry, on the basis of technical approaches used for the identification of cells present at low frequencies. Moreover, once identified, mast cells are suitable of being characterized from the phenotypic and the functional point of view, facilitating the comparison between normal and abnormal mast cells. Materials Source of BM samples Phosphate-buffered saline (PBS; APPENDIX 2A) 0.5% (w/v) toluidine blue in methanol, pH 3.0 Phycoerythrin (PE)-conjugated CD117 monoclonal antibody (CD117-PE MAb), appropriately titered Fluorescein isothiocyanate (FITC)-conjugated anti-IgE monoclonal or polyclonal Ab (anti-IgE-FITC Ab) or FITC-conjugated CD33 MAb(CD33-FITC), appropriately titered FITC-conjugated CD34 MAb (CD34-FITC), appropriately titered PE-cyanine 5 (PE-Cy5 tandem fluorochrome)–conjugated CD45 (CD45-PE-Cy5 MAb) or peridin chlorophyll protein (PerCP)–conjugated CD45 MAb (CD45-PerCP MAb), appropriately titered 1× ammonium chloride lysing solution (APPENDIX 2A), freshly prepared, or 1× Quicklysis (IMICO) PBS containing 1% paraformaldehyde (optional) 14- to 8-G biopsy needle Vacutainer tubes containing either tripotassium EDTA or heparin 25-G needle and syringe Hematology analyzer 12 × 75–mm polystyrene tubes Flow cytometer with at least three fluorescence detectors and appropriate filter sets for detection of FITC, PE, and PE-Cy5 or PerCP Software for analysis of flow cytometry data: e.g., Paint-A-Gate Pro (Becton Dickinson) NOTE: Each vial of anti-IgE Ab should be stored frozen in aliquots. Each aliquot should be titrated before use (basophils should be used as target positive control cells) and kept at 4° to 8°C until finished. Collect and prepare samples 1. Perform BM aspiration in the posterior iliac spina, using a 14- to 8-G biopsy needle in order to obtain enough BM particles. Collect 1.5 to 2 ml of bone marrow using tubes containing tripotassium EDTA or heparin as anticoagulant. Pass the aspirate 2 or 3 times through a 25-G gauge needle in order to disaggregate the bone marrow particles. Since MCs are attached to the stroma and stromal cells, BM aspiration should be performed firmly and quickly. The harvest of higher volumes of sample in a single aspirate will not increase the number of MCs collected.
2. Prepare two BM smears according to conventional methods. 3. Assess the nucleated cell count in a conventional hematology analyzer. If necessary, adjust final cell concentration with PBS to 7.5 × 109 nucleated cells/liter. Immunophenotypic Analysis of Human Mast Cells
4. Stain an air-dried BM smear containing BM particles by immersing for 4 min in 0.5% toluidine blue and then washing in water.
6.6.2 Supplement 12
Current Protocols in Cytometry
The staining makes it possible to morphologically control for the relative BMMC content of the sample and get an overall impression on whether it is low, normal, or high.
Stain samples 5. Label two 12 × 75–mm polystyrene tubes: one to receive a control for the FITC-FL1 (green fluorescence) detector-associated baseline autofluorescence levels of MCs (AutoFL-green/CD117-PE/CD45-PerCP or -PE-Cy5), and the other to contain a triple-staining combination for the specific identification and enumeration of BMMCs (anti-IgE-FITC/CD117-PE/CD45-PerCP or -PE-Cy5). Although this is usually not necessary, in those BM samples with a high proportion of CD34+ HPC expressing high amounts of CD117 and having low CD45 levels, a third triple-staining combination of MAbs (CD34-FITC/CD117-PE/CD45-PerCP or CD45-PECy5) should be used, to specifically exclude the presence of CD34+ within the MC gate. If the number of MCs seems to be low after examination of the smear stained with toluidine blue, the above tubes should be prepared in duplicate or even in triplicate in order to get a minimum number of BMMCs for analysis. If anti-IgE reagents are not available in one’s laboratory, they may be replaced with an anti-CD33-FITC monoclonal antibody.
6. Add 200 µl sample, containing ∼1.5 × 106 nucleated cells, to each tube. 7. To every tube, add saturating amounts of each of the following MAbs: CD117-PE and CD45-PerCP or CD45-PE-Cy5. Finally, add anti-IgE-FITC to tube 2. If tube 3 (CD34-FITC/CD117-PE/CD45-PerCP or CD45-PE-Cy5) is used, add CD34FITC to it. If anti-IgE reagents are not available, use CD33-FITC in tube 2.
8. Incubate tubes 15 min at room temperature, protected from light. 9. Add 2 ml fresh 1× ammonium chloride lysing solution to each tube. Vortex vigorously 5 sec and incubate all tubes another 15 min at room temperature, protected from light. 10. Centrifuge 5 min at 540 × g, room temperature, discard the supernatant, and resuspend the cell pellet from each tube in 4 ml PBS. 11. Centrifuge again 5 min at 540 × g, discard the supernatant, and resuspend each cell pellet in 500 µl PBS. Samples can be run immediately. Alternatively, they can be resuspended in PBS containing 1% paraformaldehyde and left at 4°C for a maximum of 6 hr.
Samples should be gently mixed immediately prior to data acquisition. Set up instrument and acquire data 12. Position forward scatter (FS), side scatter (SS), and fluorescence (FL) windows of analysis and adjust electronic correction for spectral overlap according to standard procedure for a lyse-then-wash immunophenotyping assay. 13. Acquire data from a total of 50,000 events for the first tube (AutoFL-green/CD117PE/CD45-PerCP or CD45-PE-Cy5). Create bivariate histograms (dot plots) as follows (see Fig 6.6.1). Histogram 1: FS versus SS (Fig. 6.6.1A) Histogram 2: CD117-PE (orange fluorescence) versus SS (Fig. 6.6.1B) Histogram 3: CD117-PE (orange fluorescence) versus CD45-PerCP or CD45-PE-Cy5 (red fluorescence) (Fig. 6.6.1C) Histogram 4: green fluorescence versus CD117-PE (orange fluorescence) (Fig. 6.6.1D) For the Becton Dickinson FACS series of instruments, fluorochromes emitting red fluorescence such as PE-Cy5 are assessed in the fluorochrome detector identified as FL3, while
Phenotypic Analysis
6.6.3 Current Protocols in Cytometry
Supplement 12
B 1000
1000
800
800
600
600
SS
SS
A
400
400
200
200 0
0 0
200 400
600 FS
800 1000
C
101
102 CD117
103
104
100
101
102 FITC
103
104
103
104
D
104
104
CD 117 BD
103 CD45
100
102 101
103 102 101
100
100 100
101
102
103
104
100
101
CD117-PE
E
1000
SS
800 600 400 200 0 102 CD117
Figure 6.6.1 Representative dot plots of data acquisition for the identification of BMMCs using Basic Protocol 1. In this example, first a total of 20,000 events/tube were acquired, corresponding to the total nucleated BM cells (A to D). In the second step, acquisition through a live gate drawn on SS/CD117+/++ was performed (B) and only the events included in this region (R1) were stored (E). Data were acquired and analyzed using CellQuest software.
for Beckman Coulter Epics XL instruments these fluorochromes should be measured in the fluorescence channel identified as FL4. Immunophenotypic Analysis of Human Mast Cells
14. Create live gate regions: in the CD117-PE versus SS histogram (tube 1), display all events. Draw a rectangular gate (R1) to include all CD117+ events (see Figs. 6.6.1B and 6.6.1E).
6.6.4 Supplement 12
Current Protocols in Cytometry
15. Acquire a minimum of 100 CD117+++ events in the region described above, for tube 1 (AutoFL-green/CD117-PE/ CD45-PerCP or -PE-Cy5). Due to the low frequency at which BMMCs are usually found (<0.1% of total nucleated cells), to acquire information on a minimum of 100 BMMCs a total of >100,000 events should be measured. Because of the limitations on computer memory, it is recommended that in this second acquisition step only the information for those events contained in the BMMC region (R1), as shown in Figures 6.6.1B and 6.6.1E, be stored. When acquiring data through a live gate, the total number of events analyzed during the acquisition of this file should be specifically recorded. The total number of cells to be acquired depends on the percentage of BMMCs present in the sample. To assure statistically reliable results, 0.5 × 106 (BMMC frequency of 0.1%) to 1.5 × 106 (BMMC frequency of 0.001%) total events should be acquired. If the expected percent of BMM’s is lower than 0.05%, the authors recommend preparation of duplicate tubes. Be sure that CD117++/+++ cells can be clearly discriminated from CD117dim+ cells (i.e., CD34+ HPC). See Critical Parameters for discussion of choice of MAb and fluorochromes.
16. Before the next sample tube, run a tube containing PBS for 1 min to flush fluids and prevent specimen carryover. Flushing the system with PBS is not required between duplicates, but should be done before another pair of duplicate samples is run.
17. Acquire data from a total of 50,000 events for the second tube (anti-IgEFITC/CD117-PE/CD45-PerCP or -PE-Cy5). Use the same bivariate histograms (dot plots) described in step 13. The gating region created in step 14 could also be applied for this second tube. If anti-IgE is not available, CD33-FITC can be used. If a third tube was prepared, create bivariate histograms (dot plots) as follows: histogram 1, FS versus SS; histogram 2, CD117-PE (orange fluorescence) versus SS; histogram 3, CD117-PE (orange fluorescence) versus CD45-PerCP or CD45-PE-Cy5 (red fluorescence); histogram 4, CD34-FITC (green fluorescence) versus CD117-PE (orange fluorescence).
18. Repeat step 15 for tube 2. If a third tube was prepared (CD34-FITC/CD117-PE/CD45-PerCP or CD45-PE-Cy5), after step 18 follow with steps 16 to 18 for this specific tube.
Analyze data Data analysis can be performed with any of the software programs commercially available for the analysis of flow cytometry FCS files. In the examples shown, the Paint-A-Gate Pro software program (BD) was used. 19. Analyze the first tube (AutoFL-green/CD117-PE/CD45-PerCP or CD45-PE-Cy5). Create bivariate histograms as follows. Histogram 1: FS versus transformed SS Histogram 2: MAb-PE (orange fluorescence) versus SS Histogram 3: MAb-FITC (green fluorescence) versus MAb-PE (orange fluorescence) Histogram 4: MAb-PE (orange fluorescence) versus MAb-PerCP or MAbPE-Cy5 (red fluorescence) The Paint-A-Gate Pro software allows a polynomial transformation of SS (transformed SS), which improves the discrimination between cell populations with low/intermediate SS values.
Phenotypic Analysis
6.6.5 Current Protocols in Cytometry
Supplement 12
20. Draw a polygonal region (R2) to remove debris and platelets from the FS/SS histogram (histogram 1) for the data file corresponding to the total acquisition of events from tube 1 (see Fig. 6.6.2A). If one’s software program does not include the option of SS polynomial transformation, it may be used in linear scale.
21. Calculate the percentage of events corresponding to nucleated cells as the percent of events from this file contained in R2 (Fig. 6.6.2A). The following steps are referred to the second file acquired for the first tube after applying region R1 to specifically select the CD117+ events (Fig. 6.6.2B).
22. Draw a rectangular region (R3) to include cells with a high expression of CD117 in histogram 2 (SS versus CD117-PE) (see Fig. 6.6.2B). 23. In histogram 4, select cells that fill the criteria of high expression of CD117 and homogeneous positive expression of CD45 in a new region (R4) (see Fig. 6.6.2C). 24. In histogram 1, select BMMCs by drawing a region (R5) on a FS/SS bivariate dot plot to include cells with intermediate/high FS/SS characteristics (see Fig. 6.6.2D). 25. In histogram 3, display events fulfilling the criteria of R1 and R2 and R3 and R4. Draw a region around CD117+++ cells (R6) as shown in Figure 6.6.2E. These should correspond to BMMCs.
26. Calculate the absolute number of events that fulfill the characteristics described for MCs. Calculate the mean channel and the coefficient of variation (CV) of the green autofluorescence and the mean fluorescence intensity and CV for both CD117-PE and CD45-PerCP or CD45-PE-Cy5 antigens. 27. Analyze the files acquired for the second tube. Repeat steps 19 to 24 for the two files stored for this tube. 28. In histogram 3, select CD117strong+ and IgE+ events (R7) as those being specifically identified as corresponding to BMMCs (see Fig. 6.6.2F). 29. Repeat step 26 for the second file stored for tube 2. If CD33 was used instead of anti-IgE in step 28, in R6 select only cells that are strongly positive for CD33. If a third tube (CD34-FITC/CD117-PE/CD45-PerCP or CD45-PECy5) was acquired, repeat steps 19 to 24 and then exclude CD34+ cells from the calculation of the total number of events corresponding to BMMCs in a CD34-FITC/CD117-PE bivariate plot (histogram 3).
Enumerate BMMCs 30. Calculate the proportion of MCs present in each sample as a fraction of the total nucleated cells using the following formula. %BMMC =
number of BMMCs × 100 number of nucleated cells
where the number of BMMCs corresponds to the events identified as BMMCs in step 28 and the number of nucleated cells corresponds to the total number of events acquired in the second file stored for tube 2 upon excluding cell debris and platelets, as follows. Immunophenotypic Analysis of Human Mast Cells
Total number of events acquired in file 2 of tube 2 × % of events in R2 100
6.6.6 Supplement 12
Current Protocols in Cytometry
A
B 1024
1024
768
Transformed SS
Transformed SS
RI/GATE
512 256
R2
0 0
256
C
512 FS
768
256
R3
0 100
D
101 102 103 CD117-PE
104
1024
Transformed SS
103 CD45-PerCP
512
1024
104
102 101
R4
100 100
E
101
102 103 CD117-PE
768 512 256
R5
0
104
0
256
F
104
104
103
103 CD117-PE
CD117-PE
768
102 101
512 FS
768
1024
102 101
R6
100 100
101 102 103 Control-FITC
104
R7
100 100
101
102 103 lgE-FITC
104
Figure 6.6.2 Representative dot plots for data analysis of BMMCs using the Paint-A-Gate Pro software program (Becton Dickinson). Dot plot A shows the total acquisition, in which black dots correspond to nucleated cells. Dot plots B to F correspond to gated cells selected on SS/CD117+ live gate R1 (Basic Protocol 1). Panel C displays the events fulfilling the criteria of R1 and R2 and R3; in dot plot D the events represented are those fulfilling the criteria for R1 and R2 and R3 and R4; events shown in dot plots E and F are those fulfilling the criteria for R1 and R2 and R3 and R4 and R5 for the AutoFL-green/CD117/CD45 and anti-IgE/CD117/CD45 tubes, respectively, and correspond to BMMCs.
Phenotypic Analysis
6.6.7 Current Protocols in Cytometry
Supplement 12
BASIC PROTOCOL 2
IDENTIFICATION AND ENUMERATION OF ASCITIC FLUID MAST CELLS This protocol allows the identification and enumeration of ascitic fluid MCs using multiparametric flow cytometry. The methodological approach is similar to that used in Basic Protocol 1, except for the sample preparation. The goal of this protocol is to determine MC content in ascitic fluid in different pathological conditions, including mastocytosis. Materials Ascitic fluid Phosphate-buffered saline (PBS; APPENDIX 2A) 50-ml centrifuge tubes (e.g., Falcon, Becton Dickinson) Hematology analyzer Additional reagents and equipment for identification and enumeration of BMMCs (see Basic Protocol 1) 1. Divide the sample into 40- to 45-ml aliquots in 50-ml tubes and centrifuge 10 min at 540 × g, room temperature. 2. Discard the supernantants and resuspend each pellet in 1 ml PBS. Transfer all the pellets to a single 50-ml tube. 3. Assess the nucleated cell count of the concentrated sample in a hematological cell analyzer. If the nucleated cell count exceeds 10 × 109 nucleated cells/liter, dilute the sample with PBS to reach a final concentration of 10 × 109 nucleated cells/liter. 4. Stain sample (see Basic Protocol 1, steps 5 to 11). 5. Set up instrument and acquire data (see Basic Protocol 1, steps 12 to 18). 6. Analyze data (see Basic Protocol 1, steps 19 to 29). An example of how MCs are identified using the data analysis procedure described in Basic Protocol 1 is shown in Figure 6.6.3.
BASIC PROTOCOL 3
IDENTIFICATION AND ENUMERATION OF LYMPHOID TISSUE MAST CELLS This protocol makes it possible to identify and enumerate lymphoid tissue MCs by multiparametric flow cytometry. Because of tissue characteristics, mechanial disaggregation is required to obtain a single-cell suspension. Materials Fresh samples from lymphoid tissue (tonsils, adenoids, lymph nodes) Phosphate-buffered saline (PBS; APPENDIX 2A) 90-mm petri dishes Surgical blades Tweezers Mechanical disaggregation system (e.g., Medimachine with 50-µl Medicons; Dako) Additional reagents and equipment for identification and enumeration of BMMCs (see Basic Protocol 1)
Immunophenotypic Analysis of Human Mast Cells
1. Prepare a 90-mm petri dish containing 5 to 10 ml PBS. Place a fragment of fresh tissue (obtained after surgical removal) in the prepared petri dish. Without delay, cut the tissue into small pieces (1 to 2 mm2) using surgical blades and tweezers.
6.6.8 Supplement 12
Current Protocols in Cytometry
A
B 1024
1024
768
Transformed SS
Transformed SS
RI/GATE
512 256
R2
0 0
256
C
512 FS
768
256
R3
0 100
D
101 102 103 CD117-PE
104
1024
Transformed SS
103 CD45-PerCP
512
1024
104
102 101
R4
100 100
101
102 103 CD117-PE
768 512 256
R5
0
104
E
0
256
512 FS
768
1024
F 104
104
103
103 CD117-PE
CD117-PE
768
102 101
102 101
R6
100 100
101 102 103 Control-FITC
104
R7
100 100
101
102 103 lgE-FITC
104
Figure 6.6.3 Representative dot plots for data analysis of MCs from ascitic fluid. Dot plot A shows the total acquisition, in which black dots correspond to nucleated cells. Dot plots B to F correspond to gated cells selected on SS/CD117+ live-gate R1 (Basic Protocol 2). Panel C displays the events fulfilling the criteria of R1 and R2 and R3; in dot plot D the events represented are those fulfilling the criteria for R1 and R2 and R3 and R4; events shown in dot plot E and F are those fulfilling the criteria for R1 and R2 and R3 and R4 and R5 for the AutoFL-green/CD117/CD45 and antiIgE/CD117/CD45 tubes, respectively, and correspond to MCs from ascitic fluid. Data were analyzed using the Paint-A-Gate Pro software program (Becton Dickinson).
Phenotypic Analysis
6.6.9 Current Protocols in Cytometry
Supplement 12
2. Immediately introduce the small tissue pieces into the tissue disgregation machine and perform mechanical disaggregation according to the manufacturer’s instructions. This procedure works extremely well with friable tissue, especially lymph nodes, and yields high number of single cells in a few minutes. Although manual mechanical disaggregation procedures can be applied, the authors recommend use of the Medimachine system.
3. Assess the nucleated cell count in a hematological cell analyzer and adjust cell concentration with PBS to 7.5 × 109 nucleated cells/liter. 4. Stain samples (see Basic Protocol 1, steps 5 to 11). 5. Set up instrument and acquire data (see Basic Protocol 1, steps 12 to 18). 6. Analyze data (see Basic Protocol 1, steps 19 to 29). An example of how MCs are identified in hyperplastic adenoids, using the data analysis procedure described in Basic Protocol 1, is shown in Figure 6.6.4. SUPPORT PROTOCOL
IMMUNOPHENOTYPICAL CHARACTERIZATION OF MAST CELLS Based on the flow cytometric characterization of human MCs, this immunophenotypic procedure allows one to distinguish between normal and pathological MCs. Any other markers of interest can be analyzed using this protocol, including either surface or intracellular proteins. Based on the authors’ experience, good results for the intracellular staining of bcl-2 are obtained using Fix & Perm (Caltag), Intrastain (Dako), and Intraprep (Immunotech) fixation and permeabilization reagents. The procedure described below can be applied to any of those three reagents. Materials Appropriately titered FITC-conjugated monoclonal antibodies: CD2, CD25, CD35, CD69, and anti-bcl2 Samples of bone marrow (see Basic Protocol 1, steps 1 to 4), ascites fluid (see Basic Protocol 2, steps 1 to 3), or lymphoid tissue (see Basic Protocol 3, steps 1 to 3) Reagents for cell fixation and permeabilization to assess the cytoplasmic expression of the bcl-2 protein (e.g., Fix & Perm, Caltag; Intrastain, Dako; or Intraprep, Immunotech) 12 × 75–mm polystyrene tubes Perform surface staining 1. Label eight 12 × 75–mm polystyrene tubes as follows. Tube 1 (after surface stainings): AutoFL-green/CD117-PE/CD45-PerCP (or CD45-PE-Cy5) Tube 2 (after surface plus intracellular staining): AutoFL-green/CD117PE/CD45-PerCP (or CD45-PE-Cy5) These two tubes are for the assessment of baseline autofluorescence levels of MCs.
Immunophenotypic Analysis of Human Mast Cells
Tube 3: anti-IgE-FITC/CD117-PE/CD45-PerCP or CD45-PE-Cy5 Tube 4: CD2-FITC/CD117-PE/CD45-PerCP or CD45-PE-Cy5 Tube 5: CD25-FITC/CD117-PE/CD45-PerCP or CD45-PE-Cy5 Tube 6: CD35-FITC/CD117-PE/CD45-PerCP or CD45-PE-Cy5 Tube 7: CD69-FITC/CD117-PE/CD45-PerCP or CD45-PE-Cy5 Tube 8: anti-bcl-2-FITC/CD117-PE/CD45-PerCP or CD45-PE-Cy5 If the amount of MCs in the sample seems to be low in the toluidine blue–stained BM smear, or is expected to be low in other samples, prepare tubes in duplicate or triplicate.
6.6.10 Supplement 12
Current Protocols in Cytometry
B 1024
1024
768
768
Transformed SS
Transformed SS
A
512
256 R2
0 0
256
512 FS
C
768
R2 100
1024
Transformed SS
CD45-PerCP
103 102
101
102 103 CD117-PE
104
1024
768
512
101
256
100
0 101
102 103 CD117-PE
104
F
104 103 102 101
R5 0
CD117-PE
CD117-PE
256
R4
100
E
512
0
D
104
R1/GATE
256
512 FS
768
1024
104 103 102 101
R6
100 100
101 102 103 Control-FITC
104
R7
100 100
101 102 LGE-FITC
103
104
Figure 6.6.4 Representative dot plots of data analysis of MCs from hyperplastic reactive adenoids Dot plot A shows the total acquisition, in which black dots correspond to nucleated cells. Dot plots B to F correspond to gated cells selected on SS/CD117+ live gate R1 (Basic Protocol 3). Panel C displays the events fulfilling the criteria of R1 and R2 and R3. In dot plot D the events represented are those fulfilling the criteria for R1 and R2 and R3 and R4. Events shown in dot plots E and F are those fulfilling the criteria for R1 and R2 and R3 and R4 and R5 for the AutoFL-green/CD117/CD45 and anti-IgE/CD117/CD45 tubes, respectively, and correspond to MCs from hyperplastic reactive adenoids. Analysis was performed using the Paint-A-Gate Pro software program (Becton Dickinson). Phenotypic Analysis
6.6.11 Current Protocols in Cytometry
Supplement 12
2. Add cells to tube 1 and to tubes 3 to 7 and stain with the appropriate reagents (see Basic Protocol 1, steps 6 to 11). Stain for intracellular bcl-2 protein 3. Add 200 µl cell suspension containing ∼1.5 ×106 nucleated cells, respectively, to tube 2 and to tube 8. 4. Add saturating amounts of each of the following MAbs to tubes 2 and 8: CD117-PE and CD45-PerCP or CD45-PE-Cy5. Gently mix and incubate both tubes 15 min at room temperature, protected from light. 5. Add 4 ml PBS to each tube, vortex 5 sec, and centrifuge both tubes 5 min at 540 × g, room temperature. 6. Using a Pasteur pipet, remove and discard the supernatant, resuspend the cell pellet, and add 100 µl fixation solution. Gently mix. The commerical fixation/permeabilization kits from Caltag, Dako, and Immunotech (see materials list above) contains two reagents, reagent A which is a fixation medium and reagent B which is a permeabilization medium. Reagent A, the fixation solution, is formalin-based.
7. Incubate 15 min at room temperature, protected from light. 8. Wash once in 4 ml PBS as described in step 5. 9. Using a Pasteur pipet, remove and discard the supernatant, leaving a volume of <200 µl at the bottom of the tube. 10. Add 100 µl permeabilization solution to each of the two tubes; additionally place saturating amounts of the anti-bcl-2 MAb into tube 8. Gently mix. The permeabilization solution is reagent B of the commercial kits (see annotation to step 6), and is saponin-based.
11. Incubate 15 min at room temperature, protected from light. 12. Wash once with 4 ml PBS/tube as described in step 5. Discard the supernantant and resuspend the cell pellet in 500 µl PBS. 13. Set up instrument and acquire data (see Basic Protocol 1, steps 12 to 18). 14. Analyze data (see Basic Protocol 1, steps 19 to 29) for the 8 different tubes. See Figure 6.6.5 for representative results. To evaluate the expression of the different antigens analyzed after staining with specific FITC-conjugated monoclonal antibody reagents, calculate the mean fluorescence intensity (MFI) as described in Basic Protocol 1 (step 26). A marker should be considered as positive on MC once its MFI is higher than 3 standard deviations above the MFI found for the baseline autofluorescence obtained in the green fluorescence detector for FITC-unstained MCs processed using the same sample-preparation protocol (either tube 1 for surface stainings or tube 2 for intracellular bcl-2 staining).
Immunophenotypic Analysis of Human Mast Cells
6.6.12 Supplement 12
Current Protocols in Cytometry
G
1024
Transformed SSA
Transformed SS
A
768 512 256 0 0 104 103 102 101 100
104 103 102 101
I
100
104 103
J
102 101
104 103
K CD117-PE
CD117-PE
100 101 102 103 104 CD25-FITC
102 101 100 100 101 102 103 104 CD35-FITC
256 512 768 1024 FS
104 103 102 101 100 101 102 103 104 CONTROL-FITC 104 103 102 101 100 101 102 103 104 CD2-FITC 104 103 102 101
104 103 102 101 100 100 101 102 103 104 CD69-FITC
100 101 102 103 104 CD25-FITC 104 103 102 101 100
L CD117-PE
CD117-PE
F
0
100
100
E
256
100 100 101 102 103 104 CD2-FITC
CD117-PE
CD117-PE
D
512
100
CD117-PE
CD117-PE
C
100 101 102 103 104 CONTROL-FITC
768
0
H CD117-PE
CD117-PE
B
256 512 768 1024 FS
1024
100 101 102 103 104 CD35-FITC 104 103 102 101 100 100 101 102 103 104 CD69-FITC
Figure 6.6.5 Representative dot plots showing the light scatter and immunophenotypic differences between BMMCs (black dots) from a reactive bone marrow (dot plots A to F) and a case of indolent systemic mast cell disease (dot plots G to L). The events represented in each plot correspond to the CD117+ cells live-gated through R1.
Phenotypic Analysis
6.6.13 Current Protocols in Cytometry
Supplement 12
COMMENTARY Background Information Mast cells (MCs) are long-living cells derived from precursors that originate in the bone marrow (Kitamura et al., 1977, 1978, 1979). MCs are present at very low frequencies in normal human BM. In histological sections, MCs are distributed especially near the endosteum, in the periosteum, in association with small blood vessels, and at the periphery of lymphoid nodules or aggregates (Johnstone, 1954). Indeed, MCs are found in close relationship with the stroma and stromal cells, which may impose great limitations on the flow cytometric immunophenotypical identification, enumeration, and characterization of BM suspensions. Quantitative evaluation of MCs has been attempted in several tissues and pathological conditions using different techniques. In any case, the enumeration of the MCs present in a sample has never been achieved with accuracy, in spite of the fact that the existence of increased numbers of MCs in different tissues, including BM, may have clinical significance in several pathologic conditions (Yoo et al., 1978; Prokocimer and Polliack, 1981; Sale and Marmont, 1981; Yoo and Lessin, 1982; McKenna, 1994). The low frequency at which MCs are usually present in human BM and the lack of MC-specific antigens have been important stumbling blocks to accurate MC identification. Flow cytometry is well suited for the analysis of singlecell suspensions, even when such cells are present at very low frequencies (San Miguel et al., 1997; Ciudad et al., 1998). In recent years, it has been shown that the combined use of multiple stainings and flow cytometry allows the sensitive detection of rare cells (Gross et al., 1993; Orfao et al., 1994; Macedo et al., 1995). In this sense, it has also been shown (Orfao et al., 1994) that the use of a double-acquisition procedure allows the detection of small populations of leukemic cells in complete-remission human bone marrow samples, even when their frequency is as low as one leukemic cell in 106 normal cells. In spite of the fact that in the last decade flow cytometry immunophenotypic studies have rapidly expanded from research to clinical diagnostic laboratories, only recently has this technique been applied to the analysis of BMMCs. A major limitation for the flow cytometric study of BMMCs was, at least from
the theoretical point of view, the low frequency at which these cells are present in BM samples. In spite of the fact that BMMCs represent only a very small proportion of all nucleated cells present in normal BM, it has recently been shown that they can be specifically identified and accurately enumerated using multiparametric flow cytometry (Orfao et al., 1996). In the authors’ experience, BMMCs are clearly identifiable on the basis of their light-scatter properties and strong CD117 expression. These cells are negative for the CD34, CD38, and CD138 antigens. In addition, they are CD33+ and display a high reactivity for the anti-IgE monoclonal antibody. This methodological approach allows a systematic analysis of the immunophenotypic characteristics of normal human BMMCs. Normal BMMCs are virtually always positive for the CD9, CD11c, CD29, CD33, CD43, CD44, CD45, CD49d, CD49e, CD51, CD54, CD71, CD117, and FcεRI antigens, while other markers such as CD11b, CD13, CD18, CD22, CD35, CD40, and CD61 display a variable expression in normal individuals. Furthermore, BMMCs from patients suffering from different hematological malignancies show phenotypes identical to those for BMMCs from normal individuals (Escribano et al., 1998). When this flow cytometric approach was applied to the analysis of BMMCs from mastocytosis patients, clear immunophenotypical differences were found between BMMCs from normal individuals and from adult patients suffering from malignant mastocytosis (Escribano et al., 1995, 1997a) or indolent systemic mast cell disease (SMCD; Escribano et al., 1997b). Accordingly, the most characteristic immunophenotypic feature of malignant mastocytosis was the coexpression of CD2 and CD25 antigens on the surface of BMMCs (Escribano et al., 1995, 1997a). Neither of these two molecules is present in normal MCs (Escribano et al., 1998). These findings suggest that the expression of both antigens could be considered aberrant; interestingly, BMMCs from malignant mastocytosis patients also displayed a clear reactivity for CD35, CD63, CD69, and C D71 activatio n- asso ciated markers (Escribano et al., 1997a). Furthermore, the overexpression of bcl-2 protein seems to be characteristic of BMMCs from malignant mas-
Immunophenotypic Analysis of Human Mast Cells
6.6.14 Supplement 12
Current Protocols in Cytometry
tocytosis but not of normal BMMCs or BMMCs from patients suffering from indolent mastocytosis or different hematological and nonhematological diseases (Cerveró et al., 1999). Further flow cytometric immunophenotypical studies of BMMCs from adult patients with cutaneous mastocytosis and BM involvement, as assessed by histological analysis, confirmed previous findings on malignant mastocytosis. Accordingly, both CD2 and CD25 antigens are expressed on BMMCs from all adult indolent SMCD, supporting the notion that coexpression of both markers is an aberrant hallmark of BMMCs from adult mastocytosis. In line with these observations, other activation-associated antigens such as CD35 (Escribano et al., 1997b) and CD69 (Díaz Agustín et al., 1999) are overexpressed in BMMCs from adult SMCD. The above protocols constitute a methodological approach that allows not only an accurate identification and enumeration of MCs from different sources, but also their immunophenotypical characterization and the diagnosis of tissue involvement on the basis of the aberrant expression of CD2 and CD25 antigens, together with the overexpression of the CD35 and CD69 activation-associated antigens. Variations on these protocols could also be made in the future for the functional analysis of MCs.
Critical Parameters and Troubleshooting Sample quality Since BMMCs are closely attached to the stroma and stromal cells, BM aspiration should be performed firmly and quickly, in order to obtain a sufficient number of BM fragments. Once obtained, the BM sample should be passed several times through a 25-G needle to assure a total dispersion of BM particles. BM samples obtained without particles could be adequate for other immunophenotypical studies but not for MC enumeration. For the enumeration of BMMCs, fresh samples with high cell viability (>95%) should be used. MC counts in samples processed more than 3 hr after BM aspiration should be considered as potentially worthless. Regarding immunophenotypical studies for characterization of BMMCs, the time from sample collection to technical processing may not represent a limiting factor if the MCs present in the sample are >0.1%.
This time parameter is especially critical for the enumeration of MCs from ascitic fluid as well as from lymph nodes, for which both the MC count and the immunophenotypical characterization studies should be performed immediately after the drainage of ascitic fluid or the lymph node biopsy. Samples obtained from tonsils or other solid tissues could be considered adequate for the identification and immunophenotypical characterization of MCs, but probably not for their enumeration. It should be noted that, at present, information on this subject is lacking in the literature. In order to obtain adequate training in immunophenotypic studies of MCs on the flow cytometer, a variety of samples should be examined, including, among others, samples having MC hyperplasia, such as reactive BM, BM from patients with Waldenström macroglobulinemia, and some cases of chronic lymphocytic leukemia. The identification of BMMCs may be difficult in some circumstances, mainly when samples contain a very low number of MCs, as well as when cells in the sample other than MCs overexpress CD117. These latter cases include some patients suffering from acute myeloid leukemia, myelodysplastic syndromes, or multiple myeloma (reviewed in Escribano et al., 1999). In these cases, the use of a triple MAb combination—including the anti-IgE Ab for the specific identification of BMMCs—together with the analysis of the expression on CD117+++ cells of antigens specific for other neoplastic cell populations in the sample that overexpress CD117 (i.e., CD34 or CD138)— may be of great help for accurate distinction between the two cell populations, MCs and neoplastic cells. Choice of MAb and fluorochromes Since CD117, CD45, and anti-IgE are the key antigens for the sensitive and specific identification of MCs, clones and fluorochromeconjugated Ab reagents directed against those markers should be carefully selected on the basis of intensity of staining on MCs and the absence of nonspecific staining of other cells such as monocytes. For that purpose, several CD117-PE, anti-IgE-FITC, and CD45-PE-Cy5 or CD45-PerCP antibody reagents are currently available. The authors have had satisfactory experience with the following as reference reagents for the detection of those three antigens: Phenotypic Analysis
6.6.15 Current Protocols in Cytometry
Supplement 12
(1) the 104D2 clone for CD117-PE (Becton Dickinson); (2) the J33 clone for CD45-PECy5 (Immunotech) and the 2D1 clone for CD45-PerCP (Becton Dickinson); and (3) FITC-conjugated polyclonal anti-IgE Ab (Caltag Laboratories). An alternative polyclonal anti-IgE-FITC that has performed well is the one from Binding Site, which unfortunately is not directly available at present. Titration of this reagent can be performed on normal peripheral blood basophils as mentioned above (see Basic Protocol 1). The choice of FITC-conjugated CD34 MAb should be restricted to anti-class III reagents; the 8G12-HPCA-2 clone is a good choice as reference reagent. In the authors’ experience, good results are obtained for the analysis of CD2, CD25, CD35, CD69, and anti-bcl-2 expression on MCs with the following FITC-conjugated MAb clones: CD2, clone S5.2 (Becton Dickinson); CD25, clone 2A3 (Becton Dickinson); CD35, clone E11 (Serotec, CLB, Cymbus Biosciences); CD69, clone LT8 (Becton Dickinson); and antibcl-2, clone 124 (Dako). In all cases, these mAbs or ones from other sources should be used only after adequate testing has been performed. New lots of material should always be compared against the old. Assessment of nonspecific fluorescence of human MCs Assessment of nonspecific fluorescence in the green fluorescence detector for human MCs is mandatory. In spite of the absence of nonspecific binding of the above mentioned reagents to human MCs, these cells display relatively high levels of green autofluorescence, specifically in adults suffering from systemic mast cell disease. This should also be taken into account if quantitative measurements of antigen expression in either fluorescence/fluorochrome or antibody binding capacity units are performed. Number of events to be acquired To assess the statistical reliability of MC counts obtained by flow cytometry, it is very important to acquire a minimum number of events corresponding to mast cells. As MCs are usually rare events, in the protocols described above for the enumeration of human MCs in different samples, the authors recommend acquiring and storing information on at least 100 events corresponding to MCs. For that purpose,
the total number of events to be acquired may vary from 0.3 × 106 to 3.5 × 106, depending on the proportion of MCs present in the sample. With this approach, the coefficient of variation for replicates in the authors’ experience is always lower than 10%.
Anticipated Results Proportion of mast cell present in different tissues The percentage of MCs present in human BM samples from normal individuals varies between 0.002% and 0.08%. Higher BMMC numbers could be expected in BM samples from Waldenström macroglobulinemia, reactive BM, BM hypoplasia, and a variable percentage of chronic lymphocytic leukemia and myelodysplastic syndromes, among others. Regarding other types of samples, the expected proportion of MCs in ascitic fluid varies from 0.001 to 0.005% (R. Navalón and L. Escribano, unpub. observ.) and in hyperplastic adenoids it ranges from 0.02 to 0.07% (R. Núñez and L. Escribano, unpub. observ.). Immunophenotypic characteristics of human MCs The technical procedure described above represents a sensitive and specific method for the study of MC immunophenotype in a systematic way. Furthermore, it is a useful tool for the diagnosis of tissue involvement in mastocytosis on the basis of the aberrant expression of the CD2 (95%) and CD25 (100%) antigens, both molecules never being present in BMMCs from either normal individuals or subjects suffering from different hematological and nonhematological disease conditions. In a similar way, overexpression of the CD35 and CD69 activation-associated antigens has been reported exclusively for BMMCs from mastocytosis (Fig. 6.6.5). From the phenotypic point of view, malignant mastocytosis is associated with overexpression of the bcl-2 protein, which is never detected in BMMCs from nonmalignant cases, including indolent systemic mast cell disease.
Time Considerations
Disaggregation of the sample requires ∼5 min for BM samples and ∼10 min for lymphoid tissue samples. The preparation of ascitic fluid samples before staining requires ∼20 min.
Immunophenotypic Analysis of Human Mast Cells
6.6.16 Supplement 12
Current Protocols in Cytometry
Toluidine blue staining of BM smears requires 5 min. Staining requires 45 min, and instruments may be set up during this time. The dead time for the acquisition of the samples, both in the basic and support protocols, depends on the mast cell frequencies in the BM specimens and can vary from 1 to 7 min per tube.
Literature Cited Cerveró, C., Escribano, L., Orfao, A., Díaz Agustín, B., Bravo, P., Villarrubia, J., García-Sanz, R., Velasco, J.L., Herrera, P., Vargas, M., et al. 1999. Expression of bcl-2 by human bone marrow mast cells and its overexpression in mast cell leukemia. Am. J. Hematol. 60:191-195. Ciudad, J., San Miguel, J.F., López-Berges, M.C., Vidriales, B., Valverde, B., Ocqueteau, M., Mateos, G., Caballero, M.D., Hernández, J., Moro, M.J., et al. 1998. Prognostic value of immunophenotypic detection of minimal residual disease in acute lymphoblastic leukemia. J. Clin. Oncol. 16:3774-3781. Díaz Agustín, B., Escribano, L., Bravo, P., Herrero, S., Nuñez, R., Navalón, R., Torrelo, A., Cantalapiedra, A., Del Castillo, L., Villarrubia, J., et al. 1999. The CD69 early activation molecule is overexpressed in human bone marrow mast cells from adults with indolent systemic mast cell disease. Br. J. Haematol. 106:400-405. Escribano, L., Orfao, A., Villarrubia, J., Cerveró, C., Velasco, J.L., Martin, F., San Miguel, J.F., and Navarro, J.L. 1995. Expression of lymphoid-associated antigens in mast cells: Report of a case of systemic mast cell disease. Br. J. Haematol. 91:941-943. Escribano, L., Orfao, A., Villarrubia, J., Martín, F., Madruga, J.I., Cuevas, M., Velasco, J.L., Rios, A., and San Miguel, J.F. 1997a. Sequential immunophenotypic analysis of mast cells in a case of systemic mast cell disease evolving to a mast cell leukemia. Cytometry 30:98-102. Escribano, L., Orfao, A., Díaz Agustín, B., Villarrubia, J., Cerveró, C., López, A., García Marcos, A., Bellas, C., Fernández Cañadas, S., Cuevas, M., et al. 1997b. Indolent systemic mast cell disease in adults. Immunophenotypic characterization and its diagnostic implications. Blood 91:2731-2736. Escribano, L., Orfao, A., Villarrubia, J., Díaz Agustín, B., Cerveró, C., Ríos, A., Velasco, J.L., Ciudad, J., Navarro, J.L., and San Miguel, J.F. 1998. Immunophenotypic characterization of human bone marrow mast cells: A flow cytometric study of normal and pathological bone marrow samples. Ann. Cell. Pathol. 16:151-159. Escribano, L., Díaz Agustín, B., Bravo, P., Navalón, R., Almeida, J., and Orfao, A. 1999. Immunophenotype of bone marrow mast cells in indolent
systemic mast cell disease in adults. Leuk. Lymphoma 35:227-235. Gross, H.J., Verwer, B., Houck, D., and Recktenwald, D. 1993. Detection of rare cells at a frequency of one per million by flow cytometry. Cytometry 14:519-526. Johnstone, J.M. 1954. The appearance and significance of tissue mast cells in the human bone marrow. J. Clin. Pathol. 7:275-280. Kitamura, Y., Shimada, M., Hatanaka, K., and Miyano, Y. 1977. Development of mast cells from grafted bone marrow cells in irradiated mice. Nature 268:442-443. Kitamura, Y., Go, S., and Hatanaka, K. 1978. Decrease of mast cells in W/Wv mice and their increase by bone marrow transplantation. Blood 52:447-452. Kitamura, Y., Hatanaka, K., Murakami, M., and Shibata, H. 1979. Presence of mast cells precursors in peripheral blood of mice demonstrated by parabiosis. Blood 53:1085-1088. Macedo, A., Orfao, A., Martínez, A., Vidriales, M.B., Valverde, B., López-Berges, M.C., and San Miguel, J.F. 1995. Immunophenotype of c-kit cells in normal human bone marrow: Implications for the detection of minimal residual disease in AML. Br. J. Haematol. 89:338-341. McKenna, M.J. 1994. Histomorphometric study of mast cells in normal bone, osteoporosis, and mastocytosis using a new stain. Calcif. Tissue Int. 55:257-259. Orfao, A., Ciudad, J., Lopez-Berges, M.C., Lopez, A., Vidriales, B., Caballero, M.D., Valverde, B., Gonzalez, M., and San Miguel, J.F. 1994. Acute lymphoblastic leukemia (ALL): Detection of minimal residual disease (MRD) at flow cytometry. Leuk. Lymphoma 13 Suppl. 1:87-90. Orfao, A., Escribano, L., Villarrubia, J., Velasco, J.L., Cerveró, C., Ciudad, J., Navarro, J.L., and San Miguel, J.F. 1996. Flow cytometric analysis of mast cells from normal and pathological human bone marrow samples. Identification and enumeration. Am. J. Pathol. 149:1493-1499. Prokocimer, M. and Polliack, A. 1981. Increased bone marrow mast cells in preleukemic syndromes, acute leukemia, and lymphoproliferative disorders. Am. J. Clin. Pathol. 75:34-38. Sale, G.E. and Marmont, P. 1981. Marrow mast cell counts do not predict bone marrow graft rejection. Hum. Pathol. 12:605-608. San Miguel, J.F., Martínez, A., Macedo, A., Vidriales, M.B., López-Berges, C., González, M., Cabaliero, D., García-Marcos, M.A., Ramos, F., Fernández-Calvo, J., et al. 1997. Immunophenotyping investigation of minimal residual disease is a useful approach for predicting relapse in acute myeloid leukemia patients. Blood 90:2465-2470.
Phenotypic Analysis
6.6.17 Current Protocols in Cytometry
Supplement 12
Yoo, D. and Lessin, L.S. 1982. Bone marrow mast cell content in preleukemic syndrome. Am. J. Med. 73:539-542. Yoo, D., Lessin, L.S., and Jensen, W.N. 1978. Bonemarrow mast cells in lymphoproliferative disorders. Ann. Intern. Med. 88:753-757.
Contributed by Luis Escribano, Raquel Navalón, Rosa Núñez, Beatriz Díaz Agustín, and Pilar Bravo Hospital Ramón y Cajal, Mast Cell Unit Madrid, Spain
The authors wish to acknowledge the support by grants from the Fondo de Investigaciones Sanitarias de la Seguridad Social (FIS 98/1345) and Fundación Oftalmológica J. Cortés Martínez. R. Navalón and R. Núñez are recipients of a grant from Comunidad de Madrid, Spain.
Immunophenotypic Analysis of Human Mast Cells
6.6.18 Supplement 12
Current Protocols in Cytometry
Measurement of CD40 Ligand (CD154) Expression on Resting and In Vitro–Activated T Cells Measurement of the ability of in vitro–activated T lymphocytes to express CD40 ligand (CD154) is a valuable tool for diagnosis of the X-linked hyper-IgM syndrome (XHIM), for detection of carriers of XHIM, and for investigating abnormalities in the costimulatory functions of T lymphocytes. CD154 is expressed primarily on activated CD4+ T cells and a small fraction of CD8+ cells. Resting peripheral blood lymphocytes do not normally express any appreciable levels of CD154, necessitating in vitro activation in order to detect it on the cell surface. The Basic Protocol details the in vitro activation procedure, the three-color monoclonal antibody panels used to label the appropriate lymphocyte subsets, the controls required to interpret the assay, the setup and acquisition of listmode data files, and the gating and analysis protocols used to interpret the data. CD154 is identified with a phycoerytherin-conjugated monoclonal antibody directed against CD154 on T helper cells. Although CD154 is primarily expressed on CD4+ T cells, CD4 cannot be used as a gating marker because it is actively down-modulated with this in vitro activation protocol (i.e., CD4 cannot be reliably detected following in vitro activation with phorbol myristate acetate and ionomycin). The protocol describes how to measure CD154 on CD4+ T cells using a negative gating strategy, i.e., labeling T cells with CD3 and CD8 and then drawing a gate on the T cells (CD3+) that do not express the CD8 antigen (CD3+CD8−). This negative gate contains the CD4+ T cells (as well as the CD4/CD8 double-negative cells).
UNIT 6.7
BASIC PROTOCOL
As with any in vitro functional procedure where an induced response is normal, the absence of an induced response is indicative of either an inherent biological abnormality or a technical problem. In order to ascertain whether it is the former or the latter, it is critical to incorporate a procedure control for the in vitro activation process and a known positive normal control. CD69 is a very early activation antigen that is expressed on all lymphocytes following in vitro activation. The up-regulation of CD69 is incorporated into the assay as a positive procedure control to confirm in vitro activation. If the test is being developed as a diagnostic assay, then it is strongly recommended that a normal healthy control sample be obtained and run simultaneously with the positive normal control. Materials 1 mg/ml PMA solution (see recipe) Complete RPMI (APPENDIX 2A) without serum 1 mg/ml calcium ionophore solution (see recipe) Sample: Venous blood drawn into sodiul heparin Vacutainer tube, or Peripheral blood mononuclear cells (PBMC): separated from whole blood by ficoll-hypaque density-gradient centrifugation (UNIT 5.1), washed twice in complete RPMI, and brought to 107 cells/ml Ca2+- and Mg2+-free PBS (APPENDIX 2A) Monoclonal antibodies: Phycoerythrin-labeled anti-CD40 ligand (CD40L-PE), clone TRAP1 (Pharmingen) PE-labeled mouse IgG1 (MsIgG1-PE;Pharmingen) isotype control Fluorescein isothiocyanate–labeled anti-CD8 (CD8-FITC; Pharmingen) FITC-labeled anti-CD3 (CD3-FITC; Pharmingen) PE-labeled anti-CD69 (CD69-PE: Becton Dickinson) PE-cyanine-5-labeled anti-CD3 (CD3-PE-Cy5; Pharmingen) Contributed by Maurice R.G. O’Gorman Current Protocols in Cytometry (2000) 6.7.1-6.7.10 Copyright © 2000 by John Wiley & Sons, Inc.
Phenotypic Analysis
6.7.1 Supplement 12
1× FACS lysing solution: 1/10 aqueous dilution of 10× stock (Becton Dickenson Immunocytometry), stable 1 month at room temperature Flow cytometry wash solution (see recipe) 1% (w/v) paraformaldehyde fixative (see recipe) 12 × 75–mm polystyrene tubes (Falcon, Becton Dickinson Labware) Flow cytometer with a 488-nm line, able to detect simultaneously at least three different emission wavelengths plus forward and right-angle light-scatter signals, and with the ability to store listmode data Flow cytometric analysis software (e.g., Cellquest, Becton Dickinson; Winlist, Verity Software House) CAUTION: Phorbol 12-myristate 13-acetate is toxic and a potential carcinogen. Handle, store, and dispose of appropriately. Stimulate peripheral-blood T cells 1. Thaw 1 mg/ml PMA solution on ice, and add 3 µl to 5 ml complete RPMI without serum to make a 0.6 µg/ml PMA solution. Mix 1 ml of the 0.6 µg/ml solution with 1 ml complete RPMI to make a 0.3 µg/ml solution. Mix 1 ml of the 0.3 µg/ml solution with 1 ml complete RPMI to make a 0.15 µg/ml working PMA solution. Keep on ice. 2. Prepare working concentration of calcium ionophore by combining 9 µl of 1 mg/ml calcium ionophore solution with 3 ml complete RPMI (final concentration 3 µg/ml). Keep on ice. 3. For each sample tested, prepare two 12 × 75–mm polystyrene tubes, labeling one U for unstimulated and the other S for stimulated. 4. In the unstimulated tube, combine 800 µl complete RPMI with either 200 µl well-mixed venous blood or 200 µl PBMCs (2 × 106 cells). 5. To the stimulated tube add: 600 µl complete RPMI 200 µl well-mixed venous blood or 200 µl PBMCs (2 × 106 cells) 100 µl 0.15 µg/ml PMA (final 15 ng/ml) 100 µl 3 µg/ml calcium ionophore (final 300 ng/ml). This protocol was optimized with Sigma’s calcium ionophore (free acid form).
6. Vortex tubes gently, cap loosely, and place in a humidified 5% CO2 incubator at 37°C for 4 hr. This incubation period is a good time to make up the 1× FACS lysing solution and the flow cytometry wash solution, if they have not already been prepared.
7. Vortex gently, add 3 ml Ca2+- and Mg2+-free PBS, and centrifuge 5 min at 300 × g, room temperature. 8. Aspirate supernatant completely. Resuspend pellet in Ca2+- and Mg2+-free PBS to a final volume of 300 µl. Stain with monoclonal antibody 9. Label two sets of three tubes “A,” “B,” and “C”, one set for the stimulated (S) sample and one set for the unstimulated (U) sample. Add the following monoclonal antibodies to the appropriate tubes: CD40 Ligand Expression on Resting and In Vitro–Activated T Cells
SA and UA tubes: CD8-FITC/MsIgG1-PE/CD3-PE-Cy5 SB and UB tubes: CD8-FITC/CD40L-PE/CD3-PE-Cy5 SC and UC tubes: CD3-FITC/CD69-PE.
6.7.2 Supplement 12
Current Protocols in Cytometry
Use saturating amounts of each monoclonal antibody; alternatively, use the amounts recommended by the manufacturer. Clone TRAP1 is highly recommended for the CD40L monoclonal antibody. Other antibodies may be substituted with appropriate testing. The SA tube will provide the stimulated isotype-control cells; the UA tube is the the unstimulated isotype sample. The SB tube will be used for determining the percentage of stimulated CD40 ligand–positive cells, and the UB will provide the percentage of unstimulated cells that are positive for the CD40 ligand. The SC and UC tubes are the positive and negative in vitro stimulation controls, which will be evaluated for the expression of CD69. The cells of interest that express CD40L are CD4+ T cells. Unfortunately, CD4 is lost with activation. The CD3+CD4+ population can, however, be analyzed by gating on cells that express the T cell receptor complex CD3 but that are CD8 negative. Therefore, the analysis actually involves determining the percentage of CD3+CD8− T cells that express CD40L.
10. Add 100 µl stimulated and unstimulated samples to the appropriate tubes, vortex, and incubate at room temperature 20 min in the dark. 11. Add 2 ml of 1× FACS lysing solution to each tube, vortex, and incubate 10 min at room temperature. 12. Centrifuge tubes 5 min at 400 × g, room temperature. Decant and wash twice with flow cytometry wash solution: add 1 ml wash solution, vortex, centrifuge, decant, and repeat. 13. Add 0.5 ml of 1% paraformaldehyde fixative to each tube, vortex, and store at 4°C until analyzed on the flow cytometer. Samples should not be kept longer than 6 hr.
Acquire data 14. Using standard daily protocol, optimize the alignment, compensation, and sensitivity of a flow cytometer for the analysis of FITC, PE, and PE-Cy5 on human leukocytes. 15. Acquire 15,000 events on the flow cytometer using standard three-color whole-blood immunophenotyping settings for FITC, PE, and PE-Cy5. Alternatively, prior to acquisition, draw a gate around all CD3-PE-Cy5-positive events, adjust the flow cytometer to acquire all events, and stop when 2500 gated events have been acquired. Analyze data 16. Set up a dot plot of right-angle light scatter (side scatter, SS; y axis) versus CD3-PECy5 (see Fig. 6.7.1A) and draw a gate around the CD3+ lymphocyte cluster (gate 1). 17. Set up a second dot plot of red fluorescence (CD3-PE-Cy5) on the x axis versus green fluorescence (CD8-FITC) on the y axis, displaying only the events from gate 1. Draw a second gate around the events that are CD3+ and CD8− (gate 2; Fig. 6.7.1B). The majority of the events in gate 2 represent CD4+ T cells.
18. Create a one-parameter histogram (orange fluorescence versus cell count) for all events that satisfy both gate 1 and gate 2. These histograms represent IgG1 (the isotype control) expressed on CD3+CD8− cells in tubes UA and SA, and CD40L expressed on CD3+CD8− cells in tubes UB and SB. See Figure 6.7.1, panels C through F.
Phenotypic Analysis
6.7.3 Current Protocols in Cytometry
Supplement 12
B
gate 2
200
101
400
gate 1
102
CD8-FITC
SS 600
103
800
1000
104
A
R1 100
101
102
103
104
100
0
R2 100
101
CD3-PE-cy5
102
103
104
CD3-PE-cy5
Counts 10 20 30 40 50
D
Counts 10 20 30 40 50
C <1% positive
M1
0
0
M1
<1% positive
100
101 102 103 Isotype control-PE
104
Counts 10 20 30 40 50
4% positive
104
90% positive
M1
0
M1
0
100
101 102 103 Isotype control-PE
F
Counts 10 20 30 40 50
E
100
101
102 CD40L-PE
103
104
100
101
102 CD40L-PE
103
104
Figure 6.7.1 Gating and analysis protocol for the analysis of CD154 (CD40L) on unstimulated and in vitro–activated CD3+CD8− T cells. (A) Dot plot illustrates the gate drawn around the CD3+ T cells (gate 1). The events in this gate are then sent to a dot plot of CD3 versus CD8 (B), and a second gate is drawn around the CD3+ T cells that are CD8− (gate 2). (C) The events that fall into both of these gates (predominantly the CD4+ T cells) are then analyzed for background fluorescence with an isotype-control monoclonal antibody for the unstimulated sample (tube UA in the protocol) so that <2% of the cells are positive. This discriminator setting (M1) is maintained for the analysis of all subsequent tubes, whether unstimulated (C, E) or stimulated (D, F). (D) Stimulated isotype control (tube SA; the percentage of positive cells in this sample should differ <2% from the unstimulated isotype control tube); (E) CD154 expression on the unstimulated cells (tube UB); (F) the percentage of stimulated CD3+CD8− T cells expressing CD154 (tube SB; in normal peripheral blood samples, this percentage should be >80%).
CD40 Ligand Expression on Resting and In Vitro–Activated T Cells
6.7.4 Supplement 12
Current Protocols in Cytometry
Counts 10 20 30 40 50
1000
4% positive
SS 600
0
800
unstimulated
101
400
100
102 CD69-PE
103
104
103
104
101
102 CD3-FITC
103
104
stimulated 99.5% positive
0
100
Counts 10 20 30 40 50
0
200
gate 3
100
101
102 CD69-PE
Figure 6.7.2 Gating and analysis protocol for the evaluation of in vitro–induced expression of CD69 (tubes UC and SC). In the dot plot of CD3 versus side scatter, a gate is drawn around the CD3+ T cells (gate 3, left panel). The events that fall into this gate are then analyzed for their expression of CD69 using the same discriminator set (M1) for the unstimulated isotype control tube. In a normal peripheral blood sample, >90% of the stimulated T cells should express CD69.
19. For tube UA, gated as above, set a positive/negative discriminator such that <2% of the cells in gates 1 and 2 are positive, i.e., >98% of the cells are negative. With the same positive/negative discriminator that was drawn for tube UA (unstimulated isotype sample), note the percentage of positive cells in tube UB (unstimulated CD40L+ cells), tube SA (stimulated isotype-control+ cells), and tube SB (stimulated CD40L+ cells). Report the results as the percentage of CD3+CD8− cells that express CD40L fluorescence at a level that is greater than that of the isotype control. 20. Create a dot plot with side scatter on the y axis versus CD3-FITC (green fluorescence). Gate on CD3+ T lymphocytes (gate 3; Fig. 6.7.2). 21. Create a one-parameter histogram (orange fluorescence versus cell count; Fig. 6.7.2) and measure the percentage of events from gate 3 (CD3+ T cells) that express CD69-PE above the level of the isotype control (set using tube UA) in both unstimulated (UC) and stimulated (SC) samples. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Calcium ionophore solution, 1 mg/ml Add 1 ml dimethyl sulfoxide (DMSO) to 1 mg calcium ionophore (Sigma, free acid form) for a stock concentration of 1 mg/ml. Make 50-µl aliquots and store at −70°C for up to 1 year. To use, thaw quickly and keep on ice. Discard aliquots after thawing. Phenotypic Analysis
6.7.5 Current Protocols in Cytometry
Supplement 12
Flow cytometry wash solution PBS (APPENDIX 2A) 1% (v/v) fetal bovine serum (heat-inactivated, Life Technologies) 0.25% (w/v) sodium azide Store up to 6 months at 4°C Paraformaldehyde fixative, 1% (w/v) Dilute 16% EM-grade formaldehyde (EM Science) 1/16 in 0.9% (w/v) sodium chloride. Filter sterilize through a 0.45-µm filter. Store up to 3 months at 2° to 8°C. Phorbol 12-myristate 13-acetate (PMA) solution, 1 mg/ml Add 1 ml dimethyl sulfoxide (DMSO) to a 5-mg vial of PMA (Sigma) and mix to dissolve. Add an additional 4 ml DMSO for a stock concentration of 1 mg/ml. Store as 100-µl aliquots for up to 6 months at −70°C. To use, thaw quickly and keep on ice. Discard aliquots after thawing. COMMENTARY Background Information
CD40 Ligand Expression on Resting and In Vitro–Activated T Cells
CD154, previously referred to as gp39 and now commonly referred to as the CD40 ligand (CD40L), is a tumor necrosis factor (TNF) family glycoprotein expressed on the surface membrane of activated CD4+ T cells and a small subset of activated CD8+ T cells (Armitage et al., 1992; Hollenbaugh et al., 1992). Binding of CD40L to the CD40 receptor on B cells induces B cell proliferation, promotes immunoglobulin class switching (in the presence of the appropriate cytokines), and prevents apoptosis of germinal-center B cells (Liu et al., 1989; Rousset et al., 1991; Armitage et al., 1993). In addition, CD40L is intimately involved in T cell proliferation and cytokine secretion in response to antigen-presenting cells (Grewal et al., 1996; Yang and Wilson, 1996). The X-linked form of the human hyper-IgM syndrome (XHIM), a rare humoral immunodeficiency in which patients are prone to recurrent bacterial and opportunistic infections, neutropenia, lymphoid hyperplasia, and autoimmune manifestations (Conley, 1992; Notarangelo et al., 1992), is the result of mutations in the gene encoding CD40L (Allen et al., 1993; Arrufo et al., 1993; DiSanto et al., 1993; Fuleihan et al., 1993; Korthäuer et al., 1993). In addition to XHIM, abnormalities in the level of CD40L expressed on activated T helper cells have been observed in common variable immunodeficiency syndrome (Farrington et al., 1994; O’Gorman et al., 1997) and in human immunodeficiency virus (HIV)–infected children (DuChateau et al., 1998).
A rapid clinical flow cytometry procedure was developed to screen for abnormalities in the expression of CD40L directly in whole blood using optimal stimulation conditions and commercially available monoclonal antibodies. Using this assay, it was observed that the percentage of in vitro–activated CD3+CD8− T cells expressing CD40L was severely reduced (i.e., not expressed) in whole blood samples obtained from seven out of seven XHIM patients tested and moderately reduced (i.e., reduced to ∼50% positive) in four out of four mothers of XHIM patients tested, and in one out of three common variable immunodeficiency patients as compared to the expression of CD40L on in vitro–activated T cells obtained from healthy control volunteers (O’Gorman et al., 1997).
Critical Parameters The assay described was originally designed to be performed in a clinical laboratory within ∼6 hr. In this regard, it is important to note that although the percentage of cells that will upregulate their surface expression of CD40L (following in vitro activation with PMA and ionomycin, as described herein) stabilizes after ∼3 hr, the level of expression of CD40L per cell (measured as the mean or median fluorescent channel) continues to rise after a 5-hr incubation (Fig. 6.7.3). Therefore, if comparing the level of expression of CD40L on a per-cell basis between experiments, extreme care should be taken with respect to the length of incubation used for T cell activation. The level of CD40L expressed is critically dependent on a rise in
6.7.6 Supplement 12
Current Protocols in Cytometry
160
70
140
60
120
50
100
40
80
30
60
20
40
10
20
mfc CD40L+
% CD40L+
80
0
0 baseline
1
2
3
4
5
Time (hr) healthy control #1 positive healthy control #1 mfc
healthy control #2 positive healthy control #2 mfc
Figure 6.7.3 Temporal response of CD40L up-regulation on the cell surface of CD3+ T cells following in vitro activation in whole blood with optimal concentrations of PMA and ionomycin in the peripheral blood of two healthy individuals. Two sets of data were collected; one set shows the percentage of T cells expressing CD40L, and the other shows the relative amount of CD40L per cell, collected as the mean fluorescent channel (mfc). Note that although the percentage of T cells expressing the CD40L increases very little after 3 hr, the level of CD40L expressed per cell continues to rise rapidly and is still increasing after a 5-hr incubation.
intracellullar free calcium. Therefore, the assay is more sensitive to the appropriate concentration of the calcium ionophore than to the concentration of PMA (Fig. 6.7.4).
Troubleshooting Problems with the measurement of the upregulation of CD40L on activated T cells can occur at all stages of the procedure. It is very important to include both a positive and a negative control for each experiment. Table 6.7.1 may help in the event that unexpected results occur. As with any flow cytometer–based experiment, one must first test that all aspects of the cytometer are quality controlled and functioning properly before problems with the assay can be resolved.
Anticipated Results Figures 6.7.1 and 6.7.2 illustrate the expected results for each of the six tubes used in this protocol to measure CD40L and CD69 up-regulation on CD3+CD8− (T helper) and
CD3+ T cells, respectively. In normal individuals, >75% of CD3+CD8− T cells should stain positively using the PE-conjugated antiCD40L monoclonal antibody (TRAP1) described in this protocol. The mean percentage of CD40L-positive CD3+CD8− T cells in the in vitro–activated peripheral blood specimens obtained from 20 normal healthy adult volunteers was 88%. In patients with XHIM syndrome (n = 7), the mean percentage of in vitro–activated CD3+CD8− T cells expressing CD40L was only 3%, and in the XHIM carrier mothers (n = 4) of these patients the mean percentage of CD3+CD8− T cells expressing CD40L after stimulation was 53%. CD40L expression on in vitro–activated T cells may also be reduced in patients infected with HIV-1 (DuChateau et al., 1998) and in patients with the common variable immunodeficiency syndrome (Farrington et al., 1994; O’Gorman et al., 1997).
Phenotypic Analysis
6.7.7 Current Protocols in Cytometry
Supplement 12
A 80.00
% of CD3+ expressing CD40L-PE
70.00 60.00 50.00 PMA 0 ng/ml 40.00
PMA 15 ng/ml PMA 30 ng/ml
30.00
PMA 60 ng/ml PMA100 ng/ml
20.00 10.00 0.00 0
25
50
100
200
300
400
[Calcium ionophore] (ng/ml)
B 500.00
mfc (1-10,000) of CD40L-PE on CD3+
450.00 400.00 350.00
PMA 0 ng/ml
300.00
PMA 15 ng/ml PMA 30 ng/ml
250.00
PMA 60 ng/ml 200.00
PMA100 ng/ml
150.00 100.00 50.00 0.00 0
25
50
100
200
300
400
[Calcium ionophore] (ng/ml)
Figure 6.7.4 (A) The effect of increasing concentrations of PMA and calcium ionophore on the percentage of CD3+ T cells expressing CD40L. Without the addition of PMA, very few T cells are induced to express CD40L unless high levels of calcium ionophore (>200 ng/ml) are present. With high concentrations of calcium ionophore, increasing the concentration of PMA above 15 ng/ml has little effect on the percentage of T cells expressing CD40L. (B) The effect of increasing concentrations of PMA and calcium ionophore on the level of CD40L expressed, where the mean fluorescent channel (mfc) provides data on a per-CD3+-T-cell basis. The level of CD40L expressed per cell is very dependent on a critical concentration of calcium ionophore, i.e., >200 ng/ml. PMA is required to induce maximal CD40L expression on the cell surface, although the amount of PMA above 15 ng/ml has little effect; without PMA the level of CD40L expressed per cell is very low.
CD40 Ligand Expression on Resting and In Vitro–Activated T Cells
6.7.8 Supplement 12
Current Protocols in Cytometry
Table 6.7.1
Troubleshooting Guide
Problem
Possible cause
No fluorescence signal generated with anti-CD3 and/or with anti-CD8 CD40L is up-regulated but well below expected in both patient and control CD40L is not up-regulated when expected, and CD69 is increased in the in vitro stimulation control No CD40L or CD69 up-regulation
An antibody was not added to the Repeat the experiment tube The sample was anticoagulated in Repeat with freshly collected EDTA instead of heparin blood using heparin PMA solution has deteriorated or calcium ionophore has not been addeda PMA was bad or not added
Solution
Make up fresh PMA and repeat the experiment Make fresh batch of PMA and repeat experiment
aCD69 up-regulation is not as sensitive as CD40L up-regulation to the condition of PMA.
Time Considerations The stock solutions of PMA and calcium ionophore should be prepared, divided into aliquots, and frozen before the actual experiment is performed. The complete RPMI medium must be made up or available on the day of the experiment. The Ca2+- and Mg2+-free PBS, the 1× FACS lysing solution, and the flow cytometry wash solution can be made up during the 4-hr incubation. It will take ∼20 to 30 min to prepare the working concentrations of PMA and the calcium ionophore, to label the tubes, and to add the blood and stimulants to the tubes prior to the incubation. If PBMCs are used instead of whole blood, it will take ∼1 hr to separate, wash, and count the PBMCs. After the 4-hr incubation, it will take ∼1 hr to wash, label, lyse, and fix the cells. It will take another 15 to 30 min to run the samples and analyze them on the flow cytometer.
Literature Cited Allen, R.C., Armitage, R.J., Conley, M.E., Rosenblatt, J., Jenkins, N.A., Copeland, N.G., Bedell, M.A., Edelhoff, S., Disteche, C.M., Simoneaux, D.K., et al. 1993. CD40 ligand gene defects responsible for X-linked hyper IgM syndrome. Science 259:990-993. Armitage, R.J., Fanslow, W.C., Strockbine, L., Sato, T.A., Clifford, K.N., Macduff, B.M., Anderson, D.M., Gimpel, S.D., Davis-Smith, T., Maliszewski, C.R., et al. 1992. Molecular and biological characterization of a murine ligand for CD40. Nature 357:80-82.
Armitage, R.A., Macduff, B.A., Spriggs, M.K., and Fasnslow, W.C. 1993. Human B cell proliferation and Ig secretion induced by recombinant CD40 ligand are modulated by soluble cytokines. J. Immunol. 150:3671-3680. Arrufo, A., Farrington, M., Hollembaugh, D., Li, X., Milatovich, A., Nonoyama, S., Bajorath, J., Grosmaire, L.S., Stenkamp, R., Neubauer, M., et al. 1993. The CD40 ligand, gp39, is defective in activated T cells from patients with X-linked hyper-IgM syndrome. Cell 72:291-300. Conley, M.E. 1992. Molecular approaches to the analysis of X-linked immunodeficiencies. Annu. Rev. Immunol. 10:215-238. DiSanto, J.P., Bonnefoy, J.Y., Gauchat, J.F., Fischer, A., and De Saint Basile, F. 1993. CD40 ligand mutations in X-linked immunodeficiency with hyper IgM. Nature 361:539-543. DuChateau, B.K., Yogev, R., and O’Gorman, M.R.G. 1998. Evaluation of CD154 expression on the surface of CD4+ lymphocytes obtained from pediatric HIV-1 infected patients. Cytometry Suppl. 9:101. Farrington, M., Grosmaire, L.S., Nonoyama, S., Fischer, S.H., Hollenbaugh, D., Ledbetter, J.A., Noelle, R.J., Aruffo, A., and Ochs, H.D. 1994. CD40 ligand expression is defective in a subset of patients with common variable immunodeficiency. Proc. Natl. Acad. Sci. U.S.A. 91:10991103. Fuleihan, R., Ramesh, N., Loh, R., Jabara, H.F., Rosen, S., Chatila, T., Fu, S.M., Stamenkovic, I., and Geha, R.S. 1993. Defective expression of the CD40 ligand in X chromosome linked immunoglobulin deficiency with normal or elevated IgM. Proc. Natl. Acad. Sci. U.S.A. 90:2170-2173.
Phenotypic Analysis
6.7.9 Current Protocols in Cytometry
Supplement 12
Grewal, I.S., Foellmer, H.G., Grewal, K.D., Xu, J., Hardardottir, F., Baron, J.L., Janeway, C.A., and Flavell, R.A. 1996. Requirement for CD40 ligand in costimulation induction, T cell activation, and experimental allergic encephalomyelitis. Science 272:1864-1867. Hollenbaugh, D., Grosmaire, L.S., Kullas, C.D., Chalupny, N.J., Braesch-Andersen, S., Noelle, R.J., Stamenkovic, I., Ledbetter, J.A., and Aruffo, A. 1992. The human T cell antigen gp39, a member of the TNF gene family, is a ligand for the CD40 receptor: Expression of a soluble form of p39 with B cell co-stimulatory activity. EMBO J. 11:4313-4321. Korthäuer, U., Graf, D., Mages, H.W., Briere, F., Padayachee, M., Malcolm, S., Ugazio, A.G., Notarangelo, L.D., Levinsky, R.J., and Kroczek, R.A. 1993. Defective expression of T-cell CD40 ligand causes X-linked immunodeficiency with hyper-IgM. Nature 361:539-541. Liu, Y.-D., Joshua, D.E., Williams, G.T., Smith, C.A., Gordon, J., and MacLennan, I.C.M. 1989. Mechanisms of antigen-driven selection in germinal centers. Nature 342:929-931.
Notarangelo, L.D., Duse, M., and Ugazio, A.G. 1992. Immunodeficiency with hyper IgM (HIM). Immunodef. Rev. 3:101-122. O’Gorman, M.R.G, Zaas, D., Paniagua, M., Corrochano V., Scholl, P.R., and Pachman L.M. 1997. Development of a rapid whole blood flow cytometry procedure for the diagnosis of X-linked hyper-IgM syndrome patients and carriers. Clin. Immunol. Immunopathol. 85:172-181. Rousset, F., Garcia, E., and Banchereau, J. 1991. Cytokine-induced proliferation and immunoglobulin production of human B lymphocytes triggered through their CD40 antigens. J. Exp. Med. 173:705-710. Yang, Y. and Wilson, J.M. 1996. CD40 ligand-dependent T cell activation: Requirement of B7CD28 sign aling th rough CD40. Science 273:1862-1864.
Contributed by Maurice R.G. O’Gorman Northwestern University Medical School Children’s Memorial Hospital Chicago, Illinois
CD40 Ligand Expression on Resting and In Vitro–Activated T Cells
6.7.10 Supplement 12
Current Protocols in Cytometry
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
UNIT 6.8
At present the enumeration of absolute cell numbers is relevant in different research settings (e.g., cell culture) as well as in the clinical laboratory. Absolute counting of cells or cell subsets by flow cytometry is an established technique in at least three major clinical settings: 1. Enumeration of residual leukocytes as part of the quality control of leukoreduced blood products; 2. Enumeration of CD4+ and CD8+ T cells in HIV disease monitoring; 3. Enumeration of CD34+ hematopoietic stem and progenitor cells. Accordingly, counting absolute numbers of CD4+ T cells, CD34+ hematopoietic precursor cells (HPC), and residual leukocytes in whole blood by flow cytometry represents diverging clinical aspects of the same technical challenge. Most efficient procedures include immunofluorescence cell analysis linked to absolute cell enumeration, as well as to some stringent gating strategy in order to identify relevant and eliminate irrelevant cell populations with high efficiency. However, as the percentages and absolute cell-level ranges in these varied clinical settings are widely different, diverging technologies are necessary to achieve the necessary corresponding sensitivity, precision, and counting accuracy. The importance and technical requirements of CD34+ hematopoietic stem and progenitor cell enumeration are discussed in UNIT 6.4. In the case of blood bank quality control, the enumeration of residual white blood cells (WBCs) in filtered blood products is a significant assay. The composite technology includes detection of cells based on cell morphology, nuclear fluorescence, and flow rate count with fluorospheres. The enumeration procedure is usually accomplished by flow cytometry. An accurate count is critical to prevent febrile reactions, microrganism transfer, and alloimmunization. Enumeration of absolute CD4+ T lymphocyte number continues to be the hallmark laboratory test for staging HIV-infected patients. This is a critical surrogate marker for assessing immunodeficiency. The T cell subset value is an independent marker, yet it complements HIV plasma viral load data. As potent anti-HIV therapies are becoming more effective and complex, the CD4+ T cell levels for diagnostic/prognostic staging of patients and therapeutic/prophylactic intervention will continue to shift (Johnson et al., 1995; Lane, 1994). However, the utility of the CD4+ T cell count remains unchallenged and critical (Nicholson et al., 1994). The absolute and percent CD4+ T cell count is also of clinical relevance in other immunodeficiency conditions. These include solid-organ transplantation, the post-chemotherapy period, the recovery phase following bone marrow (or stem cell) transplants, therapy with purine nucleosides like 2-chlorodeoxyadenosine (cladribine) for hairy-cell leukemia, cytomegalovirus infection in immunocompromised patients, and protein-calorie malnutrition. Initially, in the early 1980s, design limitations of flow cytometry restricted the availability of accurate CD4+ T cell count to the immunophenotypical identification and enumeration of lymphocyte frequencies (UNIT 6.5). This technical deficiency, at the time, appeared to be just a minor compromise. The additional information required to obtain absolute numbers was readily available from the hematology department. More recently, it has become possible to obtain absolute count with just a flow cytometer. It is now no longer necessary to combine data with a hematology workstation to generate absolute counts with flow cytometry. These two clinical methods for obtaining absolute counts are Contributed by Frank Mandy and Bruno Brando Current Protocols in Cytometry (2000) 6.8.1-6.8.26 Copyright © 2000 by John Wiley & Sons, Inc.
Phenotypic Analysis
6.8.1 Supplement 13
referred to as dual-platform (flow combined with hematology) or single-platform (flow cytometry alone) techniques, respectively. The dual-platform method: The flow cytometer provides the percentage of a given cell subset of the total lymphocyte population (i.e., in this example the lymphocyte population is the chosen reference denominator; see UNIT 6.5). The second platform then provides absolute WBC counts with a 3- or 5-part leukocyte differential count, which includes a lymphocyte count. In this example, the lymphocyte count is the reference denominator. The single-platform method: This technology provides the absolute cell number by counting only the cells of interest (i.e., the CD4+ T cells or CD34+ populations) in a precisely determined blood volume. With this method, acquisition of a “reference denominator” is unnecessary. Exact cell identification is accomplished with a logical electronic gating algorithm capable of identifying lineage-specific immunofluorescent markers. With this technology, the exclusion of undesirable (contaminant) cells is automatic. The “logical gating strategy” may vary between different protocols for the single-platform counting methods depending on the selection of lineage-specific markers. The single-platform technology has emerged as the method of choice for absolute cell enumeration in clinical applications. The advantage of the single-platform strategy is the inherent capacity to positively identify all cells of interest exclusively, and thus exclude contaminating cells. The identified cell (subset) number is extracted directly from the original blood volume. The analyzed blood volume required for the single platform by the cytometer is determined by either a volumetric or a microsphere-based method. During the practical application of the single-platform method, the quality of the pipetting step(s) is the precision-determining factor. The most reliable dispensing method for absolute cell counting assays is described in UNIT 6.4. The role of the reverse pipetting technique is critical. A further technical issue that influences the precision of absolute counting is sample preparation (see UNIT 6.2). Only the “lyse-no-wash” technique is compatible with the single-platform absolute count method. The essential original relationship between blood cell count and unit volume must be retained by using known dilution factors (volumetric approach) or by adding a known number of counting beads in a known sample volume (microsphere counting approach). When volumetric counting is employed, it is essential that all dispensing steps be handled with consistent precision throughout (i.e., sample, antibody, lysing, and fixative volume); the appropriate dilution factor must be applied to correct for the final sample volume. Conversely, using microsphere counting, the precision-determining step is limited to the preparation of the sample/microsphere mixture. The results are not influenced by the additional volume of antibody and lysing reagents. Throughout the following protocols the terms microspheres, fluorospheres, fluorescent beads, fluorescent microspheres, microbeads, and counting beads are used interchangeably and always refer to commercially available fluorescing plastic microspheres. The general relationships between various protocols covered in this chapter are outlined in Figure 6.8.1. The fundamental difference between volumetric and flow rate-based cytometry will be explained in some detail. A brief synopsis of the two parallel approaches to single-platform absolute counting is depicted in Figure 6.8.2. BASIC PROTOCOL 1
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
SINGLE-PLATFORM ENUMERATION OF ABSOLUTE NUMBERS OF RESIDUAL WBCs IN LEUKOREDUCED BLOOD PRODUCTS USING FLOW-COUNT FLUOROSPHERES Typically, a leukoreduced 300-ml blood-product bag must contain <5 × 108 WBCs (<1600 WBCs/µl) to prevent febrile reactions, <5 × 106 WBCs (<16 WBCs/µl) to minimize the transfer of microorganisms, and <1 × 106 WBCs (<3 WBCs/µl) to be considered virtually leukocyte free, as can be obtained by double-filtering the product. This implies the detection of WBCs in a cell concentration range much lower than the sensitivity range of
6.8.2 Supplement 13
Current Protocols in Cytometry
Single Platform Technology Protocols (universal approach)
flow rate cytometry
volumetric cytometry
absolute residual WBC
Basic Protocol 1
Alternate Protocol 1
absolute leukocyte
Basic Protocol 2
Alternate Protocol 3
absolute residual WBC
absolute leukocyte
Alternate Protocol 2
Alternate Protocol 5
Alternate Protocol 4
Figure 6.8.1 Relationship between the various protocols covered in this unit.
whole blood
flow-rate cytometry ml
volumetric cytometry
measured
ml
ml
blood
fluorospheres
Side scatter
Side scatter
measured blood
CD8
T gate
CD4
CD45 Gate
CD3
CD3
Phenotypic Analysis Figure 6.8.2 Two approaches to single-platform absolute counting. On the left side is the volumetric approach. On the right side is the flow-rate technology.
6.8.3
Current Protocols in Cytometry
Supplement 13
hematology analyzers and other traditional counting techniques. The analysis of such rare numbers of WBCs is accomplished by detection of the fluorescence signal from propidium iodide incorporated into the nucleus of permeabilized WBCs. RNA staining of platelets and other cells is prevented by ribonuclease (RNase) treatment. The absolute WBC count can be calculated by the microsphere-based technique. Since the final volume of the blood product is known, the total WBC number per bag can be calculated. The results must comply with performance specifications of a given filtering technique. To enumerate the absolute number of residual WBCs by single-platform flow cytometry, an accurately measured blood product sample is pipetted into a test tube, and an accurately measured number of fluorescent counting microspheres are added. A direct lyse-no-wash procedure is employed to avoid cell or microsphere loss. With this protocol, by monitoring the cell and microsphere events and by multiplying the number of beads added per unit of volume, one can calculate the absolute number of residual WBCs. To obtain precise and accurate sampling of the blood product and of the counting beads, reverse pipetting is recommended (UNIT 6.4). Terminology, gating parameters and display features unique to specific instruments (i.e., Becton Dickinson FACS, Beckman-Coulter XL) are used in this protocol. However, the described procedure can be applied universally to other flow cytometers. Materials Sample of interest: leukoreduced blood sample or leukoreduced platelet sample Staining buffer (see recipe) Flow-Count beads (aqueous suspension of microspheres; Beckman-Coulter) 12 × 75–mm polystyrene tubes Flow cytometer with at least two fluorescence detectors and appropriate filter set to detect red flurorescence from propidium iodide (>650 nm) and green fluorescence from beads (530 ± 15 nm). Prepare sample 1. For each sample, label one 12 × 75–mm polystyrene tube. Mix the blood product well and add 100 µl sample to each tube. The definition of a representative sample varies according to national/regional regulations.
2. Add 2 ml staining buffer, vortex gently, and incubate 15 min at room temperature, protected from light. 3. Immediately prior to use, resuspend Flow-Count beads manually to avoid bubble formation. Do not vortex. Add 100 µl Flow-Count beads to each tube, with the same pipet used for the sample dispensing, and mix gently. NOTE: The pipet tip should have been tilled and drained several times in advance.
Set up instrument 4. Position a log green fluorescence versus log red fluorescence dot-plot window and set the trigger and threshold on red fluorescence. Set another dot-plot window on linear side scatter versus log red fluorescence (Fig. 6.8.3). 5. Adjust the gain on the red fluorescence detector to keep the leukocyte cluster around the center of the log scale. Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
6. Adjust the red fluorescence threshold about half a log decade lower than the mean of the leukocyte cluster.
6.8.4 Supplement 13
Current Protocols in Cytometry
100
104 104
103 102 101
Log red fluorescence 101 102 103 Log green fluorescence
B R2
100
104 103 102
R1
101 100
Log red fluorescence
A
0
200
400 600 Side scatter
800
1000
Figure 6.8.3 Assessment of leukoreduction with Flow-Count fluorospheres. (A) Residual leukocytes are included in R1. Flow-Count microspheres appear as a diagonal cluster at the upper right corner; singlets are better resolved. (B) In gate (R2), only singlets are included for counting with the Flow-Count system.
7. Adjust the green fluorescence detector to keep the Flow-Count bead cluster on scale and to resolve the WBC cluster from vertical axis. The instrument is now ready for acquisition.
Create gating regions and acquire data In the following steps, instrument manufacturer’s software-specific terminology for regions and gates is identified inside parentheses for Beckman-Coulter and Becton Dickinson instruments, respectively (see UNIT 6.4), separated by a semicolon (e.g., A;R1). To see how the microspheres will look in the appropriate gates see Figures 6.8.4 and 6.8.5. 8. On the green versus red fluorescence dot plot, display all events and draw a rectangular gate [A;R1) to include leukocyte events. In the side scatter versus red fluorescence window, draw another rectangular region (B;R2) around the singlet bead population. 9. Collect enough events for statistical reliability. A product that is relatively WBC rich (i.e., >100 WBCs/ìl) may give enough cell events in the leukocyte gate (A;R1) to obtain reliable statistics with 1000 beads (∼1 ìl) counted in the bead gate (B;R2). Conversely, when the residual WBC numbers are very low (<20 WBCs/ìl), a much greater number of bead events must be acquired (i.e., 10,000 to 20,000) to ensure reliable statistics for the WBC events.
Calculate absolute number of residual WBCs 10. If an appropriate singlet bead number has been collected (in region B;R2), calculate absolute number of WBCs according to the following formula: Residual WBCs/µl = [number of events in region (A;R1)/number of events in region (B;R2) × number of beads/µl].
Phenotypic Analysis
6.8.5 Current Protocols in Cytometry
Supplement 13
B
800
240
300
1000
A
Flow-Count Counts 120 180
Side scatter 400 600
R2 Flow-Count TruCount R1
M1
0
0
60
200
M2 TruCount
200
0
400 600 Forward scatter
800
1000
100
101 102 103 Log green fluorescence
104
A
504
1024
Figure 6.8.4 Histograms of two commercially available counting fluorospheres as they appear on the FACSCalibur flow cytometer. (A) Dual-light scatter image of the TruCount and Flow-Count mircospheres in R1 and R2 gates, respectively. (B) The peaks depict the relative green fluorescence intensity of the two types of fluorospheres, TruCount and Flow-Count, respectively, as M1 and M2 regions.
B Flow-Count
Count
Side scatter
TruCount
A
TruCount Flow-Count
G
B
0
0
F 0
Forward scatter
1024
.1
Log green fluorescence
1000
Figure 6.8.5 Histograms of two commercially available counting fluorospheres as they appear on the Epics XL flow cytometer. (A) A dual light-scatter image of the TruCount and Flow-Count microspheres in F and A gates, respectively. (B) The peaks depict the relative green fluorescence intensity of the two types of fluorospheres, TruCount and Flow-Count, respectively, as G and B regions.
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
6.8.6 Supplement 13
Current Protocols in Cytometry
SINGLE-PLATFORM ENUMERATION OF ABSOLUTE NUMBERS OF RESIDUAL WBCs IN LEUKOREDUCED BLOOD PRODUCTS USING FLUOROSPHERE-CONTAINING TruCount TUBES
ALTERNATE PROTOCOL 1
This technique differs from Basic Protocol 1 in that the blood sample is added directly to the TruCount tubes containing a lyophilized fluorosphere pellet. Moreover, TruCount and Flow-Count beads differ in their light scatter and fluorescence intensity features, as shown in Figures 6.8.4 and 6.8.5. Additional Materials (also see Basic Protocol 1) TruCount absolute count tubes (Becton Dickinson Immunocytometry) Prepare sample 1. For each sample, label one TruCount tube. Mix the blood product well and add 100 µl to each tube. 2. Add 2 ml staining buffer, vortex gently, and incubate 15 min at room temperature, protected from light. Set up instrument 3. Position a log green fluorescence versus log red fluorescence dot-plot window and set the trigger and threshold on red fluorescence (Fig. 6.8.6). 4. Adjust the gain on the red fluorescence to keep the leukocyte cluster around the center of the log scale. 5. Adjust the red fluorescence threshold about half a log decade lower than the mean of the leukocyte cluster. 6. Adjust the green fluorescence detector to keep the TruCount bead cluster on scale and to resolve the WBC cluster from the vertical axis. The instrument is now ready for acquisition.At the same detector settings used in Basic Protocol 1, the TruCount beads appear as a diagonal cluster with fluorescence intensity approximately one log decade lower than that of Flow-Count beads.
Create gating regions and acquire data In the following steps, instrument manufacturer’s software-specific terminology for regions and gates is identified inside parentheses for Beckman-Coulter and Becton Dickinson instruments respectively (see UNIT 6.4), separated by a semicolon (e.g., A;R1). To see how the microspheres will look in the appropriate gates see Figures 6.8.4 and 6.8.5. 7. On the green versus red fluorescence dot plot, display all events. Draw a rectangular gate (A;R1) to include leukocyte events. In the same display draw another polygonal region (B;R2) to include all TruCount bead events (i.e., singlets plus all aggregates). 8. Collect enough events for statisical reliability. A product that is relatively WBC rich (i.e., >100 WBCs/ìl) may give enough cell events in the leukoyte gate (A;R1) to obtain reliable statistics with 1000 beads (∼2 ìl) counted in the bead gate (B;R2). Conversely, when the residual WBC numbers are very low (<20 WBCs/ìl), a much greater number of bead events must be acquired (i.e., 10,000 to 20,000) to ensure reliable statistics for the WBC events.
9. If an appropriate number of bead events has been collected, calculate the absolute number of residual WBCs according to the following formula: Residual WBCs/µl = (number of events in region A;R1 /number of events in region B;R2) × number of microspheres/µl.
Phenotypic Analysis
6.8.7 Current Protocols in Cytometry
Supplement 13
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
6.8.8
SS
1024
1000
CD3 PECy5
1000
lymphocyte %
absolute count
CD45 FITC
6: AB
.1
CD45 FITC
CD3 PECy5
1000
1000 .1
7: G
CD3 PECy5
CD3 PECy5
4: AB
1000
1000
5: beads
.1
.1
.1
8: G
1000
CD4 RD1
CD45 FITC
CD3 PECy5
.1
.1
1000
1000
Figure 6.8.6 Universal template on an Epics XL instrument with 4-color panel in one tube. The upper panel describes absolute count with bright CD45 gating (Region R1). The lower panel decribes percentage lymphocyte number with T gate (Region R2).
.1
2: AB
.1
3: AB
1024 SS 0
1:
1000 CD4 RD1 .1
0
1024
SS
0
1000 CD4 RD1 .1
1000 CD8 ECD .1
60 CD8 ECD .1
1000 CD4 RD1 .1 60 Count 0
Supplement 13
Current Protocols in Cytometry
SINGLE-PLATFORM ENUMERATION OF ABSOLUTE NUMBERS OF RESIDUAL WBCs IN LEUKOREDUCED BLOOD PRODUCTS USING THE PARTEC-PAS/DAKO GALAXY VOLUMETRIC FLOW CYTOMETER The Partec-PAS/DAKO flow cytometer is equipped with two sensing electrodes for volumetric measurement of 200 µl. As for any volumetric technique, all pipetting steps must be performed with high precision.
ALTERNATE PROTOCOL 2
Additional Materials (also see Basic Protocol 1) Partec-PAS/DAKO Galaxy flow cytometer with at least two fluorescence detectors and appropriate filter set to detect red fluorescence from propidium iodide (>650 nm) and green fluorescence from beads (530 ± 15 nm) Prepare sample 1. For each sample, label one 12 × 75–mm polystyrene tube. Mix the blood product well and add 100 µl to each tube. 2. Add 1 ml staining buffer, vortex gently, and incubate 15 min at room temperature, protected from light. With this procedure, the sample dilution factor is 11.0 (i.e., 100 ìl sample in a final volume of 1100 ìl).
Set up instrument 3. Position a log green fluorescence versus log red fluorescence dot plot window and set the trigger and threshold on red fluorescence (Figs. 6.8.7 and 6.8.8.). 4. Adjust the red fluorescence detector to keep the leukocyte cluster around the center of the log scale. 5. Adjust the red fluorescence threshold about half a log decade lower than the mean of the leukocyte cluster. 6. Adjust the green fluorescence detector to keep the WBCs resolved from the vertical axis. A red fluorescence histogram can be also used to better define the WBC nuclear fluorescence cluster. The instrument is now ready for acquisition.
Create gating regions and acquire data In the following steps, features unique to the Partec-PAS/DAKO Galaxy flow cytometer are described. 7. On the green fluorescence versus red fluorescence dot plot display all events. Draw a rectangular gate (R1) to include the major cluster of leukocyte events. 8. Begin acquisition. Counting starts when the sample level falls below the tip of the upper sensing electrode. Counting ends when the sample level falls below the tip of the lower electrode. At this point, 200 ìl of sample has been aspirated, including about 18.1 ìl of the initial blood sample.
Analyze data Region R1 contain the residual WBC events. The polygonal region statistics in the Partec-PAS/DAKO Galaxy analysis software provides an “area/ml” count, which means the number of WBC events in 1 ml of the aspirated sample. 9. Calculate the final WBC number per µl of the original blood sample according to the following formula, where 11 is the dilution factor (see Figs. 6.8.7 and 6.8.8): Residual WBCs/µl = (R1 area/ml × 11)/1,000 Phenotypic Analysis
6.8.9 Current Protocols in Cytometry
Supplement 13
Figure 6.8.7 Leukoreduction with Partec-PAS/DAKO Galaxy Instrument. Leukocytes in an unfiltered red blood cell unit. In R1 the area/ml count is 111540, or 1227 WBCs/µl.
Figure 6.8.8 Leukoreduction with Partec-PAS/DAKO Galaxy Instrument. Residual leukocytes in the same red blood cell unit shown in Figure 6.8.7 after filtering. In R1 the area/ml count is 30, or 0.3 WBCs/µl.
BASIC PROTOCOL 2
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
UNIVERSAL SINGLE-PLATFORM METHOD FOR ABSOLUTE LEUKOCYTE SUBSET COUNTING Since the majority of clinical instruments sold in the past decade were not equipped with a volumetric delivery system, the most frequently used method for reporting absolute leukocyte subset values is based on a double-platform protocol requiring both a hematology analyzer and a flow cytometer. Single-platform technology for obtaining absolute values for leukoytes, including lymphocytes, has been available for clinical flow cytometers since 1996. Single-platform results are obtained by adding fluorescent microspheres
6.8.10 Supplement 13
Current Protocols in Cytometry
of known concentration to the sample preparation. This approach is generally referred to as the flow-rate method (a graphic illustration of this technology is shown in Fig. 6.8.2). The fluorospheres function as an internal standard for blood volume calculation. By collecting both the number of events in the microsphere cluster and the number of events in the cell cluster region, it is possible to extract the number of cells per unit of blood as an absolute number. For more detailed information on the specific immunophenotypic identification and characterization of lymphocytes and other leukocyte subsets, see other units in Chapter 6. This method is suitable for clinical flow cytometers with capacity for three- or four-color analysis and works with any commercial fluorosphere counting system. Materials TruCount absolute count tubes (lyophylized pellet of beads in a test tube; Becton Dickinson Immunocytometry) or Flow-Count beads (aqueous suspension of microspheres; Beckman-Coulter) Sample of interest: whole blood anticoagulated with EDTA Commercial “lyse-no-wash” erythrocyte lysing solution (e.g., Beckman-Coulter or Becton Dickinson) 1% to 2% (v/v) formaldehyde or paraformaldehyde in PBS (see APPENDIX 2A for PBS) Premixed monoclonal antibody combinations (available from a number of suppliers; also see UNIT 6.2) 4 color panel: CD45/CD3/CD4/CD8 3 color panel: CD45/CD3/CD4 and CD45/CD3/CD8 12 × 75–mm polystyrene tubes Flow cytometer Additional reagents and equipment for immunophenotyping (UNIT 6.2) and setting up the universal template (see Support Protocol) Prepare sample For TruCount absolute count tubes 1a. Select immunophenotyping panel (UNIT 6.2). Label an adequate number of TruCount tubes for each specimen. 2a. Place 50 or 100 µl of anticoagulated whole blood (volume depending on which commercial lyse-no-wash solution is used) in each TruCount tube. Lyse erythrocytes using lyse-no-wash lysis solution according to the manufacturer’s instructions. Stain with premixed 3- or 4-color panel of monoclonal antibodies. Process specimen directly in the TruCount test tube. Several manufacturers provide premixed MAb cocktails. Use premixed three- or four-color monoclonal antibodies at concentrations recommended by the manufacturer. If singlecolor reagents are combined, each reagent must be titered (UNIT 4.1) and the cocktail volume must be adjusted in order to achieve optimal concentration.
3a. After staining and lysing, fix all samples with 1% to 2% buffered paraformaldehyde or formaldehyde. Let sample sit for 30 min. As a universal precaution, this should be done even if the commercial lysing/fixing reagents contain a fixative. With single-platform technology there is no washing step. Phenotypic Analysis
6.8.11 Current Protocols in Cytometry
Supplement 13
4a. Resuspend the cell-bead preparation before acquisition. For Flow-Count fluorospheres 1b. Select immunophenotyping panel (UNIT 75–mm test tubes for each specimen.
6.2).
Label an adequate number of 12 ×
2b. Place 50 or 100 µl of anticoagulated whole blood (volume depending on which commercial lyse-no-wash solution is used) in each tube. Lyse erythrocytes using lyse-no-wash lysis solution according to the manufacturer’s instructions. Stain with premixed 3- or 4-color panel of monoclonal antibodies. There are only a limited number of sources for 3- and 4-color cocktails for single-platform absolute counting protocols. The total volume of the reagent cocktail is critical. Several manufacturers provide premixed MAb cocktails. Use premixed three- or four-color monoclonal antibodies at concentrations recommended by the manufacturer. If singlecolor reagents are combined, each reagent must be titrated (UNIT 4.1) and the cocktail volume must be adjusted in order to achieve optimal concentration.
3b. After staining and lysing, fix all sample tubes with 1% to 2% buffered paraformaldehyde or formaldehyde. Let sample sit for 30 min. As a universal precaution, this should be done even if the commercial lysing/fixing reagents contain a fixative. With single-platform technology there is no washing step.
4b. Gently mix Flow-Count fluorospheres for 10 to 12 seconds (avoid excessive mixing to minimize air bubble formation). Add 100 µl Flow-Count fluorospheres to the stained preparation. Resuspend the cell-bead preparation before acquisition. Analyze samples within two hours of adding Flow-Count fluorospheres. The pipetting of the blood and of the fluorospheres is critical. Reverse pipetting technique is the key to reliable absolute numbers (UNIT 6.4). Verify the accuracy of the pipet, as well as the precision of the pipetting technique. For best results, use the same pipet to dispense blood and microspheres and refill the tip several times before pipetting. Immediately after processing the specimens, store all stained samples at 4°C in the dark until ready for data acquisition.
Set up instrument IMPORTANT NOTE: The principles behind the applications of the TruCount and FlowCount fluorospheres are the same. While both brands of microbeads can be used on any conventional flow cytometer, there are some significant differences in the actual recommended protocols of these two products. For further explanations see specific protocols and see Critical Parameters and Troubleshooting. 5. Follow manufacturer’s instrument setup procedure. Run an unstained specimen and adjust PMT voltages to position. 6. Prior to acquisition, set up proper autofluorescence compensation. Set up universal template (see Support Protocol), setting threshold or discriminator on the CD45 signal detector (UNIT 1.14). Set linear side scatter adjustment in order to visualize clearly all of the leukocyte clusters.
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
Acquire data 7. To obtain a reliable absolute count value, collect a minimum of 1000 events in the fluorosphere region. At the same time, collect a minimum of 2500 to 3000 bright CD45+ cells or 2000 to 2500 T cells.
6.8.12 Supplement 13
Current Protocols in Cytometry
8. Calculate the absolute number of cells from the following formula: no. of events in cell subset region total no. of beads per test × = no. of cells/µl no. of events in bead region volume of blood For example, for a Flow-Count bead concentration of 1040 beads per µl, and extracting all other values from the FACSCalibur template (Fig. 6.8.9), the absolute number of cells is: 1510 100 µl × 1040 beads/µl × = 1435 CD4 cells/µl 1094 100 µl Instrument vendors have software packages that will calculate the absolute number directly from analysis. Both CD3+CD4+ and CD3+CD8+ lymphocyte percentages can be extracted directly from the quadrant (upper right; H2 and I2) of dot plot no. 7 and no. 8, respectively (see Fig. 6.8.6 and Fig. 6.8.9). SETTING UP THE UNIVERSAL TEMPLATE Figure 6.8.6 illustrates the universal template configurations for the FACSCalibur (Becton Dickinson) and Figure 6.8.9 illustrates the universal template configuration for the Epics-XL (Beckman-Coulter). The templates are described below in detail. Numbers referring to dot plots and histograms correspond to those in Figure 6.8.6 and Figure 6.8.9. The instrument manufacturer’s specific terminology related to the software for regions and gates is identified inside parentheses for the two different makes of instruments, separated by a semicolon. For example under dot plot no. 1 “(R1;A)” refers to the corresponding terms used by Becton Dickinson and Beckman-Coulter, respectively.
SUPPORT PROTOCOL
Dot plot no. 1: CD45 versus side scatter, ungated 1. To include 100% of the lymphocytes for analysis: a. Set threshold or discriminator on CD45 detector; as close to the leukocytes as possible; b. Adjust side scatter to visualize all leukocytes; c. Create a region (R1;A) around the cells which have bright CD45-positive expression and low side scatter properties. This region may include some monocyte contaminants. Dot plot no. 2: CD3 versus side scatter, gated on (R1;A) 2. To establish an exclusive gate around T lymphocytes for absolute count determination, set a region (R2;B) around the CD3+ cluster. Dot plot no. 3: CD3 versus CD4, gated on (R1*R2;AB); dot plot no. 4: CD3 versus CD8, gated on (R1*R2;AB) 3. To collect T cell subset events for absolute count calculation: a. Set quadrants; b. Collect CD3+CD4+ and CD3+CD8+ events from the quadrant (upper right; C2 and D2) on dot plot no. 3 and no. 4, respectively. Both dot plots result from logical CD45 and CD3 gating and display only T cells.
Phenotypic Analysis
6.8.13 Current Protocols in Cytometry
Supplement 13
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
6.8.14
Supplement 13
Current Protocols in Cytometry
Figure 6.8.9 Universal Template on a FACSCalibur. Analysis with a 4-color panel in one tube. The upper panel describes the absolute count with bright CD45 gating (Region A). The lower panel illustrates the percentage lymphocyte number with T-gate (Region B).
Dot plot no. 5: CD3 versus CD4, ungated; histogram no. 5: CD3, ungated 4. To define a fluorosphere region to permit the calculation of absolute cell numbers, create a fluorosphere region (R4;E) on dot plot no. 5 or set cursor (M1;E) on histogram no. 5 The fluorescence signal emitted by the counting microspheres will appear in several fluorescence detectors (PMTs). Flow-Count and TruCount microspheres emit different levels of fluoresence intensity. The operator must select the most appropriate color PMT or combination of PMTs that provides the best fluorescence resolution between microspheres and cells of interest.
Dot plot no. 6: CD45 versus side scatter, gated on (R1*R2;AB) 5. To assist the operator with the analysis of difficult specimens which may present poor resolution between lymphocytes and monocytes, create a region (R3;G) around the CD3 clusters. (R3;G) should include at least 97% of the T cells. The single cluster that appears on dot plot no. 6 is used to draw the contour of the gate. Dot plot no. 6 represents only T cells, as it is gated on CD45 and CD3. Care must be exercised when drawing the gate region (R3; G) around the cluster.
Dot plot no. 7: CD3 versus CD4, gated on (R3;G) 6. To report CD3+CD4+ lymphocyte percentages: a. Set quadrants; b. Collect from quadrant (upper right; H2) the CD3+CD4+ lymphocyte percentages. Monocyte contamination can be monitored from the quadrant (upper left; H1).
Contamination should not exceed 3%. If it exceeds 3%, the region (R3;G) of dot plot no.6 must be redrawn. Dot plot no. 8: CD3 versus CD8, gated on (R3;G) 7. To report CD3+CD8+ lymphocyte percentages: a. Set quadrants b. Collect from quadrant (upper right;I2) the CD3+CD8+ lymphocyte percentages SINGLE-PLATFORM METHOD FOR ABSOLUTE LEUKOCYTE SUBSET COUNTING USING FLOW-COUNT FLUOROSPHERES WITH EPICS XL INSTRUMENTS
ALTERNATE PROTOCOL 3
Materials Premixed monoclonal antibody: Cyto-Stat reagents (Beckman-Coulter); 2-color, 3-color, or 4-color: CD3-FITC/T4-RD1 and CD3-FITC/T8-RD1 CD45-FITC/CD4-RD1/CD3-PC5 and CD45-FITC/CD8-RD1/CD3-PC5 CD45-FITC/CD4-RD1/CD8-ECD /CD3-PC5 Sample of interest: whole blood anticoagulated with EDTA ImmunoPrep Reagent System for whole blood lysis (Beckman-Coulter) ImmunoPrep solution A: formic acid ImmunoPrep solution B: sodium carbonate/sodium chloride/sodium sulfate ImmunoPrep solution C: paraformaldehyde Flow-Count beads (aqueous suspension of microspheres; Beckman-Coulter) 1% to 2% (v/v) formaldehyde or paraformaldehyde in PBS (see APPENDIX 2A for PBS) continued Phenotypic Analysis
6.8.15 Current Protocols in Cytometry
Supplement 13
12 × 75–mm polystyrene test tubes Q-Prep, Multi-Q-Prep, or TQ-Prep immunocytometry workstation (Beckman-Coulter) Flow cytometer with at least 3 (or 4) fluorescence detectors. 1. Label an adequate number of 12 × 75–mm test tubes for each specimen. 2. Add 10 µl of appropriate monoclonal antibody combination to each tube. 3. Pipet 100 µl blood into each tube. The pipetting of both the blood and the microspheres is critical. Reverse pipetting technique is the key to reliable absolute numbers. Verify the accuracy of the pipets as well as the precision of the pipetting technique. For best results, use the same pipet to dispense blood and microspheres and refill the tip several times before pipetting.
4. Incubate 10 min at room temperature. 5. Proceed with lysis using Q-Prep, MultiQ-Prep, or TQ-Prep workstation. 6. After staining and lysing, fix all sample tubes with 1% to 2% paraformaldehyde in PBS. As part of a universal precaution, this should be done even if the commercial lysing/fixing reagents contain a fixative.
7. Gently mix Flow-Count microspheres 10 to 12 sec. Avoid excessive mixing to minimize air bubble formation. Add 100 µl of the microsphere suspension to each of the tubes of stained cells. Immediately after processing the specimens, store all stained samples in the dark at 4°C. Resuspend the cell-bead preparation before acquisition. Analyze within two hours of adding Flow-Count fluorospheres. 8. Set up instrument following manufacturer’s instructions and run protocol compatible with color combination used. Acquire data. Refer to the tetraONE System (Beckman Coulter) guide for System II Software version 3.0. ALTERNATE PROTOCOL 4
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
SINGLE-PLATFORM METHOD FOR ABSOLUTE LEUKOCYTE SUBSET COUNTING WITH TruCount TUBES USING FACScan/FACScalibur INSTRUMENTS The need for accurate and precise monitoring of T lymphocyte subsets has challeged manufacturers to develop simpler and more efficient methods for measuring absolute cell numbers. This alternate protocol is Becton Dickinson’s solution for performing singleplatform immunophenotyping; it incorporates recent innovations in three areas; instrumentation, reagents, and software. The internal bead standard is included in each TruCount tube as a pellet, hence one possible pipetting error is eliminated. Materials TruCount absolute count tubes (lyophilized pellet of beads in a test tube; Becton Dickinson Biosciences) TriTest premixed monoclonal antibody combinations (Becton Dickinson): CD45-PerCP/CD3-FITC/CD4-PE CD45-PerCP/CD3-FITC/CD8-PE Sample of interest: whole blood anticoagulated with EDTA 1× FACS lysing solution (Becton Dickinson) 1% to 2% (v/v) formaldehyde or paraformaldehyde in PBS (see APPENDIX 2A for PBS) FACScan or FACScalibur flow cytometer (Becton Dickinson)
6.8.16 Supplement 13
Current Protocols in Cytometry
1. Label one TruCount tube for each appropriate TriTEST reagent. 2. Add 20 µl TriTEST reagent directly to the TruCount tube. 3. Add 50 µl whole blood to each test tube. The pipetting of the blood is critical. Reverse pipetting technique is the key to reliable absolute numbers (UNIT 6.4). Verify the accuracy of the pipets as well as the precision of the pipetting technique. For best results, use the same pipet to dispense blood and microspheres and refill the tip several times before pipetting.
4. Vortex and incubate 5 min at room temperature. 5. Add 450 µl 1× FACS lysing solution. 6. Incubate 15 min at room temperature. 7. After staining and lysing, fix all samples with 1% to 2% paraformaldehyde in PBS. As part of a universal precaution, this should be done even if the commercial lysing/fixing reagents contain a fixative.
8. Immediately after processing the specimens, store all stained samples in the dark at 4°C. 9. Set up FACScan or FACScalibur instrument following manufacturer’s instrument setup procedure. (Set up protocol outlined in the TriTEST three-color reagent application note.) 10. Acquire data. SINGLE-PLATFORM METHOD FOR ABSOLUTE LEUKOCYTE SUBSET COUNTING USING A PARTEC-PAS/DAKO GALAXY VOLUMETRIC FLOW CYTOMETER
ALTERNATE PROTOCOL 5
This protocol was developed for the Partec-PAS/DAKO Galaxy volumetric flow cytometer. As with any volumetric technique, this protocol requires high-precision dispensing for all pipetting steps. Materials Sample of interest: whole blood anticoagulated with EDTA 3-color antibody mixture including CD4-FITC/CD8-PE/CD3-PE-Cy5 (or PerCP) or 4-color antibody mixture including CD4-FITC/CD8-PE/CD3-PE-Cy5 (or -PerCP)/CD19-APC Staining buffer (see recipe) Ammonium chloride lysing buffer, pH 7.2 (APPENDIX 2A) 12 × 75–mm polystyrene test tubes Partec-PAS/DAKO Galaxy flow cytometer with at least two fluorescence detectors and appropriate filter sets Prepare sample 1. For each sample, label one 12 × 75–mm polystyrene tube 2. Mix each blood sample well and add 50 µl to corresponding tubes. 3a. For 3-color immunofluorescence: Add 10 µl antibody mixture containing CD4FITC/CD8-PE/CD3-PE-Cy5 or -PerCP. 3b. For 4-color immunofluorescence: Add 10 µl antibody mixture containing CD4FITC/CD8-PE/CD3-PE-Cy5 or -PerCP/CD19-APC.
Phenotypic Analysis
6.8.17 Current Protocols in Cytometry
Supplement 13
4. Incubate 20 min at room temperature, protected from direct light. 5. Add 1 ml staining buffer. Vortex gently and incubate 15 min at room temperature, protected from light. With the 3-color procedure, the sample dilution factor is 21.2 (i.e., 50-ìl sample in a final volume of 1,060 ìl). With the 4-color procedure, the sample dilution factore is 21.4 (i.e., 50-ìl sample in a final volume of 1070 ìl).
Set up instrument 6a. For 3-color immunofluorescence: Position a set of six dot plot diagrams (Fig. 6.8.10): Upper row: side scatter (SS) versus CD4; SS versus CD8; SS versus CD3; Lower row: CD3 versus CD4; CD3 versus CD8; CD4 versus CD8. 6b. For 4-color immunofluorescence: Position a set of eight dot plot diagrams (Fig. 6.8.11): Upper row: side scatter (SS) versus CD4; SS versus CD8; SS versus CD3; SS versus CD19; Lower row: CD3 versus CD4; CD3 versus CD8; CD4 versus CD8; CD3 versus CD19. 7. Adjust fluorescence PMT voltages. Compensation is best kept off during acquisition and activated during analysis.
8. Set threshold and trigger on forward scatter. The instrument is now ready for acquisition.
Acquire data, create gating regions, and analyze data 9. Start acquisition. When the sample level falls below the tip of the upper electrode, the count begins. The counting ends when the sample level falls below the tip of the lower electrode. At this point 200 ìl of sample have been aspirated.
10a. For 3-color immunofluorescence: On the SS versus CD3 dot plot, draw an amorphous polygonal gate (R1) to include all CD3-positive T cell events. 10b. For 4-color immunofluorescence: On the SS versus CD3 dot plot, draw an amorphous polygonal gate (R1) to include all CD3-positive T cell events. On the SS versus CD19 dot plot draw another amorphous polygonal gate (R2) to include all CD19-positive B cell events. 11. Appropriately compensate overlapping fluorescence using the software option “Compensate Crosstalk.” For 3-color immunofluorescence: 12a. Use R1 to gate the events in the three lower row diagrams (Fig. 6.8.10). 13a. Set three quadrant regions to identify CD3+CD4+ cells (Q1 to Q4), CD3+CD8+ cells (QA1 to QA4) and CD4+CD8+ cells (QB1 to QB4), respectively (Fig. 6.8.10).
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
14a. Correct the cell/ml count number appearing in each quadrant by the dilution factor (i.e., CD3+CD4+ cells in Q2 = 13340/ml × 21.2 = 282.8 CD3+CD4+ cells/µl). Calculate the absolute CD3+ cell count by summing CD3+CD4+ cells (Q2) and CD3+CD4− cells (Q4), or CD3+CD8+ cells (QA2) and CD3+CD8− cells (QA4). The absolute levels of CD3+CD4+CD8+ cells (QB2) and CD3+CD4−CD8− cells (QB3) can be also calculated.
6.8.18 Supplement 13
Current Protocols in Cytometry
6.8.19
0
50 100 150 Side scatter
200
250
1
10 CD3
0
50
100 150 Side scatter
200
250
1000
0
1
QA3: 0.00% 0/ml
Gate: R1 Count: 8281 QA1: 0.00% 0/ml
10 CD3
100
1000
QA4: 38.30% 15860/ml
QA2: 61.70% 26645/ml
0.1
1
10
100
1000
0.1
1
10
0
0
50
1
QB3: 7.39% 3060/ml
10 CD4
100 150 Side scatter Gate: R1 Count: 8281 QB1: 60.71% 25135/ml
R1
250
100
1000
QB4: 30.97% 12825/ml
QB2: 0.93% 385/ml
200
Figure 6.8.10 PARTEC/DAKO Galaxy 3-color volumetric configuration. The 3-color template with CD4-FITC/CD8-PE/CD3-PE-Cy5. Upper row: Ungated side scatter versus fluorescence diagrams. Lower row: R1 (CD3+ cells) acts as gate for the three fluorescence displays. The numbers per ml in each quadrant are multiplied by 21.2 (dilution factor) and divided by 1000 to obtain absolute counts per µl.
100
0.1
0
10
100
1000
0.1
Q1: 67.77% 18960/ml
Q2: 32.22% 13340/ml
1
Q3: 0.01% 5/ml
Gate: R1 Count: 8281 Q1: 0.00% 0/ml
0.1
1
1
10
100
1000
0.1
1
10
100
100
100
10
1000
1000
1000
CD4
CD4
CD8 CD8
CD3 CD8
Current Protocols in Cytometry
Supplement 13
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
6.8.20
Supplement 13
Current Protocols in Cytometry
Figure 6.8.11 PARTEC/DAKO Galaxy 4-color volumetric configuration. The 4-color template with CD4-FITC/CD8-PE/CD3-PE-Cy5/CD19-APC. Upper row: Ungated side scatter versus fluorescence diagrams. R1 (CD3+ cells) and R2 (CD19+ cells) are set. Lower row: The first three diagrams are gated on R1 and define CD4 and CD8 T cell subsets. In the lower right histogram a logical gate (G1 = R1 or R2) is used to display and count B (QC1) and T (QC4) cells. Original Partec display is graphically reprocessessed for clarity.
For 4-color immunofluorescence: 12b. Use R1 to gate the events in the first three diagrams of the lower row of Figure 6.8.11 (i.e., CD3 versus CD4; CD3 versus CD8, and CD4 versus CD8). Use the command “define G1 = R1 or R2" to gate the events in the fourth lower-row diagram (CD3 versus CD19). 13b. Set 3 quadrant regions to identify CD3+CD4+ cells (Q1 to Q4), CD3+CD8+ cells (QA1 to QA4), and CD4+CD8+ cells (QB1 to QB4), respectively. Set another quadrant to identify CD3+ and CD19+ cells (QC1 to QC4) (Fig. 6.8.11). 14b. Correct the cell/ml count number appearing in each quadrant by the dilution factor (i.e., CD3+CD4+ cells in Q2 = 24,206/ml × 21.4 = 518 CD3+CD4+ cells/µl). Calculate the absolute CD3+ cell count by summing CD3+CD4+ cells (Q2) and CD3+CD4− cells (Q4), or CD3+CD8+ cells (QA2) and CD3+CD8− cells (QA4). The absolute levels of CD3+CD4+CD8+ cells (QB2) and CD3+CD4−CD8− cells (QB3) can be also calculated. The absolute levels of CD19+ B cells can be calculated from the QC1 region. Ensure that the CD3+ T cell number in QC4 matches that in Q2 + Q4 and QA2 + QA4.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Sodium citrate buffer 1g/liter sodium citrate 50 mg/liter propidium iodide 200 µl/liter Nonidet P-40 (NP-40) 6 mg/liter ribonuclease A Staining buffer Sodium citrate buffer (see recipe) containing: 50 mg/liter propidium iodide (Sigma) 200 µl/liter Nonidet P-40 (NP-40; Sigma) 6 mg/liter ribonuclease A (RNase A; Sigma) Adjust pH to 7.6 Store at 4°C protected from light; stable up to 6 months COMMENTARY Background Information Instrument development Although the clinical application of single platform technology is recent, the history of this technology is quite extensive. Almost 20 years ago, commercial flow cytometers were first designed for a lyse-no-wash whole blood technique (Mercolino et al., 1995; Connely et al., 1995; Schlenke et al., 1998). The ORTHO CytoronAbsolute was in production between 1989 and 1996. Just recently, another volumetric flow cytometer known as the Galaxy was introduced to the clinical market. Manufactured by Partec-PAS and distributed by DAKO A/S, the
Galaxy instrument utilizes sample preparation technology similar to that developed for the CytoronAbsolute. The Galaxy counting procedure includes two sensing electrodes at different positions which monitor sample movement and hence sample volume. The sample volume contained between the two electrodes is 200 µl. As with any volumetric technique, high-precision dispensing must be used for all pipetting steps. A final sample dilution factor must be used to correct for the absolute cell count. Two other instruments are dedicated to deliver absolute CD4/CD8 and CD3 counts—the FACSCount and IMAGN 2000 instruments, both sold by Becton Dickinson. Only the FACPhenotypic Analysis
6.8.21 Current Protocols in Cytometry
Supplement 13
SCount is a true flow cytometer, and neither one provides lymphocyte percentage values. This unit provides protocols that work with the most commonly used flow cytometers.
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
Fluorospheres In 1982, C. Stewart described a simple procedure for cell concentration measurement with fluorescent latex particles. Another report was published by G. Valet (Valet, 1984). He too utilized fluorospheres as reference counting media for volume calculations of blood cells. More recently, the concept has been reintegrated into clinical flow cytometric applications, and hence has undergone stability and consistency assessment. The application principle is simple. Known amounts of fluorescent microbeads are added to a known volume of stained blood in a lyse-no-wash technique and the beads are counted along with cells. The absolute number of cells is obtained by relating the number of cell events concurrent to the number of microsphere events. By simultaneously counting both fluorospheres and cells of interest and multiplying the cell events by the number of fluorospheres per unit of volume, a ratiometric value is obtained. Remarkably, this method is independent of lysing and antibody volumes, so that high-precision dispensing is limited only to the blood and microsphere pipetting step. The two essential requirements for the microsphere counting technology are (1) that relevant cells be defined by their immunofluorescence, and (2) that a threshold or trigger can be set on the same fluorescence channel that includes all the positively stained cells and the fluorosphere cluster. Since the customary working microsphere concentration is around 1000 particles/µl in the starting blood, the acquisition of 1000 fluorosphere events means that about 1 µl of original blood sample has been analyzed. Moreover, in a representative immunophenotyping analysis, at least 1000 fluorosphere events must be acquired to ensure statistical robustness. Commercial fluorospheres for this application vary in size from 4 to 10 µm in diameter, display a high side scatter, and emit a strong signal in several fluorescence channels. Therefore, an instrument must be set up with a trigger and threshold on a fluorescence channel. The inclusion of all relevant cell events must be ensured. The basic assumption of bead-based counting techniques is the consistency between “fluorosphere events.” Unfortunately, particles in suspension tend to aggregate. This means that if microspheres are not well suspended, a “fluorosphere
event” may include more than one particle, thus altering the basic particle distribution. TruCount and Flow-Count procedures require different dispensing procedures. Fluorospheres other than Flow-Count or TruCount might certainly be used; however, they should be tested and calibrated in advance whenever they have not been maufactured for this purpose. Leukoreduced blood product absolute counting All currently available filtering methods used to deplete blood products leave a low but detectable number of residual WBCs. The counting of residual WBCs in leukoreduced red blood cell and platelet bags is not performed on every single filtered product (Rebulla and Dzik, 1995). The spot check for residual WBCs is performed as part of a random lot-sampling program to monitor the consistency of the filtering devices. The evaluation of virtually leukocyte-free blood products represents a challenge for flow cytometer operators. This application introduces a notable exception to the basic concepts of statistical robustness associated with cytometry. In conventional rare-event analysis it is generally accepted that at least 100 positive events must be acquired to obtain adequate data representation for robustness. Following those assumptions, the analysis of a virtually leukocyte-free product may require the analysis of a large blood volume over an unacceptably long acquisition time. This is an example where the concept of mathematical limits cannot be applied to practical flow cytometry. The monitoring of the absence of a rare population may include a different algorithm compared with the monitoring of the presence of a rare-cell subset. When microsphere-based methods are used, the number of beads acquired is a good indicator of the approximate blood volume analyzed. An appropriate acquisition gate counter can be set in most instruments to evaluate this parameter. In the protocols described, the acquisition of 20 µl blood (i.e., 20,000 and 10,000 fluorospheres with FlowCount and TruCount, respectively) seems an adequate upper limit to stop the analysis, in the case of virtually leukocyte-free blood products. In the case of volumetric analysis, the protocol developed and described here for the PartecPAS/DAKO Galaxy ensures the acquisition of some 18 µl blood in a 200-µl sample, and it is obtained by increasing the concentration of the blood sample. Care is required in pipetting and dispensing viscous samples such as packed red cells and platelet concentrates. For future de-
6.8.22 Supplement 13
Current Protocols in Cytometry
velopment, better staining and gating strategies are required to discriminate true cell nuclei events from debris. Enumeration of leukocyte T cell subsets and other cell subsets Ever since the introduction of clinical flow cytometers in the early 1980s, the most frequently used instruments were designed to analyze the relative ratios of positive cells in a selected cell population. The assumption all along was that automated hematology instruments would provide the required complementary information to generate absolute numbers. However, most guidelines for immunophenotyping of HIV infection recommend reporting absolute numbers for CD4+ and CD8+ T cells (see UNIT 6.5). The dual-platform protocol for absolute counting prevailed, and still remains the predicate method. The dual-platform results obtained for CD4+ and CD8+ lymphocytes are obtained by taking the percentage results from a flow cytometer and then multiplying them by the combined white blood cell (WBC) count and the WBC differential results to calculate absolute numbers for the T cell subsets. As flow cytometry–based quality assurance programs were instituted, they focused on percentage of lymphocyte subsets. The complementary data, WBC and differential results, were obtained from automated hematology instruments. Hematology instrument data are acceptable for flow cytometry–based calculations only if the blood samples were processed fresh. Hematology results from samples more than 24 hr old are unreliable. When the absolute count data are generated entirely from a flow cytometer, it is described as a single-platform technology (SPT). With current flow cytometry-based SPT, it is quite simple to obtain an absolute count. Nicholson et al. (1997) reported the precision for the SPT. However, the true accuracy of SPT is difficult to measure, because there is no gold standard for leukocyte subset counts. Hence, a paradox exists for method evaluation of immunophenotyping protocols. Although Basic Protocol 2 is devoted specifically to the enumeration of lymphocyte subsets, the principles are generally applicable to the derivation of absolute counts of other leukocytes and even nonhematopoietic cells, once they are in the form of single-cell fluid suspensions. Nonetheless, the analysis of different types of cells and/or non-blood biological flu-
ids (e.g., spinal fluid) may necessitate specific changes to the protocol. Validation of new clinical methods It is very difficult to introduce and then validate a new absolute counting technology that offers substantially better performance characteristics. Tradition requires that any new technology must be compared with the conventional or predicate method. The assumption is that a good new method will compare well to the predicate method. Yet it is also clear that a new method by definition can not be significantly different from (better than) the predicate method. Hence validation of a new single-platform (SPT) method for cell count determination provides a considerable challenge. If the new SPT is actually better (more accurate and precise) than the currently accepted technology, the challenge is to demonstrate the superiority of the new method. Bland and Altman (1986) outlined the reasons why statistical methods traditionally utilized to validate such clinical comparisons have been inappropriate. If the intent of a validation is to examine the degree of agreement between two methods, then r2 determination is an inappropriate method. After all, the test of significance may show that the two methods are related; however, one could have a considerable bias over the other. O’Gorman and Gelman (1997) suggested that pairwise statistical analysis and use of bias plots are more appropriate in showing the differences between SPT and multiplatform technology. SPT is still new but it is emerging as the more reliable and hence suitable method, especially for multi-center clinical studies.
Critical Parameters and Troubleshooting Sample quality Generally, fresh samples are easier to analyze. For single-platform enumeration of absolute numbers of residual WBCs, various regulations will dictate the maximum age of the sample. These time limitations are primarily out of concern for the protection of the recipient patient’s health. From a technical perspective, in the case of the single-platform method for absolute leukocyte subset counting, samples 48 hr old or less will perform with all protocols discussed. The use of cryopreserved/thawed samples is not recommended. Phenotypic Analysis
6.8.23 Current Protocols in Cytometry
Supplement 13
Choice of MAbs and fluorochromes The preferred option is to purchase cocktails that are specifically adjusted to perform as a three- or four-color combination. The alternative is to titrate and test each MAb that will be used in a homemade multicolor cocktail. The quality of the anti-CD45 MAb is critical. Be certain that the CD45 MAb stains brightly with lymphocytes. Both FITC and PerCP work well with CD45. Make sure that the optical filters match the fluorochrome combination selected.
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
Quality of fluorosphere suspensions The basic assumption of bead-based counting techniques is the consistency of the “fluorosphere event.” Unfortunately, particles in suspension tend to aggregate. This implies that if microspheres are not well suspended, a “fluorosphere event” may include more than one particle, thus altering the basic particle distribution. It has been reported that up to 15% of microspheres may aggregate. Therefore, steps must be taken to keep the microspheres in suspension to assure minimal sedimentation. TruCount and Flow-Count beads require different dispensing procedures. TruCount beads come as a lyophilized preparation in a 12 × 75–mm tube ready for addition of blood sample. Conversely, Flow-Count particles come as a suspension which must be pipetted with the same device used for blood dispensing. There are two approaches to counting fluorosphere events. Beckman-Coulter recommends gating on singlets only, whereas Becton Dickinson recommends including all fluorospheres. The universal template decribed in Basic Protocol 2 was designed to include both TruCount and Flow-Count fluorospheres. There are also fluorescence intensity and size differences between the two types of fluorospheres (see Fig. 6.8.4 and Fig. 6.8.5). To eliminate the risk of losing beads during the acquisition process, it is important to set the discriminator or threshold on fluorescence rather than on light scatter. Care is required in handling, storage, dispensing, and resuspension of microspheres. There is also the possible interference between cell and microsphere fluorescence. This may be an issue with TruCount fluorospheres in a situation where cells of interest are at the bright end of the fluorescence scale, near where the microspheres are observed. As already mentioned, other types of fluorospheres than those discussed here might be used, but they have to be tested and calibrated in advance.
Quality control A comprehensive quality control program covers instrument performance, instrument setup, and sample processing (UNITS 1.3, 1.4, & 6.1). Please note that not all CD45 MAbs are produced the same way. With multicolor immunophenotyping, the relative fluorescence intensity of the MAb may be a concern. As the technology discussed in this unit does rely to a large extent on the consistency of the fluorochrome label tagging the anti-CD45, it is important to pay special attention to this reagent. The stability of anti-CD45 can be monitored over time with a Levey-Jennings plot (UNIT 3.2) for a specific lot number. Pipetting technique All single-platform leukocyte subset counting requires high-precision dispensing techniques for both volumetric and flow count rate systems. It is recommended that reverse pipetting be used for all critical dispensing. Calibrated automated pipettors are also acceptable. The CVs for both precision and accuracy testing of reverse pipets should be less than 2% both for blood and fluorosphere suspension with 10 replicates. Gating considerations In general, as long as the CD3 and CD45 MAbs are of good quality, the setting of the appropriate heterogeneous gates is not difficult. However, it is critical that the protocol template selected be preserved as a template and be the same one applied, for as long as the protocol remains the choice for immunophenotyping. Integrity of leukoreduced blood product counts A prerequisite for quality rare-event analysis is the perfect cleaning of the fluidic system and the filtering of staining buffer through 0.22µm filters. A general warning must be issued about the care, experience, and patience needed in pipetting and dispensing viscous samples such as packed red cells and platelet concentrates. Despite good maintenance of the fluidic system, some background fluorescence events still pass the threshold and gate. WBCs are permeabilized by the staining buffer, and their light-scatter properties often decrease, so the light scattering parameter can be used to discriminate cell events from background. Spot checks for residual WBCs must be performed as part of a random lot-to-lot consistency of
6.8.24 Supplement 13
Current Protocols in Cytometry
sampling program for monitoring the filtering devices. The evaluation of virtually leukocytefree blood products introduces a notable exception to the basic concepts of robustness associated with cytometry data as discussed earlier.
Anticipated Results This unit deals with a relatively new technology; hence few published data are available on technology performance. When comparing the coefficient of variation (%CV’s) for CD4 T lymphocyte percentage and absolute count in an NIH quality assessment program, the %CV’s are 1.5 times higher for absolute counts (O’Gorman et al., 2000). This is in part due to the fact that the CD4 T lymphocyte percentage values are derived from a flow cytometer alone, whereas for the generation of absolute numbers, both a hematology instrument and a flow cytometer are required. This type of statistical information encourages the rapid implementation of the single-platform technology for absolute enumeration of leukocytes. The National External Quality Assessment Schemes for Leukocyte Immunophenotyping (NEQAS) has conducted a comparison of interlaboratory variability for both single platform and multiplatform technology. That study indicated that, for single and multiple platform, the precision of absolute count was 13.7% and 23.4%, respectively. Recently there have been two additional studies conducted in the U.S. One looked at the Tr uC ount absolute count tube (Schnizlein-Bick et al., 2000) and the other, separate study looked at the utilization of FlowCount fluorospheres (Reinmann et al., 2000) for absolute counting. Both were multisite studies comparing the single-platform bead technology with conventional multi-platform methods. In both cases, the single-platform technology improved interlaboratory precision for CD4 T cell determinations. Both studies indicated reduction of interlaboratory variability. Interlaboratory variation with TruCount and FlowCount system for CD4 absolute count was reduced by 8% and 8.6%, respectively. In conventional rare-event analysis, it is generally accepted that at least 100 positive events must be acquired to obtain adequate data representation for robust analysis. Following those assumptions, the analysis of a virtually leukocyte-free product may require the evaluation of a large blood volume over an unacceptably long acquisition time. This is an example where the concept of mathematical limits cannot be applied to practical flow cytometry. The monitoring of the absence of a rare population may
include a different algorithm compared with the monitoring of the presence of a rare cell subset. When microsphere-based methods are used, the number of beads acquired is a good indicator of the approximate blood volume analyzed. An appropriate acquisition gate counter can be set in most instruments to evaluate this parameter.
Time Considerations In general, analyses of leukoreduced blood products take longer, as they are rare-event analyses. Sample preparation is about 30 min with an incubation time of 15 min. An average sample run will take about 5 min. Result interpretation is another 5 min per sample. Absolute leukocyte subset enumeration has similar time requirements, except that these analyses take ∼1 min.
Literature Cited Bene, M.C., Kolopp Sarda, M.N., Kaissouni, J.E., De March Kennel, A., Mole, C. Kohler, C., and Faure, G. 1998. Automated cell counting in flow cytometry. Am J. Clin. Path. 110:321-326. Bland, J.M. and Altman, D.G. 1986. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet I:307310. Connely, M.C.M., Knight, M., Giorgy, J.V., Kagan, J., Landay, A.L., Parker, J.W., Page, E., Spino, C., Wilkening C., and Mercolino, T.J. 1995. Standardization of Absolute CD4+ lymphocyte count across laboratories: An evaluation of the Ortho CytoronAbsolute flow cytometry system on normal donors. Cytometry 22:200-210. Johnson, D., Hirschkorn, D., and Bush, M.P. 1995. National Heart Lung and Blood Institute retrovirus epidemiology donor study evaluation of four alternative methodologies for determination of absolute CD4 lymphocyte counts. J. AIDS Hum. Retrovir. 10:522-530. Lane, T.A. 1994. Leukocyte depletion of cellular blood components. Curr. Opin. Hematol. 1:443451. Mercolino, T.J., Connely, M.C. Meyer, E.J., Knight, M.D., Parker, J.W., Stelzer, G.T., and De Chirico, G. 1995. Immunologic differentiation of absolute lymphocyte count with an integrated flow cytometry system: a new concept for absolute T cell subset determination. Cytometry 22:48-59 Nicholson, J.K.A., Velleca, S., Jubert, S., Green, T.A., and Bryan, L. 1994. Evaluation of alternative CD4 technologies for the enumeration of CD4 lymphocytes. J. Immunol. Methods 177:4354. Nicholson, J.K.A., Stein, D., Mui, T., Mack, R., Hubbard, M., and Denny, T. 1997. Evaluation of a method for counting absolute numbers of cells with a flow cytometer. Clin. Diag. Lab. Immunol. 4:309-313.
Phenotypic Analysis
6.8.25 Current Protocols in Cytometry
Supplement 13
O’Gorman, M.R.G. and Gelman, R. 1997. Inter- and intrainstitutional evaluation of automated volumetric capillary cytometry for the quantitation of CD4- and CD8-positive T lymphocytes in the peripheral blood of persons infected with human immunodeficiency virus. Clin. Diag. Lab. Immunol. 4:173-179. O’Gorman, M.R.G., and Nicholson, J.K.A. 2000. Adoption of single-platform technology for enumeration of absolute T-lymphocyte subsets in peripheral blood. Clin. Diag. Lab. Immunol. 7:3,333-335. Rebulla, P. and Dzik, W. 1995. Multicenter evaluation of methods for counting residual white cells in leukocyte-depleted red blood cells. Vox Sang. 66:25-32. Reinmann, K.A., O’Gorman, M.R.G., Spritzler, J., Wilkening, C., Sabath, D. E., Helm, K., Campbell, D. E. 2000. Multisite comparison of CD4 and CD8 T-lymphocyte counting by single-versus multi-platform methodologies: Evaluation of Beckman Coulter Flow-Count fluorospheres and tetraONE System. Clin. Diag. Lab. Immunol. 7:344-351. Schlenke, P., Frohn, C., Klueter, H., Saballus, M., Hammers, H.J., Zajac, S.R., and Kirchner, H. 1998. Evaluation of a flow cytometric method
for simultaneous leukocyte phenotyping and quantification by fluorescent microspheres. Cytometry 33:310-317. Schnizlein-Bick, C.T., Spritzler, J., Wilkening, C., Nicholson, J.K.A., O’Gorman, R.G. 2000. Evaluation of the TruCount absolute-counting tubes for detrminating CD4 and CD8 cell numbers in human immunodeficiency virus-positive adults. Clin. Diag. Lab. Immunol. 7:336-343 Stewart, C.C. and Steinkamp, J.A. 1982. Quantitation of cell concentration using the flow cytometer. Cytometry 2:238-243. Valet, G. 1984. A new method for fast blood cell counting and partial differentiation by flow cytometry. Blut 49:83-90.
Contributed by Frank Mandy Health Canada Ottawa, Canada Bruno Brando Niguarda-Ca’ Granda Hospital Milano, Italy
Enumeration of Absolute Cell Counts Using Immunophenotypic Techniques
6.8.26 Supplement 13
Current Protocols in Cytometry
Immunophenotypic Identification, Enumeration, and Characterization of Human Peripheral Blood Dendritic Cells and Dendritic-Cell Precursors
UNIT 6.9
Difficulties in studying human dendritic cells (DCs) mainly relate to their low frequency in the sample and to the fact that, at present, no specific markers are available for the sensitive and specific identification of all DCs. In addition, due to the low frequency at which DCs are usually present in human samples, the collection of an adequate number of DCs has been a prerequisite for their further phenotypic analysis. Thus, the analysis of DCs has usually been performed either on mononuclear cells obtained through density-gradient centrifugation and other isolation procedures, or on DC-enriched cell cultures. In addition, DCs do not represent a homogeneous cell population, but different cell subsets have been described within DCs, further increasing the difficulties related to their phenotypic identification and characterization. Flow cytometry has been shown to be a well-suited technology for the analysis of single-cell suspensions, even when cells are present at very low frequencies (UNIT 6.3), as are the different DC subsets in normal human peripheral blood samples. Well-established applications of flow cytometry for the analysis of cells present at low frequencies are the enumeration of CD34+ hematopoietic progenitor cells (UNIT 6.4), the identification and phenotypic characterization of mast cells (UNIT 6.6), and minimal residual disease monitoring in patients with hematologic malignancies who achieved morphologically complete remission. Previous reports have shown that the use of a two-step acquisition procedure (i.e., acquisition of cells through a “live gate,” which increases the number of cells under study) represents an optimal approach for the analysis of rare cells. On the other hand, currently available techniques allow for the staining of erythrocyte-lysed whole peripheral blood samples; the analysis of all nucleated cells and not only of the mononuclear cell fraction would prevent any negative or positive selection of a particular cell subset. This unit describes procedures that allow the identification and enumeration of DCs and their subsets present in human erythrocyte-lysed whole peripheral blood (PB) samples using multiparametric flow cytometry (see Basic Protocol and Alternate Protocol). Special attention is paid to the combinations of reagents used for the identification of both all DCs and each DC subset, as well as on the multiparametric gating strategies used for the specific analysis of DCs. The Support Protocol describes the flow cytometric methods that can be applied for the immunophenotypical analysis of each DC subset. An extensive phenotypic characterization of the different DC subsets is crucial for a better understanding of the differential functional roles of each DC subset in the regulation of the immune responses, which, among other utilities, may help on the definition of the most efficient DC subpopulation to be used in future immunotherapy protocols. IMMUNOPHENOTYPIC IDENTIFICATION AND ENUMERATION OF PERIPHERAL BLOOD DENDRITIC CELLS BY FLOW CYTOMETRY The following method allows one to identify and enumerate peripheral blood (PB) dendritic cells (DCs) and the different DC subsets present in fresh blood samples. The monoclonal antibodies (mAbs) employed for these purposes should be conjugated with fluorescein isothiocyanate (FITC), phycoerythrin (PE), peridinin chlorophyll protein (PerCP; or any fluorochrome tandem, such as PE-Cy5 or PerCP-Cy5.5) as a third color, and either allophycocyanin (APC) or PE-Texas Red. Contributed by Julia Almeida and Clara Bueno Current Protocols in Cytometry (2001) 6.9.1-6.9.13 Copyright © 2001 by John Wiley & Sons, Inc.
BASIC PROTOCOL
Phenotypic Analysis
6.9.1 Supplement 17
Materials Whole anticoagulated peripheral blood (PB) containing ∼1.5 × 106 nucleated cells Fluorochrome-conjugated monoclonal antibody mixture (UNIT 4.2): Fluorescein isothiocyanate (FITC)-conjugated anti-CD3, -CD19, -CD56, and -CD14 Phycoerithrin (PE)-conjugated anti-CD16 Peridinin chlorophyll protein (PerCP)-, PE-cyanine 5 (PE-Cy5)- or PerCP-cyanine 5.5 (PerCP-Cy5.5)-conjugated anti-HLA-DR Allophycocyanin (APC)- or PE-Texas Red-conjugated anti-CD33 1× ammonium chloride lysing solution (APPENDIX 2A), freshly prepared Phosphate buffered saline (PBS; APPENDIX 2A) PBS containing 0.5% to 1% paraformaldehyde (optional) 12 × 75–mm polystyrene tubes Flow cytometer with either a single 488-nm emission laser or both 488- and 635-nm laser lines, four fluorescence detectors, and appropriate filter sets for detection of FITC, PE, and PerCP, PE-Cy5, or PerCP-Cy5.5, in addition to either APC or PE-Texas Red Software program for the analysis of flow cytometry FCS files (e.g., Paint-A-Gate Pro; Becton Dickinson) Additional reagents and equipment for assessing absolute leukocyte counts (number of cells per microliter; APPENDIX 3A) Prepare sample 1. Assess the nucleated cell count (i.e., the absolute leukocyte count) of the whole anticoagulated peripheral blood sample in a conventional hematology cell analyzer (APPENDIX 3A). 2. Label one 12 × 75–mm polystyrene tube for the specific identification and enumeration of DCs with fluorochrome-conjugated antibodies (see step 4). 3. Pipet a volume of whole anticoagulated peripheral blood containing ∼1.5 × 106 nucleated cells into the tube. 4. Add saturating amounts of each of the fluorochrome-conjugated mAbs: Anti-CD3-, anti-CD19-, anti-CD56-, and anti-CD14-FITC Anti-CD16-PE Anti-HLA-DR-PerCP, anti-HLA-DR-PE-Cy5, or anti-HLA-DR-PerCP-Cy5.5 Anti-CD33-APC or anti-CD33-PE-Texas Red Gently mix 5 sec. Note that a mixture of FITC-conjugated mAbs directed against T lymphocytes (CD3), B lymphocytes (CD19), NK cells (CD56), and monocytes (CD14) is used in order to exclude these cells (i.e., lineage+ cells) and to identify DCs by their lack of these markers (i.e., lineage− cells).
5. Incubate 15 min at room temperature in the dark. 6. Add 2 ml of 1× ammonium chloride lysing solution to each tube. Vortex vigorously 5 sec.
Phenotypic Analysis of Human PB Dendritic Cells
Alternatively, use commercially available reagents for erythrocyte lysis and fixation—i.e., Quicklysis (IMICO), FACS Lysing Solution (Becton Dickinson), Lysing Kit (Coulter), Lyse and Fix Reagent or OptiLyse (Immunotech), Whole Blood Lysing Solution (Caltag), PharMLyse (PharMingen), or Easy-Lyse (Leinco). Instructions for use are supplied with the reagents.
6.9.2 Supplement 17
Current Protocols in Cytometry
7. Incubate 10 min at room temperature in the dark. 8. Centrifuge 5 min at 540 × g, room temperature. 9. Discard the supernatant, resuspend the cell pellet, and add 4 ml phosphate buffered saline (PBS). 10. Centrifuge again 5 min at 540 × g, room temperature. 11. Discard the supernatant and resuspend the cell pellet in 500 µl PBS per tube. Mix gently and run on the flow cytometer. If samples cannot be acquired immediately, resuspend cells in 500 ìl PBS containing 0.5% to 1% paraformaldehyde per tube and store ≤24 hr at 4° to 8°C.
Set up instrument and acquire data 12. Perform acquisition with a flow cytometer equipped with either a single 488-nm laser or with both 488- and 635-nm laser lines, four fluorescence detectors, and appropriate filter sets for detection of FITC, PE, and PerCP, PE-Cy5, or PerCP-Cy5.5, in addition to either APC or PE-Texas Red. NOTE: The combinations of fluorochromes will vary depending on the excitation wavelengths of the lasers used—i.e., use FITC/PE/PerCP/APC or FITC/PE/PE-Cy5/APC for dual-laser (488 nm and 635 nm) instruments, and FITC/PE/PE-Texas Red/PE-Cy5 for single-laser (488 nm) instruments.
13. Position forward scatter (FS), side scatter (SS), and fluorescence (FL) windows of analysis and adjust electronic correction for spectral overlap according to standard procedures for lyse-and-then-wash immunophenotyping assays (UNIT 6.2). 14. Acquire data from a total of 30,000 events corresponding to the whole sample cellularity. Set up the following bivariate plots for data acquisition representation: FS versus SS (Fig. 6.9.1A) [CD3, CD19, CD56, and CD14] versus HLA-DR (Fig. 6.9.1B) SS versus HLA-DR (Fig. 6.9.1C) 15. Create an acquisition live gate in the SS versus HLA-DR dot plot. Display all events and draw a rectangular gate (R1) to include all HLA-DR-positive events (see Fig. 6.9.1C and D). 16. Acquire the events included in R1. This is done because DCs are present in peripheral blood samples at low frequency. A minimum of 3–5 × 105 events corresponding to the whole peripheral blood cellularity must be acquired in a normal human PB sample to obtain information on a minimum of 3000 to 5000 DCs. The total number of cells to be acquired in each case depends on the percentage of DCs present in the sample. It should be taken into account that DCs are not a homogeneous cell population and different DC subsets can be identified; therefore, the number of DCs acquired must be high enough to allow the unequivocal identification and analysis of the distribution of each DC subpopulation.
Analyze data NOTE: Any of the software programs commercially available for the analysis of flow cytometry FCS files can be used. Figures shown here have been analyzed by the Paint-A-Gate Pro software program. Phenotypic Analysis
6.9.3 Current Protocols in Cytometry
Supplement 17
A
B 104
HLA-DR-PerCP
Transformed SS
1024
768
512
256
103
102
101
100
0 0
256
512
768
1024
100
FS-HEIGHT
C
102
103
104
D
1024
1024
768
Transformed SS
Transformed SS
101
CD3-, CD14-, CD19-, and CD56-FITC
R1
512
256
768
R1
512
256
0
0 100
101
102
103
HLA-DR-PerCP
104
100
101
102
103
104
HLA-DR-PerCP
Figure 6.9.1 (A-C) Representative bivariate plots of flow cytometric data on dendritic cells, using the Basic Protocol. A total of 30,000 events/tube were acquired, corresponding to the total nucleated PB cells. (D) Events acquired through a live gate drawn on SS/HLA-DR+ (R1 on plot C). Only these events in region R1 were stored. Data were acquired using the CellQuest software program (Becton Dickinson).
17. Analyze data file 1, corresponding to the acquisition of all events present in the sample (first step of the acquisition procedure; step 14). Create the following bivariate dot plot histograms: FS versus transformed SS Transformed SS versus HLA-DR Transformed SS versus [CD3, CD19, CD56, and CD14] [CD3, CD19, CD56, and CD14] versus HLA-DR HLA-DR versus CD16 If the software program does not include the option of SS logarithmic transformation of the data, a linear scale can be used.
18. Draw a polygonal region (R2) in the FS versus transformed SS dot plot (Fig. 6.9.2A), in order to exclude debris and platelets. Events included in this region will be considered as total nucleated cells. Phenotypic Analysis of Human PB Dendritic Cells
19. Select all cells that are positive for HLA-DR in a new region (R3) in the transformed SS versus HLA-DR dot plot (Fig. 6.9.2B).
6.9.4 Supplement 17
Current Protocols in Cytometry
20. Among events contained in R3, draw another polygonal region (R4) to remove cells that are positive for any of the antigens recognized by the mixture of FITC-conjugated mAbs (Fig. 6.9.2C). The transformed SS together with the known fluorescence intensity of each cell subset for the different markers can help on the identification of the different cell populations that will be removed. The purpose of this step is to remove the following HLA-DR+ cell populations (selected in R3), which do not correspond to DCs: (1) monocytes, by their high expression of CD14, (2) B lymphocytes, by their positivity for CD19, and (3) activated T cells (CD3+).
21. Create another region (R5), which includes CD16+/HLA-DR+ events (Fig. 6.9.2D). Note that in R4 (Fig. 6.9.2C), events with low expression of CD14 and SS values between monocytes and lymphocytes should not be removed, since they correspond to an antigenpresenting cell subset which dimly expresses CD14. As this cell subpopulation is CD16+ and mature monocytes are CD16−, the HLA-DR versus CD16 dot plot must be used for a better discrimination between both cell populations.
22. From cells contained in R3 and R4 and R5, select only those with light-scatter values between monocytes and lymphocytes, by drawing a region (R6) on the FS/SS bivariate plot (Fig. 6.9.2E). Identify HLA-DR+/lineage− cells 23. Display events fulfilling the criteria of R2 and R3 and R4 and R6 or R2 and R3 and R5 and R6. These events correspond to the total DCs/CD16+ antigen-presenting cells present in the sample and can be better seen in the [CD3, CD19, CD56, CD14] versus HLA-DR dot plot (Fig. 6.9.2F).
Enumerate HLA-DR+/lineage− cells 24. Calculate the percent HLA-DR+/lineage− cells, which equals the total DCs/CD16+ multiplied by 100 divided by the total number of cells: Percent HLA-DR+/lineage− cells = events contained in both (R2 and R3 and R4 and R6) and (R2 and R3 and R5 and R6) × 100/events contained in R2. Calculate the absolute number of HLA-DR+/lineage− cells (number of HLADR+/lineage− cells per microliter peripheral blood): Absolute no. HLA-DR+/lineage− cells = percentage HLA-DR+/lineage− cells × absolute leukocyte count/100 Alternatively, calculation of absolute numbers of DCs can be directly derived from the flow cytometric measurement as defined in the Alternate Protocol. The following steps refer to the second file acquired after applying region R1 to select HLA-DR+ events.
25. Analyze data file 2, corresponding to the events included in R1 (step 16). Create the following bivariate dot plot histograms: FS versus transformed SS Transformed SS versus HLA-DR Transformed SS versus [CD3, CD19, CD56, and CD14] [CD3, CD19, CD56, and CD14] versus HLA-DR CD33 versus CD16 26. Follow steps 18 to 23, in order to select all HLA-DR+/lineage− cells present among the events contained in R1.
Phenotypic Analysis
6.9.5 Current Protocols in Cytometry
Supplement 17
B 1024
768
768
Transformed SS
Transformed SS
A 1024
512
256
R2
R3 512
256
0
0 0
256
512
768
100
1024
768
103
512
256
103
104
102
R5 101
R4
100
0 100
101
102
103
104
100
101
102
103
104
HLA-DR-PerCP
CD3-, CD14-, CD19-, and CD56-FITC
F
E
104
768
HLA-DR-PerCP
Transformed SS
102
D 104
CD16 PE
Transformed SS
C 1024
1024
101
HLA-DR-PerCP
FS-HEIGHT
512
R6 256
0
103
102
101
100 0
256
512
768
100
1024
101
102
103
104
CD3-, CD14-, CD19-, and CD56-FITC
FS-HEIGHT
G 104
R8 CD33 APC
103
102
R7 R9
101
100
Phenotypic Analysis of Human PB Dendritic Cells
100
101
102
103
104
CD16-PE
6.9.6 Supplement 17
Current Protocols in Cytometry
Figure 6.9.2 (on facing page) Representative bivariate dot plots containing information on PB dendritic cells analyzed with the Paint-A-Gate Pro software program (Becton Dickinson; see Basic Protocol). (A) FS/SS characteristics of all nucleated cells present in the PB (R2). (B) Events fulfilling the criteria of R2 and R3. (C) Events fulfilling the criteria for R2 and R3 and R4. (D) Events fulfilling the criteria for R2 and R3 and R5. (E) Events fulfilling the criteria for R2 and R3 and R4 and R6, as well as those contained in R2 and R3 and R5 and R6; all together these HLA-DR+/lineage− events correspond to total PB dendritic cells. (F) Total PB dendritic cells. (G) Identification of the three subsets of HLA-DR+/lineage− PB cells, based on their different reactivity for CD33 and CD16: R7 contains the CD16+ antigen-presenting cells (CD16+ black dots), R8 contains the CD16− /CD33strong+ DC subset (CD16− black dots), and the events included in R8 correspond to CD16− /CD123strong+ DCs (CD16− gray dots).
Identify HLA-DR+/lineage− cell subsets 27. Create three new regions (R7, R8, and R9) in the CD33 versus CD16 dot plot, to define the different minor subsets of antigen-presenting cells present in peripheral blood samples (Fig. 6.9.2G). R7 includes the CD16 + /HLA-DR + cell subset, R8 contains the CD16 − /CD33 strong+ DC subset, and R9 corresponds to the CD16 − /CD33 −/low+ DC subset. Note that these three cell subsets have a different reactivity (or intensity of fluorescence) for the CD33 antigen (Fig. 6.9.2G).
Enumerate HLA-DR+/lineage− cell subsets 28. After identifying the three cell subpopulations, calculate the percentage of each cell subset from the total number of peripheral blood HLA-DR+/lineage− cells. The relative number of each of these three subpopulations of antigen-presenting cells from the total number of peripheral blood nucleated cells (NCs) can be calculated as follows: Relative no. HLA-DR+/lineage− cell subset = (percentage HLA-DR+/ lineage− cell subset of total HLA-DR+/lineage− cells × percentage total HLA-DR+/lineage− cells of total NCs)/100 Calculate the absolute number of each DC subset (number of each DC subpopulation per microliter PB), by multiplying the relative number HLA-DR+/lineage− cell subset (obtained above), by the absolute number HLA-DR+/lineage− cells per microliter obtained in step 24 as: No. HLA-DR+/lineage− cell subset/µl = (relative no. HLA-DR+/lineage− cell subset × no. HLA-DR+/lineage− cells/µl)/100.
Alternatively, calculation of absolute numbers of each HLA-DR+/lineage− cell subset can be directly derived from the flow cytometric measurement (see Alternate Protocol).
FLOW CYTOMETRIC ENUMERATION OF PB DENDRITIC CELLS USING MICROBEADS AS AN INTERNAL REFERENCE TO DERIVE ABSOLUTE CELL COUNTS
ALTERNATE PROTOCOL
This protocol allows one to obtain absolute counts directly from the flow cytometer. Once this approach is preferred, internal volume-control microbeads and an erythrocyte-lysenonwash technique may be used as described below. Additional Materials (also see Basic Protocol) Flow-Count beads (Beckmann Coulter) with fluorescence emission in green, orange, and red (i.e., 515 to 650 nm) once excited with a 488-nm laser line Additional reagents and equipment for reverse pipetting (UNIT 6.4)
Phenotypic Analysis
6.9.7 Current Protocols in Cytometry
Supplement 17
Prepare sample 1. Stain anticoagulated whole blood samples (see Basic Protocol, steps 1 to 5). 2. Lyse erythrocytes by adding 2 ml 1× ammonium chloride lysing solution to each tube. Vortex gently 5 sec and incubate 10 min at room temperature in the dark. Alternatively, any other lysing solution—i.e., Quicklysis (IMICO), FACS Lysing Solution (Becton Dickinson), Lysing Kit (Coulter), Lyse and Fix Reagent (Immunotech), OptiLyse (Immunotech), Whole Blood Lysing Solution (Caltag), PharMLyse (PharMingen), EasyLyse (Leinco), or Lysing Solution (ORTHO-MUNE) can be used. Integrity of the beads once placed in a specific erythrocyte lysing solution should be tested prior to the analysis.
3. Immediately prior to use, resuspend Flow-Count beads manually by end-over-end rotation, to avoid generation of air bubbles. If beads have been sitting 12 hr or more, vortexing might be necessary. In this case, beads should be vortexed ∼2 hr prior to use so that any air bubbles generated during this process will not be present when beads are pipetted into the sample. Otherwise, do not vortex.
4. Add the same accurately measured volume of Flow-Count beads per tube as anticoagulated whole PB sample pipetted (i.e., if 100 µl PB were added per tube add 100 µl Flow-Count beads) by reverse pipetting (UNIT 6.4, also see Critical Parameters and Troubleshooting). Mix gently, but do not vortex. This procedure establishes a ratio of counting beads to volume of sample, so that the known number of added counting beads allows direct calculation of the absolute count of DCs per unit sample volume. In order to avoid bead loss, do not perform washing steps.
Set up instrument and acquire data 5. Position FS, SS, and fluorescence windows of analysis, and adjust electronic correction for spectral overlap according to standard procedures for lyse-no-wash immunophenotyping assays (UNIT 6.2). Since Flow-Count beads are detected on FACScan and FACSCalibur instruments at lower FS levels than lymphocytes, the FS threshold must be lowered so that singlet beads are not excluded from acquisition.
6. For data acquisition and analysis, follow a strategy similar to that described above (see Basic Protocol, steps 14 to 23), but in this case draw a region (R10) in SS versus HLA-DR+ red fluorescence in order to calculate the number of beads acquired during the first acquisition step. Determine absolute number of HLA-DR+/lineage− cells per ìl PB sample 7. To derive the absolute number of HLA-DR+/lineage− cells per µl PB sample, use the following formula (for more detailed information see UNIT 6.8): Number HLA-DR+/lineage− cells/µl PB = number of reference beads/µl added into the tube × (number of events acquired corresponding to HLADR+/lineage− cells/number of events acquired corresponding to beads) SUPPORT PROTOCOL
Phenotypic Analysis of Human PB Dendritic Cells
IMMUNOPHENOTYPIC CHARACTERIZATION OF PB DENDRITIC CELLS BY FLOW CYTOMETRY This protocol is a basic immunophenotypic procedure, which can be used for the phenotypic characterization of the different DC subsets present in nonmanipulated erythrocyte-lysed whole PB samples. The mAbs used should also be conjugated with FITC, PE, PerCP (or any fluorochrome tandem commercially available—e.g., PE-Cy5 or PerCP-Cy5.5), and either APC or PE-Texas Red.
6.9.8 Supplement 17
Current Protocols in Cytometry
Any marker of interest can be analyzed using this protocol. Since DCs have an important function as immunomodulatory cells, molecules involved in antigen presentation (such as HLA-ABC, HLA-DP, or HLA-DQ), cell adhesion, costimulatory molecules (i.e., CD40, CD80, CD86), immunoglobulin and complement receptors, as well as cytokine receptors (including chemokine receptors), can be analyzed. Additional Materials (also see Basic Protocol) Additional fluorochrome-conjugated mAbs (UNIT 4.2): PE-conjugated mAbs directed against the antigen under study Prepare sample 1. Assess the nucleated cell count in a conventional hematology analyzer (APPENDIX 3A). 2. Label as many 12 × 75–mm polystyrene tubes as necessary (e.g., [CD3, CD19, CD56, and CD14]-FITC/antimolecule under study-PE/anti-HLA-DR-PerCP/CD33-APC, all in the same tube). Other dye combinations (e.g., [CD3, CD19, CD56, and CD14]-FITC/antimolecule under study-PE/anti-HLA-DR-PE-Cy5 or -PerCP-CY5.5/CD33-APC or -PE-Texas Red), can also be used.
3. Pipet a volume of whole anticoagulated peripheral blood containing ∼1.5 × 106 nucleated cells into each tube. 4. Stain and process the tubes as described above (see Basic Protocol, steps 4 to 11). NOTE: If a mAb directed against the molecule under study is not available in a PE-conjugated format, a FITC-conjugated reagent can be used, but then the mixture of CD3, CD19, CD56, and CD14 should be PE conjugated.
Set up instrument and acquire data 5. Set up the instrument as described (see Basic Protocol, steps 12 to 14). 6. Create an acquisition live gate as described (see Basic Protocol, step 15), by drawing a rectangular region (R11) on the HLA-DR+ events. 7. Acquire the events included in R11 as described (see Basic Protocol, step 16). Analyze data 8. Derive the percentage of HLA-DR+/lineage− cells present in the sample (see Basic Protocol, steps 17 to 20). Note that in this protocol the HLA-DR versus CD16 dot plot described in step 17 of the Basic Protocol cannot be created, since CD16 has been replaced by the mAb directed against the specific molecule of interest. Likewise, the identification of HLA-DR+/lineage− cells (see Basic Protocol, step 23) is now performed by displaying the events fulfilling the criteria of R2 and R3 and R4 and R6 (but not R5).
9. Analyze the data file displaying the events included in R11. Create the following bivariate plots: FS versus transformed SS Transformed SS versus HLA-DR Transformed SS versus [CD3, CD19, CD56, and CD14] [CD3, CD19, CD56, and CD14] versus HLA-DR CD33 versus HLA-DR CD33 versus mAb-PE 10. Select all HLA-DR+/lineage− cells present in R11 (see Basic Protocol, steps 18 to 20 and 22 to 23). The identification of HLA-DR+/lineage− cells (see Basic Protocol, step 23) is now performed by displaying the events fulfilling the criteria of R2 and R3 and R4 and R6 (but not R5).
Phenotypic Analysis
6.9.9 Current Protocols in Cytometry
Supplement 17
Identify HLA-DR+/lineage− cell subsets 11. Create three new regions (i.e., R12, R13, and R14) in the CD33 versus HLA-DR dot plot in order to identify the different HLA-DR+/lineage− cell subpopulations present in the sample. R12 includes the CD33intermediate+/HLA-DR+ cell subset (corresponding to the CD16+/HLADR+/lineage− cell subset), R13 contains the CD33strong+/HLA-DRstrong+ DCs (CD16− /CD33strong+ DC subset), and R14 corresponds to the CD16−/CD33−/low+ DC subset, now identified as CD33low+/HLA-DRintermediate+ cells.
Characterize phenotype 12. To evaluate the expression of the different antigens analyzed after staining with specific PE-conjugated monoclonal antibody reagents in each antigen-presenting cell subset (CD33 versus mAb-PE dot plot), assess their mean fluorescence intensity (MFI) by calculating the mean channel, as well as the coefficient of variation for the PE-conjugated monoclonal antibody under study. Alternatively, for more precise and standardized quantitation of the mean fluorescence intensity for the protein under study, a mixture of beads stained with well-known amounts of PE (e.g., Quantibrite reagent kit; Becton Dickinson) is recommended. Results are then expressed as mean number of molecules of equivalent soluble fluorochrome (MESF). Instructions for use and how to transform the MFI values in MESF units are supplied by the manufacturers.
COMMENTARY Background Information
Phenotypic Analysis of Human PB Dendritic Cells
In recent years, there has been increasing interest in the study of dendritic cells (DCs), since they play a crucial role in the immune system for the initiation of responses (Banchereau and Steinman, 1998; Lane and Brocker, 1999) and such immunomodulatory functions make DCs attractive tools for their possible application in therapeutic protocols (Steinman, 1996). In spite of the large number of papers about DCs published in the last decade, the reported results on their phenotypic and functional characteristics are not always uniform. Such variability can be related, at least in part, to the use of different methodological approaches for the study of DCs. Additionally, the relatively low frequency at which DCs are present in normal human tissues (Schäkel et al., 1998) has been an important limitation to their study. Thus, the collection of an adequate number of DCs, either by cell isolation procedures (Williams et al., 1994) or by in vitro generation of DCs following culture of CD34+ cells (Bernhard et al., 1995), mononuclear cells, or monocytes (Pickl et al., 1996), has been a prerequisite for their further phenotypic analysis; however, methods that have been used for the enrichment of DCs may induce either positive or negative selection of particular dendritic cell subsets, as well as changes in the phenotype of the initial DC subsets present in the sample. In addition, the information currently available on the num-
bers of DCs present in human samples is usually based on DC-enriched samples, resulting in an inaccurate estimation of their real frequency. The lack of DC-specific antigens has also increased the difficulty of DC identification. Accordingly, the currently used phenotypic criteria for identifying DCs are based on their positivity for MHC class II molecules in the absence of reactivity for T (CD3), B (CD19), NK (CD56), and monocytic (CD14hi) lineage-associated markers (Young and Steinman, 1996; Hart, 1997). Furthermore, the DC system represents a heterogeneous group of antigen-presenting cells, in which different DC subsets (O’Doherty et al., 1994; Thomas and Lipsky, 1994; Olweus et al., 1997; Schäkel et al., 1998; Almeida et al., 1999a) that may play different roles in the immune responses (Rissoan et al., 1999; Arpinati et al., 2000) have been described, representing an additional limitation for their study. For several years, flow cytometry has proved to be an optimal approach for the analysis of cells present at very low frequencies in different human tissue samples (Orfao et al., 1996; Almeida et al., 1999b). From the methodological point of view, it has been shown that the use of a double-step acquisition procedure allows the unequivocal and sensitive detection of small populations of both malignant and normal hemopoietic cells, even when their frequency is as low as one malignant cell in 106 normal
6.9.10 Supplement 17
Current Protocols in Cytometry
cells (Almeida et al., 1999b). By applying a similar strategy, it has been shown that the use of multiple stainings analyzed by flow cytometry using appropriate multiple gating strategies allows accurate identification and analysis of large numbers of DCs, while avoiding extensive sample manipulation, such as that occurring once isolation procedures or in vitro cell cultures are used. In the authors’ experience, normal human PB DCs can be clearly identified by their positivity for HLA-DR in the absence of reactivity for lineage-associated markers (CD3, CD19, CD56, and strong reactivity for CD14 antigens) and by their typical light-scatter characteristics, with values between those of monocytes and lymphocytes (Almeida et al., 1999a). The accurate identification of DCs in nonmanipulated whole PB samples further allows the precise enumeration of DCs and even the direct calculation of DC subset absolute numbers from flow cytometric measurement once appropriate reference materials are used (see Alternate Protocol and UNIT 6.8 for more detail). This is due to the fact that all cell populations initially present in the sample are considered in the analysis. The combination of mAbs described in the Basic Protocol also allows the identification of three different minor subsets of antigen-presenting cells within the HLA-DR+/lineage− cells on the basis of their different reactivity for the CD16, CD33, and HLA-DR antigens. A first subset of antigen-presenting cells is characterized by expression of CD16+/CD33intermediate+/HLA-DR+ in the absence of lineage-associated antigens; a second subset is constituted by DCs, which are identified as CD16− /CD33strong+/HLA-DRstrong+, and the third subpopulation includes CD16−/CD33low+/HLADRintermediate+ DCs. Because these three subsets of antigen-presenting cells show different levels of CD33 and HLA-DR expression, both markers can be used to distinguish among them in normal human PB samples, with no need for using CD16. Accordingly, an extensive phenotypic characterization of each subpopulation of PB antigenpresenting cells can be performed, using PEconjugated mAbs directed against specific molecules of interest; therefore, the Support Protocol allows a systematic analysis of the immunophenotypic characteristics of normal human PB subsets of antigen-presenting cells, including DCs, and has definitively contributed to establishing that these cell subpopulations
are phenotypically different (Almeida et al., 1999a). As an example, the authors have shown that, compared to CD16− DCs, CD16+/HLADR+/lineage− cells show dim positivity for CD14 together with a higher expression of both Ig (CD16, CD32) and complement (CD11c, CD88) receptors, and lower adhesion molecules (CD29, CD62L), while reactivity for CD86 is higher (Almeida et al., 2001). In turn, CD16−/CD33strong+/HLA-DRstrong+ DCs display higher expression of the CD2 and CD5 costimulatory molecules as well as higher levels of CD38 and CD62L compared to the other two subpopulations of antigen-presenting cells, and they are CD45RA negative. Finally, CD16− /CD33low+/HLA-DRintermediate+ DCs show higher reactivity for the IL3-receptor α-chain (CD123+++), CD4, and CD45RA, and negativity for the myelomonocytic antigen CD13 compared to the other two cell subsets (Almeida et al., 1999a), and they are BDCA2+ and BDCA4+ (Dzionek et al., 2000). R ecen tly, CD16−/HLAstrong+ strong+ DR /CD33 DCs have been further su bd ivid ed into a m ajo r sub set of CD11c+/BDCA4+/CD1c+ and a minor population of CD11c+/BDCA4−/CD1c−/BDCA3+ DCs, which still deserve further characterization (Dzionek et al., 2000).
Critical Parameters and Troubleshooting Sample quality and preparation For the enumeration of DCs, fresh samples with high cell viability (>95%) should be used. It has been shown that samples which are not immediately processed after collection are not in optimal condition for calculating the absolute numbers of a given cell population, since they contain varying numbers of dead cells; therefore, samples are ideally processed within 3 hr after being obtained in order to perform reliable identification, enumeration, and phenotypic characterization of antigen-presenting cells, including DCs. Sample storage, handling, and preparation are also important variables that might affect the accuracy of DC enumeration. Accordingly, the accuracy of pipetting, which should be performed by reverse pipetting (UNIT 6.4), is of critical importance in single-platform assays, where the result is directly affected by the amount of both sample and counting beads added. Likewise, the inclusion of washing steps is not recommended when microbeads are used
Phenotypic Analysis
6.9.11 Current Protocols in Cytometry
Supplement 17
for enumerating DCs, as washing could result in selective loss of either cells or counting beads. Regarding immunophenotypical studies for the characterization of each DC subset, the time from sample collection to its technical processing may not be so limiting as for DC enumeration; however, it should be noted that at present, information in the literature in this respect is scanty. Thus, processing of samples as soon as possible is recommended for the phenotypic characterization of DC subsets.
Phenotypic Analysis of Human PB Dendritic Cells
Choice of mAbs and fluorochromes Currently, no specific markers exist that can identify all human DC subsets. Thus, it is recommended to use a mixture of mAbs directed against T and B lymphocytes, NK cells, and monocytes in order to be able to select for the minor populations of antigen-presenting cells present in the sample, including DCs, on the basis of their negativity for these markers; therefore, all the mAbs included in the cocktail must be conjugated with the same fluorochrome molecule. In the authors’ experience, the best combination of such mAbs for normal human PB includes: CD3 (to exclude all mature T cells), CD19 (for B lymphocytes), CD56 (for NK cells; alternatively, CD16 could be used but would prevent the identification of CD16+ antigen-presenting cells), and CD14 (mature monocytes can be excluded by their strong positivity for this antigen). In principle, all these mAbs should be FITC-conjugated, since PE is better used for either specific identification of CD16 reactivity or the expression of other molecules of interest. Positive identification of peripheral blood DCs requires the presence of anti-HLA-DR, and CD33 allows for the specific discrimination between the different subsets of DCs and the CD16+/HLA-DR+/lineage− antigen-presenting cells. Additionally, it should be noted that the identification and enumeration of DCs in PB samples that contain abnormally increased numbers of other HLA-DR+ cells (e.g., malignant blast cells, CD34+ hemopoietic progenitors) may need special consideration in order to exclude these events from the DC region (i.e., inclusion of CD34 mAb in the exclusion cocktail). If only a three-color instrument is available, the study of DCs would then be, in principle, limited to their identification and enumeration, since at present the fourth color would be necessary for their characterization.
Number of events to acquire To obtain statistically reliable DC subset counts by flow cytometry, it is crucial to acquire a large number of events included in the fraction in which DCs are located. As the different DC subsets are considered to be rare events, in the protocols described above for the enumeration of peripheral blood DC subsets, including the CD16+ subset of antigen-presenting cells, the authors recommend acquiring and storing information on a minimum of 3000 to 5000 total DCs/CD16+ antigen-presenting cells. Since these cells represent ∼1% of all nucleated cells in normal human peripheral blood, it is necessary to acquire information on at least 3–5 × 105 events, corresponding to the whole peripheral blood cellularity. Although information can be stored for all events acquired, the authors recommend use of a two-step acquisition procedure, information from the second step being stored exclusively through a live gate on HLADR+ events, to prevent computer-memory problems with an over-large listmode file.
Anticipated Results Enumeration of DC subpopulations present in normal human peripheral blood samples The overall percentage of PB DCs and CD16+ antigen-presenting cells from normal human individuals is ∼1% of all nucleated cells (1.03% ± 0.33%), ranging between 0.46% and 1.99%. The most frequent cell subset is CD16+, representing 0.72% ± 0.35% of all nucleated cells. From the other two subsets, CD16− /CD33strong+ DCs represent ∼0.2% (0.19% ± 0.08%) of all PB nucleated cells and the CD16− /CD123strong+ DC sub po pu lation ∼1.0% (0.11% ± 0.04%; Almeida et al., 2001). In absolute numbers the distribution of these cell subsets is as follows: 39.9 ± 17.4 CD16+ antigen-presenting cells/µl, 10.5 ± 3.9 CD16− /CD33strong+ DCs/µl, and 6.3 ± 2.6 CD16− /CD123strong+ DCs/µl. Immunophenotypic characteristics of DC subpopulations present in normal human peripheral blood samples The technical procedure described above represents a sensitive and specific method for the systematic study of the immunophenotypic characteristics of the different DC subsets. As an example, the authors have seen that the three different DC subsets can be distinguished by their distinct reactivity for Fcγ and complement receptors, certain adhesion and costimulatory molecules, in addition to their different pattern
6.9.12 Supplement 17
Current Protocols in Cytometry
of expression of CD33 and HLA-DR (Almeida et al., 2001). In principle, any molecule of interest present either on the surface or inside the DCs can be specifically studied and precisely measured using the four-color flow cytometric protocols described in this unit.
Time Considerations
Lane, P.J. and Brocker, T. 1999. Developmental regulation of dendritic cell function. Curr. Opin. Immunol. 11:308-313. O’Doherty, U., Peng, M., Gezelter, S., Swiggard, W.J., Betjes, M., Bhardwaj, M., and Steinman, R.M. 1994. Human blood contains two subsets of dendritic cells, one immunologically mature and the other immature. Immunology 82:487493.
Cell preparation takes 45 to 60 min. Acquisition on the flow cytometer will take 10 to 15 min per sample. The Alternate Protocol will take 1 hr total.
Olweus, J., BitMansour, A., Warnke, R., Thompson, P.A., Carballido, J., Picker, L.J., and Lund-Johansen, F. 1997. Dendritic cell ontogeny: A human dendritic cell lineage of myeloid origin. Proc. Natl. Acad. Sci. U.S.A. 94:12551-12556.
Literature Cited
Orfao, A., Escribano, L., Villarrubia, J., Velasco, J.L., Cerveró, C., Ciudad, J., Navarro, J.L., and San Miguel, J.L. 1996. Flow cytometric analysis of mast cells from normal and pathologic human bone marrow samples. Identification and enumeration. Am. J. Pathol. 149:1493-1499.
Almeida, J., Bueno, C., Algueró, M.C., Sanchez, M.L., Cañizo, M.C., Fernandez, M.E., Vaquero, J.M., Escribano, L., Laso, F.J., San Miguel, J.F., and Orfao, A. 1999a. Extensive characterization of the immunophenotype and the pattern of cytokine production by distinct subpopulations of normal human peripheral blood dendritic cells. Clin. Exp. Immunol. 118:392-401. Almeida, J., Orfao, A., Ocqueteau, M., Mateo, G., Corral, M., Caballero, M.D., Blade, J., Moro, M.J., Hernandez, J., and San Miguel, J.F. 1999b. High sensitive immunophenotyping and DNA ploidy studies for the investigation of minimal residual disease in multiple myeloma. Br. J. Haematol. 107:121-131. Almeida, J., Bueno, C., Algueró, M.C., Sanchez, M.L., Escribano, L., Diaz-Agustin, B., Vaquero, J.M., Laso, F.J., San Miguel, J.F., and Orfau, A. 2001. Comparative analysis of the morphological, cytochemical, immunophenotypical and functional characteristics of normal human per ip h er al bloo d lineag e−/CD16+/HLADR+/CD14−/low cells, CD14+ monocytes and CD16− dendritic cells. Clin. Immunol. In press. Arpinati, M., Green, C.L., Heimfeld, S., Heuser, J.E., and Anasetti, C. 2000. Granulocyte-colony stimulating factor mobilizes T helper 2-inducing dendritic cells. Blood 95:2484-2490. Banchereau, J. and Steinman, R.M. 1998. Dendritic cells and the control of immunity. Nature 392:245-252. Bernhard, H., Disis, M.L., Heimfeld, S., Hand, S., Gralow, J.R., and Cheever, M.A. 1995. Generation of immunostimulatory dendritic cells from human CD34+ hematopoietic progenitor cells of the bone marrow and peripheral blood. Cancer Res. 55:1099-1104. Dzionek, A., Fuchs, A., Schmidt, P., Cremer, S., Zysk, M., Miltenyi, S., Buck, D.W., and Schmitz, J. 2000. BDCA-2, BDCA-3, and BDCA-4: Three markers for distinct subsets of dendritic cells in human peripheral blood. J. Immunol. 165:6037-6046.
Pickl, W.F., Majdic, O., Kohl, P., Stöckl, J., Riedl, E., Scheinecker, C., Bello-Fernández, C., and Knapp, W. 1996. Molecular and functional characteristics of dendritic cells generated from highly purified CD14+ peripheral blood monocytes. J. Immunol. 157:3850-3859. Rissoan, M.-C., Soumelis, V., Kadowaki, N., Grouard, G., Briere, F., Waal Malefyt, R., and Liu, Y.-J. 1999. Reciprocal control of T helper cell and dendritic cell differentiation. Science 283:1183-1186. Schäkel, K., Mayer, E., Federle, C., Schmitz, M., Riethmüller, G., and Rieber, P. 1998. A novel dendritic cell population in human blood: Onestep immunomagnetic isolation by a specific mAb (M-DC8) and in vitro priming of cytotoxic T lymphocytes. Eur. J. Immunol. 28:4084-4093. Steinman, R.M. 1996. Dendritic cells and immunebased therapies. Exp. Hematol. 24:859-862. Thomas, R. and Lipsky, P.E. 1994. Human peripheral blood dendritic cell subsets. Isolation and characterization of precursor and mature antigen-presenting cells. J. Immunol. 153:40164028. Williams, L.A., Egner, W., and Hart, D.N. 1994. Isolation and function of dendritic cells. Int. Rev. Cytol. 153:41-103. Young, J.W. and Steinman, R.M. 1996. The hematopoietic development of dendritic cells: A distinct pathway for myeloid differentiation. Stem Cells 14:376-387.
Contributed by Julia Almeida and Clara Bueno Universidad de Salamanca Salamanca, Spain
Hart, D.N.J. 1997. Dendritic cells: Unique leukocyte populations which control the primary immune response. Blood 90:3245-3287. Phenotypic Analysis
6.9.13 Current Protocols in Cytometry
Supplement 17
Immunophenotypic Analysis of Platelets
UNIT 6.10
With an average diameter of 3 µm, platelets are the smallest circulating cellular component in peripheral blood. The primary role of circulating platelets is to maintain hemostasis. The evaluation of platelets by flow cytometry has proven beneficial in the investigation of many disease states, including inherited defects such as Bernard-Soulier syndrome, Glanzmann thrombasthenia, and storage pool disease (Michelson et al., 2001). Flow cytometric techniques have been used in blood bank applications such as quality control of platelet concentrates, immunophenotyping of platelet surface receptor polymorphisms, platelet crossmatching, and detection of feto-maternal anti-platelet antibodies. Platelet hyporeactivity may result in potentially life-threatening bleeding including intracranial hemorrhage, while platelet hyperreactivity may result in intravascular thrombosis resulting in potentially life-threatening acute myocardial infarction or stroke. Consequently, antiplatelet therapies designed to reduce platelet responsiveness in vivo are now common practice in clinical cardiovascular medicine. Perhaps most common is the use of flow cytometry to study the role of platelet function and platelet activation in cardiovascular disease. A more complete list of the applications of flow cytometry to the study of platelets is shown in Table 6.10.1 and discussed in Michelson et al. (2001). Resting platelets constitutively express many surface glycoproteins that are easily identified by flow cytometry. Upon platelet activation, many surface receptors are modulated in both copy number and conformation, while others, absent from the resting platelet surface, are newly expressed. This unit describes several strategies to evaluate platelet function by evaluating surface receptor expression on resting and activated platelets using flow cytometry. Three methods are described here in detail: determination of resting platelet surface receptor expression (see Basic Protocol 1 and Alternate Protocol); determination of platelet activation using P-selectin (CD62P) expression (see Basic Protocol 2), which reflects platelet α-granule release (McEver, 2001), or PAC1 binding, which detects the activated conformation of glycoprotein (GP) IIb-IIIa (integrin αIIbβ3; Shattil et al., 1985); and determination of procoagulant platelets and platelet-derived microparticles using annexin V binding or monoclonal antibodies specific for coagulation factors V/Va or X/Xa (see Basic Protocol 3; Furman et al., 2000). The methods described here are performed using the more physiologically relevant milieu of whole blood, which has the following advantages over platelet-rich plasma or washed platelet systems: (1) red cells and leukocytes are present, both of which affect platelet activation; (2) minimal sample manipulation minimizes artifactual in vitro activation and potential loss of platelet subpopulations; (3) both the activation state of circulating platelets and the reactivity of circulating platelets can be determined; (4) only minuscule volumes (∼5 µl) of blood are required, making whole-blood flow cytometry particularly advantageous for neonatal studies; and (5) platelets of patients with profound thrombocytopenia can also be accurately analyzed. This unit does not specifically address platelet-associated IgG, heparin-induced thrombocytopenia, monitoring of GPIIb-IIIa receptor antagonists, reticulated platelets (see UNIT 7.10), leukocyte-platelet aggregate formation, or platelet counting. For a review of platelet-associated analysis techniques not covered by this unit see Schmitz et al. (1998).
Phenotypic Analysis Contributed by Lori A. Krueger, Marc R. Barnard, A.L. Frelinger III, Mark I. Furman, and Alan D. Michelson
6.10.1
Current Protocols in Cytometry (2002) 6.10.1-6.10.17 Copyright © 2002 by John Wiley & Sons, Inc.
Supplement 19
Table 6.10.1 Applications of Flow Cytometry to the Study of Plateletsa
Measurement of platelet activationb Activation-dependent monoclonal antibodies/reagents Modulation of constitutively expressed surface receptors Procoagulant platelet-derived microparticles Leukocyte-platelet aggregates Platelet-platelet aggregates Diagnosis of specific disorders Bernard-Soulier syndrome Glanzmann thrombasthenia Storage pool disease Heparin-induced thrombocytopenia Immune thrombocytopenias Monitoring of antiplatelet agents GPIIb-IIIa antagonists Thienopyridines Monitoring of thrombopoiesis Reticulated platelets Blood bank applications Quality control of platelet concentrates Identification of leukocyte contamination in platelet concentrates Immunophenotyping of platelet HPA-1a Detection of maternal and fetal anti-HPA-1a antibodies Platelet cross-matching Platelet counting Research applications Platelet survival, tracking, and function in vivo Platelet recruitment Bacteria-platelet interactions Calcium flux Cytoskeletal rearrangement Fluorescence resonance energy transfer Signal transduction aTable order reflects the most commonly studied, and relevant applications. bIncludes circulating activated platelets, platelet hyperreactivity, or platelet hyporeactivity.
STRATEGIC PLANNING Blood Collection Careful experimental planning is required for accurate and consistent results when immunophenotyping platelets by flow cytometry. To minimize ex vivo platelet activation, blood samples should be processed within ∼30 min after drawing blood for many assays. The act of drawing blood is itself a potential source of artifactual platelet activation; therefore, the following recommendations are suggested (Michelson et al., 2001): Use a light tourniquet or none at all Use a 21-G (or larger bore) needle Ensure a smooth draw (i.e., good flow) Discard the first 2 ml of blood drawn Immunophenotypic Analysis of Platelets
Adhering to these recommendations will minimize tissue thromboplastin contamination of blood samples and red cell hemolysis that could lead to artifactual platelet activation.
6.10.2 Supplement 19
Current Protocols in Cytometry
Table 6.10.2
Anticoagulants Used in the Study of Platelets
Anticoagulant
Mechanism of action
Weak Ca2+ chelator Chelates Ca2+ and increases intracellular cAMP, keeping platelets “quiet” Activated coagulation factor XII inhibitor Strong Ca2+ chelator, dissociates GPIIb-IIIa complex Combines with anti-thrombin III to inhibit thrombin activity Hirudin Direct thrombin inhibitor D-Phenylalanyl-L-prolyl-L-arginine Direct thrombin inhibitor chloromethyl ketone (P-PACK) Sodium citrate Weak Ca2+ chelator
Acid citrate dextrose (ACD) Citrate theophylline adenosine dipyridimole (CTAD) Corn trypsin inhibitor EDTAa Heparina
aThese anticoagulants should be avoided for evaluation of platelet function studies by flow cytometry (see Strategic
Planning).
Each laboratory should determine whether their method of collection, including the drawing of samples through angioplasty and other catheters, results in artifactual in vitro platelet activation, as determined by the binding of activation-dependent monoclonal antibodies. Choice of Anticoagulant Although sodium citrate (a weak calcium chelator) is the most common anticoagulant used for platelet studies, others have been successfully used. EDTA (a strong calcium chelator) should be avoided, because it causes dissociation of the integrin αIIbβ3 (GPIIbIIIa) complex. Heparin should also be avoided because it binds to, and may activate, platelets. Nonchelating anticoagulants such as P-PACK (a direct thrombin inhibitor) may be preferable for the monitoring of GPIIb-IIIa antagonist therapy. The anticoagulants listed in Table 6.10.2 have reportedly been used in studies of platelets. Sample Handling The length of time between sample draw and sample preparation should be minimized to reduce spontaneous platelet activation. The blood should be properly mixed with anticoagulant, avoiding unnecessary agitation prior to testing. The whole-blood samples should be collected and maintained in a container with nonwettable surfaces such as siliconized glass or polypropylene. Experimental Design Many platelet surface receptors are modulated during platelet activation (see Table 6.10.3). For example, GPIb-IX-V may be cleaved and/or internalized to the surface-connected canalicular system upon platelet activation (Michelson et al., 1996a). This must be taken into consideration in experimental design. For example, if using a GPIbα-specific antibody (i.e., anti-CD42b) as a platelet identifier, then an adjustment in the instrument threshold (or discriminator) may be necessary when evaluating activated versus resting platelets. Alternatively, if the reduction in platelet GPIbα is being used as an indicator of platelet activation, then the maximal change in receptor expression will be observed only if platelets are labeled with the GPIbα-specific antibody after platelet activation has taken place. If a directly-conjugated GPIb-IX-specific test monoclonal antibody is added to the platelets prior to activation, then the activation-induced redistribution of GPIb-IX to the surface-connected canalicular system will not result in a significant decrease in platelet fluorescence, because fluorescence will be detected irrespective of whether the conjugated antibody is on the surface of or within the platelet; therefore, in flow cytometric
Phenotypic Analysis
6.10.3 Current Protocols in Cytometry
Supplement 19
Table 6.10.3 Activation-Dependent Changes in Platelet Surface Labeling of Monoclonal Antibodies and Annexin Va
Activation-dependent platelet surface change Changes in surface receptor expression CD36 GPIb-IX GPIIb-IIIa Conformational changes in GPIIb-IIIa (integrin αIIbβ3) Ligand-induced binding sites (LIBS) PAC1 Receptor-induced binding sites on fibrinogen (RIBS) Development of a procoagulant surface Factor VIII binding Factor V/Va binding Factor X/Xa binding Phosphatidylserine expression (detected by annexin V) Exposure of granule membrane proteins CD40L (or CD154) CD63 (lysosomes) LAMP-1 (lysosomes) LAMP-2 (lysosomes) Lectin-like oxidized LDL receptor-1 (LOX-1) P-selectin (CD62P, α-granules) Platelet surface binding of secreted platelet proteins Multimerin Thrombospondin
Resting platelet
Activated platelet
+ ++ ++
++ + +++
− − −
+++ +++ +++
− − − −
+++ +++ +++ +++
− − − − − −
+ ++ ++ ++ + +++
− −
+ +
aAnnexin V is a 35 to 36 kDa protein that binds to phosphatidylserine in the presence of Ca2+.
assays examining platelet surface GPIb-IX modulation, GPIb-IX-specific antibodies that are directly conjugated must be added to the assay after the addition of the agonist. BASIC PROTOCOL 1
Immunophenotypic Analysis of Platelets
IMMUNOPHENOTYPING OF PLATELET SURFACE RECEPTORS This procedure may be used to qualitatively and/or quantitatively evaluate a platelet surface receptor such as GPIIb-IIIa (reduced or absent in Glanzmann thrombasthenia) or GPIb-IX-V (reduced or absent in Bernard-Soulier syndrome). Additionally, this protocol may be used to evaluate the in vivo activation status of circulating platelets. The optimal final concentration of platelet-specific monoclonal antibody reagents must be determined by titration. At a minimum, two antibodies, each conjugated to a different fluorochrome (e.g., fluorescein and phycoerythrin), are used. The fluorescent conjugate of the platelet identifier determines the thresholding parameter used in the flow cytometric analysis. The choice of the platelet identifier is determined by the analysis being performed. For example, anti-CD42a or anti-CD42b (GPIX- and GPIbα-specific, respectively) are used as platelet identifiers when measuring surface expression of GPIIb-IIIa (CD41 and CD61) in the investigation of Glanzmann thrombasthenia, an inherited deficiency of GPIIb-IIIa. Anti-CD41 or anti-CD61 may be used as platelet identifiers when investigating Bernard-Soulier syndrome, an inherited deficiency of GPIb-IX-V (CD42b, CD42a, and CD42d). Anti-CD41, -CD61, -CD42a, or -CD42b may be used as platelet identifiers when evaluating activation-dependent receptors such as P-selectin
6.10.4 Supplement 19
Current Protocols in Cytometry
(recognized by anti-CD62P) and PAC1; however, surface expression of GPIIb-IIIa and GPIb-IX is modulated upon platelet activation, which must be considered when selecting a platelet identifier. Monoclonal antibody reagents are prepared in modified HEPES/Tyrode’s (HT) buffer. Materials Whole blood (WB) containing anticoagulant (see Strategic Planning) or isolated platelets (see Support Protocol) Modified HT buffer (see recipe) Platelet-specific antibody cocktail titrated in modified HT buffer (minimum two antibody specificities each conjugated to a different fluorochrome): Specific platelet identifier: monoclonal anti-CD41, -CD61, -CD42a, or -CD42b Marker of platelet activation: monoclonal anti-CD62P or PAC1 Negative control: antibody isotype-, concentration-, fluorochrome-, and F:P ratio-matched to the activation marker, or blocking agent inhibiting platelet-specific marker binding 1% formalin fixative (see recipe) 1. Within 30 min of blood draw, dilute whole blood (WB) containing anticoagulant 1/10 in modified HT buffer (e.g., 10 µl WB and 90 µl HT). In the diagnosis of Glanzmann thrombasthenia and Bernard-Soulier syndrome it is wise to use blood from a normal healthy donor analyzed in parallel with patient blood to differentiate normal and abnormal expression of GPIIb-IIIa or GPIb-IX, respectively.
2. Immediately mix diluted whole blood (dWB) with appropriately titrated platelet-specific antibody cocktail (e.g., 10 µl dWB and 30 µl antibody cocktail). Include a negative control if a platelet activation marker is used. Incubate at room temperature (e.g., 20 min). Individual laboratories must determine the optimal antibody concentration and incubation time.
3. Fix labeled cells with 10× to 20× total assay volume (e.g., 400 to 800 µl) of 1% formalin solution. Let fixation proceed 15 min at room temperature, then place tubes at 4°C until analysis. Stability of fixed preparations must be determined by individual laboratories.
4. Analyze by flow cytometry using the following setup: a. Use the platelet identifier as the thresholding (or discriminating) parameter (Fig. 6.10.1A). b. Collect light-scatter parameters in logarithmic mode. (Heterogeneity in platelet size and platelet morphology contributes to a relatively broad distribution of light scattering detectable by flow cytometry.) c. Analyze diluted fixed sample using low flow rate (e.g., 150 to 250 platelet events per second) to minimize, as far as possible, coincidence between two or more events. d. Identify single platelet events by their characteristic light scatter and positive labeling with platelet-specific identifier (Fig. 6.10.1B). e. Measure the platelet surface receptor or receptors of interest from gated events (Fig. 6.10.1C). Platelets are smaller than lymphocytes and erythrocytes; therefore, high-voltage settings used for leukocyte and erythrocyte studies may not be appropriate for platelet analysis.
Phenotypic Analysis
6.10.5 Current Protocols in Cytometry
Supplement 19
100
103
102
103
104 102
101
Forward scatter 101
B
100
104 103 102 101 100
Platelet identifier fluorescence
A
104
100
101
0
101
102
103
P-selectin fluorescence
104
103
104
103
104
102
103
D
101 100
20
30
Forward scatter
104
C
100
102
Side scatter
10
Count
40 50 60
Side scatter
100
101
102
FX/Xa fluorescence
Figure 6.10.1 Evaluation of platelet activation by whole-blood flow cytometry. (A) Platelets labeled with specific platelet identifier. The instrument threshold (or discriminator) is set on the fluorescence parameter corresponding to the conjugate of the platelet identifier. (B) Characteristic light-scatter profile of a platelet population in diluted whole blood. Data are collected and displayed using logarithmic-orthogonal and logarithmic-forward light scatter. Single platelet events are identified by their characteristic light-scatter properties and positive labeling with a platelet-specific monoclonal antibody reagent. (C) Single-parameter fluorescence histogram of the platelet activation marker P-selectin (CD62P). The positive analysis region is determined by the negative isotype control (thin solid line). Events displayed are generated from gated events in both panel A and panel B. (D) Determination of procoagulant platelets and platelet-derived microparticles. Two-parameter histogram displaying Factor X/Xa binding versus forward-angle light scatter. The analysis region for platelets capable of binding coagulation protein–-activated factors V and X or expressing phosphatidylserine is established on this two-parameter histogram.
ALTERNATE PROTOCOL
IMMUNOPHENOTYPING OF PLATELET SURFACE RECEPTORS Whole blood may be fixed first and subsequently labeled with platelet-specific reagents. Many platelet surface receptors are well preserved and recognized by specific monoclonal antibodies after formalin fixation; however, labeling intensity may diminish with prolonged fixation. Antibody concentrations required for optimal labeling may be different for fixed versus unfixed cells; therefore, optimal antibody concentrations should be verified by titration using fixed cells. Some antigen epitopes such as that recognized by PAC1 may not label after formalin fixation.
Immunophenotypic Analysis of Platelets
6.10.6 Supplement 19
Current Protocols in Cytometry
Additional Materials (also see Basic Protocol 1) 2% formalin fixative (see recipe) 1. Within 30 min of blood draw, dilute WB 1:1 with 2% formalin fixative. (i.e., 100 µl WB and 100 µl of 2% formalin). Fix 15 to 60 min at room temperature. Store samples at 4°C. Antibody labeling may decrease in intensity with prolonged fixation; therefore, the length of time that fixed samples can be stored prior to labeling must be determined by individual laboratories for each antibody clone being used.
2. Dilute fixed WB 1:10 in modified HT buffer. Dilution can be performed prior to storage at 4°C.
3. Mix dWB with appropriately titrated antibody cocktail (e.g., 20 µl dWB and 20 µl antibody cocktail) and incubate at room temperature (e.g., 20 min). Individual laboratories must determine the optimal antibody concentration and incubation time.
4. Dilute labeled cells with 10× to 20× total assay volume (e.g., 400 to 800 µl) of 1% formalin solution. Place tubes at 4°C until analysis. Stability of fixed preparations must be determined by individual laboratories.
5. Analyze by flow cytometry as described above (see Basic Protocol 1, step 4) DETERMINATION OF PLATELET ACTIVATION USING P-SELECTIN OR PAC1 EXPRESSION
BASIC PROTOCOL 2
This procedure may be used to evaluate platelet reactivity in response to agonist. Materials Whole blood (WB) containing anticoagulant (see Strategic Planning) or isolated platelets (see Support Protocol) Modified HT buffer (see recipe) Platelet-specific antibody cocktail titrated in modified HT buffer (minimum two antibody specificities each conjugated to a different fluorochrome): Specific platelet identifier: e.g., monoclonal anti-CD41, -CD61, -CD42a, -CD42b Specific marker of platelet activation: e.g., monoclonal anti-CD62P or -PAC1 Platelet agonist (see recipe): e.g., ADP, epinephrine, human α-thrombin, thrombin receptor-activating peptide (TRAP) Negative controls: antibody isotype-, concentration-, fluorochrome-, and F:P ratio-matched to the specific activation marker, or blocking agent that inhibits the platelet-specific marker binding 10 mM GPRP (see recipe) 1% formalin fixative (see recipe) 1. Within 30 min of blood draw, dilute WB 1:10 in modified HT buffer (i.e., 10 µl WB and 90 µl HT). 2. Immediately mix dWB with appropriately titrated platelet-specific antibody cocktail and with platelet agonist (e.g., 10 µl dWB, 20 µl antibody cocktail, and 10 µl platelet agonist). Include a negative control if a platelet activation marker is used. Incubate at room temperature (e.g., 20 min). If thrombin is being used as the platelet agonist, add GPRP to a final concentration of 2.5 µm to prevent fibrin polymerization and clot formation (Michelson, 1994).
Phenotypic Analysis
6.10.7 Current Protocols in Cytometry
Supplement 19
The fluorescent conjugate of the platelet identifier determines the thresholding parameter used in the flow cytometric analysis. Surface expression of GPIIb-IIIa and GPIb-IX are modulated upon platelet activation, which must be considered when selecting a platelet identifier, establishing instrument thresholds, and gating platelet populations. Individual laboratories must determine the optimal antibody concentration and incubation time.
3. Fix labeled cells with 10× to 20× total assay volume (e.g., 400 to 800 µl) of 1% formalin fixative. Let fixation proceed 15 min at room temperature, then place tubes at 4°C until analysis. Stability of fixed preparations must be determined by individual laboratories.
4. Analyze by flow cytometry as described above (see Basic Protocol 1, step 4). BASIC PROTOCOL 3
DETERMINATION OF PROCOAGULANT PLATELETS USING ANNEXIN V BINDING OR MONOCLONAL ANTIBODIES SPECIFIC FOR COAGULATION FACTOR V/Va OR X/Xa This procedure may be used to evaluate procoagulant platelets and the ability of platelets to generate procoagulant microparticles in response to an agonist (Furman et al., 2000). Materials Whole blood (WB) containing anticoagulant (see Strategic Planning) or isolated platelets (see Support Protocol) Modified HT buffer (see recipe) containing 5 mM GPRP (see recipe) Coagulation factor V/Va or X/Xa Platelet agonist (see recipe) supplemented with 6 mM CaCl2: e.g., collagen, combined thrombin/collagen mixture, or calcium ionophore A23183 Modified HT buffer Platelet-specific antibody cocktail titrated in modified HT buffer—minimum two antibodies, or one identifier and annexin V—each conjugated to a different fluorochrome: Specific platelet identifier: e.g., anti-CD41, anti-CD61, anti-CD42a, or anti-CD42b Marker of platelet procoagulant activity: e.g., annexin V, monoclonal anti-coagulation factor V/Va or X/Xa 1% formalin fixative (see recipe) 1. Within 1 hr of draw, dilute WB 1:10 in modified HT buffer containing 5 mM GPRP (i.e., 10 µl WB and 90 µl HT/GPRP). If coagulation factor V/Va or X/Xa is to be detected, include these factors in the HT/GPRP diluent. Sources of V/Va or X/Xa include autologous platelet-poor plasma or purified coagulation factors. Each individual laboratory must titrate the optimal concentration of autologous plasma or purified coagulation factor to add back for optimal platelet surface detection.
2. Immediately combine dWB with an equal volume of platelet agonist supplemented with 6 mM CaCl2, or modified HT buffer alone—e.g., 15 µl dWB/GPRP and 15 µl agonist or buffer alone (negative control). Incubate 20 min at 37°C. The optimal incubation time must be determined by individual laboratories.
Immunophenotypic Analysis of Platelets
The final concentration of GPRP is 2.5 mM during the assay incubation; therefore, for a 1:1 (dWB:agonist) mix in the assay, the GPRP concentration in the dWB buffer is 2× or 5 mM. Similarly, HT supplemented with 3 mM Ca2+ (final concentration) used alone or as the agonist diluent is prepared 2× or 6 mM.
6.10.8 Supplement 19
Current Protocols in Cytometry
The negative control for this assay incorporates HT diluent (rather than HT supplemented with CaCl2) in place of agonist. Annexin V and coagulation factors V/Va and X/Xa will not bind in the absence of Ca2+.
3. Label with titrated platelet-specific antibody cocktail (e.g., 30 µl dWB/GPRP/agonist or buffer, and 10 µl antibody reagent cocktail) 20 min at room temperature. As a minimum, two antibody specificities (or one platelet identifier and annexin V) each conjugated to a different fluorochrome (such as fluorescein and phycoerythrin) are used. The fluorescent conjugate of the platelet identifier determines the thresholding (or discriminating) parameter used in the flow cytometric analysis. Surface expression of GPIIb-IIIa and GPIb-IX is modulated upon platelet activation, which must be considered when selecting a platelet identifier. The optimal final concentration of platelet-specific monoclonal antibody reagents must be determined by titration.
4. Fix labeled cells 15 min with 10× to 20× total assay volume (e.g., 400 to 800 µl) of 1% formalin fixative at room temperature. Place tubes at 4°C until analysis. Stability of fixed preparation must be determined by individual laboratories.
5. Analyze by flow cytometry as described (see Basic Protocol 1, step 4). Procoagulant platelets often display dramatic light-scatter changes requiring adjustments to fluorescence and to light scatter gates. Additionally, the distinction between the procoagulant (positive labeling for annexin V, anti-V/Va, or anti-X/Xa) and nonprocoagulant (negative labeling for annexin V, anti-V/Va, or anti-X/Xa) phenotype may be difficult to define on a single-parameter fluorescence histogram; therefore, a two-parameter histogram displaying fluorescence (annexin V, anti-V/Va, or anti-X/Xa) versus forward-angle light scatter may better define the procoagulant platelet population (Fig. 6.10.1, panel D).
PREPARATION OF ISOLATED PLATELETS Some studies may require the isolation of platelets from other cellular and/or plasma components. Platelets may be separated from other cellular components by centrifugation and further isolated from plasma components by gel filtration or washing.
SUPPORT PROTOCOL
Materials Anticoagulated whole blood collected in 5-ml vacutainer tubes or 15-ml conical tubes if drawn by syringe Sepharose 2B beads Modified HT buffer (see recipe) Citrate wash buffer (see recipe) Benchtop centrifuge with rotors for 5-ml Vacutainer tubes and 15-ml conical tubes Polypropylene or siliconized glass test tube 10-ml syringe column 15-ml conical tubes Preparation of platelet-rich plasma 1. Within 30 min of blood draw, prepare platelet-rich plasma (PRP) by centrifuging anticoagulated whole blood collected in 5-ml vacutainer tubes or 15-ml conical tubes if drawn by syringe in a benchtop centrifuge with an appropriate rotor 10 to 15 min at 150 to 200 × g, room temperature. Remove PRP to a clean polypropylene or siliconized glass test tube, being careful not to disturb the buffy coat and red cell layers (also see UNIT 5.1).
Phenotypic Analysis
6.10.9 Current Protocols in Cytometry
Supplement 19
Isolation of platelets by gel filtration: 2a. Pack a 10-ml syringe column with Sepharose 2B beads. 3a. Equilibrate column with 10 vol deionized water, followed by 3 vol modified HT buffer. 4a. Layer PRP (step 1) on the top of the column. 5a. Let PRP fully enter the column and carefully layer with 30 ml modified HT. 6a. Collect gel-filtered platelets (i.e., eluent). Isolation of platelets by centrifugation: 2b. Fill a 15-ml conical tube containing 1 to 8 ml PRP (step 1) with citrate wash buffer. 3b. Centrifuge in a benchtop centrifuge with an appropriate rotor 10 min at 1,200 × g, at room temperature. 4b. Aspirate the supernatant and gently resuspend platelet pellet. 5b. Repeat wash procedure (steps 2b to 4b). 6b. After final wash, resuspend platelet pellet in modified HT. Use isolated platelets as soon as possible after preparation, regardless of preparation technique. Isolated platelet preparations may be substituted for whole blood in any of the preceding basic protocols.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. Pass all reagents through 0.2- to 0.4-µm filters prior to use. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Citrate wash buffer 11 mM glucose 128 mM NaCl 4.3 mM NaH2PO4 7.5 mM Na2HPO4 4.8 mM sodium citrate 2.4 mM citric acid 0.35% (w/v) BSA Adjust pH to 6.5 with 0.1 M NaOH or 0.1 M HCl Store up to 1 year at −20°C. Bring to room temperature and add 1 mg/ml prostaglandin PGE1 (see recipe) to a final concentration of 50 ng/ml immediately prior to use. Formalin fixative, 1%, 2% Dilute 10% (v/v) ultrapure methanol-free formalin (Polysciences) to 1% or 2% (v/v) in HEPES buffered saline (HBS; see recipe). Store up to 1 month at 4°C. Bring to room temperature prior to use. GPRP, 10 mM Dilute GPRP (Gly-Pro-Arg-Pro) in modified HT buffer (see recipe) to a concentration of 10 mM. Store up to 1 week at 4°C or 1 year at −20°C. Bring to room temperature prior to use. Immunophenotypic Analysis of Platelets
Use for diluting thrombin and in assays containing CaCl2 to prevent fibrin polymerization and clot formation (Michelson, 1994).
6.10.10 Supplement 19
Current Protocols in Cytometry
HEPES buffered saline (HBS) 10 mM HEPES 0.15 mM NaCl Adjust pH to 7.4 with 0.1 M NaOH or 0.1 M HCl Store up to 6 months at 4°C Bring to room temperature prior to use Modified HEPES/Tyrode’s (HT) buffer 10 mM HEPES 137 mM NaCl 2.8 mM KCl 1 mM MgCl2 12 mM NaHCO3 0.4 mM Na2HPO4 0.35% (w/v) BSA 5.5 mM glucose Adjust pH to 7.4 with 0.1 M NaOH or 0.1 M HCl Store up to 1 week at 4°C or 1 year at −20°C Bring to room temperature prior to use pH may need readjustment after storage at 4°C.
Prostaglandin PGE1 Prepare 1 mg/ml stock solution in 100% ethanol. Store in small aliquots at −80°C until immediately prior to use. Avoid multiple freeze/thaw cycles. Platelet agonists Prepare all platelet agonist working solutions (concentrations determined empirically) by diluting stocks in modified HT buffer (see recipe), with or without 6 mM CaCl2 as appropriate, just prior to use. Bring to room temperature just before use. Discard left-over working solutions daily. ADP: Adenosine diphosphate (ADP) is typically used at concentrations of 20 µM (maximal platelet activation in diluted whole blood) to submaximal doses of 0.5 µM. Store stock ADP according to manufacturer’s instructions (often frozen at −20°C). Calcium ionophore A23187: Prepare 10 mM stock A23187 in DMSO and store up to 1 year at −80°C. A23187 is typically used at concentrations of 10 to 20 µM (maximal platelet activation in diluted whole blood). Dilute stock at least 1:100 (final DMSO concentration <1%) in modified HT buffer (see recipe) supplemented with 6 mM CaCl2. Use in the evaluation of procoagulant platelets and platelet-derived microparticles. Collagen: Collagen is typically used at concentrations of 10 to 20 µg/ml alone or in combination with submaximal doses of thrombin (see below). Dilute in modified HT buffer (see recipe) supplemented with 6 mM (2×) CaCl2. Use in the evaluation of procoagulant platelets and platelet-derived microparticles in the presence of 2.5 mM (final concentration) GPRP (see recipe). Store stock collagen according to manufacturer’s instructions. Epinephrine: Epinephrine is often used in combination with ADP (see above) at concentrations of 10 to 20 µM (maximal platelet activation in diluted whole blood). Store stock epinephrine according to manufacturer’s instructions. continued
Phenotypic Analysis
6.10.11 Current Protocols in Cytometry
Supplement 19
Human α-thrombin: Thrombin is typically used at concentrations of 1 to 2 U/ml (maximal platelet activation in diluted whole blood) to submaximal doses of 0.1 U/ml. The concentration of thrombin required to achieve maximal platelet activation in plasma-free systems is ∼0.1 U/ml. Dilute working concentration of thrombin in modified HT buffer (see recipe) containing sufficient 10 mM GPRP (see recipe) to give a final assay concentration of 2.5 mM. Store 200 U/ml stock concentration up to 6 months at −80°C or up to one week at 4°C. Thrombin receptor-activating peptide (TRAP/SFLLRN): TRAP is typically used at concentrations of 20 to 50 µM (maximal platelet activation in diluted whole blood) to submaximal doses of 1.5 to 5 µM. Dilute SFLLRN (Ser-Phe-Leu-Leu-Arg-Asp) in modified HT buffer (see recipe). Store up to one week at 4°C or 1 year at −80°C. COMMENTARY Background Information
Immunophenotypic Analysis of Platelets
The use of flow cytometry for the study of platelet function and platelet activation encompasses multiple assays for multiple purposes (Table 6.10.1). While this unit does not describe procedures for all the assays outlined in Table 6.10.1, many facets of platelet function can be evaluated using the protocols described herein. In the absence of an exogenously added platelet agonist (see Basic Protocol 1), the activation state of circulating platelets in vivo, as judged by the binding of an activation-dependent monoclonal antibody or similar reagent, can be determined. Circulating activated platelets have been detected in patients with stable and unstable angina, acute myocardial infarction, acute cerebrovascular ischemia, peripheral arterial occlusive disease, diabetes mellitus, preeclampsia, hemodialysis, systemic inflammatory response syndrome, septic multiple organ dysfunction syndrome, myeloproliferative disorders, and Alzheimer disease. Platelet-derived microparticles are increased in acute coronary syndromes, cardiopulmonary bypass, transient ischemic attacks, and patients with prosthetic heart valves. Platelet hyporeactivity has been reported in very-low-birth-weight preterm neonates and may contribute to the propensity of intraventricular hemorrhage in that patient group. Bernard-Soulier syndrome is an inherited deficiency of the GPIb-IX-V complex. Flow cytometric analysis with GPIb-, GPIX-, and GPV-specific monoclonal antibodies (i.e., antiCD42b, -CD42a, and -CD42d, respectively) provides a rapid and simple means for the diagnosis of the homozygous and heterozygous states of Bernard-Soulier syndrome (Michelson, 1987). Whole-blood flow cytometry allows analysis of platelets without attempting the technically difficult procedure of physically separating the giant Bernard-Soulier syndrome
platelets from similarly sized red and white blood cells. Because light scatter (especially forward light scatter) correlates with platelet size, light-scatter gates may need to be adjusted in the flow cytometric analysis of giant platelet syndromes such as Bernard-Soulier syndrome. This adjustment may result in overlap of the light scatter of giant platelets with red and white blood cells. It is therefore essential to include in the assay a platelet-specific monoclonal antibody as a platelet identifier. For BernardSoulier syndrome platelets, this identifier antibody obviously cannot be GPIb, GPIX, or GPV specific. Glanzmann thrombasthenia is an inherited deficiency of integrin αIIbβ3 (GPIIb-IIIa). Flow cytometric analysis with anti-CD41- and antiCD61-specific monoclonal antibodies provides a rapid and simple means for the diagnosis of the homozygous and heterozygous states of Glanzmann thrombasthenia (Jennings et al., 1986). In addition, a panel of activation-dependent monoclonal antibodies can be used to evaluate patients with defects in platelet aggregation, secretion, or procoagulant activity. The inclusion of an agonist in the assay enables evaluation of the reactivity of circulating platelets in vitro (see Basic Protocol 2). Thus, whole-blood platelet flow cytometry may be used as a physiological assay of platelet function in that a platelet agonist results in a specific functional response (i.e., a change in the surface expression of a physiological receptor or bound ligand) as determined by the binding characteristics of a monoclonal antibody or similar reagent. Activation-dependent monoclonal antibodies Laboratory markers of platelet activation include activation-dependent conformational changes in the GPIIb-IIIa complex (integrin αIIbβ3, CD41/CD61), exposure of granule
6.10.12 Supplement 19
Current Protocols in Cytometry
membrane proteins, platelet surface binding of secreted platelet proteins, and development of a procoagulant surface (Table 6.10.2). The two most widely studied types of activation-dependent monoclonal antibodies are those directed against granule membrane proteins and those directed against conformational changes in GPIIb-IIIa. P-selectin is a component of the α-granule membrane of resting platelets that is expressed on the platelet surface membrane only after α-granule secretion, which occurs during platelet activation (McEver, 2001). Thus, a P-selectin-specific monoclonal antibody binds only to degranulated platelets, not to resting platelets. The activation-dependent increase in platelet surface P-selectin is not reversible over time in vitro; however, in vivo circulating degranulated platelets rapidly lose their surface P-selectin, but continue to circulate and function (Michelson et al., 1996b). Platelet surface P-selectin is therefore not an ideal marker for the detection of circulating degranulated platelets, unless the blood sample is drawn immediately distal to the site of platelet activation, the blood sample is drawn within 5 min of the activating stimulus, or there is continuous activation of platelets. The length of time that other activation-dependent surface markers remain expressed on the platelet surface in vivo has not yet been definitively determined. GPIIb-IIIa is a receptor for fibrinogen and von Willebrand factor that is essential for platelet aggregation. Whereas most monoclonal antibodies directed against GPIIb-IIIa (antiCD41 and -CD61) bind to resting platelets, monoclonal antibody PAC1 is directed against the fibrinogen binding site exposed by a conformational change in GPIIb-IIIa (Shattil et al., 1985). Thus, PAC1 binds only to activated platelets, not to resting platelets. Other GPIIbIIIa-specific activation-dependent monoclonal antibodies are directed against either ligand-induced conformational changes in GPIIb-IIIa— ligand-induced binding sites (LIBS; Frelinger et al., 1988)—or receptor-induced conformational changes in the bound ligand (i.e., fibrinogen)—receptor-induced binding sites (RIBS; Zamarron et al., 1990; Table 6.10.3). Rather than GPIIb-IIIa-specific monoclonal antibodies, fluorescein-conjugated fibrinogen can also be used in flow cytometric assays to detect the activated form of platelet surface GPIIb-IIIa (Faraday et al., 1994), but the concentration of unlabeled plasma fibrinogen and unlabeled fibrinogen released from platelet α granules must also be considered in these assays.
As determined by flow cytometry, in vitro activation of platelets by some agonists (e.g., collagen, collagen/thrombin, C5b-9, calcium ionophore A23187) in the presence of extracellular calcium ions results in platelet-derived microparticles (defined by low forward-angle light scatter and binding of a platelet-specific monoclonal antibody; see Basic Protocol 3) that are procoagulant (determined by binding of monoclonal antibodies to activated factors V, X, or VIII, or by annexin V; Gilbert et al., 1991; Holme et al., 1995; Furman et al., 2000). These findings suggest that procoagulant platelet-derived microparticles may have an important role in the assembly of the “tenase” and “prothrombinase” components of the coagulation system in vivo. A flow cytometric method for the direct detection of procoagulant platelet-derived microparticles in whole blood has been developed (Rajasekhar et al., 1993).
Critical Parameters and Troubleshooting Minimizing platelet aggregates Platelet aggregates can be minimized in the preparation of platelets for whole-blood flow cytometry by following these recommendations (Michelson et al., 2001): 1. Prepare reagents in advance and avoid delays in procedure. 2. Use appropriate blood collection technique (see Strategic Planning). 3. Use polypropylene (or siliconized glass) tubes and syringes. 4. Mix blood immediately with the anticoagulant 5. Do not include washing, centrifugation, gel filtration, vortexing, or stirring steps. 6. Reduce the platelet count by diluting the samples. 7. Include the synthetic tetrapeptide GPRP in the assay if thrombin is the agonist. 8. Mix gently after addition of agonist, then incubate undisturbed. 9. Fix cells. Arg-Gly-Asp (RGD)-containing peptides have also been used to minimize platelet aggregates, but these peptides may interfere with the binding of detecting antibodies (e.g., PAC1) and result in exposure of LIBS. Each sample should be monitored for evidence of platelet aggregation—i.e., “smearing” of the platelets into the upper right quadrant of the log side (orthogonal) light scatter versus log forward light scatter histogram.
Phenotypic Analysis
6.10.13 Current Protocols in Cytometry
Supplement 19
Platelet activation by thrombin, one of the most physiologically important platelet activators, can be directly measured in whole blood through the use of GPRP (Michelson, 1994). In the absence of GPRP, addition of thrombin to whole blood results in a fibrin clot, thereby precluding the use of thrombin as an agonist in the whole-blood assay. Furthermore, thrombin is a potent inducer of platelet-to-platelet aggregation, which precludes analysis by flow cytometry of activation-dependent changes in individual platelets; however, addition to whole blood of GPRP together with thrombin inhibits both fibrin polymerization and, to a lesser extent, platelet-to-platelet aggregation, without affecting thrombin-induced platelet activation. An alternative to the use of thrombin and GPRP in a whole-blood flow cytometric assay is the use of TRAP, a peptide fragment of the protease-activated receptor 1 (PAR1) “tethered ligand” receptor for thrombin (Vu et al., 1991). Without the need for GPRP, TRAP directly activates platelets in whole blood without resulting in a fibrin clot; however, TRAP may not reflect all aspects of thrombin-induced platelet activation, because PAR1 is not the only platelet receptor for thrombin. TRAP is also useful in patient populations where therapeutic heparin (a direct thrombin inhibitor) has been administered. Whole-blood flow cytometric assays frequently employ a GPIb- or GPIX-specific monoclonal antibody to identify platelets. The activation-induced decrease in the platelet surface expression of GPIb-IX generally does not result in fluorescence below the threshold used to distinguish platelets from other cells and debris. Thus, no subpopulations of platelets are excluded. A method of avoiding the activationinduced decrease in binding of a GPIb-specific monoclonal antibody is to add a direct conjugate of the GPIb-specific antibody before addition of the agonist. To specifically analyze the activation-induced decrease in the platelet surface expression of the GPIb-IX complex in whole blood, a CD41- or CD61-specific monoclonal antibody can be employed as the platelet-identifying reagent and the GPIb-IX specific reagent should be added after platelet activation has occurred (i.e., after the addition of agonist).
Immunophenotypic Analysis of Platelets
Lag time Perhaps the single biggest obstacle to the widespread use of flow cytometry in the study of platelet activation is the inconvenience of initiating testing within 30 min of drawing
blood. To circumvent this problem, platelets may be fixed first with 1% formalin prior to labeling with monoclonal antibody reagents. Fix-first methods do not lead to artifactual platelet activation in the authors’ hands. The biggest disadvantages are that some activationdependent epitopes are no longer recognized by their specific monoclonal reagents and that fix-first methods preclude the ability to study platelet responsiveness to agonists. Other elapsed-time considerations are between: fix-first and labeling; fixation after labeling and analysis; and use of PRP or purified platelets instead of whole blood. There is generally a time-dependent decrease in the ability to label fixed platelets. Additionally, the fluorescence intensity after labeling and subsequent fixation generally decreases in a timedependent manner; therefore, the stability of fixed samples prior to flow cytometric analysis must be determined by individual laboratories. Fixation Sample fixation is advantageous in a clinical setting where there may not be immediate access to a flow cytometer. Fixation prevents subsequent artifactual in vitro platelet activation. For most antibodies, the “antibody labeling before fixation” method described above results in no significant differences in fluorescence intensity between samples analyzed immediately and samples analyzed within 24 hr of fixation. A “fixation before antibody labeling” method also results in no significant differences in fluorescence intensity between samples analyzed immediately and samples analyzed within 24 hr of antibody labeling; however, fixation is an important variable to be controlled for, especially in a “fixation before antibody labeling” method, because the binding of activation-dependent monoclonal antibodies to fixed platelets is often decreased compared to that of unfixed platelets (Michelson et al., 1995). Furthermore, the binding of some antibodies further decreases after fixation in a time-dependent manner. The optimal fixation method for each new monoclonal antibody must therefore be defined by each laboratory. A compelling argument in favor of immediate sample fixation is that activation-dependent changes are often time-dependent, at least in vitro. For example, the platelet surface expression of the GPIb-IX-V complex decreases within 30 sec of platelet activation, reaching a nadir at ∼5 min, but over the next ∼45 min the platelet surface expression of the GPIb-IX-V complex returns to normal (Michelson et al.,
6.10.14 Supplement 19
Current Protocols in Cytometry
1996a). The activation-dependent increase in the platelet surface expression of GPIIb-IIIa and CD40 ligand (CD40L, CD154) is also reversible with time (Ruf and Patscheke, 1995; Henn et al., 1998). In contrast, although circulating degranulated platelets rapidly lose their surface P-selectin in vivo, the activation-dependent increase in platelet surface P-selectin is not reversible over time in vitro. Choice of antibodies Because the expression of different antigens reflects different aspects of platelet activation, it may be preferable to use a panel of monoclonal antibodies and/or reagents. Platelet-specific monoclonal antibodies are available from several commercial sources, and can be purchased directly conjugated to fluorescein, biotin, phycoerythrin, PerCP, APC, or tandem conjugates (e.g., phycoerythrin-Cy5 or RED-670). Alternatively, unlabeled antibodies can be FITC conjugated or biotinylated easily using commercially available kits. The use of antibodies that are directly conjugated with fluorochrome eliminates the requirement for the addition of secondary antibodies and/or reagents, thereby avoiding time-consuming additional incubations and washing procedures which, in unfixed samples, may result in artifactual in vitro activation of platelets. Furthermore, the use of secondary antibodies may result in increased background fluorescence and decreased sensitivity of the assay. Finally, the use of directly conjugated antibodies allows multiple color analysis with, for example, a number of differently conjugated murine antibodies. Platelets can be detected in diluted whole blood by light scatter alone using flow cytometry; however, some of the particles with characteristic light scatter for platelets may not be bona fide platelet events. It is therefore recommended that a minimum two-color/twoantibody technique be used for whole-blood platelet flow cytometry, with one monoclonal antibody (e.g., anti-CD41, anti-CD61, antiCD42a, anti-CD42b) to identify a particle as a platelet, and a second (or more) monoclonal antibody or reagent conjugated with alternative fluorochrome (or fluorochromes) to quantitate the expression of the glycoprotein of interest. Laboratories need to confirm that “home-brew” antibody cocktails of multiple antibody specificities do not interfere with the labeling of each individual monoclonal antibody.
FcγRIIa receptor-mediated activation FcγRIIa receptors are constitutively expressed on the platelet surface. Some murine monoclonal antibodies such as those specific for CD9 and, under some conditions, a few murine monoclonal antibodies directed toward GPIIb-IIIa (CD41/CD61) can induce or augment platelet activation via FcγRIIa receptor clustering. Unintentional FcγRIIa receptor– mediated platelet activation can be avoided by using Fab murine monoclonal antibody fragments, using fix-first methods, or by blocking FcγRIIa-mediated platelet activation with the receptor-specific murine monoclonal antibody IV.3 (Medarex).
Anticipated Results Normal healthy donors should exhibit minimal markers of platelet activation (P-selectin expression and PAC1 binding) when whole blood is handled according to the procedures described above. Antibody binding can be expressed as mean fluorescence intensity (MFI) or as the percent of platelets staining positive for a particular antibody (based on a positive analysis region placed just to the right of the negative (usually isotype) control fluorescence histogram). Depending on the experimental circumstances and the physiologic nature of the antigen being measured, either MFI or percent positive platelets may have more relevance than the other. The “percent positive platelets” method may detect subpopulations of platelets arising from a local in vivo insult; however, it is important to recognize that “antibody-positive” platelets may have very little antigen expressed on their surface. For example, in a given clinical setting, the data may be reported as 20% circulating activated platelets, based on P-selectin positivity; however, if each P-selectinpositive platelet expresses only 10% of maximal platelet surface P-selectin, then the overall increase in percent P-selectin-positive platelets may be only 2%. If the goal is to determine the relative increase of a platelet surface antigen, MFI is the preferred method of data presentation. Submaximal doses of platelet agonists are often useful to detect subtle differences in platelet reactivity between different donor groups. Maximal doses of platelet agonist are used to determine maximal potential for platelet activation and receptor expression. Maximally expressed markers of activation serve as a benchmark for comparing platelet activation between different donors and data sets. For example, P-selectin expression may be reported as 100%
Phenotypic Analysis
6.10.15 Current Protocols in Cytometry
Supplement 19
maximal expression rather than as the raw MFI value derived from the flow cytometer. Less than maximal expression may then be reported as a fraction, or percentage of maximal expression.
steps of 20 min each. Each sample can be analyzed in usually <1 min; therefore, the total time for whole-blood flow-cytometric platelet analysis is generally ∼1 hr.
Literature Cited Controls For activation-dependent antibodies, inclusion of a positive control sample (e.g., maximally activated by thrombin, TRAP, or phorbol myristate acetate) assists in the evaluation of surface antigen per platelet and in the comparison of platelet responsiveness between different donors. Positive control samples also ensure that platelet agonists and activation-dependent monoclonal antibodies and/or reagents are functioning properly. Negative controls include the “blank” or isotype control commonly used in many flow cytometric immunophenotyping assays, as well as the “no-agonist” control. The isotype control defines the fluorescent events that are considered negative versus those considered positive. This negative control takes into account cellular autofluorescence, Fc-mediated antibody binding, and nonspecific antibody binding. Defining the negative and positive regions may be determined by a fluorochromematched, concentration-matched, F:P ratiomatched, and isotype-matched antibody to an irrelevant receptor. To determine the negative versus positive regions for PAC1 binding, an appropriate negative control may include an RGD-based peptide in sufficient concentration to completely block specific PAC1 binding. For annexin V, factor V/Va, and/or factor X/Xa binding, omitting the addition of Ca2+ may be used to define negative and positive regions. The “no-agonist” control provides information about technique as well as about donor or patient in vivo platelet activation status, since improper sample handling will lead to artifactual platelet activation. Other controls might include establishing normal ranges for activation markers.
Time Considerations
Immunophenotypic Analysis of Platelets
To minimize delays, prepare working concentrations of monoclonal antibody reagents, platelet agonists, buffers, and fixatives just prior to receiving whole blood for platelet testing. Reagent preparation should take ∼15 min. Platelet activation and labeling with monoclonal antibodies should proceed for ∼20 min and subsequent fixation for ∼15 min. Note that for the determination of procoagulant platelets (see Basic Protocol 3) there are two incubation
Faraday, N., Goldschmidt-Clermont, P., Dise, K., and Bray, P.F. 1994. Quantitation of soluble fibrinogen binding to platelets by fluorescence-activated flow cytometry. J. Lab. Clin. Med. 123:728-740. Frelinger, A.L., Lam, S.C., Plow, E.F., Smith, M.A., Loftus, J.C., and Ginsberg, M.H. 1988. Occupancy of an adhesive glycoprotein receptor modulates expression of an antigenic site involved in cell adhesion. J. Biol. Chem. 263:12397-12402. Furman, M.I., Krueger, L.A., Frelinger, A.L. III, Barnard, M.R., Mascelli, M.A., Nakada, M.T., and Michelson, A.D. 2000. GPIIb-IIIa antagonist-induced reduction in platelet surface factor V/Va binding and phosphatidylserine expression in whole blood. Thromb. Haemost. 84:492-498. Gilbert, G.E., Sims, P.J., Wiedmer, T., Furie, B., Furie, B.C., and Shattil, S.J. 1991. Platelet-derived microparticles express high affinity receptors for factor VIII. J. Biol. Chem. 266:1726117268. Henn, V., Slupsky, J.R., Grafe, M., Anagnostopoulos, I., Forster, R., Muller-Berghaus, G., and Kroczek, R.A. 1998. CD40 ligand on activated platelets triggers an inflammatory reaction of endothelial cells. Nature 391:591-594. Holme, P.A., Brosstad, F., and Solum, N.O. 1995. Platelet-derived microvesicles and activated platelets express factor Xa activity. Blood Coagul. Fibrinolysis 6:302-310. Jennings, L.K., Ashmun, R.A., Wang, W.C., and Dockter, M.E. 1986. Analysis of human platelet glycoproteins IIb-IIIa and Glanzmann’s thrombasthenia in whole blood by flow cytometry. Blood 68:173-179. McEver, R.P. 2001. P-selectin/PSGL-1 and other interactions between platelets, leukocytes, and endothelium. In Platelets (A.D. Michelson, ed.). Academic Press, New York. In press. Michelson, A.D. 1987. Flow cytometric analysis of platelet surface glycoproteins: Phenotypically distinct subpopulations of platelets in children with chronic myeloid leukemia. J. Lab. Clin. Med. 110:346-354. Michelson, A.D. 1994. Platelet activation by thrombin can be directly measured in whole blood through the use of the peptide GPRP and flow cytometry: Methods and clinical studies. Blood Coagul. Fibrinolysis 5:121-131. Michelson, A.D., Barnard, M.R., Benoit, S.E., Mitchell, J., Knowles, C., and Ault, K.A. 1995. Characterization of platelet binding of blind panel mAb. In Leucocyte Typing V (S.F. Schlossman, L. Boumsell, W. Gilks, J.W. Harlan, T. Kishimoto, C. Morimoto, J. Ritz, S. Shaw, R.L. Silverstein, T.A. Springer, T.F. Tedder, and
6.10.16 Supplement 19
Current Protocols in Cytometry
R.F. Todd, eds.) pp. 1207-1210. Oxford University Press, Oxford. Michelson, A.D., Benoit, S.E., Furman, M.I., Barnard, M.R., Nurden, P., and Nurden, A.T. 1996a. The platelet surface expression of glycoprotein V is regulated by two independent mechanisms: Proteolysis and a reversible cytoskeletal-mediated redistribution to the surface-connected canalicular system. Blood 87:1396-1408. Michelson, A.D., Barnard, M.R., Hechtman, H.B., MacGregor, H., Connolly, R.J., Loscalzo, J., and Valeri, C.R. 1996b. In vivo tracking of platelets: Circulating degranulated platelets rapidly lose surface P-selectin but continue to circulate and function. Proc. Natl. Acad. Sci. U.S.A. 93:1187711882. Michelson, A.D., Barnard, M.R., Krueger, L.A., Frelinger, A.L., Furman, M.I. 2001. Flow cytometry. In Platelets (A.D. Michelson, ed.). Academic Press, New York. In press. Rajasekhar, D., Barnard, M.R., Bednarek, F.J., Benoit, S.E., and Michelson, A.D. 1993. Procoagulant activity of platelet-derived microparticles in whole blood: Differences between neonates and adults. Blood 82:63a-163a. Ruf, A. and Patscheke, H. 1995. Flow cytometric detection of activated platelets: Comparison of determining shape change, fibrinogen binding, and P-selectin expression. Semin. Thromb. Hemost. 21:146-151.
Schmitz, G., Rothe, G., Ruf, A., Barlage, S., Tschöpe, D., Clemetson, K.J., Goodall, A.H., Michelson, A.D., Nurden, A.T., and Shankey, T.V., for the European Working Group on Clinical Cell Analysis. 1998. European Working Group on Clinical Cell Analysis: Consensus protocol for the flow cytometric characterisation of platelet function. Thromb. Haemost. 79:885896. Shattil, S.J., Hoxie, J.A., Cunningham, M., and Brass, L.F. 1985. Changes in the platelet membrane glycoprotein IIb-IIIa complex during platelet activation. J. Biol. Chem. 260:1110711114. Vu, T.-K.H., Hung, D.T., Wheaton, V.I., and Coughlin, S.R. 1991. Molecular cloning of a functional thrombin receptor reveals a novel proteolytic mechanism of receptor activation. Cell 64:10571068. Zamarron, C., Ginsberg, M.H., and Plow, E.F. 1990. Monoclonal antibodies specific for a conformationally altered state of fibrinogen. Thromb. Haemost. 64:41-46.
Contributed by Lori A. Krueger, Marc R. Barnard, A.L. Frelinger III, Mark I. Furman, and Alan D. Michelson University of Massachusetts Medical School Worcester, Massachusetts
Phenotypic Analysis
6.10.17 Current Protocols in Cytometry
Supplement 19
Immunophenotypic Analysis of PNH Cells
UNIT 6.11
Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired hematopoietic stem-cell disorder in which somatic mutation of the X-linked phosphatidylinositol glycan complementation class A (pig-a) gene results in a partial or absolute deficiency of all proteins normally linked to the cell membrane by a glycosylphosphatidylinositol (GPI) anchor (Takeda et al., 1993; Bessler et al., 1994). Clonal expansion of this cell population frequently occurs in patients with aplastic or hypoplastic anemia in which normal hematopoiesis has failed. The classical clinical features of this condition are intravascular hemolysis, bone marrow failure, and a thrombotic tendency, though patients show a wide spectrum of clinical presentation (Hillmen et al., 1995; Socie et al., 1996). Diagnosis and screening of patients for this rare condition have improved significantly since the development of a flow cytometry–based method, which in the appropriate clinical setting is specific for PNH (Hall and Rosse, 1996; Richards et al., 2000a; Richards and Hillmen, 2001). The absence or partial expression of GPI-linked antigens is specific for all PNH patients and accounts for many of the clinical features of the disease. Absence of two GPI-linked complement regulatory molecules, CD55 and CD59, from red blood cells in patients with PNH is responsible for the increased sensitivity of patient red blood cells to activated complement and the ensuing bouts of intravascular hemolysis characteristic of the condition. PNH is a rare disease, and patients may go undiagnosed for many months. Consequently, the availability of a diagnostic flow cytometry assay means that in patients with suspected PNH, a definitive diagnosis can be rapidly established, resulting in improved patient management and prognosis. As PNH is a rare condition, it is essential to demonstrate deficiency of at least two GPI-linked antigens from both red blood cells and neutrophils to confidently establish a diagnosis. For neutrophils, this is best achieved by multicolor cytometry, though more complex combinations permit neutrophils and monocytes to be analyzed in a single combination of monoclonal antibodies. For red-cell analysis, single-antibody staining remains the most satisfactory and reliable approach, as multicolor labeling of red blood cells encourages agglutination, particularly in the presence of albumin or other protein support in diluents and washing solutions. Similarly, IgM antibodies should be avoided for red-cell analysis. There are circumstances when both red-cell and neutrophil analysis cannot be achieved— i.e., patients who have severe neutropenia or who have received multiple transfusions for severe anemia or hemolysis. Furthermore, a small but significant proportion of patients have only neutrophil PNH clones (van der Schoot et al., 1990; Hillmen and Richards, 1999). Protocols are provided for erythrocytes (see Basic Protocol 1), neutrophils (see Basic Protocol 2), and monocytes (see Basic Protocol 3) as the main diagnostic cell lineages. In addition, a protocol for simultaneous analysis of neutrophils and monocytes is also provided (see Basic Protocol 4). Due to the longevity of lymphocytes, it is recommended that these cells not be studied for diagnostic purposes; however, in patients who have undergone spontaneous remission of PNH, GPI-deficient lymphocytes can be detected for many years afterwards (Nakakuma et al., 1994; Richards et al., 1998, 2000b).
Phenotypic Analysis Contributed by Stephen J. Richards and Peter Hillmen Current Protocols in Cytometry (2002) 6.11.1-6.11.16 Copyright © 2002 by John Wiley & Sons, Inc.
6.11.1 Supplement 20
BASIC PROTOCOL 1
IMMUNOPHENOTYPIC ANALYSIS OF GPI-LINKED ANTIGEN EXPRESSION ON ERYTHROCYTES For analysis of red blood cells in PNH, the most frequently studied GPI-anchored antigens are CD55 and CD59 (van der Schoot et al., 1990; Alfinito et al., 1996; Navenot et al., 1996; Hall and Rosse, 1996), though CD58 is also GPI-linked. Erythrocytes are labeled with directly conjugated antibodies and as single-color reagents. Red blood cells are identified on bivariate plots of forward scatter (FS) versus side scatter (SS) with detectors set in logarithmic amplification mode. Examination of red blood cells in the untransfused PNH patient provides the clearest definition of type III (complete deficiency), type II (partial deficiency), and type I (normal expression) and is illustrated in Figure 6.11.1. The distributions of these populations show wide variation from patient to patient, and
B 300
120
240
90
I
II
Counts
Counts
A 150
III
60
180 120 60
30 0
I
II III
0 100
101
102 CD59-FITC
103
104
C
100
101
102 CD59-FITC
103
104
D 120 120 100 I
III
80
Counts
Counts
100 II
60
I
II 60
40
40
20
20
III
0
0 100
101
102 CD59-FITC
103
104
E
100
101
102 CD59-FITC
103
104
F 150
150
120 I
90
Counts
Counts
120
II III
60 30
I
90
II III
60 30
0
0 100
Immunophenotypic Analysis of PNH Cells
80
101
102 CD59-FITC
103
104
100
101
102 CD59-FITC
103
104
Figure 6.11.1 Examples of CD59 expression on (A) normal red blood cells (all type I cells) and (B to F) five cases of PNH. (B) Three clearly defined populations of red blood cells: type III (complete deficiency), type II (partial deficiency), and type I (normal expression). (C) Type III and type I cells only. (D) Type II and type I cells. (E and F) Small type III and type II PNH clones, respectively.
6.11.2 Supplement 20
Current Protocols in Cytometry
delineation between the various types is not always clear-cut (Richards et al., 2000a). Red-cell analysis is important in PNH because accurate determination of the distribution of type I, II, and III cells can predict clinical phenotype in that patients with greater than 20% type III (complete deficiency) almost always show clinical evidence of hemolysis (Hillmen and Richards, 1999). Materials <48-hr-old (room temperature) or <7-day-old (4°C) EDTA-anticoagulated peripheral blood sample Washing solution without protein support: phosphate buffered saline (PBS; APPENDIX 2A), FACSFlow, or equivalent isotonic cell washing solution not containing protein support FITC- and PE-conjugated isotype-matched negative controls, appropriately titered
A
B 104
104
Orange fluorescence
R1
SS
103
102
101
100
103
102
101
100 100
101
102
103
104
100
101
FS
102
103
104
Green fluorescence
C
D 104
200
M1
Counts
Counts
160 103
M1
120
M2
102
M2
80 40
101 100
101 102 103 Green fluorescence
Marker Events All 27200 M1 27180 M2 20
% Total 100.00 99.93 0.07
104
0 100
101
102
103
104
Orange fluorescence
Mean 1.61 1.58 39.71
Marker Events All 27200 M1 27177 M2 23
% Total 100.00 99.92 0.08
Mean 2.37 2.35 34.49
Figure 6.11.2 Example analysis of GPI-linked antigen expression on red blood cells. (A) Unstained red blood cells used to optimize FS/SS settings and to define an acquisition region (R1). (B) Fluorescent characteristics of control red blood cells as they are acquired. The position of these control cells acts as a guide to setting markers on plots (C) and (D). Markers M1 and M2 are set to delineate the positive/negative threshold on each histogram and show <0.1% positive events.
Phenotypic Analysis
6.11.3 Current Protocols in Cytometry
Supplement 20
PE-conjugated monoclonal antibodies to CD235a (glycophorin A; Cymbus Biotechnology) and CD55 (DAF; Cymbus Biotechnology) FITC-conjugated monoclonal antibodies to CD59 (MIRL; Cymbus Biotechnology) 96-well round-bottom microtiter plates Microtiter plate mixer 12 × 75–mm polystyrene tubes as required by flow cytometer Flow cytometer with 488-nm argon laser, at least two fluorescence detectors, and appropriate filter sets for detection of FITC and PE Label red blood cells with monoclonal antibodies 1. Prepare a 1:100 dilution of EDTA-stabilized peripheral blood sample by adding 10 µl sample to 1 ml washing solution without protein support and mixing. Blood samples can be up to 48 hr old if stored at room temperature or up to 7 days old if stored at 4°C.
2. For each sample to be tested, use a row of four wells in a 96-well round-bottom microtiter plate. Add 10 µl FITC- and PE-conjugated (combined) isotype control antibodies to well 1, 10 µl CD235a-PE to well 2, 10 µl CD59-FITC to well 3, and 10 µl CD55-PE to well 4. 3. Mix the diluted sample and add 10 µl to each antibody in the microtiter plate, including controls. 4. Mix using a microtiter plate mixer and incubate 60 min at room temperature protected from light, mixing gently every 15 min. 5. Resuspend cells on the microtiter plate mixer, add 150 µl washing solution without protein support to each well, and centrifuge 1 min at 400 × g, room temperature. Discard the supernatant by rapid inversion of the microtiter plate. NOTE: Do not shake the plate.
6. Repeat the washing step. 7. Resuspend the cells in 300 µl washing solution without protein support and transfer to four 12 × 75–mm tubes labeled with mAb designations (i.e., corresponding to wells). Use a 100-µl pipet to break up any red-cell agglutination. Analyze by flow cytometry as soon as possible after antibody staining. Set up flow cytometer and acquire data 8. Load standard instrument settings for two-color analysis of red cells using a flow cytometer with 488-nm argon laser, at least two fluorescence detectors, and appropriate filter sets for detection of FITC and PE. Set FS and SS detectors to log amplification. 9. Create bivariate plots of FS versus SS and green fluorescence (FITC) versus orange fluorescence (PE), as well as one-parameter histograms of green and orange fluorescence (Fig. 6.11.2). 10. Use unlabelled red blood cells (i.e., EDTA-anticoagulated peripheral blood) to optimize light-scatter detector settings on the bivariate plot of FS versus SS. Set an acquisition region R1 around the red blood cell population.
Immunophenotypic Analysis of PNH Cells
11. Run the negative control (tube 1). Acquire and store 25,000 events gated on R1. Run a tube of washing solution for 1 min after each sample tube to prevent carryover of red blood cells into the next tube. 12. Repeat step 11 for CD235a, CD59, and CD55 (tubes 2, 3, and 4, respectively).
6.11.4 Supplement 20
Current Protocols in Cytometry
Analyze listmode data files 13. Create the following five histograms (Fig. 6.11.3): Histogram 1: Histogram 2: Histogram 3: Histogram 4: Histogram 5:
A
Green fluorescence (FITC; tube 1) Orange fluorescence (PE; tube 1) CD235a-PE (Glycophorin A; tube 2) CD59-FITC (tube 3) CD55-PE (tube 4).
250
Counts
200 M1
150
M2
100 50
Marker All M1 M2
Left,Right 1, 9910 1, 10 10, 9910
Events 25000 24949 51
% Total 100.00 99.80 0.20
Geo Mean 1.82 1.80 41.20
Marker All M1 M2
Left,Right 1, 9910 1, 10 10, 9910
Events 25000 24841 160
% Total 100.00 99.36 0.64
Geo Mean 2.63 2.56 188.97
Marker All M1 M2
Left,Right 1, 9910 1, 10 10, 9910
Events 25000 64 24936
% Total 100.00 0.26 99.74
Geo Mean 687.66 3.74 696.93
Marker All III II I
Left,Right 1, 9910 1, 10 10, 99 97, 9910
Events 25000 7955 2300 14759
% Total 100.00 31.82 9.20 59.04
Geo Mean 53.81 2.41 34.05 307.83
Marker All III II I
Left,Right 1, 9910 1, 10 10, 97 97, 9910
Events 25000 7493 2236 15291
% Total 100.00 29.97 8.94 61.16
Geo Mean 84.05 2.66 31.51 526.04
0 100
B
101
103
104
250 200
Counts
102 Mouse IgG1-FITC
M1
150 M2
100 50 0 100
101
102
103
104
Mouse IgG1-PE
C
450
Counts
360 M1
270
M2
180 90 0 100
101
102
103
104
CD235a-PE (Glycophorin A)
D
200
Counts
160
I
III
120
M2
II
80 40 0 100
101
102
103
104
CD59-FITC
E
200
Counts
160
III
120
I II
80 40 0
100
101
102
103
104
CD55-PE
Figure 6.11.3 Example of red-cell analysis. (A and B) Negative isotype controls used to define the position of type III cells. (C) CD235a (Glycophorin A) acts as control both for efficiency of gating and for non-GPI-linked antigen expression. In this case 99.7% of events in the region are CD235a positive, defining the threshold of sensitivity of the technique. (D) PNH patient with clearly defined type II and III cells, with a major component of normal cells. (E) The diagnosis of PNH is confirmed by a similar pattern of expression for the GPI-linked antigen CD55.
Phenotypic Analysis
6.11.5 Current Protocols in Cytometry
Supplement 20
14. Histograms 1 and 2: Recall file 1 (negative control) and use the position of the negatively stained peaks to delineate positive and negative expression. Record percentages of negative and positive cells. 15. Histogram 3: Recall file 2 (CD235a) and use the marker settings from the negative control histogram to determine the percentage of CD235a-positive red blood cells. This tube not only checks the efficiency of gating, but also confirms normal expression of a non-GPI linked antigen by red blood cells.
16. Histogram 4: Recall file 3 (CD59). Set three markers on this histogram: a. Define type III cells (complete deficiency of GPI-antigens) using the marker setting from the negative control. b. Define type I cells (normal expression) from processing a normal control or from the normal red blood cells present in the sample. c. Classify type II cells (partial GPI-deficiency) as those cells that fall between type III and type I cells. Record percentages of type I, II, and III cells. The total PNH clone = percentage type II cells + percentage type III cells. Ideally the fluorescence peak from the normal control should cover the third logarithm fluorescence decade.
17. Histogram 5: Recall file 4 (CD55). For this antigen define type I, type II, and type III cells as established in step 16. Type III cells will be clearly identified and can be defined using the marker settings from the PE-negative control antibody. The percentage of type III cells from the CD59 and CD55 plots should correspond. The PNH cell populations may not be as clearly defined with this antibody, particularly type II cells.
18. Report PNH clone size and distribution of type II and type III cells. BASIC PROTOCOL 2
ANALYSIS OF GPI-LINKED ANTIGEN EXPRESSION ON PERIPHERAL BLOOD NEUTROPHILS There is no single standardized method for immunophenotypic analysis of PNH neutrophils, and virtually all published and individual institution methods differ in their approaches (van der Schoot et al., 1990; Alfinito et al., 1996; Navenot et al., 1996; Hall and Rosse, 1996). Much of the variability reflects personal choice of antibodies and familiarity with antigen expression profiles. Similarly, protocol method—i.e., prestain followed by lysis or lysis followed by staining—is largely dependent upon personal choice, though antigens that are expressed on red blood cells can significantly reduce fluorescent staining intensities on neutrophils if prestaining followed by lysis methods are used. The most important technical considerations are that (1) multicolor analysis (3or 4-color) is used, as this greatly enhances identification of small PNH clones; (2) antibodies are directly conjugated and appropriately titered; and (3) peripheral blood is the material of choice for testing. The complexity of myeloid differentiation in the bone marrow and the maturation-related expression of some GPI-linked antigens such as CD16 and CD66 render interpretation of results very difficult and potentially misleading.
Immunophenotypic Analysis of PNH Cells
In the protocol presented here, a three-color combination of CD66abce-FITC, CD55-PE, and CD16-PE-Cy5 is used in a prestain followed by differential red blood cell lysis with a commercial lysing reagent, FACSLyse. This particular combination allows neutrophil
6.11.6 Supplement 20
Current Protocols in Cytometry
B
A
104
1000
800
R1
CD55-PE
103
SS
600
400
102
101 200
100
0 0
200
400
600
800
1000
0
200
400
D 104
normal cells
CD16-PE-CY5
CD55-PE
103 102 101 100 100
PNH cells
101 102 103 CD66abce-FITC
104
% Gated X Geo mean Y Geo mean 5.73 74.25 84.34 74.43 468.04 392.62 19.66 67.28 6.00 0.18 595.38 4.93
800
1000
E 104
104
103
normal cells
102 101 100 100
PNH cells
101 102 103 CD66abce-FITC
104
% Gated X Geo mean Y Geo mean 0.23 32.65 563.09 72.15 464.39 3212.20 24.83 68.35 75.94 2.79 523.79 26.51
CD16-PE-CY5
C
600 SS
FS
103
normal cells
102 101 100 100
PNH cells
101 102 103 CD55-PE
104
% Gated X Geo mean Y Geo mean 0.07 14.81 597.81 72.26 402.46 3205.08 20.34 6.20 79.05 7.32 115.81 45.85
Figure 6.11.4 Example analysis of GPI-linked antigen expression on PB neutrophils. (A) Neutrophils (region R1) are identified on the basis of physical characteristics on the bivariate plot of FS/SS. (B) Normal and PNH neutrophils clearly separated from the monocytic component with low/intermediate SS characteristics. (C, D, and E) R1-gated data clearly showing normal and PNH neutrophils.
identification by light-scatter characteristics and accurate detection of PNH clones by simultaneous demonstration of deficiency of three GPI-linked antigens. This combination is particularly effective in that normal eosinophils with a CD66+CD16– phenotype can be clearly separated from small neutrophil PNH clones. The antibody combination can be greatly improved by the addition of a fourth-color reagent, CD33-APC, which allows neutrophils and monocytes to be identified in a single combination (see Basic Protocol 4). Materials <24-hr-old EDTA-stabilized peripheral blood sample Washing solution with protein support: FACSFlow containing 0.1% (w/v) BSA, or equivalent isotonic cell washing solution containing protein support FITC-, PE-, and PE-Cy5-conjugated (directly) monoclonal antibodies to CD66abce (Dako), CD55 (Cymbus Biotechnology), and CD16 (Cymbus Biotechnology), appropriately titered Lysing reagent—e.g., FACSLyse (Becton Dickinson), Uti-Lyse (Dako), QuickLysis (Quest Biomedical)
Phenotypic Analysis
6.11.7 Current Protocols in Cytometry
Supplement 20
CellFIX reagent (Becton Dickinson) or 1% (v/v) paraformaldehyde in PBS (APPENDIX 2A) 12 × 75–mm polystyrene tubes suitable for the flow cytometer Flow cytometer with 488-nm argon laser, at least three fluorescence detectors, and appropriate filter sets for detection of FITC, PE, and PE-Cy5 NOTE: PerCp or PerCP-Cy5.5 conjugates can be used in place of PE-Cy5. Additional reagents and equipment for counting leukocytes (APPENDIX 3A) Label neutrophils with monoclonal antibodies 1. Assess the leukocyte count of an EDTA-stabilized peripheral blood sample (APPENDIX 9 3A). In the unlikely event of a leukocyte count >20 × 10 /liter, dilute the sample using washing solution with protein support to obtain a leukocyte count of <20 × 109/liter. Testing should be performed as soon as possible after collection, but no longer than 24 hr.
2. Label a single 12 × 75–mm polystyrene tube with sample reference number. To the bottom of the tube, add 10 µl of each FITC-, PE-, and PE-Cy5-conjugated (directly) monoclonal antibody to CD66abce, CD55, and CD16. 3. Mix the blood sample well and add 100 µl to the tube, pipetting the blood directly onto the antibody combination. Gently mix. 4. Incubate the tube 20 min at room temperature, protected from light, mixing gently every 5 min. 5. Add 2 ml lysing reagent. Vortex gently and incubate 10 min at room temperature, protected from light. 6. Centrifuge 5 min at 400 × g, room temperature. Discard supernatant by rapidly inverting the tube. Remove excess lysing reagent by blotting the tube rim on a paper towel. NOTE: Do not shake the tube.
7. Gently resuspend the cell pellet and add 2 ml washing solution with protein support. Repeat step 6. 8. Add 300 µl CellFIX reagent or 1% paraformaldehyde in PBS. Incubate ≥15 min at room temperature. Analyze immediately on a flow cytometer, or store ≤24 hr at 4°C in the dark before processing. Set up flow cytometer 9. Load standard instrument settings for three-color analysis using a flow cytometer with 488-nm argon laser, at least three fluorescence detectors, and appropriate filter sets for detection of FITC, PE, and PE-Cy5. 10. Create bivariate dot plots of FS versus SS, SS versus CD55, CD66abce versus CD55, CD66abce versus CD16, and CD55 versus CD16 (Fig. 6.11.4). 11. Set the instrument for acquisition and storage of 50,000 ungated events. 12. Run sample in setup mode and set threshold on FS/SS plot to eliminate debris. 13. Acquire data and store as listmode file. Immunophenotypic Analysis of PNH Cells
6.11.8 Supplement 20
Current Protocols in Cytometry
Analyze listmode data files 14. Draw a nonrectangular region R1 around the neutrophil population based on FS/SS characteristics (Fig. 6.11.4A). 15. Apply this region to the CD55/SS plot in multicolor gating mode. The position of R1 can then be adjusted so as not to include any monocytes (Fig. 6.11.4B). 16. Apply this analysis region to the three fluorescence plots CD66 versus CD55 (Fig. 6.11.4C), CD66 versus CD16 (Fig. 6.11.4D), and CD55 versus CD16 (Fig. 6.11.4E). Set quadrant markers on all these plots and record gated quadrant statistics for each plot. 17. Define normal and PNH neutrophils. Residual normal neutrophils are almost always present in PNH samples and serve as an internal positive control. These can be used to define normal expression of GPI-linked antigens. PNH neutrophils are usually clearly separated from the normal component but often show weaker staining of GPI-linked antigens rather than fall into a region that would be regarded as genuinely negative. Panels C and E of Figure 6.11.4 show a PNH clone that is CD55 negative. Though expression of CD66 and CD16 is clearly reduced, it does not appear to be completely negative.
B 104 Monocytes (R1)
104
800
103
SS
CD64-PE
103 102
C
1000
CD64-PE
A
600 400
101
200
400 600 SS
800 1000
0
200 400 600 800 1000 FS
100
101
102
103
104
CD14-FITC
60 50
Counts
100
0 0
D
101
200
100
102
PNH monocytes
40
Normal monocytes
30 20 10 0 100
101
102
103
104
CD14-FITC
Marker All PNH Monocytes Normal Monocytes
Left,Right 1, 9910 1, 106 106, 9910
Events 5623 5513 110
% Gated % Total Geo mean 100.00 11.25 7.63 98.04 11.03 7.11 1.96 0.22 265.27
Figure 6.11.5 PNH monocytes. (A) Monocytes are identified on the basis of CD64 expression and SSlow/intermediate characteristics, region R1. (B) This region is then shown on the FS/SS plot. (C) CD64 expression versus the GPI-linked antigen CD14 for the R1-gated monocytes. (D) Histogram analysis of CD14 expression for this population shows a PNH monocyte population of 98% and a small residual population (2%) of normal monocytes.
Phenotypic Analysis
6.11.9 Current Protocols in Cytometry
Supplement 20
18. In cases of PNH, report the neutrophil PNH clone size. For normal results or negative screens, report that neutrophils express normal levels of GPI-linked antigens and that no PNH clones were detected. BASIC PROTOCOL 3
ANALYSIS OF GPI-LINKED ANTIGEN EXPRESSION ON PERIPHERAL BLOOD MONOCYTES As with neutrophils, there is no single standardized method for immunophenotypic analysis of PNH monocytes, or any clear consensus on which GPI-linked antigens provide the clearest resolution of PNH population. The protocol described employs a CD14/CD64 antibody combination, using CD64 versus SS to identify monocytes and CD14 expression to quantitate the PNH clone. Materials <24-hr-old EDTA-stabilized peripheral blood sample Washing solution with protein support: FACSFlow containing 0.1% (w/v) BSA, or other equivalent isotonic cell-washing solution containing protein support FITC- or PE-conjugated (directly) monoclonal antibodies to CD14 (Dako) and CD64 (Becton Dickinson) Lysing solution—e.g., FACSLyse (Becton Dickinson), Uti-Lyse (Dako), QuickLysis (Quest Biomedical) CellFIX reagent (Becton Dickinson) or 1% paraformaldehyde in PBS 12 × 75–mm polystyrene tubes suitable for the flow cytometer Flow cytometer with 488-nm argon laser, two fluorescence detectors, and appropriate filter sets for detection of FITC and PE. Additional reagents and equipment for counting leukocytes (APPENDIX 3A) Label monocytes with monoclonal antibodies 1. Assess the leukocyte count of an EDTA-stabilized peripheral blood sample (APPENDIX 9 3A). In the unlikely event of a leukocyte count >20 × 10 /liter, dilute the sample using washing solution with protein support to obtain a leukocyte count of <20 × 109/liter. Testing should be performed as soon as possible after collection, but no longer than 24 hr.
2. Label a single 12 × 75–mm polystyrene tube with sample reference number. To the bottom of the tube, add 10 µl of each FITC- or PE-conjugated (directly) monoclonal antibody to CD14 and CD64. 3. Mix the blood sample well and add 100 µl to the tube, pipetting the blood directly onto the antibody combination. Gently mix. 4. Incubate the tubes 20 min at room temperature, protected from light, mixing gently every 5 min. 5. Add 2 ml lysing solution. Vortex gently and incubate 10 min at room temperature, protected from light. 6. Centrifuge 5 min at 400 × g, room temperature. Discard supernatant by rapidly inverting tube. Remove excess lysing reagent by blotting the tube rim on a paper towel. NOTE: Do not shake the tube.
7. Gently resuspend the cell pellet and then add 2 ml washing solution with protein support. Repeat step 6. Immunophenotypic Analysis of PNH Cells
8. Add 300 µl CellFIX reagent or 1% paraformaldehyde in PBS. Incubate ≥15 min at room temperature. Analyze immediately on a flow cytometer, or store ≤24 hr at 4oC in the dark before processing.
6.11.10 Supplement 20
Current Protocols in Cytometry
Set up flow cytometer 9. Load standard instrument settings for two-color analysis using a flow cytometer with 488-nm argon laser, two fluorescence detectors, and appropriate filter sets for detection of FITC and PE. 10. Create bivariate dot plots of FS versus SS, SS versus CD64, and CD14 versus CD64, and a single-histogram plot of CD14 expression. 11. Set the instrument for acquisition and storage of 50,000 ungated events. 12. Run sample in setup mode and set threshold on FS/SS plot to eliminate debris. 13. Acquire data and store as listmode file. Analyze data 14. Draw region around monocytic cells identified by SSlow/intermediate and CD64high expression on the SS versus CD64 bivariate dot plot. Apply this region to a histogram plot of CD14 expression. Then set histogram markers (Fig. 6.11.5D) to discriminate PNH monocytes (CD14–) from normal monocytes (CD14+). SIMULTANEOUS ANALYSIS OF GPI-LINKED ANTIGEN EXPRESSION ON PERIPHERAL BLOOD NEUTROPHILS AND MONOCYTES This is a four-color modification of the procedure for analyzing neutrophils (see Basic Protocol 2) that permits simultaneous analysis of monocytes and neutrophils in a single tube, using CD33 expression combined with SS characteristics. The combination of antibodies used is CD66/CD55/CD16/CD33, although CD45 may be substituted for CD33.
BASIC PROTOCOL 4
Materials <24-hr-old EDTA-stabilized peripheral blood sample Washing solution with protein support: FACSFlow containing 0.1% (w/v) BSA, or other equivalent isotonic cell-washing solution containing protein support FITC-, PE-, and PE-Cy5-conjugated (directly) monoclonal antibodies to CD66abce, CD55 and CD16 APC-conjugated monoclonal antibody to CD33 Lysing reagent (e.g., FACSLyse (Becton Dickinson), Uti-Lyse (Dako), QuickLysis (Quest Biomedical) CellFIX reagent (Becton Dickinson) or 1% paraformaldehyde in PBS 12 × 75–mm polystyrene tubes suitable for flow cytometer Flow cytometer with 488-nm argon laser and 635-nm red diode laser, at least four fluorescence detectors, and appropriate filter sets for detection of FITC, PE, PE-Cy5, and APC. Additional reagents and equipment for counting leukocytes (APPENDIX 3A) NOTE: PerCP or PerCP-Cy-5.5 conjugates can be used in place of PE-Cy5. Label leukocytes with monoclonal antibodies 1. Assess the leukocyte count of an EDTA-stabilized peripheral blood sample (APPENDIX 9 3A). In the unlikely event of a leukocyte count >20 × 10 /liter, dilute the sample using washing solution with protein support to obtain a leukocyte count of <20 × 109/liter. Testing should be performed as soon as possible after collection, but no longer than 24 hr.
Phenotypic Analysis
6.11.11 Current Protocols in Cytometry
Supplement 20
A
C
B
1000
104
800
103
Granulocytes (R1 gated)
1000
600 400 200
R2
600
101
400 200 400 600 800 1000 FS Granulocytes
104
normal granulocytes
102 PNH clone
101 100
200 400 600 800 1000 SS
200
F
Granulocytes
104
CD16-PE-CV5
103
0
E
normal granulocytes
103 102 PNH clone
101 100
100
G
101 102 103 CD66abce-FITC % Gated 5.41 8.36 86.12 0.11
104
400
600 FS
800
Granulocytes
104
CD16-PE-CV5
0
D
normal granulocytes
103 102 PNH clone
101 100
100
101 102 103 CD66abce-FITC % Gated 0.02 8.17 91.52 0.29
104
100
101 102 103 CD55-PE % Gated 0.09 8.19 85.90 5.82
104
H
Monocytes (R3 gated)
40
1000
PNH monocytes
R4
600
Counts
30
800 SS
800
100
0
CD55-PE
R1
102
SS
CD33-APC
SS
R3
101
20
Normal monocytes
10
400
0
200
100
101
0 0
200 400 600 800 1000 FS
Marker All PNH Monocytes Normal Monocytes
102 CD55-PE
Left,Right 1, 9910 1, 449 449, 9910
103
Events % Gated 4393 100.00 3801 86.52 592 13.48
104
Geo mean 14.12 6.97 1308.37
Figure 6.11.6 Example analysis of PNH neutrophils and monocytes. (A) Ungated data with debris removed by increasing the FS threshold. (B) Identification of neutrophils (R1) and monocytes (R3) based on SS characteristics and CD33 expression. (C) FS/SS profile of R1-gated neutrophils. A further region is set to exclude debris and dead/dying cells (R2). This analysis region is then combined with R1 and applied to the three bivariate fluorescence plots of the GPI-linked antigens (D) CD66abce, (D) CD55, and (E) CD16. Normal neutrophils in this example are present in the upper right quadrants of these plots and comprise 8.24% of total (mean value). The PNH clone on the same plots comprises 91.76% of total neutrophils. A small proportion of the PNH cells (5%) express weak CD55 antigen. (G) FS/SS profile of R3-gated monocytes. An additional region R4 is set to exclude debris and dying/dead cells. (H) Analysis of monocytes (region R3 and region R4) shows a PNH clone of 86.5% based on CD55 expression.
2. Label a single 12 × 75–mm polystyrene tube with sample reference number. To the bottom of the tube, add 10 µl of each fluorochrome-conjugated (directly) monoclonal antibody to CD66abce, CD55, CD16, and CD33, and 5 µl APC-conjugated monoclonal antibody to CD33. 3. Mix the blood sample and add 100 µl to the tube, pipetting the blood directly onto the antibody combination. Gently mix. Immunophenotypic Analysis of PNH Cells
4. Incubate the tube 20 min at room temperature, protected from light, mixing gently every 5 min.
6.11.12 Supplement 20
Current Protocols in Cytometry
5. Add 2 ml lysing reagent. Vortex gently and incubate 10 min at room temperature, protected from light. 6. Centrifuge 5 min at 400 × g, room temperature. Discard supernatant by rapidly inverting tube. Remove excess lysing reagent by blotting the tube rim on a paper towel. NOTE: Do not shake the tube.
7. Gently resuspend the cell pellet and then add 2 ml washing solution with protein support. Repeat step 6. 8. Add 300 µl CellFIX reagent or 1% paraformaldehyde in PBS. Incubate ≥15 min at room temperature. Analyze immediately on a flow cytometer, or store ≤24 hr at 4°C in the dark before processing. Set up flow cytometer 9. Load instrument settings for four-color analysis using a flow cytometer with 488-nm argon laser and 635-nm red diode laser, with at least four fluorescence detectors, and appropriate filter sets for detection of FITC, PE, PE-Cy5, and APC. 10. Create bivariate dot blots of FS versus SS, SS versus CD33, CD66abce versus CD55, CD66abce versus CD16, and CD55 versus CD16, and a single-parameter plot of CD55 (Fig. 6.11.6). 11. Set the instrument for acquisition and storage of 50,000 ungated events. 12. Run sample in setup mode and set threshold on FS/SS plot to eliminate debris. 13. Acquire data and store as list mode file. Analyze data for neutrophils 14. On the SS versus CD33 bivariate dot plot, draw a region R1 around neutrophil cells identified by CD33+SSintermediate/high expression. Apply this region to a bivariate dot plot of FS versus SS and set a further region R2 around the neutrophil population. 15. Create an analysis region of (R1 × R2) and apply this to the three bivariate dot plots of CD66abce versus CD55, CD66abce versus CD16, and CD55 versus CD16. 16. Then set quadrant markers to discriminate normal and PNH cells. For more detailed discussion of result interpretation see Basic Protocol 2, step 17. Analyze data for monocytes 17. On the SS versus CD33 bivariate dot plot draw a region R3 around monocytes identified by CD33strong+SSintermediate expression. Apply this region to a bivariate dot plot of FS versus SS and set a further region R4 around the monocyte population. 18. Create a combined analysis region of (R3 × R4) and apply this to a single-parameter histogram of CD55. 19. Then set histogram markers (Fig. 6.11.6H) to discriminate PNH monocytes (CD55–) from normal monocytes (CD55+) in order to determine the size of the PNH monocyte clone.
Phenotypic Analysis
6.11.13 Current Protocols in Cytometry
Supplement 20
COMMENTARY Background Information PNH is a rare form of acquired hemolytic anemia characterized by intravascular hemolysis, bone marrow failure, and a thrombotic tendency. It is a hematopoietic stem-cell disorder; the molecular mutation responsible for the clinical phenotype of the disease occurs within the X-linked pig-a gene, and results in an inability of cells to synthesize GPI anchors that are used to attach numerous proteins to the cell membrane (Takeda et al., 1993; Bessler et al., 1994). As a consequence, all GPI-linked antigens, including the complement regulatory molecules CD55 and CD59, are deficient in PNH cells. Method choice Analysis of GPI-linked antigen expression on red blood cells and peripheral blood neutrophils by flow cytometry provides a rapid, sensitive, and specific method for screening and diagnosis of PNH (van der Schoot et al., 1990; Alfinito et al., 1996; Navenot et al., 1996; Hall and Rosse, 1996). Although requiring expensive equipment and specialist knowledge, flow cytometry has become the “gold standard” for PNH testing. Traditional laboratory methods for screening and diagnosis of PNH are based upon demonstrating an increased susceptibility of patient red blood cells to lysis by complement, activated by acidification of serum (Ham test). There have been numerous modifications to this technique with little increase in sensitivity or specificity. Sensitivity in experienced hands remains around 5% for type III cells and 20% for type II cells (Regan et al., 2001). A gel-based card test system (Diamed) similar to that used in blood grouping has also been developed for PNH screening; the reported sensitivity is similar to that of the Ham test. These laboratory methods have largely been superseded by flow cytometry as the method of choice for screening and diagnosis of PNH. PNH is a rare condition and patients may go undiagnosed for many months; however, the availability of a diagnostic flow cytometry assay means that in patients with suspected PNH, a definitive diagnosis can be rapidly established. This consequently has a considerable impact on patient management and prognosis.
Immunophenotypic Analysis of PNH Cells
General considerations The assays can be undertaken on most benchtop flow cytometers equipped with a 488-
nm argon laser. The four-color method may require an additional red diode laser.
Critical Parameters A fresh (<24 hr old) peripheral blood sample is the most suitable specimen for immunophenotyping for PNH. Ideally, neutrophils should be tested as soon as possible following collection. As samples age, neutrophil viability falls, and by two days nonviable neutrophils may cause significant difficulty in the identification of small PNH clones due to the increased nonspecific uptake of antibodies. For a number of reasons it is important that, whenever possible, both red blood cells and neutrophils be screened. Firstly, in multiply transfused patients red-cell testing is effectively carried out on transfused blood and the proportion of patients’ own cells present may be very small. Secondly, a small proportion of patients have normal red blood cells with only small neutrophil PNH clones (van der Schoot et al., 1990; Hillmen and Richards, 1999). Finally, in patients with severe hypoplastic anemia there may be insufficient numbers of neutrophils for reliable analysis in peripheral blood. Bone marrow aspirates are not recommended for screening for PNH. The spectrum of neutrophil-cell differentiation seen in bone marrow and the maturation-dependent expression of some GPIlinked antigens such as CD16 mean that data analysis can be unnecessarily complex. The antibody combinations and conjugates described in this unit are by no means prescriptive and there clearly is a large number of possible combinations, particularly for multicolor analysis of neutrophils. However, the authors strongly recommend that directly conjugated antibodies be used and that any antibody be appropriately titered for the method used and the cellular target (i.e., neutrophils or red blood cells). The use of isotype-negative controls remains a controversial area. The authors continue to use both positive CD235a (Glycophorin A) and negative controls for red blood cells, primarily to define the position of type III red blood cells, but also to confirm gating efficiency and normal expression of non-GPI linked antigens. They also continue to use single-color analysis for red-cell analysis. Although PNH red blood cells can be identified by multicolor analysis, normal red blood cells can agglutinate and potentially escape analysis
6.11.14 Supplement 20
Current Protocols in Cytometry
regions. This leads to an overestimation of the size of the PNH red-cell clone. For neutrophil and monocyte analysis, the use of controls is less of an issue, as residual populations of normal cells act as internal positive controls. For systematic investigation of suspected PNH cases, it would be useful to perform Basic Protocols 1, 2, and 3, or 1 and 4. This allows red blood cells, neutrophils, and monocytes to be studied, though monocyte analysis is not essential. Although PNH is a stem-cell disorder, analysis of lymphocytes and lymphocyte subsets for the expression of GPI-linked antigens is not recommended for screening purposes and establishing a diagnosis of PNH. The relative longevity of lymphoid cells in comparison to myeloid cells and the effect of aging on production of lymphocytes mean that even in PNH patients with large neutrophil clones, involvement of the lymphoid lineage may be minimal and highly variable (Richards et al., 1998). However, analysis of lymphocytes in patients who have undergone spontaneous remission of PNH has shown that GPI-deficient lymphocytes of both T and B cell lineage can persist for many years following normalization of redcell and neutrophil populations (Nakakuma et al., 1994; Richards et al., 2000b).
Troubleshooting Prior to screening samples, it is essential to establish whether the patient has received a recent blood transfusion. This is important as neutrophil analysis is the only realistic approach to screening heavily pretransfused patients for PNH. When patients are screened for PNH clones, the presence of hypogranular neutrophils in peripheral blood can potentially cause difficulties with gating strategies. These cells can be found in known PNH patients who are developing myelodysplastic syndrome (MDS) or in de novo MDS patients. Because of the altered SS characteristics, hypogranular neutrophils show significant overlap with monocytes; therefore, gating strategies based upon light-scatter characteristics alone will no longer be suitable and a combined lineage marker/SS approach is required. Similarly, neutrophils from aged samples (>24 hr old) with falling viability will show altered FS/SS characteristics that may result in increased nonspecific uptake of antibodies. Another potential pitfall with neutrophil analysis is the presence of immature forms that have significantly weaker expression of CD16. In patients with low neutrophil numbers and relatively high
proportions of eosinophils, separation of normal and GPI-deficient neutrophils and eosinophils may be difficult unless multicolor analysis is used (see Basic Protocol 2 or 4).
Anticipated Results Flow cytometry in PNH is different from many other applications, in that diagnosis depends upon demonstrating absence of relevant antigens. It is therefore important that at least two GPI-linked antigens per cell lineage be studied in order to exclude rare congenital deficiencies of single antigens, particularly CD55 and CD59, and polymorphisms within individual antigens (e.g., CD16) that render them undetectable by some monoclonal antibody clones (Yamashina et al., 1990; Reid et al., 1991). Furthermore, interpretation and reporting of results is dependent on a detailed knowledge of the distribution of GPI-linked antigens and their expression at different stages of hematopoietic cell differentiation. Once a diagnosis of PNH is established on the basis of clinical findings and flow cytometry results, serial monitoring of the peripheral blood PNH clone provides important information (Richards and Hillmen, 2001). Analysis of red blood cells (Fig. 6.11.1) shows typical results for 1 normal and 5 PNH patients. CD59 expression provides the clearest definition of PNH type I, II, and III cells. Moreover, patients with >20% type-III PNH red blood cells will almost always show signs of intravascular hemolysis. Typical results for neutrophil analysis in a PNH patient are shown in Fig. 6.11.4. The GPI-deficient PNH clone is clearly separated from normal neutrophils and shows deficiency of expression of all three GPI-linked antigens studied. If neutrophils and monocytes are studied in the same patient, the PNH component of each cell lineage is almost always identical in size. In general terms the neutrophil PNH clone is often larger then the red-cell PNH clone. This is mainly due to loss of PNH red blood cells from hemolysis and the differences in life span of normal red blood cells, PNH red blood cells, and neutrophils. Patients with neutrophil PNH clones (>50%) have a significant risk of thrombosis, and prophylactic anticoagulation is an important consideration. It is important to emphasize that PNH is a very rare condition and that when no PNH clones are detected in patients with unexplained hemolytic anemia, pancytopenia, or mesenteric/hepatic vein thrombosis, further investigations relevant to the more common causes of these clinical conditions are to be recom-
Phenotypic Analysis
6.11.15 Current Protocols in Cytometry
Supplement 20
mended. The authors frequently find hairy cell leukemia, acute leukemia, or faulty artificial heart valves in patients inappropriately referred for PNH screening.
Reid, M.E., Mallinson, G., Sim, R.B., Poole, J., Pausch, V., Merry, A.H., Liew, Y.W., and Tanner, M.J. 1991. Biochemical studies on red blood cells from a patient with the Inab phenotype (decay-accelerating factor deficiency). Blood 78:3291-3297.
Time Considerations
Richards, S.J. and Hillmen, P. 2001. Advances in the laboratory diagnosis of paroxysmal nocturnal hemoglobinuria. Clin. Applied Immunol. Rev. 1:315-330.
The protocols need to be completed with strict adherence to incubation times. Red blood cells must be run as soon as possible after labeling. For leukocyte analysis, the addition of CellFIX or paraformaldehyde to the antibody-labeled cells means that samples can be stored ≤24 hr at 4°C in the dark before processing. Basic Protocol 1 takes 75 min to complete. Basic Protocol 2, 3, and 4 take 60 min to complete.
Literature Cited Alfinito, F., Del Vecchio, L., Rocco, S., Boccuni, P., Musto, P., and Rotoli, B. 1996. Blood cell flow cytometry in paroxysmal nocturnal hemoglobinuria: A tool for measuring the extent of the PNH clone. Leukemia 10:1326-1330. Bessler, M., Mason, P.J., Hillmen, P., Miyata, T., Yamada, N., Takeda, N., Luzzatto, L., and Kinoshita, T. 1994. Paroxysmal nocturnal hemoglobinuria (PNH) is caused by somatic mutations in the PIG-A gene. E.M.B.O. J. 13:110117. Hall, S.E. and Rosse, W.F. The use of monoclonal antibodies and flow cytometry in the diagnosis of paroxysmal nocturnal hemoglobinuria. Blood 87:5332-5340. Hillmen, P., Lewis, S.M., Bessler, M., Luzzatto, L., and Dacie, J.V. 1995. Natural history of paroxysmal nocturnal hemoglobinuria. N. Engl. J. Med. 333:1253-1258. Hillmen, P. and Richards, S.J. 1999. Flow cytometry in PNH: Serial analysis and the prediction of outcome. Blood 94 :412a. Nakakuma, H., Nagakura, S., Kawaguchi, T., Iwamoto, N., Hidaka, M., Horikawa, K., Kagimoto, T., Tsuruzaki, R., and Takatsuki, K. 1994. Persistence of affected T lymphocytes in longterm clinical remission in paroxysmal nocturnal hemoglobinuria. Blood 84:3925-3928. Navenot, J.M., Bernard, D., Harousseau, J.L., Muller, J.Y., and Blanchard, D. 1996. Expression of glycosyl-phosphotidylinositol-linked glycoproteins in blood cells from paroxysmal nocturnal hemoglobinuria patients. A flow cytometry study using CD55, CD58 and CD59 monoclonal antibodies. Leuk. Lymphoma 21:143-151. Regan, F., Newlands, M., and Bain, B. 2001. Acquired haemolytic anaemias. In Dacie and Lewis Practical Haematology (S.M. Lewis, B.J. Bain, and I. Bates, eds.) pp. 199-229. Harcourt Health Sciences, St. Louis.
Richards, S.J., Norfolk, D.R., Swirsky, D.M., and Hillmen, P. 1998. Lymphocyte subset analysis and glycosylphosphatidylinositol phenotype in patients with paroxysmal nocturnal hemoglobinuria. Blood 92:1799-1806. Richards, S.J., Rawstron, A.C., and Hillmen, P. 2000a. The application of flow cytometry to the diagnosis of paroxysmal nocturnal hemoglobinuria. Cytometry (CCC) 42:223-233. Richards, S.J., Morgan G.J, and Hillmen, P. 2000b. Immunophenotypic analysis of B cells in PNH: Insights into the generation of circulating naive and memory B cells. Blood 96:3522-3528. Socie, G., Mary, J.Y., de Gramont, A., Rio, B., Leporrier, M., Rose, C., Heudier, P., Rochant, H., Cahn, J.Y., and Gluckman, E. 1996. Paroxysmal nocturnal haemoglobinuria: Long term followup and prognostic factors. French society for haematology. Lancet 348:573-577. Takeda, J., Miyata, T., Kawagoe, K., Iida, Y., Endo, Y., Fujita, T., Takahashi, M., Kitani, T., and Kinoshita, T. 1993. Deficiency of the GPI anchor caused by a somatic mutation of the PIG-A gene in paroxysmal nocturnal hemoglobinuria. Cell 73:703-711. van der Schoot, C.E., Huizinga, T.W., van’t VeerKorthof, E.T., Wijmans, R., Pinkster, J., and von dem Borne, A.E. 1990. Deficiency of glycosylphosphatidylinositol-linked membrane glycoproteins of leukocytes in paroxysmal nocturnal hemoglobinuria, description of a new diagnostic cytofluorometric assay. Blood 76:1853-1859. Yamashina, M., Ueda, E., Kinoshita, T., Takami, T.., Ojima, A., Ono, H., Tanaka, H., Kondo, N., Orii, T., Okada, N., Okada, H., Inoue, K., and Kitani, T. 1990. Inherited complete deficiency of 20kilodalton homologous restriction factor (CD59) as a cause of paroxysmal nocturnal hemoglobinuria. N. Engl. J. Med. 323:1184-1189.
Contributed by Stephen J. Richards and Peter Hillmen Haematological Malignancy Diagnostic Service Leeds, United Kingdom
Immunophenotypic Analysis of PNH Cells
6.11.16 Supplement 20
Current Protocols in Cytometry
Quantitative Flow Cytometric Analysis of Membrane Antigen Expression
UNIT 6.12
Cellular phenotyping by use of antibody staining methods (immunophenotyping) is one of the most sensitive and specific phenotyping techniques (see other units in this chapter for more examples). Of the different methods available to perform immunophenotyping, flow cytometry (FCM) offers several unique advantages. FCM allows simultaneous analysis of multiple antigens and analysis of a large number of cells in a relatively short time, and moreover provides the best approach for quantitative estimation of the number of binding sites on cells or other particles. For more than three decades flow cytometers have been used for immunological analysis of the presence or absence of antigens on the cell membrane and/or inside the cell in a qualitative manner; however, over the years it became more and more obvious that quantitation of the number of antigens per cell, or other related measurements, provides unique and useful information that should be considered an intrinsic part of the cellular phenotype process. One answer to this need for quantitative techniques is the so-called quantitative flow cytometry (QFCM). Despite all the variables involved, calibration of instruments for quantitation of fluorescence measurements by FCM should be considered a trivial procedure. The quantitative measurement still is based on the calibration of the fluorescence axis into the number of fluorochrome molecules attached to the cell using Type IIIb standards, or directly into antibody binding capacity (ABC) using Type IIIc standards. Assuming a known fluorochrome/antibody ratio, the Type IIIb calibrator allows the determination of the number of antigens per cells as directly obtained with the Type IIIc standards. For an explanation of bead standard classification, see Schwartz et al. (1998). The protocols described in this unit include those most commonly used and cover the major methods of analysis of the number of antibody binding sites on either cells or particles. Two protocols describe this type of experiment using indirect assays with QIFIKIT (see Basic Protocol) or CELLQUANT (see Alternate Protocol 3) calibrator beads. A third technique provides a direct method using QuantiBRITE PE calibrator beads (see Alternate Protocol 2), while a fourth uses Quantum Simply Cellular calibrator beads for direct or indirect immunofluorescent assays (see Alternate Protocol 1). Two support protocols are provided for titrating antibodies for use with the Quantum Simply Cellular (see Support Protocol 1) or CELLQUANT (see Support Protocol 2) systems. Other assays exist, but either their use is typically restricted to very specific applications (http://cyto.mednet.ucla.edu; see Internet Resources) or they require the use of a “secondary” calibration step (Janossy et al., 1998). QUANTITATIVE DETERMINATION OF CELL SURFACE ANTIGENS BY INDIRECT IMMUNOFLUORESCENCE ASSAY USING QIFIKIT CALIBRATOR BEADS
BASIC PROTOCOL
In this procedure, the cell specimen is labeled with a saturating concentration of either a primary mouse monoclonal antibody (MAb) directed against the antigen of interest or an irrelevant mouse monoclonal antibody (control). Saturation conditions are determined through titration experiments for each MAb investigated, using a fluorescein-conjugated anti-mouse secondary antibody. The primary antibody may be of any mouse IgG isotype. Under these conditions, the number of bound primary antibody molecules corresponds to the number of antigenic sites present on the cell surface. Phenotypic Analysis Contributed by Jean-Luc D’hautcourt Current Protocols in Cytometry (2002) 6.12.1-6.12.22 Copyright © 2002 by John Wiley & Sons, Inc.
6.12.1 Supplement 22
Next, the cell specimen and the setup and calibration beads of the kit are labeled in parallel with a fluorescein-conjugated anti-mouse secondary antibody. The secondary antibody is also used at saturating concentrations. Consequently, fluorescence is correlated with the number of bound primary antibody molecules on the cells and/or the beads. Finally, thanks to the saturating conditions of both primary and secondary antibodies, the amount of fluorescence obtained directly correlates to the number of antigenic sites present on the cell surface. In other words, the cell specimen is analyzed on the flow cytometer and the ABC calculated by interpolation of the calibration curve. Two types of beads are used in this protocol. Setup beads are a mixture of blank and high-level fluorescence beads. Autofluorescence of the beads may be different from that of the cells under analysis; therefore, it might be necessary to adjust the voltage of the fluorescence detector (PMT) to assure that both negative cells and the two populations of beads are displayed simultaneously on the fluorescence scale. Thus, the setup beads are used to establish the fluorescence windows of analysis of the flow cytometer. Calibration beads are a mixture of different beads with well known numbers of antibodybinding sites per bead; fluorescence data corresponding to each of the five bead peaks are used for construction of the calibration curve of mean fluorescence intensity (MFI) against antibody-binding capacity (ABC). Materials Cell suspension: whole blood, cell lines, or isolated cells Unconjugated primary antibody: mouse monoclonal IgG antibody specific for the cell surface antigen under evaluation Unconjugated negative control antibody: irrelevant mouse monoclonal antibody of the same isotype as the unconjugated primary antibody PBS/BSA/azide: PBS (APPENDIX 2A) with 0.5% (w/v) BSA and 0.1% (w/v) NaN3 QIFIKIT (Dako): Setup beads Calibration beads F(ab′) 2 fragment of FITC-conjugated goat anti-mouse (GAM) IgG PBS/BSA: 1× PBS with 0.5% BSA 1× ammonium chloride lysing solution (APPENDIX 2A) or equivalent (for whole blood only) 1% paraformaldehyde in 1× PBS PBS/azide: 1× PBS with 0.1% (w/v) NaN3 Computer with spreadsheet software (e.g., MS Excel) TallyCAL software for Windows: dedicated software for automatic calculation of antigen density (optional) Label cells with primary antibody 1. For each antigen to be tested, prepare two 12 × 75–mm test tubes labeled MAb Specificity and Isotype, respectively. Add 100 µl cell suspension to each tube. If more than one cell surface antigen is to be determined using MAb of the same isotype and concentration, only a single isotype control tube will be necessary.
2. Add 10 µl unconjugated primary antibody to the sample in the MAb Specificity tube. Quantitative Analysis of Membrane Antigen Expression
3. Add 10 µl unconjugated negative control antibody, adjusted to the same concentration as the primary antibody, to the sample in the Isotype (control) tube.
6.12.2 Supplement 22
Current Protocols in Cytometry
4. Incubate 30 to 60 min at 4°C. Incubation time and temperature may depend upon the titration conditions used for staining with the specific monoclonal antibody reagent.
5. Add 3 ml PBS/BSA/azide to each tube. Vortex gently to mix. 6. Centrifuge 5 min at 300 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in each tube. 7. Repeat steps 5 and 6. Label QIFIKIT beads 8. Prepare two test tubes labeled Setup and Cal. 9. Add 100 µl of a homogeneous suspension of setup beads to the Setup tube. Setup beads are a mixture of two different populations of beads: blank beads and beads with high numbers of monoclonal antibody molecules attached to each bead.
10. Add 100 µl of a homogeneous suspension of calibration beads to the Cal tube. Calibration beads are a mixture of five different populations of beads, each population bearing different numbers of monoclonal antibody molecules. Lot-specific information for each population on the exact mean number of monoclonal antibody molecules attached per bead is included with each kit.
11. Add 3 ml PBS/BSA to each tube. Vortex gently to mix. 12. Centrifuge 5 min at 300 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in each tube. From this point on, the same procedure should be followed for both beads and cells.
Stain with secondary antibody to detect antibody binding 13. Dilute an appropriate volume of FITC-conjugated of F(ab′) 2 fragment goat antimouse (GAM) IgG (secondary antibody) 1:50 in 1× PBS. Vortex gently to mix. Each tube to be analyzed will require 100 ìl secondary antibody.
14. To each tube, add 100 µl diluted FITC-conjugated secondary antibody. 15. Incubate 45 min at 4°C in the dark. 16. If the sample is whole blood, lyse erythrocytes by adding 1× ammonium chloride lysing solution to the tube. Incubate 10 min in the dark at room temperature. Alternatively, reagents are commercially available for manual erythrocyte lysis. Follow instructions provided with the reagents.
17. Centrifuge 5 min at 300 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in each tube. If fixation is not required 18a. Add 3 ml PBS/BSA/azide to each tube. Vortex gently to mix. 19a. Centrifuge 5 min at 300 × g, room temperature. 20a. Remove supernatant, leaving ∼50 µl fluid in each tube. 21a. Repeat steps 18a to 20a. 22a. Resuspend the pellet in 500 µl PBS/BSA/azide. 23a. Store tubes in the dark at 4°C no more than 2 hr before analyzing on a flow cytometer.
Phenotypic Analysis
6.12.3 Current Protocols in Cytometry
Supplement 22
If fixation is required 18b. Add 3 ml PBS-azide to each tube. Vortex gently to mix. 19b. Centrifuge 5 min at 300 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in each tube. 20b. Repeat steps 18b and 19b.
0
0
R1
B
Counts 40 80 120 160 200
A
SS
1000
21b. Resuspend pellet in 500 µl of 1% paraformaldehyde in 1× PBS and incubate 2 hr at room temperature or overnight at 4°C. Add 3 ml PBS/azide to each tube and vortex gently to mix.
0
100
200 400 600 800 1000 FS
0
0
103
104
D
M5 M2
M3
M4
M1
100
200 400 600 800 1000 FS
101
102 103 Green fluorescence
104
Marker
Left, Right
Events
% Gated
Mean
Geo Mean
All M1 M2 M3 M4 M5
1, 9910 10, 32 39, 122 137, 426 638, 1894 2054, 5425
30051 3866 4515 5383 6865 8061
100.00 12.86 15.02 17.91 22.84 26.82
1268.78 18.22 69.91 261.36 1084.84 3047.58
403.19 17.75 68.59 256.65 1065.53 3005.25
Counts
SS
M1
M2
0
0
R2
40 80 120 160 200
F
1000
E
0
Quantitative Analysis of Membrane Antigen Expression
102
0
SS
1000
Counts 40 80 120 160 200
C
101
200 400 600 800 1000 FS
100
101
102 103 Green fluorescence
104
Marker
Left, Right
Events
% Gated
Mean
Geo Mean
All M1 M2
1, 9910 1, 6 52, 227
30225 15619 14291
100.00 51.68 47.28
49.85 2.35 102.40
13.75 2.22 100.64
Figure 6.12.1 (A) Gate around the singlet population of setup beads. (B) Corresponding FITCassociated histogram showing both lower- and higher-intensity populations simultaneously. (C) Gate around the singlet population of the calibration beads. (D) Corresponding FITC-associated fluorescence histogram showing the five bead populations. Note that histogram statistics, including geometric means, are displayed. (E) Gate around the cell population of interest. (F) Corresponding histogram plot of FITC-associated fluorescence. Histogram statistics are again displayed.
6.12.4 Supplement 22
Current Protocols in Cytometry
22b. Centrifuge 5 min at 300 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in each tube. Resuspend the pellet in 500 µl PBS/azide. 23b. Store tubes up to 5 days in the dark at 4°C until analyzed on a flow cytometer. It is recommended to check the stability of each of the studied parameters after storage.
Acquire flow cytometry data 24. Create an acquisition document containing one bivariate histogram of forward scatter (FS) versus side scatter (SS) in linear mode and one FITC fluorescence histogram in log mode. 25. Position FS and SS windows of analysis according to standard procedures, allowing a clear gating of the population of bead singlets. 26. Run the Setup tube and gate on the population of bead singlets on the FS/SS plot, collecting at least 10,000 total events or at least 5,000 gated events (Fig. 6.12.1A). 27. Use a histogram plot to show the FITC-associated fluorescence of the population of bead singlets gated from the FS versus SS dot plot (Fig. 6.12.1B). 28. Adjust the voltage of the FITC fluorescence detector to assure that both negative cells and the higher- and lower-intensity populations of setup beads are all displayed simultaneously on the FITC-associated histogram (Fig. 6.12.1B). Keep instrument settings constant for subsequent measurements of calibrator beads and cells. 29. Run the Cal tube and gate on the population of bead singlets on the FS/SS plot, collecting at least 10,000 total events or at least 5,000 gated events (Fig. 6.12.1C). 30. Analyze FITC-associated fluorescence for the population of singlet beads using a histogram plot. Adjust markers around the five bead peaks (Fig. 6.12.1D). Show histogram statistics, making sure that geometric means are displayed. 31. Analyze MAb Specificity and Isotype tubes containing the cells. For cell acquisition, adjust FS and SS windows to clearly display the cells of interest (Fig. 6.12.1E). 32. Analyze the FITC fluorescence for the cell population of interest using a histogram plot. Adjust markers around the positive and/or the negative cell population peaks (Fig. 6.12.1F). Show histogram statistics, making sure that geometric means are displayed. Construct the calibration curve 33. View the histogram statistics of the Cal tube. Either MFI arbitrary units or channel numbers are useful.
34. On a computer with spreadsheet software, enter either the MFI or the channel numbers as the dependent variable y. 35. Enter the lot-specific values for the ABC molecules for each of the five bead peaks as the independent variable x. 36. If MFI is used, calculate log10 for the FITC-associated MFI values and log10 for the ABC levels per bead. 37. Calculate and plot the linear regression line of log10 ABC per bead against log10 MFI or against channel numbers, using the equation y = mx + c, where y equals log10 MFI or channel numbers and x equals log10 ABC per bead. Phenotypic Analysis
6.12.5 Current Protocols in Cytometry
Supplement 22
Calculate the ABC of the cell population under analysis 38. To determine an unknown ABC for a given cell population, introduce the log10 FITC MFI or channel number obtained for those specific cells as y in the linear regression equation and derive the log ABC value. Calculate the anti-log of the value obtained to get the ABC value. 39. Optional. If necessary, correct the ABC of the cell population under analysis for the ABC obtained with the isotype control on the blank bead tube by subtracting the ABC of the Isotype from that of the MAb Specificity tube. The ABC of the Isotype is generally negligible, but in case of low antigen densities it might be significant and therefore should be subtracted from the ABC value obtained for the MAb Specificity tube. ABC values rather than raw values are subtracted because log-transformed data cannot be subtracted directly. ALTERNATE PROTOCOL 1
QUANTITATIVE DETERMINATION OF CELL SURFACE ANTIGENS BY DIRECT OR INDIRECT IMMUNOFLUORESCENCE ASSAY USING QUANTUM SIMPLY CELLULAR CALIBRATOR BEADS Quantum Simply Cellular (Bangs Laboratories) is a mixture of four uniform-size microbead populations. Each bead population is labeled with a different calibrated amount of goat anti-mouse (GAM) IgG specific for the Fc portion of mouse IgG antibodies. The GAM is balanced against IgG1, IgG2a, and IgG2b isotypes and can be used as a universal calibrator for all mouse IgG monoclonal antibodies. The Quantum Simply Cellular beads are stained with the same fluorochrome-conjugated antibody and under the same conditions as the cell samples under study. Each bead population binds varying amounts of the fluorochrome-conjugated monoclonal antibody, producing a corresponding intensity of fluorescence which is analyzed on the flow cytometer. A calibration curve is generated by plotting the mean fluorescence intensity of each bead population versus its assigned antibody binding capacity (ABC). The cell specimen is then analyzed on the flow cytometer using the same instrument settings as for the calibration beads and the ABCs of the cell populations are calculated by interpolation from the calibration curve. Quantum Simply Cellular provides a convenient way to quantitate the ABC of specific cell populations with a flow cytometer. The protocol designed for whole blood samples could be easily adapted for cell lines, or other single cell suspensions. Additional Materials (also see Basic Protocol) Quantum Simply Cellular calibration beads (Bangs Laboratories) 1× PBS (APPENDIX 2A) Fluorochrome-conjugated mouse IgG monoclonal antibody or antibodies (see Support Protocol 1) Computer with spreadsheet software (e.g., MS Excel) QuickCal data diskette to be used with QuickCal v2.1 (optional) QuickCal v2.1 for Windows and Macintosh platforms (optional) Additional reagents and solutions for determining optimal labeling conditions of cells with conjugated mouse IgG (see Support Protocol 1)
Quantitative Analysis of Membrane Antigen Expression
6.12.6 Supplement 22
Current Protocols in Cytometry
Prepare microbeads 1. Label a 12 × 75–mm tube Cal. 2. Add one drop (50 µl) Quantum Simply Cellular calibration beads and one drop (50 µl) 1× PBS to the bottom of the Cal tube. Calibration beads are a mixture of four highly uniform microbead populations of the same size which have varying capacities to bind mouse IgG monoclonal antibodies. Included in the mixture is a population of blank microbeads which has no specific binding capacity for mouse IgG.
3. Add the appropriate amount of conjugated mouse IgG monoclonal antibody recommended for staining 1,000,000 cells. Vortex gently to mix. 4. Incubate 60 min at room temperature in the dark. 5. Add 2 ml PBS to the tube. Vortex gently to mix. 6. Centrifuge 1 min at 540 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in the tube. 7. Repeat steps 5 and 6 twice. 8. Resuspend pellet in 500 µl PBS. The calibration tube is ready for analysis on the flow cytometer. The above protocol can be adapted for indirect staining of Quantum Simply Cellular microbeads using a primary mouse IgG antibody and a secondary fluorochrome-conjugated anti-mouse antibody.
For single-color assay 9a. Label a 12 × 75–mm tube Sample, and add 100 µl cell suspension to the tube. 10a. Add one drop (50 µl) Quantum Simply Cellular microbeads directly to the sample. 11a. Add enough fluorochrome-conjugated mouse IgG monoclonal antibody specific for the cell surface antigen to label both the microbeads and cells (determined by separate titration; see Support Protocol 1). Vortex gently to mix. 12a. Incubate 60 min at room temperature in the dark. 13a. If the sample is whole blood, add 1× ammonium chloride lysing solution to the tube and incubate 10 min in the dark at room temperature. The lysing procedure is unnecessary for specimens other than whole blood (i.e., which are erythrocyte free). Alternatively, reagents are commercially available for manual erythrocyte lysis. Follow the instructions provided with the reagents.
14a. Centrifuge 1 min at 540 × g, room temperature. 15a. Remove supernatant, leaving ∼50 µl fluid in the tube. 16a. Add 2 ml PBS to the tube. Vortex gently to mix. 17a. Centrifuge 1 min at 540 × g, room temperature. 18a. Remove supernatant, leaving ∼50 µl fluid in the tube. 19a. Repeat steps 16a through 18a twice. 20a. Resuspend the pellet in 500 µl PBS. The cells are ready for analysis.
Phenotypic Analysis
6.12.7 Current Protocols in Cytometry
Supplement 22
For multicolor assays 9b. Prepare one tube labeled Cal-MAb for each conjugated monoclonal antibody used to stain the cells. 10b. Place one drop (50 µl) Quantum Simply Cellular microbeads at the bottom of each Cal-MAb tube and add 50 µl PBS. 11b. Add an appropriate amount of each fluorochrome-conjugated mouse IgG monoclonal antibody (MAb) as recommended for 1,000,000 cells to the corresponding tube (see Support Protocol 1). Vortex gently to mix. 12b. In parallel, prepare a test tube labeled Sample. Add 100 µl cell suspension to this tube. 13b. Label the sample with combinations of the above conjugated mouse IgG monoclonal antibodies (MAb) under the same conditions as those used to label Quantum Simply Cellular microbeads. 14b. Incubate 60 min at room temperature in the dark. 15b. If the sample is whole blood, add 1× ammonium chloride lysing solution to the tube. Add 2 ml PBS to the Cal-MAb tubes. Vortex gently to mix. The lysing procedure is unnecessary for specimens other than whole blood (i.e., which are erythrocyte free). Alternatively, reagents are commercially available for manual erythrocyte lysis. Follow the instructions provided with the reagents.
16b. Centrifuge the Cal-MAb tubes and the Sample tube 1 min at 540 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in each tube. 17b. Add 2 ml PBS to each tube. Vortex gently to mix. 18b. Centrifuge 1 min at 540 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in each tube. 19b. Repeat steps 17b and 18b twice. 20b. Resuspend the bead pellets and the cell pellet with 500 µl PBS each. Both the calibration tubes and the sample tube are now ready for analysis on the flow cytometer.
Acquire data 21. Position FS, SS, and fluorescence (FL) windows of analysis according to standard procedures. 22. Create an acquisition document containing at least one bivariate histogram (FS versus SS) and one or more fluorescence histograms. For subsequent measurements of calibrator beads and cells, fluorescence instrument settings must be kept constant.
23. Define a gate containing bead singlets, and excluding debris, on a FS versus SS dot plot and adjust the gate around the population of bead singlets. 24. Using a histogram plot of fluorescence in linear values for the singlet bead population, place and adjust markers around each of the four bead peaks. Quantitative Analysis of Membrane Antigen Expression
25. View the histogram statistics.
6.12.8 Supplement 22
Current Protocols in Cytometry
Construct the calibration curve 26. On a computer with spreadsheet software, enter the histogram statistics of the Cal tube, expressed either as MFI (arbitrary units) or channel numbers, for each fluorochrome-conjugated MAb as the dependent variable y. 27. Enter the lot-specific values for the ABC molecules of the four bead peaks as the independent variable x. 28. If MFI values are used, calculate log10 for the mean fluorescence intensity-associated values and log10 for the mean ABC per bead. 29. Calculate and plot the linear regression line of log10 mean ABC level per bead against log10 MFI or against mean channel number, using the equation y = mx + c, where y equals log10 MFI or mean channel number and x represents log10 ABC value per bead. Calculate mean ABC of the cell population under analysis 30. To determine the mean ABC value for a given cell population, substitute log10 MFI or mean channel number as y in the linear regression equation and derive the log ABC value. 31. Calculate the anti-log of the value obtained to determine ABC. TITRATION OF MONOCLONAL ANTIBODIES TO USE WITH QUANTUM SIMPLY CELLULAR CALIBRATOR BEADS
SUPPORT PROTOCOL 1
The amount of antibody to be used needs to be optimized. This is done by obtaining the maximum separation of the blank and the first labeled population while having a minimal shift of the blank population to higher fluorescence levels. Verification of the saturation equilibrium can be assessed using small aliquots of Quantum Simply Cellular beads and adding increasing amounts of antibody to them—e.g., add additional 50% antibody, until an increase in fluorescence intensity of <10% is obtained after adding the excess antibody. FLOW CYTOMETRIC QUANTITATIVE DETERMINATION OF CELL SURFACE ANTIGENS BY DIRECT IMMUNOFLUORESCENCE ASSAY USING 1:1 PE-LABELED ANTIBODIES AND QUANTIBRITE PE CALIBRATOR BEADS
ALTERNATE PROTOCOL 2
Each QuantiBRITE PE tube contains a lyophilized pellet of four different bead populations. Each population is conjugated with a pre-established well-known mean number of PE molecules attached per bead. QuantiBRITE PE tubes are designed for use with PE-labeled monoclonal antibodies only for the purpose of estimating ABC by flow cytometry. Analyzing QuantiBRITE PE tubes at the same instrument settings and conditions as the assays allows the conversion of the PE-associated fluorescence values into the number of PE molecules bound per cell. By using known ratios of PE conjugated to antibodies, the number of PE molecules per cell can then be converted into the number of antibodies bound per cell. It is believed that the ratio of PE to antibodies is one among several factors which impact the results of quantitation. The use of PE conjugates purified to maximize the content of a one-to-one PE/monoclonal antibody immunoglobin molecule ratio ensures optimal staining and direct reading of ABC values. QuantiBRITE PE tubes are intended for use with any cell suspension provided that the cells are stained with PE-labeled monoclonal antibodies with a known PE/protein ratio. Phenotypic Analysis
6.12.9 Current Protocols in Cytometry
Supplement 22
Additional Materials (also see Basic Protocol) Phycoerythrin (PE) Fluorescence Quantitation kit (BDB): QuantiBRITE PE tubes; store up to the expiration date at 2° to 4°C in manufacturer’s foil pouch PE-conjugated monoclonal antibody with 1:1 PE/protein ratio QuantiQuest v1.0 for Macintosh platforms (optional) Additional reagents and equipment for direct immunostaining (e.g., UNIT 6.2) Prepare calibrator and cell sample 1. Remove the QuantiBRITE PE tube from the foil pouch just prior to use. Reconstitute using 0.5 ml PBS/BSA/azide buffer and vortex.
M3 M4 M2
M1
C
101 102 103 Orange fluorescence
104
Marker
Left, Right
Events
% Gated
Mean
Geo Mean
All M1 M2 M3 M4
1, 9910 8, 26 253, 562 750, 1486 1747, 3338
14791 3545 3685 3738 3584
100.00 23.97 24.91 25.27 24.23
893.32 15.01 366.95 1031.83 2151.82
339.76 14.78 365.01 1025.90 2144.59
D
M2
M1
0
0
R2
100
Counts 40 80 120 160 200
0 200 400 600 800 1000 FS
SS
1000
0
0
Quantitative Analysis of Membrane Antigen Expression
B
0
R1
Counts 40 80 120 160 200
A
SS
1000
The reconstituted pellet is stable 24 hr when protected from light and stored at 2° to 8°C.
200 400 600 800 1000 FS
100
101 102 103 Orange fluorescence
104
Marker
Left, Right
Events
% Gated
Mean
Geo Mean
All M1 M2
1, 9910 1, 5 396, 2267
10798 4309 6334
100.00 39.91 58.66
569.16 1.55 967.28
67.71 1.43 949.81
Figure 6.12.2 (A) Gate around bead singlets. (B) Corresponding PE-associated fluorescence histogram in linear value. Four distinct peaks representing the fluorescence for the four bead populations are shown. (C) Gate around cell population of interest. (D) Corresponding histogram plot of PE-associated fluorescence.
6.12.10 Supplement 22
Current Protocols in Cytometry
QuantiBRITE PE tubes contain a lyophilized pellet of beads conjugated with four levels of phycoerythrin (PE) held in the bottom of the tube by a stainless steel retainer. Each tube is packaged in a foil pouch and must be stored at 2° to 8°C. The number of PE molecules per bead at each level varies among lots; lot-specific information is included with each kit.
2. Add 100 µl of sample to a tube and follow a classical method of direct immunostaining (e.g., lyse-no-wash-format, UNIT 6.2) for the sample, but use PE-conjugated monoclonal antibody with a 1:1 PE/protein ratio for the quantitative analysis of surface antigen expression. Acquire flow cytometric data on the calibrator 3. Create an acquisition document containing at least one bivariate plot (FS versus SS) and one PE-associated fluorescence histogram in linear values. 4. Position FS, SS, and fluorescence (FL) windows of analysis and adjust electronic compensation according to standard procedures for lyse-no-wash immunophenotyping assays. 5. Analyze the QuantiBRITE PE tube, thresholding on FS or SS. If necessary, the FS and/or SS parameter settings can be adjusted to gate more easily on bead singlets. These modifications of FS/SS setting do not alter results of fluorescence quantitation; however, all instrument settings for fluorescence and compensation must be kept constant during subsequent cellular measurements.
Log10 of bead peak geometric mean
4
3
2
y = 1.0125x − 1.6184
1
2
3
4
5
Log10 of no. PE molecules/bead
Figure 6.12.3 Linear regression line (calibration curve) where x equals the log10 of the number or PE molecules per bead and y equals the log10 of the bead peak (PE fluorescence) geometric mean using the equation y = mx + c.
Phenotypic Analysis
6.12.11 Current Protocols in Cytometry
Supplement 22
Table 6.12.1
Calibration Curve Constructiona
Bead population no.
No. PE molecules/bead
Geometric mean of bead peaks
Log10 PE molecules/bead
Log10 geometric mean of bead peaks
1 2 3 4
570 13400 37000 78000
14.78 365.01 1025.9 2144.99
2.7559 4.1271 4.5682 4.8921
1.1697 2.5623 3.0111 3.3313
aSee Figure 6.12.3 for regression line.
6. Define and adjust a gate around bead singlets (Fig. 6.12.2A) on the FS versus SS dot plot and collect at least 10,000 total events or at least 5,000 gated events. 7. Analyze the fluorescence distribution of the singlet population on a histogram plot of PE-associated fluorescence in linear values. Adjust markers around each bead peak (Fig. 6.12.2B). 8. Record histogram statistics (lower part of Fig. 6.12.2B), making sure that geometric means are displayed. Construct the calibration curve Use the following procedure to calculate the mean number of PE molecules per cell for the cell population of interest. 9. On a computer with spreadsheet software, enter the geometric mean value obtained for each of the four bead peaks from the PE-associated fluorescence histogram (Table 6.12.1). 10. For the four bead peaks, enter the lot-specific values corresponding to the mean number of PE molecules per bead. 11. For each of the four bead populations, calculate log10 for both the geometric mean values of PE fluorescence and the mean number of PE molecules per bead (Table 6.12.1). 12. Plot a linear regression of log10 PE molecules per bead (y) against log10 geometric mean fluorescence for the four bead peaks (x), using the equation y = mx + c (Fig. 6.12.3). Acquire sample data and calculate mean number of PE molecules/cell 13. Using the same instrument settings, run the sample and analyze the PE-associated fluorescence for the cell subsets of interest. 14. Substitute log10 geometric MFI of the cell peak for y in the linear regression equation and solve for log10 PE molecules. 15. Calculate the anti-log of the value obtained to get the mean number of PE molecules per cell. Quantitative Analysis of Membrane Antigen Expression
6.12.12 Supplement 22
Current Protocols in Cytometry
QUANTITATIVE EVALUATION OF CELL SURFACE ANTIGEN EXPRESSION ON LEUKOCYTE SUBSETS BY INDIRECT IMMUNOFLUORESCENCE ASSAY USING COUNTER-STAINING REAGENTS AND CELLQUANT CALIBRATOR BEADS
ALTERNATE PROTOCOL 3
A whole-blood specimen or other type of sample is labeled by indirect immunofluorescence with any primary mouse monoclonal antibodies of the IgG isotype specific for the antigen of interest. An aliquot of the same specimen is also labeled with an irrelevant mouse monoclonal antibody of the same IgG isotype and at the same concentration (i.e., a control). A stain no-wash technique is used and the primary monoclonal bound to the cell is detected through a secondary fluorescein-conjugated anti-mouse IgG antibody. After a single wash, a neutralization step allows the counter-staining of the specific leukocyte subsets of interest with specific monoclonal antibody reagents conjugated to a second compatible fluorochrome of a different color (e.g., PE). The primary antibody used to label the cells from the specimen is used at saturating concentrations. Under these conditions, the number of bound primary antibody molecules directly corresponds to the number of antigenic sites present on the cell surface. The secondary antibody is also used at saturating concentrations. Consequently, the overall fluorescence associated with a given cell directly reflects the number of bound primary antibody molecules on its surface. Calibrator beads are stained separately under the same conditions and analyzed on the flow cytometer. The fluorescence data obtained are used for the construction of the calibration curve—(mean fluorescence intensity against antibody binding per cell (AB/C). The stained cell specimen is then analyzed on the flow cytometer using the same instrument settings. The AB/C for the cells of interest is calculated by interpolation from the calibration curve and reflects the number of antigen sites per cell. The CELLQUANT Calibrator kit is intended for quantitation of any antigens on the surface of human leukocytes or leukocyte subsets in whole-blood cell samples, with or without counter-staining. The method could easily be adapted to other cellular suspensions such as cell cultures or separated cells. Additional Materials (also see Basic Protocol) CELLQUANT Calibrator (BioCytex): FITC-conjugated polyclonal anti-mouse IgG 1× dilution buffer: prepare appropriate volume (∼7 ml/tube) by diluting 10× concentrate 1:10 with distilled water; store up to 15 days at 2°C to 8°C Neutralizing solution: purified mouse IgG Calibration beads: Four different populations of beads bearing increasing and precise amounts of Mouse IgG MAb molecules. Lot-specific information on the exact number of MAb is provided with each kit. Unconjugated primary antibody (see Support Protocol 2): mouse monoclonal antibody directed to a cell surface antigen of interest of the IgG1 or IgG2a isotype Isotypic control antibody (see Support Protocol 2): irrelevant mouse monoclonal antibody of the same isotype as the unconjugated primary mouse antibody Monoclonal antibody conjugated to a different fluorochrome (e.g., PE) for counterstaining Additional reagents and equipment for compensation (UNIT 1.14) NOTE: Use all reagents according to manufacturer’s instructions. NOTE: Any fluorochrome compatible with FITC can be employed instead of PE.
Phenotypic Analysis
6.12.13 Current Protocols in Cytometry
Supplement 23
Label with primary antibody 1. Label a test tube Cal. For each antigen to be tested, label two additional test tubes Mab and Ctl, respectively, followed by the sample number (e.g., MAb1 and Ctl1, MAb2 and Ctl2). 2. Add 50 µl homogeneous calibration bead suspension to the Cal tube. 3. Add 10 µl saturating dilution of the unconjugated primary antibody of each specific MAb to the appropriate MAb test tube and the corresponding amount of isotypic control antibody to the associated Ctrl test tube. Add 10 µl diluent buffer to the Cal tube. Saturating dilution must be determined by titration for each antibody.
4. Add 50 µl cell suspension to each MAb and Ctl tube. 5. Mix the tubes by vortexing gently 2 sec. 6. Incubate all tubes 10 min at room temperature in the dark. 7. Add 10 µl FITC-conjugated polyclonal anti-mouse IgG to each MAb, Ctl, and Cal tube. 8. Mix all tubes by vortexing gently 2 sec. 9. Incubate all tubes 10 min at room temperature in the dark. If a single-color procedure is used, go directly from here to step 19.
10. Add 3 ml of 1× dilution buffer to each tube. 11. Mix all tubes by vortexing gently 2 sec. 12. Centrifuge 5 min at 300 × g, room temperature. Remove the supernatant, leaving ∼50 µl fluid in each tube. 13. Add 50 µl neutralizing solution to each tube. Neutralizing solution consists of purified mouse IgG reacts with the free antibody sites of the FITC-conjugated polyclonal anti-mouse IgG used to stain the cells under investigation.
14. Vortex all tubes gently 2 sec. 15. To each MAb and Ctl tube, add the amount of counter-staining PE-labeled monoclonal antibody as recommended by the manufacturer. 16. Vortex all tubes gently 2 sec. 17. Incubate all tubes 10 min at room temperature in the dark. 18. If erythrocytes are present (i.e., whole-blood specimens), add 1× ammonium chloride lysing solution to each tube and incubate 10 min in the dark at room temperature. The lysing procedure is necessary only if red cells are present. Alternate lysing reagents are commercially available for manual erythrocyte lysis.
19. Centrifuge 5 min at 300 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in each tube. 20. Add 3 ml dilution buffer to each tube. Vortex gently to mix. Quantitative Analysis of Membrane Antigen Expression
21. Centrifuge 5 min at 300 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in each tube.
6.12.14 Supplement 23
Current Protocols in Cytometry
22. Resuspend pellet in 500 µl dilution buffer. 23. Store tubes in the dark ≤2 days at 2° to 8°C before analyzing on a flow cytometer. It is recommended to check the stability of each of the studied parameters after storage.
Set fluorescence compensation 24. Create an acquisition document containing at least one bivariate plot (FS versus SS) in linear mode and one bivariate plot of FITC versus PE fluorescence in log mode. Position FS and SS windows of analysis according to standard procedures. 25. For optimum analysis, adjust the FITC fluorescence detector (PMT) such that the bead population with the higher fluorescence intensity is positioned at the beginning of the fourth log decade. 26. Set the PE fluorescence detector (PMT) according to the standard procedure. 27. On the bivariate plot of FITC versus PE fluorescence, set a first gate around the bead population displaying the lower FITC-associated intensity and a second gate around the bead population showing the higher FITC-associated intensity, then adjust compensation for PE-associated fluorescence until the mean PE fluorescence intensity of the two gated populations becomes equal (UNIT 1.14). 28. Adjust compensation of FITC fluorescence in the same manner by analyzing the sample stained with the PE-conjugated counter-staining antibody. Set regions around the PE-positive and-negative cell populations and adjust compensation for the FITC-associated fluorescence until the MFI for the two populations becomes equal. Keep all instrument setting for fluorescence and compensation constant for subsequent measurement of Cal tubes and cellular samples.
Acquire calibration data 29. Define and set a gate around the bead singlets on the FS versus SS dot plot. To help in visualization of the population of bead singlets, FS and/or SS gains can be adjusted without altering the fluorescence measurements.
30. Condition the fluorescence acquisition on the bead gate defined in the FS versus SS dot plot. 31. Analyze the Cal tube, gating the population of bead singlets on FS or SS, and collect at least 10,000 total events or at least 5,000 gated events. 32. Analyze the fluorescence of the calibration bead populations using a histogram plot of FITC fluorescence. Adjust markers around each of the four bead peaks. 33. Show histogram statistics for each bead population. Acquire cellular data 34. Gate the cell population of interest using the appropriate gating strategy. To help in gating, FS and/or SS gains can be adjusted without altering the fluorescence measurements.
35. Run MAb and Ctl tubes. 36. Analyze the cell population of interest using a histogram plot of FITC-associated fluorescence, adjusting markers around the positive and/or the negative peaks. 37. Show histogram statistics for each cell population. Phenotypic Analysis
6.12.15 Current Protocols in Cytometry
Supplement 22
Construct calibration curve 38. On a computer with spreadsheet software, enter the MFI values or the mean channel numbers obtained for FITC-associated fluorescence for each of the four bead peaks as the dependent variable y. 39. Enter the lot-specific values for the AB/C for each of the four bead peaks as the independent variable x. 40. If MFI values are used, calculate log10 for the FITC-associated MFI and log10 for the AB/C obtained for each bead population. 41. Calculate and plot the linear regression line of log10 mean AB/C against log10 MFI or against mean channel numbers, using the equation y = mx + c, where y equals log10 MFI or mean channel numbers and x equals log10 AB/C. Calculate the mean AB/C for the cell population of interest 42. Substitute log10 MFI values or mean channel number for this population for y in the linear regression equation and derive the log AB/C valve. 43. Calculate the anti-log of the value obtained to determine AB/C. SUPPORT PROTOCOL 2
SELECTION AND TITRATION OF MONOCLONAL ANTIBODY REAGENT FOR INDIRECT IMMUNOFLUORESCENCE STAINING OF CELL SURFACE ANTIGENS The primary mouse monoclonal antibody of the IgG1 or IgG2a isotype must be used at saturating concentration. Ideally, this concentration will be below 5 µg/ml (final). Additional Materials (also see Alternate Protocol 3) Unconjugated primary antibody Cellular sample containing both negative and positive cells for the antigen detected by the unconjugated primary antibody FITC-conjugated GAM antibody 1. Prepare a series of 1:1 serial dilutions (i.e., 1:1, 1:2, 1:4, 1:8, etc.) of each unconjugated primary antibody. 2. Incubate dilutions 10 min with 50 to 100 µl of the sample containing cells both negative and positive for the antigen detected by the unconjugated primary antibody. The concentration of sample used must be similar to that in the experiment for which the titration is being performed.
3. Add 3 ml of 1× dilution buffer and vortex gently to mix. 4. Centrifuge 5 min at 300 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in each tube. 5. Add 10 µl FITC-conjugated GAM antibody to each tube. 6. Add 3 ml dilution buffer and vortex gently to mix. 7. Centrifuge 5 min at 300 × g, room temperature. Remove supernatant, leaving ∼50 µl fluid in each tube. Quantitative Analysis of Membrane Antigen Expression
8. Resuspend the pellet in 500 µl dilution buffer and analyze on the flow cytometer. 9. Use the minimal saturating concentration as determined on the fluorescence versus monoclonal antibody concentration curve.
6.12.16 Supplement 22
Current Protocols in Cytometry
COMMENTARY Background Information Since the first observations under the fluorescence microscope of cells stained with fluorescently labeled antibody molecules, it has been recognized that variations in fluorescence intensity reflect intrinsic biological characteristics of those cells. Based on intensity, fluorescence expression was initially classified as “dim” or “bright.” The need for objective interpretation of the results and for standardization of both staining and analytical methods has led to the development of more precise quantitative assays. In the early eighties, several flow cytometric methods were developed to fulfill this goal. Titus and coworkers (1981) initially described a microfluorometric method for quantitating the number of Fc receptors on individual cells. Later this method was extended to determine the number of any type of monoclonal antibody molecule bound to cells (Poncelet and Carayon, 1985). Using direct radiobinding assays, Poncelet and coworkers demonstrated that the mean fluorescence intensity of a given cell population stained for a specific monoclonal antibody is linearly related to the mean number of monoclonal antibody molecules bound per cell. Such a linear relationship between the two techniques allows the use of simple calibration curves to determine the mean number of monoclonal antibody molecules bound per cell in a given cell population. Thus, the quantitative indirect immunofluorescence (QIFI) method was born. From that point on, several new methods were developed and extended to the clinical area. Quantitation of several cell surface molecules on both normal and pathologic hematopoietic cells (Lenkei and Anderson, 1995; Bikoue et al., 1996; Poncelet et al., 1996) has been shown to provide useful information. A recent exhaustive review (Poncelet et al., 2000) illustrates the potential applications of antigen quantitation with special emphasis on the clinical applications of QFCM. For example, at present it is well established that in hematological malignancies, very few, if any, markers are specific to neoplastic cells; however, antigen over- and underexpression are frequently found. Underexpression of CD45 has been claimed to be a useful marker for the recognition of acute leukemia blast cells (Lacombe et al., 1997); at the same time it provides prognostic information in acute lymphoblastic leukemia (ALL; Borowitz et al., 1997). In a high
proportion of precursor-B ALL patients, leukemic cells overexpress CD10, this feature being used for the detection of minimal residual disease. Recently, quantitative cytometry of blast cells using multiple staining with CD38, CD34, CD13, and CD10 has proved useful for the rapid identification of BCR/ABL gene rearrangements in adults with precursor B-ALL (Tabernero et al., 2001). Chronic lymphocytic leukemia (CLL) of the B cell type can be distinguished from other B cell chronic lymphoproliferative malignancies as well as from normal B cells by quantitation of CD5, CD22, and surface immunoglobulin expression (Matutes et al., 1994). In addition, other rare diseases are easily and precisely diagnosed by quantitation of surface molecules. In Glanzmann’s thrombasthenia, quantitation of CD41 and CD61 expression permits detection and classification of the disease (George et al., 1990; UNIT 6.10). Similarly, paroxysmal nocturnal hemoglobinuria (PNH), a rare acquired clonal disorder, is detected and monitored by the quantitative evaluation of CD55 and/or CD59 expression on red cells, platelets, and/or leukocytes (Richards et al., 2000; UNIT 6.11). In some infectious diseases, particularly in human immunodeficiency virus (HIV)-infected individuals, the level of expression of CD38 molecules by CD8+ lymphocytes has been shown to be a powerful factor for monitoring disease activity and the development of AIDS (Autran et al., 1997). Quantitative evaluation of other markers has also been useful for the assessment of the activation status of different types of cells; however, despite the amount of information accumulated through the estimation of the number of cellular binding sites for their respective fluorescent probes, quantitative cytometry remains largely under-exploited for a number of different reasons. A general belief exists that until now, most of the studies using quantitative cytometry have been performed in a single center and no clear application of QFCM has proved relevant enough to impose the need for multicentric studies (Poncelet et al., 2000). Obviously, there are also many important technical issues, particularly the use of highly different staining methods, data-analysis protocols, calibrators (Serke et al., 1998), instrument setup and standardization procedures (Lenkei et al., 1998), or even different units to quantitate fluorescence
Phenotypic Analysis
6.12.17 Current Protocols in Cytometry
Supplement 22
intensity. In addition, steric hindrance and quenching (D’hautcourt and Isaac, 1999) should be considered a major problem in multicolor analysis, and compensation between fluorescent emissions of different fluorochromes is an added problem in this case. Finally, the choice of the monoclonal antibody clone used for quantitation of its self-associated binding sites (Poncelet et al., 2000) and both binding properties and saturating conditions of this clone are also now recognized as factors that need to be standardized. Other specific methodological, instrumental, or biological limitations (Davis et al., 1998; Gratama et al., 1998) have been identified as potential sources of variability. Improvement of standardization procedures is also needed.
Quantitative Analysis of Membrane Antigen Expression
Terminology Efforts to improve methodology often begin with definition of precise terminology and standardization of nomenclature. Such key work was started more than twenty years ago and pursued more recently by the production of a National Committee for Clinical Laboratory Standards document on standard terminology (NCCLS, 1998). Of course, as an evolving technique, quantitative fluorescence cytometry is still using unique terms as well as several different terms for a unique concept. Recently, Henderson and coworkers (1998) proposed new guidelines following the format required for NCCLS documents for quantitative flow cytometry. Some of the most important key words with respect to quantitative fluorescence measurements are defined below. More details are available in the original manuscript. Quantitative fluorescence cytometry (QFCM). The calibrated measurement of fluorescence intensity from labeled particles. QFCM is also used for Quantitative Flow Cytometry, but the first definition is more precise, because calibration refers to fluorescence only. Fluorescence intensity. The portion of the total radiant flux (W) emitted by excited fluorochromes that is detected by the cytometer or other detector. Molecules of equivalent soluble fluorochrome (MESF). Stoichiometric units for the fluorescence intensity of fluorochrome-labeled particles expressed as the number of fluorochrome molecules in solution required to produce the same fluorescence intensity as that measured on the labeled particles. Antibody binding capacity (ABC). The number of antibody molecules bound by a particle when specific binding sites are saturated. The
ABC acronym has also been used for antibody bound per cell. In this case AB/C has been used to emphasize that the “C” is for cell, not capacity. Relative fluorescence intensity (RFI). Expression of a fluorescence intensity relatively to another fluorescence intensity measured under identical conditions. RFI is also used to convert histogram channel distribution from logarithmically amplified fluorescence measurements into values that can be easily compared for relative fluorescence intensity. Comparison of measurement expressed in this way needs caution, and RFI set by manufacturers should not be used for primary calibration.
Critical Parameters and Troubleshooting Fluorescence units Measurements of cell antigen–associated fluorescence may be expressed in different units depending on the software used. Most software from Becton Dickinson Biosciences displays fluorescence intensity on a scale of 1 to 10,000 relative linear values of log data arbitrary units (a.u.), whereas the major software from Beckman Coulter expresses fluorescence intensity in the range of 0.1 to 1000 relative linear values of log data arbitrary units (a.u.). Other software, and in general all flow cytometry software, provides MFI as expressed by the mean channel number (Ch) using a nominal scale of 0 to 255, 0 to 1023, or 0 to 2047. All these scale units can be mathematically interconverted (Schwartz et al., 1996). Qifikit or Cellquant calibrators allow, for the measurement under study, translation of this measured a.u. into antibody binding capacity (ABC) or antibody molecules bound per cell (AB/C). The Quantum Simply Cellular system uses a different principle and a restriction on the interpretation due to the different way that the antibody reacts with either the sample or the calibrator. Accordingly, once a cell sample is stained, the fluorochrome-conjugated antibody reacts with its corresponding antigen through the “ab” fragment. In contrast, the Quantum Simply Cellular calibrator, once incubated with the fluorochrome-conjugated antibody, is “captured” through its Fc fragment by the ab fragment of the goat anti-mouse (GAM) immunoglobulin attached to the bead surface. For QuantiBRITE, the primary translation of the fluorescence arbitrary units is in the number of fluorochrome molecules per cells.
6.12.18 Supplement 22
Current Protocols in Cytometry
deviation are the most commonly used for normal or Gaussian distribution. In biology, for some or many body constituents the distribution is log-normal rather than normal. In flow cytometry, distribution of antigen expression on a given cell population for many surface proteins follows the log-normal curve and by chance is “normalized” by the log amplification. Statistical calculation of the mean and standard deviation is robust enough to derive
When the ratio of fluorochrome molecules to antibody molecules (or preferably the apparent ratio) for the particular antibody reagent used is known, the number of fluorochrome molecules per cell obtained can be further translated into either ABC or AB/C as well. Statistical parameters One of the features of parametric statistics is the ability to describe a distribution with only a few numerical parameters. Mean and standard
1000
A
SS
R1
0
R2 0
1000
FS
100
103 102
Green fluorescence
R3
100
101
103 101
102
R3
100
Green fluorescence
104
D 104
B
101 102 103 Orange fluorescence
104
C
100
101 102 103 Orange fluorescence
104
101 102 103 Orange fluorescence
104
100
0
0
40
40
Counts 80 120 160 200
Counts 80 120 160 200
E
101 102 103 Orange fluorescence
104
100
Figure 6.12.4 Effect of compensation on quantitation. (A) FS/SS containing the calibration beads (R1) and the selected population (R2) of a sample stained with CD3-FITC and CD4-PE. (B) Correct compensation. CD3+CD4– and CD3+CD4+ cells are in the R3 region (black). Lymphocytes and monocytes CD4dim are with the calibrite beads (gray) along the x axis. (C) PE histogram with the four calibrite peaks (gray), CD3–CD4– and CD3–CD4dim+ cells (black), and CD3+CD4– and CD3+CD4+ cells from R3 (filled). (D and E) The same data set as panels B and C with under-compensation of PE. Only the FITC-positive cells are affected. As the calibration beads are not stained with FITC they are not affected.
Phenotypic Analysis
6.12.19 Current Protocols in Cytometry
Supplement 22
informative parameters on a distribution measured in log mode. The problem arises with interpretation of the data. Unfortunately, the mean value of a fluorescent cell population in the real world is not simply a mathematical transformation of the mean directly calculated from the histogram distribution. The various software programs currently available use different units and different methods of calculation. In any case, including those with very complex histogram distributions in which a true picture is not given by arithmetic or geometric mean, the use of median values allows the calculation of the median of the sample population as well as those of the calibrator populations. Additionally, the median value can be mathematically transformed without changing its significance; therefore, it is the most robust statistical tool for multi-site, multi-platform comparison of quantitative antigen expression measurements.
Quantitative Analysis of Membrane Antigen Expression
Calibration curve: linear regression Regression is a technique for the calculation of an equation describing the relationship among scattered data points. When data fit a straight-line equation it is called linear regression. Linear regression is the most commonly used regression technique in the clinical laboratory. It not only characterizes the relationships among different values, but also allows prediction of future data points. The most popular method used for linear regression calculation is the least-square method. The mathematics of linear equations is so simple to use that in many cases nonlinear functions are converted into linear functions prior to the regression calculation. In the clinical laboratory one often finds exponential curves or power curves. These are transformed to linear expressions by conversion to logarithms. By convention, x represents the independent variable while y refers to the dependent variable or variables. The typical equation of linear regression is y = mx + b. In order to make the plot appear similar to routine flow cytometric histograms, an x,y switch for graphical representations of the calibration plot has been proposed (Schwartz et al., 1996). This system uses the x axis for instrument response and the y axis for plotting fluorescence-associated units per calibrator bead. Nevertheless, linear regression suffers from some pitfalls related to range, frequency, or imprecision of x values. Different weighted regression techniques exist to compensate for these limitations. As a general rule, the ratio of
the highest value to the lowest value needs to be higher than three to provide reliability. A few isolated low or high values added to values concentrated near the center position of the distribution do not solve this problem. An easy way to avoid the problem linked to extreme values is to limit the use of calibration points to three standard deviations around the mean. Moreover, because the extremes of the regression line are not so certain as the middle points, caution must be taken for prediction of y values much beyond the lowest or the highest calibration values. Like the arithmetic mean, the regression equation without an estimate of the variance gives information which is useful but incomplete (Weisbrot, 1979). This variability is estimated by:
S xy =
∑ ( y − yc ) 2 n−2
where y is the observed dependent variable and yc the calculated value, n is the number of points used to calculate the regression line, and Sxy describes the dispersion of data points around the fitted regression line. This indicates that the 95% confidence interval of any calculated new value of y could be ±2Sxy. Effect on compensation All the technical aspects linked to the antigen-antibody reaction, the measurement of fluorescence emission and statistical analysis, and the quantitative measurement of antigen expression further increase in complexity because of the compensation for fluorescence overlap in multicolor measurements. Although it has been shown that the proportionality of fluorescence is kept for any compensation factor, once quantitative measurements are done it becomes impossible to measure the absolute fluorescence in multicolor analysis accurately without a perfect compensation setting. In fact, fluorescence reaching a detector is the overall sum of the fluorescence directly related to the specific fluorochrome to be measured in this detector and a fraction of the fluorescence of the other fluorochromes. This fraction is actually the one for which compensation corrects. In the case of multicolor analysis, as for single-color, calibration particles are always singly stained particles and the added fluorescence from another fluorochrome could be considered as negligible; however, in a stained sample, the goal of a second monoclonal anti-
6.12.20 Supplement 22
Current Protocols in Cytometry
body conjugated to another fluorochrome is to help in the selection of the population of interest. In this case a fraction of the intensity of this second marker is added to the fluorescence of the population for the first fluorochrome-conjugated reagent fluorescence emission. If not correctly compensated, this can provide erroneous values during the measurement (Fig. 6.12.4).
Time Considerations In quantitative flow cytometry, the method used for fluorescence calibration is the most significant factor affecting the duration of the whole procedure. The shortest time is obtained by use of Type IIIb calibrators and specific fluorochrome-conjugated monoclonal antibodies with known fluorochrome/antibody ratio. In this case the fluorescence histogram is directly converted into number of fluorochrome molecules or into AB/C units, and the duration equals that of the classical immunofluorescence analysis. For the other procedures, using Type IIIb calibrators, there is a large time saving if the direct technique is used instead of the indirect one. The number of samples to be tested do not significantly increase the total time until very large series of samples are analyzed. The time required for specific protocols (assuming standard numbers of samples) are as follows: indirect immunofluorescence using QIFIKIT (see Basic Protocol) takes ∼2 hr, direct or indirect immunofluorescence using Quantum Simply Cellular calibrator beads (see Alternate Protocol 1) requires ∼1.5 hr; direct immunofluorescence using PE labeled-antibodies and QuantiBRITE PE calibrator beads (see Alternate Protocol 2) takes ∼30 min, and indirect immunofluorescence using counterstaining reagents and CELLQUANT calibrator beads, ∼1 hr.
Borowitz, M.J., Shuster, J., Carroll, A.J., Nash, M., Look, A.T., Camitta, B., Mahoney, D., Lauer, S.J., and Pullen, D.J. 1997. Prognostic significance of fluorescence intensity of surface marker expression in childhood B-precursor acute lymphoblastic leukemia. A pediatric oncology group study. Blood 89:3960-3966. Davis, A.K., Abrams, B., Iyer, S.B., Hoffman R.A., and Bishop, J.E. 1998. Determination of CD4 antigen density on cells: Role of antibody valency, avidity, clones and conjugation. Cytometry 33:197-205. D’hautcourt, J.-L. and Isaac, J. 1999. Mean fluorescence intensity of dual stained cells. Cytometry 38:44-45. George, J.N., Caen, J.P., and Nurden, A.T. 1990. Glanzmann’s thrombasthenia, the spectrum of clinical disease. Blood 75:1383-1395. Gratama, J.W., D’hautcourt, J.-L., Mandy, F., Rothe, G., Barnett, D., Janossy, G., Papa, S., Schmitz, G., and Lenkei, R. 1998. Flow cytometric quantitation of immunofluorescence intensity: Problems and perspectives. European Working Group in Clinical Cell Analysis. Cytometry 33:166-178. Henderson, L.O., Marti, G.F., Galgalas, A., Hannon, W.H., and Vogt, R. Jr. 1998. Terminology and nomenclature for standardization in quantitative fluorescence cytometry. Cytometry 33:97-105. Janossy, G., Bikoue, A., Tilling, R.S., Reilly, J.T., Granger, V., and Barnett, D. 1998. Stabilised cellular immunofluorescent analysis (SCIFA) a new concept for quantitative immunocytometry in routine immunohematology. Cytometry Supp. 9:145-146. Lacombe, F., Durrieu, F., Briais, A., Dumain, P., Belloc, F., Bascans, E., Reiffers, J., Boisseau, M.R., and Bernard, P. 1997. Flow cytometry CD45 gating for immunophenotyping of acute myeloid leukemia. Leukemia 11:1878-1886. Lenkei, R. and Anderson, B. 1995. Determination of the antibody binding capacity of lymphocyte membrane antigens by flow cytometry in 58 blood donors. J. Immunol. Methods 183:267277.
Literature Cited
Lenkei, R., Gratama, J.W., Rothe, G., Schmitz, G., D’hautcourt, J-L., Arekrans, A., Mandy, F., and Marti, G. 1998. Performance of calibration standards for antigen quantitation with flow cytometry. Cytometry 33:188-196.
Autran, B., Carcelain, G., Li, T.S., Blanc, C., Mathez, D., Tubiana, R., Katlama, C., Debre P., and Leibowitch, C. 1997. Positive effects of combined antiretroviral therapy on CD4+ T cells homeostasis and function in advanced HIV disease. Science 277:112-116.
Matutes, E., Owusu-Ankomah, K., Morilla, R., Garcia Marco, J., Houlihan, A., Que, T.H., and Catovsky, D. 1994. The immunological profile of B-cell disorders and proposal of a scoring system for the diagnosis of CLL. Leukemia 8:16401645.
Bikoue, A., George, F., Poncelet, P., Mutin, M., Janossy, G., and Sampol, J. 1996. Quantitative analysis of leukocyte membrane antigen expression: Normal adult values. Cytometry 26:137147.
National Committee for Clinical Laboratory Standards. 1998. Terminology and Definitions for Use in NCCLS Documents; Approved Standard. NCCLS Publication NRSCL8-A. National Committee for Clinical Laboratory Standards, Wayne, Pa. Phenotypic Analysis
6.12.21 Current Protocols in Cytometry
Supplement 22
Poncelet, P. and Carayon, P. 1985. Cytofluorometric quantification of cell-surface antigens by indirect immunofluorescence using monoclonal antibodies. J. Immunol. Meth. 85:65-74. Poncelet, P., George, F., Papa, P., and Lanza, F. 1996. Quantitation of hematopoietic cell antigens in flow cytometry. Eur. J. Histochem. 40:15-32. Poncelet, P., Besson-Faure, I., and Lavabre-Bertrand, T. 2000. Clinical applications of quantitative immunophenotyping. In Immunophenotyping (C. Stewart and J.K.A. Nicholson, eds.) pp. 105-132. Wiley Liss, New York. Richards, J.S., Rawstron, C.A., and Hillmen, P. 2000. Application of flow cytometry to the diagnosis of paroxysmal nocturnal hemoglobinuria. Cytometry 42:223-233. Schwartz, A., Fernández-Repollet, E., Vogt, R., and Gratama, J.W. 1996. Standardizing flow cytometry: Construction of a standardized fluorescence calibration plot using matching spectral calibrators. Cytometry 26:22-31.
Titus, J.A., Sharrow, S.O., Connolly, J.M., and Segal, D.M. 1981. Fc (IgG) receptor distribution in homogeneous and heterogeneous cell populations by flow microfluorometry. Proc. Natl. Acad. Sci. U.S.A. 78:519-523. Weisbrot, I.M. 1979. Linear regression. In Clinical Laboratory Statistics, 2nd Ed. (R.N. Barnett, ed.) pp 45-60. Little, Brown, and Company, Boston.
Key References Cytometry Volume 33, Number 2, October 1, 1998. A special issue of Cytometry: Quantitative Fluorescence Cytometry: An Emerging Consensus. This issue contains summaries of three international conferences held in Europe and the United States in late 1997, review papers that provide a general background on QFCM, original reports on clinical relevance of CD38 expression, use of QFCM in multi-center multi-platform situations, as well as papers on more fundamental aspects of QFCM. Poncelet et al., 2000. See above.
Schwartz, A., Marti, G.E., Poon, R., Gratama, J.W., and Fernandez-Repollet, E. 1998. Standardizing flow cytometry: A classification system of fluorescence standards used for flow cytometry. Cytometry 33:106-114.
A recent and exhaustive review, illustrating the potential of antigen quantitation with emphasis on clinical applications of QFCM. With more than one hundred references.
Serke, S., van Lessen, A., and Huln, D. 1998. Quantitative fluorescence flow cytometry: A comparison of the three techniques for direct and indirect immunofluorescence. Cytometry 33:179-187.
Internet Resources
Tabernero, M.D., Bortoluci, A.M., Alaejos, I., Lopez-Berges, M.C., Rasillo, A., Garcia-Sanz, R., Garcia, M., Sayagues, J.M., Gonzalez, M., Mateo, G., San Miguel, J.F., and Orfao, A. 2001. Adult precursor B-ALL with BCR/ABL gene rearrangements displays a unique immunophenotype based on the pattern of CD10, CD34, CD13, and CD38 expresssion. Leukemia 15:406-414.
http://cyto.mednet.ucla.edu Provides a protocol written by L.E. Hultin and J.V. Giorgi for estimating the number of CD38 molecules on the CD8+ T lymphocytes of HIV-infected individuals. Protocol for general distribution developed with support from U.S. Public Health Service Award U01-A1-37613:
Contributed by Jean-Luc D’hautcourt Centre Hospitalier Régional Mons Warquignies Boussu, Belgium
Quantitative Analysis of Membrane Antigen Expression
6.12.22 Supplement 22
Current Protocols in Cytometry
Immunophenotyping Using a Laser Scanning Cytometer
UNIT 6.13
In hematology, efforts have to be made to reduce the blood volume needed for immunophenotyping of patients such as neonates or preterm neonates and the critically ill. Slide-based cytometry is a useful approach to achieve this task. This unit describes a simple and reliable preparation and analysis method for four-color analysis of peripheral blood leukocytes (PBLs) on the laser scanning cytometer (LSC). Cells are recognized by DNA staining with 7-aminoactinomycin D (7-AAD; Basic Protocol) and are further stained for surface markers by three different monoclonal antibodies that are directly conjugated to fluorochromes. After analysis, the sample is stained by conventional cytological techniques and the cells are relocalized to yield additional morphological information. The Alternate Protocol describes four-color immunophenotyping without nuclear staining. Because slide-based cytometry is nondestructive and nonconsumptive, samples can be archived and re-analyzed. The Support Protocol describes the setup of the LSC. IMMUNOPHENOTYPING OF PERIPHERAL BLOOD LEUKOCYTES Whole blood is collected using a tube containing EDTA as the anticoagulant. Whereas a conventional blood draw is 7 to 10 ml or more, these protocols require only 500 µl or less. The blood cells in suspension are stained with three monoclonal antibodies that are directly conjugated to different fluorochromes. The erythrocytes are lysed, and the sample is centrifuged and resuspended in PBS. The stained leukocytes are then mounted on conventional uncoated glass microscope slides and air dried. Cells are fixed in acetone and stained for nuclear DNA by incubation with 7-aminoactinomycin D (7-AAD). Analysis in the laser scanning cytometer (LSC) is then triggered on the nuclear fluorescence, and a variety of parameters are measured for the three surface protein–associated fluorescence emissions, DNA-associated fluorescence emission, and forward light scatter. To facilitate data interpretation, the slide is counterstained with hematoxylin and eosin (H & E). The morphology of the different subpopulations of leukocytes and their discrimination from cell debris and other contamination are verified and documented with the built-in CCD camera, or any other camera, by taking micrographs after relocalization of single cells on the slide with the LSC.
BASIC PROTOCOL
The advantage of the Basic Protocol is that because of the DNA staining, all cell types can be easily recognized by and triggered on the nuclear fluorescence. DNA staining provides a very stable assay for leukocyte recognition. Triggering on forward scatter (see Alternate Protocol) can be hampered by low scatter signals in particular lymphocytes. For three-color immunophenotyping, the Basic Protocol is preferred. If four different fluorochromes are required, however, the Alternate Protocol muse be used. Materials Peripheral blood sample (≤500 µl) collected in 1.5-ml syringe containing EDTA (Kabe Labortechnik) Monoclonal antibodies (e.g., CD3, CD4, CD8, CD14, CD45 from BD Biosciences, Caltag Laboratories, Beckman Coulter, DakoCytomation, or other commercial sources) conjugated to fluorescein isothiocyanate (FITC), phycoerythrin (PE), and allophycocyanin (APC) Lysing solution: ammonium chloride lysing solution (APPENDIX 2A) or commercial equivalent (e.g., FACS lysing solution; BD Biosciences) PBS (APPENDIX 2A), room temperature and 4°C Contributed by Attila Tárnok and Andreas O.H. Gerstner Current Protocols in Cytometry (2003) 6.13.1-6.13.15 Copyright © 2003 by John Wiley & Sons, Inc.
Phenotypic Analysis
6.13.1 Supplement 23
0.5% (w/v) paraformaldehyde in PBS (APPENDIX 2A) PBS/BSA: 0.5% (w/v) BSA/0.016% (w/v) sodium azide in PBS (APPENDIX 2A), pH 7.40 Acetone, 4°C Fluorescence mounting medium (e.g., DakoCytomation) 5 µg/ml 7-aminoactinomycin D (7-AAD; Sigma) in PBS/BSA Hematoxylin and eosin (H & E): purchase a commercial kit or see Taboas and Ceremsak (1967) 70%, 80%, 90%, and 100% (v/v) ethanol Xylene (e.g., E. Merck) Permanent mounting medium for cytological slides (e.g., Eukitt; Kindler) 1.5-ml microcentrifuge tubes or 12 × 75–mm polystyrene tubes (e.g., Falcon) Grease pencil to mark slides Uncoated glass microscope slides Micropipettor and plastic tips Laminar flow chamber Coplin jars Humidified chamber for incubating slides 24 × 24–mm coverslips Laser scanning cytometer (LSC; CompuCyte) equipped with argon (Ar) and helium/neon (HeNe) laser; standard filter settings for the measurement of FITC, PE, APC, and 7-AAD fluorescence; and WinCyte software Additional reagents and equipment for setting up LSC (see Support Protocol) and H & E staining (Taboas and Ceremsak, 1967) NOTE: Common EDTA syringes are available in sizes >1.5 ml and provide far more material than is required for the assay. Subdivide the EDTA solution from the syringe into small neutral syringes and reduce the volume of blood drawn accordingly. Prepare sample and stain for cell surface markers 1. Pipet 10 or 20 µl of a peripheral blood sample into a 1.5-ml microcentrifuge tube or a 12 × 75–mm polystyrene tube. The smaller and larger solution volumes given in subsequent steps correspond to these two sample sizes. The volume used depends on the amount of blood available. Using the smaller blood volume for staining and analysis reduces the antibody volume needed; however, the proportion of cells lost during the washing steps and the inaccuracy of pipetted volumes increases with smaller blood volumes.
2. Add 1 µl (for 10 µl blood) or 2.5 µl (for 20 µl blood) of each appropriate monoclonal antibody. Mix well and incubate 15 min in the dark. Exact antibody concentrations have to be optimized by titration (see Critical Parameters, discussion of quality control). There is no limitation to the antibodies, antibody classes, or antibody suppliers that can be used in this protocol (Gerstner et al., 2000). Control antibodies (i.e., those that do not stain leukocytes) should always be used to stain a second sample aliquot as a negative control. From this step on, samples must be protected from light to prevent bleaching of the fluorochromes. Staining is best performed within 1 hr after collecting the blood. Although the blood samples may be stored in EDTA solution overnight, this can lead to some problems, such as increased background fluorescence, and is not recommended. Immunophenotyping Using a Laser Scanning Cytometer
3. Add 0.5 ml lysing solution and incubate 15 min at room temperature in the dark. 4. Centrifuge 5 min at 250 × g, discard supernatant, and resuspend cell pellet in 0.5 ml PBS. Centrifuge again and discard supernatant.
6.13.2 Supplement 23
Current Protocols in Cytometry
5. Resuspend cell pellet in 50 or 100 µl of 0.5% paraformaldehyde in PBS and incubate 15 min at room temperature in the dark. 6. Centrifuge 5 min at 250 × g, discard supernatant, and resuspend cell pellet in 250 or 500 µl PBS/BSA solution. At this point, stained samples may be stored at 4°C overnight in the dark. However, many tandem fluorochromes (e.g., PE-Cy5, PE-Cy7) are sensitive to aldehydes; if conjugates of these dyes are used, preparation and analysis should be carried out within the next 2 to 3 hr.
7. Centrifuge 5 min at 250 × g, discard supernatant, and resuspend cell pellet in the remaining volume of buffer (∼50 or 100 µl). 8. Use a grease pencil to designate two small areas on an uncoated glass slide. A template should be used for drawing the areas on the slide to ensure that cells are placed in defined areas of the slide with low variation. Alternatively, microchambers can be constructed on a slide using Parafilm M foil (Pechiney Plastic Packaging). For this purpose, foil pieces are cut to the size and shape of the slide. Microchambers of the desired shape and size are then cut in the foil with a scalpel (e.g., two separate squares each measuring 1.2 cm by 1.2 cm). The foil is placed on the glass slide and heated on a hotplate at 80°C for ∼5 min until the foil starts to melt. The slide is removed from the heat and allowed to cool to room temperature before it is used.
9. Use a micropipettor with a plastic tip to transfer the cells to the glass side. Move the tip over the designated areas of the slide to disperse the cells. The number of cells placed on the slide has to be within a certain range for optimal analysis. As a rule of thumb, adjust the volume placed on the slide or dilute the cell suspension in such a way that cells lie separated from each other by ∼1 cell diameter. This should be checked by direct visualization through the eyepiece of the microscope. If cells are too close, too many events will be triggered as doublets; if cells are too far apart, analysis will take longer (see Critical Parameters).
Stain cellular DNA on the slide 10. Allow slide to air dry 15 min in a laminar flow chamber or, alternatively, overnight at room temperature in a dry and dark place. The way in which air drying is performed is crucial for the results. If air drying is too rapid (i.e., performed by heating with a fan), cells may be disrupted and fluorescence will be strongly affected.
11. Incubate slide 10 min in a Coplin jar filled with cold acetone at room temperature in the dark. Remove slide from jar and air dry 15 min in the dark in the laminar flow chamber. 12. Place slide flat on a horizontal surface and pipet 200 µl PBS/BSA onto each designated area of the slide. Incubate 15 min at room temperature in a humidified chamber in the dark to rehydrate cells. A simple humidified chamber can be made using a 1-liter beaker containing paper towels moistened with distilled water, and covered with aluminum foil.
13. Discard PBS/BSA from slide by inversion and pipet 100 µl of 5 µg/ml 7-AAD in PBS/BSA onto each designated area of the slide. Incubate 30 min at room temperature in the humidified chamber in the dark. 14. Wash slide twice with 500 µl cold PBS.
Phenotypic Analysis
6.13.3 Current Protocols in Cytometry
Supplement 23
15. Pipet 40 µl fluorescence mounting medium onto each designated area of the slide and place a separate coverslip on each area. At this point slides should be either immediately analyzed or stored ≤36 hr at 4°C in the dark. Storage will affect some fluorochrome-conjugated reagents. Under optimal conditions, samples may be analyzed even after weeks of storage. 7-AAD and APC are stable for weeks, but FITC is pH sensitive. For extended storage and reanalysis, FITC-conjugated reagents can be replaced by reagents conjugated with Alexa green. Because 7-AAD binds noncovalently to DNA, 5 ìg/ml 7-AAD (final) should be added to the mounting medium to avoid washing out the dye.
Set up LSC and acquire data 16. Set up the LSC (see Support Protocol). Steps 17 to 21 describe the optimization of the LSC settings. If similar samples have already been run, instrument settings are set by loading an already measured data file for the appropriate fluorescence color combination. In any case, before starting measurements the optimal threshold setting should be verified (steps 20 to 21).
17. Place the slide containing the stained cells (step 15) onto the motorized stage of the LSC under the 20× objective in a standardized and reproducible manner. Alternatively, the 10× or 40× objective can be used. With the 10× objective, the same area is measured four times faster, but measured fluorescence values will be lower than with the 20× objective. With the 40× objective, spatial resolution is higher, but analysis takes four times longer than with the 20× objective. The LSC has a high focal depth so that exact focusing is not crucial. Placement of the slide on the stage is crucial for exact relocalization of single cells of interest.
18. Focus on the cells and adjust the scan area so that cells in one or more designated areas of the glass slide are measured. 19. Run sample slide with the setup conditions established (see Support Protocol, steps 2 and 3a), trigger the analysis on the 7-AAD signal (far-red fluorescence; contouring parameter), and use the set sensor option to adjust the PMT (%) settings to avoid saturated pixels. 20. View scan data image and contours in FL4 (7-AAD). Check that all nuclei are contoured, that cells are not fused to doublets, and that non-nucleated cells and debris are excluded. Adjust the contouring parameter’s minimum area and threshold, if necessary. Parts of the cell should not be excluded from the measurement, nor should too much of the cell-free area be included in it. The setting on the scan data image should be checked.
21. View scan data images and contours in FL1 (Ar laser excitation), FL2 (Ar laser excitation), and FL4 (HeNe laser excitation) and check that the contour region covers the whole cytoplasmic region of the cells. Adjust the “add pixel to threshold” value so that the analysis contour around the cell includes the whole cell area, if necessary. The typical “add pixel to threshold” value is 10.
22. Stop the measurement and save the data (i.e., the instrument settings) as an appropriate fcs file (*.fcs). Reload this data file for subsequent use. Continue analyzing cells. Areas on the slide that have been used for setting up the instrument should be excluded from the analysis, because the fluorescence in these regions may have been affected by photobleaching. Immunophenotyping Using a Laser Scanning Cytometer
23. When measurements are finished, save data file under an appropriate file name (*.fcs).
6.13.4 Supplement 23
Current Protocols in Cytometry
A
3
300 sq. µ
A1 1000
1000000
Area
10000000 CD45-APC
CD3+CD4+
CD4- & CD14-PE
5
1000000 CD4- & CD14-PE
1000000
5011872 CD45-APC
A2 1000
CD3-FITC
4
CD3− CD3+CD4−
B
C1 1000
1000000
100000
6 1000
7-AAD
1000000
C2 1000 CD4- & CD14-PE
10000000
1000000
CD3-FITC
CD45-APC
CD4- & CD14-PE
7
1000
8 D1 1000
1000000
D2 1000
1000000 CD4- & CD14-PE
E1 1000
1000000 CD3-FITC
100000
1000
CD45-APC
CD4- & CD14-PE
10000000
1000000
CD3-FITC
from gate 2
1000000
100000
501187
100000
1000
1
CD45-APC
CD4- & CD14-PE
Fwd scatter max pixel
10000000
1000000
from gate 1
2
E2 1000
1000000 CD4- & CD14-PE
Figure 6.13.1 Typical example of analysis and gating strategy on the laser scanning cytometer for three-color plus DNA-stained peripheral blood leukocytes from lysed whole blood according to the Basic Protocol. Dot plot A shows the forward scatter maximum pixel value (max pixel) versus the measured cell area for all cells. Gate 1 (large rectangle), single cells. Cell aggregates appear to the right of this gate (open arrow). Gate 2 (oval), eosinophils (cells with high forward scatter max pixel); gate 3 (small rectangle), other non-eosinophil single cells. Cells in gate 1 are displayed in dot plots A1 and A2. Dot plot B shows all cells in gate 3; the 7-AAD fluorescence is plotted versus the CD45-APC fluorescence. Gate 4, lymphocytes; gate 5, neutrophils and monocytes; gate 6, basophils. Dot plots C1 and C2 show all cells in gate 4 (lymphocytes) as displayed in A1 and A2. Dot plots D1 and D2 show all cells in gate 5. In D1, gate 7 identifies monocytes and gate 8 identifies neutrophils. Data are displayed on a log scale except forward scatter max pixel, area, and 7-AAD fluorescence, which are displayed on a linear scale. Only fluorescence integral values are displayed. Figure reprinted from Gerstner et al. (2000), with modifications, with permission from Elsevier Science.
6.13.5 Current Protocols in Cytometry
Supplement 23
Stain cells for morphologic assessment 24. Remove slide from the microscope stage and place in an upright position inside a Coplin jar filled with PBS. Incubate slide 15 min and let the coverslip slip off the slide. 25. Stain cells with H & E according to the stain manufacturer’s instructions (or see Taboas and Ceremsak, 1967). The authors use a 5-min incubation at room temperature in Meyer’s Hemalun Solution (Merck), followed by a 1-min rinse in water, and a 5-min incubation in water in a Coplin jar. The slide is then transferred to a jar with a 2% (w/v) eosin solution (Merck) for 2 min and is again rinsed in water.
26. Dehydrate cells using a series of 5-min sequential washes in 70%, 80%, 90%, and 100% ethanol. 27. Incubate slide 3 min in xylene at room temperature. 28. Place one drop of permanent mounting medium per designated area on the slide and cover each section with a coverslip. Store slides as normal cytological samples at room temperature in the dark. Analyze immunophenotypic data 29. For the analysis of leukocyte subsets with the proprietary software WinCyte, create a display (*.dpr) protocol as shown in Figure 6.13.1. The listmode files generated by the LSC are in FCS3.0 format and are most easily analyzed with the proprietary WinCyte software. Alternatively, the data can be exported, downsized to the FCS2.0 standard, and analyzed with other software for flow cytometry (that can analyze FCS2.0 data files).
30. Create dot plot A of area versus forward scatter maximum pixel value (max pixel) on linear scales and display the data. Load the data file. Three major populations should be present: one with high forward scatter max pixel (eosinophils) and two with low forward scatter max pixel and smaller (lymphocytes) or larger (neutrophils) areas.
31. For dot plot A, create gates 1 to 3 for cells with high or low forward scatter max pixel values in the cell area range of 10 to 100 µm2. 32. Place the corresponding H & E-stained sample on the stage of the LSC, relocalize cells from gates 1 through 3, and visualize them with the built-in CCD camera (Fig. 6.13.2). Adjust gate positions and sizes if necessary. If gates are located correctly, all leukocyte subsets (gate 1), eosinophils (gate 2), and residual leukocytes (gate 3) will be included and doublets will be excluded.
33. Create dot plot A1 showing CD3-FITC versus CD4- & CD14-PE integral fluorescence on a logarithmic scale. Gate dot plot A1 on gate 1 and display the data. Four major populations should be visible: two with low CD3-FITC and either low CD4& CD14-PE (neutrophils and natural killer [NK] and B cells) or high CD4- & CD14-PE (monocytes), and two with high CD3-FITC and either low CD4- & CD14-PE (cytotoxic T cells) or high CD4- & CD14-PE (helper T cells).
34. Create dot plot B with 7-AAD integral fluorescence (linear) versus CD45-APC integral fluorescence (logarithmic). Gate dot plot B on gate 3 from dot plot A and reload the data file. Three major populations should be present. Immunophenotyping Using a Laser Scanning Cytometer
6.13.6 Supplement 23
Current Protocols in Cytometry
A
B
C
D
E
Figure 6.13.2 Typical example of morphological images of leukocytes identified according to the strategy described in the Basic Protocol. Dot plots correspond to those in Figure 6.13.1. (A) Relocalized eosinophils. Hematoxylin and eosin–stained cells were relocalized using gate 2 (dot plot A). (B) Relocalized neutrophils from both gate 5 (dot plot B) and gate 8 (dot plot D1). (C) Relocalized monocytes from both gate 5 (dot plot B) and gate 7 (dot plot D1). (D) Relocalized lymphocytes from gate 4 (dot plot B) and gate CD3+CD4−. (E) Relocalized basophils from gate 6 (dot plot B). Images were taken with a 100× objective and a digital camera (Olympus Camedia C-2020 Zoom; A,E) or a 40× objective and the built-in CCD-camera (B,C,D) and show the first eight subsequently relocalized cells. Figure reprinted from Gerstner et al. (2000), with modifications, with permission from Elsevier Science.
35. Create gates 4, 5, and 6 in dot plot B and relocalize the cells in the respective gates on the H & E-stained specimen (Fig. 6.13.2). Adjust gate positions and sizes if necessary. Lymphocytes (gate 4; low 7-AAD and high CD45), neutrophils and monocytes (gate 5; high 7-AAD and low CD45), and basophils (gate 6; low 7-AAD and low CD45) should be present.
36. Create dot plot C1 of CD3-FITC versus CD4- & CD14-PE integral fluorescence (both logarithmic). Gate dot plot C1 on gate 4 and display the data. Three major populations should appear: one with low CD3-FITC and low CD4- & CD14-PE (CD3−; B and NK cells), and two with high CD3-FITC and either low CD4- & CD14-PE (CD3+CD4−; cytotoxic T cells) or high CD4- & CD14-PE (CD3+CD4+; helper T cells).
37. Create dot plot D1 of CD3-FITC versus CD4- & CD14-PE integral fluorescence (both logarithmic). Gate dot plot D1 on gate 5 and display the data. Adjust gate 5 in dot plot B to minimize the population with high CD3 integral fluorescence (contaminating T lymphocytes). Two major populations with low CD3 integral fluorescence should be present. Phenotypic Analysis
6.13.7 Current Protocols in Cytometry
Supplement 23
38. Create gates 7 and 8 for dot plot D1 and relocalize cells (Fig. 6.13.2). Adjust gates if necessary. Monocytes (gate 7; low CD3-FITC and high CD4- & CD14-PE) and neutrophils (gate 8; low CD3-FITC and low CD4- & CD14-PE) should be present. NOTE: Eosinophilic granulocytes have a high affinity for unbound FITC, which is found in any vial of FITC-conjugated antibodies. Therefore, when FITC-conjugated antibodies are used, eosinophilic cells within the sample would also be stained and could be mistaken for a FITC-positive subpopulation (Bedner et al., 1999). Additionally, eosinophils are highly autofluorescent and will fluoresce in all fluorescence channels when they are excited with the blue laser line, even in the absence of staining. Autofluorescence occurs minimally when the red laser line is used. ALTERNATE PROTOCOL
FOUR-COLOR IMMUNOPHENOTYPING WITHOUT NUCLEAR STAINING The use of a DNA dye such as 7-AAD prevents the use of a fourth fluorochrome-labeled antibody reagent with an overlapping fluorescence emission spectrum (e.g., phycoerythrin [PE]-Cy5-conjugated monoclonal antibodies). Furthermore, 7-AAD-associated fluorescence spills over into the detector for PE fluorescence and reduces sensitivity. If 7-AAD staining is omitted, then a directly conjugated PE-Cy5 antibody can be added. This allows four-color immunophenotyping and increases PE sensitivity. In the absence of a DNA dye, data acquisition has to be triggered on light scatter and/or on a specific fluorescence emission. Conveniently, the filter settings in both protocols are identical; both 7-AAD and PE-Cy5 are measured in FL4-Ar. Additional Materials (also see Basic Protocol) PE-Cy5-conjugated monoclonal antibody Prepare sample and stain for cell surface markers 1. Prepare blood sample and stain cells as described (see Basic Protocol, steps 1 to 10), adding a PE-Cy5-conjugated monoclonal antibody in step 2. The tandem dye (PE-Cy5)-conjugated antibodies have to be carefully selected. In general, all tandem dyes will also emit in the frequency of their donor fluorochrome (here, PE). This fluorescence depends on the quality of the sample and should be minimal (<5% of the cognate signal in Cy5) in order to reduce undesired spectral overlap with the specific PE signal.
2. Pipet 40 µl mounting medium onto each designated area of the slide and place a separate coverslip on each area. At this point, slides should be either immediately analyzed or stored ≤36 hr at 4°C in the dark. Storage will affect some fluorochrome-conjugated reagents.
Set up LSC 3. Set up the LSC (see Support Protocol). Steps 4 to 11 describe the alignment of the microscope condenser and optimization of the LSC settings. If similar samples have already been run, instrument settings are set by loading an already measured data file for the appropriate fluorescence color combination. The condenser, however, should be aligned before each sample is analyzed, and the optimal threshold setting should be verified (see Basic Protocol, steps 20 to 21).
Immunophenotyping Using a Laser Scanning Cytometer
6.13.8 Supplement 23
Current Protocols in Cytometry
4. Place the slide containing the stained cells (step 2) onto the motorized stage of the LSC under the 20× objective in a standardized and reproducible manner. Alternatively, the 10× or 40× objective can be used. With the 10× objective, the same area is measured four times faster, but measured fluorescence values will be lower than with the 20× objective. With the 40× objective, spatial resolution is higher, but analysis takes four times longer than with the 20× objective. The LSC has a high focal depth so that exact focusing is not crucial. Placement of the slide on the stage is crucial for exact relocalization of single cells of interest.
5. Focus on the cells and adjust the scan area so that cells in one designated area on the glass slide are measured. Align microscope condenser 6. To optimize the forward-scatter signal, run the sample and display the scan data image. 7. Open the aperture of the condenser. Move the condenser carefully up or down and check if the contrast increases. Best results are obtained if cells appear as white or gray dots on a black background in the scan data display. Improper settings may result in either triggering on noncellular events (debris, inconsistencies in the medium) or a lack of cell recognition.
8. If the cells still appear dark with a bright background, then adjust the obscuration bar of the forward-scatter detector by removing the cover from the forward-scatter detector and turning on the Ar laser. If the light from this laser is not seen as a bright line on the obscuration bar, position the condenser so the laser hits the obscuration bar and only a small amount of light is going to the photodiode below. Run the sample and fine tune the scatter signal by carefully turning the obscuration bar. 9. Repeat steps 6 to 8 until the scan data image is acceptable. Acquire and analyze data 10. Run sample slide with the setup conditions established (see Support Protocol, steps 2 and 3b), triggering the analysis on the forward-scatter signal of the Ar laser or on a combination of the forward scatter signal with one to three of the fluorescence signals generated by the Ar laser (see Support Protocol, step 3b) and using the set sensor option to adjust the PMT (%) settings to avoid saturated pixels. 11. Continue to optimize settings and analyze cells as described (see Basic Protocol, steps 20 to 38). The filters used in the Basic Protocol and the Alternate Protocol are identical. For data acquisition and analysis, a display protocol as shown in Figure 6.13.1 can be used. Obviously, because 7-AAD is replaced by the PE-Cy5-conjugated antibody, the gating has to be modified (as an example, see Gerstner et al., 2002).
Phenotypic Analysis
6.13.9 Current Protocols in Cytometry
Supplement 23
SUPPORT PROTOCOL
SETTING UP A LASER SCANNING CYTOMETER This protocol describes how to set up the laser scanning cytometer (LSC) for measurement of peripheral blood leukocytes (PBLs) stained for surface antigens and stained for nuclear DNA with 7-aminoactinomycin D (7-AAD; see Basic Protocol). In the standard instrument setup, fluorescence is collected with the following filters: FL1, 530/DF30 band-pass (green fluorescence; FITC, Alexa green); FL2, 580/DF30 band-pass (orange fluorescence; PE); FL3, 625/DF28 band-pass (red fluorescence; PE-Cy3); and FL4, 650EFLP long-pass (far-red fluorescence; 7-AAD or PE-Cy5 with Ar laser excitation or APC with HeNe laser excitation). For a more detailed description of the LSC, please refer to the review literature (Kamentsky and Kamentsky, 1991; Kamentsky et al., 1997; Tárnok and Gerstner, 2002). Materials Laser scanning cytometer (LSC; CompuCyte) equipped with argon (Ar) and helium/neon (HeNe) laser; standard filter settings for the measurement of FITC, PE, APC, and 7-AAD fluorescence; and WinCyte software 1. Start the WinCyte software of the LSC and load an appropriate display protocol (.dpr). The instrument comes with a default display protocols (*.dpr), which must be loaded and then further adapted to the experimental needs.
2. Adjust photomultiplier (PMT) settings (PMT [%]/offset/gain values) for each fluorescence channel. For the Ar laser, set FL1 to 30/2075/255, FL2 to 30/2075/255, and FL4 to 65/2075/255. (FL3 is not active.) For the HeNe laser, set FL4 to 65/2075/255. (FL1 through FL3 are not active.) 3a. For 7-AAD-labeled cells (see Basic Protocol): Set “contouring parameter” (i.e., threshold parameter) to FL4-Ar, “add pixels to threshold” to 10, “minimum area” to 8 µm2, and “threshold” to 1200. These settings depend on the PMT settings and must be checked initially before starting data acquisition.
3b. For cells without nuclear staining (see Alternate Protocol): Set “contouring parameter” (i.e., threshold parameter) to forward-scatter signal, “add pixels to threshold” 4 to 6, “minimum area” to 8 µm2, and “threshold” to 1000. The alignment of the microscope condenser is crucial for optimal triggering on the forward scatter signal and has to be carefully adjusted (see Alternate Protocol, steps 6 to 9). NOTE: The light-scatter signal is highly dependent on the refraction index of the mounting medium. Best contrast is obtained for intact cells in PBS. Cells in PBS, however, have to be measured shortly after preparation. Alternatively, a specific fluorescence signal (e.g., CD45 for all leukocytes) can be used for triggering or a virtual new trigger parameter can be created by adding the forward-scatter signal and up to three of the fluorescences generated by the argon laser. (To do this, click the “add parameter” check box in the instrument settings menu and activate the “added parameter” channel.)
COMMENTARY Background Information
Immunophenotyping Using a Laser Scanning Cytometer
In hematology, many efforts have to be made to reduce the blood volume needed for immunophenotyping. This is important, for example, for patients with a lymphoma that is yet undefined or with leukemia and for monitoring the course of disease under therapy. There are many
more situations, however, in which minimizing volumes will prove beneficial for the patient either by making immunophenotyping possible at all, as in neonates or preterm neonates, or by allowing a continuous analysis of cell subtypes, as in the critically ill patient.
6.13.10 Supplement 23
Current Protocols in Cytometry
The analysis of the complexity of the immune system can be rapidly performed by flow cytometry (FCM) with routinely applied threeand four-color immunophenotyping. The amount of data acquired from a single sample is increased by using novel fluorescent dyes for multiple-color staining (Roederer et al., 1996; Gerstner et al., 2002). Furthermore, three-color analysis of lymphocyte subsets can be combined with measurements of cell proliferation by FCM (Schmid et al., 2000). Using FCM, however, does not allow the analysis of cell morphology or re-staining of cells of interest unless cell sorting is performed. Sorting is technically demanding, time consuming, and not applicable for routine analytical purposes, especially in small-volume samples. To reliably detect a rare population of cells (≤1% of the total) in peripheral blood leukocytes (PBLs), ≥10,000 cells have to be analyzed. Although slide-based cytometry techniques are applicable for rare-cell detection (Bajaj et al., 2000), a drawback of immunophenotyping by scanning laser cytometry is the speed of analysis; the same number of cells would be analyzed within 10 to 30 sec by FCM. With scanning laser cytometry, however, cells are not lost and further examination of morphology or staining by other means without the need for physical cell sorting dramatically increases the amount of information acquired per cell. For these reasons, analysis by slide-based cytometry has the potential to become a powerful tool for clinical diagnosis. In general, many assays originally developed for FCM analysis can be adapted for the LSC. The main difference is that some incubation steps have to be performed on a slide (e.g., nuclear DNA staining in the Basic Protocol). Nevertheless, staining characteristics and reaction parameters of nucleic-acid dyes and antibodies as well as other probes (fluorescence in situ hybridization, mitochondrial potential) are identical. Clinical applications of scanning laser cytometry have recently been reviewed (Tárnok and Gerstner, 2002). The limitation of scanning laser cytometry compared with FCM is the absence of a sidescatter signal, which, in combination with the forward-scatter signal in FCM, allows a fourpart differential of leukocytes (lymphocytes, monocytes, neutrophils, and eosinophils). In the laser scanning instrument, however, the combination of cell area with forward scatter maximum pixel value (max pixel) and forward scatter integral values allows a three-part differential of lymphocytes, monocytes plus neu-
trophils, and eosinophils (Gerstner et al., 2002; Fig. 6.13.1). In multicolor applications, there is always the problem of fluorescence emission spillover into more than one fluorescence detector. For example, the FITC emission spills over into the phycoerythrin (PE) detector. (For a detailed example of a spillover matrix of the most commonly used fluorochromes for monoclonal antibodies in the LSC, see Gerstner et al., 2002.) Therefore, multicolor analysis requires crosscompensation between all fluorescence detector channels of the instrument. This feature is not yet incorporated in the LSC software. Therefore, for the optimal setting of analysis gates, it is important to run samples in which each fluorochrome is sequentially left out. Measurement of antigens with low expression levels requires a substantial sensitivity for dimly expressed markers. Based on analysis of calibration beads (see Critical Parameters, discussion of quality control; Tárnok and Gerstner, 2002), the sensitivity of the LSC with respect to the recognition of antigens of low abundance is comparable to that of benchtop flow cytometers. The fluorescence detector channels with the highest sensitivity are the PE channel (FL2Ar) and the APC channel (FL4-HeNe). As a rule of thumb, weakly expressed antigens should be labeled with PE or APC.
Critical Parameters The most prominent feature of the LSC is the ability to observe cells directly during analysis (scan data image) and to document their morphology. The tools to accomplish this include the “scan data” key in the WinCyte display and the relocalization process. The “scan data” option (available only during data acquisition) shows the pixel map of the actual scanning step in process at that moment; it can also show the user-defined contours as they are applied for analysis. This is helpful in adjusting the threshold levels and in visualizing stainings that are too weak to be judged by eye. To make full use of the relocation capability, two points are critical. First, it is crucial to place the slide on the microscope stage in a standardized and reproducible way prior to data acquisition. Changes in slide position on the stage during data acquisition will make it impossible to relocalize a cell. Best results in terms of preserving the x-y coordinates are usually obtained by relocalizing the cells of interest immediately after scanning; however, only fluorescence stainings are present at this point. More frequently, scanned slides are
Phenotypic Analysis
6.13.11 Current Protocols in Cytometry
Supplement 23
Immunophenotyping Using a Laser Scanning Cytometer
stained again by conventional cytopathological methods in a second independent step and cells are relocalized afterwards. This staining must be performed without shifting the cells on the slide and with minimal cell loss. For this purpose, the coverslip should not be dragged from the slide but should be allowed to slip off by gravity in a container filled with PBS. Nevertheless, cell loss will always occur to a certain extent when cells are relocalized, but it should neither exceed 5% nor show preference for a specific cell type. As for any other cytometric assay, the process of triggering and distinguishing a cell from noise is of fundamental importance. First a triggering parameter (contouring parameter) should be selected, which in the Basic Protocol is 7-AAD fluorescence and in the Alternate Protocol is forward scatter or a virtual “added parameter.” Events on the slide will be represented on a gray-scale pixel map where the brightness of a pixel represents the fluorescence intensity or the light-scatter intensity detected at this spot on the slide; a cell is represented as a cluster of pixels. Next, the criteria are set that must be fulfilled to include a cluster of pixels as a cell in the analysis. A minimal area covered continuously by pixels with a minimal brightness (i.e., threshold level) has to be defined. This will generate a triggering contour around a cluster of pixels, which in the Basic Protocol will include the cell nucleus and in the Alternate Protocol will include the whole cell. To include the cytoplasmic rim, a second contour will be drawn at a user-defined distance from the triggering contour (typically ∼10 pixels in the Basic Protocol and 4 to 6 pixels in the Alternate Protocol); this is the actual analysis contour. To subtract background staining for each individual cell, two additional contours are drawn in the periphery at user-defined distances outside the analysis contour. All these settings should be such that the entire cell is included in the analysis, only cells (but neither larger nor smaller particles) are triggered, and cells are triggered as single cells and are not fused to doublets or cell multiplets. (For further discussion, refer to Tárnok and Gerstner, 2002.) Obviously, these settings depend on the density of the cells on the slide. If cells are too close to one another, singlets are almost impossible to discriminate. In general, doublets or larger cell clusters are excluded from the analysis. If cells are too scattered, data acquisition will take a long time because a large area has to be scanned. For the Basic Protocol, the frequency
of cell aggregates does not generally exceed 2% to 3% of all cells. During the staining steps, it is important that each designated area be entirely covered by the staining solution without bubbles. This is of greatest importance for 7-AAD staining. Cells that are not properly covered are incubated with a lower concentration of 7-AAD (or with none at all) and will therefore not be triggered properly. Nuclear staining intensity with 7-AAD depends on the degree of DNA condensation (Loborg et al., 1995). This allows differentiation of neutrophils from mononuclear cells solely on the basis of 7-AAD staining intensity (Gerstner et al., 2000). When staining across the population of cells is uneven, this resolution is lost. Changes in 7-AAD concentration could even take place after the staining is completed, by dissociation of the dye from the DNA. To maintain an equilibrium without uncontrolled shifting of 7-AAD from the nucleus to the extracellular space (which could be a problem at the edge of the designated area), extra 7-AAD should be added to the mounting medium used to cover the slide. To determine whether areas of discontinuous 7-AAD staining have occurred on a slide, 7-AAD intensity should be plotted versus y position of each cell on the slide without gating. This should result in two parallel vertical lines if the staining is even across the designated area. The line representing cells with a low 7-AAD intensity corresponds to neutrophils, and the line representing those with a high intensity corresponds to lymphocytes. Improperly stained regions will show up as irregularities (i.e., drifting of the intensities to higher or lower values) and are excluded by gating (Gerstner et al., 2000). Forward scatter is an important parameter in immunophenotyping by scanning laser cytometry. It allows one to detect cells without additional staining and discriminates between eosinophils and cell debris based on the max pixel signal. In order to yield optimal forwardscatter results, the optical alignment of the condenser lens, shutter, and the photodiode underneath the condenser that detects the scattered light should be carefully adjusted. The quality of the forward-scatter signal should be checked prior to any measurement; cells or particles should appear as white or gray objects against an evenly dark background. Quality control The strategy for optimal discrimination of positive and negative signals is comparable to
6.13.12 Supplement 23
Current Protocols in Cytometry
that usually used in FCM. Discrimination and resolution are quantified using published methods for FCM (Roederer et al., 1996; Chase and Hoffman, 1998; Wood, 1998; Gerstner et al., 2002). In brief, two parameters, resolution and relative brightness, are calculated as follows: Resolution = (Fs − Fu)/(SDs + SDu) Relative brightness = Fs/Fu where Fs and Fu are the mean integral fluorescence of stained and unstained cells, respectively; and SDs and SDu are the respective standard deviations of the fluorescence measurements. Both resolution and relative brightness have to be maximized for optimal results. Instrument performance is best checked with fluorescence alignment beads such as Rainbow Beads (Spherotech) for alignment of the Ar laser and APC beads (Spherotech) for alignment of the HeNe laser. Rainbow Beads and APC beads consist of eight and six bead populations, respectively, of different fluorescence intensities that correspond to different numbers of molecules of equivalent soluble fluorochrome (MESF) per bead. The MESFs of the eight Rainbow Beads for FITC-associated fluorescence are 100, 600, 1800, 4700, 15,000, 40,000, 140,000, and 300,000; for PEassociated fluorescence they are 48, 400, 1200, 3800, 12,000, 34,000, 124,000, and 300,000. For APC-associated fluorescence with the APC beads, the MESFs are 336, 7880, 17,400, 53,100, 130,000, and 208,000. The beads are diluted tenfold in PBS and measured on the LSC with forward scatter as the trigger and a minimum particle area setting of 3 µm2. On a well-aligned instrument, all bead populations are resolved (examples are shown in Tárnok and Gerstner, 2002). The measured fluorescence integral values can be fitted to the respective MESF values by a four-parametric sigmoid function. This calibration can then be used to calculate the MESF values of a cell for the same fluorochrome. The most important factors that influence discrimination of stained cells are the setting of the background correction and the quality of staining. In the laser scanning instrument, background fluorescence is determined for each individual cell. This value is then subtracted from the respective fluorescence intensity values obtained for that cell. Background fluorescence is measured in a ring outside the contouring region around each individual cell. The distance of the background measurement from the contouring region is preset by the operator. Measurement is optimal if background is always acquired outside the cell and does not
overlap with adjacent cells. The background setting can be modified in the background setup of the WinCyte software and can be changed from dynamic (automatic) to manual, if appropriate. An incorrect background setting may result either in high fluorescence intensity of unstained cells or in subtraction of too-high values, leading to artificially negative fluorescence integral values. Antibody concentrations have to be optimized by titration to avoid nonspecific staining. Antibody concentrations are optimal if the resolution parameter is maximal. If cells display abnormally low staining, check the antibodies on a flow cytometer with the same cells. Abnormally low staining or resolution can also be the effect of a contouring region that is too small. Check the size of the contouring region on the scan data display and increase the “add pixels to threshold” value if necessary. The settings and antibody concentration need to be carefully adjusted with real samples (e.g., CD3-stained PBLs) to maximize resolution and relative brightness.
Troubleshooting If too few cells are triggered by the software, the slide has to be directly visualized using the fluorescence microscope. Dark red nuclei should be visible. It should be verified that the filter settings of the fluorescence microscope are correct for visualizing 7-AAD (e.g., XF35 filter block from Omega Optical). If nuclei cannot be seen, but the cells have been properly prepared, cells might have been lost during fixation; either the slides were not air dried sufficiently or the glass of the microscope slide was inadequate for the cell type. Sialanized slides or slides for frozen sections should be tried instead. If cells on the slide are visible by bright field but not by fluorescence, the incubation with 7-AAD might have been faulty. If, however, there are enough cells on the slide but they are not displayed in the “scan data” window, the analysis protocol should be checked in the following sequence. (1) Is the 7-AADassociated fluorescence channel (FL4-Ar) activated? (2) Is the photomultiplier (PMT) set correctly with pixels neither too bright nor too weak? (Check by running the sample with the set sensor function activated.) (3) Are the trigger settings adequate for this specimen? Preparation artifacts can result from loss of 7-AAD intensity towards the edge of the designated area (see Critical Parameters) and contamination by bacteria, yielding irregular changes in 7-AAD intensity.
Phenotypic Analysis
6.13.13 Current Protocols in Cytometry
Supplement 23
With the Basic Protocol, neutrophils and lymphocytes should be discriminated based on their different CD45-APC-associated fluorescence expression. If this is not the case in the LSC, although resolution with the antibody is acceptable in FCM, then the APC fluorescence is too low. In most cases, this results from misalignment of the Ar and HeNe lasers. Check the laser alignment by running Rainbow Beads in dual-laser mode. The scan data images of the beads in FL1-Ar and FL4-HeNe should overlap completely. (This is best visualized in the Compucolor display of the WinCyte software.) If the lasers are misaligned, correct the alignment setup or call technical service. A quick solution is to increase the “add pixels to threshold” value until both the signals from the Ar and the HeNe laser are covered by the contouring threshold. In some cases, cell losses between the start of the analysis and relocalization after hematoxylin and eosin (H & E) staining might be unacceptably high. In general, this is a mechanical problem: either cells were stripped off the slide by harsh removal of the coverslip or they slipped off during H & E staining procedures. In the latter situation let slides air dry sufficiently or try other cytological staining procedures. Instead of real cell loss, however, there could also be a virtual relocalization cell loss. This means that cells actually are still in place but cannot be found because the slide is not exactly in the same position as it was during data acquisition or because the instrument setup is not correct. Relocalize rare but morphologically distinct cells in the sample, such as eosinophilic granulocytes. Check where the cells are on the CCD image and adjust the temporary offset of the CCD image. In the case of incorrect parameter settings, change the settings or call technical service.
Anticipated Results
Immunophenotyping Using a Laser Scanning Cytometer
The protocol should enable one to perform four-color (three antibodies plus DNA or four antibodies) immunophenotyping of PBLs on a slide. With highly expressed surface antigens such as CD3, CD4, CD8, or CD14, among others, stained cells are clearly discriminated. Staining antigens of low abundance on the cell may result in negative results. This may be the case for some activation antigens. The percentages of distinguishable leukocyte subsets should be identical with results obtained by FCM. Expected percentage values in healthy adults are ∼50% to 70% neutrophils, 5% to 10% monocytes, 20% to 30% lymphocytes, 2% to 5% eosinophils, and 0% to 1% basophils. More
than half the T cells are CD4+ and ∼60% to 75% are CD+.
Time Considerations The Basic Protocol requires 4 hr and the Alternate Protocol requires 2.5 hr. The time required for individual steps is as follows: surface antigen staining of whole blood, 15 min; lysing erythrocytes, 15 min; washing steps, 5 min; fixing and washing cells, 25 min; transferring cells to slides, 5 min; and air drying, 15 min (or overnight without laminar flow chamber). The following preparation steps are performed only with 7-AAD staining. Cell permeabilization and subsequent drying take 10 min and 15 min, respectively. Finally, cell rehydration and DNA staining take 45 min. Measuring the slides on the LSC and data analysis require ∼30 min and the optional cytological counterstaining requires 60 min (depending on the protocol). Relocalization and documentation take ∼15 to 30 min.
Literature Cited Bajaj, S., Welsh, J.B., Leif, R.C., and Price, J.H. 2000. Ultra-rare-event detection performance of a custom scanning cytometer on a model preparation of fetal nRBCs. Cytometry 39:285-294. Bedner, E., Halicka, H.D., Cheng, W., Salomon, T., Deptala, A., Gorczyca, W., Melamed, M.R., and Darzynkiewicz, Z. 1999. High affinity binding of fluorescein isothiocyanate to eosinophils detected by laser scanning cytometry: A potential source of error in analysis of blood samples utilizing fluorescein-conjugated reagents in flow cytometry. Cytometry 36:77-82. Chase, E.S. and Hoffman, R.A. 1998. Resolution of dimly fluorescent particles: A practical measure of fluorescence sensitivity. Cytometry 33:267279. Gerstner, A.O.H., Laffers, W., Bootz, F., and Tárnok, A. 2000. Immunophenotyping of peripheral blood leukocytes by laser scanning cytometry. J. Immunol. Methods 246:175-185. Gerstner, A.O.H., Lenz, D., Laffers, W., Hoffman, R.A., Steinbrecher, M., Bootz, F., and Tárnok, A. 2002. Near-infrared dyes for six color immunophenotyping by LSC. Cytometry 48:115-123. Kamentsky, L.A. and Kamentsky, L.D. 1991. Microscope-based multiparameter laser scanning cytometer yielding data comparable to flow cytometry data. Cytometry 12:381-387. Kamentsky, L.A., Burger, D.E., Gershman, R.J., Kamentsky, L.D., and Luther, E. 1997. Slidebased laser scanning cytometry. Acta Cytol. 41:123-143. Loborg, H., Linden, E., Lonn, A., Skoglund, P., and Rundquist, I. 1995. High affinity binding of 7aminoactinomycin D and 4′,6-diamidino-2phenylindole to human neutrophilic granulocytes and lymphocytes. Cytometry 20:296-306.
6.13.14 Supplement 23
Current Protocols in Cytometry
Roederer, M., Kantor, A.B., Parks, D.R., and Herzenberg, L.A. 1996. Cy7PE and Cy7APC: Bright new probes for immunofluorescence. Cytometry 24:191-197. Schmid, I., Cole, S.W., Zack, J.A., and Giorgi, J.V. 2000. Measurement of lymphocyte subset proliferation by three-color immunofluorescence and DNA flow cytometry. J. Immunol. Methods 235:121-131. Taboas, J.O. and Ceremsak, R.J. 1967. A rapid hematoxylin and eosin stain. Tech. Bull. Regist. Med. Technol. 37:119-120.
Tárnok, A. and Gerstner, A.O.H. 2002. Clinical applications of laser scanning cytometry. Cytometry 50:133-143. Wood, J.C. 1998. Fundamental flow cytometer properties governing sensitivity and resolution. Cytometry 33:260-266.
Contributed by Attila Tárnok and Andreas O.H. Gerstner University of Leipzig Leipzig, Germany
The authors would like to acknowledge Wiebke Laffers and Dominik Lenz for their help with the development of these protocols.
Phenotypic Analysis
6.13.15 Current Protocols in Cytometry
Supplement 23
Enzymatic Amplification Staining for Cell Surface Antigens
UNIT 6.14
Flow cytometry is a remarkable and powerful technology because it has made possible the simultaneous analysis of multiple attributes of tens of thousands of cells on a single-cell basis in a short period of time. Nevertheless, one of the most vexing problems associated with flow cytometry is the poor sensitivity of detection for molecules expressed in low amounts (UNIT 6.3). Molecules that are functional at low levels include cytokines, chemokines, cytokine receptors, chemokine receptors, apoptosis-inducing ligands, other surface-bound ligands, viral proteins, transcription factors, specific messenger RNA, activated signal-transduction molecules, enzymes, organellar molecules, cyclins, and Bcl-2-related proteins. These molecules have been difficult to detect with the standard procedures used to stain cells for flow cytometric analysis; however, since they function at low levels, they are still biologically important. In order to address this deficiency, the author has developed a catalyzed reporter deposition system to stain cells for analysis on a flow cytometer (Kaplan and Smith, 2000). This system amplifies the fluorescent signal by adding an enzymatic step that catalyzes the abundant deposition of a labeled molecule onto the cell being stained. Enzymatic amplification staining (EAS) is qualitatively different from the technique of including additional layers of binding molecules since background levels of fluorescent are not increased along with the specific signal. EAS can amplify the separation of the fluorescent signals between specific and control treatments by as much as 100-fold. This added sensitivity has been useful in detecting the presence of molecules not detectable by standard staining procedures. STRATEGIC PLANNING Some cellular molecules are expressed abundantly and do not require any specific amplification procedures for their detection by flow cytometry. For instance, CD4 on T lymphocytes and CD20 on B lymphocytes can be easily and efficiently observed with directly labeled antibodies. Molecules that can be detected by direct or indirect staining methods should be ascertained before turning to the EAS procedure. Molecules that are expressed at >10,000 copies per cell are readily observed by direct or indirect staining techniques. Expression between 2000 and 10,000 copies per cell may be detectable by these methods, depending on the affinity of the probe being used, the fluorophores used, the reaction medium, the incubation temperature, and the sensitivity of the detection flow cytometer. Molecules expressed at <2000 copies per cell are usually not detected by these staining techniques and would consequently be candidates for detection by the EAS technology. The number of copies per cell is not known for most molecules; moreover, populations of cells to be assessed are often heterogeneous and may express different levels of the molecule of interest. It is preferable to first assess staining for a given molecule of unknown cellular abundance with standard staining procedures using directly labeled probes (direct staining) or labeled secondary reagents (indirect staining). Indirect staining gives more sensitivity than direct staining but still may not give a definitive staining pattern. Unacceptable results include the failure to demonstrate that known positive cells give a positive histogram compared to histograms obtained with isotype/subtype control immunoglobulin and/or with negative cells, and the appearance of shoulders or smears (Fig. 6.14.1). Positive histograms should be clearly distinct from control histograms with Phenotypic Analysis Contributed by David Kaplan Current Protocols in Cytometry (2003) 6.14.1-6.14.10 Copyright © 2003 by John Wiley & Sons, Inc.
6.14.1 Supplement 23
Standard amplification
EAS
EAS 100
100
100 CEM
80
CEM
80
60
60
60
40
40
40
20
20
20
0
0 100
101
102
103
104
100
Jurkat
80
0 101
102
103
104
100
101
102
103
104
Fluorescence intensity
Figure 6.14.1 CEM and Jurkat cells were stained with control murine IgG1 (filled histograms) or murine monoclonal anti-CD132 (open histograms). CEM cells were processed for flow cytometric analysis by standard amplification (indirect staining) or by EAS. Jurkat cells were processed by EAS only.
only minimal overlap. For heterogeneous populations, bimodal distributions (peaktrough-peak) demonstrate a definitive staining pattern but broad unimodal distributions may or may not be accurate. BASIC PROTOCOL
EAS FOR CELL SURFACE MOLECULES If the standard staining procedures do not result in a definitive staining pattern, EAS can be used to provide high-resolution immunophenotyping. After the primary antibody binds to the antigen of interest, the secondary reagent is used to specifically tag the cells with horseradish peroxidase. With the specially formulated amplification medium, this enzyme catalyzes the deposition of biotinylated tyramide onto the cells specifically tagged with the enzyme. Streptavidin-FITC binds to biotinylated tyramide and cellular fluorescence is analyzed on a flow cytometer. Materials Cells in a single-cell suspension in phosphate buffered saline (PBS), pH 7.2 to 7.4 (APPENDIX 2A for PBS) Staining buffer: PBS, pH 7.2 to 7.4, with 2% fetal bovine serum, sterile filtered Unconjugated, biotinylated, or fluorescein isothiocyanate (FITC)-conjugated antibody specific for the antigen of interest in PBS, pH 7.2 to 7.4, with 2% fetal bovine serum Isotype/subtype control immunoglobulin unconjugated, biotinylated, or FITC-conjugated in PBS, pH 7.2 to 7.4, with 2% fetal bovine serum EAS kit for unconjugated, biotinylated, or FITC-conjugated primary antibodies (Flow-Amp Systems) containing horseradish peroxidase–conjugated secondary reagents (anti-immunoglobulin antibody, streptavidin, or anti-FITC antibody), amplification medium, amplification reagent, and streptavidin-FITC
Enzymatic Amplification Staining for Cell Surface Antigens
12 × 75–mm polystyrene round-bottom test tubes Flow cytometer with 488-nm excitation and filters for collection of green fluorescence
6.14.2 Supplement 23
Current Protocols in Cytometry
Perform primary antibody labeling 1. Suspend 0.5-1.0 × 106 cells in 50 µl staining buffer in 12 × 75–mm polystyrene tubes. The cells must be in a single-cell suspension so that they can be run in the flow cytometer. Many companies recommend lysing buffers to remove erythrocytes. Most of these lysing buffers contain detergents that permeabilize the cells and consequently they should not be used in this procedure. Erythrocytes can be removed by Ficoll/hypaque discontinuous gradient centrifugation, or alternatively they can be lysed with VitaLyse (BioErgonomics), which does not contain detergent. Other means of removing erythrocytes without permeabilizing nucleated cells, such as ammonium chloride lysis or QuickLysis (Cytognos), have not been assessed with this procedure but are likely to be compatible with EAS. The EAS procedure has not been successfully accomplished in the presence of erythrocytes.
2. Add the appropriate amount of primary antibody to every tube but the negative control. To this tube add the equivalent amount of isotype/subtype control immunoglobulin. Incubate all tubes 10 min at room temperature. Concentrations for each antibody must be determined empirically for the EAS procedure. Optimal concentrations for standard staining procedures may not be appropriate for the EAS method. In many circumstances, the optimal concentration for the EAS procedure is less than the optimal concentration for standard staining techniques. In other situations, the optimal concentrations are identical for both procedures. Instead of an isotype/subtype control immunoglobulin group, a negative cell can be used as the control. For instance, for the detection of a viral glycoprotein, uninfected cells can be used as control for infected cells. For detection of a protein expressed via transfection, nontransfected parental cells can be used as the negative control. Use of both immunoglobulin and cellular negative controls is preferred. At least one negative control is necessary for each experiment. The incubation can be performed on ice for 20 min and azide may be included in the staining buffer. These maneuvers may be used for surface antigens whose expression may be modulated by immunoglobulin binding.
3. After the incubation, wash cells by adding 2 ml staining buffer, vortex, and centrifuge cells 6 min at 200 × g, at room temperature. Remove the supernatant. Wash cells one additional time with 2 ml staining buffer and remove the supernatant. Perform enzymatic amplification 4. Incubate cells 10 min at room temperature with the recommended amount of the secondary reagent from the Flow-Amp EAS kit in 50 µl staining buffer. The incubation can be performed 20 min on ice and azide may be included in the staining buffer. These maneuvers may be used for surface antigens whose expression may be modulated by immunoglobulin binding.
5. Wash three or four times as described in step 3. 6. Incubate cells 10 min at room temperature with the recommended amount of amplification reagent from the Flow-Amp EAS kit in 50 µl amplification buffer from the kit. This step requires enzymatic activity and consequently will not proceed in the cold. The time can vary. Shorter times can be used to sharpen and more clearly delineate peaks without significantly decreasing the level of amplification.
7. Wash two times as described in step 3.
Phenotypic Analysis
6.14.3 Current Protocols in Cytometry
Supplement 23
8. Incubate cells 10 min at room temperature with the recommended amount of streptavidin-FITC from the Flow-Amp EAS kit in 50 µl staining buffer. Streptavidin conjugated with other fluorochromes can be substituted for streptavidin-FITC in this step. An appropriate concentration must be empirically determined. Extravidin conjugates (Sigma) have also been used successfully. Formulations of some streptavidin conjugates (e.g., from BD Pharmingen) are not compatible with EAS (although they are successful with standard staining procedures) and should not be substituted in this procedure.
9. Wash two times as described in step 3. Acquire and analyze cellular fluorescent measurements 10. Collect forward-scatter and side-scatter measurements for the cells on a flow cytometer with a 488-nm laser and gate on the appropriate subpopulation. 11. Collect green fluorescence at 525 nm on a log or linear scale for 5000 to 20,000 gated cells. Data for more cells can be obtained if the proportion of positive cells in the gated subpopulation is low.
12. Assess specific fluorescence by comparing measurements for specifically stained cells with those for cells stained with the isotype/subtype matched control immunoglobulin. Comparison can be made with peak or mean channel numbers. It is essential that the flow cytometric instrument settings be kept the same for cells stained with specific primary antibodies and cells stained with control immunoglobulin. Positive cells are represented by events that possess more fluorescence than events collected from the negative control population. Small shifts of a few channels are not convincing because they may represent variation in the technique. The most stringent and defensible criterion for positivity is minimal overlap (<10%) with negative control events. For analysis in log mode, when the negative control population is in the first decade (mean channel number 5), minimally positive cells would be observed in the second decade (mean channel number >20). Shoulders of cells with increased fluorescence intensity represent incomplete separation of negative cells and potentially positive cells and cannot be used as evidence of positivity. Smears of cells can represent true heterogeneity of expression. In this case, mean or peak channel numbers are not that useful, but minimal overlap can be used as a criterion for determining positivity. ALTERNATE PROTOCOL 1
MULTICOLOR STAINING WITH AMPLIFICATION OF A SINGLE MARKER WITH EAS A major advantage of flow cytometric analysis is the potential to assess multiple parameters on single cells. Commonly used flow cytometers can accommodate the assessment of at least three fluorescent markers on each cell, and specially designed instruments can simultaneously analyze single cells for ten or more fluorescent molecules. EAS is compatible with multicolor staining (Kaplan et al., 2001a,b,c). For this procedure, it is convenient to use antibodies to various cell surface molecules that are directly labeled with fluorescent moieties. In that way, anti-immunoglobulins, which could detect more than one of the antibodies, would not be used. Alternatively, it is possible to use antibodies of different species and develop the stains with species-specific anti-immunoglobulins.
Enzymatic Amplification Staining for Cell Surface Antigens
Additional Materials (also see Basic Protocol) Monoclonal antibodies directly conjugated with fluorochromes Filters for collection of fluorescence emission from fluorophores used 1. Follow the Basic Protocol, steps 1 through 7.
6.14.4 Supplement 23
Current Protocols in Cytometry
2. Incubate cells 10 min at room temperature with the recommended amount of streptavidin-FITC from the Flow-Amp EAS kit and simultaneously with the recommended amounts of monoclonal antibodies specific for informative cell surface antigens (the antibodies are directly conjugated with unique fluorophores). Because the marker to be analyzed uses FITC, the other antibodies should not be conjugated with FITC; however, if it is desired to reserve FITC for another marker, the molecule that needs to be amplified can be developed with the addition of a different fluorochrome conjugate of streptavidin, such as streptavidin-PE-Cy5 or streptavidin-APC. The use of alternative streptavidin conjugates in this procedure requires prior titration for optimal results. Incubation conditions for alternative streptavidin conjugates are identical to the conditions for the incubation of streptavidin-FITC.
3. Wash two times as described in Basic Protocol, step 3. 4. Run on a flow cytometer with excitation at 488 nm. Collect fluorescence at 525 nm (green), 575 nm (orange), 675 nm (red), and/or additional wavelengths as required by the choice of fluorochromes. Analysis of the results with multicolor staining should include setting compensation levels (UNIT 1.14). Compensation levels need to be adjusted based on the added fluorescence intensity that is obtained with EAS. Setting compensation levels with controls obtained via standard staining will usually result in under-compensation for samples treated by EAS.
MULTIPLE AMPLIFICATIONS WITH EAS ON A SINGLE CELL POPULATION
ALTERNATE PROTOCOL 2
It is possible to perform multiple amplifications on the same cell population. If it is desirable to amplify the signal from two different molecules expressed on cells in the same population, EAS can be used serially but not simultaneously. Adding a peroxide incubation between each incubation, the investigator can perform multiple amplifications via the EAS procedure; however, the number of amplifications is limited by the cell loss that occurs with each wash. Two amplifications are readily performed, three amplifications are possible, and four are difficult. Replicate tubes can be pooled to provide appropriate cell numbers for analysis. Additional Materials (also see Basic Protocol) 1% hydrogen peroxide in PBS 1. Perform Basic Protocol, steps 1 through 8, to amplify the signal associated with one of the molecules. 2. Before proceeding to amplify the signal associated with the second molecule, incubate cells 10 min at room temperature in 1 ml of 1% hydrogen peroxide in PBS. 3. After an additional wash, repeat Basic Protocol, steps 1 to 8, with a second antibody. Be sure to use a different streptavidin conjugate for the second amplification. COMMENTARY Background Information Many cell surface molecules are expressed in quantities that are too low to be definitively detected by standard staining procedures. These molecules include cytokine receptors, chemokine receptors, apoptosis-inducing ligands, other surface-bound ligands, and viral
glycoproteins; moreover, lineage markers expressed in abundance in one cell may be expressed in low levels on other cells (Kaplan et al., 2001b). Small particles represent another difficulty. Since fluorescence is read cumulatively per event, the threshold for detection for small events is more stringent than for larger
Phenotypic Analysis
6.14.5 Current Protocols in Cytometry
Supplement 23
600 EAS indirect staining
Mean fluorescence intensity
500 400 300 200 100 0 0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Human IgG conjugated to beads (µg) Figure 6.14.2 Carboxylated polystyrene spheres were conjugated with various amounts of human IgG and processed for flow cytometric analysis by either indirect staining or EAS. The mean fluorescence intensity of the unimodal histograms obtained was determined by flow cytometric analysis.
Enzymatic Amplification Staining for Cell Surface Antigens
events. Thus, a molecule expressed in small and large cells with the same density is more difficult to detect on the small cell. The EAS method is a new technology that significantly amplifies the fluorescent signal for flow cytometric analysis. It is based on the catalyzed deposition of a tagged molecule. In the presence of peroxide and peroxidase, oxygen radicals are produced and cause the production of a radical form of another molecule, tyramide. When tyramide is radicalized, it becomes highly reactive and covalently binds with nearby molecules. The tyramide radical is short-lived, so that its reactivity is proximityregulated by the enzyme. Tyramide can be derivatized (Bobrow et al., 1989; Bobrow et al., 1991). The EAS method uses a biotinylated form of the tyramide molecule. EAS is qualitatively different from including additional layers of binding molecules, since background levels of fluorescence are not increased along with the specific signal. EAS can amplify the separation of the fluorescent signals between specific and control treatments for the detection of cell surface molecules by as much as 100-fold.
Using EAS technology, the author has published several interesting findings. EAS has been used to provide a higher-resolution phenotype of leukemic cells from patient samples (Kaplan et al., 2001a,b) and to demonstrate that all B lymphocytes express CD5 constitutively in a unimodal fashion, indicating that the absence/presence of CD5 expression can no longer be used to distinguish subsets among B cells (Kaplan et al., 2001c). Results obtained with EAS kits have also been published in studies of human immunodeficiency virus gp120 expression (Kaplan, 2002) and of cytokine receptors (Kaplan et al., 2001b). Other scientists have also published results with this technology (Bernard et al., 2001; Horner et al., 2001; Jabara et al., 2001; Kennedy et al., 2001; Xinhluo et al., 2001). Staining obtained with EAS has been validated to be functionally significant (Kaplan and Smith, 2000). Peripheral-blood mononuclear cells were induced to express Fas ligand functional activity by stimulation with a lectin and a cytokine followed by a phorbol ester and a calcium ionophore. Although after this extensive induction procedure these cells possessed Fas ligand functional activity, the presence of
6.14.6 Supplement 23
Current Protocols in Cytometry
Fas ligand could not be detected by indirect staining and flow cytometric analysis. The cells were also stained for Fas ligand expression and processed by the EAS technique. With EAS processing, 10% to 15% of the lymphocytic cells expressed enough Fas ligand to define a clearly separate positive subpopulation of cells. It should be noted that the only difference between the indirect staining procedure and EAS was the inclusion of the enzymatic amplification steps in the EAS method. To ascertain that the staining observed with EAS was functionally significant, the positive and negative cells were sorted and their cytotoxic potential against Fas-expressing target cells assessed. The Fas ligand–positive cells identified by EAS possessed almost all the activity observed whereas the Fas ligand–negative cells exhibited little cytotoxic activity. Staining of cell surface molecules by EAS has also been verified by correlating annexin V binding to apoptosis or cell death and by correlating expression with the presence of the correct-size band after electrophoresis, transfer to a filter, and immunoblotting (Kaplan and Smith, 2000). The staining that is obtained by EAS is approximately linear even at analyte concentrations below the level of detection by standard staining procedures. The author covalently conjugated variable amounts of human IgG to carboxylated polystyrene microparticles. The conjugated microspheres were stained with anti-human IgG and then processed either by indirect staining or by EAS. The results demonstrate that the mean fluorescence intensity of the staining obtained by EAS at concentrations of bead-conjugated human IgG below the threshold of detection by the indirect staining method was a linear correlate to the amount of analyte conjugated (Fig. 6.14.2). The linearity of the response at low amounts of a specific molecule further indicates that EAS is a useful and reliable technology for analysis of analytes at these low levels of expression.
Critical Parameters Several important considerations should be included in planning experiments involving staining by the EAS technique. First, the concentration of the primary antibody is important. For some antigens, high concentrations can catalyze the deposition of so much flurochrome that quenching can occur. It is necessary to determine the optimal concentration using the EAS procedure. Optimal concentrations obtained with standard staining may not translate well. With an antibody that has not been pre-
viously tested, the range of recommended antibody concentrations is 0.2 to 20 µg/ml. Tenfold dilutions within this range can be used to find a concentration that gives good peak separation between control and specific stains, and then fine-tuning can be used to optimize the distinction. Second, the addition of sodium azide to the diluent is not absolutely forbidden. Azide inhibits the enzymatic action of peroxidase and it should be thoroughly removed before proceeding to the amplification step. The addition of azide to the incubations of primary and secondary antibodies and the use of low-temperature (4°C) incubations are acceptable. For some antigens, azide and low temperature incubations may be desired to prevent modulation, but for most antigens, the absence of azide and room-temperature incubations do not present a problem. Modulation would be observed as a failure to obtain specific staining. Third, wash volumes after the incubation with the peroxidase-containing reagent and before the amplification step are crucial. Any residual peroxidase, even if it is in the fluid phase and not bound to cells, will catalyze the deposition of label and lead to inappropriate staining. A total wash volume of 6 to 8 ml is recommended. Staining cells in microtiter plates does not allow for large enough wash volumes at this step and is consequently not recommended. Fourth, endogenous peroxidases are theoretically a problem with the EAS procedure; however, several considerations tend to mitigate this concern. Peroxidases have not been found to be expressed on cell surfaces, but instead localize in secretory granules, endoplasmic reticulum, and nuclear envelope (DeSombre et al., 1975; Deimann, 1984). Moreover, the formulations of the components in the EAS kits include a peroxidase inhibitor. Nevertheless, if inappropriate endogenous peroxidase activity is suspected, preincubation of the cells with 0.3% hydrogen peroxide in staining buffer with azide (0.1%) for 10 min at room temperature irreversibly inhibits peroxidase activity while leaving antigens unaffected (Li et al., 1987). Fifth, with the EAS technique, there is the possibility of bystander staining–transfer of stain from true positive cells to true negative cells. This phenomenon is more likely to be significant with excessively bright positive cells and/or with small proportions of negative cells. Since EAS has been designed for antigens that are not abundant and therefore do not give
Phenotypic Analysis
6.14.7 Current Protocols in Cytometry
Supplement 23
cells incubated 1 hr at 37°C
cells ex vivo 100
100
gated on CD3+ cells 0 100
101
102
103
0 100
104
101
102
103
104
100
100
Number of events
gated on CD19+ cells 0 100
101
102
103
0 100
104
101
102
103
104
100
100
gated on CD16+ cells 0 100
101
102
103
0 100
104
101
102
103
104
100
100
gated on CD56+ cells 0 100
101
102
103
104
0 100
101
102
103
104
Fluorescence intensity
Figure 6.14.3 Human peripheral blood mononuclear cells were stained ex vivo with anti-human IgM (filled histograms) or without anti-human IgM (open histograms), and the cells were processed by enzymatic amplification staining (EAS). The stained cells were analyzed for IgM expression by flow cytometry. Some of the peripheral blood mononuclear cells were incubated 1 hr in complete medium at 37°C to allow for exogenously bound IgM to be cleared from the cell surface. These cultured cells were stained and processed in the same manner as the cells ex vivo. The cells were also double-stained with PE-conjugated antibodies specific for the various lineages of mononuclear cells in the blood and were gated as indicated. The results show that T cells passively absorb some IgM which is lost after incubation. B cells express endogenously produced IgM and natural killer cells bind exogenous IgM via Fc receptors but do not produce it.
Enzymatic Amplification Staining for Cell Surface Antigens
excessive bright signals, this problem is usually not seen. Nevertheless, in the event that the negative subpopulation is being “dragged” into the positive region, the parameters of the reaction can be altered to decrease the staining intensity. Decreasing the amplification time and decreasing the concentration of primary antibody are likely to eliminate bystander staining. Alternatively, preparative cell separation to isolate the positive and negative cells can be used. Weak positivity of a small subpopulation of cells in the face of bright positivity of a large
subpopulation of cells will require cell separation to ascertain the staining status of the small subpopulation. Finally, it is critical in establishing the EAS procedure to work from positive results. Start by enzymatically amplifying a signal on cells that give a weak positive stain by standard staining technology. The cells that stain positive by the standard procedure give assurance that the cells being assessed actually express the molecule and that the antibody is capable of binding to that molecule.
6.14.8 Supplement 23
Current Protocols in Cytometry
Troubleshooting
Anticipated Results
If the results obtained have a high level of background staining, there are several possible ways to address this situation. These experimental maneuvers tend to enhance the specific staining by decreasing levels of background. First, decrease the amount of primary antibody used and/or increase the concentration of serum in the diluent. Second, decrease the concentration of the peroxidase-containing reagent. Third, increase the number of washes after incubation of the cells with the peroxidase-containing reagent. Fourth, decrease the time of the amplification step. Fifth, wash the cells more thoroughly prior to the procedure or incubate the cells 30 to 60 min at 37°C to eliminate molecules passively adsorbed on the cell surface. Sixth, to eliminate inappropriate binding through Fc receptors, use F(ab′)2 primary antibodies or block Fc receptor binding with unlabeled immunoglobulin or inhibitory antibodies to specific receptors. Incubations of 30 to 60 min at 37°C prior to staining can also be used to distinguish between endogenous synthesis and binding to Fc receptors (Fig. 6.14.3). Seventh, substitute another streptavidin fluorochrome conjugate for the one supplied. Low signals obtained with the EAS procedure can be addressed by first decreasing the serum concentration in the diluent and/or increasing the concentration of the primary antibody. Second, increase the concentration of the peroxidase-containing reagent. Third, increase the time of the amplification step. Fourth, substitute streptavidin conjugated to a more potent fluorochrome (such as APC or PE) for streptavidin-FITC. Fifth, substitute a different form of the primary antibody. The EAS procedure is compatible with unconjugated, FITC-conjugated, and biotinylated primary antibodies. If the results obtained with the EAS technique fail to clearly delineate peaks of different fluorescent intensity, it is possible that bystander staining is occurring. Bystander staining is the potential for tyramide deposition being catalyzed by enzyme on a positive cell to occur on neighboring negative cells. This possibility is most likely when the population of cells being analyzed includes a large proportion of highly positive cells and a low proportion of negative cells. To address this possibility, try the first four steps for decreasing background staining (see above). The most likely maneuvers to help this situation are decreasing primary antibody concentration and/or decreasing the time of the amplification step.
The EAS procedure gives 10- to 100-fold enhancement in the specific fluorescent signal compared to standard staining procedures. The specific fluorescent signal is defined as the difference between the mean channel numbers of histograms obtained from cells stained with specific antibody compared to isotype/subtype control immunoglobulin. Alternatively, it refers to the difference between the mean channel numbers of histograms obtained from cells expressing the targeted molecule and cells not expressing this molecule, stained with specific antibody. The presence of molecules expressed at levels too low to be detected by standard staining procedures can be detected by EAS. Distinct populations and subpopulations of positive cells can be delineated with the EAS procedure.
Time Considerations Once the cells are ready, the EAS staining procedure for 10 to 30 tubes takes 2 to 3 hr. For 30 to 50 tubes, the procedure requires 3 to 4 hr. These estimates do not include the time needed for flow cytometric analysis.
Literature Cited Bernard, J., Treton, D., Vermot-Desroches, C., Boden, C., Horellou, P., Angevin, E., Galanaud, P., Wijdenes, J., and Richard, Y. 2001. Expression of interleukin 13 receptor in glioma and renal carcinoma: IL13α2 as a decoy receptor for IL13. Lab. Invest. 81:1223-1231. Bobrow, M., Harris, T., Shaughnessy, K., and Litt, G. 1989. Catalyzed reporter deposition, a novel method of signal amplification. Application to immunoassays. J. Immunol. Methods 125:279285. Bobrow, M., Shaughnessy, K., and Litt, G. 1991. Catalyzed reporter deposition, a novel method of signal amplification. II. Application to membrane immunoassays. J. Immunol. Methods 137:103-112. Deimann, W. 1984. Endogenous peroxidase activity in mononuclear phagocytes. Prog. Histochem. Cytochem. 15:1-58. DeSombre, E., Anderson, W., and Kang, Y. 1975. Identification, subcellular localization, and estrogen regulation of peroxidase in 7,12-dimethylbenz(a)anthracene-induced rat mammary tumors. Cancer Res. 35:172-179. Horner, A., Widhopf, G., Burger, J., Takabayashi, K., Cinman, N., Ronaghy, A., Spiegelberg, H., and Raz, E. 2001. Immunostimulatory DNA inhibits IL-4-dependent IgE synthesis by human B cells. J. Allergy Clin. Immunol. 108:417-423. Phenotypic Analysis
6.14.9 Current Protocols in Cytometry
Supplement 23
Jabara, H., Broder, S., and Geha, R. 2001. Glucocorticoids upregulate CD40 ligand expression and induce CD40L-dependent immunoglobulin isotype switching. J. Clin. Invest. 107:371-378. Kaplan, D. and Smith, D. 2000. Enzymatic amplification staining for flow cytometric analysis of cell surface molecules. Cytometry 40:81-85. Kaplan, D. 2002. Flow cytometric analysis of cells from persons with HIV-1 disease by enzymatic amplification staining. In Cellular Aspects of HIV Infection (A. Cossarizza and D. Kaplan, eds.) pp. 351-369. John Wiley & Sons, New York. Kaplan, D., Meyerson, H., and Lewandowska, K. 2001a. High resolution immunophenotypic analysis of chronic lymphocytic leukemia cells by enzymatic amplification staining. Amer. J. Clin. Pathol. 116:429-436. Kaplan, D., Husel, W., and Meyerson, H. 2001b. Immunophenotypic analysis with enhanced sensitivity of detection by enzymatic amplification staining. Clin. Lab. Med. 21:763-778. Kaplan, D., Smith, D., Meyerson, H., Pecora, N., and Lewandowska, K. 2001c. CD5 expression by B lymphocytes and its regulation upon Epstein-Barr virus transformation. Proc. Natl. Acad. Sci. U.S.A. 98:13850-13853.
Kennedy, N., Russell, J., Michail, N., and Budd, R. 2001. Liver damage by infiltrating CD8+ cells is Fas dependent. J. Immunol. 167:6654-6662. Li, C., Ziesmer, S., and Lazcano-Villareal, O. 1987. Use of azide and hydrogen peroxide as an inhibitor for endogenous peroxidase in the immunoperoxidase method. J. Histochem. Cytochem. 35:1457-1460. Xinhluo, L., Bai, X., Wen, J., Gao, J., Liu, J., Lu, P., Wang, Y., Zheng, P., and Liu, Y. 2001. B7H costimulates clonal expansion of, and cognate destruction of tumor cells by, CD8+ T lymphocytes in vivo. J. Exp. Med. 194:1339-1348.
Key Reference Kaplan, D. and Smith, D. 2000. See above. Validation of the EAS procedure.
Contributed by David Kaplan Case Western Reserve University Cleveland, Ohio
Enzymatic Amplification Staining for Cell Surface Antigens
6.14.10 Supplement 23
Current Protocols in Cytometry
Whole Blood Analysis of Leukocyte-Platelet Aggregates
UNIT 6.15
Platelets play a critical role in the initial reaction to bleeding by rapidly forming a plug at sites of vascular injury. In inflammatory and thrombotic syndromes, platelets aggregate with circulating leukocytes, especially monocytes and neutrophils (Michelson et al., 2002). The binding of platelets to monocytes and neutrophils is initiated primarily through platelet surface expression of P-selectin (CD62P) following activation-dependent degranulation (McEver, 2002). P-selectin in turn binds to PSGL-1 (P-selectin glycoprotein ligand-1, also known as CD162), which is constitutively expressed on the leukocyte surface. Leukocyte-platelet aggregates (LPAs) are therefore formed, and these are stabilized by several mechanisms, including the binding of leukocyte Mac-1 (CD11b/CD18, integrin αMβ2) and platelet glycoprotein (GP) Ibα (McEver, 2002). Circulating monocyte-platelet aggregates have been demonstrated to be a more sensitive marker than platelet surface P-selectin for in vivo platelet activation (Michelson et al., 2001). Although the cellular interactions involved are complex, detection of LPAs by whole blood flow cytometry is relatively simple. While direct measurement of low-level platelet P-selectin expression is often difficult, even low-level P-selectin expression can initiate leukocyte binding. The presence of platelets associated with leukocytes is then easily detected using immunostaining for abundant platelet-specific markers. Flow cytometric analysis of LPAs involves gating on the desired leukocyte population through light scatter and at least one leukocyte-specific reagent such as CD14 to identify monocytes (bright) and neutrophils (dim). A second antibody that is platelet specific, typically CD41 (GPIIb), CD61 (GPIIIa), or CD42a (GPIX), will distinguish distinct platelet-positive and platelet-negative leukocyte subpopulations due to the high copy number of CD41, CD61, and CD42a on platelets. Platelets are quite heterogeneous in size (1 to 5 µm in diameter) and platelet-derived microparticles (PDMP) also express P-selectin, which may explain the appearance of dim platelet-positive leukocytes whose fluorescence is just above that of the isotypic negative control antibody. LPAs of this type should not be overlooked, as circulating PDMP are often elevated in clinically relevant disease states (Michelson et al., 2002; Nieuwland and Sturk, 2002). Review of UNIT 6.10 is strongly recommended, as many of the important considerations outlined therein are also critical in LPA analysis. For example, both choice of anticoagulant and blood sample handling can affect the measurement of LPAs. Expedient processing is also important due to the sensitivity of the assay and a high likelihood of in vitro platelet activation, resulting in an artifactually high number of LPAs. WHOLE BLOOD ANALYSIS OF LEUKOCYTE-PLATELET AGGREGATES Monocytes and neutrophils can be distinguished based on linear forward versus side light-scatter properties, but identification of gated populations is improved by using a leukocyte-specific antibody. Platelet-platelet aggregates in the sample can also interfere with light-scatter gates, particularly with the monocyte population. For these reasons, it is recommended that a monoclonal anti-leukocyte reagent be used in addition to the anti-platelet monoclonal antibody. These antibodies must be titrated by each laboratory for determination of the optimal concentration and incubation times. CD14 is a common choice for identifying monocytes, although it can be shed upon activation and may be dim on immature cells. CD45 negativity is effective for eliminating non-leukocyte-associated platelets, but care must be exercised to avoid basophil or large granular lymphocyte (LGL) inclusion in the monocyte gate. LGLs have light scatter similar to that of monocytes, but
BASIC PROTOCOL
Phenotypic Analysis Contributed by Marc R. Barnard, Lori A. Krueger, A.L. Frelinger III, Mark I. Furman, and Alan D. Michelson
6.15.1
Current Protocols in Cytometry (2003) 6.15.1-6.15.8 Copyright © 2003 by John Wiley & Sons, Inc.
Supplement 24
stain more brightly for CD45. Basophils have CD45 staining resembling that of monocytes, but have light scatter like lymphocytes. Therefore, a combination of side light scatter and leukocyte-specific labeling provides the best monocyte LPA gating. Neutrophils are less problematic to gate thanks to their high side light scatter, which should be used in conjunction with the leukocyte marker. Forward light scatter is not useful for leukocyte gating due to the large potential increase in this signal when platelets are bound, particularly when platelets are highly activated. The platelet GPIIb-IIIa complex (CD41/CD61, integrin αIIbβ3) is often used as the platelet-specific antibody target because of the high copy number of this receptor (80,000/platelet). Clones directed against this complex should be screened with caution, since CD41/61 is an important platelet receptor capable of binding a wide variety of adhesive glycoprotein ligands (Hato et al., 2002), which may influence normal LPA development and stability. It should also be noted that CD42b is susceptible to proteolysis, and both CD42a and CD42b down-regulate with platelet activation (LaRosa et al., 1994). Phycoerythrin is the fluorophore of choice for the platelet-specific antibody because of its high signal and reduced neutrophil autofluorescence compared to fluorescein. Neutrophil isotypic background fluorescence (which includes autofluorescence) can be very high in the FITC channel, often resulting in incorrect placement of the platelet-positive LPA histogram marker. Materials Platelet-specific antibody: anti-CD42a, -CD42b, -CD41, -CD61 (one or more of these antigens may be missing in rare inherited disorders, e.g., Bernard-Soulier syndrome and Glanzmann thrombasthenia; see UNIT 6.10) Leukocyte-specific antibody: monoclonal anti-CD14, -CD64, -CD33, or -CD45 (this reagent is often diluted in an antibody cocktail with the platelet-specific antibody) Negative control: an isotype-, fluorophore-, concentration-, and F:P ratio-matched non-specific antibody for background fluorescence determination (to be used in conjunction with the platelet-specific antibody) Platelet agonist: such as ADP or thrombin receptor-activating peptide (TRAP; see recipes) Modified HEPES/Tyrode’s (HT) buffer (see recipe) Source of human whole blood (patient or donor) Fix/lyse solution: any red-cell lysis reagent optimized for flow cytometry and containing formaldehyde, e.g., FACS lysing solution, Becton Dickinson, or Optilyse, Beckman Coulter (the performance of these products must be assessed by the individual laboratory) Blood collection tubes containing buffered sodium citrate or other suitable anticoagulant (see UNIT 6.10) 21-G (or larger bore) needles Flow cytometer with 488-nm excitation and suitable filters for collection of fluorophore emissions NOTE: Individual laboratories must determine optimal assay conditions and sample stability. 1. Prior to drawing blood, label assay tubes and add 20 µl platelet-specific antibody (or negative control antibody), 20 µl leukocyte-specific antibody, and 20 µl platelet agonist (or HT buffer, for an unstimulated sample) to each tube for a total of 60 µl. Whole Blood Analysis of Leukocyte-Platelet Aggregates
6.15.2 Supplement 24
Current Protocols in Cytometry
2. Draw blood directly into blood collection tube containing buffered sodium citrate or other suitable anticoagulant using a 21-G (or larger bore) needle and a light tourniquet. Ensure a smooth free-flowing blood draw and discard the first 2 to 5 ml drawn. 3. Within 20 min of drawing blood, add 20 µl whole blood to each tube prepared in step 1 and gently mix. Incubate undisturbed 15 min at room temperature. Higher concentrations of whole blood may be used if care is taken to avoid platelet aggregation, which can interfere with light scatter gates.
4. Add 800 µl fix/lyse solution and incubate ∼10 min at room temperature to allow red cell lysis. Store samples at 4°C. 5. Analyze by flow cytometry using the following setup: a. Run samples on medium flow setting (see Critical Parameters and Troubleshooting). Due to the high number of platelets in the unwashed sample, leukocyte-platelet coincident events will increase if the sample is run at high flow rates. Collect light-scatter data in linear mode, and set the threshold on forward light scatter. Collect data on a minimum of 1000 CD14-positive monocytes, which constitute 2% to 15% of all leukocytes. b. Create a dot plot of side light scatter versus log CD14 (or other leukocyte identifier) fluorescence (Fig. 6.15.1A). c. Create regions (gates) containing the populations of interest (e.g., monocytes and neutrophils). d. Prepare gated histograms for each population of interest. See Figure 6.15.1B and 6.15.1C—the arrows indicate the gate designation. e. Set the positive marker just above the isotype-negative control peak on an unstimulated sample (Figs. 6.15.1B and 6.15.1C dotted lines). f. Run unstimulated (buffer) and platelet agonist-stimulated samples. Figure 6.15.1 shows TRAP-stimulated cells. g. Record percent of monocytes or neutrophils positive for platelet-specific antigen staining. WHOLE BLOOD ANALYSIS OF LEUKOCYTE-PLATELET AGGREGATES IN PRE-FIXED SAMPLES
ALTERNATE PROTOCOL
Although fresh whole blood antibody labeling generally yields the best results, there are situations where it is desirable to fix prior to labeling. Fixation allows more lag time between blood collection and subsequent sample processing. If a number of time points are clustered together during a procedure, fixation may be the only practical approach. Possible concerns about the interaction of drugs present in a patient sample in combination with the monoclonal antibodies being used can be avoided by pre-fixation. Also, platelets are very sensitive to spontaneous activation in vitro, and fixation is a convenient method to “freeze” a sample in a state more reflective of the in vivo activation level. Many antibodies label with less intensity or even fail to bind after formaldehyde fixation. Therefore, individual laboratories should titrate antibody reagents specifically for fixed specimens. For materials, see Basic Protocol. 1. Within 20 min of blood draw, fix 200 µl anticoagulated whole blood in a 5-ml tube with 1 ml fix/lyse solution (e.g., FACS lysing solution from Becton Dickinson) 10 min at room temperature. Samples may be activated prior to fixation to study platelet-leukocyte reactivity to agonists.
Phenotypic Analysis
6.15.3 Current Protocols in Cytometry
Supplement 24
B monocyte-platelet aggregates 40
Counts
30
A 104
20
10
anti-CD14-FITC
monocytes 103
0 100
101
neutrophils
102
101
C
100 0
200
400 600 Side scatter
800
102 CD41-PE
103
104
neutrophil-platelet aggregates
250
1000
Counts
200 150 100 50 0 100
101
102 CD41-PE
103
104
Figure 6.15.1. Whole blood analysis of leukocyte-platelet aggregates in a normal donor after activation with 20 µM thrombin receptor-activating peptide (TRAP). (A) Leukocytes are labeled with anti-CD14-FITC and plotted versus side light scatter. The threshold is set on forward light scatter (not shown). (B) A gated histogram of CD14-bright monocytes is plotted and a marker is placed just above the negative control antibody peak (dotted lines). The platelet-positive monocytes stain brightly with anti-CD41-PE (filled histogram). (C) A gated histogram of high side-scatter, CD14-dim neutrophils is plotted and analyzed in the same manner as the monocytes. Reproduced with permission from Figure 20-2 in Platelets (A.D. Michelson, ed.), Academic Press/Elsevier Science, New York.
2. Dilute by adding 3 ml HT buffer and store at 4°C. Undiluted fixed samples can also be stored at 4°C and diluted just prior to step 3. Antibody staining may decrease with prolonged fixation.
3. Centrifuge all samples to be labeled 5 min at 250 to 400 × g, room temperature. Remove all supernatant and resuspend cells in 400 µl HT buffer with gentle vortexing. 4. In separate tubes, label 40 µl of the above suspension with 20 µl leukocyte-specific antibody and 20 µl platelet-specific antibody or isotype control reagent. Incubate 15 min at room temperature. Whole Blood Analysis of Leukocyte-Platelet Aggregates
5. Add 800 µl fix/lyse solution and store at 4°C. 6. Analyze by flow cytometry using the setup outlined in Basic Protocol, step 5.
6.15.4 Supplement 24
Current Protocols in Cytometry
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
GPRP, 10 mM Dilute GPRP (Gly-Pro-Arg-Pro) in modified HT buffer (see recipe) to a concentration of 10 mM. Store up to 1 week at 4°C or 1 year at –20°C. Bring to room temperature prior to use. Use for diluting thrombin and in assays containing CaCl2 to prevent fibrin polymerization and clot formation (Michelson, 1994).
Modified HEPES/Tyrode’s (HT) buffer 10 mM HEPES 137 mM NaCl 2.8 mM KCl 1 mM MgCl2 12 mM NaHCO3 0.4 mM Na2HPO4 0.35% bovine serum albumin 5.5 mM glucose Adjust pH to 7.4 with 0.1 M NaOH or 0.1 M HCl Store up to 1 week at 4°C or 1 year at –20°C Bring to room temperature prior to use The pH may need readjustment after storage at 4°C.
Platelet agonists Prepare all platelet agonist working solutions (concentrations determined empirically) by diluting stocks in modified HT buffer (see recipe), with or without 6 mM CaCl2 as appropriate, just prior to use. Discard left-over working solutions daily. ADP: Adenosine diphosphate (ADP) is typically used at concentrations of 20 µM (maximal platelet activation in diluted whole blood) to sub-maximal doses of 0.5 µM. Store stock ADP according to manufacturer’s instructions (often frozen at –20°C). Collagen: Collagen is typically used at concentrations of 10 to 20 µg/ml alone or in combination with sub-maximal doses of thrombin (see below). Dilute in modified HT buffer (see recipe) supplemented with 6 mM (2×) CaCl2. Use in the evaluation of procoagulant platelets and platelet-derived microparticles in the presence of 2.5 mM (final concentration) GPRP (see recipe for 10 mM). Store stock collagen according to manufacturer’s instructions. Epinephrine: Epinephrine is often used in combination with ADP at concentrations of 10 to 20 µM (maximal platelet activation in diluted whole blood). Store stock epinephrine according to manufacturer’s instructions. Human α-thrombin: Thrombin is typically used at concentrations of 1 to 2 U/ml (maximal platelet activation in diluted whole blood) to sub-maximal doses of 0.1 U/ml. The concentration of thrombin required to achieve maximal platelet activation in plasma-free systems is ∼0.1 U/ml. Dilute working concentration of thrombin in modified HT buffer (see recipe) containing sufficient 10 mM GPRP (see recipe for 10 mM) to give a final assay concentration of 2.5 mM. Store 200 U/ml stock concentration up to 6 months at –80°C or up to 1 week at 4°C. Phenotypic Analysis
6.15.5 Current Protocols in Cytometry
Supplement 24
Thrombin receptor-activating peptide (TRAP/SFLLRN): Dilute in modified HT buffer (see recipe). TRAP is typically used at concentrations of 20 to 50 µM (maximal platelet activation in diluted whole blood) to sub-maximal doses of 1.5 to 5 µM. Store frozen at –80°C or up to 1 week at 4°C . Bring to room temperature prior to use. COMMENTARY Background Information The role of platelets in the recruitment of monocytes and neutrophils to sites of vascular injury has long been recognized (Palabrica et al., 1992). The appearance of circulating LPAs is a dynamic process involving initial formation, vascular adhesion, potential sequestration to elements of the reticuloendothelial system, leukocyte activation, and LPA disaggregation via granulocyte proteinases (Gardiner et al., 2001). Circulating LPAs are increased in stable coronary artery disease (Furman et al., 1998), unstable angina (Ott et al., 1996), acute myocardial infarction (Furman et al., 2001; Michelson et al., 2001), chronic venous insufficiency (Powell et al., 1999), and during cardiopulmonary bypass (Rinder et al., 1992). Circulating LPAs are also increased after coronary angioplasty, with a greater magnitude in patients experiencing late clinical events (Mickelson et al., 1996).
Critical Parameters and Troubleshooting
Whole Blood Analysis of Leukocyte-Platelet Aggregates
A clean blood draw and gentle handling of specimens are required to avoid spontaneous platelet activation. If agonists are being added during processing to study cell reactivity, platelet aggregation could occur and interfere with analysis. It is important, therefore, that the samples be left undisturbed during any activation steps, followed immediately by fixation. Even in the absence of added agonist, LPA will form relatively quickly once blood is drawn into anticoagulant; therefore, the time to processing should be kept to a minimum (<20 min) to avoid artifactual LPA formation. Large platelet-platelet aggregates can exhibit high light scatter properties, and can interfere with analysis. These can generally be gated out, however, by taking advantage of their high platelet-specific and low leukocyte-specific fluorescence. EDTA-anticoagulated whole blood should never be used for LPA analysis, as the EDTA can cause in vitro dissociation of platelets from leukocytes. To minimize delays, all tubes should be labeled and reagents prepared prior to obtaining
the blood sample. Anticoagulants such as CTAD (citrate theophylline adenosine dipyridimole) or added platelet activation inhibitors, such as prostacyclin or apyrase, can minimize formation of spontaneous LPAs, but these methods will alter agonist responsiveness and may even result in dissaggregation of existing LPA. Samples should never be exposed to temperatures <15°C, as rewarming will result in platelet degranulation and LPA formation. The precise percent LPAs obtained using the Basic Protocol may differ depending on the antibodies chosen as leukocyte- and plateletspecific markers. CD14 was chosen here since it does not appear to play a role in leukocyteplatelet aggregation. The influence of leukocyte-specific antibodies on LPA formation can be determined by comparing LPAs found by gating on the leukocyte-specific antibody signal versus side light scatter to those found by gating on forward versus side light scatter in the absence of the leukocyte antibody. The monocyte population in particular will be different (lower purity) but differences due to the presence of antibody can be estimated. The platelet-specific antibody can also influence LPA development. For example, antibodies directed toward the Mac-1 (CD11b/18) binding region of platelet CD42b should be avoided as this interaction is important in the stabilization of LPAs (Simon et al., 2000). Blocking antibodies for P-selectin (CD62P) or its counterreceptor PSGL-1 (CD162) are useful for preparing baseline control samples as they are very effective in blocking formation of LPAs. These blocking agents will not dissociate stable LPAs, however, so these can not be considered true negative controls. Significant differences in the baseline LPA level in the presence or absence of blocking antibody may indicate an activating effect by the chosen leukocyte or platelet antibody. The pre-fixation method described in the Alternate Protocol is less susceptible to antibody influences, but sample fixation and washing can also modify the measurement of LPA. Fixation can reduce or eliminate binding of
6.15.6 Supplement 24
Current Protocols in Cytometry
some antibodies. As a general rule, anti-platelet antibodies, which label with greater intensity, will result in an increase in the apparent number of LPAs. This is particularly noticeable in the baseline sample, in which few platelets are bound per LPA. The pre-fixation technique requires washing to remove the fixative and concentrate the sample for labeling. Care must also be taken during wash steps since even fixed CD62P-positive platelets can bind leukocytes. The low G force described in the Alternate Protocol does not pellet most free platelets, so this effect is minimized in the method described. Monocytes are the time-limiting cell type during data acquisition; 2 to 7 min are needed to acquire 1000 to 2000 monocyte events. Samples should be run on a low to medium flow rate, depending on dilution. It is tempting to concentrate samples or increase flow rates; however, leukocyte-platelet coincident events (particularly monocyte-platelet events) are a distinct problem due to the large number of platelets in the unwashed whole blood samples described in this method. To determine the extent to which coincidence is contributing to increased leukocyte-platelet events, accumulate 1000 monocyte events at high, medium, and low flow settings with the same sample. Typically, the baseline (no added agonist) percent of monocytes that are platelet positive will be considerably increased for samples run at high flow rates. In some circumstances, a low flow rate should be used, particularly if the platelet count in the sample is abnormally high, but at the final blood dilution described in this protocol (1:44) medium flow is acceptable for most samples. Sheath pressures at medium and low flow settings determine absolute flow rate and core stream diameter and will differ from one instrument to another. Therefore, each laboratory should independently determine the optimal flow rate for a given instrument. Samples should be run as soon as possible after fixation, although LPAs labeled and subsequently fixed are quite stable up to 72 hours when stored at 4°C.
Anticipated Results LPA results are usually expressed as the percentage of the gated leukocyte subset that stains positively for the platelet-specific antibody. In addition to neutrophils and monocytes, which constitute the majority of leukocytes in circulating LPAs, other leukocyte subsets such as eosinophils, basophils, and NK cells are
capable of binding platelets (Bruijne-Admiraal et al., 1992). Approximately 4% to 14% of monocytes from healthy normal blood donors (citrate-anticoagulated whole blood, no agonist added) will be positive for platelet markers (i.e., have attached platelets), while ∼2% to 12% of neutrophils in the same sample will be positive for platelet markers. Even a carefully drawn sample from a patient with a recent thrombotic event, such as an acute myocardial infarction, can have as many as 80% circulating monocyteplatelet aggregates. A normal donor sample stimulated in vitro with 20 µM TRAP will become nearly 100% monocyte-platelet positive with neutrophils becoming 80% to 95% platelet positive in the same sample. The mean fluorescence of a given LPA type generally reflects the relative number of platelets bound to that leukocyte subtype. However, it is difficult to translate this into the exact number of platelets per LPA due to several factors. Normal platelets and platelet-derived microparticles vary widely in size and expression of surface antigens, and thus the amount of fluorescence for a single bound platelet varies widely. In addition, the membrane-binding domain between the leukocyte and platelet can cause loss of available surface for staining, and the presence of a large number of fluorophore molecules in proximity can result in quenching of emitted fluorescence. For these reasons, relative LPA fluorescence should be considered semi-quantitative.
Time Considerations A significant portion of the time commitment for LPA study is devoted to the necessary reagent optimization and generation of a normal range for a given assay configuration. Sample stability must also be confirmed in clinical studies in which immediate access to a flow cytometer is not always practical. Once a procedure has been established, the assay should be set up in anticipation of obtaining the blood sample (∼20 to 30 min). Whole blood labeling with or without activation should take <20 min at 22° to 37°C. Subsequent fixation can be done in 10 min. The most time-consuming hands-on portion of LPA analysis may be data acquisition, which can take 2 to 7 min per sample to acquire 1000 to 2000 monocytes.
Phenotypic Analysis
6.15.7 Current Protocols in Cytometry
Supplement 24
Literature Cited Bruijne-Admiraal, L.G., Modderman, P.W., Von dem Borne, A.E., and Sonnenberg, A. 1992. Pselectin mediates Ca(2+)-dependent adhesion of activated platelets to many different types of leukocytes: Detection by flow cytometry. Blood 80:134-142. Furman, M.I., Barnard, M.R., Krueger, L.A., Fox, M.L., Shilale, E.A., Lessard, D.M., Marchese, P.J., Frelinger, A.L., III, Goldberg, R., and Michelson, A.D. 2001. Circulating monocyteplatelet aggregates are an early marker of acute myocardial infarction. J. Am. Coll. Cardiol. 38:1002-1006. Furman, M.I., Benoit, S.E., Barnard, M.R., Valeri, C.R., Borbone, M.L., Becker, R.C., Hechtman, H.B., and Michelson, A.D. 1998. Increased platelet reactivity and circulating monocyteplatelet aggregates in patients with stable coronary artery disease. J. Am. Coll. Cardiol. 31:352358. Gardiner, E.E., De Luca, M., McNally, T., Michelson, A.D., Andrews, R.K., and Berndt, M.C. 2001. Regulation of P-selectin binding to the neutrophil P-selectin counter-receptor P-selectin glycoprotein ligand-1 by neutrophil elastase and cathepsin G. Blood 98:1440-1447. Hato, T., Ginsberg, M.H., and Shattil, S.J. 2002. Integrin alphaIIb-beta3. In Platelets (A.D. Michelson, ed.), pp. 105-116. Academic Press/Elsevier Science, New York. LaRosa, C.A., Rohrer, M.J., Benoit, S.E., Barnard, M.R., and Michelson, A.D. 1994. Neutrophil cathepsin G modulates the platelet surface expression of the glycoprotein (GP) Ib-IX complex by proteolysis of the von Willebrand factor binding site on GPIba and by a cytoskeletal-mediated redistribution of the remainder of the complex. Blood 84:158-168. McEver, R.P. 2002. P-selectin/PSGL-1 and other interactions between platelets, leukocytes, and endothelium. In Platelets (A.D. Michelson, ed.), pp. 139-155. Academic Press/Elsevier Science, New York. Michelson, A.D. 1994. Platelet activation by thrombin can be directly measured in whole blood through the use of the peptide GPRP and flow cytometry: Methods and clinical studies. Blood Coagul. Fibrinolysis 5:121-131.
Michelson, A.D., Barnard, M.R., Krueger, L.A., Valeri, C.R., and Furman, M.I. 2001. Circulating monocyte-platelet aggregates are a more sensitive marker of in vivo platelet activation than platelet surface P-selectin: Studies in baboons, human coronary intervention, and human acute myocardial infarction. Circulation 104:15331537. Mickelson, J.K., Lakkis, N.M., Villarreal-Levy, G., Hughes, B.J., and Smith, C.W. 1996. Leukocyte activation with platelet adhesion after coronary angioplasty: A mechanism for recurrent disease? J. Am. Coll. Cardiol. 28:345-353. Nieuwland, R. and Sturk, A. 2002. Platelet-derived microparticles. In Platelets (A.D. Michelson, ed.), pp. 255-265. Academic Press/Elsevier Science, New York. Ott, I., Neumann, F.J., Gawaz, M., Schmitt, M., and Schomig, A. 1996. Increased neutrophil-platelet adhesion in patients with unstable angina. Circulation 94:1239-1246. Palabrica, T., Lobb, R., Furie, B.C., Aronovitz, M., Benjamin, C., Hsu, Y.M., Sajer, S.A., and Furie, B. 1992. Leukocyte accumulation promoting fibrin deposition is mediated in vivo by P-selectin on adherent platelets. Nature 359:848-851. Powell, C.C., Rohrer, M.J., Barnard, M.R., Peyton, B.D., Furman, M.I., and Michelson, A.D. 1999. Chronic venous insufficiency is associated with increased platelet and monocyte activation and aggregation. J. Vasc. Surg. 30:844-851. Rinder, C.S., Bonan, J.L., Rinder, H.M., Mathew, J., Hines, R., and Smith, B.R. 1992. Cardiopulmonary bypass induces leukocyte-platelet adhesion. Blood 79:1201-1205. Simon, D.I., Chen, Z., Xu, H., Li, C.Q., Dong, J., McIntire, L.V., Ballantyne, C.M., Zhang, L., Furman, M.I., Berndt, M.C., and Lopez, J.A. 2000. Platelet glycoprotein Ibα is a counterreceptor for the leukocyte integrin Mac-1 (CD11b/CD18). J. Exp. Med. 192:193-204
Contributed by Marc R. Barnard, Lori A. Krueger, A.L. Frelinger III, Mark I. Furman, and Alan D. Michelson University of Massachusetts Medical School Worcester, Massachusetts
Michelson, A.D., Barnard, M.R., Krueger, L.A., Frelinger, A.L., III, and Furman, M.I. 2002. Flow cytometry. In Platelets (A.D. Michelson, ed.), pp. 297-315. Academic Press/Elsevier Science, New York.
Whole Blood Analysis of Leukocyte-Platelet Aggregates
6.15.8 Supplement 24
Current Protocols in Cytometry
Flow Cytometric Assessment of HLA Alloantibodies
UNIT 6.16
Antibodies against HLA molecules are formed in response to exposure to foreign HLA molecules, which can occur as a result of blood transfusion, pregnancy, or transplant. Blood components, particularly those containing cellular elements, are the most common cause of anti-HLA antibodies. For a number of reasons, the accurate and sensitive detection of HLA antibodies is of critical clinical importance. The detection of donor-directed HLA antibodies in the serum of potential transplant recipients is generally considered to be a significant risk factor for allograft rejection and/or loss. For transplant recipients, the strength (i.e., titer) of the antibody can be used to provide a measure of risk associated with poor graft outcomes. Newer methods, specifically flow cytometry, have significantly enhanced the ability to detect low levels of HLA antibody. Unfortunately, cell-based flow cytometric assays, although very sensitive, lack specificity. Recently, solid-phase flow cytometric assays utilizing microparticles coated with purified HLA molecules have been introduced. These assays measure HLA-specific alloantibodies in a very precise fashion and have revolutionized the ability to detect and document the HLA specificity of an alloantibody. This unit describes technical aspects of the flow cytometric crossmatch (FCXM; see Basic Protocol 1), flow cytometric microparticle assays (see Alternate Protocol and Basic Protocol 2), and cell-based flow cytometric screening assays (see Basic Protocol 3). The collective goal for these assays is to clearly identify the presence of HLA antibodies, determine the titer of antibody, and elucidate the specificities (i.e., HLA antigens) to which they will react. Knowledge of this information is critical for organ allocation and accurate assessment of the immunological risk for a patient at the time of transplantation. In addition, the identification of HLA antibodies in blood components may be useful in planning appropriate transfusion support strategies for selected patients. DETECTION OF HLA ANTIBODIES BY FLOW CYTOMETRIC CROSSMATCH
BASIC PROTOCOL 1
The flow cytometric crossmatch (FCXM) is performed by incubating donor cells with serum from a potential transplant recipient, followed by addition of a fluorescently labeled anti-human immunoglobulin reagent. For most studies, this secondary antibody is a polyclonal reagent with specificity for the Fc portion of IgG. IgM antibodies, which may be important in transplantation, are best detected by cytotoxicity assays. On average, <10 IgM molecules/cell are sufficient to activate complement and produce a positive result. This quantity is far below the current limit of detection by cytometry. In addition to measuring the amount of bound alloantibody, it is important to identify specificity of the antibody and thereby predict the cell types that will bind the antibody. Antibodies directed against class I HLA molecules will bind to all leukocytes. In contrast, antibodies against class II HLA molecules will react only with B cells, monocytes, and activated T cells. Thus, for most standard FCXMs, a multicolor approach is utilized to discriminate T cells from B cells. If the second antibody is conjugated to fluorescein (FITC), then a phycoerythrin (PE)-labeled monoclonal antibody that detects B cells (CD19 or CD20) and a PerCP (peridinin chlorophyll protein)-conjugated monoclonal antibody that detects T cells (CD3) are added. A number of other CD3 antibodies that are tandem conjugates can also be substituted provided they have excitation at 488 nm and emission >650 nm. Alternatively, if a dual-laser flow cytometer is available, one may use Contributed by Robert A. Bray, Howard M. Gebel, and Thomas M. Ellis Current Protocols in Cytometry (2004) 6.16.1-6.16.24 Copyright © 2004 by John Wiley & Sons, Inc.
Phenotypic Analysis
6.16.1 Supplement 27
an allophycocyanin (APC) conjugate. This three-color combination allows for the simultaneous detection of alloantibodies reacting with T cells and B cells and eliminates background binding due to natural killer (NK) cells and monocytes. The results are then analyzed by flow cytometry and expressed as positive or negative based on the observed change in median fluorescence intensity of the test serum with respect to a negative control or autologous serum. A significant increase in fluorescence above the control would indicate the presence of donor-directed alloantibodies. Materials Donor sample: freshly purified lymphocytes or mononuclear cells with >80% viability, isolated from lymph node, spleen, or peripheral blood; or thawed cryopreserved mononuclear cells Patient sample: serum from potential recipient(s) Negative control: normal human or pooled human sera from non-sensitized individuals (NHS) Positive control: pooled positive serum from sensitized patients with high anti-HLA antibody titer (PPS) Flow wash buffer (FWB; see recipe), ice cold FITC-conjugated goat anti-human IgG [F(ab′)2 fragment, Fcγ-specific; Jackson Laboratories; PE conjugate also available] Anti-human CD3-PerCP (or other conjugated fluorochrome with excitation at 488 nm and emission >650 nm) Anti-human CD19-PE (or CD20-PE) 1% paraformaldehyde in PBS, pH 7.2 ± 0.2, ice cold; store protected from light up to 2 weeks at 4°C 6 × 50–mm glass tubes (Kimble), 1.5-ml conical plastic tubes, or 96-well microtiter plates Tabletop centrifuge Vacuum aspirator Flow cytometer with 488-nm excitation and filters for collection of green (525 ± 15 nm), orange (575 ± 15 nm), and red (>650 nm) fluorescence Analysis software Additional reagents and equipment for pronase treatment (Vaidya et al., 2001), and cell counting (APPENDIX 3A) Prepare cells 1. Flash-freeze all fresh (unfrozen) serum (donor sample, patient sample, and negative and positive controls) 10 min in liquid nitrogen, then centrifuge all serum samples as indicated below. NOTE: Do not use a cell preparation that has been treated with Lympho-Kwik or Percoll. Centrifuge patient sample (serum from potential recipient) 10 min at ≥10,000 × g prior to use to remove particulate matter that may interfere with the assay. Ultracentrifugation (10 to 20 min at 100,000 × g) may also help reduce non-specific background by removing large immune complexes. The flash freezing maintains consistency with other frozen serum samples that may be used.
2. (Optional) Treat with pronase according to Vaidya et al. (2001). Pronase cleaves Fc receptors from the cell surface, thereby reducing non-specific background staining. CD20 will also be cleaved by pronase. Flow Cytometric Assessment of HLA Alloantibodies
6.16.2 Supplement 27
Current Protocols in Cytometry
Table 6.16.1 Suggested Setup for Tube Labeling
Tube no.
Primary antibody/serum
Secondary antibody
1
Negative or autologous no. 1
2
Pooled positive serum
3
Negative or autologous no. 2
4 - na
Patient sample(s)
IgG-FITC CD3-PerCP and CD19-PE IgG-FITC CD3-PerCP and CD19-PE IgG-FITC CD3-PerCP and CD19-PE IgG-FITC CD3-PerCP and CD19-PE
aRepeat setup as in tube no. 4 for each additional serum.
3. Count all nucleated cells in lymphocyte or mononuclear cell preparation (APPENDIX 3A). Mix the cell preparation well to ensure consistency and place at least 100,000 but no more than 250,000 cells/labeled 6 × 50–mm glass tube (or other appropriate vessel). See Table 6.16.1 for suggested tube setup. Do not use 12 × 75–mm tubes. All nucleated cells, not just lymphocytes, possess HLA class I antigens on their surface. The number of tubes required will depend upon the number of serum samples to be tested.
4. Centrifuge cells 2 min at 1000 × g, room temperature, in the tabletop centrifuge. 5. Aspirate all supernatant, being careful not to aspirate cell pellet. Leave a residual fluid volume in the tube (<10 µl). In this and subsequent aspirations, be careful not to contaminate sample with pipet tips. Be sure to aspirate positive control last to decrease the chance of carryover. Rinse tip after aspirating positive control.
Incubate cells with serum 6. Add 25 µl of the appropriate serum directly to the cell pellet in each corresponding tube; ensure that all the serum is deposited at the bottom of the tube. Vortex the sample to ensure proper mixing of serum and cells. Serum volumes from 20 to 50 ìl/test are optimal. 7. Incubate tubes 30 min at 4°C, protected from light. 8. Add 400 µl (or appropriate volume) ice-cold FWB to the cells. Centrifuge cells 2 min at 1000 × g, room temperature, to a pellet in a tabletop centrifuge. 9. Aspirate supernatant leaving <10 µl and repeat wash step (step 8) at least two additional times for a total of three (or four) washes. Adequate washing is critical here to ensure the removal of unbound IgG. Excess free IgG will bind the secondary antibody and may produce a false-negative result.
Stain cells 10. Add 20 to 50 µl pre-titered FITC-conjugated goat anti-human IgG (or PE-conjugated if applicable) to each tube and incubate 10 min at 4°C in the dark. 11. Add either 20 µl CD3-PerCP and 20 µl CD19-PE or 40 µl CD3-PerCP/CD19-PE premixed reagent to each tube, vortex gently, and incubate an additional 20 min at 4°C in the dark. CD3-PerCP and CD19-PE may be combined using equal volumes of the appropriate dilution of each reagent if a pre-made cocktail is not purchased.
Phenotypic Analysis
6.16.3 Current Protocols in Cytometry
Supplement 27
12. Wash cells with 400 µl ice-cold FWB and centrifuge as in step 8. Aspirate supernatant and repeat the wash. 13. Aspirate supernatant and resuspend cell pellet in 200 µl ice-cold FWB by vortexing. 14. While vortexing, add 200 µl ice-cold 1% paraformaldehyde to each tube. Cells can then be analyzed or held ≤1 week at 4°C in the dark. If samples are to be run immediately, then this fixation step may be omitted, but the volume must be adjusted to a final volume of 400 ìl for analysis. Unfixed cells should be stored at 4°C, protected from light, and analyzed within 8 hr. Samples may be transferred to 12 × 75–mm tubes if the cytometer will take nothing else. CAUTION: Paraformaldehyde is a carcinogen. Use personal protective equipment and take appropriate safety measures.
Acquire and analyze data 15. Run samples on the flow cytometer. Create a forward scatter (FS) versus side scatter (SS) dot plot. 16. Initially, set the first acquisition gate (physical gate = R1) around the lymphocyte population from the FS versus SS data plot (Fig. 6.16.1A). Next, generate a two-parameter dot plot of CD3 (red fluorescence) versus CD19 or CD20 (orange fluorescence; Fig. 6.16.1B). On this dot plot, create two additional physical gates (R2 and R3). 17. Using the analysis software, create two logical gates, R4 = R1 and R2, and R5 = R1 and R3. Next, display the results of R4 and R5 as independent single-parameter histograms with the anti-IgG antibody as the x-axis (Fig. 6.16.1C and D). This gating strategy should produce clear histograms of T cells (R4) and B cells (R5) to the exclusion of other contaminating cells such as monocytes or natural killer (NK) cells (Fig. 6.16.1E). Once the single-parameter histograms for T and B cells have been generated, analysis of results is based on the change (increase) in fluorescence of the test sample compared to the negative control sample. This may be accomplished by assessing channel values or linear values or by converting to molecules of equivalent soluble fluorochrome (MESF) units. Independent of the analytical method used to assess a positive reaction, care should be taken to ensure reproducibility of all controls used. That is, the negative and positive control samples should demonstrate a stable and reproducible fluorescence. For the negative control, a specific level of fluorescence should be determined. For the positive control, a consistent increase in fluorescence above the negative control should be established. ALTERNATE PROTOCOL
Flow Cytometric Assessment of HLA Alloantibodies
DETECTION OF HLA ANTIBODIES USING A MICROPARTICLE SCREENING ASSAY Panel-reactive antibody (PRA) screening is performed to assess the level of presensitization of a given transplant candidate. Historically, PRA testing has been accomplished by testing a patient sample against a panel of HLA-typed cells. The resulting percent PRA could be used as an approximation of the positive crossmatch frequency to be expected for that patient. That is, a PRA of 50% would indicate that one of two donors would have a positive crossmatch with this patient. However, cell-based assays are not as sensitive as flow cytometric crossmatch methods and therefore create a “disconnect” between the ability to screen for HLA antibody and the ability to detect HLA antibody via the FCXM. Moreover, attempting to perform classical PRA testing by flow cytometry is neither financially nor logistically possible. To address these and other issues, researchers developed microparticles coated with purified HLA antigens. Using this approach, a PRA can be obtained by incubating the HLA antigen–coated beads with patient serum. As in
6.16.4 Supplement 27
Current Protocols in Cytometry
C A Counts
1000
Side scatter
800 600 400
R1 * R2 T cells
101
D
200
102 FITC-anti-lg
103
104
40 R1 0
200
400 600 800 Forward scatter
B 104 103
0 100 R6
Counts 101
102 103 CD3-PerCP
101
E
R2
101 100 100
B cells 20 10
R3
102
R1 * R3
30
1000
Counts
0
CD20-PE
90 80 70 60 50 40 30 20 10 0 100
104
90 80 70 60 50 40 30 20 10 0 100
102 FITC-anti-lg
103
104
R1 lymphocytes NK cells
101
102 FITC-anti-lg
103
104
Figure 6.16.1 Recommended gating strategy for the multicolor flow cytometric crossmatch. Mononuclear cells are prepared from peripheral blood (shown), lymph nodes, or spleen. In this example, cells were incubated with a negative control serum. Since the flow crossmatch is focused on assessing antibodies bound to lymphocytes, an initial lymphocyte gate R1 is set (A). (B) Two-color dot plot of cells contained in R1. Cells in this example are stained with CD3-PerCP (x-axis) and CD20-PE (y-axis). Additional gates are set around the CD3 (R2) and CD20 (R3) populations. To correctly analyze the populations of interest, additional logical gates (combination of two or more physical gates) must be established as follows: R4 = R1 and R2, for display of only the CD3+ lymphocytes (C), and R5 = R1 and R3, for display of only the CD20+ lymphocytes (D). (E) The results of plotting only R1, the lymphocyte gate. As shown, this histogram contains additional brightly staining cells. These cells are natural killer (NK) cells, that by their expression of CD16 (FcγRIII) bind IgG-class immunoglobulin molecules and display a bright fluorescence when stained with an anti-immunoglobulin reagent. Since the bright fluorescence observed with NK cells precludes their analysis for HLA antibodies, it is important to exclude them from analysis.
Phenotypic Analysis
6.16.5 Current Protocols in Cytometry
Supplement 27
the crossmatch, a secondary anti-IgG reagent is used. Control beads consisting of the same plastic beads coated with human serum albumin are used as an additional control. The microparticles are constructed such that class I antigens are coated on one set of particles and class II antigens are coated on a second set. These particles are discernable by differences in size (i.e., forward scatter) as well as fluorescence. Specifically, the class II beads are impregnated with a red fluorescent dye that emits at the same wavelength as PE. Thus, with a single sample of patient serum, testing for both class I and class II antibodies can be done simultaneously. Samples are analyzed by flow cytometry and the results are expressed as percent positive or negative based on the shift in fluorescence intensity of the test serum with respect to the negative control serum and the control microparticles. The bead procedure is used primarily for detecting IgG HLA antibodies. The class of antibody detected is dependent on the specificity of the secondary antibody used in the assay. When testing for IgM antibodies, the same reagents and controls are used as in the setup for IgG antibodies, provided that the sample contains sufficient IgM alloantibody for detection. However, one must use an anti-IgM-labeled secondary antibody. Materials Sera to be tested Negative control: normal human or pooled human sera from non-sensitized individuals (NHS; One Lamda or Pel-Freez) Positive control: pooled positive serum from sensitized patients with high antibody titer (PPS; One Lamda or Pel-Freez) Class I and class II FlowPRA beads (a combined kit or separate products from One Lambda). 1:10 (v/v) diluted flow bead wash buffer (provided with the kit from One Lambda) in water; use at room temperature FITC-conjugated goat anti-human IgG [F(ab′)2 fragment, Fcγ specific; Jackson Laboratories; PE conjugate also available] Tabletop centrifuge (e.g., Fisher or Beckman) 6 × 50–mm glass tubes, 5-ml plastic conical tubes, or 96-well tissue culture plates Repeat pipettor (e.g., Gilson Pipetman) with DNA gel-loading tips Vacuum aspirator Flow cytometer with 488-nm excitation and filters for collection of green fluorescence Prepare sera samples and beads 1. Thaw and vortex all sera to be tested. Thaw and mix the appropriate controls. Both negative and positive control can be prepared locally or obtained from a commercial source.
2. Centrifuge all sera 5 min at 5000 × g, room temperature, in a tabletop centrifuge. Sera may then be removed directly and used for testing or transferred into microultracentrifuge tubes to aid in multi-channel pipetting for 96-well plate setups.
3a. If using 6 × 50–mm glass tubes, label the appropriate number of tubes needed to run the test.
Flow Cytometric Assessment of HLA Alloantibodies
Tube 1 = NHS Tube 2 = NHS (duplicate) Tube 3 = PPS Tubes 4 through n = patient samples
6.16.6 Supplement 27
Current Protocols in Cytometry
3b. If using a 96-well tissue culture plate, label the plate with the batch number. Well A1 = NHS Well A2 = NHS (duplicate) Well A3 = PPS Wells A4 through H12 = patient samples 4. Vortex class I and class II FlowPRA beads to resuspend completely. To prepare a working combination, combine one thawed vial of class I and one thawed vial of class II beads. Mix well before use.
Add beads to samples 5. Using a repeat pipettor with DNA gel-loading tips, add 10 µl class I/class II bead mixture to each of the above labeled tubes or wells. A Gilson Pipetman with DNA gel-loading tips (these very long tips allow the addition of serum directly to the bottom of the staining tubes) may also be used to dispense the beads.
6. Using regular pipet tips, add 50 µl control or patient serum to the beads. 7. Vortex each tube or the plate with cover replaced and incubate 30 min at room temperature in the dark. Gently vortex the tubes/plate to resuspend the beads after 10 min and again after 20 min. 8a. If using 6 × 50–mm glass tubes, wash two times as follows: a. Add 400 µl of 1:10 flow bead wash buffer/water and vortex. b. Centrifuge 1 min at 5000 × g, room temperature, in a tabletop centrifuge. c. Aspirate the supernatant to a dry button, being careful not to aspirate the beads. d. Repeat. 8b. If using a 96-well plate, wash three times as follows: a. Add 75 µl of 1:10 flow bead wash buffer/water to each well. Use a multi-channel adapter for the repeat pipettor. b. Vortex the plate to mix the beads well. c. Add another 75 µl of 1:10 flow bead wash buffer/water to the wells. Do not vortex at this point because of the risk of splashover between wells. d. Replace plate cover and centrifuge 3 min at 900 × g with the brake on. e. Remove the plate from the centrifuge and flick the plate to remove the supernatant. To flick 96-well plates, remove the cover and quickly turn the plate upside down so that the supernatant is forcibly ejected from the wells. Place the plate upside down on an absorbent pad to pull the last remaining liquid from the wells.
f. Replace the cover and gently run the plate across the vortex in order to loosen the bead pellet. g. Repeat wash steps two additional times. 9. Add 20 µl appropriately titered FITC-conjugated anti-human IgG to the dry bead button. 10. Vortex samples and incubate 30 min in the dark at room temperature. After 15 min, gently vortex to resuspend the beads. 11. Wash samples as in step 8a or 8b. Phenotypic Analysis
6.16.7 Current Protocols in Cytometry
Supplement 27
A
C BEADSQC8.001
BEADSQC8.001
104
R1 class I
Side scatter
103
class II
102
class I SS
class II
101 100 0
200
400
600
800
B
D BEADSQC8.001
BEADSQC8.001
104 Orange fluorescence
FS
1000
class II
R3
103
class II class I
102
Orange fluorescence
class I
101 R2 100 0
200
400 600 800 Forward scatter
1000
FS
Figure 6.16.2 Analysis of the FlowPRA microparticles. (A) Forward-scatter versus side-scatter properties of these microparticles. It is sometimes easier to gate on the beads by using log side scatter, as displayed here. As with the FCXM, logical gating strategies are required for correct analysis of the microparticles. Initially, a region (R1) is established around an area containing both the class I and class II beads (A). Since the beads are of two different sizes, it is easy to identify both sets. Next, a dot plot is constructed that displays forward scatter versus orange fluorescence (B). Since the class II beads are impregnated with a dye that emits in the orange fluorescence detector of most cytometers, a clear discrimination of the two bead types is now easy. Two additional physical gates R2 and R3 are now constructed. A logical gating strategy is next employed and the logical gates R4 = R1 and R2 and R5 = R1 and R3 are constructed. The results of R4 and R5 are then plotted on a single-parameter histogram (See Fig. 6.16.3). C and D are the 3-D plots of A and B, respectively.
12a. If using 6 × 50–mm glass tubes, resuspend in 200 µl of 1:10 diluted flow bead wash buffer/water and vortex. 12b. If using a 96-well plate, resuspend in 75 µl of 1:10 diluted flow bead wash buffer/water, vortex, then add another 75 µl flow bead wash buffer. Transfer the volume from the wells to prenumbered 6 × 50–mm glass tubes. 13. Analyze on the flow cytometer.
Flow Cytometric Assessment of HLA Alloantibodies
Samples may be transferred to 12 × 75–mm tubes if the cytometer will take nothing else. However, the 6 × 50–ml tubes hold only 200 ìl, a volume that may be too small in a 12 × 75–tube for some instruments. Alternatively, for BD instruments and the Beckman Coulter FC500, the 6 × 50–ml tubes can be fitted into 12 × 75–mm tubes and will work well if care is taken to get the sample probe into the smaller tube when the sample is run. This configuration permits staining of very small volumes of sample and low cell numbers yet maintains reasonable acquisition times. For these assays, particle (cell) to antibody (serum) relationships are critical. Since antibody concentration is unknown, keeping the number of particles constant and minimal maximizes the potential sensitivity.
6.16.8 Supplement 27
Current Protocols in Cytometry
class I A
C
negative control 60
30 20
M2
Counts
40 Counts
patient sample
M1
M1
10 0
0 0
200
400 600 800 Anti-lgG-FITC
1000
0
class II
200
400 600 Anti-lgG-FITC
800
1000
D
B 30
120 Counts
Counts
6%
M1
20 M1
10
M2
0
0 0
200
400 600 Anti-lgG-FITC
800
1000
0
200
400 600 Anti-lgG-FITC
800
1000
Figure 6.16.3 Sample histograms of the class I (A, C) and class II (B, D) microparticles. (A) and (B) are negative control samples. (C) and (D) are positive patient samples, with a clear increase in fluorescence intensity of microparticle clusters. The PRA value is calculated as the percentage of beads that have been displaced with respect to the negative control (M2 marker). The M1 marker is maintained in each histogram to indicate the boundaries of the negative control. Positivity of a single bead would be reflected by a single population of beads that constitute 3% (1 out of 30) of the total events. For a sample to be considered as positive, a cluster of beads must be observed to be displaced along the x-axis. The arrow in D indicates a cluster accounting for ∼6% of the total events, or two positive beads.
Acquire and analyze data 14. First, set the instrument to collect log SS versus linear FS (Fig. 6.16.2A,C). Next, generate a plot of FS versus orange fluorescence (Fig. 6.16.2B,D). This plot is necessary to elucidate bead aggregates as well as bead fragments.
15. Build histogram plots for class I beads and class II beads (Fig. 6.16.3). The gating strategy for the histograms is similar to the logical gating strategy used in Basic Protocol 1. For the class I beads, the logical gate should be R4 = R1 and R2. For the class II beads, the logical gate should be R5 = R1 and R3. The interpretation of a positive result is made by observing increases in fluorescence above background (i.e., negative control serum).
DETERMINATION OF HLA SPECIFICITY (CLASS I AND CLASS II) USING FLOW MICROPARTICLES Antibody detection and specificity identification have always been major functions of an HLA laboratory. An individual who has experienced sensitizing events such as transfusion, pregnancy, or loss of a previous transplant is at risk for developing anti-HLA antibodies. Thus, it is important for the HLA laboratory to detect and identify these antibodies prior to the patient’s receiving an organ for transplant or re-transplant. FlowPRA testing can determine the absence or presence of an antibody and assign a PRA value; however, this is insufficient. In addition to PRA, the assignment of HLA antibody specificity is essential for appropriate organ allocation and for helping predict crossmatch results. The flow-specific-bead assay allows for the identification of HLA specificity for both class I and class II antibodies.
BASIC PROTOCOL 2
Phenotypic Analysis
6.16.9 Current Protocols in Cytometry
Supplement 27
Briefly, the flow-specific-bead assay utilizes microparticles that have been coated with purified HLA antigen. Beads may be coated with either class I or class II antigens, allowing for the determination of either of these classes of antibodies. Antibody screening using flow specificity beads provides a positive or negative, allows for the determination of a percent PRA, and most importantly, permits the assignment of antibody specificity (class I or class II depending on the test setup). The specificity beads are similar to the screening beads in that they are coated with HLA antigens and that the test is performed by incubating the beads with patient sera, followed by the addition of fluoresceinated anti-human immunoglobulin reagent; either IgG or IgM antibodies can be detected depending on the type of secondary antibody used. The added advantage of the specificity beads is that there are four individual pools with each pool containing 8 (or 11) beads (for a panel size of 32 or 44), rather than a single pool of 30 beads. Within each pool, the individual beads are discriminated by the level of red fluorescence. This characteristic permits the assessment of individual beads within the pool. Since the HLA antigens expressed on each of the beads are known, anti-HLA antibody specificity can be determined based on pattern analysis of the positive reactions. More recently, single-antigen-coated microparticles have been developed that have only one HLA allele coated on their surface, permitting fine specificity analysis of patient alloantisera. The combination of FlowPRA for screening with flow-specific beads and single-antigen beads provides the best approach for characterization of HLA-specific antibodies. Materials Patient serum Negative control sera Positive control sera Liquid nitrogen FlowPRA specific class I beads and/or flowPRA specific class II beads (One Lambda); provided in 4 groups of 8 beads/group 1:10 (v/v) diluted flow bead wash buffer (included in specificity bead kits from One Lambda) in water; use at room temperature FITC-conjugated goat anti-human IgG [F(ab′)2 Fc specific] or FITC-conjugated goat anti-human IgM [F(ab′)2 Fc specific] Tabletop centrifuge (e.g., Fisher or Beckman) Polyethylene microultracentrifuge tubes 6 × 50–mm glass tubes or 96-well tissue culture plate Vortexer 5-µl single-shooter syringe (e.g., Hamilton) Vacuum aspirator Repeat pipettor (e.g., Eppendorf) with a multi-channel adapter Flow cytometer with 488-nm excitation and filter set for detection of green and orange fluorescence Prepare sera and beads 1. If the serum has not been frozen, quick-freeze 10 min in liquid nitrogen. Thaw and mix all sera to be tested.
Flow Cytometric Assessment of HLA Alloantibodies
2. Centrifuge sera 5 min at 3000 × g, room temperature, in a tabletop centrifuge. Remove sera directly from the tubes for testing or dispense aliquots into microultracentrifuge tubes to aid in multichannel pipetting for plate setups. Centrifugation of the sample is necessary prior to testing to remove aggregates and large immune complexes, which may interfere with the assay.
6.16.10 Supplement 27
Current Protocols in Cytometry
Table 6.16.2 Tube Format for Assay Using Flow Microparticles
1
2
3
4 ————— through N samples ——————————————
N H S | 1
P P S
N H S | 2
S E R U M 1
S E R U M 2
S E R U M 3
S E R U M 4
S E R U M 5
S E R U M 6
S E R U M 7
S E R U M 8 ——-
S E R U M
3a. If using 6 × 50–mm glass tubes, for each pool (total of 4) of specificity beads, set up and label the appropriate number of glass tubes following the format given in Table 6.16.2. 3b. If using a 96-well tissue culture plate, label the plate with the batch number. Plate the bead pools in rows and serum in columns, as shown in Table 6.16.3. 4. Mix each group (1 through 4) of beads very well by vortexing until the beads are completely resuspended. 5. Using a 5-µl single-shooter syringe, add 5 µl flow specificity beads (class I or class II) to each of the above labeled tubes or wells. See Tables 6.16.2 and 6.16.3 for formats. A Gilson Pipetman with gel loading tips may also be used to dispense the beads into the glass tubes.
6. Add 50 µl control or patient serum to each pool of beads. Incubate sera and wash 7. Vortex each tube or the plate (replace cover) and incubate 30 min at room temperature in the dark. Gently vortex the tubes/plate to resuspend the beads after 10 min and again after 20 min. 8a. If using 6 × 50–mm glass tubes, wash samples two times as follows: a. Add 400 µl of 1:10 diluted flow bead wash buffer/water at room temperature and vortex. b. Centrifuge 1 min at 3000 × g, room temperature, in a tabletop centrifuge. c. Aspirate the supernatant to a dry button (do not aspirate beads). d. Repeat. Table 6.16.3 Plate Format for Flow Microparticles Assay 1
2
NHS-1 PPS A B C D E F G H
3
4
5
6
7
8
9
10
11
12
NHS-2 Samples —————————————————————->
————————————Group 1———————————————————-> ————————————Group 2———————————————————-> ————————————Group 3———————————————————-> ————————————Group 4———————————————————-> ————————————Group 1———————————————————-> ————————————Group 2———————————————————-> ————————————Group 3———————————————————-> ————————————Group 4———————————————————->
Phenotypic Analysis
6.16.11 Current Protocols in Cytometry
Supplement 27
8b. If using a 96-well tissue-culture plate, wash samples three times as follows: a. Using a repeat pipettor, pipet 75 µl of 1:10 diluted flow bead wash buffer/water to each well. b. Replace cover and vortex the plate to mix the beads well. c. Add another 75 µl of 1:10 diluted flow bead wash buffer/water to the wells. Do not vortex at this point because of the risk of splashover between wells. d. Replace plate cover and centrifuge 3 min at 900 × g with the brake on. e. Remove the plate from the centrifuge and flick the plate to remove the supernatant. To flick, remove the plate cover and quickly invert the plate over an appropriate receptacle. The inversion should be forceful enough that the supernatant is removed from the well, but not the beads. While still holding the plate upside down, place it on an absorbent pad to wick off the remaining liquid.
f. Replace the cover and gently run the plate across the vortexer in order to loosen the bead pellet. g. Repeat wash two additional times. Stain with secondary antibody 9. Add 20 µl FITC-conjugated anti-human IgG to the dry button. If testing for IgM antibodies use FITC-conjugated anti-human IgM.
10. Vortex samples and incubate 30 min in the dark at room temperature. After 15 min, gently vortex to resuspend beads. 11. Wash samples as in step 8a or 8b. 12a. If using 6 × 50–mm glass tubes, resuspend in 200 µl of 1:10 diluted flow bead wash buffer/water and vortex. 12b. If using a 96-well tissue culture plate, add 75 µl of 1:10 diluted flow bead wash buffer/water, vortex, and add another 75 µl of 1:10 diluted flow bead wash buffer/water. Transfer the volume from the wells to prenumbered 6 × 50–mm glass tubes using the multi-channel pipettor. 13. Run samples immediately on a flow cytometer and analyze, or store ≤24 hr at 4°C. Samples may be transferred to 12 × 75–mm tubes if the cytometer will take nothing else. However, the 6 × 50–ml tubes hold only 200 ìl, a volume that may be too small in a 12 × 75–tube for some instruments. Alternatively, for BD instruments and the Beckman Coulter FC500, the 6 × 50–ml tubes can be fitted into 12 × 75–mm tubes and will work well if care is taken to get the sample probe into the smaller tube when the sample is run. This configuration permits staining of very small volumes of sample and low cell numbers yet maintains reasonable acquisition times. For these assays, particle (cell) to antibody (serum) relationships are critical. Since antibody concentration is unknown, keeping the number of particles constant and minimal maximizes the potential sensitivity.
Acquire and analyze data 14. Set up flow cytometer. The flow cytometer should be calibrated daily using the appropriate control particles for each instrument. Calibration practices should be designed for optimal detection of low fluorescence values, i.e., maximizing signal to noise is essential.
Flow Cytometric Assessment of HLA Alloantibodies
6.16.12 Supplement 27
Current Protocols in Cytometry
15. Gate on the bead population in the forward scatter versus side scatter histogram and acquire 15,000-20,000 events. Plot green versus orange (bead) fluorescence (Fig. 6.16.5). Since results for a given sample are a summation of four separate tubes, data analysis is best performed as a separate post-acquisition step. Assign the individual beads as positive or negative based on their shift relative to the negative control markers. Refer to Anticipated Results for more detail.
CELL-BASED FLOW CYTOMETRIC PANEL REACTIVE ANTIBODY (FC-PRA) ASSAY The flow cytometric crossmatch is the most sensitive method for detecting anti-HLA antibodies in the sera of potential allograft recipients. The fact that the FC-PRA is more sensitive than the antiglobulin-enhanced-complement-dependent cytotoxicity (AHGCDC) indicates that situations may arise wherein the antiglobulin-enhanced-complement-dependent cytotoxicity (CDC) crossmatch is negative but the flow cytometric (FC) crossmatch is positive, thereby precluding transplantation in certain instances. In order to better identify and define alloantibodies, the FC-PRA was developed to address routine antibody screening for selected patients. Such patients include new transplant candidates who have a history significant for sensitization (i.e., multiple transfusions or pregnancies) and currently active patients in whom the antibody titer, by AHG-CDC or flow screening beads, has significantly declined.
BASIC PROTOCOL 3
The FC-PRA is performed by using pools of panel cells (7 pools with 4 cells/pool). Patient sera are tested undiluted against the pools and the reactivity patterns are recorded. From the reaction patterns, a percent PRA can be calculated and in some instances specificities can be assigned. The nature of the pools is such that broadly cross-reactive antibodies (CREGs) can be identified rather than multiple unique specificities. The goal of the FC-PRA is to determine the presence or absence of alloantibodies in selected patients and to assign specificity, albeit broadly reactive or CREGs. Cell-based flow PRA is useful in identifying predominantly HLA class I antibodies. As such, only T lymphocyte reactivity is assessed. Materials Sera to be tested Negative controls: normal human and pooled human sera (NHS) Positive control: pooled positive serum titered for the sensitivity of the flow (PPS) Frozen cell pools, frozen using techniques that preserve maximum viability; appropriate concentration is ∼8 × 106/ml RPMI containing 20% FBS Flow wash buffer (FWB, see recipe), ice cold FITC-conjugated goat anti-human IgG [F(ab′)2, Fc specific; Jackson Laboratories] Phycoerythrin (PE)-conjugated anti-human CD3 monoclonal antibody (Becton-Dickinson), diluted according to manufacturer’s specification 1% paraformaldehyde in PBS, pH 7.2 ± 0.2 Airfuge (e.g., Beckman) with microultracentrifuge tubes and protective caps 15-ml conical tubes Colored tape Benchtop centrifuge (e.g., Beckman GP) 6-ml polypropylene tubes (e.g., Falcon) 96-well U-bottom cell culture plate (Corning-Costar) Repeat pipettor (Eppendorf) and 8-channel attachment with appropriate tips Transfer pipet (e.g., Brinkman Transferpette-12) with appropriate tips Absorbent pad
Phenotypic Analysis
6.16.13 Current Protocols in Cytometry
Supplement 27
Flow cytometer with 488-nm excitation and filters for collection of green and orange fluorescence 6 × 50–mm glass tubes (Baxter diSPo culture tubes) Additional reagents and equipment for counting cells (APPENDIX 3A) Prepare serum 1. Remove frozen patient sera (maximum of nine patients) and positive (PPS) and negative (NHS) controls. Allow to thaw completely at room temperature. If serum has not been frozen, flash-freeze 10 min in liquid nitrogen.
2. Vortex all samples. Airfuge all patient sera and controls 10 min at 28 psi. A total volume of 250 ìl per sample is needed. Vortex two microultracentrifuge tubes with 125 ìl of serum in each tube. Thaw frozen cell pools 3. For each pool, label one 15-ml conical tube with colored tape (select a different color for each pool). Record the lot number for each pool. 4. Add 1 ml RPMI containing 20% FBS to each tube. 5. For each pool, obtain the required volume of cells and place on dry ice in a styrofoam container. For a concentration of 8 × 106/ml, one vial of each pool is sufficient. If the cell concentration is less, more than one vial will be needed.
6. Thaw one pool at a time. Follow the laboratory procedure for thawing frozen cells or see APPENDIX 3B. 7. Place the thawed cells into the appropriately labeled 15-ml conical tube containing RPMI with 20% FBS. Slowly add more RPMI with 20% FBS, drop by drop until the tube is filled. 8. Repeat steps 6 and 7 for the remaining pools, one at a time. Wash the cell pools 9. Cap the tubes containing the thawed cells and centrifuge 1 min at 1000 × g, room temperature (2400 rpm in a Beckman-GP centrifuge). 10. Decant the supernatant and resuspend the cells in 10 ml RPMI with 20% FBS. 11. Repeat the centrifugation in step 9 and decant the supernatant. 12. Label 6-ml polypropylene tubes, one for each pool, with the appropriate colored tape. To each tube, add 2 ml FWB. 13. Transfer each cell preparation from the 15-ml conical tube to the appropriately labeled 6-ml polypropylene tube. 14. Check the viability of all pools and perform a cell count (APPENDIX 3A). Cleanup of the flow PRA pool cells is usually not needed, but if viability is <70%, one may use DNase. Do not use Percoll or Lympho-Kwik to clean up the cell preparation.
15. Adjust the cell count with FWB to 2.5 × 106/ml (a final minimum volume of 1.4 ml is needed). 16. Label a 96-well U-bottom cell culture plate with the setup date and the batch number. Flow Cytometric Assessment of HLA Alloantibodies
17. Using the repeat pipettor, add 100 µl of each of the cell pools to the 96-well plate using the format described in Table 6.16.4.
6.16.14 Supplement 27
Current Protocols in Cytometry
Table 6.16.4 Plate Format for Cell Pools
1
2
3
4
5
6
7
8
9
10
11
12
A B C D E F G H
<——————————————Pool 1—————————————————-> <——————————————Pool 2—————————————————-> <——————————————Pool 3—————————————————-> <——————————————Pool 4—————————————————-> <——————————————Pool 5—————————————————-> <——————————————Pool 6—————————————————-> <——————————————Pool 7—————————————————->
18. Replace the plate cover and centrifuge 3 min at 900 × g, room temperature, with the brake on. 19. Remove the plate from the centrifuge and remove the supernatant by quickly turning the plate upside down so the supernatant is forcibly ejected from the plate. Next, while still holding the plate upside down, place it on an absorbent pad to pull the last remaining liquid from the wells. 20. Replace the cover and gently run the plate across a vortexer to loosen the cell pellet. 21. Using a transfer pipet, add 25 µl/well of patient serum and any controls, using the plating format described in Table 6.16.5. 22. Replace the plate cover and mix the cell pools and serum by gently running the plate over a vortexer. Incubate the plate 30 min at 4°C. Wash the plate 23. Remove the plate cover and use the repeat pipettor with 8-channel attachment to add 75 µl cold flow wash buffer to each well. Replace the cover and vortex. 24. Add another 75 µl to each well to complete the wash, but do not vortex at this point since splashover may occur. 25. Replace the plate cover and centrifuge 3 min at 900 × g with the brake on to pellet the cells.
Table 6.16.5 Plating Format for Flow Cell PRA
A B C D E F G H
1
2
3
4
5
6
7
8
9
10
11
12
N H S ↓ ↓ ↓ ↓
P P S
S E R U M 1 ↓
S E R U M 2 ↓
S E R U M 3 ↓
S E R U M 4 ↓
S E R U M 5 ↓
S E R U M 6 ↓
S E R U M 7 ↓
S E R U M 8 ↓
S E R U M 9 ↓
P H S # 2
↓ ↓
↓ Phenotypic Analysis
6.16.15 Current Protocols in Cytometry
Supplement 27
26. Remove the plate cover and flick the plate to remove the supernatant. Turn plate upside down on an absorbent pad in order to drain the remaining liquid from the wells. 27. Replace the plate cover and gently vortex the plate to loosen the cell pellet. 28. Repeat steps 23 to 26 two additional times for a total of three washes. 29. After the last wash, make sure that the cell pellet is dry by the absence of any visual liquid and vortex the plate to loosen the cell pellet. Add fluorochrome and monoclonal antibody 30. Add 20 µl properly diluted FITC-conjugated IgG to every well containing cell pools. 31. Replace the plate cover and vortex the plate to mix the cells and fluorochrome reagent. Incubate 10 min at 4°C. 32. Add 20 µl properly diluted PE-conjugated CD3 monoclonal antibody to every well containing cell pools. CD3 should be titered prior to use to determine optimal dilution.
33. Replace the cover and vortex the cell/reagent mixture. Incubate 20 min at 4°C. 34. Wash the 96-well plate three times as described in steps 23 to 26. 35. After the final wash, make sure that the cell pellet is dry by the absence of any visual liquid and vortex the plate to loosen the cell pellet. 36. Using the repeat pipettor, add 75 µl FWB to each of the 96 wells. Replace the cover on the plate and gently vortex. 37. Using the repeat pipettor, add 75 µl of 1% paraformaldehyde to each well containing cells. Replace the cover and gently vortex. The plate may be stored ≤3 days in the dark at 4°C.
38. When ready to run on the flow cytometer, transfer the sample preparations to pre-labeled 6 × 50–mm glass tubes. Samples may be transferred to 12 × 75–mm tubes if the cytometer will take nothing else. However, the 6 × 50–ml tubes hold only 200 ìl, a volume that may be too small in a 12 × 75–tube for some instruments. Alternatively, for BD instruments and the Beckman Coulter FC500, the 6 × 50–ml tubes can be fitted into 12 × 75–mm tubes and will work well if care is taken to get the sample probe into the smaller tube when the sample is run. This configuration permits staining of very small volumes of sample and low cell numbers yet maintains reasonable acquisition times. For these assays, particle (cell) to antibody (serum) relationships are critical. Since antibody concentration is unknown, keeping the number of particles constant and minimal maximizes the potential sensitivity.
Acquire and analyze data 39. Plot FS versus SS and set a region based on the lymphocytes. Next, construct a plot of FS versus CD3 and set a second region around the CD3-positive T lymphocytes. Collect a minimum of 5000 CD3 cells for analysis. 40. Once CD3 cells have been acquired, plot a single-parameter histogram of anti-IgG fluorescence.
Flow Cytometric Assessment of HLA Alloantibodies
Since each cell pool is a mixture of four individual cells, each individual cell comprises 25% of the total events. Hence, positive reactions should be graded in increments of 25% (Fig. 6.16.7).
6.16.16 Supplement 27
Current Protocols in Cytometry
TREATMENT OF CELL PREPARATION CONTAINING LYMPHO-KWIK OR PERCOLL
SUPPORT PROTOCOL
Lympho-Kwik (which contains Percoll) and Percoll may produce false-negative crossmatch results. This protocol removes Percoll from cell preparations if it is absolutely necessary to use them. Materials Cell preparation treated with Percoll or Lympho-Kwik RPMI medium with 10% FBS Flow wash buffer (FWB, see recipe) 37°C incubator or water bath 1. Incubate cell preparation ≥1 hr in RPMI with 10% FCS at 37°C. 2. Wash three times with flow wash buffer by centrifuging 5 min at 1000 × g. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Flow wash buffer (FWB) Phosphate buffered saline (PBS; APPENDIX 2A) with Ca2+ and Mg2+ 2% FBS 0.1% sodium azide (NaN3) Store up to 30 days at 4°C COMMENTARY Background Information Antibodies against HLA molecules are formed in response to exposure to foreign HLA molecules. Exposure to HLA antigens can occur as a result of blood transfusion, pregnancy, or transplant. Blood components, particularly those containing cellular elements, are the most common cause of anti-HLA antibodies. However, even components that are mostly devoid of cellular elements (i.e., plasma, serum, and cryoprecipitate) may induce an HLA response as a result of soluble HLA proteins. During pregnancy, the mother is exposed to the paternal HLA antigens that are carried by the fetus. The occurrence of HLA antibodies in parous females is related to the number of pregnancies. Lastly, organ and stem cell transplantation, particularly when an allograft is rejected, can result in the production of HLA antibodies. The detection of donor-directed HLA antibodies in the serum of potential transplant recipients is generally considered to be a significant risk factor for allograft rejection and/or loss. This aspect applies to both solid organ and stem cell transplantation. More recently, anti-HLA antibodies in blood components that are transfused into patients have been associated with the life-
threatening condition called TRALI (Transfusion-Related Acute Lung Injury). A recent survey of blood components determined that HLA antibodies can be found in ∼22% of all blood components (Harris et al., 2003). Of possible clinical importance is the observation that blood component support (primarily fresh-frozen plasma and cryoprecipitate) is required for patients undergoing plasmapheresis protocols for the removal of HLA antibodies. Such protocols are performed preemptively for highly sensitized patients or as rescue therapy for graft rejection. Thus, for a growing number of reasons, the accurate and sensitive detection of HLA antibodies is of critical clinical importance. For transplant recipients, the strength (i.e., titer) of the antibody can be used to provide a measure of risk associated with poor graft outcomes. The level of risk ranges from negligible (i.e., little or no change required in post-transplant immunosuppression) to extreme (the transplant is contraindicated). Assessment of HLA antibodies in serum (or any blood component) generally addresses two questions. (1) Is there an HLA antibody present? (2) Is this antibody of clinical relevance? The operating
Phenotypic Analysis
6.16.17 Current Protocols in Cytometry
Supplement 27
Flow Cytometric Assessment of HLA Alloantibodies
premise behind the first question is essentially one of sensitivity. Newer methods, specifically flow cytometry, have significantly enhanced the ability to detect low levels of HLA antibody. The underlying principles for the second question are far more complicated and encompass issues related to antibody specificity and affinity, immunoglobulin class, and titer. First and foremost, however, is the ability to confirm that a donor-reactive antibody is HLA specific. Historically, HLA antibody detection methods utilized viable donor lymphocytes as targets. The assumption was that any positive reaction was caused by HLA antibodies and was, therefore, clinically significant. In many instances, this assumption did not hold true. Transplant candidates may possess many types of donor-reactive antibodies that are not HLA specific and hence do not add significant risk to the transplant. Unfortunately, cell-based flow cytometric assays, although very sensitive, lack specificity. As a result, many transplant programs adopted a conservative approach and interpreted all positive results as a contraindication for transplant. Recently, solid-phase flow cytometric assays utilizing microparticles coated with purified HLA molecules have been introduced. These assays measure HLA-specific alloantibodies in a very precise fashion and have revolutionized the ability to detect and document the HLA specificity of an alloantibody. The “gold standard” for detecting HLA antibodies is the complement-dependent cytotoxicity (CDC) assay. This assay has been utilized as a primary crossmatch method because it detects high-titer antibodies that are responsible for hyperacute graft rejection. However, this methodology does not detect low concentrations of antibodies that have been associated with accelerated graft rejection or other rejection-related immunological complications. Consequently, more sensitive antibody detection methods have been developed that utilize flow cytometry. The first of these tests is the flow cytometric crossmatch (FCXM; Garovoy et al., 1983). Although numerous studies have demonstrated that low levels of HLA alloantibody can be associated with severe clinical consequences (Chapman et al., 1985; Cook et al., 1987; Thistlethwaite et al., 1987; Lazda et al., 1988; Talbot et al., 1988; Bray et al., 1989; Ogura et al., 1993; Scornik et al., 1994; Utzig et al., 1997; Pei et al., 1998; Piazza et al., 1998; Gebel and Bray, 2000; Muller-Steinhardt et al., 2000; Gebel et al., 2001; Karpinski et al., 2001; Piazza et al., 2001; Scornik et al., 2001; Vaidya
et al., 2001; Kerman et al., 2002; Le Bas-Bernadet et al., 2003; Pei et al., 2003), the fact that the FCXM is significantly more sensitive than standard cytotoxic crossmatch techniques poses a serious problem for the transplant community. That is, patients presumed to be devoid of HLA antibody, as determined by a CDC method, give an unexpected positive crossmatch when tested by flow cytometry. Moreover, this discrepancy can be explained in two ways: (1) flow cytometry is more sensitive for detecting HLA-specific antibody, and (2) flow cytometry is too sensitive and detects irrelevant antibodies. Such discrepancies impact the ability to prospectively assess the risk of transplantation since a negative CDC crossmatch does not necessarily signify that the FCXM will be negative and a positive FCXM can not be confirmed as HLA-specific. The ability to assess risk is predicated on the belief that a positive crossmatch is the result of HLA-specific antibody. Such predictive capability is essential for appropriate risk assessment and efficient allocation of organs. Together, the combination of solid-phase flow cytometric microparticle assays and the FCXM permits the most accurate assessment of HLA-specific alloantibodies.
Critical Parameters and Troubleshooting Whole blood is not routinely used for the FCXM due to the high concentration of immunoglobulin in the plasma. Additionally, since the target cell of interest is the lymphocyte, contaminating cells of other lineages, specifically neutrophils, could bind antibody and produce a false-negative result. Cell preparations should not be treated with Lympho-Kwik or Percoll. Both these products increase the background fluorescence of target cells, which can significantly reduce the sensitivity of the assay (i.e., decreased signal-tonoise ratio) and may result in a false-negative result. Because flow cytometry is highly sensitive, all cell concentrations, serum dilutions, and volumes must be exact and accurate. Small fluctuations in cell numbers will result in erroneous and inconsistent results. The biggest potential for error lies in performing accurate cell counts. Excess cell numbers can produce falsenegative results. It is also important to remember to keep all reagents cold and protected from light.
6.16.18 Supplement 27
Current Protocols in Cytometry
A
100 80 M2
Class I beads
60 M1 40 20 0
CNTRBDS.001
0
1000
B
FL2-H
800
R3
200
R6
0 R2 200
1000
800
1000
400 600 800 FITC fluorescence
1000
M2 M1
5
200
0
600 M3
Class II beads 10
0
800
15
600 400
400
20
400 600 800 Forward scatter
1000
C
0
200
400
600
10 8
Control beads
6 4
M1
2 0
0
200
Figure 6.16.4 Example of a patient sample in which there is both an anti-HLA antibody and an antibody directed against the plastic bead. To analyze samples that contain anti-bead antibodies, a control bead is added (R6). The results of this gate are plotted on a single-parameter histogram like the class I and class II beads (C). Class I beads display a distinct and homogeneous increase in fluorescence compared to the control (A). However, the control beads display a similar increase in fluorescence, indicating the presence of an anti-bead antibody. Thus, the class I PRA = 0%. Owing to the shift in the control beads, the correct marker for analysis of the class II beads in histogram B is M3.
Anticipated Results Interpretation of FCXM results should be performed by a qualified individual experienced in both flow cytometry and histocompatibility. The question to be answered is: Does the patient sample produce a fluorescence value significantly greater than the negative control and is it clinically significant? This so-called “cut-off” value represents a level of fluorescence with respect to the negative control, above which it is believed that the change in fluorescence is attributable to the binding of an HLA alloantibody. Unfortunately, the most appropriate negative control does not exist. From a comparative standpoint, the best negative control is the serum of the patient with only the donor-directed HLA antibody adsorbed out. This then would provide the most accurate assessment of fluorescence background. Since this control is unavailable, the next best control
is a pool of serum from healthy individuals that are devoid of HLA antibody. Additionally, this control should have a protein and immunoglobulin concentration that approximates an average patient serum. Hence, based on a population average, the pooled negative control will be appropriate. However, for any given patient, this control may be suboptimal, i.e., will have a protein/immunoglobulin concentration that is generally higher than that of a dialysis patient awaiting transplantation. Thus, the final interpretation of the crossmatch must take into consideration the clinical and immunological history of the patient. Calculation of results For the flow cytometric crossmatch, the calculation of results can be achieved through at least three approaches. For any calculation of results, the main premise is, does the test sam-
Phenotypic Analysis
6.16.19 Current Protocols in Cytometry
Supplement 27
A
B
1000
1000 1 2 3 4 5 6 7
600
800
Specific beads (PE)
Specific beads (PE)
800
400
600
400
200
200 8 0
0 0
200
400 600 Anti-lgG-FITC
800
1000
Negative control
0
200
400 600 Anti-lgG-FITC
800
1000
Positive control
Figure 6.16.5 Histograms of class I (or class II) flow specific beads. (A) Beads tested with a negative control serum. (B) Beads stained with a positive control serum. Any bead that shows significant (>50% of the beads) increase in fluorescence intensity is considered positive.
Flow Cytometric Assessment of HLA Alloantibodies
ple exhibit fluorescence intensity that is significantly above the negative control? Contained within this statement are two difficult determinations. First is establishing a negative control value and range, and second is the determination of what is “significant” fluorescence above the control. Of these, the most critical determination is establishing a negative control. The negative control serum should approximate the background fluorescence of a patient sample that is devoid of HLA antibodies. For this to be true the negative serum must exhibit an average fluorescence. If the control serum underestimates average background fluorescence, then test results may be skewed towards false-positive results. In contrast, overestimating background fluorescence will lead to false-negative results. For these reasons, it is recommended that the control sera be from a pool of individuals and that two control sera be used. Once appropriate negative control sera are identified, calculation of test results can proceed. Determining a positive value is, in theory, very straightforward in that one must decide if the fluorescence value of the test serum is greater than that of the negative control. Such a determination would then indicate that HLA antibodies have bound to the target cell. For this calculation, the simplest approach is to utilize
channel values. The binding of an HLA antibody will produce an increase in fluorescence that will be measured by a channel displacement. If this displacement is significant, the test result will be deemed “positive.” How one decides what is “significant” will be dependent upon the variability of the negative control (e.g., c.v. of the negative control) as well as the sensitivity of the instrument. On average, using a 1024-channel scale, a displacement of ≥40 channels compared to the negative control may be considered as positive. Such a channel displacement, if converted to molecules of equivalent soluble fluorochrome (MESF), represents ∼1000 to 1500 MESF above background. Obviously, because of the inherent biological variability in this assay, a 40-channel displacement can be considered only a guideline. Therefore, displacements of 41 channels would not always be positive, nor would displacements of 39 channels always be negative. For values very close to the cut-point, patient history must be taken into consideration. Other approaches that may be used to represent the change in fluorescence would be (1) recording the actual linear fluorescence value and determining what a “significant” increase in fluorescence should be, and (2) converting fluorescence values to MESF. Within an indi-
6.16.20 Supplement 27
Current Protocols in Cytometry
A
B BATCH#5865.004
FBID98001.024
1000
1000
A1 A2
Specific beads (PE)
600
400
200
800 Specific beads (PE)
A2, 11; B13, 62 A2, 29; B7, 46 A3, 24; B55, 61 A3, 68; B7, 65 A3, 32; B50, 56 A3, 23; B18, 71
800
A3 A69 A66
600
A29
control beads
400
A30
200 A26
A32, 36; B53, 61
0 0
200
400 600 Anti-lgG-FITC
800
1000
0 0
200
400 600 Anti-lgG-FITC
800
1000
Figure 6.16.6 Class I flow-specific beads (A) and class I single-antigen beads (B) tested with patient sera containing anti-HLA antibodies. In plot A, the second bead from the top is considered positive, although it is not displaced across the line determined by the negative control. In this instance, the background fluorescence of this bead was less than that of the other beads when stained with a negative control serum. To accommodate these differences in background fluorescence, a box was drawn to delineate the position of this bead with the negative control. As shown, 100% of the beads clearly moved out of the box. (B) A positive patient serum with the specificities of anti-A2, A69, and A29. Note that the A26 bead is contained in the box, which has a fluorescence intensity that is less than that of the other beads but consistent with the negative control.
vidual laboratory the method used is not important. The most important requirement is that the laboratory set up its range and cut-point properly and perform appropriate quality control to ensure consistency in results reporting. As with the FCXM, analysis and interpretation of data for HLA detection by microparticle screening should be made by an individual who is appropriately qualified in both clinical histocompatibility and flow cytometry. Analysis of results, while not difficult, does require a good degree of expertise. The interpretation of a positive result is made by observing increases in fluorescence above background (i.e., negative control serum). In addition to a mere shift in fluorescence, the architecture of the histogram plot can also be of importance. Figures 6.16.3 and 6.16.4 show representative examples of results obtained with actual patient serum. Figure 6.16.3 shows class I and class II beads stained with either negative control serum (Fig. 6.16.3A, B) or serum from a patient who possesses both class I and class II antibodies (Fig. 6.16.3C, D). Figure 6.16.4 also shows a third
set of beads that have been coupled with human serum albumin and that serve as a control for anti-bead reactivity. The logical gate for this bead should be R7 = R1 and R6. In this example, the patient possesses antibodies against HLA class II antigens as well as an antibody that reacts with the bead plastic. Thus, the fluorescence increase seen with the class I beads is the result of an anti-bead antibody and not an HLA antibody. Failure to use a control bead may result in the incorrect reporting of HLA antibody. For each sample, both a percent PRA and antibody specificity can be determined from the flow specificity bead assay. After printing the results for a patient sample, assign the individual beads as positive or negative based on their shifts relative to the negative control markers. Any significant shift of the bead population to the right of the negative control marker is positive. A significant shift is considered as >50% of the bead population moving to the right of the negative control marker. If the population remains to the left of the negative control marker, then it is negative. Any population that
Phenotypic Analysis
6.16.21 Current Protocols in Cytometry
Supplement 27
C
FC99066.011
100
A
negative control 0% positive
82811.001 1000 800
Counts
80 60
M1
40
Side scatter
20 600
0 0
400
D
200
R1 0
200 400 600 800 1000 Forward scatter
B
Counts
0
82811.004
104 103 CD3-PCP
200
100 0
200 400 600 800 1000 Forward scatter
25%
1 cell positive
M1
200
E
Counts
101
M2
0
102
1000
FC99066.047
90 80 70 60 50 40 30 20 10 0
R2
800
400 600 Anti-lgG-FITC
70 60 50 40 30 20 10 0
400 600 Anti-lgG-FITC
800
1000
FC99066.059
0
2 cells positive
50% M2
M1
200
400 600 Anti-lgG-FITC
800
1000
Figure 6.16.7 Sample results from the flow cytometric PRA. Pooled lymphocytes are incubated with patient serum, then stained for IgG binding as well as for identification of CD3-positive cells. (A) Initially, a scatter gate is drawn around the apparent lymphocyte population (R1). (B) Next, a plot of FS versus CD3 is generated. From this plot, the CD3 cells are clearly delineated and an analysis gate is set (R2). A logical gate R3 = R1 and R2 is then generated and single-parameter histograms are displayed from this gate. Representative histograms are shown. (C) Results with a negative control serum. (D) Sample for which 1 cell (25% of the total events) was positive. (E) Sample for which 2 cells (50% of the total events) were positive. Depending upon the makeup of each pool and the distribution of positive and negative results, HLA specificities can be assigned.
Flow Cytometric Assessment of HLA Alloantibodies
straddles the line should be considered indeterminate until specificity is analyzed. Based on antibody assignment, the undetermined bead population may then be reassessed for positive or negative interpretation. The percent PRA may then be calculated based on the number of positive reactions divided by the panel size. Specificity analysis is based on the positive reaction pattern noted for a specific sample. Figure 6.16.5 illustrates an example of negative and positive serum samples. The quadrant marker for the negative control is placed to the right of the negative bead population. Note that the top bead (no. 1) in Figure 6.16.5 is an outlier (i.e., high background), as it appears to straddle the negative control line. To adjust for this difference, a box is drawn around this population to denote the
difference and indicate that when a patient sample is tested, these beads will be considered positive only when they have moved outside the delineated area. This type of adjustment is needed since it is not possible to control the background fluorescence of all bead preparations. Once the negative markers are established, they are not moved during the remainder of the patient analysis since they are used as points of reference for the shift in FITC fluorescence. The positive control shows that all bead populations have shifted to the right of their negative control markers, thus validating the test setup. This evaluation is repeated for each group of beads used. Figure 6.16.6 shows two illustrations of patient samples tested with the class I specificity beads (Fig. 6.16.6A) and the class I single-antigen beads (Fig. 6.16.6B).
6.16.22 Supplement 27
Current Protocols in Cytometry
Figure 6.16.7 shows some representative examples of negative and positive results for flow PRA analysis. Once positive reactions have been assigned, the probability of an HLA antibody can be evaluated by knowing the HLA phenotype of the cells included in each of the seven pools.
Time Considerations The time required for any of the above flow cytometric evaluations will vary depending upon the number of samples tested. As a guideline, the flow cytometric crossmatch can be completed in ∼2.5 to 3 hr provided that all samples are ready for testing. Flow cytometric antibody screening can be performed in about the same time frame. For flow cytometric specificity testing, the time required is generally longer due to the increase complexity of set up and analysis. In general, 4 to 6 hr would be required to complete specificity testing of ∼10 serum samples.
Literature Cited Bray, R.A. 1994. Flow cytometric crossmatching in solid organ transplantation. In Methods in Cell Biology: Flow Cytometry, Vol. 41 (Z. Darzynkiewicz, H.A. Crissman, and J.P. Robinson, eds.) pp. 103-119. Academic Press, New York. Bray, R.A. and Gebel, H.M. 2000. Transplantation immunophenotyping. In Immunophenotyping: Cytometric and Cellular Analysis (C. Stewart and J. Nicholson, eds.) pp. 321-332. Academic Press, New York. Bray, R.A., Lebeck, L.L., and Gebel, H.M. 1989. The flow cytometric crossmatch: Dual-color analysis of T and B cells. Transplantation 48:834-840. Bray, R.A., Cook, D.J., and Gebel, H.M. 1997. Flow cytometric detection of HLA alloantibodies using class I coated microparticles. Hum. Immunol. 55:36. Bray, R.A, Sinclair, D.A, Wilmoth-Hosey, L., Lyons, C., Chapman, P., and Holcomb, J. 1998. Significance of the flow cytometric PRA (FCPRA) in the evaluation of patients awaiting renal transplantation. Hum. Immunol. 59:121. Chapman, J.R., Deierhoi, M.H., Carter, N.P., Ting, A., and Morris, P.J. 1985. Analysis of flow cytometry and cytotoxicity crossmatches in renal transplantation. Transplant. Proc. 17:2480. Cook, D.J., Terasaki, P.I., Iwaki, Y., Terashita, G.Y., and Lau, M. 1987. An approach to reducing early kidney transplant failure by flow cytometry crossmatching. Clin. Transplant 1:253. Garovoy, M.R., Rheinschmilt, M.A., Bigos, M., Perkins, H., Colombe, B., Feduska, N., and Salvatierra, O. 1983. Flow cytometry analysis: A high technology crossmatch technique facilitating transplantation. Transplant. Proc. 15:1939.
Gebel, H.M. and Bray, R.A. 2000. Sensitization and sensitivity: Defining the unsensitized patient. Transplantation 69:1370-1374. Gebel, H.M., Bray, R.A., Ruth, J.A., Zibari, G.B., McDonald, J.C., Kahan, B.D., and Kerman, R.H. 2001. Flow PRA to detect clinically relevant HLA antibodies. Transplant. Proc. 33:477. Harris, S.B., Bray, R.A., Josephson, C.D., Hillyer, C.D., and Gebel, H.M. 2003. Presence of HLA antibodies in blood components: An unappreciated risk factor for transplant patients? Am. J. Transplant. 3:558. Karpinski, M., Rush, D.R., Jeffery, J., Exner, M., Regele, H., Dancea, S., Pochinco, D., Birk, P., and Nickerson, P. 2001. Flow cytometric crossmatching in primary renal transplant recipients with a negative anti-human globulin enhanced cytotoxicity crossmatch. J. Am. Soc. Nephrol. 12:2807-2814. Kerman, R., Gebel, H., Bray, R., Garcia, C., Renna, S., Branislav, R., Knight, R., Katz, S., Van Burren, C., and Kahan, B. 2002. HLA antibody and donor reactivity define patients at risk for rejection or graft loss. Amer. J. Transplant. 2:258. Lazda, V.A., Pollak, R., Mozes, M.F., and Jonasson, O. 1988. The relationship between flow cytometer crossmatch results and subsequent rejection episodes in cadaver renal allograft recipients. Transplantation 45:562. Le Bas-Bernardet, S., Hourmant, M., Valentin, N., Paitier, C., Giral-Classe, M., Curry, S., Follea, G., Soulillou, J.P., and Bignon, J.D. 2003. Identification of the antibodies involved in B-cell crossmatch positivity in renal transplantation. Transplantation 75:477-482. Muller-Steinhardt, M., Fricke, L., Kirchner, H., Hoyer, J., and Kluter, H. 2000. Monitoring of anti-HLA class I and II antibodies by flow cytometry in patients after first cadaveric kidney transplantation. Clin. Transplant. 14:85-89. Ogura, K., Terasaki, P.I., Johnson, C., Mendez, R., Rosenthal, J.T., Ettenger, R., Martin, D.C., Dainko, E., Cohen, L., Mackett, T., et al. 1993. The significance of a positive flow cytometric crossmatch test in primary renal transplantation. Transplantation 56:294-298. Pei, R., Wang, G., Tarsitani, C., Rojo, S., Chen, T., Takemura, S., Liu, A., and Lee, J. 1998. Simultaneous HLA class I and class II antibodies screening with flow cytometry. Hum. Immunol. 5:313-322. Pei, R., Lee, J.-H., Shih, N.-J., Chen, M., and Terasaki, P.I. 2003. Single human leukocyte antigen flow cytometry beads for accurate identification of human leukocyte antibody specificities. Transplantation 75:43-49. Piazza, A., Adorno, D., Poggi, E., Borrelli, L., Buonomo, O., Pisani, E., Valeri, M., Torlone, N., Camplone, C., Monaco, P.I., Fraboni, D., and Casciani, C.U. 1998. Flow cytometry crossmatch: A sensitive technique for assessment of acute rejection in renal transplantation. Transplant. Proc. 30:1769-1771.
Phenotypic Analysis
6.16.23 Current Protocols in Cytometry
Supplement 27
Piazza, A., Poggi, E., Borrelli, L., Servetti, S., Monaco, P.L., Buonomo, O., Valeri, M., Torlone, N., Adorno, D., and Casciani, C.U. 2001. Impact of donor-specific antibodies on chronic rejection occurrence and graft loss in renal transplantation: Posttransplant analysis using flow cytometric techniques. Transplantation 71:1106-1112. Scornik, J.C., Brunson, M.E., Schaub, B., Howard, R.J., and Pfaffa, W.W. 1994. The crossmatch in renal transplantation. Evaluation of flow cytometry as a replacement for standard cytotoxicity. Transplantation 57:621-625. Scornik, J.C., Clapp, W., Patton, P.R., Van der Werf, W.J., Hemming, A.W., Reed, A.I., and Howard, R.J. 2001. Outcome of kidney transplants in patients known to be flow cytometry crossmatch positive. Transplantation 71:1098-1102. Talbot, D., Givan, A.L., Shenton, B.K., Stratton, A., Proud, G., and Taylor, R.M.R. 1988. Rapid detection of low levels of donor specific IgG by flow cytometry with single and dual color fluorescence in renal transplantation. J. Immunol. Methods 112:279-283.
Thistlethwaite, J.R., Buckingham, M., Stuart, J.K., Garber, A.O., Mayes, J.T., and Stuart, F.P. 1987. T cell immunofluorescence flow cytometry cross-match results in cadaver donor renal transplantation. Transplant. Proc. 19:722. Utzig, M.J., Blumke, M., Wolff-Vorbeck, G., Lang, H., and Kirste, G. 1997. Flow cytometry crossmatch: A method for predicting graft rejection. Transplantation 63:551-554. Vaidya, S., Cooper, T.Y., Avandsalehi, J., Barnes, T., Brooks, K., Hymel, P., Noor, M., Sellers, R., Thomas, A., Stewart D., Daller, J., Fish, J.C., Gugliuzza, K.K., and Bray, R.A. 2001. Improved flow cytometric detection of HLA alloantibodies using pronase. Transplantation 71:422-428.
Contributed by Robert A. Bray and Howard M. Gebel Emory University Atlanta, Georgia Thomas M. Ellis The Blood Center of Southeastern Wisconsin Milwaukee, Wisconsin
Flow Cytometric Assessment of HLA Alloantibodies
6.16.24 Supplement 27
Current Protocols in Cytometry
Enumeration of Fetal Red Blood Cells, F Cells, and F Reticulocytes in Human Blood
UNIT 6.17
Recent advances in analytical cytometry have improved diagnostic tools for the study of erythropoiesis in anemic patients and resolution of the differential diagnosis in diseases of the erythron. This unit presents three applications of red blood cell (RBC) analysis— quantitation of fetal red cells (see Basic Protocol 1), F cell enumeration (see Basic Protocol 2), and F reticulocyte analysis (see Basic Protocol 3)—that improve diagnostic precision, sensitivity, and specificity, and provide better laboratory indicators of therapeutic efficacy in a variety of hematologic and obstetric disorders. Such advances also include the measurement and quantitation of RBC hemoglobins and their relative nucleic acid levels. These advances not only promise to improve diagnostic accuracy and laboratory precision over techniques such as the traditional manual reticulocyte counting method and the Kleihauer-Betke stain method for evaluating fetomaternal hemorrhage (FMH), but also serve as tools for newer assays of anemia diagnosis and improved clinical outcomes. In addition to the primary methods, supporting techniques for preparing spiked controls (see Support Protocol 1), setting up a fetal hemoglobin acquisition protocol (see Support Protocol 2), and assaying reticulocytes using thiazole orange (see Support Protocol 3) are also presented. QUANTITATION OF FETAL RED CELLS BY FLOW CYTOMETRY Flow cytometric methods are more sensitive and precise than routine visual counting and reliably detect low levels of fetal RBCs in circulating maternal blood. The following protocol for the detection of fetal hemoglobin (HbF)–containing RBCs uses a monoclonal antibody conjugated to fluorescein isothiocyanate (FITC), a method for the intracytoplasmic staining of red blood cells, and WinList or any other equivalent Software (Davis et al., 1998; Chen et al., 2000); however, the method can be easily adapted for use with anti-HbF monoclonal antibody reagents conjugated to other fluorochromes and with other data-analysis software.
BASIC PROTOCOL 1
Materials Whole-blood samples anticoagulated with EDTA or other suitable anticoagulant (test within 4 hr of collection or store ≤72 hr at 4°C) Fetaltrol stabilized three-level controls (Caltag Laboratories) or spiked control blood samples (see Support Protocol 1) PBS/0.1% BSA (see recipe) 0.05% glutaraldehyde (see recipe) 0.1% Triton X-100 (see recipe) Anti-HbF antibody dilution (see recipe) 1% formaldehyde fixative (see recipe) 12 × 75–mm disposable polystyrene tubes (Falcon) with rack DAC II automatic cell washer (Baxter) or equivalent (optional) Multiparameter flow cytometer with 488-nm excitation and filters for detection of green and orange fluorescence WinList Software (Verity Software House; http://www.vsh.com) or equivalent Additional reagents and equipment for counting cells (APPENDIX 3A) and acquisition of data (see Support Protocol 2) Phenotypic Analysis Contributed by Bruce H. Davis and Kathleen Thompson Davis Current Protocols in Cytometry (2004) 6.17.1-6.17.16 Copyright © 2004 by John Wiley & Sons, Inc.
6.17.1 Supplement 28
Table 6.17.1
Volume to Fix for a Given RBC Concentration
RBC X (µl) (× 106/µl)
RBC X (µl) (× 106/µl)
RBC X (µl) (× 106/µl)
5.0 4.9 4.8 4.7 4.6 4.5 4.4 4.3 4.2 4.1 4.0
3.9 3.8 3.7 3.6 3.5 3.4 3.3 3.2 3.1 3.0 2.9
2.8 2.7 2.6 2.5 2.4 2.3 2.2 2.1 2.0 1.9 1.8
5.0 5.1 5.2 5.3 5.4 5.6 5.7 5.8 6.0 6.1 6.3
6.4 6.6 6.8 6.9 7.1 7.4 7.6 7.8 8.1 8.3 8.6
8.9 9.3 9.6 10.0 10.4 10.9 11.5 11.9 12.5 12.9 13.5
1. Mix anticoagulated whole blood samples and spiked control blood samples or Fetaltrol stabilized three-level controls by rolling the tubes ten times between the hands. Place tubes on a rocker to continue mixing and allow to reach ambient temperature if chilled. Run all three controls (negative, low positive, and high positive) with each batch of patient specimens.
2. Measure the RBC count for each patient sample and each control (APPENDIX 3A). If the RBC count is >5.0 × 106 cells/µl, dilute 1:2 with PBS/0.1% BSA and count again. 3. Using Table 6.17.1, determine the volume of blood for processing (denoted by X). X contains ∼2.5 × 107 RBCs. This volume of RBCs in the fixation tube will result in ∼5 × 105 cells in the staining tube.
4. Label one 12 × 75–mm disposable polystyrene tube for each control and two for each patient sample. Place in a rack. Patient samples are run in duplicate.
Fix cells 5. Fix X µl whole blood or control in 1 ml cold 0.05% glutaraldehyde 10 min at room temperature. Vortex immediately for 15 sec at high speed. In subsequent steps, when instructed to vortex, do so using these conditions. Note that other fixatives may reduce red cell lysis.
6a. For washing with a cell washer: Load the tubes into a DACII automatic cell washer and select three wash cycles using flow cytometric isotonic sheath fluid. 6b. For washing by centrifugation: Wash cells with 2 ml PBS/BSA and centrifuge 5 min at 200 × g, at room temperature. 7. Resuspend the cell pellet by vortexing. Add 500 µl of 0.1% Triton X-100. Mix by vortexing. Enumeration of Fetal Red Blood Cells, F Cells, and F Reticulocytes in Human Blood
8. Incubate 3 to 5 min at room temperature. 9. Load tubes into the DACII automatic cell washer and select one wash cycle (see step 6a). Alternatively, perform a manual wash as described (see step 6b).
6.17.2 Supplement 28
Current Protocols in Cytometry
10. Resuspend the cell pellet by vortexing. Add 500 µl PBS/BSA and mix by vortexing. 11. During the above incubation and washes, prepare tubes for the staining step. Label two 12 × 75–mm tubes for each control and three for each duplicate patient sample. A total of six tubes is labeled for each patient.
Stain cells 12. To each tube add 10 µl anti-HbF antibody dilution and 80 µl PBS/BSA. 13. Add 10 µl (∼1–2 × 105 cells) of the suspensions made in step 10 to the tubes containing Ab. Mix by vortexing. 14. Incubate 15 min in the dark at room temperature. 15. Load the tubes into a DACII automatic cell washer, select two wash cycles, and start the machine (step 6a). Alternatively, perform two manual washes (step 6b). 16. Resuspend the cell pellet by vortexing. Add 500 µl of 0.1% formaldehyde. Vortex to mix. 17. Store tubes up to 24 hours at 4°C until ready for flow cytometric acquisition. Acquire sample data 18. Perform cytometer start up and run daily quality control to verify proper operation. If all instrument checks are satisfactory, launch the acquisition protocol (see Support Protocol 2). Instrument settings must be checked and/or adjusted if there are any changes in the laser. A change in the laser will require new voltage settings for the detectors and PMTs. When selecting new voltage settings, place the negative population fully within the first decade.
19. Run tubes in the following order: Patient sample tubes Blank tube of water or buffer Fetaltrol Level 1 or negative control Fetaltrol Level 2 or low-positive (0.15%) control Fetaltrol Level 3 or high-positive (1.5%) control. If there are specimens from more than one patient, run a blank (water) for 30 sec between patient sets (six tubes) to prevent carryover. It is not necessary to run a blank between the three levels of controls since they are known levels arranged from lowest to highest. Carryover will not affect the results.
20. Collect at least 50,000 ungated events at high flow rate for each specimen. 21. Name and save listmode files. Analyze data 22. Select the HbF macro in WinList software to bring up the analysis template. 23. Gate on RBCs (R1) using the orange fluorescence versus SS parameters to exclude autofluorescent (orange positive) events of intensity greater than the first decade (Fig. 6.17.1A). 24. Using the Fetaltrol Level 3 or high-positive control files, set region markers closely around the fetal RBC peak to define the fetal cell region R4 (Fig. 6.17.1B). Include all 50,000 events by clicking on the Data Source bar.
Phenotypic Analysis
6.17.3 Current Protocols in Cytometry
Supplement 28
B 104
250
103
200
adult cells
150
R2 F cells
Number
Autofluorescence
A
102 R1
fetal cells R4
101
100 100
R3
100 50
101
102 103 Side scatter
104
0 100 101 102 103 104 Green fluorescence (HbF-FITC)
Figure 6.17.1 Analysis of anti-HbF stained Fetaltrol sample with 1.5% fetal cells. Listmode files are collected with parameters of light scatter, green fluorescence (anti-HbF FITC signal), and orange fluorescence (autofluorescence signal). Analysis is performed by gating events to exclude small particles and autofluorescent nucleated cells as shown as the R1 region in panel A with the R1 gated events displayed as a single fluorescence histogram of the HbF distribution as shown in panel B. The distribution of HbF in red cells typically shows three regions of adult red cells lacking HbF content (R2), adult F cells with an intermediate level of HbF (R3), and fetal red cells (if present) with high levels of HbF (R4), as shown in panel B.
25. With the R4 region marker unchanged, open the Fetaltrol Level 2 and Level 1 or low-positive and negative control files. 26. Report the % gate value for R4 as the value representing the percentage of HbF-positive cells. 27. Evaluate the control values. Results must fall within the reference (Fetaltrol) or established laboratory range (in-house controls). If they fall outside, troubleshoot the procedure and repeat the batch. 28. If the controls are within range, open the patient files. The R4 region marker will remain unchanged, but the R1 gate may need to be adjusted. Include all events. 29. Report the % gate value for R4 as the percentage of HbF-positive fetal red blood cells. Normal values for adult blood: HbF (fetal cells) ≤0.07%
Patient results which fall outside the range cannot be reported. BASIC PROTOCOL 2
Enumeration of Fetal Red Blood Cells, F Cells, and F Reticulocytes in Human Blood
F-CELL ENUMERATION BY FLOW CYTOMETRY The procedure for sample staining and flow cytometric analysis is identical to that used for fetal red cell enumeration. A no-wash modification of the assay has also been reported (Mundee et al., 2000); however, it is recommended that laboratories first gain experience with the original method, which has been more extensively validated. Refer to the detailed description of the fetal red cell enumeration method (see Basic Protocol 1), paying particular attention to the instrument setup instructions on how to adjust the fluorescence detection settings such that the peak positions of cells lacking HbF and autofluorescent leukocytes are similarly positioned on both fluorescence parameters. This instrument setup then allows use of the autofluorescence signal to determine the thresholds or cursor positions for counting F cells based upon the anti-HbF antibody fluorescence with a high level of precision.
6.17.4 Supplement 28
Current Protocols in Cytometry
Data are analyzed using WinList (Verity Software House) or other equivalent software . The instrument calibration and adjustment for the quantitation of F cells result in the placement of the predominant RBC population fully within the first decade of the log fluorescence scale at an equal amount of green and orange intensity (see Fig. 6.17.2). The color compensation is adjusted so that the autofluorescence signals of reticulocytes, leukocytes, and other nucleated cells in an unstained glutaraldehyde-fixed blood sample fall on the 45° angle or line of equivalence in the green versus orange fluorescence histogram (Fig. 6.17.2A). Data analysis for F-cell enumeration on any sample is performed to initially exclude the autofluorescent nucleated cells, which are primarily leukocytes, from gates set on histograms showing the fluorescent cutoff between F cells and other erythrocytes (Fig. 6.17.2B). The orange fluorescence is used to define the cursor position or threshold for F cells with the FITC-conjugated anti-HbF antibody by finding the channel at which ≥99.8% of the cells are to the left of the cursor or ≤0.02% of the orange fluorescent cells are to the right of the cursor (Fig. 6.17.2C). The same channel position of the cursor from the orange autofluorescence histogram is then used as the cursor position on the green fluorescence histogram, thus defining the F cells as all events to the right of that cutoff position (Fig. 6.17.2D).
C 104
500
103
400 Number
Orange fluorescence
A
102
300 200
101
100
100 100
101 102 103 Green fluorescence
B
0 100
104
D 104
1100 900
HbF-
103 700 leuk
Number
Orange fluorescence
101 102 103 Orange fluorescence
102
F cells
500 300
101
RBC 100
100
101
102 103 Side scatter
104
100
102 103 101 Green fluorescence
Figure 6.17.2 Histogram setup for analysis of F cells as described above in the text. Normal values: F cells in 40 adult normals = 3.8% ± 3.6%.
Phenotypic Analysis
6.17.5 Current Protocols in Cytometry
Supplement 28
SUPPORT PROTOCOL 1
PREPARATION OF SPIKED CONTROL BLOOD SAMPLES It is imperative that multilevel assayed control samples be used as a means of assessing both the procedure and the analysis. Low staining intensity, increased F-cell levels, and improper compensation and gating can seriously impact the validity of the generated results. There are two sources of quality-control samples: Fetaltrol and in-house preparations. Fetaltrol, a stabilized three-level blood control product produced by Trillium Diagnostics (http://www.trilliumdx.com), is a convenient alternative to “home-brew” controls. It contains vials of negative, low-level positive (∼0.15% fetal RBCs), and high-level positive (∼1.5% fetal RBCs) controls, and has a 3-month shelf life. This product is used exactly like whole blood in the protocols. Alternatively, spiked control blood samples may be prepared in house following the procedure outlined here. Additional Materials (also see Basic Protocols 1 and 2) Peripheral blood from a healthy nonpregnant adult Alsever’s solution (see recipe) Umbilical-cord or newborn blood Anticoagulant (e.g., EDTA)-containing vacutainer Additional reagents and equipment for collection of EDTA-anticoagulated blood and determining blood ABO grouping Prepare negative control 1. Using an anticoagulant-containing vacutainer, collect a whole-blood sample from a healthy nonpregnant adult. 2. Take 1 ml of this normal sample and immediately add 1 ml Alsever’s solution. Mix well. 3. Measure and record the new RBC count (APPENDIX 2A). 4. Store ≤1 week at 4°C. Prepare positive controls 5. Obtain whole blood from an umbilical cord or newborn, in an anticoagulant-containing vacutainer. Determine the ABO grouping. 6. Wash cells three times with 5 ml PBS/0.1% BSA. 7. From a healthy nonpregnant adult, obtain a fresh ABO-compatible whole blood sample, anticoagulated with EDTA, with normal hematology CBC parameters. 8. Spike 1 ml adult whole blood with the washed cord cells (see step 4) to make 1.5% and 0.1-0.2% mixtures, which can serve as high and low level fetal RBC controls. 9. Immediately add 1 ml Alsever’s solution to each spiked mixture, mix well, and determine and record the new RBC count. Store ≤2 weeks at 4°C.
SUPPORT PROTOCOL 2
Enumeration of Fetal Red Blood Cells, F Cells, and F Reticulocytes in Human Blood
SETTING UP A FETAL HEMOGLOBIN ACQUISITION PROTOCOL The acquisition protocol for the fetal red cell assay requires a multi-parameter instrument with a 488-nm excitation laser and the ability to collect light scatter and at least two fluorescence parameters. This procedure describes the optimal instrument setup to allow for both fetal red cells and adult F cells. Although the procedure uses only a single fluorescently labeled antibody (anti-HbF), the collection of a second fluorescent signal for autofluorescence is important for both exclusion of nucleated cells from analysis and accurate enumeration of F cells (Chen et al., 2000). Once the acquisition protocol is established and stored on the instrument as described below, it does not need to be performed again unless major service alignment adjustments, PMT replacements, or laser replacements are performed.
6.17.6 Supplement 28
Current Protocols in Cytometry
Additional Materials (also see Basic Protocol 1) Blood sample, preferably with high leukocyte count (>25 × 109 cells/liter) Fetaltrol Level 3 or adult blood spiked with washed ABO-compatible cord cells (see Support Protocol 1) 1. Select two blood samples, one preferably with a high WBC and the other either Fetaltrol Level 3, or adult blood spiked with washed ABO-compatible cord cells. 2. Stain as described (see Basic Protocol 1, steps 1 to 17) and run on the flow cytometer. 3. Draw the following histograms: Log FS versus log SS HbF-FITC (green fluorescence) versus log FS Autofluorescence (orange fluorescence) versus log FS Autofluorescence (orange fluorescence) versus HbF-FITC (green fluorescence) Number versus HbF-FITC (green fluorescence). 4. While running the tube containing the stained spiked blood (or Fetaltrol), adjust FS and SS so that the RBC population lies midscale on both axes. 5. Adjust threshold on FS to eliminate unwanted events (platelets, cell debris) with a lower signal than that of the red cell population. 6. Adjust FITC (green fluorescence) and autofluorescence (orange fluorescence) so that the red cell peaks are in the center of the first decade for both parameters and the entire peak can be visualized on the histogram. 7. While running the tube containing the stained sample with the high WBC, adjust orange minus green compensation so that the (stained) RBCs along the green fluorescence axis will fall mostly within the first decade of the orange fluorescence axis, and WBCs will line up on a 45° diagonal. 8. Set acquisition to collect at least 50,000 events in a listmode file with all parameters (FS, SS, green fluorescence, and orange fluorescence). Name and save the instrument settings and the protocol template. F-RETICULOCYTE ANALYSIS Reliable reticulocyte analysis on blood samples is considered to be the least expensive and fastest way to evaluate human bone marrow erythropoietic production. Flow cytometry is presently the preferred and most precise method for reticulocyte measurement (Davis, 2001a,b). Hence the measurement of F-cell production can be achieved by combining the monoclonal HbF assay with the thiazole orange reticulocyte assay. The cells that are the newly released precursors to mature circulating F cells are called F reticulocytes. This measurement of F reticulocytes can be used to specifically monitor the rate of F-cell production, such as in various anemias or myelodysplasia or in evaluation of new therapies in patients with sickle cell disease or other hemoglobinopathies (Nagel et al., 1993; Maier-Redelsperger et al., 1998a,b; Bohmer et al., 2000; Mundee et al., 2001). Materials Whole-blood sample, anticoagulated with EDTA or other suitable anticoagulant (test within 6 hr of collection or store ≤72 hr at 4°C) PE-Cy5 (Tri-color)–conjugated anti-HbF monoclonal antibody (Caltag Laboratories) 1% formaldehyde fixative (see recipe)
BASIC PROTOCOL 3
Phenotypic Analysis
6.17.7 Current Protocols in Cytometry
Supplement 28
Retic-Count kit (thiazole orange; BD Biosciences) Sheath fluid: filtered PBS (0.40 µm pore) Retic-COUNT solution (BD Biosciences) or an alternative thiazol orange solution in filtered PBS Multiparameter flow cytometer with 488-nm excitation and filters for collection of green and red fluorescence Additional reagents and equipment for staining blood samples (see Basic Protocol 1, steps 1 to 17) and thiozole orange reticulocyte analysis (see Support Protocol 3).
A
B 104
Autofluorescence
104
Side scatter
103 102 101
103
leukocytes
R4
102 R2
101
100 100
101 102 103 Forward scatter
104
C
101
102 103 Side scatter
104
D 104 103 102
700 leukocytes
600
R3
Number
Autofluorescence
100
101
500 400 300 200
fetal cells (R4)
100 100 100 101 102 103 104 Green fluorescence (HbF-FITC)
Enumeration of Fetal Red Blood Cells, F Cells, and F Reticulocytes in Human Blood
100 101 102 103 104 Green fluorescence (HbF-FITC)
Figure 6.17.3 Representative histograms following optimal setup of protocol for analysis of fetal red cells in blood sample stained with anti-HbF monoclonal antibody. The positioning of the “negative” population of red cells lacking HbF should be within the first decade for both the anti-HbF and autofluorescence signals, and the leukocyte population with high level of autofluorescence should be positioned on a 45° diagonal. Panel A shows the typical light scatter signal of red cells. Panel B shows the higher level of autofluorescence shown by leukocytes (and other nucleated cells) gated in R1 and the red cells lacking significant autofluorescence (large cluster), which is the population gated in R2 for eventual analysis for fetal cells. Panel C demonstrates the optimal instrument setup such that the two fluorescent signals have balanced and equal intensities for the autofluorescent leukocyte or nucleated cell cluster (R3). Panel D demonstrates the HbF expression of the red cell population gated as shown in panel B with the fetal red cell region of analysis indicated by the bracketed region (R4).
6.17.8 Supplement 28
Current Protocols in Cytometry
Red fluorescence (HbF-PE-Cy5)
Red fluorescence (HbF-PE-Cy5)
104 103
102
101
100 0
10
1
2
3
10 10 10 Green fluorescence
4
10
104 103
102
101
100 100
101 102 103 Thiazole orange
104
Figure 6.17.4 F-reticulocyte analysis using dual staining for HbF and RNA content. Blood samples stained with anti-HbF (PE-Cy5) to identify F cells and non-F cells based upon level of red fluorescence are seen in both panels. The reticulocyte component of both red cell types can be calculated by comparing the green fluorescence of autofluorescence in samples without thiazole orange staining (left panel) to that seen with co-staining with thiazole orange (right panel).
Stain specimens 1. Stain two aliquots of each blood sample as described (see Basic Protocol 1, steps 1 to 17) using PE-Cy5 (Tri-color)-conjugated anti-HbF monoclonal antibody. Before beginning fixation, perform a standard thiazole orange reticulocyte analysis (see Support Protocol 3). 2. Resuspend the cell pellet in one tube by vortexing in 500 µl of 1% formaldehyde fixative. Store at 4°C shielded from light until ready to run on the flow cytometer. This tube is now ready for data acquisition.
3. Resuspend the cell pellet in the remaining tube by vortexing in 0.5 ml Retic-COUNT reagent (or an alternative thiazol orange solution) from the Retic-COUNT kit. Incubate 30 min in the dark at room temperature. Acquire data 4. Prepare a calibrated and standardized flow cytometer that has been properly quality controlled according to laboratory procedure. 5. Run the tube not exposed to thiazole orange first, while the second tube is incubating with the Retic-COUNT reagent (or an alternative thiazole orange solution). Be sure to run a blank tube containing sheath fluid after every patient specimen. 6. Run the thiazole orange/1Hb-F dual-stained tubes after the formaldehyde-fixed tubes and within 1 hr of the initial incubation with thiazole orange. Be sure to run a blank tube containing sheath fluid after every patient specimen to avoid specimen carryover. Use the formaldehyde-fixed sample as a guide to adjust cursor or quadstat positions to define regions for F reticulocytes (thiazole orange and HbF dual-positive) and reticulocytes of the non-F cell population (thiazole orange positive and HbF negative).
Phenotypic Analysis
6.17.9 Current Protocols in Cytometry
Supplement 28
7. Collect log FS, log SS, log green fluorescence (thiazole orange), and log red fluorescence (HbF-PE-Cy5) in listmode. Set the threshold on the FS signal and collect 50,000 events at high flow rate for each specimen. The same voltage settings for the detectors are used consistently for standardizing results. A change in the laser, the FS detector, the SS PMT, the green fluorescence PMT, or the red fluorescence PMT will require adjustment of these settings. When selecting a new voltage setting, place the negative population for both green fluorescence and red fluorescence fully within the first decade. It will be necessary to adjust the green from red compensation to achieve histogram displays as shown in Figure 6.17.4. SUPPORT PROTOCOL 3
RETICULOCYTE ASSAY USING THIAZOLE ORANGE A standard thiazole orange reticulocyte analysis is performed for each patient specimen just prior to beginning the fixation process for the HbF staining procedure. This provides assurance that the result of the F-reticulocyte assay correlates with standard reticulocyte analysis results. See Basic Protocols 1 and 3 for materials. Stain samples 1. Label two 12 × 75–mm polystyrene tubes for each patient specimen. One tube is for the stained sample and one for the unstained (control) sample.
2. Add 1 ml Retic-COUNT reagent (or alternative thiazol orange solution) to the stained sample tube and 1 ml PBS/0.1% BSA to the unstained (control) sample tube. 3. Add 5 µl well-mixed whole blood from the same specimen to each tube. Cap and vortex gently. 4. Incubate both tubes 30 min in the dark at room temperature. Samples are now ready to run on the flow cytometer.
Acquire data 5. Perform daily quality control for the flow cytometer according to laboratory procedure, ensuring the instrument is in acceptable working order. 6. Set up a protocol that collects log FS, log SS, and log green fluorescence. 7. Adjust FS and SS such that the red blood cells are in the upper right of the histogram and fully on scale. 8. Using a control tube, adjust the green fluorescence signal such that the peak is in the first decade of the single-parameter histogram (Fig. 6.17.5). If necessary, increase the FS threshold to exclude noise and debris. 9. Acquire 50,000 events from the unstained sample tube using a low flow rate. Save the listmode data. Run all unstained tubes before stained tubes. Do not change any settings. 10. Acquire 50,000 events from the stained sample tube using a low flow rate. Save the listmode data. Analyze data 11. Gate on the red cell population based on light-scatter properties. Define the reticulocyte population using the unstained sample by setting the cursor such that 0.1% of the cells lie to the right of the cursor on the thiazole orange fluorescence signal. Enumeration of Fetal Red Blood Cells, F Cells, and F Reticulocytes in Human Blood
12. Use this cursor position to analyze the thiazole orange–stained sample.
6.17.10 Supplement 28
Current Protocols in Cytometry
500
103
400 Number
Forward scatter
104
102
101
100 100
300 200 100 0
101
102
Side scatter
103
104
100
101 102 103 Green fluorescence
104
Figure 6.17.5 Thiazole orange staining of blood samples for reticulocytes. Thiazole orange can label both the platelets (cluster in middle) and red cells (upper right cluster) shown by light scatter parameters in the left panel. The red-cell cluster is gated to analyze the green fluorescence distribution or RNA content of the cells, with reticulocytes being defined by autofluorescence of red cells as shown in the right panel.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Alsever’s solution Dissolve 0.50 g citric acid (C6H8O7), 20.50 g dextrose (C6H12O6), 4.20 g sodium chloride (NaCl), and 8.00 g trisodium citrate (C6H5Na3O7⋅2H2O) in ∼600 ml water. Add 0.33 g chloramphenicol, 2.00 g inosine, and 0.50 g neomycin sulfate; mix well. Adjust to 1 liter with water and sterilize using a 0.2-µm vacuum filter. Adjust pH to 7.2, using aliquots to determine pH and aseptic technique to avoid contaminating the filtered solution. Store ≤3 months at 4°C; discard if any visible growth is detected. Anti-HbF antibody dilution FITC-conjugated anti-HbF antibody (Caltag Laboratories or Maine Biotechnology Services): use 5 or 10 µl undiluted antibody per test as determined by titration. Anti-HbF optimal concentration is usually ∼2.5 µg/ml, but it is determined by examining a series of dilutions (two fold dilutions between 0.1 and 10 µg/ml) with samples with a mixture of fetal red cells and adult red cells containing at least 1% fetal red cells. The antibody concentration should be sufficient to provide the optimal fluorescence separation between the negative adult RBCs and the positive fetal cells. Additionally, the concentration should be sufficient to have no change in fetal red cell fluorescence intensity for samples up to 10% fetal cells. Glutaraldehyde, 0.05% 50 µl 25% glutaraldehyde (Sigma; store at −20°C) 25 ml PBS, pH 7.4, without BSA (CellGro or Sigma) Mix well and keep at 4°C Make fresh working solution each day just prior to use and use cold Stability past 2 hr is uncertain
Phenotypic Analysis
6.17.11 Current Protocols in Cytometry
Supplement 28
Formaldehyde fixative, 1% Combine 15 ml of 10% formaldehyde (methanol free; Polysciences) and 135 ml PBS/0.1% BSA (see recipe). Sterilize using a 0.2-µm vacuum filter. Adjust pH to 7.4 when solution reaches room temperature, using aliquots to measure pH and aseptic technique to avoid contaminating the filtered solution. Store ≤1 week at 4°C. PBS/BSA, 0.1% 1 packet phosphate-buffered saline powder (Sigma; available through Diamedix) 1000 ml H2O 1.0 g BSA (Sigma) Mix well and sterilize using a 0.2-µm vacuum filter Adjust pH to 7.4 when solution reaches room temperature Store ≤1 week at 2° to 8°C Triton X-100, 0.1% 50 µl Triton X-100 (Sigma) 49.95 ml PBS/0.1% BSA (see recipe) Mix well and store ≤1 month at 4°C Discard at detection of visible growth Use cold COMMENTARY Background Information
Enumeration of Fetal Red Blood Cells, F Cells, and F Reticulocytes in Human Blood
Fetomaternal hemorrhage and fetal RBC detection The most important clinical use of fetal RBC detection is the diagnosis and quantitation of fetomaternal hemorrhage (FMH; Sebring and Polesky, 1990; Davis, 2001a,b ; Davis et al., 2001). FMH occurs normally throughout pregnancy in minute amounts, with increasing volumes during the later stages of gestation (Giacoia, 1997). Any significant difference in the RBC antigenicity between fetus and mother can result in allosensitization of the maternal immune system either before or after parturition. The maternal antibodies to the fetal RBC antigens may be clinically silent or cause lifethreatening autoimmune sequelae for current or subsequent pregnancies (e.g., erythroblastosis fetalis, early abortion). Such sensitization can occur with any RBC antigen mismatch, but the highest frequency and profound clinical consequences occur with Rh or D-antigen mismatches. Detection and enumeration of fetal RBCs is an essential part of the management of those patients with FMH treated with Rh immune globulin (RhIG) preparations, such as RhoGam (Ortho Pharmaceutical; Polesky and Sebring, 1981; Sebring and Polesky, 1990). The use of Rh immune globulin prophylaxis is a universal practice, but dosing amounts and schedules have regional variations (Hartwell, 1998; Lee et al., 1999). Hence, the sensitivity
and specificity of detection assays for FMH is a critical factor in therapeutic efficacy and subsequent clinical outcome. Heretofore, the most widely used assay for FMH detection has been the visual microscopic counting Kleihauer-Betke (KB) method, which is based upon the differences in solubility properties in acid conditions between fetal hemoglobin (HbF) and adult hemoglobin (Kleihauer et al., 1957). The KB technique is easily performed by most clinical laboratories, but lacks sensitivity and exhibits poor reproducibility and precision (CVs of 50% to 100%). The problematic nature of the KB assay was demonstrated in the recent proficiency surveys (HBF) conducted by the College of American Pathologists, in which two surrogate blood samples were selected at values either side of the 0.6% fetal RBC level, indicative of a FMH of ∼30 ml, typically used in the U.S. as the trigger level for additional Rh immune globulin therapy. The performance by laboratories using the KB method in this survey was less than optimal for a laboratory test used for therapeutic monitoring, as ∼12% of labs reported results of ≤0.6% for the sample with >0.8% fetal RBC and >45% of labs reported results of ≥0.6% for the sample with ≤0.4% fetal RBC. Such performance in this laboratory survey indicates that clinical practice using the KB assay is likely associated with a significant amount of both under- and over-administration of Rh immune globulin to patients with signifi-
6.17.12 Supplement 28
Current Protocols in Cytometry
cant FMH. Similar reported experiences with the KB assay have also brought into question the clinical validity of this manual method (Emery et al., 1995; Ducket and Constantine, 1997). Flow cytometric methods have recently been developed that improve both sensitivity and precision by using monoclonal antibodies to HbF, i antigen, and D antigen (Nance et al., 1989; Thorpe et al., 1994; Lloyd-Evans et al., 1996; Davis et al., 1998; Navenot et al., 1998; Nelson et al., 1998; Campbell et al., 1999; Chen et al., 2002). These methods reliably detect low levels of fetal RBCs in maternal circulation (<0.05%) and can more reliably determine the necessity for and dosage of Rh immune globulin in clinical practice. Use of the anti-HbF method for FMH testing may be preferred, as there are potential drawbacks to anti-D and anti-i immunophenotypic methods. These antigens, although expressed on the RBC surface, are not specific for fetal RBCs and may not be applicable in all clinical situations. For example, the monoclonal anti-D reagents, although clinically useful in the case of Rh incompatibility between a D-negative mother and a D+ fetus, are not applicable for the evaluation of FMH in cases of D+ and Du women following trauma or amniocentesis. The i antigen is a neoantigen that can be expressed on adult RBCs during some viral infections and paroxysmal nocturnal hemoglobinuria (PNH), as well as related disorders (Cooling, 1997; Navenot et al., 1997). The more recent approach to fetal RBC enumeration detects FMH using flow cytometry and anti-HbF monoclonal antibodies, which, as sold by Caltag Laboratories, are cleared for use by the U.S. Food and Drug Agency, and have broader clinical applicability (Thorpe et al., 1994; Davis et al., 1998; Nelson et al., 1998; Chen et al., 2002). Fetal RBCs are uniquely identified on the basis of their high intracellular content of HbF and can be readily distinguished from F cells (adult RBCs containing both HbF and other adult hemoglobins) and other adult RBCs on the basis of quantitative HbF expression (Fig. 6.17.1; Davis et al., 1998; Chen et al., 2000; Mundee et al., 2000). The flow cytometric anti-HbF methods exhibit excellent intermethod correlations with the KB test and are capable of detecting <0.02% fetal RBCs with precision considerably superior to that of the KB method (CVs <15%; Bromilow and Duguid, 1997; Campbell et al., 1998; Davis et al., 1998). The identification of fetal nucleated RBCs in maternal circulation also represents a rela-
tively noninvasive means of prenatal genetic testing using a maternal blood sample (Steele et al., 1996; Bianchi, 1999). Initial studies utilized multiparametric flow cytometry for the identification and cell-sorted purification of fetal cells through cell surface antigen differences from maternal cells; however, more recent advances have used intracellular HbF expression to obtain purified fetal nucleated RBCs (Cheung et al., 1996; Bischoff et al., 1998; Bianchi, 1999). Nucleated fetal RBCs can then be studied by fluorescence in situ hybridization or other molecular techniques to identify genetic abnormalities in the fetus. This diagnostic approach to screening for well-characterized genetic abnormalities is still a rapidly evolving field, but promises to provide a safer alternative to amniocentesis or chorionic villus sampling procedures early in gestation. Intracellular hemoglobin detection in RBCs can be exploited for other applications, since monoclonal antibodies exist for many of the hemoglobin types, such as hemoglobin S (HbS). Thus, the flow cytometric method can be used to look at patients with hemoglobin variants, such as HbSS disease and HbS trait (Thorpe et al., 1994; Campbell et al., 1999). Cost considerations make it unlikely that this approach will replace standard hemoglobinopathy evaluation using electrophoretic and high-pressure liquid chromatography methods; however, the multiparametric capability of flow cytometry does allow for a cell-by-cell determination of hemoglobin content, which may be valuable in some clinical research studies. In addition to the improved detection of fetal red cells, another important methodological improvement afforded by the anti-HbF flow cytometric technique is the detection of fetal hemoglobin–containing RBCs, called F cells, produced throughout life by all individuals (Davis et al., 1998; Maier-Redelsperger et al., 1998a; Thein and Craig, 1998; Campbell et al., 1999; Chen et al., 2000; Garner et al., 2000; Davis, 2001b). F-cell quantitation is potentially useful in the evaluation and therapeutic monitoring of sickle cell disease, thalassemias, other hemoglobinopathy conditions, myelodysplastic syndromes, and hemolytic anemias. Adult F cells are defined as red cells having HbF as only one of the cellular hemoglobin component of the cell along with other hemoglobin forms. F cells usually represent <5% of the RBC population in adult blood (Thein and Craig, 1998; Chen et al., 2000; Garner et al., 2000). The number of F cells increases in acquired anemias or other conditions of increased hematopoietic
Phenotypic Analysis
6.17.13 Current Protocols in Cytometry
Supplement 28
activity, as well as in a normal genetic variation known as hereditary persistence of HbF (HPHF). Most studies of F cells in various hematologic conditions use imprecise microscope-based assays (Thein and Craig, 1998; Chen et al., 2000; Garner et al., 2000); however, the precision and sensitivity of F-cell enumeration are significantly improved with the antiHbF flow cytometric methods, with coefficients of variation being <7.5% (Paterakis et al., 1998; Chen et al., 2000). F-cell enumeration to date has been utilized in clinical research studies, rather than in routine clinical practice, due in large part to methodological problems associated with microscopic immunofluorescence counting and the lack of assay availability in most clinical practice settings. However, as the flow cytometric F-cell enumeration methods and reported no-wash technique variations are more standardized and less labor intensive, they promise to make clinical usage more routine and potentially even automated on blood cell counters (Mundee et al., 2000). This has important implications for patients with sickle cell disease, where the standard of care has evolved to the use of therapies, such as hydroxyurea, having the specific therapeutic intent of raising F-cell levels (Maier-Redelsperger et al., 1998a; Atweh et al., 1999; Iyamu et al., 2000). The increase in F-cell or HbF levels parallels an improved clinical course in both sickle cell disease and thalassaemia (Nagel et al., 1993; Breymann et al., 1999); however, since the therapies that induce increased F cells and HbF all have toxic side effects, it follows that therapeutic monitoring by F-cell enumeration should facilitate more optimal dosing for these patients. Additionally, F-cell assays could identify patients not requiring such therapies by virtue of preexisting constitutive production of F cells due to their genetic constitution. Additionally, studies have indicated that F-cell or HbF levels may be of prognostic value for evaluation of patients with myelodysplasia or thalassemia (Craig et al., 1996; Liu et al., 1997; Reinhardt et al., 1998; Luna-Fineman et al., 1999). The finding of elevated F-cell levels has been associated with shortened survival in myelodysplasia, and F-cell enumeration may be of greater clinical utility and less costly than cytogenetic or bone marrow morphologic monitoring of such patients. Enumeration of Fetal Red Blood Cells, F Cells, and F Reticulocytes in Human Blood
permeablization. The glutaraldehyde must be properly stored and diluted just prior to use in the fixation step in the first method (see Basic Protocol 1). If the cells are not adequately fixed by the glutaraldehyde, they will lyse when the Triton X-100 is added. If the Triton X-100 step is not performed properly, the conjugated HbF antibody will not be able to reach its target within the cell. Improper washing of cells (frequency or speed of centrifugation) can result in either loss of cells or cell lysis. Poor data can result from failure to optimize instrument setup (gains) and compensation or failure to use a positive region for defining fetal cells based on a positive control, such as Fetaltrol. Failure to exclude autofluorescent nucleated cells during data analysis will give a false elevation in fetal red cell level. Autofluorescence of leukocytes or nucleated cells may be of the same intensity as fetal cells. Hence data analysis must assure that such populations are excluded from fetal-cell HbF-positive events. Another source of error is carryover. Run positive control samples last and always run blank tubes with water or buffer (or backflush) between samples. This is especially important because the assay involves rare-event detection. Other problems include RBC aggregates, especially with poor mixing after washes, red cell coincidence due to high flow rate or high cell concentration, or immunologically mediated cell aggregates in post-transfusion samples. The following clinical conditions may result in an increased level of HbF due to elevated levels of adult F cells and should not be confused with a fetomaternal hemorrhage: severe anemia, hereditary persistence of HbF, and sickle cell disease in crisis or under therapy with hydroxyurea, butyrate, or other drugs that elevate F cells.
Anticipated Results Fetal cell in blood of non-pregnant adult: ≤0.07%. Adult F cells in 40 adults normals: mean (±SD) = 3.8 ± 3.6%.
Time Considerations The total assay time is less than an hour. The hands-on technical time can be significantly reduced using an automated cell washer in place of manual pipetting and standard centrifugal washes.
Critical Parameters and Troubleshooting Inadequate staining and poor separation are usually a result of improper fixation and/or
6.17.14 Supplement 28
Current Protocols in Cytometry
Literature Cited Atweh, G., Suton, M., Nassif, I., Boosalis, V., Dover, G.J., Wallenstein, S., Wright, E., McMahon, L., Stamatoyannopoulos, G., and Faller, D.V. 1999. Sustained induction of fetal hemoglobin by pulse butyrate therapy in sickle cell disease. Blood 93:1790-1797. Bianchi, D. 1999. Fetal cells in the maternal circulation: Feasibility for prenatal diagnosis. Br. J. Haematol. 105:574-583. Bischoff, F.Z., Lewis, D.E., Nguyen, D.D., Murrell, S., Schober, W., Scott, J., Simpson, J.L., and Elias, S. 1998. Prenatal diagnosis with use of fetal cells isolated from maternal blood: Fivecolor fluorescent in situ hybridization analysis on flow-sorted cells for chromosomes X, Y, 13, 18, and 21. Am. J. Obstet. Gynecol. 179:203-209. Bohmer, R., Campbell, T., and Bianchi, D. 2000. Selectively increased growth of fetal hemoglobin-expressing adult erythroid progenitors after brief treatment of early progenitors with transforming growth factor beta. Blood 95:29672974.
Davis, B.H. 2001a. Diagnostic advances in defining erythropoietic abnormalities and red cell diseases. Semin. Hematol. 38:148-159. Davis, B.H. 2001b. Diagnostic utility of red cell flow cytometric analysis. Clin. Lab. Med. 21:829840. Davis, B.H. and Bigelow, N.C. 1989. Flow cytometric reticulocyte quantification using thiazole orange provides clinically useful reticulocyte maturity index. Arch. Pathol. Lab. Med. 113:684689. Davis, B.H., Olsen, S., Bigelow, N.C., and Chen, J.C. 1998. Detection of fetal red cells in fetomaternal hemorrhage using a fetal hemoglobin monoclonal antibody by flow cytometry. Transfusion 38:749-756. Davis, B.H., et al. 2001. Fetal red cell detection; approved guideline. NCCLS Document H52-A. Duckett, J.R. and Constantine, G. 1997. The Kleihauer technique: An accurate method of quantifying fetomaternal haemorrhage? Br. J. Obstet. Gynaecol. 104:845-846.
Breymann, C., Fibach, E., Visca, E., Huettner, C., Huch, A., and Huch, R. 1999. Induction of fetal hemoglobin synthesis with recombinant human erythropoietin in anemic patients with heterozygous beta-thalassemia during pregnancy. J. Matern. Fetal Med. 8:1-7.
Emery, C.L., Morway, L.F., Chung-Park, M., WyattAshmead, J., Sawady, J., and Beddow, T.D. 1995. The Kleihauer-Betke test. Clinical utility, indication, and correlation in patients with placental abruption and cocaine use. Arch. Pathol. Lab. Med. 119:1032-1037.
Bromilow, I.M. and Duguid, J.K. 1997. Measurement of feto-maternal haemorrhage: A comparative study of three Kleihauer techniques and two flow cytometry methods. Clin. Lab. Haematol. 19:137-142.
Garner, C., Tatu, T., Reittie, J.E., Littlewood, T., Darley, J., Cervino, S., Farrall, M., Kelly, P., Spector, T.D., and Theun, S.L. 2000. Genetic influences on F cells and other hematologic variables: A twin heritability study. Blood 95:342346.
Campbell, T.A., Ware, R.E., and Mason, M. 1998. Detection of fetal hemoglobin in erythrocytes by flow cytometry. Ann. N. Y. Acad. Sci. 850:446448.
Giacoia, G.P. 1997. Severe fetomaternal hemorrhage: A review. Obstet. Gynecol. Surv. 52:372380.
Campbell, T.A., Ware, R.E., and Mason, M. 1999. Detection of hemoglobin variants in erythrocytes by flow cytometry. Cytometry 35:242-248.
Hartwell, E.A. 1998. Use of Rh immune globulin: ASCP practice parameter. Am. J. Clin. Pathol. 110:281-292.
Chen, J., Bigelow, N., and Davis, B. 2000. Proposed flow cytometric reference method for the determination of erythroid F-cell counts. Cytometry 42:239-246.
Iyamu, W.E., Adunyah, S.E., Fasold, H., Horiuchi, K., Elford, H.L., Asalsura, T., and Turner, E.A. 2000. Enhancement of hemoglobin and F-cell production by targeting growth inhibition and differentiation of K562 cells with ribonucleotide reductase inhibitors (didox and trimidox) in combination with streptozotocin. Am. J. Hematol. 63:176-183.
Chen, J.C., Davis, B.H., Wood, B., and Worzynski, M.J. 2002. Multicenter clinical experience with flow cytometric method for fetomaternal hemorrhage detection. Cytometry 50:285-290. Cheung, M., Goldberg, J., and Kan, Y. 1996. Prenatal diagnosis of sickle cell anaemia and thalassaemia by analysis of fetal cells in maternal blood. Nat. Genet. 14:264-268. Cooling, L. 1997. Increased expression of i on HEMPAS red cells: A flow cytometric study. Transfusion 37:1102-1104. Craig, J.E., Sampietro, M., Oscier, D.G., Contreras, M., and Thein, S. 1996. Myelodysplastic syndrome with karyotype abnormality is associated with elevated F-cell production. Br. J. Haematol. 93:601-605.
Kleihauer, E., Braun, H., and Betke, K. 1957. Demonstration of fetal hemoglobin in erythrocytes of a blood smear. Klin. Wochenschr. 35:637-638. Lee, D., Contreras, M., Robson, S.C., Rodock, C.H., and Whittle, M.J. 1999. Recommendations for the use of anti-D immunoglobulin for Rh prophylaxis. Transfus. Med. 9:93-97. Liu, T., Seong, P., and Lin, T. 1997. The erythrocyte cell hemoglobin distribution width segregates thalassemia traits from other nonthalassemic conditions with microcytosis. Am. J. Clin. Pathol. 107:601-607. Phenotypic Analysis
6.17.15 Current Protocols in Cytometry
Supplement 28
Lloyd-Evans, P., Kumpel, B.M., Bromeloio, I., Austin, E., and Taylor, E. 1996. Use of a directly conjugated monoclonal anti-D (BRAD-3) for quantification of fetomaternal hemorrhage by flow cytometry. Transfusion 36:432-437. Luna-Fineman, S., Shannon, K.M., Atwater, S.K., Davis, J., Masterson, M., Ortega, J., Sanders, J., Steinherz, P., Weinberg, V., and Lange, B.J. 1999. Myelodysplastic and myeloproliferative disorders of childhood: A study of 167 patients. Blood 93:459-466. Maier-Redelsperger, M., de Montalembert, M., Flahault, A., Neonato, M.G., Ducrocq, R., Masson, M.P., Girot, R., and Eiron, J. 1998a. Fetal hemoglobin and F-cell responses to long-term hydroxyurea treatment in young sickle cell patients. The French study group on Sickle Cell Disease. Blood 91:4472-4479. Maier-Redelsperger, M., Elion, J., and Girot, R. 1998b. F reticulocytes assay: A method to evaluate fetal hemoglobin production. Hemoglobin 22:419-425. Mundee, Y., Bigelow, N.C., Davis, B.H., and Porter, J.B. 2000. Simplified flow cytometric method for fetal hemoglobin containing red blood cells. Cytometry 42:389-393. Mundee, Y., Bigelow, N.C., Davis, B.H., and Porter, J.B. 2001. Flow cytometric method for simultaneous assay of foetal haemoglobin containing red cells, reticulocytes and foetal haemoglobin containing reticulocytes. Clin. Lab. Haematol. 23:149-154. Nagel, R.L., Vichinsky, E., Shah, M., Johnson, R., Spadacino, E., Fabry, M.E., Mangahas, L., Abel, R., and Stamatoyannopoulos, G. 1993. F reticulocyte response in sickle cell anemia treated with recombinant human erythropoietin: A doubleblind study. Blood 81:9-14. Nance, S.J., Nelson, J.M., Arndt, P.A., Lam, H.C., and Garratty, G. 1989. Quantitation of fetal-maternal hemorrhage by flow cytometry. A simple and accurate method. Am. J. Clin. Pathol. 91:288-292. Navenot, J.M., Muller, J.Y., and Blanchard, D. 1997. Expression of blood group i antigen and fetal hemoglobin in paroxysmal nocturnal hemoglobinuria. Transfusion 37:291-297.
Navenot, J.M., Merghoub, T., Duarocq, R., Muller, J.-Y., Krishnamoorthy, R., and Blanchard, D. 1998. New method for quantitative determination of fetal hemoglobin-containing red blood cells by flow cytometry: Application to sicklecell disease. Cytometry 32:186-190. Nelson, M., Zarkos, K., Popp, H., and Gibson, J. 1998. A flow-cytometric equivalent of the Kleihauer test. Vox Sang. 75:234-241. Paterakis, G.S., Thein, S.L., Fibach, E., and Cappellini, M.D. 1998. Cross evaluation of three flow cytometric F cell counting methods performed by different laboratories. Hemoglobin 22:427444. Polesky, H. and Sebring, E. 1981. Evaluation of methods for detection and quantitation of fetal cells and their effect on RhIgG. Am. J. Clin. Pathol. 76:525-529. Reinhardt, D., Haase, D., Schoch, C., Wollenweber, S., Hinkelmann, E., V Heyden, W., Lentini, G., Worman, B., Schroler, W., and Pelcrun, A. 1998. Hemoglobin F in myelodysplastic syndrome. Ann. Hematol. 76:135-138. Sebring, E. and Polesky, H. 1990. Fetomaternal hemorrhage: Incidence, risk factors, time of occurrence, and clinical effects. Transfusion 30:344-357. Steele, C.D., Wapner, R.J., Smith, J.B., Haynes, M.K., and Jackson, L.G. 1996. Prenatal diagnosis using fetal cells isolated from maternal peripheral blood: A review. Clin. Obstet. Gynecol. 39:801-813. Thein, S.L. and Craig, J.E. 1998. Genetics of Hb F/F cell variance in adults and heterocellular hereditary persistence of fetal hemoglobin. Hemoglobin 22:401-414. Thorpe, S.J., Thein, S.L., Sampietro, M., Craig, J.E., Mahon, B., and Huehns, E.R. 1994. Immunochemical estimation of haemoglobin types in red blood cells by FACS analysis. Br. J. Haematol. 87:125-132.
Contributed by Bruce H. Davis and Kathleen Thompson Davis Maine Medical Center Research Institute Scarborough, Maine
Enumeration of Fetal Red Blood Cells, F Cells, and F Reticulocytes in Human Blood
6.17.16 Supplement 28
Current Protocols in Cytometry
Identification of Human Antigen-Specific T Cells Using MHC Class I and Class II Tetramers
UNIT 6.18
Methods for detecting and enumerating antigen-specific T lymphocytes (T cells) include ELISPOT, cytokine flow cytometry (CFC, described in UNIT 9.9), and the use of major histocompatibility complex (MHC) tetramers. MHC tetramers typically consist of a fluorophore-streptavidin complex and biotinylated soluble MHC molecules that carry a peptide of interest. Tetramers bind to T cell receptors (TCRs) in association with CD4 or CD8 molecules expressed on the surface of T cells, which recognize the MHC molecule/peptide combination with high specificity. Hence, MHC tetramer binding is a direct consequence of the affinity of the specific TCR/CD4 or TCR/CD8 complex for its ligand (i.e., MHC/peptide complexes), independent of readout for T cell antigen recognition (e.g., cytokine production as is the case in ELISPOT and CFC). Native MHC molecules are expressed as cell-surface glycoproteins capable of binding a variety of peptides generated from the degradation of self and non-self proteins for display to T cells. The human MHC gene locus, also known as human leukocyte antigen (HLA) locus, is highly polymorphic, with over 800 MHC class I and over 500 MHC class II alleles cuurently identified (see the immunogenic HLA sequence database; http://www.ebi.ac.uk/imgt/hla/). This heterogeneity contributes to the uniqueness of each person’s immune system, and circulating T cells recognize specific peptides bound to an HLA allele that the individual carries. The protocols described here are used for detecting and enumerating human antigenspecific T cells by flow cytometry, for research purposes only. Basic Protocol 1 describes the procedure for labeling CD8+ T cells with MHC class I tetramers. Basic Protocol 2 describes the procedure for labeling CD4+ T cells with MHC class II tetramers. Both CD8+ and CD4+ antigen-specific T cells are rare events and require that sufficient numbers of cells be evaluated. For detection of CD4+ antigen-specific cells with class II tetramers, in vitro cell expansion (described in UNIT 9.11) is often necessary. To minimize nonspecific tetramer binding contributed by irrelevant cell populations, a cumulative Boolean gating strategy using positive selection and/or exclusion gating should be employed. These protocols have been developed and validated on human samples, i.e., peripheral blood mononuclear cells (PBMCs) and whole blood, but are also applicable to murine specimens. NOTE: MHC tetramers are for research use only and are not to be used in diagnostic procedures.
STRATEGIC PLANNING TCR recognition of peptide-loaded MHC molecules is extremely specific. The specificity of the MHC allele used to create a MHC tetramer reagent must exactly match the HLA class I or class II allele of the donor sample. Therefore, the donor’s HLA type should be known prior to testing. High-resolution genotyping is recommended.
IDENTIFICATION OF MHC CLASS I ANTIGEN-SPECIFIC T CELLS CD8+
Basic Protocol 1 describes how to evaluate class I tetramer–positive T cells using PE-Cy5-conjuated anti-CD3 antibody and FITC-conjugated anti-CD8 antibody as gating reagents. Antigen-specific, steady-state, memory CD8+ T cells are present in
BASIC PROTOCOL 1
Phenotypic Analysis Contributed by Lori A. Krueger, C. Thomas Nugent, and Johannes Hampl Current Protocols in Cytometry (2004) 6.18.1-6.18.12 C 2004 by John Wiley & Sons, Inc. Copyright
6.18.1 Supplement 30
low frequencies (typically less than 1 in 1000 leukocytes) in the peripheral blood of most healthy normal donors. These cells may be detected using MHC class I tetramer reagents and multiparametric flow cytometry. A minimum of two fluorescent reagents is required. Anti-CD8-FITC (fluorescein isothiocyanate) and MHC class I tetramer–PE (phycoerythrin) may be used in combination to identify antigen-specific CD8+ T cells in whole-blood or PBMC preparations. However, the ability to resolve true antigen-specific CD8+ T cells from those labeled by nonspecific cell binding is increased by using additional antibody gating reagents. For example, by adding anti-CD3-PE-Cy5 and gating on CD3+ events, irrelevant CD8+ /CD3− cells are excluded from the analysis. CD8− /CD3− cells are incapable of binding MHC class I tetramers via antigen-specific TCRs, but may contribute to nonspecific tetramer binding. Exclusion gating may also be used. Cells that express CD4 (helper T cells), CD19 (B cells), CD13 (cells of myeloid lineage), or CD33 (cells of myeloid lineage) are not capable of binding MHC class I tetramer in an antigen-specific manner. Addition of monoclonal antibodies (MAbs) that recognize these cells can be used to eliminate irrelevant cell populations from the analysis region. Exclusion-gating reagents may be conjugated to the same fluorochrome. To further reduce nonspecific binding contributed by nonviable cells, reagents such as 7-AAD that differentiate permeable (usually dead and dying) from nonpermeable (intact, live) cells can also be added. Extended phenotyping can also be performed with the addition of lineage-specific and/or activation-dependent reagents. MHC tetramer analysis can be combined with CFC to determine the number of antigenspecific T cells present in the sample and their ability to generate cytokines in response to specific antigen stimulation. However, during the incubation periods of the combined assay, the antigen-receptor TCR/CD3 complex is rapidly down-regulated upon cell activation. Therefore, to perform this type of assay, cells are first labeled with tetramer reagent prior to antigen stimulation followed by processing for CFC. Refer to UNIT 9.9 for protocols describing intracellular cytokine measurements. Alternatively, a kit containing the reagents required for measuring intracellular IFN-γ in combination with tetramers can be purchased from Beckman Coulter. If tetramers are used in conjunction with other vendors’ kits, individual laboratories are required to validate the use of these kits with the tetramer reagents.
Materials Whole blood collected into EDTA or heparin anticoagulant (UNIT 5.1) Labeling buffer: e.g., PBS (APPENDIX 2A) Anti-CD8-FITC (e.g., Beckman Coulter) Anti-CD3-PE-Cy5 (e.g., Beckman Coulter) Irrelevant (negative) MHC class I tetramer conjugated to PE (Beckman Coulter) Specific MHC class I tetramer conjugated to PE (Beckman Coulter) Phosphate-buffered saline (PBS; APPENDIX 2A) 0.5% paraformaldehyde (see recipe) or 0.5% formalin in PBS 12 × 75–mm polypropylene test tubes Centrifuge capable of 400 × g Flow cytometer Additional reagents and equipment for preparing PBMCs and lysis of erythrocytes (UNIT 5.1) and titering antibodies (UNIT 4.1)
Identification of Human Antigen-Specific T Cells Using MHC Tetramers
Prepare samples 1. Prepare peripheral blood mononuclear cells (PBMCs) from anticoagulated whole blood by density-gradient centrifugation (UNIT 5.1). Resuspend cells at a concentration of 5 × 106 cells/ml in appropriate labeling buffer (i.e., PBS or other cell-labeling buffer suitable for flow cytometry).
6.18.2 Supplement 30
Current Protocols in Cytometry
PBMCs may be used freshly isolated or thawed after frozen storage. Alternatively, whole blood collected into EDTA or heparin anticoagulant may be used. Blood should be processed within 24 hr of collection. Optimal results are achieved when blood is processed within 8 hr of collection.
2. For each sample being tested, label two 12 × 75–mm test tubes: one control and one test. 3. Add the appropriate amounts of optimally titered (UNIT 4.1) anti-CD8-FITC and antiCD3-PE-Cy5 reagents to each tube. 4. Add 10 µl optimally titered (as for monoclonal antibodies; see UNIT 4.1) irrelevant (negative) tetramer reagent to the tube labeled “control” and 10 µl optimally titered specific tetramer reagent to the tube labeled “test.” Irrelevant tetramers are reagents that are not recognized by any TCR in the context of the experiment. These reagents may carry irrelevant peptide sequences or they may represent an HLA allele mismatch.
5. Add 200 µl PBMCs (i.e., 1 × 106 cells) or 200 µl whole blood to each test tube and vortex gently. 6. Incubate 30 min at room temperature protected from light. 7. If using whole blood, lyse red blood cells according to established procedures (UNIT 5.1). Individual laboratories must validate red blood cell lysis procedures for use with tetramer reagents.
8. Add 3 ml PBS. Centrifuge tubes 5 min at 400 × g (or as directed by erythrocyte lysis procedure) at room temperature. 9. Aspirate the supernatant and resuspend the cell pellet in 500 µl PBS with 0.5% paraformaldehyde or 0.5% formalin. 10. Analyze immediately or store prepared samples at 4◦ C in the dark up to 24 hr until analysis by flow cytometry.
Set up flow cytometer and run samples 11. Perform instrument quality control according to manufacturer’s recommendations. 12. Establish PMT voltages and appropriate fluorescence color compensation for the acquisition protocol using samples that have been processed in the same way as the assay tubes. Set the primary instrument discriminator (threshold) on forward-angle light scatter. 13. Acquire a minimum of 30,000 to 50,000 CD3+ /CD8+ events and a minimum of 100 tetramer+ /CD8+ events. These numbers of events are required to provide an adequate level of precision appropriate for the evaluation of rare tetramer+ /CD8+ events. A minimum of four color parameters, forward scatter, and orthogonal scatter are required. In the protocol described here, instrument filter configurations appropriate for the collection of CD3-FITC, CD4-ECD, CD8/CD13/CD19-PE-Cy5 and PE-conjugated tetramer are required.The standard filter configuration on most three- and four-color instruments should be appropriate for the collection of these parameters. Alternate compatible fluorochromes may be used if the instrument configuration cannot accommodate these fluorochromes. MHC class I-tetramer positive events are usually reported as a percent of total CD3+ /CD8+ events. Refer to the gating strategy in Figure 6.18.1 for further details. If an absolute count of tetramer-positive events is desired, the investigator may combine the percentage of
Phenotypic Analysis
6.18.3 Current Protocols in Cytometry
Supplement 30
Identification of Human Antigen-Specific T Cells Using MHC Tetramers
Figure 6.18.1 Gating strategy used in Basic Protocol 1 to include the cell population of interest and exclude irrelevant events and debris. (A) Anti-CD3-PE-Cy5 versus SS. Region A is drawn to include all CD3+ T cells with low/intermediate side scatter. (B) Anti-CD8-FITC versus SS. Gated on region A; region B is drawn to include all CD8+ T cells. (C) SS versus FS. Gated on regions A and B; region C is drawn to include all CD3+ /CD8+ T cells with characteristic lymphocyte light scatter. (D) MHC class I negative tetramer–PE versus CD-FITC. Gated on regions A, B, and C; region D indicates the level of non-antigen-specific background binding. (E) MHC class I B0702 CMV–specific tetramer–PE versus CD8-FITC. Gated on regions A, B, and C; region D includes all antigen-specific CD8+ T cells. Region D is drawn to include all tetramer positive events (panel E) above background binding determined using an irrelevant tetramer (panel D). Abbreviations: CMV, cytomegalovirus; FITC, fluorescein isothiocyantate; FS, forward scatter; PE, phycoerythrin; SS, side scatter.
6.18.4 Supplement 30
Current Protocols in Cytometry
tetramer-specific CD3+ /CD8+ events with an appropriately determined absolute count of total CD3+ /CD8+ positive cells. Figure 6.18.1 illustrates the cumulative gating strategy used to include the cell population of interest and exclude all other irrelevant events and debris.
IDENTIFICATION OF MHC CLASS II ANTIGEN-SPECIFIC T LYMPHOCYES
BASIC PROTOCOL 2
The frequency of circulating antigen-specific CD4+ T cells is much lower than the typical frequencies observed with CD8+ T cells. This presents a challenge for their detection. To increase chances for direct ex vivo detection, nonspecific fluorescent events and background binding must be minimized. An optimized flow cytometric assay should ideally have a lower limit of detection of 0.01% of CD4+ T cells (i.e., 1 tetramer+ /CD4+ event in 10,000 CD4+ events). However, despite optimal assay conditions, biologically relevant frequencies of antigen-specific CD4+ T cells may fall below the lower level of detection. PBMCs may be cultured in vitro for several days in presence of antigen to selectively expand antigen-specific T cells. Cell-tracking dyes such as carboxyfluorescein diacetate succinimidyl ester (CFSE; UNIT 9.11) can be used in conjunction with tetramers to provide information regarding the proliferative history of labeled cells while in culture, permitting back-calculation of antigen-specific CD4+ precursor frequencies in the original sample (Novak et al., 1999). Recently, a new approach with increased sensitivity (i.e., 1 tetramer+ /CD4+ event in 100,000 CD4+ events) has been described in which PE-conjugated tetramer–labeled cells were isolated ex vivo from PBMCs with anti-PE magnetic beads (Day et al., 2003). This basic protocol describes the detection of antigen-specific CD4+ T cells with MHC class II tetramers using anti-CD3-FITC, tetramer-PE, and anti-CD4-ECD (energycoupled dye), along with the exclusion gating MAbs anti-CD8-PE-Cy5, anti-CD13-PECy5, and anti-CD19-PE-Cy5. If CFSE-labeled cultured cells are used, anti-CD3 conjugated to PE-Cy5 may be employed. The exclusion gating strategy is not as critical when using CSFE-labeled and expanded cells, because the frequency of bona fide tetramerpositive events is sufficiently increased over nonspecific binding. In general, the procedure described here is appropriate for the detection of antigen-specific CD4+ T cells using MHC class II tetramers. Extended incubation periods of tetramer with cells in culture medium at 37◦ C, as recommended here, allows internalization of tetramers by cells, and hence practically permanent labeling of the cells. This procedure has been designed to maximize labeling efficiencies of all antigen-specific CD4+ T cells, including those with low functional avidity. However, labeling conditions (including tetramer concentration, incubation time, and temperature) may require optimization depending on the tetramer specificity.
Materials Whole blood collected into EDTA or heparin anticoagulant (UNIT 5.1) or ex vivo–expanded lymphocytes Culture medium for ex vivo–expanded lymphocytes (see recipe) Irrelevant MHC class II tetramer conjugated to PE (Beckman Coulter) Specific MHC class II tetramer conjugated to PE (Beckman Coulter) Anti-CD3-FITC (e.g., Beckman Coulter) Anti-CD4-ECD (e.g., Beckman Coulter) Anti-CD8-PE-Cy5 (e.g., Beckman Coulter) Anti-CD13-PE-Cy5 (e.g., Beckman Coulter) Anti-CD19-PE-Cy5 (e.g., Beckman Coulter) Phosphate-buffered saline (PBS; APPENDIX 2A) 0.5% paraformaldehyde (see recipe) or 0.5% formalin in PBS
Phenotypic Analysis
6.18.5 Current Protocols in Cytometry
Supplement 30
12 × 75–mm polypropylene test tubes 5% CO2 incubator, equilibrated at 37◦ C Centrifuge capable of 400 × g Flow cytometer Additional reagents and equipment for preparing PBMC and lysis of erythrocytes (UNIT 5.1) and titering antibodies (UNIT 4.1) Prepare samples 1. Prepare peripheral blood mononuclear cells (PBMCs) from anticoagulated whole blood by density-gradient centrifugation (UNIT 5.1). Resuspend cells in culture medium at a concentration of 5–15 × 106 cells/ml. Blood should be collected into EDTA or heparin anticoagulant and processed within 24 hr of collection. PBMCs may be used freshly isolated or thawed after frozen storage. Alternatively, whole blood collected into EDTA or heparin anticoagulant or ex vivo– expanded lymphocytes may be used. Optimal results are achieved when blood is processed within 8 hours of collection.
2. For each sample being tested, label two 12 × 75–mm test tubes: one control and one test. If whole blood is used, prepare tubes in duplicate (to be pooled prior to data acquisition). 3. Add 10 µl optimally titered (UNIT 4.1) irrelevant tetramer reagent to the tube labeled “control” and 10 µl optimally titered specific tetramer reagent to the tube labeled “test.” Titering of the tetramer reagents is done as for monoclonal antibodies (see UNIT 4.1). Irrelevant tetramers are reagents that are not recognized by any TCR in the context of the experiment. These reagents may carry irrelevant peptide sequences or they may represent an HLA allele mismatch.
4. Add 200 µl PBMCs or cultured lymphocytes (i.e., 1–3 × 106 cells) or 200 µl whole blood to each test tube and vortex gently. Prepare whole-blood tubes in duplicate.
5. Incubate tubes 2 hr at 37◦ C in a humidified incubator with 5% CO2 atmosphere, protected from light. 6. Chill tubes ∼5 min on ice, protected from light. 7. Add the appropriate amount of optimally titered (UNIT 4.1) anti-CD3-FITC, anti-CD4ECD, anti-CD8-PE-Cy5, anti-CD13-PE-Cy5, and anti-CD19-PE-Cy5 to each tube. Vortex gently. 8. Incubate tubes 20 min at room temperature, protected from light. 9. If using whole blood, lyse red blood cells according to established procedures (UNIT 5.1). Individual laboratories must validate red blood cell lysis procedures for use with tetramer reagents.
10. Add 3 ml PBS. Centrifuge tubes 5 min at 400 × g (or as directed by erythrocyte lysis procedure) at room temperature.
Identification of Human Antigen-Specific T Cells Using MHC Tetramers
11. Aspirate the supernatant and resuspend the cell pellet in 500 µl PBS with 0.5% paraformaldehyde or 0.5% formalin. For whole blood samples, resuspend each of the duplicate tubes in 250 µl PBS with 0.5% paraformaldehyde or formalin and pool the tubes.
6.18.6 Supplement 30
Current Protocols in Cytometry
Figure 6.18.2 Gating strategy used in Basic Protocol 2 to include the cell population of interest and exclude irrelevant events and debris. (A) Multiple MAbs (anti-CD8, anti-CD19)–single color (PE-Cy5) versus SS. Region A is drawn to include all cells that do not stain with anti-CD8, antiCD19. The majority of monocytes are also excluded because of higher autofluorescence and SS. While recommended as an option in Basic Protocol 2, anti-CD13-PE-Cy5 was not included in this example. (B) Anti-CD3-FITC versus anti-CD4-ECD. Gated on region A; region B is drawn to include all CD3+ CD4+ T cells. The CD3 gate should be generous on the lower side because of tetramer-induced TCR/CD3 down-regulation. (C) FS versus SS. Gated on regions A and B; region C is drawn to include all CD3+ /CD4+ T cells with characteristic lymphocyte light scatter. Gate can be tight when using PBMCs directly. Gate should be generous when using cultured cells, to include cell blasts. (D) CD4-ECD versus MHC class II negative tetramer–PE. Gated on regions A, B, and C; region D2 indicates the level of non-specific binding. Region D2 should be drawn to exclude tetramer-negative events. (E) CD4-ECD versus MHC class II positive tetramer–PE. Gated on regions A, B, and C; region D2 indicates the level of antigen-specific binding.
Phenotypic Analysis
6.18.7 Current Protocols in Cytometry
Supplement 30
12. Analyze immediately or store prepared samples at 4◦ C in the dark up to 24 hr until analysis by flow cytometry.
Set up flow cytometer and run samples 13. Perform instrument quality control according to manufacturer’s recommendations. 14. Establish PMT voltages and appropriate fluorescence color compensation for the acquisition protocol using samples that have been processed in the same way as the assay tubes. Set the primary instrument discriminator (threshold) on forward-angle light scatter. 15. Acquire data. The minimum number of CD3+ /CD4+ events to be collected is dependent on the frequency of CD4+ /tetramer+ events present. A minimum of 100,000 CD4+ events is desirable. A minimum of four color parameters, forward scatter, and orthogonal scatter are required. In the protocol described here, instrument filter configurations appropriate for the collection of CD3-FITC, CD4-ECD, CD8/CD13/CD19-PE-Cy5 and PE-conjugated tetramer are required. The standard filter configuration on most three- and four-color instruments should be appropriate for the collection of these parameters. Alternate compatible fluorochromes may be used if the instrument configuration cannot accommodate these fluorochromes. Refer to the gating strategy in Figure 6.18.2 for further details. Figure 6.18.2 illustrates the cumulative gating strategy used to include the cell population of interest and exclude all other irrelevant events and debris.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Culture medium RPMI 1640 (e.g., Life Technologies) containing: 10% heat-inactivated human AB serum 2 mM L-glutamine 50 µM 2-mercaptoethanol 1 mM sodium pyruvate 5 mM HEPES 1% nonessential amino acids 100 U/ml penicillin 100 g/ml streptomycin Paraformaldehyde, 0.5% Dilute electron microscopy–grade 20% paraformaldehyde solution (i.e., 20% formaldehyde solution prepared from paraformaldehyde; Electron Microscopy Sciences) 1:40 in PBS (APPENDIX 2A) to prepare 0.5% paraformaldehyde. COMMENTARY Background Information
Identification of Human Antigen-Specific T Cells Using MHC Tetramers
T cells recognize MHC/peptide complexes expressed on the cell surface of a multitude of nucleated cells via their antigen-specific TCRs. Peptides originate from self and nonself (e.g., viral, bacterial, or parasitic) proteins that are degraded intracellularly and bound to MHC molecules. The MHC/peptide complex– TCR interactions are inherently of low affinity (Kd = 10–4 to 10–6 ), and MHC/peptide
monomers dissociate within seconds from bound TCRs (Davis et al., 1998). Multimerized MHC/peptide complexes have increased avidity for TCRs, resulting in sufficiently stable binding to T cells (Altman et al., 1996). MHC-tetramer reagents are most often conjugated to phycoerythrin (PE) and detected by flow cytometry when bound to antigen-specific T cells. These reagents are used to study immune responses to treatment
6.18.8 Supplement 30
Current Protocols in Cytometry
in viral infection, cancer, and autoimmune disease. MHC tetramers are created from either class I or class II MHC molecules. Class I tetramers bind to CD8+ T cells (cytotoxic T cells or CTLs). CTLs destroy their target cells either directly by cytolytic activity, or indirectly by cytokine secretion. Class I MHC tetramers are used to evaluate antigen-specific CTLs that recognize bacteria, viruses, tumors, self-antigens, and parasites. Class II tetramers identify antigen-specific CD4+ T cells (helper T cells). Helper T cells regulate the response of other immune-system cell populations, including B cells, antigen-presenting cells, and other T cells, through receptor-ligand interactions and/or cytokine secretion. The majority of the scientific literature regarding MHC tetramers to date has focused on identification of antigen-specific CD8+ T cells with MHC class I tetramers. Far fewer reports exist concerning the use of MHC class II tetramers to monitor CD4+ T cells. MHC tetramers are used in research studies to directly identify antigen-specific T cells in a variety of sample types, including whole blood, PBMCs, and expanded cell lines. Other single-cell methods of identifying antigenspecific T cells (i.e., ELISPOT and CFC) rely on the ability of specific cells to be stimulated by antigen. These assays do not detect cells that are unresponsive to antigen stimulation, nor can they differentiate cells responding directly to antigen from those cells that are activated as a result of cytokines released during the assay incubations (bystander activation). Soluble MHC/peptide complexes can be multimerized and fluorescently labeled in a variety of ways. For MHC tetramer reagents, multimerization is achieved by binding four biotinylated MHC/peptide complexes to the four biotin-binding sites of a fluorophoreconjugated avidin or streptavidin molecule. This approach has been chosen by a majority of investigators and can be considered a tried-and-true method for enumeration of antigen-specific T cells. Other approaches include MHC dimers, consisting of two MHC molecules expressed as an IgG fusion protein (Schneck, 2000), MHC oligomers multimerized via synthetic, branched polypeptide chains (Cochran et al., 2000), MHC liposomes (Prakken et al., 2000; Mallet-Designe et al., 2003), MHC pentamers, and MHC hexamers. MHC dimers have been used as long as MHC tetramers and provide sufficient sensitivity in some cases. Generally, MHC dimers appear to
label specific lymphocytes with less fluorescence intensity than tetramers. Other methods described here have not yet found widespread use, and it is not clear if they offer any significant advantage over MHC tetramers.
Critical Parameters and Troubleshooting Rare-event analysis In most cases, antigen-specific T cells constitute a small fraction of total T cells. The smaller this fraction, the greater the need for discriminating bona fide positive events from noise. An optimized cell-labeling procedure, including appropriate controls and an optimized data-acquisition protocol, will assist the investigator in evaluating the validity of positive results. A negative or irrelevant tetramer reagent assists in evaluating the lower limit of detection in the flow cytometric assay. The negative (irrelevant) tetramer should be manufactured with the same lot of streptavidinPE (SA-PE) as the tetramer of interest, since different lots of SA-PE may impact nonspecific binding to PBMCs and peripheral-blood leukocytes. Detection limits can be lowered by incorporating lineage-specific antibodies to positively identify cell populations of interest while using exclusion gating MAbs (see Basic Protocol 2), and by the use of cell viability dyes. Confidence in the accuracy of positive results is partly dependent on the total number of positive cells collected (see below). Labeling patterns of cell clusters (tight and discrete versus diffuse staining) and the separation between tetramer-negative and tetramer-positive events lend further evidence that the observed tetramer-positive results are valid. Precision of rare event analysis Like other rare-event analyses, the tetramer frequency distribution is described by Poisson statistics. The essential feature of Poisson distributions is that when n positive events are observed, the standard deviation (SD) associated with n is the square root of n (Shapiro, 1995). Therefore, if at least 100 tetramerpositive events are counted, the SD is 10. If a 95% confidence interval is defined as 2SD, then the true value of tetramer-positive events is between 80 and 120. Coefficient of variation (CV) is defined by: Phenotypic Analysis
6.18.9 Current Protocols in Cytometry
Supplement 30
A CV of 10% is ensured by counting 100 positive tetramer events. Precision of the assay is increased when a greater number of positive events is counted. Nonspecific tetramer binding Properly manufactured and stored MHC tetramers should generate reliable results for the enumeration of antigen-specific T cells. Nonspecific binding may be caused by inappropriate tetramer reagent concentration, and this problem can be resolved by titration of the reagent. A special case of non-antigen-specific binding of MHC class I tetramer is mediated through its binding site for CD8 in the α3 domain of the HLA heavy chain. Possible remedies are discussed below.
Identification of Human Antigen-Specific T Cells Using MHC Tetramers
Binding of MHC Class I tetramers to CD8 Native MHC molecules interact with CD4 or CD8 coreceptors in addition to TCRs on the surface of antigen-presenting cells. MHC class I tetramers can exhibit a relatively high reactivity with CD8. A corresponding phenomenon for MHC class II tetramers and CD4 has not been reported. Mutations in the MHC class I heavy chain near the CD8 binding region have been shown to significantly reduce CD8 binding while allowing stable peptide-MHC-TCR binding; changing amino acids at or near residue 227 (226-228) of the heavy chain completely abolishes MHC interaction with CD8, while a mutation at residue 245 reduces non-antigen-specific CD8 binding 4fold (Salter et al., 1989; Salter et al., 1990). Importantly, tetramers carrying a mutation at residue 245 have been observed to abrogate CD8-mediated, but not antigen-specific, binding to T cells (observed frequencies of tetramer-positive cells remained unchanged). Mutated tetramers label antigen-specific T cells with slightly lower mean fluorescence intensity compared to wild-type tetramers (Bodinier et al., 2000; Beckman Coulter, unpub. observ.). MHC class I tetramers mutated at amino acid 245 are commercially available for research use from Beckman Coulter (http://www.immunomics.com). Monoclonal antibodies to CD8 can affect wild-type tetramer binding in both human and murine systems (Daniels et al., 2000; Campanelli et al., 2002), either blocking or enhancing tetramer staining. Several factors contribute to the phenomenon, including the choice of antiCD8 MAb clone (steric hindrance and epitope
recognized) and the specificity of the MHC allele being investigated. Cross-titration of both wild-type MHC class I tetramer and anti-CD8 antibody is recommended to optimize labeling. Alternatively, tetramers carrying the mutation at amino acid 245 can be used regardless of the choice and concentration of CD8 MAb (human systems only). Lower-than-expected tetramer binding 1. The stability of tetramer reagents is partly dependent upon the affinity of the specific peptide for the MHC molecule. Loweraffinity peptides have a tendency to dissociate over time, if in fact the specific tetramer can be manufactured in the first place. An unstable tetramer reagent will usually result in poorer-than-expected levels of specific binding and higher-than-expected levels of nonspecific binding, effectively reducing the signalto-noise ratio. 2. Poorer-than-expected levels of labeling may also be observed if the cells have been recently activated and TCRs have been internalized. This process may be detected by labeling cells with specific pan-TCR antibody reagents and/or anti-CD3. When activated cells are properly rested, surface CD3 and TCR density return to normal levels. 3. Unknown or mismatched HLA type or low-resolution HLA typing may cause difficulties in interpreting observed tetramer binding. 4. T cells in a state of unresponsiveness have been reported to lose their ability to bind MHC tetramer. This phenomenon has been observed primarily with CD4+ T cells using in vitro–cultured T cells (Cameron et al., 2001). It is not known if this finding can be extended to freshly isolated PBMCs. This point should be considered when working with samples from patients with diseases known to cause T cell unresponsiveness, i.e., late-stage chronic viral infections, tumors, and autoimmune diseases.
Anticipated Results MHC tetramers are T cell epitope–specific reagents. Investigators should expect multiple epitopes to be recognized during the course of an immune response. Considering that humans may express up to four different MHC class I alleles that restrict CD8 T cell recognition and up to 12 different MHC class II alleles that restrict CD4 T cell responses, it is not surprising that donor-to-donor variability in response to identical stimuli is observed. However, in many diseases, immunodominant
6.18.10 Supplement 30
Current Protocols in Cytometry
epitopes have been identified, and it is advisable to include those in any research study. The frequency of antigen-specific T cells depends on the disease and the status of the responder. In infectious diseases, the highest responses can be expected during the acute phase, e.g., 10% to 50% of all CD8 T cells, with lower responses expected during chronic phases (Barouch and Letvin, 2001). In healthy cytomegalovirus (CMV) seropositive donors, the authors of this unit have observed that the percentage of CD8-positive T cells specific for CMV epitopes remains reasonably constant over time within each individual (∼0.2% to 2%; L. Kruger, C.T. Nugent, and J. Hampl, unpub. observ.). Tetramer-positive CD4+ T cell responses appear to be lower than CD8 T cell responses in general. For example, steadystate influenza hemagglutinin–specific CD4+ T cell frequencies range from 1:1,000,000 to 1:10,000 in normal donors (Lucas et al., 2004). In general, the frequency of na¨ıve T cells may preclude their detection.
Time Considerations Identifying antigen-specific T cells with tetramers is rapid compared to other techniques requiring T cell activation. For MHC class I and class II tetramer evaluations, the cell-labeling times are about 1 hr and 3 hr, respectively. Additional time must be considered for cell preparation if PBMCs or cultured lymphocytes are being used. The acquisition time on the flow cytometer may be as long as 20 min per tube, owing to the large numbers of cells to be collected. Instruments equipped with autosamplers are helpful. If tetramer analysis is to be combined with other methods such as cytokine flow cytometry (CFC) or cell cultivation, then the time required to prepare additional evaluations must be factored in. Most assays for CFC take about 6 to 8 hr to complete. Cell expansion may take up to 10 days.
Literature Cited Altman, J.D., Moss, P.A., Goulder, P.J., Barouch, D.H., McHeyzer-Williams, M.G., Bell, J.I., McMichael, A.J., and Davis, M.M., 1996. Phenotypic analysis of antigen-specific T cells. Science 274:94-96.
sorting of specific T cells using multimers of MHC class I and peptide with reduced CD8 binding. Nat. Med. 6:707-10. Cochran, J. R., Cameron, T. O., and Stern L. J. 2000. The relationship of MHC-peptide binding and T cell activation probed using chemically defined MHC class II oligomers. Immunity 12:241-250. Cameron, T.O., Cochran, J.R., Yassine-Diab, B., Sekalym, R.-P., and Stern, L.J. 2001. Cutting edge: Detection of antigen-specific CD4+ T cells by HLA-DR1 oligomers is dependent on the T cell activation state. J. Immunol. 166:741-745. Campanelli, R., Palermo, B., Garbelli, S., Mantovani, S., Lucchi, P., Necker, A., Lantelme, E., and Giachino, C. 2002. Human CD8 coreceptor is strictly involved in MHC-peptide tetramerTCR binding and T cell activation. Int. Immunol. 14:39-44. Daniels, M.A. and Jameson, S.C. 2000. Critical role for CD8 in T cell receptor binding and activation by peptide/major histocompatibility complex multimers. J. Exp. Med. 191:335-46. Davis, M.M., Boniface, J.J., Reich, Z., Lyons, D. S., Hampl, J., Arden, B. and Chien, Y.H. 1998. Ligand recognition by alpha beta T cell receptors. Ann. Rev. Immunol. 16:523-44. Day, C.L., Seth, N.P., Lucas, M., Appel, H., Gauthier, L., Lauer, G.M., Robbins, G.K., Szczepiorkowski, Z.M., Casson, D.R., Chung, R.T., Bell, S., Harcourt, G., Walker, B.D., Klenerman, P., and Wucherpfennig, K.W. 2003. Ex vivo analysis of human memory CD4 T cells specific for hepatitis C virus using MHC class II tetramers. J. Clin. Invest. 112:831-842. Lucas, M., Day, C.L., Wyer, J.R., Cunliffe, S.L., Loughry, A., McMichael, A.J., and Klenerman, P. 2004. Ex vivo phenotype and frequency of influenza virus-specific CD4 memory T cells. J. Virol. 78:7284-7287. Mallet-Designe, V. I., Stratmann T., Homann, D., Carbone, F., Oldstone, M. B., and Teyton, L. 2003. Detection of low-avidity CD4+ T cells using recombinant artificial APC: Following the antiovalbumin immune response. J. Immunol. 170:123-131. Novak, E.J., Liu, A.W., Nepom, T., and Kwok, W.W. 1999. MHC class II tetramers identify peptidespecific human CD4+ T cells proliferating in response to influenza A antigen. J. Clin. Invest. 104:R63-R67. Prakken, B., Wauben, M., Genini, D., Samodal, R., Barnett, J., Mendivil, A., Leoni L., and Albani S. 2000. Artificial antigen-presenting cells as a tool to exploit the immune “synapse.” Nat. Med. 6: 1406-1410.
Barouch, D.H. and Letvin, N.L. 2001. CD8+ cytotoxic T lymphocyte responses to lentiviruses and herpesviruses. Curr. Opin. Immunol. 13:479482.
Salter, R.D., Norment, A.M., Chen, B.P., Clayberger, C., Krensky, A.M., Littman, D.R., and Parham, P. 1989. Polymorphism in the alpha 3 domain of HLA-A molecules affects binding to CD8. Nature 338:345-347.
Bodinier, M., Peyrat, M.A., Tournay, C., Davodeau, F., Romagne, F., Bonneville, M., and Lang, F. 2000. Efficient detection and immunomagnetic
Schneck, J. P. 2000. Monitoring antigen-specific T cells using MHC-Ig dimers. Immunol Invest. 29:163-169.
Phenotypic Analysis
6.18.11 Current Protocols in Cytometry
Supplement 30
Salter, R.D., Benjamin, R.J., Wesley, P.K., Buxton, S.E., Garrett, T.P.J., Clayberger, C., Krensky, A.M., Norment, A.M., Littman, D.R., and Parham, P. 1990. A binding site for the T-cell co-receptor CD8 on the a3 domain of HLA-A2. Nature 345:41-46. Shapiro, H. 1995. Practical Flow Cytometry, 3rd ed. Wiley-Liss, Inc., New York.
Contributed by Lori A. Krueger, C. Thomas Nugent, and Johannes Hampl Beckman Coulter Inc. San Diego, California
Identification of Human Antigen-Specific T Cells Using MHC Tetramers
6.18.12 Supplement 30
Current Protocols in Cytometry
ZAP-70 Staining in Chronic Lymphocytic Leukemia
UNIT 6.19
Chronic lymphocytic leukemia (CLL) is the most frequently diagnosed chronic lymphoid leukemia in the USA and Europe. The disease is diagnosed mainly in elderly patients, but about a third of patients are under 60 years of age at diagnosis. The clinical course of patients after diagnosis is extremely variable. Although a subset of patients have an indolent CLL and a probability of survival similar to that of age-matched controls, another subset of patients have an aggressive CLL and die within a few months of diagnosis. The clinical management of CLL patients depends on the stage and pace of the disease. Currently, CLL patients in advanced clinical stages receive therapy, whilst patients in early stages of the disease do not receive therapy unless the disease progresses. Nowadays, ∼80% of patients with CLL are diagnosed in the early stages of the disease. The introduction of more effective, potentially curative therapeutic options for CLL patients requires identification of patients in early stages of the disease who are at high risk of progression and who could benefit from early therapy. Different biological features have been associated with prognosis in patients with CLL, including hematological characteristics, biochemical markers, and cytogenetical abnormalities. The mutational configuration of the variable regions of immunoglobulin heavychain genes (IgVH ) has a major impact in the clinical outcome. Somatically mutated IgVH genes are detected in about half of patients, and in them the course of the disease is indolent. On the contrary, unmutated IgVH confers bad prognosis to the other half of patients with CLL. The ancillary techniques used to determine the mutational status of IgVH are routinely unavailable to general laboratories, for whom other markers able to discriminate the two clinical forms of CLL should be identified to make the estimation of prognosis possible. Gene expression profiles reveal that ZAP-70, a tyrosine kinase expressed in T and natural killer (NK) lymphocytes, is expressed in a subset of CLL cases in close relationship with an unmutated configuration of IgVH . This finding has been validated using less complex and more affordable technologies, including flow cytometry. Flow cytometry has several advantages over other methods to assess ZAP-70 in CLL cells. The multiparametric analysis of flow cytometry allows an assay using an unpurified sample of peripheral blood (either mononuclear cells or whole blood); the procedure is fast, inexpensive, and easy to introduce in routine diagnostic panels, and an internal positive control is present in the sample. The protocols described in this unit include two different staining methods for assessing the risk group in CLL patients. The Basic Protocol describes a technique that is relatively complex compared to the usual staining methods for immunophenotypical characterization of cells. The clinical information available has been mainly generated using methodology similar to the Basic Protocol, which reflects a rapid transition from the highly complex procedure of gene-expression profiling to a more affordable and less complex flow-cytometry-based technology. However, when directly labeled monoclonal antibodies against ZAP-70 have been proved reliable, even simpler direct-staining protocols such as the one described in the Alternate Protocol can replace the Basic Protocol.
Phenotypic Analysis Contributed by Neus Villamor Current Protocols in Cytometry (2005) 6.19.1-6.19.14 C 2005 by John Wiley & Sons, Inc. Copyright
6.19.1 Supplement 32
BASIC PROTOCOL
INDIRECT STAINING OF INTRACELLULAR ZAP-70 This protocol describes the technique to quantify ZAP-70 in CLL cells using indirect intracellular staining followed by direct staining to recognize normal and leukemic cell subpopulations in the sample. This technique is unusual in diagnostic laboratories performing immunophenotyping, owing to the absence (until very recently) of directly labeled monoclonal antibodies (MAbs) against ZAP-70. The protocol is described for a flow cytometer equipped with two lasers (488-nm HeNe and 635-nm red diode laser). NOTE: All incubations and centrifugations are performed at room temperature protected from light unless otherwise specified.
Materials Whole peripheral blood sample, mononuclear cells, or thawed cryopreserved cells Fixation and Permeabilization Kit formulated to study intracellular antigens in leukemia cells (Caltag, Dako) Unconjugated MAb against ZAP-70 (e.g., clone 2F3.2, available from Upstate Biotechnology, Novus Biological, LabVision, Diagnostic Biosystems, or Acris Antibodies GmbH; or clone 1E7.2, available from Caltag and Upstate Biotechnology), appropriately titered (UNIT 4.2) Phosphate-buffered saline (PBS; APPENDIX 2A) Fluorescein isothiocyanate (FITC)-conjugated polyclonal goat anti-mouse IgG F(ab )2 , appropriately titered Normal mouse serum Fluorochrome-conjugated MAbs against B lymphocytes (CD19-PerCP-Cy5.5), T&NK cells (PE-conjugated, e.g., CD3 and CD56), and CLL cells (CD5-APC), appropriately titered 12 × 75–mm polystyrene tubes Flow cytometer with a 488-nm HeNe laser and a 635-nm red diode laser Fix samples 1. Assess the number of nucleated cells in the sample using a conventional hematology cell analyzer (APPENDIX 3A). If necessary, adjust the final concentration to 10 × 106 /ml with PBS. 2. Put 50 µl sample into a 12 × 75–mm polystyrene tube labeled with the appropriate combination of MAbs. 3. Add 50 µl fixation solution, mix, and incubate 15 min. Good results for intracellular staining are obtained using Fix & Perm (Caltag), Intrastain (Dako), or reagents of similar composition. The procedure described here can be applied to any of those reagents. The fixative reagent contains formaldehyde.
4. Add 4 ml PBS and centrifuge 5 min at 350 × g. Aspirate the supernatant, leaving no more than 100 µl in the tube. 5. Resuspend the cell pellet, being especially careful that suspension is complete if whole peripheral blood is used.
Permeabilize and stain samples for ZAP-70 6. Add 50 µl permeabilization buffer and the appropriate volume of ZAP-70 MAb (as determined by titering; e.g., 1.5 µl of undiluted MAb). Mix well and incubate 15 min. The permeabilizing reagent contains saponin. ZAP-70 Staining in Chronic Lymphocytic Leukemia
Two clones recognizing ZAP-70, 2F3.2, and 1E7.2, have been reported to give good results, and both are available from several commercial sources (see above). Store MAb frozen in aliquots to avoid freeze/thaw cycles.
6.19.2 Supplement 32
Current Protocols in Cytometry
7. Add 4 ml PBS and centrifuge 5 min at 350 × g. Carefully aspirate the supernatant without disturbing the pellet, to minimize cell loss. 8. Resuspend the cell pellet. Add the appropriate volume of FITC-labeled goat antimouse immunoglobulins (e.g., 20 µl of a 1/20 dilution) and incubate 20 min. 9. Add 4 ml PBS and centrifuge 5 min at 350 × g. Carefully aspirate the supernatant without disturbing the pellet. 10. Add 5 µl normal mouse serum. Mix well and incubate 5 min. The addition of normal mouse serum removes any FITC-labeled goat anti-mouse immunoglobulins left in the tube that could bind to the directly labeled MAb.
Stain cell subpopulations 11. Add 5 µl of each of the directly labeled MAbs to recognize T&NK cells (MAb-PE, e.g., CD3 and CD56), B cells (CD19-PerCP-Cy5.5) and CLL cells (CD5-APC). Mix well and incubate 20 min. If the flow cytometer is equipped with only three fluorescence detectors, no discrimination of normal B cells and CLL B cells can be performed. An additional tube for standard surface staining using directly labeled MAbs including CD5-PE and CD19 PerCP-Cy5.5 (see UNIT 6.2) is recommended. This additional tube provides information on the percent, if any, of normal residual B cells in the sample. As normal B cells do not express ZAP70, the presence of a high percentage of normal B cells will decrease the percent of B cells expressing ZAP-70 in the triple-fluorescence assay. The magnitude of the decrease is proportional to the percent of normal residual B cells present in the sample. Design the combination of markers to detect T&NK cells to include the majority of these cells (for example CD3 and CD56, but other combinations are also useful). The volume of directly labeled MAb necessary to obtain a good stain may differ depending on the supplier. Adjust the volume according to the recommendations of the supplier. If the flow cytometer is equipped with one laser, adapt the fluorochrome-conjugated MAb combinations to the instrument.
12. Add 4 ml PBS and centrifuge 5 min at 350 × g. 13. Aspirate the supernatant and add 500 µl PBS. Samples can be run immediately or kept at 4◦ C for a maximum of 6 hr. Gently mix the samples immediately before acquisition on the flow cytometer.
Set up instrument and acquire data 14. Set up the flow cytometer according to standard procedures to ensure correct signal for forward scatter (FS), side scatter (SS), and fluorescences, and with the appropriate electronic correction for spectral overlap of the fluorochromes used. 15. Create bivariate histograms (dot plots) to ensure correct acquisition for all parameters:
FS versus SS SS versus orange fluorescence Green fluorescence versus orange fluorescence Orange fluorescence versus red fluorescence (488 nm, HeNe laser) Green fluorescence versus red fluorescence (488 nm, HeNe laser) Red fluorescence (488 nm, HeNe laser) versus red fluorescence (635 nm, red diode laser). 16. In the histogram of SS versus orange fluorescence, draw a gate around T&NK lymphocytes. Phenotypic Analysis
6.19.3 Current Protocols in Cytometry
Supplement 32
17. Acquire at least 1000 T&NK lymphocytes within a minimum of 15,000 total events. In CLL samples with very low numbers of residual T cells this could take some minutes (e.g., for samples with 1% T cells 100,000 total events must be acquired to include 1000 T&NK cells).
Analyze data Perform data analysis with any of the software programs commercially available for the analysis of flow cytometry FCS files. In the example shown CellQuest Pro software (BD) is used. 18. Create bivariate histograms as follows:
Histogram 1: FS versus SS Histogram 2: CD19-PerCP-Cy5.5 versus CD5-APC Histogram 3: SS versus MAb-PE Histogram 4: ZAP-70 (indirectly labeled with FITC) versus MAb-PE Histogram 5: ZAP-70 (indirectly labeled with FITC) versus MAb-PE Histogram 6: ZAP-70 (indirectly labeled with FITC) versus MAb-PE (optional) If the flow cytometer is equipped with only three fluorescence detectors, create histograms 1, 4, and 5 as previously described. Replace histograms 2 and 3 with a new histogram 2: CD19-PerCP-Cy5.5 (red fluorescence) versus MAb-PE (orange fluorescence).
19. Draw a polygonal region (R1) that includes lymphocytes and excludes debris, doublets, monocytes, and other cells in histogram 1 (Fig. 6.19.1A). 20. In histogram 2 and 3 display only cells from region 1 (R1), corresponding to lymphocytes. 21. In histogram 2, draw a region (R2) including CLL cells (CD19+ CD5+ ; Fig. 6.19.1B). 22. In histogram 3, draw a region (R3) including T&NK cells (CD3+ CD56+ ; Fig. 6.19.1C). 23. In histogram 4 display cells fulfilling the criteria R1 and R3 and not R2 (T&NK cells). 24. In histogram 5 display cells fulfilling the criteria R1 and R2 and not R3 (CLL cells). 25. In histogram 4, place the markers to include the cluster of cells in the upper right quadrant (Fig. 6.19.1D to 6.19.1L). Some samples present a small percent of T&NK cells with low or no expression of ZAP-70. In these cases include in the upper right quadrant only the T&NK cells that form a tight cluster. See Figures 6.19.1D and 6.19.1J.
26. Copy the quadrant marker and paste into histogram 5. Record the percent of CLL positive for ZAP-70 (lower right quadrant; Fig. 6.19.1E, 6.19.1H, and 6.19.1K). 27. In histogram 6, display events fulfilling the criterion R1, and paste the quadrant marker from histogram 4. This histogram is optional, but the green fluorescence signals for both T&NK cells and B cells are represented, giving a visual image of the relationship. When the assay is performed with triple immunofluorescence, analyze in the same way with the exception of drawing R2 and R3 in new histogram 2. R2 includes cells negative for MAb-PE and positive for CD19-PerCP-Cy5.5 (B cells), and R3 includes cells positive for MAb-PE and negative for CD19-PerCP-Cy5.5 (T&NK cells; Fig. 6.19.2). ZAP-70 Staining in Chronic Lymphocytic Leukemia
When normal residual B cells are present in the sample (as depicted in Fig. 6.19.1B, gate R4), they are always negative for ZAP-70 as are B cells in normal samples (Fig. 6.19.1L).
6.19.4 Supplement 32
Current Protocols in Cytometry
Figure 6.19.1 Analysis of indirect staining for ZAP-70 using a four-color protocol. Data were acquired and analyzed using CellQuest software. Histograms (A) to (C) show the gating regions to select lymphocytes (R1), CLL cells (R2), T&NK cells (R3), and normal residual B cells (R4). Three patients are represented [(D) to (F), (G) to (I), and (J) to (L)]. Histogram (A) shows all events acquired. Histograms (D), (G), and (J) correspond to gated cells according to the gating criteria R1 and R3 and not R2. Histograms (E), (H), and (K) represent cells according to the criteria R1 and R2 and not R3. Histograms (B), (F), and (I) represent cells of gating region R1 (lymphocytes). Histogram (L) represents cells fulfilling the criteria R1 and R4 and not R3 (normal residual B cells). Phenotypic Analysis
6.19.5 Current Protocols in Cytometry
Supplement 32
Figure 6.19.2 Analysis of indirect staining for ZAP-70 using a triple immunofluorescence assay. Data were acquired and analyzed using CellQuest software. Histogram (A) shows all events acquired. A region R1 to select lymphocytes is drawn. Histogram (B) displays cells from R1. Additional regions R2 (CD19+ , B cells) and R3 (T&NK cells) are drawn. Histogram (C) displays cells fulfilling the criteria R1 and R3. Histogram (D) displays cells fulfilling the criteria R1 and R2.
ALTERNATE PROTOCOL
DIRECT IMMUNOFLUORESCENCE STAINING OF ZAP-70 The expression of ZAP-70 in B cells was an unexpected finding and probably for this reason no directly labeled MAbs recognizing ZAP-70 were available for flow cytometry for the early studies. Recently several anti-ZAP-70 MAbs conjugated to different fluorochromes have been released. The use of these MAbs simplifies the staining protocol (see also UNIT 6.6). A simplified method using a MAb against ZAP-70 (clone 1E7.2) conjugated to Alexa Fluor 488 (emitting green fluorescence) is described below because this combination of MAb clone and fluorochrome has shown its predictive value in clinical analysis.
Additional Materials (also see Basic Protocol) Fluorochrome-conjugated MAb against ZAP-70 conjugated to Alexa Fluor 488 (Caltag), isotypic MAb conjugated to Alexa Fluor 488, MAb against B lymphocytes (CD19-PerCP-Cy5.5), MAb against T&NK cells (PE-conjugated, e.g., CD3 and CD56), and MAb against CLL cells (CD5-APC), appropriately titered ZAP-70 Staining in Chronic Lymphocytic Leukemia
6.19.6 Supplement 32
Current Protocols in Cytometry
Prepare samples 1. Assess the number of nucleated cells in the sample using a conventional hematology cell analyzer (APPENDIX 3A). If necessary, adjust the final concentration to 10 × 106 /ml with PBS. 2. Label two polystyrene tubes: one to receive an isotypic MAb-Alexa Fluor 488 (isotypic negative control) for baseline green autofluorescence (isotype-specific MAb–Alexa Fluor 488; T&NK-PE/CD19-PerCP-Cy5.5/CD5-APC), and the other to receive a combination including the ZAP-70-Alexa Fluor 488 monoclonal antibody (anti-ZAP-70-Alexa Fluor 488/T&NK-PE/CD19PerCP-Cy5.5/CD5-APC). Place 50 µl cell suspension at a concentration of 10 × 106 cells/ml in each tube.
Surface staining 3. Add 5 µl of each of the directly labeled MAbs able to recognize T&NK cells (MAbPE), B cells (CD19-PerCP-Cy5.5), and CLL cells (CD5-APC) to both tubes. Mix well and incubate 15 min. The volume of directly labeled MAb necessary to obtain a good stain may differ depending on the supplier. Adjust the volume according to the recommendations of the supplier.
4. Add 50 µl fixation medium from the Fixation and Permeabilization Kit, mix, and incubate 15 min. The fixative reagent contains formaldehyde.
5. Add 4 ml PBS and centrifuge 5 min at 350 × g. Aspirate the supernatant, leaving no more than 100 µl in the tube. 6. Resuspend the cell pellet, being especially careful if whole peripheral blood is employed.
Perform intracellular staining 7. Add 50 µl permeabilization medium. 8. Add 5 µl of either anti-ZAP-70-Alexa Fluor 488 or isotypic control MAb-Alexa Fluor 488. Mix well and incubate 15 min. 9. Add 4 ml PBS and centrifuge 5 min at 350 × g. Aspirate the supernatant. 10. Repeat step 9. 11. Resuspend cells and add 500 µl PBS. Samples can be run immediately or kept at 4◦ C for a maximum of 6 hr. Gently mix the samples immediately before sample acquisition on the flow cytometer.
Set up instrument and acquire data 12. Set up instrument as described in the Basic Protocol, steps 14-17. Acquire the sample incubated with isotypic MAb-Alexa Fluor 488 first, followed by the sample incubated with anti-ZAP-70-Alexa Fluor 488. Analyze data 13. Create the same histograms described in the Basic Protocol, step 18. The analysis could be performed as described in the Basic Protocol or by considering the green autofluorescence signal of CLL cells.
14. Create bivariate histograms as follows:
Histogram 1: FS versus SS Histogram 2: CD19-PerCP-Cy5.5 versus CD5-APC Histogram 3: MAb-green fluorescence versus CD19-PerCP-Cy5.5
Phenotypic Analysis
6.19.7 Current Protocols in Cytometry
Supplement 32
Histogram 4: MAb-green fluorescence versus CD19-PerCP-Cy5.5 Histogram 5: MAb-green fluorescence versus MAb-PE Histogram 6: SS versus MAb-PE (optional) When the flow cytometer is equipped with only three fluorescence detectors, create histograms 1, 3, 4, and 5 as previously described. Replace histogram 2 with a new histogram 2: CD19-PerCP-Cy5.5 (red fluorescence) versus MAb-PE (orange fluorescence).
15. Draw a polygonal region (R1) that includes lymphocytes and excludes debris, doublets, and other cells in histogram 1 (Fig. 6.19.1A) 16. In histogram 2, display only cells from region 1 (R1), corresponding to lymphocytes. 17. In histogram 2, draw a region (R2) including CLL cells (CD19+ CD5+ ; see Fig. 6.19.1B). It is convenient to ensure that no T&NK cells are included in this R2 region. For this purpose it is recommended to display in histogram 6 cells from region 1 and draw a polygonal region (R3) that includes T&NK cells (Fig. 6.19.1C).
18. In histogram 3, display cells fulfilling the criteria R1 and R2 (CLL cells) from the tube stained with the isotypic negative control MAb-Alexa Fluor 488 and place a marker to include cells in the left portion (Figs. 6.19.3A and 6.19.3C). When region R3 has been drawn, display in histogram 3 cells fulfilling the criteria R1 and R2 but not R3.
ZAP-70 Staining in Chronic Lymphocytic Leukemia
Figure 6.19.3 Analysis of direct staining of ZAP-70 using a four-color protocol. Data were acquired and analyzed using CellQuest software. Data from two patients are displayed [(A) and (B), and (C) and (D), respectively]. Histograms (A) to (D) include only CLL cells (CD19+ CD5+ T&NK− ; R1 and R2 and not R3), as described in Figure 6.19.1. Histograms (A) and (C) show the baseline fluorescence in a tube stained with an isotypic MAb conjugated to Alexa Fluor 488. Histograms (B) and (D) represent fluorescence of B-CLL cells after staining with anti-ZAP-70-Alexa Fluor 488.
6.19.8 Supplement 32
Current Protocols in Cytometry
19. Copy the marker. 20. In histogram 4, display cells fulfilling the criteria R1 and R2 (CLL cells) from the tube stained with anti-ZAP-70-Alexa Fluor 488. Ensure that R1 and R2 are correct for the cells of this tube. When region R3 has been drawn, display in histogram 4 cells fulfilling the criteria R1 and R2 but not R3.
21. Paste the marker and record the percent of CLL cells positive for ZAP-70 (right portion; Fig. 6.19.3B and 6.19.3D). 22. In histogram 5, display events fulfilling the criterion R1 for the tube stained with anti-ZAP-70-Alexa Fluor488. This histogram ensures that the technique has been correctly performed: T&NK must express a homogeneous strong signal for ZAP-70. When the assay is performed with triple immunofluorescence, analyze the stained cells in the same way with the exception of drawing R2 and R3 in the new histogram 2. R2 includes cells negative for MAb-PE and positive for CD19-PerCP-Cy5.5 (B cells), and R3 includes cells positive for MAb-PE and negative for CD19-PerCP-Cy5.5 (T&NK cells; see Fig. 6.19.2).
COMMENTARY Background Information Chronic lymphocytic leukemia (CLL) is a lymphoproliferative disorder of mature B lymphocytes and represents ∼90% of all chronic lymphoid leukemias in Western countries. The diagnosis of the disease is relatively straightforward and is based on the morphological, histological, and phenotypic characteristics of leukemic lymphocytes (M¨uller-Hermelink et al., 2001). In contrast, the clinical course of the disease is highly variable. The median survival of patients is ∼10 years, but a number of patients die soon after diagnosis, while the life expectancy of others is unaffected by the disease (Rozman and Montserrat, 1995). The identification of patients with different potential for survival is important, because the currently available therapeutic options have different efficacy and toxicity. Several prognostic factors for CLL patients have been described during the last 30 years based on clinical examination, histological characteristics, and laboratory data (Montserrat, 2002; Byrd et al., 2004). Clinical staging systems based on data from clinical examination and peripheral blood cell counts identify groups of CLL patients with different estimated survival. However, the most commonly employed clinical staging systems are unable to predict which patients are more likely to progress. Patients affected with CLL are more frequently diagnosed in very early stages of the disease. Thus, there is a need to identify other features able to indicate the risk of progression in such patients. In 1999
two independent groups reported that a unique cellular characteristic was able to discriminate two different clinical forms of CLL: the configuration of the immunoglobulin variable heavy-chain (IgVH ) gene (Damle et al., 1999; Hamblin et al., 1999). Patients whose CLL cells have an unmutated configuration of IgVH genes present with unfavorable prognostic features, the disease is aggressive with high risk of progression, the patients require treatment, and have shortened survival (∼8 years). In contrast, patients whose CLL cells show hypermutations in the IgVH genes present with favorable features, do not require treatment, remain stable, and have long survival (∼20 years). Unfortunately, the technology to determine the IgVH gene configuration is complex and time consuming. For this reason, at the same time as the mutational status of IgVH genes was recognized as an important prognostic parameter in CLL patients, the search began for additional, more easily obtained cellular characteristics that retain the power to segregate the two clinical forms of CLL. CD38, a transmembrane glycoprotein with a discontinuous pattern of expression in leukocytes, was proposed as a surrogate for IgVH (Damle et al., 1999). CD38 expression correlates with an unmutated configuration of IgVH. Unfortunately, a quarter of cases are misclassified using this surface antigen, and the expression of CD38 changes during the course of the disease.
Phenotypic Analysis
6.19.9 Current Protocols in Cytometry
Supplement 32
ZAP-70 Staining in Chronic Lymphocytic Leukemia
The description of two clinical forms of CLL closely related to the IgVH gene configuration also resulted in a debate over the cellular origin of CLL. The gene expression profiling technology strongly suggests that CLL cells have a unique cellular origin because a distinct and highly uniform profile characterizes CLL independently of the IgVH mutational status (Rosenwald et al., 2001). The two clinical forms differ only in the expression of a small subset of genes. The ZAP-70 gene is highly overexpressed in CLL patients with an unmutated configuration of IgVH genes, whereas it is absent in patients with somatically hypermutated IgVH genes. The overexpression of ZAP-70 and its tight relationship with the unmutated type of CLL have been corroborated by other techniques such as RT-PCR (Durig et al., 2003), real-time quantitative PCR (Wiestner et al., 2003), immunoblot (Chen et al., 2002; Crespo et al., 2003), immunohistochemistry (Carreras et al., 2005), and flow cytometry (Crespo et al., 2003; Durig et al., 2003; Orchard et al., 2004; Rassenti et al., 2004). The expression of ZAP-70 correlates better than CD38 antigen with the mutational status of IgVH , with 80% (Rassenti et al., 2004) to 90% of cases (Crespo et al., 2003; Orchard et al., 2004) showing agreement between both features. ZAP-70 is a protein closely associated with membrane cell receptors and transduces membrane signals to intracellular pathways. In agreement, CLL cells expressing ZAP-70 show greater tyrosine phosphorylation of cytosolic proteins (Chen et al., 2002), and calcium influx after B cell receptor (BCR) engagement (Chen et al., 2004), suggesting that CLL cells expressing ZAP-70 signal more effectively through BCR than CLL cells lacking ZAP-70. This different biological behavior could contribute to the higher aggressiveness of ZAP-70-positive CLL. Of note, ZAP-70 remains stable during the evolution of the disease in all the studies published so far (Crespo et al., 2003; Durig et al., 2003; Orchard et al., 2004; Rassenti et al., 2004). From the technical point of view, the high expression of ZAP-70 in T and NK cells, at both the RNA and protein levels, precludes the use of unsorted samples in methodologies that analyze the sample globally (RTPCR, real-time PCR, or immunoblot) when more than 10% of T&NK are present in a CLL sample. In contrast, multiparametric flow cytometry allows the analysis of the expression of ZAP-70 in each individual cell subpopulation, with T and NK cells being an inter-
nal positive control for the staining reaction. Different staining protocols and different cutoff values to establish a high expression of ZAP-70 have been used. So far, there is no published study in which a directly labeled commercially available MAb has been employed. Instead, indirect immunofluorescence staining (Crespo et al., 2003; Durig et al., 2003; Orchard et al., 2004) or a custom-labeled MAb (Rassenti et al., 2004) has been used. Cells are fixed using paraformaldehyde-based buffers (Crespo et al., 2003; Durig et al., 2003; Orchard et al., 2004; Rassenti et al., 2004) and permeabilized using detergent-based buffers (saponin: Crespo et al., 2003; Durig et al., 2003; Rassenti et al., 2004; or tween: Orchard et al., 2004), with minimal differences in their composition. In most cases, the staining procedure immediately follows the fixation and permeabilization step, but storage of fixed and permeabilized cells in cold ethanol (80%, −20◦ C) has also been described (Orchard et al., 2004). Two different ways to define positive or high-expressing ZAP-70 cells are employed. One takes in consideration the signal obtained in the isotypic MAb negative control staining (Orchard et al., 2004; Rassenti et al., 2004). The other considers the aberrant expression of ZAP-70 in leukemic cells when CLL cells have as much ZAP-70 as high ZAP-70-expressing T&NK cells (Crespo et al., 2003; Durig et al., 2003). No direct comparison between the two methods of analysis has been performed, but probably most cases would remain in the same ZAP-70 categorical group (Fig. 6.19.4). The cutoff level employed to assess the ZAP-70 category (positive versus negative) varies from 10% (Orchard et al., 2004) to 20% (Crespo et al., 2003; Durig et al., 2003; Rassenti et al., 2004). The criteria employed to choose the cutoff value include either the mean value plus 3 standard deviations (SD) of ZAP-70 in normal B cells (Orchard et al., 2004), or the values best predicting IgVH mutational status (Crespo et al., 2003) or the clinical outcome (Rassenti et al., 2004). The above protocols constitute a methodological approach to analyze ZAP-70 in CLL cells, a protein that correlates with the clinical outcome of the patients. In comparison with other methodologies to assess ZAP-70, flow cytometry is faster, less expensive, and less complex than other methods, is applicable to all CLL patients, and could be introduced as an additional tube in the diagnostic panel employed for patients with CLL.
6.19.10 Supplement 32
Current Protocols in Cytometry
Figure 6.19.4 Comparison of both methods of staining and analysis. Data from four different patients are shown [(A) to (D), (E) to (H), (I) to (L), (M) to (P)]. The two left columns display cells from the R1 gate region (lymphocytes), the two right columns display CLL cells (CD19+ CD5+ ; R1 and R2 and not R3, see Fig. 6.19.1). The percent values are CLL cells positive for ZAP-70. Histograms (A), (E), (I,) and (M) display staining with the indirect technique. Histograms (B), (F), (J), and (N) display the results of a direct ZAP-70 staining analyzed according to T&NK cell expression. Histograms (C), (G), (K), and (O) display the signal obtained after staining with isotype negative MAb-Alexa Fluor 488. Histograms (D), (H), (L), and (P) display the values of ZAP-70-positive CLL cells according to the isotypic negative control.
Critical Parameters Sample quality A fresh peripheral blood sample is the most suitable specimen for immunophenotyping for ZAP-70 quantification, basically for two reasons: (1) the presence of dead cells and debris hampers the identification of clean cell subpopulations and increases the green fluores-
cence in an unspecific manner; (2) no analysis on the stability of the ZAP-70 protein over long periods of time has been performed. The staining signal obtained in flow cytometry differs among MAb clones against ZAP-70. At present time, two clones recognizing ZAP-70 (2F3.2 and 1E7.2) have shown good and reproducible results. Recently, there has been a release of anti-ZAP-70 antibodies
Phenotypic Analysis
6.19.11 Current Protocols in Cytometry
Supplement 32
Figure 6.19.5 Comparison of staining of directly labeled antibodies obtained in indirect and direct staining of ZAP-70. Samples have been acquired using CellQuest software and analyzed using Paint-a-Gate software. Histograms (A) to (F) display cells from gate R1 (lymphocytes) from the same patient. Histograms (A), (B), and (C) show the fluorescence signal of directly labeled antibodies after indirect staining for ZAP-70. Histograms (D), (E), and (F) represent the same fluorescence obtained after direct staining for ZAP-70. Each cell subpopulation in the sample is identified.
ZAP-70 Staining in Chronic Lymphocytic Leukemia
conjugated to a variety of fluorochromes, but these antibodies have not yet been clinically tested. It should be stressed that prognostic information from ZAP-70 by flow cytometry has been reported only using indirect staining with FITC or using ZAP-70 MAb custom-labeled with Alexa Fluor 488. Also, preliminary reports suggest that ZAP-70 antibodies labeled with fluorochromes other than FITC or Alexa Fluor 488 could give different staining results. Therefore, depending on the application of the assay, selection of fluorochrome should be restricted until new information is available. The
staining using a directly labeled MAb against ZAP-70 results in a higher quality of staining for MAb directly labeled in the tube, as shown in Figure 6.19.5. In the Basic Protocol the sample is subjected to repeated steps of centrifugation and aspiration of the supernatant that could result in significant cellular losses. Minimize cell losses by aspirating the supernatant carefully and by leaving enough buffer volume (∼150 to 200 µl) above the pellet in noncritical steps (i.e., all but step 4). The incubation with normal mouse serum following the incubation with
6.19.12 Supplement 32
Current Protocols in Cytometry
Figure 6.19.6 ZAP-70 staining patterns observed in the different subpopulations of lymphocytes from a CLL patient whose neoplastic B cells are ZAP-70 negative. Samples have been acquired using CellQuest software and analyzed using Paint-a-Gate software. Histogram (A) displays cells from gating region R1 (lymphocytes). Each cell subpopulation in the sample is identified. Histogram (B) shows the fluorescence intensity of each cell subpopulation after direct staining for ZAP-70. The mean fluorescence intensity of each cell subpopulation is shown in parentheses.
anti-mouse-FITC reagent and before the addition of the directly labeled mouse MAb is mandatory to remove any free anti-mouse antibody that would also bind to the directly labeled antibodies. With all these considerations in mind, staining with a reliable directly labeled MAb would help to shift to a shorter, less complex, and better staining protocol for ZAP-70 in CLL patients. Two methodological approaches to establish the percent of ZAP-70-positive CLL cells are in use, both establishing a cutoff value in order to assign the patient risk category. As the percent of CLL cells positive for ZAP-70 defines the prognosis, during the analysis, establishing the cutoff markers is critical and should be carefully performed. ZAP-70 is expressed in CLL cells as a continuum in the fluorescence signal, so the position of the marker could influence the final result in both methods of analysis. When T&NK cells are used, the marker has to include a homogenous cluster of ZAP-70-positive T&NK cells. However, for samples with a very low percent of T&NK it could be difficult to establish the marker position. As mentioned, normal B cells do not express significant amounts of ZAP-70. Therefore, in samples from CLL patients containing residual normal B cells, the staining of the normal B cells should be negative in both ZAP70-negative and ZAP-70-positive CLL. This could serve as an internal control for a negative population. The gate of cell subpopulations is also an important parameter to perform a correct analysis. Debris and doublets increase green fluorescence in an unspecific
manner and should be excluded from the analysis gate. As mentioned, T&NK cells have high intracellular ZAP-70 content and it is necessary to avoid contamination with these cells in the CLL gate. Otherwise, the percent of ZAP70-positive CLL cells is inaccurately increased through T&NK cell contamination. Finally, although both methods of analysis, isotypic control or T&NK cells, are able to discriminate the two clinical forms of CLL and probably will classify most patients identically, the relationship between both methods of analysis should be addressed.
Anticipated Results In normal samples, staining with ZAP-70 MAb for lymphocytes results in a discrete population with homogeneous ZAP-70 expression, whereas a small proportion of cells do not express ZAP-70. The strongest signal is observed in NK cells, followed by T cells. Normal B cells do not express significant amounts of ZAP-70 (Fig. 6.19.6). In CLL samples the T&NK cells react strongly, whereas ZAP-70 expression in CLL cells ranges from negative to strong homogeneous positive (similar to T&NK cells). In ∼40% to 60% of CLL patients the leukemic lymphocytes have low expression (<20% of CLL cells) or no expression of ZAP-70. In the remaining patients, CLL cells stain for ZAP-70 with a continuous spectrum of intensity, obtaining values ranging from 20% to 100% positive B-CLL cells. Normal residual B cells, when present in a CLL sample that expresses ZAP-70, have the normal pattern of
Phenotypic Analysis
6.19.13 Current Protocols in Cytometry
Supplement 32
expression for B cells from normal peripheral blood (see Fig. 6.19.1L). Sequential analysis of samples from the same patient will result in the assignment to the same risk group (∼95% of patients).
Time Consideration The preparation of sample (white cell count and dilution if necessary) takes between 2 and 5 min. The staining of a sample following the Basic Protocol requires ∼1 hr 45 min. The staining of a sample following the Alternate Protocol requires ∼60 min. The cytometer is set up during one of the staining steps. The time for sample acquisition when T cells are employed as the marker for positivity depends on the proportion of T&NK cells present. The time for acquisition when an isotypic negative control is employed takes ∼1 min if ∼20,000 cells are acquired. Sample analysis takes ∼5 min, less if a template protocol for analysis is available.
Literature Cited Byrd, J.C., Stilgenbauer, S., and Flinn, I.W. 2004. Chronic lymphocytic leukemia. Hematology (Am. Soc. Hematol. Educ. Program) 2004:163183. Carreras, J., Villamor, N., Colomo, L., Moreno, C., Ramon y Cajal, S., Crespo, M., Tort, F., Bosch, F., Lopez-Guillermo, A., Colomer, D., Montserrat, E., and Campo, E. 2005. ZAP-70 expression in B-cell lymphoid neoplasms: An immunohistochemical study. J. Pathol. 205:507-513. Chen, L., Widhopf, G., Huynh, L., Rassenti, L., Rai, K.R., Weiss, A., and Kipps, T.J. 2002. Expression of ZAP-70 is associated with increased Bcell receptor signaling in chronic lymphocytic leukemia. Blood 100:4609-4614. Chen, L., Apgar, J., Huynh, L., Dicker, F., GiagoMcGahan, T., Rassenti, L., Weiss, A., and Kipps, T.J. 2004. ZAP-70 directly enhances IgM signaling in chronic lymphocytic leukemia. Blood 105:2036-2041. Crespo, M., Bosch, F., Villamor, N., Bellosillo, B., Colomer, D., Rozman, M., Marce, S., LopezGuillermo, A., Campo, E., and Montserrat, E. 2003. ZAP-70 expression as a surrogate for immunoglobulin-variable-region mutations in chronic lymphocytic leukemia. N. Engl. J. Med. 348:1764-1775. Damle, R.N., Wasil, T., Fais, F., Ghiotto, F., Valetto, A., Allen, S.L., Buchbinder, A., Budman, D., Dittmar, K., Kolitz, J., Lichtman, S.M., Schulman, P., Vinciguerra, V.P., Rai, K.R., Ferrarini, M., and Chiorazzi, N. 1999. Ig V gene mutation status and CD38 expression as novel prognostic indicators in chronic lymphocytic leukemia. Blood 94:1840-1847.
Durig, J., Nuckel, H., Cremer, M., Fuhrer, A., Halfmeyer, K., Fandrey, J., Moroy, T., KleinHitpass, L., and Duhrsen, U. 2003. ZAP-70 expression is a prognostic factor in chronic lymphocytic leukemia. Leukemia 17:2426-2434. Hamblin, T.J., Davis, Z., Gardiner, A., Oscier, D.G., and Stevenson, F.K. 1999. Unmutated Ig VH genes are associated with a more aggressive form of chronic lymphocytic leukemia. Blood 94:1848-1854. Montserrat, E. 2002. Classical and new prognostic factors in chronic lymphocytic leukemia: Where to now? Hematol. J. 3:7-9. M¨uller-Hermelink, H.K., Montserrat, E., Catovsky, D., and Harris, N.L. 2001. Chronic lymphocytic leukaemia/small lymphocytic lymphoma. In World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues (E.S. Jaffe,N.L. Harris,H. Stein, andJ.W. Vardiman, eds.) pp. 127-130. IARC Press, Lyon. Orchard, J.A., Ibbotson, R.E., Davis, Z., Wiestner, A., Rosenwald, A., Thomas, P.W., Hamblin, T.J., Staudt, L.M., and Oscier, D.G. 2004. ZAP-70 expression and prognosis in chronic lymphocytic leukaemia. Lancet 363:105-111. Rassenti, L.Z., Huynh, L., Toy, T.L., Chen, L., Keating, M.J., Gribben, J.G., Neuberg, D.S., Flinn, I.W., Rai, K.R., Byrd, J.C., Kay, N.E., Greaves, A., Weiss, A., and Kipps, T.J. 2004. ZAP70 compared with immunoglobulin heavy-chain gene mutation status as a predictor of disease progression in chronic lymphocytic leukemia. N. Engl. J. Med. 351:893-901. Rosenwald, A., Alizadeh, A.A., Widhopf, G., Simon, R., Davis, R.E., Yu, X., Yang, L., Pickeral, O.K., Rassenti, L.Z., Powell, J., Botstein, D., Byrd, J.C., Grever, M.R., Cheson, B.D., Chiorazzi, N., Wilson, W.H., Kipps, T.J., Brown, P.O., and Staudt, L.M. 2001. Relation of gene expression phenotype to immunoglobulin mutation genotype in B cell chronic lymphocytic leukemia. J. Exp. Med. 194:1639-1647. Rozman, C. and Montserrat, E. 1995. Chronic lymphocytic leukemia. N. Engl. J. Med. 333:10521057. Wiestner, A., Rosenwald, A., Barry, T.S., Wright, G., Davis, R.E., Henrickson, S.E., Zhao, H., Ibbotson, R.E., Orchard, J.A., Davis, Z., StetlerStevenson, M., Raffeld, M., Arthur, D.C., Marti, G.E., Wilson, W.H., Hamblin, T.J., Oscier, D.G., and Staudt, L.M. 2003. ZAP-70 expression identifies a chronic lymphocytic leukemia subtype with unmutated immunoglobulin genes, inferior clinical outcome, and distinct gene expression profile. Blood 101:4944-4951.
Contributed by Neus Villamor Hospital Cl´ınic de Barcelona Barcelona, Spain
ZAP-70 Staining in Chronic Lymphocytic Leukemia
6.19.14 Supplement 32
Current Protocols in Cytometry
Multiparameter Analysis of Intracellular Phosphoepitopes in Immunophenotyped Cell Populations by Flow Cytometry
UNIT 6.20
This unit presents several protocols for measuring intracellular phosphoepitopes by flow cytometry. These assays enable biochemical investigations in both human and murine primary cells, as well as in cell lines. Conventional methods that require cellular lysis, such as immunoblots or immunoprecipitations, cannot discriminate between proteins from different cellular subsets and ultimately provide averaged protein readings. The advantage of multiparameter flow cytometry is apparent in immunophenotypical analysis. Intracellular detection of signaling molecules by flow cytometry, namely the detection of phosphorylated and nonphosphorylated molecules, has recently exposed the heterogeneity that can be observed upon signal transduction. Although the advantages of intracellular staining methodologies for cytokines and cell cycle antigens have been appreciated for years, detection of phosphorylated molecules presents several new challenges. Many of these challenges, including generation of the reagents and details of the staining techniques (Perez and Nolan, 2002; Krutzik et al., 2004; Perez et al., 2004), have been addressed in this unit. As these techniques are adapted to new applications, the protocols continue to be refined. This unit describes phosphoepitope protocols for amplification of intracellular staining signals upon alcoholbased (see Basic Protocol 1) and detergent-based (see Basic Protocol 2) permeabilizations, with alternative instructions for staining adherent cells; procedures are also described for whole-blood staining (see Basic Protocol 3) and multiparameter staining of surface and intracellular antigens (see Basic Protocols 4 and 5 and Alternate Protocol).
STRATEGIC PLANNING Selection of cell types and optimization of culture conditions are critical in detecting differences in phosphoepitopes. For example, cell lines have the ability to adapt to culture conditions and can yield heterogeneous responses. The authors of this unit were able to identify eight different Jurkat lines, presumed to be from the same original source, that responded differently to stimulations. This observation ultimately required characterization based on phenotype to correlate it with functional responsiveness (Fig. 6.20.1A). Preactivated cells show a characteristic drop in some phosphoepitope staining upon stimulation (Fig. 6.20.1A). This is indicative that either the cell clone or the cell handling procedures need to be optimized. Two general strategies exist for permeabilization of cells, a detergent-based method (see Basic Protocol 2) and an alcohol-based method (see Basic Protocol 1), both of which have advantages and disadvantages (Perez et al., 2004). The use of multicolor, multiparameter flow cytometry requires primary conjugated antibodies at defined fluorophore-to-protein (FTP) ratios. It is not sufficient to give a range of FTP ratios, for example 2 to 8, such as the majority of conjugation kits for small molecule fluorophores indicate. It is necessary to quantitate the final product thoroughly, as FTP ratios differing between two molecules can represent significant decreases in phosphoepitope staining (Fig. 6.20.1B). This effect of disparities in FTP appears to be more pronounced in the case of phosphoepitopes than with intracellular antigens such as cytokines, and cannot be anticipated a priori for this type of analysis. It is also important to note that each fluorophore’s optimal FTP is unique and can differ among antibody clones directed against phosphoepitopes. Immunophenotyping Contributed by Omar D. Perez, Dennis Mitchell, Roberto Campos, Guo-Jian Gao, Li Li, and Garry P. Nolan
6.20.1
Current Protocols in Cytometry (2005) 6.20.1-6.20.22 C 2005 by John Wiley & Sons, Inc. Copyright
Supplement 32
Figure 6.20.1
Legend at right.
Multiparameter Analysis of Intracellular Phosphoepitopes
6.20.2 Supplement 32
Current Protocols in Cytometry
In the authors’ earlier work, extensive quality control in the conjugation procedures yielded maximal detection of phosphorylation (Perez and Nolan, 2002). Although many phosphospecificities are commercially available and are beginning to enter the market as direct conjugates, including the antibody clones that the authors’ laboratory has validated by flow cytometry, particular conjugations of specificities may need to be generated inhouse. Many times, the conjugation differences result in submaximal detection. For these applications, in particular for phosphoantibody conjugation, the reader is directed to several references illustrating the quality control and validation schemes for testing the reagent (Perez and Nolan, 2002; Krutzik and Nolan, 2003; Perez et al., 2004). Cell handling is also critical in human cell isolation procedures. An example of signaling differences observed upon temperature and time handling for blood from the same donor is presented in Figure 6.20.1C. These parameters are to be defined by the investigator. Since the phosphoantibodies are generated against linear phosphopeptides and then subsequently purified by phosphopeptide affinity columns, there is a concern that antibodies may not fully detect phosphoepitopes when in intracellular conformations or localized regions of the cell. This is evident in that there are antibody clones to phosphoepitopes that perform better on immunoblots than by ELISA or by immunofluorescence. As is the case in immunofluorescence microscopy, permeabilization by Triton X-100 (Basic Protocol 2) or methanol/acetate (Basic Protocol 1) can yield different results in patterns of localization. For flow cytometry, the differences in detergent-based and alcohol-based permeabilizations can yield differences in signal intensities detected using the same conjugated primary phosphoantibody, and often result in different optimal titrations (Fig. 6.20.2A). Since staining is relative to an unstimulated control sample or to a normalization control such as a non-phospho protein, detectable differences between stimulated and unstimulated are usually sufficient. However, one may want to detect both phospho- and nonphospho target proteins simultaneously in the same cell. This requires careful evaluation of antibody pairs, as all pairs for a given target do not necessarily work intracellularly (O. Perez and G. Nolan, unpub. observ.). An example of a ratiometric staining for phospho-syk and non-phospho syk is presented in Figure 6.20.2B. There is
Figure 6.20.1 (at left) Effect of cell type, FTP ratio, and primary cell handling on phosphoepitope detection by flow cytometry. (A) 1 × 106 Jurkat clones (clone 010805 and clone 020202) were left in complete medium and stimulated 15 min with 500 ng/ml PMA/1 µM ionomycin. Cells were fixed, permeabilized, and stained (Basic Protocol 2) with anti-phospho-p44/42(T202/Y204)-Alexa647 (clone 20a) at 0.125 µg/stain. Cells were immunophenotyped for CD3, CD11a, CD25, and CD69 expression. Down-shifting phospho stains post-stimulation are indicative of preactivated state. (B) 1 × 106 Jurkat cells (clone 010805) were stimulated 15 min with 500 ng/ml PMA/1 µl ionomycin or 25 ng/ml phytohemagglutin (PHA). Cells were fixed, permeabilized and stained (Basic Protocol 2) with anti-phospho-p44/42(T202/Y204) (clone 20a) conjugated to varying ratios of Alexa-488 as indicated in the figure. 0.125 µg of antibody was used for all stains. (C) Freshly drawn human blood was either Ficoll separated and stimulated directly (upper row of graphs), Ficoll separated and allowed to rest 1 hr in tissue culture plates before stimulating (middle row of graphs), or allowed to sit at 25◦ C for 1 hr, Ficoll separated, and then stimulated (lower row of graphs). Peripheral blood lymphocytes were stimulated 15 min with 500 ng/ml PMA/ 1 µl ionomycin or 100 mg/nl INF-α prior to being processed as in Basic Protocol 2. Cells were stained with anti-phosphop44/42(T202/Y204)-PE (clone 20a), anti-phospho-p38(T180/y182)-Alexa647 (clone 36), and antiphospho-JNK (T183/185)-Alexa488 (clone 41). Antibodies were used at 0.125 µg (p-p44/42) and 0.3 µg (p-p38 and p-JNK). Ficoll-Paque Plus (Amersham Biosciences) was used for cell isolation. Cells were in the presence of 5% human AB serum (Irvine Scientific) in RPMI. This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www.interscience.wiley.com/c p/colorfigures.htm.
Immunophenotyping
6.20.3 Current Protocols in Cytometry
Supplement 32
Figure 6.20.2
Legend at right.
Multiparameter Analysis of Intracellular Phosphoepitopes
6.20.4 Supplement 32
Current Protocols in Cytometry
also a concern, using the detergent-based permeabilization techniques, if intracellular phosphoepitopes are buried in protein-protein interactions. For the authors, the only phosphoproteins not detectable by detergent-based permeabilization are the STAT phosphoproteins (of 80 specificities tested to date), and this is believed to be due to the fact that STAT proteins, upon phosphorylation, form tight dimers that are maintained upon fixation.
FLOW CYTOMETRIC ANALYSIS OF INTRACELLULAR PHOSPHOEPITOPES USING ALCOHOL-BASED PERMEABILIZATION
BASIC PROTOCOL 1
This protocol describes an assay for intracellular detection of phosphoepitopes. Cells are stimulated, fixed with methanol, and permeabilized. Cells are then stained with singlecolor phospho-antibodies and washed extensively. The amount of phosphorylated-antigen target is determined by flow cytometry. References on using alcohol permeabilizations for phosphoepitope detection include Fleisher et al. (1999); Chow et al. (2001); Uzel et al. (2001); Hilger et al. (2002); and Kaech et al. (2002).
Materials 1 × 106 cells/condition of stimulation in 1 ml medium 1 × 106 cells/unstimulated condition in 1 ml medium 1 × 106 cells as blank in 1 ml of medium Stimuli of interest 16% paraformaldehyde, methanol-free (Electron Microscopy Sciences) 90% (v/v) methanol, ice-cold Staining buffer 1 (see recipe) Pretitered, fluorescently conjugated phosphospecific antibody Antibody capture beads for compensation controls (UNIT 1.14) 12 × 75–mm polystyrene tubes Refrigerated centrifuge Microscope slides and hemocytometer Flow cytometer with appropriate band-pass filters
Figure 6.20.2 (at left) Comparison of staining protocols, non-phosphoprotein normalization, and adherent cell preparation staining. (A) 1 × 106 Jurkat cells were stimulated 15 min with 500 ng/ml PMA/ 1 µl ionomycin and processed for flow cytometry by Basic Protocol 1 (methanol) or Basic Protocol 2 (saponin). Cells were stained with anti-phospho-p44/42(T202/Y204)-Alexa488 (clone 20a) at 0.125 µg/stain for the methanol protocol and 1.0 µg/stain for the saponin protocol. (B) Ratiometric measurement of phospho and non-phospho target protein simultaneously. Human CD3+ cells were stimulated 20 min with 1 mM pervanadate and stained with anti–human Syk-FITC and anti-phospho-Zap70(Y319)/Syk(Y352)-PE at optimized titers. (C) Phosphoepitope detection in adherent human monocytes. Cells were allowed to adhere for 2 hr prior to being stimulated 15 min with either IFN-γ , IL-4, or GM-CSF (100 ng/ml). Cells were then processed per Basic Protocol 2 and stained for anti-phospho-STAT1(Y701)-AX647 (clone 14) and anti-phospho-STAT6(Y641)AX488 (clone 18) at 0.8 µg per 1 × 106 cells. (D) Human monocyte staining for phosphorylated Stat3. Human monocytes were prepared as described above, and stimulated with either 100 ng/ml IL-6 or IL-10 alone, or 100 ng/ml IL-6 or IL-10 15 min prior to stimulation with 100 ng LPS. Cells were processed according to Basic Protocol 2 and stained for anti-phospho-STAT3(S727)-Ax647 (clone 49) or anti-phospho-STAT3(Y705)-AX647 (clone 4/P-stat3) at 0.1 µg per 1 × 106 cells. Histograms displayed are gated on monocyte gate. Cells were cultured in 5% human AB sera (Irvine Scientific) in RPMI. This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www.interscience.wiley.com/c p/colorfigures.htm.
Immunophenotyping
6.20.5 Current Protocols in Cytometry
Supplement 32
1. Fix experimental (treated) cells and control cells directly in tissue culture plates by adding 100 µl of 16% paraformaldehyde to 1 ml of medium containing the cells (final concentration, 1.5%). Swirl plate and incubate 15 min at 37◦ C. Methods of activation vary and should be determined by each investigator. For adherent cells, fixing directly in the culture dish is preferred over trypsinization (or even PBS/EDTA lifting) followed by fixation; however, investigators are encouraged to compare the differences of their particular cell types in this respect.
2. Collect experimental (activated) and control cells in separate 12 × 75–mm polystyrene tubes, centrifuge 5 to 10 min at 250 × g, 4◦ C, and discard supernatant. Fixation of adherent cells in situ will require mild scraping with a pipet tip and generous pipetting up and down to dislodge them for harvesting into tubes. This can damage the cells if done too vigorously. Cell scraping with small spatulas can also disrupt cell integrity. It is also important to note that activation of adherent cells (e.g., by PMA stimulation) can induce strong adhesion in the cells. It is often suggested that the plates be checked under the microscope after dislodging the cells to ensure that cells have been harvested. This procedure will not work for adherent cells in matrix components (i.e., fibronectin, Matrigel, or collagen), and will not work for cells fixed onto tissues.
3. Add 1 ml ice-cold 90% methanol and incubate 30 min on ice to permeabilize the cells. At this stage, cells can be stored at −20◦ C for up to a month.
4. Centrifuge 5 to 10 min at 250 × g, 4◦ C. Wash cells twice, each time by adding 1 ml staining buffer 1, centrifuging again as before, and removing the supernatant. Having 1% FBS in the wash buffer helps in pelleting cells after methanol permeabilization. More than 5% FBS in the wash buffer may induce precipitation.
5. Resuspend cells in staining buffer 1 at 10 × 106 cells/ml and aliquot 100 µl per tube for staining. 6. Add fluorochrome-conjugated antibody to cell suspension at optimal concentration (as determined by preliminary experiments). Incubate 1 hr at room temperature in the dark. 7. Add 1 ml staining buffer 1 to each tube, then centrifuge 5 to 10 min at 250 × g, 4◦ C. Remove supernatant and resuspend cells in 200 µl staining buffer 1. 8. Using capture beads for setting proper compensation controls, collect fluorescence emission using flow cytometer with appropriate band-pass filter.
BASIC PROTOCOL 2
FLOW CYTOMETRIC ANALYSIS OF INTRACELLULAR PHOSPHOEPITOPES USING DETERGENT-BASED PERMEABILIZATION The following protocol enhances intracellular detection of phosphoepitopes using detergent-based permeabilization. Cells are stimulated, fixed, and permeabilized, and are then stained with single-color phosphoantibodies and washed extensively. The amount of phosphorylated antigen target is determined by flow cytometry.
Multiparameter Analysis of Intracellular Phosphoepitopes
This protocol is suited for single-color analysis. If additional stains are to be used, appropriate blocking agents are necessary. For example, if a primary mouse intracellular antibody is used and detected with an anti-mouse fluorochrome-conjugated antibody, it is necessary to block with mouse IgG (4 µg/ml in permeabilization/staining buffer) for
6.20.6 Supplement 32
Current Protocols in Cytometry
15 min prior to staining with directly conjugated mouse antibodies. This is also important if receptor cross-linking is used to stimulate cells, as secondary anti-mouse antibodies are often used to cross-link bound surface antibodies to initiate cellular activation. If working with mouse cells it is also important to preblock at the permeabilization stage with an Fc blocking reagent (e.g., FC-block, clone 2.42) for 15 min prior to staining. This may also apply to human cells expressing CD16/CD32 or CD64. In the case of human cells, nonspecific IgG is sufficient to block nonspecific staining. Use of a Tyramide Signal Amplification kit (Molecular Probes) can enhance separation of weak signals. The investigator is referred to the instructions from the manufacturer of this kit. If using an anti-mouse-biotin plus streptavidin-fluorophore conjugate to amplify intracellular signals, adequate controls are required. References for detection of intracellular phosphoepitopes using detergent-based permeabilizations and/or modifications to this protocol include Rosette et al. (2001); Zell et al. (2001); Perez et al. (2002, 2003); and Zell and Jenkins (2002).
Materials 1 × 106 cells/condition of stimulation in 1 ml medium 1 × 106 cells/unstimulated condition in 1 ml medium 1 × 106 cells as blank in 1 ml medium 16% paraformaldehyde, methanol-free (Electron Microscopy Sciences) Phosphate-buffered saline, pH 7.4 (PBS; APPENDIX 2A) Permeabilization/staining buffer (see recipe) Pretitered, fluorochrome-conjugated phosphospecific antibody or pretitered unconjugated phosphospecific antibody, anti-Ig biotinylated antibody, and fluorochrome-conjugated streptavidin PBS, pH 7.4 (APPENDIX 2A) containing 1 mM EDTA (store at room temperature) Antibody capture beads for compensation controls (UNIT 1.14) Refrigerated centrifuge 12 × 75–mm polystyrene tubes Platform shaker Flow cytometer 1. Fix experimental (treated) cells and control cells directly in tissue culture plates by adding 100 µl of 16% paraformaldehyde to 1 ml of medium containing cells (final concentration, 1.5%). Swirl plate and incubate 15 min at 37◦ C. For adherent cells, fixing directly in the culture dish is preferred over trypsinization (or even PBS/EDTA lifting) followed by fixation; however, investigators are encouraged to compare the differences of their particular cell types in this respect. Fixing with 1% to 2% paraformaldehyde solution (i.e., resuspending cell pellet in 100 to 500 µl) is also performed at 37◦ C. Fixing at 4◦ C induces shrinkage in cell size and augments forward and side scatter. When using commercial fixation/permeabilization kits, it is often best to fix completely, then permeabilize. Subsequent incubation of cells in permeabilization/stain buffer significantly enhances intracellular detection. Diethylene glycol–based permeabilization kits do not perform as well as saponin-based kits. However, it is recommended that such comparisons be made by the experimenter to determine the effects on the particular specificity of interest. Immunophenotyping
6.20.7 Current Protocols in Cytometry
Supplement 32
2. Collect experimental (activated) and control cells in separate 12 × 75–mm polystyrene tubes. Centrifuge 5 min at 500 × g, 4◦ C and remove the supernatant. Wash once by adding 2 ml PBS, centrifuging again as before, then inverting the tube over the sink (or aspirating) to discard the supernatant (cell pellet should be visible). Fixation of adherent cells in situ will require mild scraping with a pipet tip and generous pipetting up and down to dislodge cells for harvesting into tubes. This can damage the cells if done too vigorously. Cell scraping with small spatulas can also disrupt cell integrity. It is also important to note that activation of adherent cells (e.g., by PMA stimulation) can induce strong adhesion in the cells. It is often suggested that the plates be checked under the microscope after dislodging the cells to ensure that cells have been harvested. This procedure will not work for adherent cells in matrix components (i.e., fibronectin, Matrigel, or collagen), and will not work for cells fixed onto tissues. Unless adherent cells require surface antigen detection, adherent cell lines can be appropriately stained for phosphoepitopes using alcohol-based permeabilization (see Basic Protocol 1). Examples of human monocyte staining are presented in Figure 6.20.2C and 6.20.2D.
3. Permeabilize cells by washing twice with permeabilization/staining buffer, each time by adding 500 µl of that buffer, incubating 15 min at 4◦ C, then centrifuging 5 min at 500 × g, 4◦ C, and removing the supernatant. Perform this step even if Cytofix/Cytoperm solution (BD Biosciences) is used.
4. Thoroughly resuspend the fixed/permeabilized cells in 50 µl permeabilization/staining buffer containing an optimal concentration (as determined by preliminary experiments) of either anti-phosphoepitope antibody (fluorochrome-conjugated or unconjugated) or an appropriate negative control. Incubate 60 min at 25◦ C in the dark on a platform shaker. The permeabilization/staining buffer can serve as an antibody diluent and cell wash buffer. Because saponin-mediated cell permeabilization is a reversible process, it is important to keep the cells in the presence of saponin during intracellular phosphoepitope staining.
5. Wash cells twice with 500 µl permeabilization/staining buffer using the technique described in step 3. If using a conjugated antibody, resuspend cell pellet in PBS/1 mM EDTA and proceed to flow cytometric analysis (step 9). If using unconjugated primary antibodies, continue with steps 6, 7, and 8. 6. Resuspend cells in 50 µl permeabilization/staining buffer containing a predetermined optimal concentration of anti-Ig biotinylated antibody. Incubate 30 min at 25◦ C with shaking on a platform shaker. 7. Wash cells twice, each time with 500 µl permeabilization/staining buffer using the technique described in step 3. Resuspend cell pellet in 50 µl permeabilization/staining buffer containing a predetermined optimal concentration of fluorochromeconjugated streptavidin. Incubate 30 min at 25◦ C with shaking on a platform shaker. 8. Wash cells twice, each time with 500 µl permeabilization/staining buffer using the technique described in step 3. Resuspend cell pellet in 50 µl PBS/1 mM EDTA buffer 9. Using capture beads for setting proper compensation controls, collect fluorescence emission using flow cytometer with appropriate band-pass filter. Multiparameter Analysis of Intracellular Phosphoepitopes
6.20.8 Supplement 32
Current Protocols in Cytometry
FLOW CYTOMETRIC ANALYSIS OF INTRACELLULAR OR SURFACE PHOSPHOEPITOPES USING WHOLE-BLOOD STAINING
BASIC PROTOCOL 3
Rapid, optimized protocols for whole-blood/buffy-coat staining are needed with clinical samples. The following protocol has been designed to directly stain blood collected directly from patients or blood-derived buffy coats. It is recommended that the time involved in blood collection and that the time and temperature at which the blood sits before being processed be monitored, as differences in the collection procedure will affect the detected intracellular signaling. It is advisable not to use blood stored for longer than 4 hr, even if it is kept at 4◦ C.
Materials Human donor Activators needed for studying phosphoepitopes of interest 5× BD PhosFlow Whole Blood Lyse/Fix Buffer (BD Biosciences) Hanks’ balanced salt solution (HBSS; APPENDIX 2A) 70% (v/v) methanol, cold Staining buffer 2 (see recipe) Pretitered fluorescently conjugated phosphospecific antibody Antibody capture beads for compensation controls (UNIT 1.14) EDTA-treated Vacutainers or syringes Centrifuge 12 × 75–mm polystyrene tubes Flow cytometer 1. Collect normal human blood in the presence of EDTA and treat with appropriate activators if applicable. 2. Dilute 5× BD PhosFlow Whole Blood Lyse/Fix Buffer to 1× with distilled water and prewarm to 37◦ C. 3. Mix one volume of blood with 20 volumes of prewarmed 1× Whole Blood Lyse/Fix Buffer. Mix well by vigorously inverting the tubes 8 to 10 times, then incubate 10 min in a 37◦ C water bath. 4. Centrifuge 8 min at 500 × g, 4◦ C. Aspirate the supernatant, then add 1 ml Hanks’ balanced salt solution (HBSS) to the pellet, centrifuge again as before, and aspirate the supernatant. 5. Vortex to loosen the cells and permeablize cells by adding 1 ml cold 70% methanol slowly while vortexing. Incubate 30 min on ice and then centrifuge cells 5 min at 500 × g, 4◦ C. Remove supernatant. 6. Wash cells twice, each time by adding 1 ml HBSS, centrifuging 5 min at 250 × g, 4◦ C, and removing the supernatant. Finally, resuspend in staining buffer 2. 7. Add appropriate antibody conjugates at optimal concentration and incubate 30 to 60 min at 25◦ C. Wash cells using the technique described in step 6, then resuspend in HBSS. 8. Using capture beads for setting proper compensation controls, collect fluorescence emission using flow cytometer with appropriate band-pass filter. Examples of data generated using this protocol are presented in Figure 6.20.3.
Immunophenotyping
6.20.9 Current Protocols in Cytometry
Supplement 32
Figure 6.20.3 Whole blood staining for phosphoepitopes. Blood was stimulated with 40 ng/ml PMA, processed according to Basic Protocol 3, and stained for anti-phospho-p44/42-AX488 (clone 20a), anti-phospho-p38-Ax647 (clone 36), and anti-CD4-PE (clone RPA-T4) at optimal concentrations.This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www.interscience.wiley.com/c p/colorfigures.htm.
BASIC PROTOCOL 4
FLOW CYTOMETRIC ANALYSIS OF CELL-SURFACE AND INTRACELLULAR PHOSPHOEPITOPES USING ALCOHOL-BASED PERMEABILIZATION This protocol describes intracellular detection of phosphoepitopes and surface antigens. Cells are stimulated, fixed with methanol, and permeabilized. Cells are then stained with single-color phosphospecific antibodies and single-color surface antibodies and washed extensively. The amount of phosphorylated antigen in surface-defined subsets is determined by flow cytometry.
Materials
Multiparameter Analysis of Intracellular Phosphoepitopes
1 × 106 cells/condition of stimulation in 1 ml medium 1 × 106 cells/unstimulated condition in 1 ml medium 1 × 106 cells as blank in 1 ml medium Stimuli of interest 16% paraformaldehyde, methanol-free (Electron Microscopy Sciences) 90% (v/v) methanol, ice-cold Staining buffer 2 (see recipe)
6.20.10 Supplement 32
Current Protocols in Cytometry
Pretitered, fluorescently conjugated phosphospecific antibody or antibodies Antibody capture beads for compensation controls (UNIT 1.14) 6-, 12-, or 24-well tissue culture plates 12 × 75–mm polystyrene tubes Refrigerated centrifuge Flow cytometer 1. Plate 1 × 106 cells/ml in standard 6-, 12-, or 24-well tissue culture plates. It is suggested that cells be allowed to sit at least 1 hr at 37◦ C before proceeding to stimulations.
2. Stimulate cells with desired stimulus. Include controls in addition to unstimulated/stimulated samples (i.e., isotype control if available, or secondary stain alone if performing indirect stains). Incubate 15 min at 37◦ C. When staining PBMCs, Fc receptor–bearing cells often bind some antibody isotypes; therefore one must add blocking agents such as nonspecific mouse IgG or Ig corresponding to the isotype recognized by the Fc receptor. This is of particular importance with secondary staining techniques, where directly conjugated (Fab)2 fragments might be considered.
3. Fix experimental (treated) cells and control cells directly in tissue culture plates by adding 100 µl of 16% paraformaldehyde to 1 ml of medium containing cells (final concentration, 1.5%). Swirl plate and incubate 15 min at 37◦ C. 4. Transfer samples to 12 × 75–mm tubes, pipet up and down to ensure complete cell removal, and place samples on ice. Activated cells will tend to stick to the plastic, as will some others owing to the presence of the fixative. Pipetting up and down dislodges the majority of these cells. Check under a microscope to determine that cells have been completely removed, as significant cell loss can occur at this step.
5. Centrifuge cells 5 min at 500 × g, 4◦ C, and remove supernatant. Permeabilize cells by adding 1 ml ice-cold 90% methanol at a reasonable rate while vortexing the tube at medium speed. Let sit 30 min on ice. Fixed and permeabilized cells can be stored in methanol at −20◦ C or can be processed directly afterwards. Samples can be stored both short-term (several days to 2 to 3 weeks) or long term (>3 months).
6. Centrifuge cells 5 min at 500 × g, 4◦ C. Wash cells three times to ensure removal of methanol, each time by adding 2 to 4 ml PBS, centrifuging again as before, and removing the supernatant. Resuspend the cells in 500 µl staining buffer 2 and let sit 1 hr at 4◦ C. Washing with staining buffer should be avoided, as it may result in precipitatation of FBS proteins if methanol is still present. The extensive washing and rehydration step increases detection of many epitopes for human surface antigens. Murine surface epitopes do not seem to present as many difficulties. Experimenters can omit the rehydration step if desired.
7. Stain with antibody (diluted in staining buffer 2) 1 hr at 4◦ C on ice (covered to protect from light). Multicolor work requires staining with individual antibody-fluorophore conjugates to be used as compensation controls. In the antibody cocktails, 1/5 of the final volume of the antibody cocktail should be staining medium containing 4% FBS (i.e., staining buffer 2). If higher amounts of diluted antibodies are used, the final volume of the antibody cocktail can be increased to 100 µl with staining buffer 2. It is recommended that the final volume of blocking agent be 0.8% to 1% of total volume (this applies to BSA and FBS blocking).
Immunophenotyping
6.20.11 Current Protocols in Cytometry
Supplement 32
Figure 6.20.4 Tabulated data for the performance of human surface antibodies under the saponin-based or methanolbased protocols. Surface marker, clone, and fluorophore-conjugated antibody tested are tabulated, along with performance in the indicated protocol. The table depicts staining results obtained for various cell surface markers using conjugated and unconjugated antibodies under the conditions described. Individual results may differ from these depending on several factors, such as donor, sample quality, and experimental conditions. Legend: (+) indicates good separation; (+/−) indicates marginal separation, i.e., results may be condition- and/or donor-dependent; (−) indicates no separation observed; (blank) indicates “not tested”; asterisk (∗) indicates donor-dependent (peak does not separate well for some donors); double asterisk (∗∗) indicates that some donors show high background.
Multiparameter Analysis of Intracellular Phosphoepitopes
6.20.12 Supplement 32
Current Protocols in Cytometry
Figure 6.20.5 Tabulated data for the performance of mouse surface antibodies under the saponin-based or methanolbased protocols. Surface marker, clone, and fluorophore-conjugated antibody tested are tabulated, along with performance in the indicated protocol. The table depicts staining results obtained for various cell surface markers using conjugated antibodies under the conditions described. Legend: (+) indicates good separation; (+/−) indicates marginal separation, i.e., results may be condition- and/or donor-dependent; (−) indicates no separation observed; (blank) indicates not tested.
Immunophenotyping
6.20.13 Current Protocols in Cytometry
Supplement 32
Table 6.20.1 Typical Example of Antibody Cocktail Guidelinesa
Reagent
Vol. per sample
No. of samples
Total vol. required
Ab-FITC
5 µl
5
25 µl
Ab-PE
6 µl
5
30 µl
Ab-PerCP
10 µl
5
50 µl
Ab-APC
2 µl
5
10 µl
Staining medium
27 µl
5
135 µl
Total:
50 µl
250 µl
a Total volume = 50 µl/1 × 106 cells. For N samples, make up sufficient reagents for N + 1.
Antibodies used for either surface or intracellular staining can be used in cocktails once an optimal concentration has been determined. Titrations are critical for determining proper antibody cocktail composition. If titration is not indicated by the manufacturer of the reagent used, it needs to be determined by the investigator. A typical example of antibody cocktail guidelines is presented in Table 6.20.1. The final volume of 50 µl per sample is adjusted using staining medium. A minimum of 10 to 15 µl staining medium is needed to ensure that blocking agents are included in the stain. If diluted antibodies are used, it might be necessary to scale up to a 100-µl staining volume. With some epitopes, enhanced intracellular signals have been detected after 1 hr at 25◦ C, with constant shaking.
8. Centrifuge 5 min at 500 × g, 4◦ C. Wash cells three times, each time by adding 1 ml PBS, centrifuging again as before, and removing the supernatant. Resuspend cell pellet in 100 µl PBS/1 mM EDTA. 9. Using capture beads for setting proper compensation controls, collect fluorescence emissions using flow cytometer with appropriate band-pass filters. Examples of data generated by this protocol are presented in Figures 6.20.4 and 6.20.5. BASIC PROTOCOL 5
FLOW CYTOMETRIC ANALYSIS OF CELL-SURFACE AND INTRACELLULAR PHOSPHOEPITOPES USING SAPONIN-BASED PERMEABILIZATION Discrimination between bright and dim and intermediate/high populations
Multiparameter Analysis of Intracellular Phosphoepitopes
As noted with methanol permeabilization, there can be a loss of staining for certain surface epitopes, as well as a loss of distinctive levels of expression between populations expressing a marker at intermediate and high levels. The staining patterns of these populations tend to merge, as antigens are affected, and nonspecific background staining increases as well. In addition, the side scatter (SS) and forward scatter (FS) properties are not always maintained with methanol or saponin permeabilization in one-step staining procedures, and this is most apparent with paraformaldehyde fixation done at 37◦ C (where shape and size are maintained) versus fixation at 4◦ C (where shape and size are augmented). Since some peripheral blood cell populations are readily discernable by size (such as monocytes and lymphocytes) these changes can cause difficulties during analysis. With these considerations in mind, the authors have applied saponin-based techniques for fine cellular subset characterization, even at the expense of submaximal phosphoepitope detection. It is also important to note that variations in the grade of saponin that is commercially available can make a difference in phosphoepitope detection by this method. To overcome this, select a saponin with saponigen content of at least 25%, or alternatively use BD Perm Wash buffer (BD Biosciences).
6.20.14 Supplement 32
Current Protocols in Cytometry
Materials 1 × 106 cells/condition of stimulation in 1 ml medium 1 × 106 cells/unstimulated condition in 1 ml medium 1 × 106 cells as blank in 1 ml medium Stimuli of interest 16% paraformaldehyde, methanol free (Electron Microscopy Sciences) Permeabilization/staining buffer (see recipe) PBS, pH 7.4 (APPENDIX 2A) containing 1 mM EDTA (store at room temperature) Pretitered, fluorescently conjugated phosphospecific antibody or antibodies Antibody capture beads for compensation controls (UNIT 1.14) 6-, 12-, or 24-well tissue culture plates 12 × 75–mm tubes Refrigerated centrifuge Platform shaker Flow cytometer 1. Plate 1 × 106 cells/ml in standard 6-, 12-, or 24-well tissue culture plates. It is suggested that cells be allowed to sit at least 1 hr at 37◦ C before proceeding to stimulations.
2. Stimulate cells as desired. Include controls in addition to unstimulated/stimulated samples (i.e., isotype control if available, or secondary stain alone if performing indirect stains). Incubate 15 min at 37◦ C. When staining PBMCs, Fc receptor–bearing cells often bind some antibody isotopes; therefore one must add blocking agents such as nonspecific mouse IgG or Ig corresponding to the isotype recognized by the Fc receptor. This is of particular importance with secondary staining techniques, where directly conjugated F(ab)2 fragments might be considered.
3. Fix experimental (treated) cells and control cells directly in tissue culture plates by adding 100 µl of 16% paraformaldehyde to 1 ml medium containing cells (final concentration, 1.5%). Swirl plate and incubate 15 min at 37◦ C. 4. Transfer samples to 12 × 75–mm tubes, pipet up and down to ensure complete cell removal, and place samples on ice. Activated cells will tend to stick to the plastic, as will some others owing to the presence of the fixative. Pipetting up and down dislodges the majority of these cells. Check under a microscope to determine that cells have been completely removed, as significant cell loss can occur at this step.
5. Centrifuge cells 5 min at 500 × g, 4◦ C, and remove supernatant. Permeabilize cells by adding 200 µl permeabilization/staining buffer and incubating 15 min on ice. 6. Add 1 ml PBS/1 mM EDTA, then centrifuge cells 5 min at 500 × g, 4◦ C. 7. Stain cells with antibody cocktails (see Basic Protocol 4, step 7), except use the saponin-based permeabilization/staining buffer as the antibody diluent and incubate 1 hr at 25◦ C, protected from light with constant shaking on a platform shaker. Cells can also be stained 1 hr at 4◦ C, although enhancement of some phosphoepitopes is seen at 25◦ C (Perez et al., 2003). It is very important that the saponin-based buffer be used for the antibody cocktail, as saponin permeabilization is reversible and the antibodies for intracellular staining will not gain access to the appropriate intracellular compartments if staining medium without saponin is used.
Immunophenotyping
6.20.15 Current Protocols in Cytometry
Supplement 32
8. Centrifuge 5 min at 500 × g, 4◦ C. Wash cells three times, each time by adding ?? ml PBS, centrifuging again as before, and removing the supernatant. Resuspend cell pellet in 100 µl PBS/1 mM EDTA. 9. Using capture beads for setting proper compensation controls, collect fluorescence emissions using flow cytometer with appropriate band-pass filters. Examples of data generated by this protocol are presented in Figures 6.20.4 and 6.20.5. ALTERNATE PROTOCOL
TWO-STEP INTRACELLULAR AND SURFACE STAINING FOR FLOW CYTOMETRIC ANALYSIS OF PHOSPHOEPITOPES The development of flow cytometry instrumentation capable of 6-, 8-, 12-, and 16-color multiparametric measurements in conjunction with phosphoepitope detection methodologies can provide correlative biochemical analysis of many different cellular subsets. Exciting possibilities will emerge, because increasing accessibility of parameters for measuring kinase activation states as a function of cell-cycle progression, apoptosis, or cellular differentiation will contribute significantly to the understanding of cellular function. It is important to mention that every measurable parameter warrants careful evaluation of how it can be combined with intracellular phosphoepitope measurements. The authors’ previous work has combined annexin V staining with intracellular phosphoepitope staining in immunophenotyped cells (Perez et al., 2002) as well as cell-cycle progression and cytokine production (Perez et al., 2003). These combinatorial approaches require extensive optimization (Perez and Nolan, 2002). The two-step protocol that was used in these investigations is described below. It is recommended that controlled experiments be proved reproducible in the hands of the investigator before going on to process a complex set of unknowns. It is also recommended that the investigator be thoroughly familiar with the basic protocols described in this unit.
Additional Materials (also see Basic Protocol 4) Phospho wash buffer (see recipe) Extracellular staining buffer (see recipe) Pretitered fluorescently conjugated antibody or antibodies against extracellular phosphoepitopes Fixation buffer: phospho wash buffer (see recipe) containing 1% paraformaldehyde (store at 4◦ C) Permeabilization buffer: (see recipe) Pretitered fluorescently conjugated antibody or antibodies against intracellular phosphoepitopes Antibody capture beads for compensation controls 96-well round-bottom tissue culture plates 96-well round-bottom vinyl plates Refrigerated centrifuge with microtiter plate carrier 12 × 75–mm polystyrene tubes Flow cytometer 1. Dispense PBMCs (or purified cells) into 96-well round-bottom tissue culture plates at 0.5–1 × 106 cells per well in 100 µl medium. PBMCs are used up to 2 hr after blood draw. Adherent cells are removed by letting the purified cells sit in a plastic dish for 1.5 hr under the optimized culture conditions (37◦ C). After that time, the adherent cells (monocytes, dendritic cells, and macrophages) will have adhered to the plastic and the lymphocytes will be in suspension. Multiparameter Analysis of Intracellular Phosphoepitopes
It is suggested that cells be allowed to sit at least 1 hr before proceeding to stimulations.
6.20.16 Supplement 32
Current Protocols in Cytometry
2. Stimulate cells with desired stimulus. Include as controls: a. Single-color controls for all colors used (both positive and negative). b. Controls for phosphoproteins (i.e., stimulated versus nonstimulated). c. Unlabeled control for autofluorescence. d. Intracellular isotype controls for background staining. 3. Harvest cells by adding 100 µl phospho wash buffer to each well of the 96-well plate (transferring cells at this point is not recommended), and centrifuging 5 min at 500 × g, 4◦ C. Flick the plate to remove the supernatants and immediately resuspend cells in 50 µl ice-cold extracellular staining buffer per 1–2 × 106 cells. Transfer cells to 96-well round-bottom vinyl plates using a multichannel pipettor and place at 4◦ C. The authors use custom-designed metallic 96-well plate holders that maintain 4◦ C in every single well of a vinyl U-bottom plate (versus 12◦ to 15◦ C for a 96-well plate in an ice bucket). These holders are stored in a freezer (in a manner very similar to restrictionenzyme holding containers) until used.
4. Dilute fluorescently conjugated antibody or antibodies against cell surface phosphoepitopes appropriately (as determined by preliminary experiments) in extracellular staining buffer. Add this antibody cocktail at 50 µl per well, and incubate samples 15 min on ice in the dark. 5. Add 150 µl ice-cold phospho wash buffer per well and centrifuge 5 min at 500 × g, 4◦ C. Flick plate to remove supernatants, then wash once by adding 200 µl phospho wash buffer per well, centrifuging again as before, and removing the supernatants. 6. Fix by adding 100 µl ice-cold fixation buffer per well and incubating 30 min on ice, in the dark. Final concentration should be between 1% to 2% if paraformaldehyde is added directly from a stock solution.
7. Add 100 µl ice-cold phospho wash buffer per well and centrifuge 5 min at 500 × g, 4◦ C. Flick plate to remove supernatants, then wash once by adding 200 µl phospho wash buffer per well, centrifuging again as before, and removing the supernatants. 8. Permeabilize by adding 200 µl ice-cold permeabilization buffer per well and pipetting up and down four to five times. Incubate 15 min at 4◦ C in the dark. 9. Centrifuge 5 min at 500 × g, 4◦ C, flick plate to remove supernatants, add 100 µl ice-cold phospho wash buffer per well, centrifuge again as before, then flick plate to remove supernatants. 10. Dilute fluorescently conjugated antibody or antibodies against intracellular phosphoepitopes appropriately (as determined by preliminary experiments) in ice-cold permeabilization buffer. Resuspend cells in 50 µl of this antibody cocktail per well and incubate at least 30 min on ice in the dark. Longer incubations (1 hr) and incubations at room temperature can increase staining in some cases.
11. Add 150 µl ice-cold permeabilization buffer per well and centrifuge 5 min at 500 × g, 4◦ C. Flick plate to remove supernatants, then wash one to two times, each time by adding 200 µl permeabilization buffer per well, centrifuging again as before, and removing the supernatants. Two washes are usually sufficient, but more washes may decrease background.
12. Resuspend each pellet in 100 to 200 µl PBS/1 mM EDTA and transfer to a 12 × 75–mm tube.
Immunophenotyping
6.20.17 Current Protocols in Cytometry
Supplement 32
13. Using capture beads for setting proper compensation controls, collect fluorescence emissions using flow cytometer with appropriate band-pass filters. There are several variations on the surface and intracellular staining protocols. For examples of multiparameter staining using this protocol, see Perez et al. (2002, 2003).
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Extracellular staining buffer (Alternate Protocol) Phospho wash buffer (see recipe) containing: 4% (v/v) fetal bovine serum (FBS) 1 protease inhibitor cocktail tablet (Boehringer Mannheim) per 1 ml buffer Store up to 1 week at 4◦ C. Permeabilization buffer (Alternate Protocol) Phospho wash buffer (see recipe) containing: 0.2% (w/v) saponin (add from 10% stock solution; see recipe) 4% (v/v) fetal bovine serum (FBS) Store up to 1 week at 4◦ C Final saponin concentration for permeabilization should be no less than 0.1% per sample. A 0.2% solution is made to account for residual volume in wells left after washing. Saponin concentrations as high as 0.5% have been tested and determined to be suitable. Higher concentrations are not recommended. This is the same saponin-containing buffer that is used for intracellular staining in the Alternate Protocol.
Permeabilization/staining buffer (Basic Protocols 2 and 5) Phosphate-buffered saline, pH 7.4 (PBS; APPENDIX 2A) containing: 0.5% (w/v) saponin (add from 10% saponin stock solution; see recipe) 4% (w/v) fetal bovine serum (FBS) 1 mM sodium azide Store up to 1 month at 4◦ C Alternatively use 1× BD Perm Wash Buffer (prepared from 10× stock; BD Biosciences).
Phospho wash buffer (Alternate Protocol) Phosphate-buffered saline, pH 7.4 (PBS; APPENDIX 2A) containing: 1 mM β-glycerolphosphate 1 mM sodium orthovanadate 1 µg/ml microcystin (500 µg vials can be purchased from Calbiochem) 1 mM sodium azide Store up to 1 month at 4◦ C This is the base buffer for the extracellular staining buffer, fixation buffer, and permeabilization buffer formulations used in the Alternate Protocol of this unit.
Saponin stock solution, 10% Make a 10% saponin stock solution by mixing 10 g saponin (containing ≥25% saponingen content; Sigma) with 100 ml PBS (APPENDIX 2A). Keep at 37◦ C with mild stirring until saponin has dissolved (solution will be yellow in appearance). Filter-sterilize through a 0.22-µl filter and store up to 1 month at 4◦ C. Multiparameter Analysis of Intracellular Phosphoepitopes
Saponin is a glycoside derived from plants, e.g., Quillaja bark, or produced synthetically. Saponins are natural surfactants and comprise several different but related molecules, continued
6.20.18 Supplement 32
Current Protocols in Cytometry
triterpenoid structures that consist of aglycone (saponingen) structures with glycosyl moieties. The purity and chemical composition of commercial preparations will vary. The authors have observed that the final concentrations of the saponin-based buffers are best achieved using saponins of ≥ 25% (w/v) saponingen content.
Staining buffer 1 (Basic Protocol 1) Phosphate-buffered saline (PBS; APPENDIX 2A) containing: 1% (v/v) fetal bovine serum (FBS) 1 mM NaN3 Store up to 1 month at 4◦ C Staining buffer 2 (Basic Protocols 3 and 4) Phosphate-buffered saline (PBS; APPENDIX 2A) containing: 4% (v/v) fetal bovine serum (FBS) 1 mM sodium azide Store up to 1 month at 4◦ C Alternatively, use BD Pharmingen Staining Buffer (BD Biosciences).
COMMENTARY Background Information Multiparameter flow cytometric analysis of protein activation states in immunophenotyped cells can measure discrete biochemical events occurring at the single-cell level. This is suitable for primary cell investigations, as complex heterogeneous populations represent population-specific cell functions. The reader is directed to Perez et al. (2004), which discusses at length the complexities of multiparameter staining of both surface and intracellular antigens, with extensive discussion on the caveats with regard to protocol and antigen selection that go into designing the staining procedure. It is also recommended that, before attempting multicolor work, the experimenter be well familiarized with single-color staining for phosphoepitopes and trained in multicolor compensation. The single most important factor in advancing to multicolor staining is identification of the proper antibody clones that will work for the surface detection under the chosen protocol. The authors’ laboratory has profiled several hundred surface antibodies under the protocols described and has directly evaluated antibody performance. This testing takes several forms. These include (1) sequential staining evaluations, i.e., comparing the performance of surface antibodies for the ability to stain in paraformaldehyde, saponin, and methanol pre-fixation, post-fixation, prepermeabilization, and post-permeabilization; (2) evaluating the effect of paraformaldehyde, saponin, and methanol on fluorochrome intensity for a panel of protein and inorganic dyes; and (3) evaluating sequential staining steps as
described in point 1, above, for several different antibody clones directed against the same antigen. An example of sequential surface antigen testing is presented in Figure 6.20.6, panels A and B, for CD3 and CD4 human surface antibodies. It is necessary to perform these procedures for every single reagent desired in a multiparameter setup. Given the infrastructure and resources necessary to compile this information, its dissemination has been slow. The authors are in the process of making the information available in Web-based tutorials (http://proteomics.stanford.edu). Tabulated lists of surface antibodies and their performance as tested are presented in Figure 6.20.4 for human and Figure 6.20.5 for mouse. Flow cytometric measurement of phosphorylation states in immunophenotyped cells enables discrete biochemical signature analysis of particular subsets that exist in heterogeneous cell populations. Each application requires evaluation of the necessary protocol required, and custom applications always require staining optimization. These methodologies are applicable to primary cell isolates from mouse and human. They are not designed to work with solid tumors or tissues, as cell disruption from solids presents new challenges yet to be overcome. At present, single-cell suspensions, as well as cells that are adherent to tissue culture plastic, are best suited for biochemical analysis by phospo-epitope detection via flow cytometry. Exciting opportunities exist in studying the molecular details of clinical samples and disease models in mice. The generation of such information, either in multidimensional formats or by scaling to hundreds
Immunophenotyping
6.20.19 Current Protocols in Cytometry
Supplement 32
Figure 6.20.6 Sequential staining of peripheral blood with CD3 and CD4 human antibody clones pre- and post-fixation and pre- and post-permeabilization. Peripheral blood was treated under the following conditions: (1) surface stained; (2) surface stained and then fixed in 1% paraformaldehyde; (3) surface stained and then methanol permeabilized; (4) surface stained, fixed in 1% PFA, then permeabilized in 0.2% saponin; (5) surface stained, fixed in 1% PFA, then permeabilized in methanol; (6) fixed in 1% PFA, then surface stained; (7) fixed in 1% PFA, methanol permeabilized, and then surface stained; (8) fixed in 1% PFA, permeabilized with 0.2% saponin, and then surface stained. There were two washes between sequential steps. (A) Staining observations for CD3-PE (clone UCHT1), CD3-APC (clone Hit3a), and CD3-Cychrome (clone SK7). (B) Staining observations for CD4-FITC (clone RPA-T4), CD4-Cychrome (clone RPA-T4), CD4-PE (clone RPA-T4), CD4-PerCP-CY5.5 (clone SK3), and CD4-APC (clone RPA-T4).
of samples, will demand the development of analytical software to process the multivariate data sets in an automated manner. The development of bioinformatics tools in this arena will greatly expedite flow cytometric data processing.
Critical Parameters and Troubleshooting
Multiparameter Analysis of Intracellular Phosphoepitopes
For functional analysis of signaling molecules in immunophenotyped cells, several areas of the experiment need to be carefully considered: (1) identification of the proper combination of antibody cocktails for the stains; (2) identification of the sequential procedure for the staining using the antigens and antibody clones of interest; and (3) thorough evaluation of the effects of cell culture conditions on cell stimulation. For example, an experimental procedure should be evaluated on the basis of whether or not surface antigens are required to identify specific subsets in the sample. If the use of
surface markers is required, the performance of individual antibody clones for an antigen of choice needs to be assessed under the various conditions of the protocols described here. A majority of common human and mouse surface antibody performance is catalogued in Figures 6.20.4 and 6.20.5; however, clones or antigens not listed will require independent evaluation. Antibody clone selection for a given human cell surface antigen is of particular importance, as different antibody clones yield different results and do not stain similarly in different protocols. Comparison of what the surface staining looks like before and after the procedures is recommended, as a population can become distorted after permeabilizations. If surface antigens are not required, Basic Protocol 1 for intracellular staining is sufficient. Identifying the antibody clones needed and the suitable protocol for a population of interest may require a series of trials. This is particularly true for certain monocyte,
6.20.20 Supplement 32
Current Protocols in Cytometry
basophil, eosinophil, granulocyte, and antigenspecific lymphocyte populations whose size and scatter properties can be altered during the fixation and permeabilization procedures. This also applies to the intracellular reagents used. The use of multicolor labels requires primary conjugated antibodies at defined fluorophore-to-protein ratios. If selfconjugation is performed for small molecules (FITC, Texas Red, Alexa dyes, biotin), fluorescent proteins (EGFP, APC, PE, tandems), or other reagents (terbium, lanthanide series dyes, or quantum dots), a proper fluorochrome-to-protein ratio needs to be assessed. This is in addition to reagent titrations to identify optimal staining amounts in controlled experiments. Stimulating conditions using surface antibodies and cross-linking antibody reagents, e.g., CD3 and CD28 cross-linking for T cell activation, require thorough evaluation of the stimulatory agents and appropriate titrations of the secondary reagent used. If antibody reagents are to be used as stimulatory (or inhibitory) agents and an anti-mouse or antirabbit cross-linking reagent is used, it is advised that a blocking IgG step be implemented using at least 1 µg block per 0.5 µg of secondary (species-matching) cross-linking agent prior to adding directly conjugated antibodies. This will help eliminate excessive background that occurs in the absence of an IgG block if antibody cross-linking reagents are used for stimulatiory purposes. The authors’ laboratory uses several methods to evaluate intracellular specificity of each reagent. These procedures include (1) identifying appropriate biological conditions to induce or eliminate the phosphoepitope of interest; (2) using knockout cells or siRNA methods when available; (3) resorting to phosphopeptide competition or intracellular phosphatase treatment; or (4) using pharmacological inactivating agents of the phosphorylated residues under analysis and comparing the results with a stimulated control in which no pharmacological agent was presented.
Anticipated Results Every experiment performed using these procedures should employ a standard stimulation control. For example, stimulating most cells with PMA/ionomycin will activate phospho-erk and can serve as an internal positive control for the staining procedures. Results for cells will vary depending on the conditions used and the state of the cells. Implementing
stimulation controls will help in determining if the relative differences observed between stimulated and unstimulated cells are significant. It is also advised that investigators familiarize themselves with the various protocols and advance to the more complex staining procedures after becoming comfortable with the simpler ones.
Time Considerations The time required to complete the protocols described is not extensive; each can be accomplished in a standard work day. This, of course, will depend on the number of samples to be stained and the time required for cell preparation. The incubations for cell staining take the most time. A significant amount of time can also be spent on data acquisition at the cytometer and subsequent analysis. The buffers needed for the protocol should be prepared a day in advance to save time. Cells should either be prepared fresh (if primary) or plated out the night before at optimal densities. Staining cocktails can be prepared during the incubation periods for fixation and permeabilization. If time is critical, samples can be stored at −20◦ C after methanol permeabilization and kept on ice or at 4◦ C after antibody staining for several hours to overnight. It is advised that the results from such samples be compared with those from samples that have not been stored, as differences can result.
Literature Cited Chow, S., Patel, H., and Hedley, D.W. 2001. Measurement of MAP kinase activation by flow cytometry using phospho-specific antibodies to MEK and ERK: Potential for pharmacodynamic monitoring of signal transduction inhibitors. Cytometry 46:72-78. Fleisher, T.A., Dorman, S.E., Anderson, J.A., Vail, M., Brown, M.R., and Holland, S.M. 1999. Detection of intracellular phosphorylated STAT-1 by flow cytometry. Clin. Immunol. 90:425-430. Hilger, R.A., Kredke, S., Hedley, D., Moeller, J.G., Bauer, R.J., Stellberg, W., Seeber, S., Scheulen, M.E., and Strumberg, D. 2002. ERK1/2 phosphorylation: A biomarker analysis within a phase I study with the new Raf kinase inhibitor BAY43-9006. Int. J. Clin. Pharmacol. Ther. 40:567-568. Kaech, S.M., Hemby, S., Kersh, E., and Ahmed, R. 2002. Molecular and functional profiling of memory CD8 T cell differentiation. Cell 111:837-851. Krutzik, P.O. and Nolan, G.P. 2003. Intracellular phospho-protein staining techniques for flow cytometry: Monitoring single cell signaling events. Cytometry 55A:61-70.
Immunophenotyping
6.20.21 Current Protocols in Cytometry
Supplement 32
Krutzik, P.O., Irish, J.M., Nolan, G.P., and Perez, O.D. 2004. Analysis of protein phosphorylation and cellular signaling events by flow cytometry: Techniques and clinical applications. Clin. Immunol. 110:206-21. Perez, O.D. and Nolan, G.P. 2002. Simultaneous measurement of multiple active kinase states using polychromatic flow cytometry. Nat. Biotechnol. 20:155-162. Perez, O.D., Kinoshita, S., Hitoshi, Y., Payan, D.G., Kitamura, T., Nolan, G.P., and Lorens, J.B. 2002. Activation of the PKB/AKT Pathway by ICAM2. Immunity 16:51-65. Perez, O.D., Mitchell, D., Jager, G.C., South, S., Murriel, C., McBride, J., Herzenberg, L.A., Kinoshita, S., and Nolan, G.P. 2003. Leukocyte functional antigen 1 lowers T cell activation thresholds and signaling through cytohesin-1 and Jun-activating binding protein 1. Nat. Immunol. 4:1083-1092. Perez, O.D., Krutzik, P.O., and Nolan, G.P. 2004. Flow cytometric analysis of kinase signaling cascades. Methods Mol. Biol. 263:67-94. Rosette, C., Werlen, G., Daniels, M.A., Holman, P.O., Alam, S.M., Travers, P.J., Gascoigne, N.R., Palmer, E., and Jameson, S.C. 2001. The impact
of duration versus extent of TCR occupancy on T cell activation: A revision of the kinetic proofreading model. Immunity 15:59-70. Uzel, G., Frucht, D.M., Fleisher, T.A., and Holland, S.M. 2001. Detection of intracellular phosphorylated STAT-4 by flow cytometry. Clin. Immunol. 100:270-276. Zell, T. and Jenkins, M.K. 2002. Flow cytometric analysis of T cell receptor signal transduction. Science STKE 2002:PL5. Zell, T., Khoruts, A., Ingulli, E., Bonnevier, J.L., Mueller, D.L., and Jenkins, M.K. 2001. Singlecell analysis of signal transduction in CD4 T cells stimulated by antigen in vivo. Proc. Natl. Acad. Sci. U.S.A 98:10805-10810.
Contributed by Omar D. Perez, Dennis Mitchell, and Garry P. Nolan Stanford University School of Medicine Stanford, California Roberto Campos, Guo-Jian Gao, and Li Li BD Biosciences Pharmingen San Diego, California
Multiparameter Analysis of Intracellular Phosphoepitopes
6.20.22 Supplement 32
Current Protocols in Cytometry
CHAPTER 7 Nucleic Acid Analysis INTRODUCTION his chapter presents the most common methods for measurement of cellular DNA and RNA content. These methods utilize a variety of fluorochromes whose spectral properties and binding characteristics are described in UNIT 4.4. Knowledge of these properties is essential for development of optimal staining conditions and selection of excitation wavelength and emission filters, as well as proper interpretation of the data. Users are strongly urged to read the relevant material on a particular fluorochrome before trying to follow the protocol in which it appears.
T
General applicability of nucleic acid analytical methods, in both basic and clinical research, is discussed in UNIT 7.1. Details of specific applications of these methods, such as studies of cell cycle and cell differentiation and determination of tumor prognosis, can be found elsewhere (Melamed et al., 1990; Bauer et al., 1993; Darzynkiewicz et al., 1994). Because flow cytometric analysis of cellular DNA content is already an established procedure for measurements of DNA ploidy of tumors and of the fraction of cells in the S phase of the cell cycle—two parameters that have prognostic utility in clinical oncology—careful quality control of this type of analysis is essential. This topic is covered in UNIT 7.2. Differential, correlated staining of DNA and RNA is discussed in UNIT 7.3, which presents one protocol that employs the metachromatic dye acridine orange (AO) and a second that uses a combination of Hoechst 33342 and pyronin Y (PY). Each method has advantages and limitations; in particular, the Hoechst 33342–PY method requires excitation at two different wavelengths, including UV light, and thus is not applicable to single-laser instruments. Also described in UNIT 7.3 is supravital cell staining with Hoechst 33348 and PY, which may be applicable for detection and sorting of live stem/progenitor cells. Still another RNA fluorochrome, thiazole orange (TO), is used to detect RNA in immature platelets. The protocol describing analysis of reticulated platelets utilizing TO alone, or in combination with antibodies reactive with platelet-specific glycoproteins such as GPIIb/IIa or GPIb, is presented in UNIT 7.10. For a comprehensive view of RNA metabolism, the measurement of cellular RNA content can be complemented by analysis of a rate of RNA synthesis. The latter, by providing information on the overall transcriptional activity of the cell, may be useful in studies of mitogenic cell activation and in other instances when the rate of transcription is of interest. RNA synthesis is analyzed after in vitro or in vivo incorporation of the halogenated RNA precursor (BrU) followed by its immunocytochemical detection. The method presented in UNIT 7.12 describes the bivariate analysis of the rate of RNA synthesis versus DNA content, which makes it possible to correlate the transcriptional activity with the cell cycle position. Simultaneous measurement of DNA content and detection of DNA strand breaks is described in UNIT 7.4. This method is used to identify apoptotic cells, because DNA breakage is a characteristic feature of apoptosis. Correlated analysis of DNA content and DNA strand breaks allows one not only to identify apoptotic cells but also to pinpoint their position in the cell cycle. By revealing the cell cycle distribution of both nonapoptotic and apoptotic cell populations, this method is uniquely designed to analyze the cell cycle Nucleic Acid Analysis Contributed by Zbigniew Darzynkiewicz Current Protocols in Cytometry (2004) 7.0.1-7.0.6 C 2004 by John Wiley & Sons, Inc. Copyright
7.0.1 Supplement 30
specificity of apoptosis. The protocol presented in UNIT 7.27 describes the detection of another type of DNA strand break, namely the double-stand (ds) breaks such as those induced by ionizing radiation or DNA-damaging drugs. Induction of these breaks leads to phosphorylation of histone H2AX. Immunocytochemical detection of the phosphorylated histone combined with differential staining of DNA, described in this unit, reveals the presence of ds DNA breaks in relation to the cell cycle phase. The most common methods of DNA content analysis are presented in UNIT 7.5. Staining of fixed cells is generally preferred when samples must be stored or transported, as fixation preserves cells for later analysis and permits sample storage for an indefinite period of time. Among the disadvantages of fixation is cell loss (fixed cells have a propensity to adhere to tubing) and somewhat lowered accuracy of DNA content analysis reflected by the greater coefficient of variation for the mean DNA content of the G1 cell population compared to that observed for detergent-treated cells. A second general approach is staining of DNA in cells permeabilized by detergents and subjected to controlled proteolysis. Analysis of DNA content of cells stained according to this protocol generally provides the most accurate results (lowest CV for the mean DNA content of the G1 cell population) and appears to be preferred by clinicians for analysis of tumor DNA content. A disadvantage is that the technique is cumbersome; an alternate protocol presents a simplified procedure that can, however, be used only with uniform cell populations. Staining of DNA in live cells (supravital staining) is a third approach that is generally attempted when one desires to sort and possibly clone live cells based on differences in DNA content. The method presented in UNIT 7.5 is based on the use of Hoechst 33342, a dye that requires UV light excitation. It should be noted, however, that there is an alternative procedure to stain DNA in live cells, which utilizes DRAQ5, a fluorochrome excited with red light (UNIT 7.25). Although the live cell staining procedures are simple, some cell types cannot be stained supravitally. This is generally the case with cells having a very active P-glycoprotein efflux mechanism (e.g., multidrug-resistant or stem cells), which rapidly removes the fluorochrome from the cell interior. A fourth approach, DNA content analysis of archival paraffin-embedded samples, is the most difficult and is subject to potential artifacts. The major problem is lack of adequate controls for DNA ploidy. Because of possible differences in the fixation rate, any external control may not be adequate. On the other hand, cells with different chromatin structure show differences in DNA stainability following fixation with formaldehyde, making it difficult to recognize the presumed DNA diploid normal host (e.g., stromal or infiltrating) cells in the sample, as a marker of DNA ploidy. The fifth approach presented, detection of apoptotic cells characterized by fractional DNA content, is based on fixation of cells in ethanol followed by controlled extraction of the fragmented low-molecular-weight DNA from apoptotic cells prior to staining. The latter are then represented as cells having DNA content significantly lower than that of G1 cells and displayed in the DNA content frequency histograms as so-called “sub-G1 peaks.” Finally, a support protocol describes agarose gel electrophoresis of the degraded DNA selectively extracted from the very same apoptotic cells that can be identified by flow cytometry. Assays of DNA content in plant cells for ploidy determination and cell cycle analysis are becoming routine in many laboratories and have found numerous applications in agricultural research and practice. These assays differ significantly in many details, especially with regard to cell or nuclei preparation, from the assays applicable to the analysis of animal cells that are presented in UNIT 7.5. UNIT 7.6, therefore, is devoted specifically to DNA analysis of material obtained from plant tissues.
Introduction
Still another protocol of DNA content analysis, designed for budding yeast, is presented in UNIT 7.23. This protocol utilizes SYTOX fluorochrome and in some instances is expected to provide more accurate DNA content determination (lower coefficient of variation of
7.0.2 Supplement 30
Current Protocols in Cytometry
G1 cell population) compared to the method described in UNIT 11.10, which is based on the use of P1. Analysis of cellular DNA content also has application in tests of mutagenicity. One of the widely used mutagenicity assays, described in UNIT 7.21, is based on quantitation of micronuclei in peripheral blood erythrocytes. This assay can be used to monitor in vivo cumulative exposure to carcinogens/mutagens. Cytometric analysis of BrdU incorporation (Gratzner, 1982) to detect DNA replication has replaced cumbersome and ecologically hazardous methods utilizing radioactive DNA precursors. Further development of this method to allow bivariate analysis of DNA content and BrdU incorporation (Dolbeare et al., 1983) revolutionized our ability to study cell proliferation and the cell cycle. This method, presented in UNIT 7.7, is currently one of the most widely used assays in cell biology; it has also found a clinical application in analysis of the kinetics of cell proliferation in tumors. The original methods of immunocytochemical detection of BrdU incorporation (UNIT 7.7) require partial DNA denaturation to make the incorporated precursor accessible to antibody. The harsh conditions of DNA denaturation by acid or heat often are incompatible with simultaneous immunocytochemical detection of other cell constituents. However, a more recent protocol based on partial DNA photolysis by UV light combined with hypotonic conditions of cell exposure to the anti-BrdU (UNIT 7.15) can be used for simultaneous detection of BrdU incorporation and other intracellular antigens. The alternative to immunocytochemical, the cytochemical method of detection of BrdU incorporation, based on quenching of Hoechst 33258 or Hoechst 33342 fluorescence, is described in UNIT 7.14. Incorporation of BrdU into DNA results in significant suppression of fluorescence when Hoechst dyes are used as the DNA fluorochromes. On the other hand, fluorescence of such dyes as ethidium bromide (EB) or 7-AAD is unaffected by the incorporated BrdU. Thus, when BrdU-labeled cells are stained with both Hoechst 33328 (or Hoechst 33342) and EB (or 7-AAD), quenching of Hoechst dye fluorescence (UV excitation, blue emission) reveals BrdU incorporation, while the intensity of EB (blue excitation, red emission) or 7-AAD (far red emission) fluorescence discloses the cell cycle position. Compared to immunocytochemical BrdU detection (UNIT 7.7) the cytochemical technique (UNIT 7.14) is less sensitive, and therefore longer incubation times with BrdU are needed for adequate labeling of the BrdU-incorporating cells. It should be noted that prolonged incubation with BrdU carries the risk of cell cycle perturbation by the incorporated precursor. However, by virtue of its simplicity and low cost (no need for anti-BrdU Ab), the cytochemical approach described in UNIT 7.14 may be preferred in many instances, particularly to distinguish cohorts of cells that either do not proliferate or have different rates of proliferation in mixed cell populations. The protocols presented in UNIT 7.14 allow one to carry out the analysis of BrdU incorporation vs. cellular DNA content in cell subpopulations selected based on their green fluorescent protein (GFP), which is of utility to study differences in cell proliferation between nontransfected cells and those transfected with particular vectors carrying the GFP marker. Green fluorescent protein (GFP) is now widely used as reporter of successful transfections and new gene activity. It is frequently desired, therefore, to combine the detection of GFP with analysis of DNA content, for example, to correlate the transfection with the cell cycle position or DNA ploidy, or to sort the transfected cells. Procedures for simultaneous DNA content analysis and GFP detection are presented in UNIT 7.16. Not only the quantity but also the conformation of nucleic acids can be analyzed by flow cytometry. This can be accomplished using AO, which under appropriate conditions (AO concentration, pH, ionic strength) can differentially stain double-stranded (ds) versus
Nucleic Acid Analysis
7.0.3 Current Protocols in Cytometry
Supplement 30
single-stranded (ss) nucleic acids. Application of AO for differential staining of DNA and RNA is described in UNIT 7.3, and its application for the study of in situ DNA denaturation in UNIT 7.8. Because DNA in chromatin of mitotic and apoptotic cells is more sensitive to denaturation than is DNA in interphase chromatin of live cells, the measurement of DNA denaturation (single-strandedness) following cell exposure to heat or acid makes possible the discrimination of mitotic and/or apoptotic cells. The acridine orange-based assay of DNA sensitivity to denaturation is also helpful in analysis of the changes in sperm-cell chromatin structure that are associated with male infertility (UNIT 7.13). Intriguingly, the chromatin abnormality of sperm cells detected by this assay shows strong resemblance to the changes in chromatin structure associated with apoptosis of somatic cells: DNA hyperchromicity, increased DNA denaturability, and extensive DNA fragmentation (Gorczyca et al., 1993). The assay described in UNIT 7.13 is now widely used in both veterinary and human fertility clinics. Analysis of the cell cycle is invariably associated with DNA content measurement. Univariate analysis of DNA content (UNITS 7.5 & 7.6) is adequate to position cells in their respective phases of the cell cycle, G1 , S, and G2 /M. By providing additional information on the cell cycle, bivariate analyses—combining measurement of either DNA and RNA content (UNIT 7.3), DNA content and BrdU incorporation (UNIT 7.7), or respective proportions of ss and dsDNA to reveal in situ DNA sensitivity to denaturation (UNIT 7.8)— complement univariate DNA content estimates. Thus, the RNA parameter discriminates cells that have the same DNA content (e.g., equivalent of G1 cells) but are functionally different, such as cycling from noncycling or differentiated from undifferentiated cells (UNIT 7.3). Detection of BrdU incorporation, which substitutes for the radioactive thymidine incorporation assay, directly reveals DNA replication and provides a marker that allows one to measure the kinetics of cell progression through the cell cycle (UNIT 7.7). Differences in DNA sensitivity to denaturation, that reflect differences in degree of chromatin condensation, provide a marker to distinguish between G2 and M cells, and to identify quiescent cells with condensed chromatin, or apoptotic cells (UNIT 7.8). The alternative method that allows one to distinguish between G2 and M cells is based on immunocytochemical detection of histone H3 phosphorylated on Ser 10 (UNIT 7.24). Concurrent differential staining of cellular DNA in this method allows one to identify cells in all four major phases of the cell cycle, G0 /G1 versus S, versus G2 , versus M. It is often desirable to reveal cell cycle distribution in a particular cell subpopulation within a sample that contains a mixture of subpopulations. A method to identify cell subpopulations based on differences in surface immunophenotype and probe their DNA content is described in UNIT 7.11. Because the presence of dead or dying cells in the sample often interferes with their immunophenotyping, this method allows one to identify such cells and exclude them from analysis. Analysis of fluorescence (excited with blue light) at three different emission wavelengths, however, is required for simultaneous analysis of cell viability, immunofluorescence, and DNA content.
Introduction
Cyclins are the key components of the cell cycle progression machinery, and their analysis provides an insight into molecular mechanisms of this machinery. The bivariate analysis of DNA content versus expression of cyclin D, E, A , or B1 is described in UNIT 7.9. Because these cyclins are expressed transiently during the unperturbed growth of normal cells, at very specific time points in the cell cycle, their presence in the cell provides a marker of a particular portion (compartment) of the cell cycle. Based on bivariate analysis of cellular DNA content and expression of these cyclins, therefore, the cell cycle can be subdivided into at least eight distinct subcompartments. This subdivision allows one to establish with a greater accuracy the points of cell arrest in the cycle caused by various antitumor drugs.
7.0.4 Supplement 30
Current Protocols in Cytometry
Infection of cells with viruses induces a variety of intracellular changes reflecting the engagement of host cell replication, transcription, and translation for synthesis of viral progeny. Because viral proteins often mimic components of cellular regulatory machinery, studies on viral infection have been particularly fruitful in providing insight on regulation of the cell cycle, and more recently also on apoptosis. UNIT 7.17 describes protocols for detection of the DNA polyoma virus SV40 in infected cells as well as of some viral proteins that play a role in regulation of the cell cycle progression. These protocols can be used for the detection of products of other viruses for which antibodies are available. It has become apparent in recent years that cell death is often an active and highly choreographed process of cell suicide, termed apoptosis. During apoptosis, the cell engages a plethora of different pathways that either promote its survival or lead to its demise. A constant crosstalk between these pathways and those that control cell proliferation and other cell activities takes place. This crosstalk is necessary to maintain tissue homeotosis, the mechanism that assures that the cell birth rate is in balance with the death rate. Disruption of this balance is the cause of many diseases, including, of course, cancer. These pathways, the subject of intense research in recents years, are described in UNIT 7.18. Complementing UNIT 7.18, UNIT 7.19 describes a variety of different methods to identify apoptotic cells and distinguish the apoptotic mode of cell death from cell necrosis. Although some of these methods do not involve analysis of nucleic acids, they are grouped all together in a single unit for convenience, because strategies in analysis of apoptosis, possible difficulties, and pitfalls are common to most if not all the methods. Furthermore, the presentation in a single unit of most of the methods designed to identify and estimate frequency of apoptotic cells may help the reader to choose the one that optimally suits his/her cell type or reveals a particular apoptotic event of interest. The laser scanning cytometer (LSC) is a microscope-based cytometer that in many applications complements flow cytometry (see review in Darzynkiewicz et al., 1999). UNIT 7.20 presents protocols for analysis of fine-needle aspirate biopsies (FNABs), an application for which LSC is more suitable than flow cytometry. Samples for this slide-based technology can be obtained by minimally invasive approaches and information on ploidy, immunophenotype, and morphology from analysis of minimal sample volumes such as hypocellular FNABs. Another advantage is the possibility of visual examination or image analysis of the measured cells upon their relocation by LSC. Another protocol of concurrent DNA content and immunofluorescence analysis of solid tumors adapted to LSC is presented in UNIT 7.22. Unlike UNIT 7.20, this protocol relies on analysis of tissue imprints rather than on FNAB. Each of these protocols can also be applied to analyze freshly prepared cytological specimens as well as cells isolated from tumors and deposited on microscope slides by cytospinning. It should be stressed here that none of the methods presented in this chapter is quantitative in analytical terms, and the claim cannot be made that all cellular DNA or RNA is stained or that the fluorescence intensity is in stoichiometric relationship to total cellular DNA or RNA content. In the case of DNA, only a fraction of nuclear DNA is accessible to each of the fluorochromes, and the size of this fraction varies depending on fluorochrome structure (size and binding mode), chromatin structure, and method of analysis (Darzynkiewicz et al., 1984). Nevertheless, the size of the stainable fraction is often proportional to total content of cellular DNA, which makes these methods semiquantitative. Similarly, quantitation of RNA is also biased by conformational differences (single vs. double strandedness) and by the possibility of extraction of low-molecular-weight RNA (e.g., tRNA) from ethanol-fixed cells. Nucleic Acid Analysis
7.0.5 Current Protocols in Cytometry
Supplement 30
LITERATURE CITED Bauer, K.D., Duque, R.E., and Shankey, V.C. (eds.) 1993. Clinical Flow Cytometry. Williams and Wilkins, Baltimore. Darzynkiewicz, Z., Robinson, J.P., and Crissman, H. (eds.) 1994. Flow Cytometry, 2nd ed., Vols. I and II. Academic Press, San Diego. Darzynkiewicz, Z., Traganos, F., Kapuscinski, J., Staiano-Coico, L., and Melamed, M.R. 1984. Accessibility of DNA in situ to various fluorochromes: Relationship to chromatin changes during erythroid differentiation of Friend leukemia cells. Cytometry 5:355-363. Darzynkiewicz, Z., Bedner, E., Li, X., Gorczyca, W., and Melamed, M.R. 1999. Laser-scanning cytometry: A new instrumentation with many applications. Exp. Cell Res. 249:1-12. Dolbeare, F., Gratzner, H., Pallavicini, M., and Gray, J.W. 1983. Flow cytometric measurement of total DNA content and incorporated bromodeoxyuridine. Proc. Natl. Acad. Sci. U.S.A. 80:5573-5577. Gorczyca, W., Traganos, F., Jesionowska, H., and Darzynkiewicz, Z. 1993. Presence of DNA strand breaks and increased sensitivity of DNA in situ to denaturation in abnormal human sperm cells: Analogy to apoptosis of somatic cells. Exp. Cell Res. 207:202-205. Gratzner, H. 1982. Monoclonal antibody against 5-bromo- and 5-iodo-deoxyuridine: A new reagent for detection of DNA replication. Science 218:474-475. Melamed, M.R., Lindmo, T., and Mendelsohn, M.L. (eds.) 1990. Flow Cytometry and Sorting. Wiley-Liss. New York.
Zbigniew Darzynkiewicz
Introduction
7.0.6 Supplement 30
Current Protocols in Cytometry
Overview of Nucleic Acid Analysis A variety of fluorochromes can be used for flow cytometric analysis of nucleic acids. Their spectral properties, chemical structures, and binding characteristics are presented in detail in UNIT 4.2. This overview, therefore, is limited to general introduction of the methods presented in UNITS 7.2-7.5, and primarily focuses on the applicability of these methods in different fields of biology and medicine. Analysis of nucleic acids is, perhaps after immunophenotyping (UNITS 6.2 & 10.4), the most common application of flow cytometry. Quantitative analysis of cellular DNA content provides information about DNA ploidy and cell cycle distribution. Cellular RNA content, on the other hand, characterizes cell phenotypes associated with differentiation, quiescence, and proliferation. Because all these cell attributes are associated with tumor malignancy and thus are of importance in tumor diagnosis and prognosis, the methodologies of nucleic acid analysis are of particular interest for researchers in the fields of basic and clinical oncology. They also are widely used by researchers in various areas of cell and molecular biology. During the early years of flow cytometry the primary focus of nucleic acid analysis was on DNA ploidy estimates. Because many malignancies (depending on the tumor type) were found to be aneuploid, cellular DNA content provided a diagnostic marker for such tumors. Progress in instrument and methods development was aimed toward highly accurate measurement of DNA content to allow identification of clones of aneuploid cells having relatively little deviation from diploid cells. The term “DNA ploidy index” (DI), paralleling cell ploidy index estimated by cell karyotyping, was introduced to define a variation in DNA value between clones of tumor cells and the diploid standard (Barlogie et al., 1983; Hiddeman et al., 1984). Quality control tests and standards were proposed to make DNA ploidy index a reproducible measure from one laboratory to another (see UNIT 7.2 for discussion of quality control in nucleic acid studies). It soon became apparent that DI is not only of diagnostic but also of prognostic value in some malignacies. A consensus is developing to use this marker as prognostic tool in clinical oncology (Bauer et al., 1993; Duque et al., 1993; Hedley et al., 1993; Shankey et al., 1993a,b; Wheeless et al., 1993). It also became
apparent that analysis of the cell cycle provides an additional prognostic parameter, which in many instances is of greater significance than DI. Thus, analysis of the S phase fraction, or so-called proliferative fraction, defined by many authors as the fraction of cells in (S+G2)/M, has also become an established clinical assay (Hall et al., 1992). Discrimination of S or (S+G2)/M phase cells on the basis of single-parameter DNA content analysis, however, has been plagued by technical problems. One is the inability, in the case of diploid tumors, to discriminate between the tumor cells and infiltrating or stromal normal host cells. For aneuploid tumors data analysis software has been developed to deconvolute DNA histograms of diploid and aneuploid populations. Immunocytochemical detection of cytokeratins, as a second parameter in addition to DNA, made it possible to discriminate the infiltrating cells in cancers of epithelial cell lineage (Cooper et al., 1985). Another problem with univariate DNA content analysis is the inability to distinguish whether cells with a DNA content equivalent to that of S phase cells indeed replicate DNA, information of importance in light of evidence of kinetically inactive S phase cells in tumors (Darzynkiewicz et al., 1980). Despite these technical problems, the single-parameter cellular DNA content measurement remains one of the most common assays in flow cytometry. This chapter covers the methods most commonly used in analysis of nucleic acids. In UNIT 7.5, DNA analytical methods are presented that are adapted not only to DNA staining in fixed cells and in cells permeabilized by detergents, but also to supravital cell staining, staining of samples of cells isolated from archival material embedded in paraffin blocks, and detection of apoptotic cells characterized by fractional DNA content. A support protocol in UNIT 7.5 describes agarose gel electrophoresis of the degraded DNA, which is selectively extracted from the same apoptotic cells that can be identified by flow cytometry (Gong et al., 1994). In addition to DNA content, DNA conformation (degree of single versus double strandedness) can also be analyzed by flow cytometry (Darzynkiewicz et al., 1975). Analysis of DNA conformation is done to assay in situ DNA sensitivity to denaturation. This assay is used to discriminate cells differing in chromatin
Contributed by Zbigniew Darzynkiewicz and Gloria Juan Current Protocols in Cytometry (1997) 7.1.1-7.1.4 Copyright © 1997 by John Wiley & Sons, Inc.
UNIT 7.1
Nucleic Acid Analysis
7.1.1
structure (such as mitotic versus interphase cells), to distinguish apoptotic cells, and to identify abnormal, infertile sperm cells with damaged chromatin (Evenson et al., 1980). Differential staining of the nondenatured (double-stranded) and denatured (singlestranded) DNA sections is achieved with use of acridine orange (AO), a metachromatic fluorochrome (Darzynkiewicz et al., 1975; see also UNIT 7.3). Still another type of analysis of DNA by flow cytometry involves detection of DNA strand breaks. Excessive DNA breakage occurs during apoptosis, and detection of DNA breaks is widely used to identify apoptotic cells (Gorczyca et al., 1992; Li and Darzynkiewicz, 1995). The method is based on use of fluorescently tagged triphosphodeoxynucleotides that are incorporated into 3′-OH termini in DNA breaks by exogenous terminal deoxynucleotidyltransferase (TdT; UNIT 7.4). Analysis of RNA content allows one to recognize cell phenotypes associated with cell withdrawal from the cycle (G0 versus G1 status) or differentiation. Because quiescent cells contain on average 5- to 10-fold fewer ribosomes than their cycling counterparts, and rRNA is the major component (>90%) of all cellular RNA, detection of G0 cells can easily be accomplished by analysis of cellular RNA content (Darzynkiewicz et al., 1976). Unfortunately, in most tumors the cells do not enter the quiescent G0 state, and therefore their RNA content is not as definite a marker of their cycling status as it is in the case of nonmalignant cells. Yet, cellular RNA content has been shown to be a significant prognostic marker in many malignancies of both hematopoietic and solid tissues (Darzynkiewicz, 1988). In addition to being of interest in oncology, the analysis of RNA content allows one to identify reticulocytes and undifferentiated platelets. Compared to DNA, however, there are relatively few fluorochromes with practical application for analysis of RNA content. Methods based on use of the metachromatic property of AO (Darzynkiewicz et al., 1976) and a combination of Hoechst 33342 and Pyronin Y (Shapiro, 1981) are presented in UNIT 7.3.
Overview of Nucleic Acid Analysis
Introduction of 5-bromodeoxyuridine (BrdU) as a marker of DNA replication allows investigators to identify cells that actually do progress through S phase. Three different methodologies for BrdU detection have been developed. One is based on fluorescence quenching of DNA fluorochromes such as Hoechst 33258 (Latt, 1973; Crissman and Steinkamp, 1987;
Poot et al., 1990) and AO (Darzynkiewicz et al., 1978) by BrdU incorporated into DNA. Following BrdU incorporation the cells are stained with propidium iodide (PI) and Hoechst 33258: the PI stains DNA stoichiometrically, thereby allowing for positioning of cells in the major phases of the cycle, while DNA-replicating cells can be discriminated by their disproportionately lower Hoechst 33258 stainability. The methods based on fluorochrome quenching are relatively simple and inexpensive but in general lack the sensitivity needed for detection of BrdU pulse labeling or for clinical assays. An exception is the procedure described by Crissman and Steinkamp (1987), which, however, cannot be applied to standard, commercially available flow cytometers without modification of the hardware or software (Crissman et al., 1990). The second group of methods is based on direct or indirect immunocytochemical detection of the incorporated BrdU (Gratzner, 1982; Dolbeare et al., 1983; Dolbeare and Selden, 1990). These methods, although offering high sensitivity, require DNA denaturation or partial DNA digestion by restriction nucleases to make the incorporated BrdU accessible to antibody. Such treatments destroy chromatin structure, often preclude immunocytochemical detection of other cell components, and affect the stoichiometry of DNA staining. Despite these limitations, immunocytochemical detection of BrdU, primarily by virtue of its high sensitivity, has become a widely used methodology to identify DNA-replicating cells. Its application in clinical oncology has been expanded by the possibility of analyzing the duration of the DNA synthesis phase and the potential doubling time of tumors (Begg et al., 1985). The third, most recent approach relies on photolysis of DNA. The photolysis is selective for DNA sections containing incorporated BrdU. DNA strand breaks induced by photolysis (SBIP) in sites adjacent to the incorporated BrdU are then directly or indirectly labeled with tagged deoxynucleotides in the reaction catalyzed by exogenous TdT (Li and Darzynkiewicz, 1995). Chromatin structure and many cell constituents are perfectly preserved, but the method is rather complex and the reagents expensive. Because it is compatible with cell immunophenotyping or detection of intracellular antigens, however, the method holds promise of wide application when DNA replication has to be correlated with other cell features detected immunocytochemically (Li et al., 1996).
7.1.2 Current Protocols in Cytometry
Because of its high sensitivity in detection of incorporated BrdU and wide applicability in different fields of biology and medicine, the immunocytochemical method of BrdU detection is becoming widespread. Readers interested in the methods based on cytochemical BrdU detection, or in the SBIP methodology, will find the necessary details elsewhere (Crissman et al., 1990; Poot et al., 1990; Li and Darzynkiewicz, 1995; Li et al., 1996).
LITERATURE CITED Barlogie, B., Raber, M.N., Schumann, J., Johnson, T.S., Drewinko, B., Schwartzendruber, D.E., Goehde, W., Andreeff, M., and Freireich, E.J. 1983. Flow cytometry in clinical cancer research. Cancer Res. 43:3982-3997.
Dolbeare, F., Gratzner, H., Pallavicini, M., and Gray, J.W. 1983. Flow cytometric measurement of total DNA content and incorporated bromodeoxyuridine. Proc. Natl. Acad. Sci. U.S.A. 80:5573-5577. Dolbeare, F. and Selden, J.R. 1990. Immunochemical quantitation of bromodeoxyuridine: Application to cell-cycle kinetics. Methods Cell Biol. 41:297-316. Duque, R.E., Andreeff, M., Braylan, R.C., Diamond, L.W., and Peiper, S.C. 1993. Consensus review of the clinical utility of DNA flow cytometry in neoplastichematopathology. Cytometry 14:492-496. Evenson, D.P., Darzynkiewicz, Z., and Melamed, M.R. 1980. Relation of mammalian sperm chromatin heterogeneity to fertility. Science 210:1131-1133.
Bauer, K.D., Bagwell, C.B., Giaretti, W., Melamed, M.R., Zarbo, R.J., Witzig, T.E., and Rabinovitch, P.S. 1993. Consensus review of the clinical utility of DNA flow cytometry in colorectal cancer. Cytometry 14:486-491.
Gong, J., Traganos, F., and Darzynkiewicz, Z. 1994. A selective procedure for DNA extraction from apoptotic cells applicable for gel electrophoresis and flow cytometry. Anal. Biochem. 218:314319.
Begg, A.C., McNally, N.J., Shrieve, D.C., and Karcher, H. 1985. A method to measure the duration of DNA synthesis and the potential doubling time from a single sample. Cytometry 6:620-626.
Gorczyca, W., Bruno, S., Darzynkiewicz, R.J., Gong, J., and Darzynkiewicz, Z. 1992. DNA strand breaks occurring during apoptosis: Their early in situ detection by the terminal deoxynucleotidyltransferase and nick translation assays and prevention by serine protease inhibitors. Int. J. Oncol. 1:639-648.
Cooper, D., Schermer, A., and Sun, T.-T. 1985. Classification of human epithelia and their neoplasms using monoclonal antibodies to keratin: Strategies, applications, and limitations. Lab. Invest. 52:243-256. Crissman, H.A. and Steinkamp, J.A. 1987. A new method for rapid and sensitive detection of bromodeoxyuridine in DNA-replicating cells. Exp. Cell Res. 173:256-261.
Gratzner, H.G. 1982. Monoclonal antibody to 5bromo- and 5-iododeoxyuridine: A new reagent for detection of DNA replication. Science 218:474-475. Hall, P.A., Levison, D.A., and Wright, N.A. (eds.) 1992. Assessment of Cell Proliferation in Clinical Diagnosis. Springer-Verlag, London.
Crissman, H.A., Oishi, N., and Habbersett, R. 1990. Detection of BrdU-labeled cells by differential fluorescence analysis of DNA fluorochromes: Pulse-chase and continuous labeling methods. Methods Cell Biol. 41:341-349.
Hedley, D., Clark G.M., Cornelisse, C.J., Killander, D., Kute, T., and Merkel, D. 1993. Consensus review of the clinical utility of DNA cytometry in carcinoma of the breast. Cytometry 14:482485.
Darzynkiewicz, Z. 1988. Cellular RNA content, a feature correlated with cell kinetics and tumor prognosis. Leukemia 2:777-787.
Hiddeman, W., Schuman, J., Andreeff, M., Barlogie, B., Herman, C.J., Leif, R.C., Mayall, B.H., Murphy, R.F., and Sandberg, A.A. 1984. Convention on nomenclature for DNA cytometry. Cytometry 5:445-446.
Darzynkiewicz, Z., Traganos, F., Sharpless, T., and Melamed, M.R. 1975. Thermal denaturation of DNA in situ as studied by acridine orange staining and automated cytofluorometry. Exp. Cell Res. 90:411-422. Darzynkiewicz, Z., Traganos, F., Sharpless, T., and Melamed, M.R. 1976. Lymphocyte stimulation: A rapid multiparameter analysis. Proc. Natl. Acad. Sci. U.S.A. 73:2881-2884. Darzynkiewicz, Z., Andreeff, M., Traganos, F., and Melamed, M.R. 1978. Discrimination of cycling and noncycling lymphocytes by BudR-suppressed acridine orange fluorescence in a flow cytometric system. Exp. Cell Res. 115:31-35. Darzynkiewicz, Z., Traganos, F., and Melamed, M.R. 1980. New cell cycle compartments identified by flow cytometry. Cytometry 1:98-108.
Latt, S.A. 1973. Microfluorimetric detection of deoxyribonucleic acid replication in human metaphase chromosomes. Proc. Natl. Acad. Sci. U.S.A. 70:3395-3399. Li, X. and Darzynkiewicz, Z. 1995. Labeling DNA strand breaks with BrdUTP. Detection of apoptosis and cell proliferation. Cell Proliferation 28:571-579. Li, X., Melamed, M.R., and Darzynkiewicz, Z. 1996. Detection of apoptosis and DNA replication by differential labeling of DNA strand breaks with fluorochromes of different color. Exp. Cell Res. 222:28-37. Nucleic Acid Analysis
7.1.3 Current Protocols in Cytometry
Poot, M., Hoehn, H., Kubbies, M., Grossman, A., Chen, Y., and Rabinovitch, P.S. 1990. Cell-cycle analysis using continuous bromodeoxyuridine labeling and Hoechst-33358–ethidium bromide bivariate flow cytometry. Methods Cell Biol. 41:327-340. Shankey, T.V., Kallionieni, O.-P., Koslowski, J.M., Lieber, M.L., Mayall, B.H., Miller, G., and Smith, G.J. 1993a. Consensus review of the clinical utility of DNA content cytometry in prostate cancer. Cytometry 14:497-500. Shankey T.V., Rabinovitch, P.S., Bagwell, B., Bauer, K.D., Duque, R.E., Hedley, D.W., Mayall, B.H., and Wheeless, L. 1993b. Guidelines for implementation of clinical cytometry. Cytometry 14:472-477.
Shapiro, H.M. 1981. Flow cytometric estimation of DNA and RNA content in intact cells stained with Hoechst 33342 and Pyronin Y. Cytometry 2:143-150. Wheeless, L.L., Badalament, R.A., de Vere White, R.W., Fradet, Y., and Tribukait, B. 1993. Consensus review of the clinical utility of DNA cytometry in bladder cancer. Cytometry 14:478481.
Contributed by Zbigniew Darzynkiewicz and Gloria Juan New York Medical College Valhalla, New York
Overview of Nucleic Acid Analysis
7.1.4 Current Protocols in Cytometry
Quality Control in Nucleic Acid Analysis All flow cytometry systems require instrument quality control procedures as described in UNIT 6.1 on phenotypic analysis. However, there are additional considerations for DNA analysis beyond those outlined for sample preparation for immunophenotyping. Here, a biological material will serve to verify the linearity of the system, verify doublet discrimination, and verify cell cycle percentages. The material should have all the characteristics necessary to allow for adequate examination of the fluorescent parameter of interest to perform DNA analysis. The priorities are as follows: the control material should have as narrow a coefficient of variation (CV) as possible, with particles as large as possible; it should form doublets and/or triplets in order to verify the linearity of the system; and it should have true cycling cells to act as a positive control for cells in S phase. This unit provides information complementary to UNITS 7.3, 7.4 & 7.5, with respect to quantitative analysis of cellular DNA content, primarily to estimate DNA ploidy of tumor samples. Historically, “normal” lymphocytes, fibroblasts, or lymphocytes from the patient have been used as the locator for correct placement of diploid location and verification of the reagent system in use. This “sample preparation control” must be handled at the same time as the sample, with the same preparation, and ideally as an admixed material (i.e., both sample and control are mixed together and then stained together in the same tube). Because the intrinsic variables in DNA analysis on a flow cytometer include (1) source and tissue state, (2) histologic architecture of tissues being studied, (3) preparation artifact, (4) degree of necrosis, and (5) inflammatory reaction, the suggested sequence for DNA sample analysis is as follows: diploid control alone ↓ diploid control admixed with tissue sample ↓ tissue sample alone
This sequence will allow for the correct location (by channel number on the histogram) of the normal diploid cells using the diploid control alone. The other samples will demonstrate any slight difference in DNA content as observed by a shift from the channel number Contributed by Anne A. Hurley Current Protocols in Cytometry (1997) 7.2.1-7.2.5 Copyright © 1997 by John Wiley & Sons, Inc.
obtained in the diploid control alone, and ploidy status is verified through the tissue sample alone. The microscope is an additional analysis tool that is extremely important and often overlooked. Indeed, the microscope is the flow cytometry laboratory’s best friend, especially in DNA analysis. Perhaps the most critical aspect of DNA analysis, for any material, is knowing the morphology of the cells to be analyzed in the flow cytometer. For example, is the cell type epithelial? Does it have a mature nucleus, and is it large? A simple microscopic check of cell morphology provides a great deal of confidence in the laboratory’s results. Although historically chicken or trout red blood cells have been used as controls in flow cytometry (UNIT 1.3; i.e., as place markers for relative interpretation), there are significant problems in using them as “interpretation controls” in DNA analysis. Because both cell types are quite different from human cells in their nucleation and ploidy (most trout species are 4N, so investigators must be extremely careful when ordering trout cells from a supplier), they can be used as an instrument control only. It is important to be consistent if using trout cells, however; other than as an instrument control only, it is best to avoid their use. When attempting to distinguish slight ploidy variations to determine if a sample is aneuploid, it is important to use material whose DNA is as close that of to the sample as possible. Virtually everything must be held constant to obtain optimum results (see variables 1 to 5 above). Even if a laboratory uses lymphocytes as a diploid control, extreme care must be taken that there are only lymphocytes in the preparation—just the slight difference in how tightly wound the DNA is in lymphocytes, monocytes, and granulocytes will cause the diploid location to vary. It has been shown (Hurley et al., 1987) that, given the same concentration of propidium iodide, each of the three white cell types will be visibly separated on a histogram. Interestingly, ethidium bromide staining shows less of a distinction. Concentrations of both nuclear dyes and nuclei or cells must also be held constant to avoid population shifts on histograms caused by differences in staining equilibria. These examples highlight the need to have cell-specific diploid controls and to be sensitive to the differences in nuclear stains. Staining
UNIT 7.2
Nucleic Acid Analysis
7.2.1
42
A
35 28 21 14 7 0
Number of cells x 10 2
24
B
20 16 12 8 4 0 10
C 8 6 4 2 0 0
5
10
15
20
25
30
DNA content x 101
Figure 7.2.1 Flow cytometric analyses (histograms) of nuclei from an aneuploid human colon tumor using propidium iodide, with donor normal diploid cells as a control. (A) Normal human nuclei (control). (B) Colon tumor nuclei mixed with normal human nuclei. (C) Colon tumor nuclei. Notice that the tumor cells are hyperploid.
Quality Control in Nucleic Acid Analysis
times become critical because of cell-to-cell differences, cellular/nuclear concentrations, stoichiometry of nuclear material, and numerous other factors. Historically, the optimum staining time is determined by using a known cell concentration with a known dye amount and by analyzing samples over time (e.g., from 5 min up to 30 min or longer). When there are no more shifts in the histogram peak, the preferred staining time has been found. Note that it is important to establish staining times for each of the DNA dyes to be used.
Quantum yield, photobleaching, and steric hindrance are additional concerns. Another potential problem is residual dye, exemplified by acridine orange (AO) used for such applications as differential staining of DNA and RNA (UNIT 7.3). AO and many other fluorochromes tend to adhere to tubing and flow systems in flow cytometers. Unless great care is taken to flush the instrument with bleach, subsequent testing will be compromised, especially for unstained or weakly stained cells. Shapiro (1995) discusses specificities and characteristics of vari-
7.2.2 Current Protocols in Cytometry
10
A 8 6 4 2 0
Number of cells x 10 2
30
B
25 20 15 10 5 0 30
C
25 20 15 10 5 0 0
5
10
15
20
25
30
DNA content x 101
Figure 7.2.2 Flow cytometric analyses (histograms) of nuclei from an aneuploid human breast tumor using propidium iodide, with donor normal diploid cells as a control. (A) Normal human nuclei (control). (B) Breast tumor nuclei mixed with normal human nuclei. (C) Breast tumor nuclei. Notice that the tumor cells are hypoploid.
ous nuclear dyes and gives excellent precautions and recommendations; these issues are also addressed in UNIT 4.3. Figures 7.2.1 and 7.2.2 demonstrate some of the pitfalls in analyzing DNA content, especially with slightly aneuploid populations, and illustrate the need for proper placement of the diploid control (by channel). In Figure 7.2.1, had the three-sample sequence not been run, the sample could have been interpreted as hypoploid/diploid simply on the basis of assuming that the largest (highest) peak encompasses
the normal cells. The converse can be seen in Figure 7.2.2, where an interpretation of hyperploidy could have easily, and erroneously, been made. The DNA Cytometry Consensus Conference (1993), from which came a series of articles discussing the role of DNA analysis in various organ systems as well as the proper control of such analysis, stressed that the normal cell or stromal component present in tumor samples best represents DNA diploidy and therefore provides the best standard or calibra-
Nucleic Acid Analysis
7.2.3 Current Protocols in Cytometry
Quality Control in Nucleic Acid Analysis
tor. This is especially important because polymorphism is readily apparent from person to person and from cell type to cell type within a given person. This information again argues for the above sequence of analysis, using the normal stromal component in tumor samples as the diploid locator. In addition to the many concerns outlined here, the actual statistical analysis of the populations (both to “calculate” diploid percentages and to determine cycling phase percentages) is of extreme importance. The analysis software should be chosen on the basis of the number of cells in the sample, the amount of debris, and the extent of overlap between populations (see Chapter 10 for more extensive discussion of data analysis). Because various models have different methods for calculating CVs and different extrapolation parameters, the mathematical model criteria must be matched with the biological system under investigation. A number of companies offer software to help users statistically “model” the S phase histograms. Again, it is important to look at the data with a clear eye and ask the following questions: 1. What is the manner in which the model determines S phase? For example, a standard type of analysis platform is RFIT (rectangle fit) which fits a rectangle to the S phase of a DNA histogram, whereas SFIT (S phase fit) uses a second-order polynomial to determine the height of the S phase at any histogram channel. In each case, the mathematics inherent in the model then calculate G0/G1, S, and G2/M. A number of studies have compared analysis models on “control” materials such as calf thymocyte nuclei (CTN; Ortho Diagnostics) to examine effects on the data. 2. Is debris present in the sample and therefore expected on the histogram? The presence of debris may be determined microscopically. 3. Could there be doublets (two cells in G0/G1) in the histogram? It is important to note that the percentage of S phase cells (which is usually the most important value to know) can be greatly overestimated or underestimated compared to the “reference” method using tritiated thymidine (e.g., Shevach, 1991). However, if the flow cytometry laboratory also performed such comparisons using different software models (without the tritiated thymidine analysis), it could only conclude that the data are different, not that one set of data is more accurate than the other. An excellent resource paper that explores this further is Kallioniemi et al. (1994).
Many software programs will allow the user to subtract out debris, either automatically or manually. Obviously, this can lead to considerable confusion as many times it is difficult to ascertain where debris ends and where real cellular/nuclear information begins. Understanding the sample (e.g., cell or tumor origin, debris, cell type) is the first step in avoiding calculation errors. Doublets, which are interrogated by the laser simultaneously, in fact “look” like a single cell in G2/M; if they are included, the largest error will be in overestimation of the G2/M population. The effect can be assessed by analysis of gated and ungated histograms examining the differences in G2/M percentages. If CTN are used for this purpose, a clear difference in percentage can be seen. To avoid this potential source of error, many manufacturers provide some sort of doublet discrimination software or hardware that allows the user to correct for such occurrences. Analyses of DNA ploidy and cell cycle distributions are important tools on which patient treatment may depend. In some cases (e.g., lymphoproliferative diseases), the DNA information is crucial for selection of the proper treatment regime. The best course a laboratory can follow is first to know what “normal” is, then to determine “abnormal.” Although DNA analysis is deceptively simple, laboratory personnel need to take a great deal of time to learn the intricacies of this all-important analysis.
LITERATURE CITED DNA Cytometry Consensus Conference. 1993. Cytometry 14:471-500. Hurley, A. A., Houston, J. A. , and Stelzer, G. T. 1987. Parameters affecting interpretation of DNA content analysis. Cytometry (Suppl.)1:19 (abstr. 105). Kallioniemi, O.-P., Visakorpi, T., Holli, K., Isola, J.J., and Rabinovitch, P.S. 1994. Automated peak detection and cell cycle analysis of flow cytometric DNA histograms. Cytometry 16:250-255. Shapiro, H.M. 1995. In Practical Flow Cytometry, 3rd ed., pp. 252-262. Wiley-Liss, New York. Shevach, E.M. 1991. Labeling cells in microtiter plates for determination of [3H] thymide uptake. In Current Protocols in Immunology (J.E. Coligan, A.M. Kruisbeek, D.H. Margulies, E.M. Shevach, and W. Strober, eds.) p. A.3D.1.
KEY REFERENCES DNA Cytometry Consensus Conference, 1993. See above. Contains a series of articles discussing role of DNA analysis in various organ systems as well as proper control of such analysis.
7.2.4 Current Protocols in Cytometry
Kallioniemi et al., 1994. See above. Excellent resource paper on comparing cell cycle analysis results.
Contributed by Anne A. Hurley Comprehensive Cytometric Consulting Ballwin, Missouri
Shapiro, 1995. See above. Excellent discussion on specificities and characteristics of nuclear dyes; gives precautions and recommendations.
Nucleic Acid Analysis
7.2.5 Current Protocols in Cytometry
Differential Staining of DNA and RNA
UNIT 7.3
Analysis of nucleic acids is a common application of flow cytometry (reviewed in UNIT 7.1). Measurement of the DNA content of cells allows investigators to determine DNA ploidy (UNIT 7.5) as well as to analyze populations of cells in various phases of the cell cycle. Such studies, along with analysis of DNA strand breaks to detect apoptotic cells (UNIT 7.4), are increasingly important in disease diagnosis. Cell cycle analysis may also be done by means of differential staining of DNA and RNA, as described in this unit. Determining the RNA content allows one to discriminate G0 versus G1 cells and detect cell differentiation. Differential staining of DNA and RNA can be accomplished using either the metachromatic dye acridine orange (AO) or a combination of pyronin Y (PY) and Hoechst 33342 (PY–Hoechst 33342). Acridine orange is a versatile dye that can be used to stain a variety of different constituents in cells (Darzynkiewicz and Kapuscinski, 1990). Following cell fixation and removal of RNA with RNase, AO can also be used for differential staining of double-stranded versus single-stranded DNA sections to analyze the sensitivity of DNA to denaturation. The mechanisms of interaction with nucleic acids, and in particular the spectral changes on binding to DNA or RNA, are very much different for AO and PY–Hoechst 33342 (Table 7.3.1). Dye binding characteristics and specific features of the method based on use of AO (see Basic Protocol 1 and Alternate Protocol 1) and the method utilizing PY (see Basic Protocol 2) are therefore described separately. A procedure for determining the specificity of cell staining with AO or PY–Hoechst 33342 is also provided (see Support Protocol; also see UNIT 7.2 for more information on quality control in nucleic acid analysis). Basic Protocol 3 describes a method for staining viable cells with Hoechst 33342 and pyronin Y; Alternate Protocol 2 describes a similar procedure, also for viable cells, that includes FITC as a cell surface stain, in addition to the two abovementioned fluorochromes. Table 7.3.1 Spectral Characteristics of Dyes and Nucleic Acid–Dye Complexes
Dye or complex AO (monomer)
Absorption Recommended Emission maximum (nm) excitation (nm) maximum (nm) 492
—
525
Recommended band-pass or long-pass filter (nm) —
a
AO–ds DNA (intercalated)
502
488
520-524
AO–ss DNA (precipitated)
426-458b
457
630-644b
>640
AO–ss RNA (precipitated)
426-458b
457
630-644b
>640
547
—
565
—
PY PY–ds DNA PY–ds RNA
b
547-563
b
488-530
520±15
565-574
b
>560
b
>560
547-563
488-530
565-574
Hoechst 33342
355
—
467
—
Hoechst 33342–DNA
351
350-360
466
480±15
a Green fluorescence of AO-DNA has a long tail toward higher wavelengths. b Value depends on base composition.
Nucleic Acid Analysis Contributed by Zbigniew Darzynkiewicz, Gloria Juan, and Edward F. Srour Current Protocols in Cytometry (2004) 7.3.1-7.3.16 C 2004 by John Wiley & Sons, Inc. Copyright
7.3.1 Supplement 30
BASIC PROTOCOL 1
DIFFERENTIAL STAINING OF DNA AND RNA OF UNFIXED CELLS WITH ACRIDINE ORANGE This protocol describes the application of AO to differential staining of DNA and RNA. The cells to be stained with AO can be either prefixed in ethanol (see Alternate Protocol 1) or permeabilized with the nonionic detergent Triton X-100 as described here. The permeabilization is done at low pH in the presence of serum proteins. At low pH, most histones dissociate from DNA, making DNA more accessible to AO and thereby improving stoichiometry and accuracy in DNA detection (Darzynkiewicz et al., 1984b). Following treatment with the permeabilizing solution, the cells are stained with AO dissolved in phosphate–citric acid buffer. The high molarity and excess of the buffer neutralize the acid maintaining the low pH in the first step. In the presence of EDTA, AO at the proper concentration interacts with cellular RNA to form condensed complexes that luminesce red, with maximum emission above 630 nm. At the same time, interactions of AO with DNA result in green fluorescence. Thus this metachromatic fluorochrome differentially stains RNA and DNA (see Support Protocol).
Materials Cells to be stained (APPENDIX 3B): ≤106 cells/ml suspended in tissue culture medium containing 10% (v/v) serum or 1% (w/v) BSA Cell permeabilizing solution (see recipe), ice cold Acridine orange (AO) staining solution (see recipe), ice cold Flow cytometer equipped either with a 488- or 457-nm argon-ion laser (or both lasers) or with a mercury arc lamp NOTE: In performing the following steps, carefully avoid pipetting, vortexing, or any mechanical agitation of cells, to prevent cell lysis. 1. Set up the flow cytometer with excitation at 488 nm, using emission filters and a dichroic mirror that discriminate green fluorescence (measured at 530 ± 15 nm) and red luminescence (measured preferably above 640 or 650 nm). Maximum absorption by AO occurs at ∼455 to 490 nm. The 488-nm line of the argon ion laser is the most commonly used excitation wavelength. In instruments having a mercury or xenon lamp, blue excitation filters can be used (for example, a BG 12 short-pass filter transmitting below 470 nm or band-pass combination fitters transmitting between 460 and 500 may be used). Optimal excitation can be achieved using two lasers, one tuned to 488 nm (DNA detection, green fluorescence) and another to 457 nm (RNA detection, red luminescence).
2. Transfer a 0.2-ml aliquot of the original cell suspension to a small glass or plastic tube (e.g., 2- or 5-ml volume). Chill on ice. The 0.2-ml aliquot should have ≤2 × 105 cells suspended in tissue culture medium containing 10% (v/v) serum or 1% (w/v) BSA. The serum or BSA protects cells from lysis by detergent in step 3.
3. Gently add 0.4 ml ice-cold cell permeabilizing solution. Wait 15 sec, keeping cells on ice. 4. Gently add 1.2 ml ice-cold AO staining solution. Keep cells on ice in the dark. 5. Measure and record cell fluorescence in the flow cytometer during the 2 to 10 min after addition of AO staining solution. Differential Staining of DNA and RNA
The sample should be kept on ice prior to and during the measurement. Vortexing or syringing cells in the permeabilizing solution, especially in the absence of any serum or proteins in the original cell suspension, results in disintegration of plasma membranes and isolation of cell nuclei. The RNA content of isolated nuclei, therefore, can be measured
7.3.2 Supplement 30
Current Protocols in Cytometry
after plasma membrane disruption in this way. Visual inspection of the nuclei under phasecontrast or UV light microscopy is essential to estimate the efficiency of the isolation, which can be controlled by selecting either an optimal time and speed of vortexing or an optimal number of syringings. DNA frequency histograms (green fluorescence) can be deconvoluted to obtain the proportion of cells in G1 versus S versus G2 /M (see UNIT 7.5).
DIFFERENTIAL STAINING OF FIXED CELLS WITH ACRIDINE ORANGE In the instances when cells have to be fixed (e.g., for storage or transportation), staining with AO is done on fixed cells according to the following protocol.
ALTERNATE PROTOCOL 1
Additional Materials (also see Basic Protocol 1) Cells to be stained PBS (APPENDIX 2A), ice cold 70% ethanol, ice cold Centrifuge, 4◦ C Additional reagents and equipment for trypsinizing adherent cells (UNIT 5.2 or APPENDIX 3B) or dissociating cells from tissues (UNIT 5.2) 1a. For cells in suspension culture or hematologic samples: Rinse cells once with ice-cold PBS and suspend in ice-cold PBS at ∼106 cells/ml. 1b. For cells attached to tissue culture plates: Collect cells from flasks or petri plates by trypsinization, pool the trypsinized cells with cells floating in the medium (mostly detached mitotic and dead cells), and rinse once with medium containing serum to inactivate the trypsin (see UNIT 5.2 or APPENDIX 3B for details of this procedure). Suspend cells in ice-cold PBS at ∼106 cells/ml. Other means of trypsin inactivation such as addition of protease inhibitors may also be used.
1c. For cells isolated from solid tumors: Rinse cells free of any enzyme used for cell dissociation and suspend in ice-cold PBS at ∼106 cells/ml. The final cell suspension should be well dispersed (no aggregates), with a density ≤5 × 106 cells/ml. The cells should not be stored on ice for longer than 30 min before fixation.
2. With a Pasteur pipet transfer 1 ml cell suspension to a 15-ml conical glass tube containing 10 ml ice-cold 70% ethanol. Fix cells ≥2 hr on ice. To minimize cell clumping, rapidly injecting the cell suspension into the fixative, rather than layering onto the surface and then mixing, is preferred. The reverse order (i.e., addition of ethanol to cell suspensions) results in more extensive cell loss due to cell adherence to the glass surface and aggregation. The time of fixation in ethanol may vary, but at least 2 hr should be allowed for cells to be fixed. Cells may also be stored in ethanol at 4◦ C for several months.
3. Centrifuge tubes 5 min at 300 × g, 4◦ C. Remove all ethanol, rinse cells once with ice-cold PBS, and suspend in ice-cold PBS at a density of <2 × 106 cells/ml. 4. Withdraw 0.2 ml cell suspension (≤2 × 105 cells) and transfer to a small tube (e.g., 2 or 5 ml volume). Chill on ice. 5. Add 0.4 ml ice-cold permeabilizing solution. Wait 15 sec, keeping cells on ice. 6. Add 1.2 ml ice-cold AO staining solution. Keep cells on ice. Nucleic Acid Analysis
7.3.3 Current Protocols in Cytometry
Supplement 30
Although the presence of Triton X-100 in the permeabilizing solution is not necessary in the case of fixed cells, it does not interfere with staining. Therefore, the same solution can be used in both this protocol and Basic Protocol 1.
7. Measure cell fluorescence as described for unfixed cells (see Basic Protocol 1, step 5). To assess the contribution of RNA to the detected luminescence, following step 3 duplicate cell samples can be incubated with RNase A (100 µg/ml) for 30 min at 37◦ C, prior to staining with AO. BASIC PROTOCOL 2
DIFFERENTIAL STAINING OF DNA AND RNA WITH HOECHST 33342 AND PYRONIN Y Shapiro (1981) first proposed a combination of Hoechst 33342 and PY for the differential staining of cellular RNA and DNA. PY interacts with double-stranded RNA and double-stranded DNA by intercalation, and in this form it fluoresces at an orange-red wavelength. However, interactions of PY with DNA are suppressed in the presence of the DNA fluorochrome Hoechst 33342, which itself stains DNA (it fluoresces blue on excitation with UV light). This property makes it possible to use PY as the RNA-specific fluorochrome by using Hoechst 33342 to prevent PY staining of DNA. In this procedure the cells are fixed in ethanol, then resuspended in a solution of Hoechst 33342 and PY and measured by flow cytometry while suspended in this solution. Because Hoechst 33342 is a more specific DNA fluorochrome than is AO, this protocol is preferred when more accurate DNA content measurements are desired.
Materials Cells to be stained PBS (APPENDIX 2A), ice cold 70% ethanol, ice cold Hanks’ balanced salt solution (HBSS) containing Mg2+ and Ca2+ (APPENDIX 2A), ice cold Pyronin Y (PY)–Hoechst 33342 staining solution (see recipe), ice cold Centrifuge, 4◦ C Flow cytometer equipped either with two lasers or with one laser and a mercury arc lamp Additional reagents and equipment for trypsinizing adherent cells (UNIT 5.2 or APPENDIX 3B) or dissociating cells from tissues (UNIT 5.2) 1a. For cells in suspension culture or hematologic samples: Rinse cells once with icecold PBS and suspend in ice-cold PBS at ∼106 cells/ml. 1b. For cells attached to tissue culture plates: Collect cells from flasks or petri plates by trypsinization, pool the trypsinized cells with cells floating in the medium (mostly detached mitotic and dead cells), and rinse once with medium containing serum to inactivate the trypsin (see UNIT 5.2 or APPENDIX 3B for details of this procedure). Suspend cells in ice-cold PBS at ∼106 cells/ml. Other means of trypsin inactivation such as addition of protease inhibitors may also be used.
Differential Staining of DNA and RNA
1c. For cells isolated from solid tumors: Rinse cells free of any enzyme used for cell dissociation and suspend in ice-cold PBS at ∼106 cells/ml. The final cell suspension should be well dispersed (no aggregates) with a density no higher than 5 × 106 cells/ml. Do not store cells on ice longer than 30 min before fixation.
7.3.4 Supplement 30
Current Protocols in Cytometry
2. With a Pasteur pipet transfer 1 ml cell suspension to a 15-ml glass tube containing 10 ml ice-cold 70% ethanol. Fix cells for ≥2 hr. To minimize cell clumping, rapidly injecting the cell suspension into the fixative, rather than layering onto the surface and then mixing, is preferable. The reverse order (i.e., addition of ethanol to cell suspensions) results in more extensive cell loss due to cell aggregation and adherence of cells to the glass surface. The time of fixation (storage) in ethanol may vary from 2 hr to several months at 4◦ C.
3. Centrifuge tubes 5 min at 300 × g, 4◦ C. Remove all ethanol, rinse cells once with icecold HBSS containing Mg2+ and Ca2+ , and suspend at a density <2 × 106 cells/ml in ice-cold HBSS containing Mg2+ and Ca2+ . 4. Mix 0.5 ml cell suspension with 0.5 ml ice-cold PY–Hoechst 33342 staining solution in a small tube. Keep sample in the dark. 5. Set up the two-laser flow cytometer with one laser providing excitation in the UV, and the other providing excitation in the blue or green. Alternatively, use a combination of a laser and mercury lamp. Collect blue Hoechst 33342 fluorescence (DNA) using a 480 ± 15-nm band-pass filter and green PY fluorescence (RNA) with a 570- or 580-nm long-pass filter. Maximal excitation of DNA-bound Hoechst 33342 fluorescence is at 350 nm; optimal excitation can be achieved with the 350/356-nm line of the krypton-ion laser, the 351/363nm line of the argon laser, or with UV filters (e.g., UG 1 short-pass filter transmitting below 390 nm) in a mercury lamp system. Optimal excitation of PY is in green light (550 nm), but PY can be excited with any wavelength between 488 and 530 nm, by several lines available in either krypton- or argon-ion lasers (Crissman et al., 1985). Maximal emission of the RNA-bound PY fluorescence is at 570 nm (see Table 7.3.1).
6. Measure cell fluorescence 20 min after addition of the PY–Hoechst 33342 staining solution.
DETERMINATION OF SPECIFICITY OF CELL STAINING This protocol is provided to estimate the specificity of DNA or RNA detection by either Basic Protocol 1 or Basic Protocol 2. The percentage loss of red luminescence and green fluorescence as a result of treatment with RNase or DNase (AO procedure, see Basic Protocol 1), or loss of red and blue fluorescence (PY–Hoechst 33342, see Basic Protocol 2), is an indication of the specificity of staining of RNA or DNA, respectively.
SUPPORT PROTOCOL
Additional Materials (also see Basic Protocol 1, Alternate Protocol 1, or Basic Protocol 2) Cells fixed in ethanol (see Alternate Protocol 1, step 2, or see Basic Protocol 2, step 2) RNase solution: 100 µg/ml RNase A in PBS containing Mg2+ DNase solution: 500 µg/ml DNase I in PBS containing Mg2+ 1a. To estimate the stainability of RNA: Centrifuge ethanol-fixed cells 5 min at 300 × g, 4◦ C, thoroughly remove all ethanol by decanting, and resuspend the cell pellet (≤106 cells) in 1 ml RNase solution. Incubate 30 min at 37◦ C. Use DNase-free RNase. Otherwise, boil the RNase solution for 5 min and cool to 37◦ C before use.
1b. To estimate the stainability of DNA: Centrifuge cells as in step 1a, but resuspend the cell pellet in 1 ml DNase solution. Incubate 30 min at 37◦ C. 2. Stain cells with AO (see Alternate Protocol 1, steps 4 to 6) or with PY–Hoechst 33342 (see Basic Protocol 2, step 4).
Nucleic Acid Analysis
7.3.5 Current Protocols in Cytometry
Supplement 30
For unfixed cells stained with AO (Basic Protocol 1), cell suspensions already treated with the permeabilizing and staining solutions (see Basic Protocol 1, steps 2 to 4) may be subsequently treated with 100 µg/ml of RNase A and incubated 30 min at 24◦ C prior to fluorescence measurements.
3. Perform flow cytometry and make appropriate measurements (see Basic Protocol 1, step 5, for AO; see Basic Protocol 2, steps 5 and 6, for PY-Hoechst 33342). BASIC PROTOCOL 3
STAINING OF VIABLE CELLS WITH HOECHST 33342 AND PYRONIN Y TO IDENTIFY/SORT HEMATOPOIETIC STEM CELLS As discussed in the introduction to Basic Protocol 2, Shapiro (1981) described a DNA/RNA staining protocol using Hoechst 33342 and pyronin Y (PY) to stain these two nucleic acids independently. Later, Ladd et al. (1997) experimented with these dye combinations using lower concentrations of PY in an attempt to identify and isolate viable human hematopoietic cells in different phases of the cell cycle. This work was prompted by the long-standing notion that hematopoietic stem cells lie dormant in adult bone marrow (Till and McCullough, 1961) and by the inability to isolate viable cells in the G0 phase of the cell cycle. Fortunately, this approach allowed for the successful cell sorting of viable human hematopoietic cells in all phases of the cell cycle for further investigations. However, for unknown reasons (see Critical Parameters and Troubleshooting), this dye combination is extremely toxic to murine hematopoietic cells even when used at lower concentrations than those employed with human cells. Specificity of RNA staining with PY under conditions used with this protocol was verified with RNase treatment. The resultant loss of PY fluorescence demonstrated that PY specifically stained RNA (Ladd et al., 1997). However, it should be noted that PY also stains mitochondrial membranes in live cells, and therefore the RNA specificity obtained with PY staining and the cell cycle status of isolated cells should be confirmed with alternative methods. Use of RNase to interfere with PY staining of RNA requires permeabilization of the cells, a process that causes mitochondrial membranes to discharge, forcing the release of PY, which then stains RNA. One of the most reliable markers to confirm the identity of isolated cells with respect to their cell cycle status is Ki-67 (Jordan et al., 1996; Gothot et al., 1998). Ki-67 is a nuclear protein expressed by cells as they traverse from G0 into G1 , and should not therefore be expressed by cells defined by the procedure described below as being in the G0 phase of the cell cycle.
Materials Viable cells to be stained Hoechst staining buffer (see recipe) Hoechst 33342 working solution (see recipe) Pyronin Y working solution (see recipe) 15-ml snap-cap tubes Flow cytometer 1. Place 1–5 × 106 cells to be stained and sorted into a 15-ml snap-cap tube (sample tube). Also prepare two control tubes (these can have the smallest number of cells possible, i.e., ∼1 × 105 cells): a Hoechst 33342 control for cells stained with only Hoechst 33342, and a pyronin Y control for cells stained with pyronin Y only. Differential Staining of DNA and RNA
If human hematopoietic cells are to be stained, selection of CD34+ cells, if required, can be done prior to staining, or CD34+ cells can be identified by surface immunostaining as a separate parameter.
7.3.6 Supplement 30
Current Protocols in Cytometry
For larger numbers of cells, a larger volume of Hoechst 33342 working solution (see step 3) and more frequent vortexing of cells while staining are recommended. If the cell number increases substantially, staining of cells in more than one tube may produce better results.
2. Wash cells in all three tubes once with Hoechst staining buffer by centrifuging 10 min at ∼300 × g, 4◦ C, and discarding the supernatant. 3. Add 0.5 ml Hoechst 33342 working solution to the Hoechst 33342 control tube and 0.5 ml Hoechst staining buffer to the pyronin Y control tube. 4. Add 1.5 ml Hoechst 33342 working solution to the sample tube. 5. Vortex all three tubes gently, and incubate 45 min in a 37◦ C water bath in the dark. Vortex gently every 15 min. Increase frequency of vortexing if larger number of cells and a larger volume of Hoechst working solution are used.
6. At the end of the incubation period, add 2.5 µl pyronin Y working solution to the pyronin Y control tube and 7.5 µl (or equivalent volume based on the ratio of 2.5 µl for every 500 µl of Hoechst working solution) to the sample tube. Add 2.5 µl Hoechst staining buffer to the Hoechst control tube. Vortex all three tubes and incubate in a 37◦ C water bath in the dark for 45 min. Vortex gently every 15 min, or more frequently. Since pyronin Y is used to stain RNA after cellular DNA has already been stained with Hoechst 33342, the pyronin Y working solution is added to cells without washing the Hoechst 33342 stain. This step constitutes the final dilution step for pyronin Y, which is used at a final concentration of 3.3 µM.
7. Add an adequate volume of ice-cold Hoechst staining buffer for washing to each tube, and centrifuge all tubes as described in step 2. Decant the supernatant and resuspend the cell pellet in an appropriate volume of Hoechst staining buffer. 8. Perform flow cytometric analysis and cell sorting. The selection criteria for cells in G0 are as follows.Viewed separately, Hoechst 33342 should yield a typical DNA histogram with a G0 /G1 peak followed by an S population and a G2+M peak whereas pyronin Y should yield a continuum of staining from dim to bright, corresponding to cells with low and high RNA content, respectively. Since cells in G0 do not require robust RNA transcription, these cells should then have 2N DNA and low pyronin Y staining. Therefore, in a dot plot depicting Hoechst 33342 staining on the x axis and pyronin Y on the y axis, cells in G0 would be in the lower left corner of the dot plot. Cells in G1 would have 2N DNA but a higher pyronin Y fluorescence while cells in S and G2+M would have a bright pyronin Y fluorescence and a higher-than-2N DNA content.
SIMULTANEOUS CELL SURFACE AND HOECHST 33342 AND PYRONIN Y STAINING
ALTERNATE PROTOCOL 2
Excitation and emission properties of Hoechst 33342 and pyronin Y allow for the use of FITC as a surface marker. In addition, other commonly used fluorochromes such as allophycocyanin (APC) can also be used in conjunction with Hoechst 33342 and pyronin Y, depending on the optical bench configuration available to the user. However, using FITC in addition to Hoechst 33342 and pyronin Y is probably the easiest application when immunostaining is required. Staining cells with a surface marker requires a modification of Basic Protocol 3, as described below. Control tubes prepared for this procedure should include a positive FITC control (cells stained with the FITC-conjugated test antibody only) to be used for compensation and
Nucleic Acid Analysis
7.3.7 Current Protocols in Cytometry
Supplement 30
determination of percent positive, and cells stained with Hoechst, pyronin Y (PY), and FITC-conjugated isotype control, to be used for establishing background FITC fluorescence.
Additional Materials (also see Basic Protocol 3) Surface immunostain (e.g., FITC-labeled anti-CD34, anti-CD38, or anti-CD133) Isotype control 1. Stain and wash cells as described in Basic Protocol 3, steps 1 through 7. 2. Stain pelleted cells in a very small volume with the required surface immunostain while maintaining the cells on ice. Make sure to include all the appropriate isoptype control tubes. Immunostaining should be carried out as usual, i.e., on ice for the appropriate time with occasional vortexing of cells.
3. At the end of the immunostaining, wash the cells with ice-cold Hoechst buffer by centrifuging 10 min at ∼300 × g, 4◦ C, and discarding the supernatant. Resuspend the cells in that buffer for analysis and sorting.
REAGENTS AND SOLUTIONS Use distilled, deionized water for the preparation of all buffers. For common stock solutions, see APPENDIX 2A for suppliers, see SUPPLIERS APPENDIX.
Acridine orange (AO) staining solution Stock solution: Dissolve 1 mg/ml AO in distilled water and store ≤6 months in the dark at 4◦ C. Staining solution 370 ml 0.1 M citric acid (37 mM final) 630 ml 0.2 M Na2 HPO4 (126 mM final) 8.77 g NaCl (150 mM final) 340 mg Na2 EDTA (1 mM final) 6.0 ml 1 mg/ml AO stock solution (see above; 6 µg/ml final) H2 O to 1 liter Store ≤6 months in dark or foil-wrapped bottles at 4◦ C The quality of the AO is essential. Highly purified AO (mol. wt. 302) is available from Molecular Probes. Stir to dissolve the sodium chloride before adding the EDTA. Continue stirring until completely dissolved. An equivalent amount of Na4 EDTA may be used; H4 EDTA can also be used but will require a longer time to dissolve. The AO staining solution may be kept in a brown-colored automatic-dispensing pipet bottle set at 1.2 ml.
Cell permeabilizing solution 1 ml Triton X-100 (0.1% final) 80 ml 1.0 M HCl (80 mM final) 8.77 g NaCl (150 mM final) H2 O to 1 liter Store ≤6 months at 4◦ C Differential Staining of DNA and RNA
The permeabilizing solution may be kept in an automatic-dispensing pipet bottle set at 0.4 ml.
7.3.8 Supplement 30
Current Protocols in Cytometry
Hoechst 33342 working solution Intermediate solution: Obtain 10 mg/ml stock solution of Hoechst 33342 in water from Molecular Probes. Prepare an intermediate solution by adding 20 µl stock solution to 1.98 ml Hoechst staining buffer (see recipe); keep in the dark and promptly dilute to prepare the working solution. Working solution: Mix 50 µl intermediate solution with 4.95 ml Hoechst staining buffer (final concentration, 1 µg/ml or 1.6 µM Hoechst 33342). Keep in the dark and use within a few hours of preparation. The source of Hoechst 33342 is very important; Molecular Probes supplies an excellent stock solution. Intermediate and working solutions of Hoechst 33342 should be prepared fresh for every application, kept in the dark, and used within a few hours of preparation. When diluted in PBS or Hoechst staining buffer (see recipe below), Hoechst 33342 precipitates at concentrations higher than 30 µM. This process is not instantaneous and therefore there is ample time to prepare the final working solution from the intermediate solution, which has a concentration >30 µM.
Hoechst staining buffer Hanks’ balanced salt solution (HBSS; APPENDIX 2A) containing: 20 mM HEPES, pH 7.2 1 g/liter glucose 10% fetal bovine serum (FBS) Store buffer with above components up to 6 weeks at 4◦ C (better to prepare more often) Immediately before use, add verapamil (from 100 mg/ml stock; see recipe) to 50 to 100 µM final concentration Recipe from Wolf et al. (1993). Hoechst staining buffer containing verapamil should not be stored.
Pyronin Y working solution Stock solution: Obtain pyronin Y as a 5-g powder preparation from Polysciences. Dissolve all 5 g in 10 ml water to prepare a 500-mg/ml stock solution. Store at 4◦ C in the dark. Intermediate solution: Mix 100 µl stock solution with 4.9 ml Hoechst staining buffer to prepare the 10-mg/ml intermediate solution (can also be stored at 4◦ C in the dark). Working solution: Mix 10 µl intermediate solution with 490 µl Hoechst staining buffer.
Verapamil stock solution, 100 mg/ml Dissolve verapamil hydrochloride (supplied in 1.0-g amounts by Sigma) in DMSO at 100 mg/ml. Store up to 1 week at 4◦ C. Pyronin Y (PY)–Hoechst 33342 staining solution 2 mg Hoechst 33342 4 mg PY Add HBSS containing Mg2+ and Ca2+ (APPENDIX 2A) to 1 liter Prepare fresh Hoechst 33342 (mol. wt. 652) is available from Molecular Probes. Pure PY is available from Polysciences. Most PY from other sources has 30% to 40% impurities and should be purified by chloroform extraction and recrystallization from methanol (Kapuscinski and Darzynkiewicz, 1987).
Nucleic Acid Analysis
7.3.9 Current Protocols in Cytometry
Supplement 30
COMMENTARY Background Information
Differential Staining of DNA and RNA
Differential staining with AO The unique ability of AO to differentially stain nucleic acids of different conformations stems from the fact that this metachromatic dye shows a large change in its absorption and emission spectra when bound to doublestranded (ds) versus single-stranded (ss) nucleic acids (Table 7.3.1). AO binds to ds nucleic acids by intercalation, and the intercalated form fluoresces green when excited by blue light. The maximum absorption of AO bound by intercalation to DNA is at 500 to 506 nm and emission is at 520 to 524 nm; this is classical fluorescence emission (S1 to S0 transitions), with short (5-nsec) lifetime (Wilson and Jones, 1982). Interaction of AO with ss nucleic acids is a complex, multistep process initiated by AO intercalation between neighboring bases, neutralization of the polymer charge by the cationic dye, and subsequent condensation and agglomeration (precipitation; solute-to-solid state transition) of the product (Kapuscinski et al., 1982). The condensation reaction is highly cooperative. In the final product, AO molecules are interspaced with DNA bases, forming stacks of alternating dye-base composition, which by virtue of their solid-state form are protected, to a limited degree, from interaction with oxygen or water molecules. The absorption spectrum of AO in these precipitated products is blueshifted compared to that of the intercalated AO, with maximum absorption ranging between 426 and 458 nm, depending on the base composition of the nucleic acid. The emission of AO in these complexes also varies, between 630 and 644 nm, depending on the base composition. The lifetime of the red emission at room temperature is >20 nsec (Darzynkiewicz and Kapuscinski, 1990). These spectral properties of AO in complexes with ss nucleic acids are suggestive of intersystem crossing (phosphorescence; T1 to S0 transition) rather than fluorescence. The solidstate nature of the complexes, similar to freezing AO in solution, may facilitate the intersystem crossing by partially eliminating collision quenching, which otherwise occurs due to the long lifetime of the T1 excited state in the presence of oxygen and solvents. In the cell, large sections of rRNA and tRNA have ds conformation. Therefore, to obtain differential staining of DNA versus RNA with AO, these sections have to be selectively denatured, under conditions in which
DNA still remains double stranded. This is accomplished by treatment of cells with AO in the presence of the chelating agent EDTA (Darzynkiewicz et al., 1976). By breaking RNA-protein interactions in ribosomes that stabilize ds RNA, EDTA promotes denaturation of ds RNA, which occurs as a result of interaction with AO. The RNA-selective denaturing properties of AO result from the fact that this ligand has higher affinity to ss RNA than ss DNA (Kapuscinski and Darzynkiewicz, 1989). The denaturation itself results from the fact that, at increased dye/phosphate (D/P) ratio, binding of AO to ss nucleic acids becomes thermodynamically preferable; the weaker but more numerous 1:1 (D/P) interactions of AO with ss sections thermodynamically dominate the stronger but fewer (1:4) sites of AO intercalation to ds regions (Kapuscinski and Darzynkiewicz, 1989). Thus, increasing D/P promotes denaturation of ds RNA sections. Cell staining with AO is generally done in salt solutions of relatively high ionic strength (0.1 to 0.2 M NaCl). Because of the significant electrostatic component in binding of AO to nucleic acids, competitive interactions between the AO+ and Na+ result, and it is the concentration of free AO in solution (rather than absolute D/P calculated on the basis of molar ratios of the dye and nucleic acid in the sample) that is of importance for selective RNA denaturation (Darzynkiewicz and Kapuscinski, 1990). To recapitulate, AO has two distinct functions in the mechanism of metachromatic staining of RNA, namely, it (1) denatures ds RNA and (2) differentially stains RNA (after its conversion to ss form) versus DNA. At a given ionic strength, selective RNA denaturation can be achieved only within a narrow concentration range of free dye. Too low an AO concentration produces incomplete RNA denaturation (both DNA and portions of RNA then stain green), whereas too high a concentration also leads to denaturation of DNA (both RNA and DNA stain red). Thus acridine orange, in contrast to most other dyes used in cytometry, requires very stringent conditions, especially with regard to concentration of AO and ionic strength of the staining solutions. The protocol of cell staining described in Basic Protocol 1 and Alternate Protocol 1 has been established after testing a variety of ionic conditions, pH, and dye concentrations (Darzynkiewicz et al., 1976; Traganos et al., 1977).
7.3.10 Supplement 30
Current Protocols in Cytometry
Differential staining with PY–Hoechst 33342 Historically, PY has been widely used in absorption microscopy as a dye that in combination with methyl green specifically stains cellular RNA (Scott, 1967). More recently, it also found an application in flow cytometry as a fluorochrome of RNA (Tanke et al., 1980; Shapiro, 1981). The interactions of PY with nucleic acids, which are responsible for its specificity to RNA, as in the case of AO (see above) also are complex (Kapuscinski and Darzynkiewicz, 1987). The following binding and spectral characteristics of PY are of importance for its role as an RNA fluorochrome: 1. Intercalary mode of binding. PY binds by intercalation to ds nucleic acids. Its binding affinity to ds RNA is severalfold higher than to ds DNA. In the intercalated form, whether bound to RNA or DNA, the dye has maximal absorption between 547 and 563 nm and fluoresces with maximum emission between 565 and 574 nm; the variation is due to differences in base composition of the nucleic acids (Kapuscinski and Darzynkiewicz, 1987). Its quantum yield also varies widely with changes in base composition. In contrast to AO, which stains total cellular RNA, PY used as fluorochrome can detect only ds sections of RNA and is sensitive to the AU/GC base ratio. 2. Condensation and precipitation of nucleic acids. Like AO binding, binding of PY to ss nucleic acids results in condensation (precipitation) of the product. The mechanism of condensation and the structure of the complexes are similar to those generated by AO. In contrast to AO, however, PY fluorescence is nearly totally quenched in these complexes (Kapuscinski and Darzynkiewicz, 1987). In absorption microscopy these products are characterized by lavender color. 3. Stoichiometry of RNA detection. The stoichiometry and thermodynamics of binding of PY to ds and ss nucleic acids are similar to those of AO. Thus, PY can denature the ds sections of nucleic acids, rendering them single-stranded and causing their condensation and agglomeration (Kapuscinski and Darzynkiewicz, 1987). At increasing PY concentration, therefore, the fluorescence of PY bound to ds RNA is suppressed because of the progressive denaturation of the ds sections (“self-extinguishing” effect of PY). Similarly, as in the case of AO, selective stainability of RNA with PY can be obtained at only a relatively narrow range of dye concentrations. 4. DNA stainability with PY. PY, having affinity to ds DNA, can also stain DNA.
Its binding to DNA, however, can be suppressed by DNA-specific ligands such as methyl green and Hoechst 33342 (Tanke et al., 1980; Shapiro, 1981). In the presence of these dyes, therefore, PY can be used as a specific RNA fluorochrome. Dual cell staining with Hoechst 33342 and PY as described in Basic Protocol 2 provides the basis for simultaneous detection of DNA and RNA in flow cytometry (Shapiro, 1981; Darzynkiewicz et al., 1987). Comparison of AO and PY–Hoechst 33342 methods Each method is characterized by different advantages and limitations that should be considered. Quantitative aspect of the methods. The stoichiometry of RNA measurement is better assured by the AO methodology (Bauer and Dethlefsen, 1981), because total RNA content is stained by AO, and because there is less variation in quantum yield resulting from differences in base composition or conformation of RNAs than in the case of PY. Differences in sensitivity of RNA detection. Because there is significant spectral overlap between AO bound to DNA and AO bound to RNA, the ability to detect small amounts of RNA under standard measuring conditions is better provided by the PY–Hoechst 33342 technique. Excitation of AO-stained cells at two different wavelengths, however, raises the sensitivity of the AO method to or above that of PY. Quantum yields and fluorescence intensities of cells stained with AO are higher than with PY. Specificity in DNA content measurement. Hoechst 33342 is a more specific DNA stain than AO; the latter stains glycosaminoglycans (Darzynkiewicz and Kapuscinski, 1990) as well as DNA and RNA. Resolution of DNA measurements by AO in cells containing excessive amounts of glycosaminoglycans (primary fibroblasts, mast cells, and keratinocytes) is therefore low. Analysis of RNA conformation. Both AO and PY can be used to reveal changes in the conformation of RNA. The PY–Hoechst 33342 technique detects changes associated with assemblage of polyribosomes (Traganos et al., 1988), whereas the AO method can be applied to measure the degree of double strandedness of RNA and its sensitivity to heat denaturation (melting profile). Instrument requirements. The AO methodology (Basic Protocol 1, Alternate Protocol 1) has the advantage of requiring simpler and less costly instruments. The protocol can be
Nucleic Acid Analysis
7.3.11 Current Protocols in Cytometry
Supplement 30
performed with excitation at a single wavelength or with a mercury lamp instead of a laser. RNA content standards RNA and DNA content measured in any cells can be expressed quantitatively by comparison with the content of standard, calibrated cells. Nonstimulated peripheral blood lymphocytes appear to offer the best standard. RNasetreated and untreated lymphocytes should be measured to establish the extent of RNasespecific red luminescence. The cells to be compared have to be measured under conditions identical to those used for the lymphocytes. Their RNA index should be expressed as a multiplicity of the lymphocyte RNA content. The lymphocytes may also serve as a calibrator of DNA content, to estimate the DNA index from the mean (modal) intensity of the green fluorescence of the G0 /G1 population.
Differential Staining of DNA and RNA
Applications of RNA analysis In the majority of cell types, ∼80% of total cellular RNA is rRNA. Most of the remaining RNA is tRNA, and only a minor fraction of total RNA is mRNA. The content of total cellular RNA, therefore, is primarily an indicator of the number of ribosomes per cell and reflects cell translational potential. Nuclear RNA, being predominantly pre-rRNA, also is associated with cell capacity to synthesize proteins, and its increase often precedes the buildup of ribosomal machinery in the cytoplasm. RNA content measurement, either of the whole cell or of the isolated nucleus, is a marker of the overall translational capacity of the cell. Differences in RNA content between individual cells have two different origins. One is associated with the tissue type– or cell differentiation–related constitutive level of the translational activity. Thus, cells known to secrete, or produce for internal use, large quantities of tissue-specific protein (e.g., plasma cells or neurons) are generally characterized by high RNA content. In these cases the cellular RNA content is a reflection of a phenotype of the differentiated cell. Its measurement, therefore, can discriminate between cells of different tissues in the sample or be a marker of cell differentiation. The most common application of RNA measurement, in this context, is to identify reticulocytes, the immature red blood cells which retain RNA; in contrast, mature red blood cells have no measurable RNA (Tanke et al., 1980). The second cause of variability in RNA content is related to cell reproduction. Divid-
ing cells double their constituents, including the number of ribosomes, during the cell cycle. Progression through the cell cycle is thus associated with an increase in cellular RNA content, which occurs throughout interphase at a relatively constant rate proportional to the rate of cell proliferation (Darzynkiewicz et al., 1984a). Cellular RNA content, therefore, is a reflection of cell maturity in the cycle, and as such allows the discrimination of earlyG1 from late-G1 cells (Darzynkiewicz et al., 1980). Because growth in cell size (number of ribosomes) and the rate of proliferation are generally coupled, the RNA parameter, therefore, is indirectly a marker of cell proliferation. Noncycling cells withdrawn from the cell cycle (quiescent cells, G0 , G1Q ) have on average 5- to 10-fold fewer ribosomes than their cycling counterparts (Johnson et al., 1974). Thus, the difference in RNA content allows one to identify the noncycling cells and can be used as a marker of their mitogenic stimulation (Darzynkiewicz et al., 1976). Tumorous transformed cells, on the other hand, even noncycling ones, are characterized by an RNA content significantly higher than that of normal quiescent cells (Stanners et al., 1979). There is a growing body of evidence that the RNA content of tumor cells has prognostic value in many malignancies (reviewed by Darzynkiewicz, 1988).
Critical Parameters and Troubleshooting Differential staining with AO 1. Preservation of intact cells during permeabilization with Triton X-100. Disintegration (lysis) of unfixed cells during permeabilization by Triton X-100 is prevented by the presence of serum or serum albumin (Darzynkiewicz et al., 1976). It is recommended, therefore, to have the cells suspended in PBS or culture medium that contains 10% to 20% (v/v) serum or 1% (w/v) bovine serum albumin prior to addition of the permeabilizing solution. Furthermore, vigorous shaking, pipeting, or vortexing cell suspensions after addition of the detergent should be avoided (Darzynkiewicz et al., 1976). 2. Critical concentration of AO. Differential staining of RNA versus DNA requires a proper concentration (∼20 mM) of free (unbound) AO in the final staining solution as well as at the time of fluorescence measurement, i.e., at the moment of cell intersection with the laser beam in the flow cytometer. The
7.3.12 Supplement 30
Current Protocols in Cytometry
Figure 7.3.1 Differential staining of RNA and DNA. Lymphocytes were analyzed (A) 0 hr, (B) 16 hr, (C) 24 hr, and (D) 48 hr after stimulation with phytohemagglutinin (PHA). The cells were stained with AO according to Basic Protocol 1. Insets represent DNA content frequency histograms (green fluorescence) of the respective cell populations.
following problems associated with this requirement may occur: (a). When the cell number (density) in the original suspension exceeds 2 × 106 cells/ml (or even less when cells are highly hyperdiploid and/or have excessive RNA content), the amount of bound AO is high and therefore the free dye concentration may be significantly reduced (the “mass action” law). RNA denaturation is then incomplete, and some RNA can stain green. In such cases, dilute the cell suspension to obtain a lower concentration. (b). With some instruments (e.g., most cell sorters) in which cell measurements take place in air outside the nozzle, a significant diffusion
of dye from the sample to the sheath fluid takes place before the cell reaches the laser beam (see Fig. 1.1.6). This breaks the equilibrium and lowers the actual AO concentration in the sample at the time of cell measurement. Dye diffusion is also a problem in some instruments that have narrow sample streams and long flow channels (e.g., Cytofluorograf 50 made by Ortho Diagnostics). The solution is to increase the AO concentration in the AO solution (up to 20 µg/ml) and to increase the sample flow rate to compensate for the diffusion. Wherever possible, use channels with favorable geometry (wider sample stream and/or shorter distance between the nozzle and intersection with
Nucleic Acid Analysis
7.3.13 Current Protocols in Cytometry
Supplement 30
Figure 7.3.2 Protocol 2.
Differential Staining of DNA and RNA
Differential staining of RNA and DNA with PY–Hoechst 33342 as described in Basic
the laser beam). The optimal dye concentration for a particular instrument can be established by preparing a series of staining solutions with increasing AO concentrations (e.g., from 5 to 20 µg/ml) and determining the concentration at which cells in G0 /G1 cell cluster have the same green fluorescence (the lowest coefficient of variation [CV] of the green fluorescence mean value, corresponding to a lack of correlation between green and red luminescence). On the bivariate DNA/RNA cytograms (see Figs. 7.3.1 and 7.3.2), the G0 /G1 cell cluster ought to be horizontal (or vertical if axes are reversed) but never skewed (diagonal). 3. Overlap of red and green emission spectra of AO. One of the limitations of the AO technique is the relatively low sensitivity of RNA detection. This is primarily due to overlap of the emission spectra: the green fluorescence of AO intercalated to DNA has a long tail toward higher wavelengths. Therefore, RNA measurements in cells (or cell nuclei) characterized by a high DNA/RNA ratio lack sensitivity, being obscured by the high component of AO bound to DNA. There are two ways to improve the sensitivity of RNA measurements: (1) Use long-pass filters transmitting above 640 or 650 nm, rather than 610 or 620 nm, to measure red luminescence. This significantly reduces the DNA-associated spectral component. (2) When possible use two lasers for excitation, one tuned to 457 nm to excite the red luminescence and another tuned to 488 nm to excite the green fluorescence of AO. 4. Contamination of sample tubing by AO. Like other cationic, strongly fluorescing fluorochromes (e.g., rhodamine 123), AO is adsorbed to surfaces of the sample flow tubing. Its subsequent release from the tubing interferes with later sample measurement, especially with cells of low fluorescence intensity.
To avoid this problem, after measuring samples stained with AO and prior to measuring samples stained with other fluorochromes, rinse the sample flow line with bleach (e.g., 10% Clorox), then 50% ethanol, then PBS, each for 10 min. Alternatively, if possible, have some sample flow tubing dedicated exclusively to AO use. Differential staining with PY–Hoechst 33342 In many respects the critical points of cell staining are similar for PY and AO. The most crucial aspect of the PY–Hoechst 33342 methodology is a stringent requirement for the appropriate PY concentration. Too low a concentration of the dye (or too dense a cell suspension) cannot ensure stoichiometry of staining because of the paucity of PY in the solution (i.e., dye/binding site ratio <1.0). Too high a concentration of PY triggers denaturation and condensation of RNA, which quenches PY fluorescence. Paradoxically, thus, by increasing the PY concentration one can completely suppress RNA fluorescence and (in the absence of Hoechst 33342) induce PY intercalation to DNA; under these conditions PY can be used as a DNA-specific fluorochrome (Portela and Stockert, 1979). Because of these denaturing properties of PY, the RNA stainability is very sensitive to native RNA conformation. Under appropriate conditions, the procedure allows one to discriminate between polyribosomal RNA (which is more resistant to denaturation and shows more extensive double strandedness) and rRNA in dispersed ribosomes (Traganos et al., 1988). Thus the staining procedure can be used to measure disaggregation of polyribosomes occurring during mitosis or hyperthermia.
7.3.14 Supplement 30
Current Protocols in Cytometry
It should be stressed that PY taken up by live cells is partially localized in mitochondria and lysosomes (Darzynkiewicz et al., 1987). The specificity of staining of RNA with PY in live cells (Shapiro, 1981), therefore, is uncertain. Staining of viable cells with Hoechst 33342 and pyronin Y It is important to determine the toxicity of the Hoechst 33342/pyronin Y dye combination for each cell type to be stained. As noted above, human hematopoietic cells from many sources (adult bone marrow, mobilized peripheral blood, umbilical cord blood, and fetal tissues) tolerate this staining protocol well, and cell viability after cell sorting is excellent. However, murine hematopoietic cells are susceptible to the stain combination even when lower concentrations are used. This may be owing to the fact that pyronin Y also stains mitochondrial membranes, which become very light sensitive (Z. Darzynkiewicz, pers. comm.). However, it is not presently known why human CD34+ cells are perhaps more resistant to this injury. Since cells stained with Hoechst and pyronin Y become photosensitive, it is therefore advisable to shield cells from direct light during staining and sorting. Commercial preparations of pyronin Y are usually contaminated with impurities. Pyronin Y preparations from Polysciences contain fewer impurities than similar preparations from other vendors.
Anticipated Results Changes in cellular DNA and RNA content during mitogenic stimulation of lymphocytes are shown in Figure 7.3.1. On the basis of differences in RNA content, it is possible to distinguish nonstimulated, quiescent cells from cells entering the cell cycle. Progression through the cell cycle is paralleled by further increases in RNA content. Correlated measurements of RNA and DNA offer a sensitive assay that provides information regarding both the initial steps of stimulation (exit from G0 ) and cell cycle progression. As stimulation of lymphocytes is a multistep process that does not always result in cell proliferation, traditional assays based on radioactive thymidine incorporation, in contrast to the differential staining technique, cannot detect the early steps of the stimulation process and thus are useless in such situations. Stainability of RNA and DNA with PY and Hoechst 33342 is shown in Figure 7.3.2. The staining reaction is very sensitive to the conformation of RNA in the cell. Thus, as is evident,
cells in mitosis (M) have lower stainability with PY than do G2 cells, despite the fact that the RNA content of M cells is somewhat higher than that of most G2 cells. The hypochromicity of RNA in M cells with PY is a consequence of the denaturing properties of the dye: RNA of polyribosomes (which are more numerous in interphase than in metaphase) is more resistant to denaturation and thus stains more intensively compared to RNA in mitotic cells. During mitosis, polyribosomes disaggregate and RNA of individual ribosomes is more extensively denatured by PY, leading to quenching of fluorescence (Traganos et al., 1988).
Time Considerations
Cell staining with AO requires <1 min. Cell fixation for PY–Hoechst 33342 staining takes ∼10 min, after which cells must remain in fixative for ≥2 hr. Subsequent cell staining takes another 10 min. Cytometric analysis takes between 1 and 10 min depending on the cell density in the sample. Analysis of the results takes, on average, 2 to 5 min.
Literature Cited Bauer, K.D. and Dethlefsen, L.A. 1981. Control of cellular proliferation of HeLa-S3 suspension cultures. Characterization of cultures utilizing acridine orange staining procedures. J. Cell Physiol. 108:99-112. Crissman, H.A., Darzynkiewicz, Z., Tobey, R.A., and Steinkamp, J.A. 1985. Correlated measurements of DNA, RNA and protein in individual cells by flow cytometry. Science 228:1321-1324. Darzynkiewicz, Z. 1988. Cellular RNA content, a feature correlated with cell kinetics and tumor prognosis. Leukemia 2:777-787. Darzynkiewicz, Z. and Kapuscinski, J. 1990. Acridine orange: A versatile probe of nucleic acids and other cell constituents. In Flow Cytometry and Cell Sorting, 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 291314. Wiley-Liss, New York. Darzynkiewicz, Z., Traganos, F., Sharpless, T., and Melamed, M.R. 1976. Lymphocyte stimulation: A rapid multiparameter analysis. Proc. Natl. Acad. Sci. U.S.A. 73:2881-2884. Darzynkiewicz, Z., Sharpless, T., Staiano-Coico, L., and Melamed, M.R. 1980. Subcompartments of G1 phase of the cell cycle identified by multiparameter flow cytometry. Proc. Natl. Acad. Sci. U.S.A. 77:6696-6699. Darzynkiewicz, Z., Crissman, H.A., Traganos, F., and Steinkamp, J. 1984a. Cell heterogeneity during the cell cycle. J. Cell Physiol. 113:465-474. Darzynkiewicz, Z., Traganos, F., Kapuscinski, J., Staiano-Coico, L., and Melamed, M.R. 1984b. Accessibility of DNA in situ to various fluorochromes: Relationship to chromatin
Nucleic Acid Analysis
7.3.15 Current Protocols in Cytometry
Supplement 30
changes during erythroid differentiation of Friend leukemia cells. Cytometry 5:355-363. Darzynkiewicz, Z., Kapuscinski, J., Traganos, F., and Crissman, H.A. 1987. Application of pyronin Y (G) in cytochemistry of nucleic acids. Cytometry 8:138-145. Gothot, A., van der Loo, J.C.M., Clapp, D.W., and Srour, E.F. 1998. Cell cycle–related changes in repopulating capacity of human mobilized peripheral blood CD34+ cells in non-obese diabetic/severe combined immune-deficient mice. Blood 92:2641-2649.
Shapiro, H.M. 1981. Flow cytometric estimation of DNA and RNA content in intact cells stained with Hoechst 33342 and pyronin Y. Cytometry 2:143-150. Stanners, C.P., Adams, M.E., Harrkins, J.L., and Pollard, J.W. 1979. Transformed cells have lost control of ribosome number through their growth cycle. J. Cell Physiol. 100:127-138. Tanke, H.J., Nieuwenhuis, I.A.B., Koper, G.J.M., Slats, J.C.M., and Ploem, J.S. 1980. Flow cytometry of human reticulocytes based on RNA fluorescence. Cytometry 1:313-320.
Johnson, L.F., Abelson, H.T., Green, H., and Penman, S. 1974. Changes in RNA in relation to growth of fibroblasts. I. Amounts of RNA in resting and growing cells. Cell 1:95-100.
Till, J.E. and McCulloch, E.A. 1961. A direct measurement of the radiation sensitivity of normal mouse bone marrow cells. Radiat. Res. 14:213222.
Jordan, C.T., Yamasaki, G., and Minamoto, D. 1996. High-resolution cell cycle analysis of defined phenotypic subsets within primitive human hematopoietic cell populations. Exp. Hemat. 24:1347-1352.
Traganos, F., Darzynkiewicz, Z., Sharpless, T., and Melamed, M.R. 1977. Simultaneous staining of ribonucleic and deoxyribonucleic acids in unfixed cells using acridine orange in a flow cytofluorometric system. J. Histochem. Cytochem. 25:46-56.
Kapuscinski, J. and Darzynkiewicz, Z. 1987. Interactions of pyronin Y (G) with nucleic acids. Cytometry 8:129-137. Kapuscinski, J. and Darzynkiewicz, Z. 1989. Structure destabilization and condensation of nucleic acids by intercalators. In Biological Structure, Dynamics, Interactions and Stereodynamics (R.H. Sarma and M.H. Sarma, eds.) pp. 267-281. Adenine Press, Schenectady, N.Y. Kapuscinski, J., Darzynkiewicz, Z., and Melamed, M.R. 1982. Luminescence of the solid complexes of acridine orange with RNA. Cytometry 2:201-211. Ladd, A.C., Pyatt, R., Gothot, A., Rice, S.J.M., Traycoff, C.M., and Srour, E.F. 1997. Orderly process of sequential cytokine stimulation is required for activation and maximal proliferation of primitive human bone marrow CD34+ hematopoietic progenitor cells residing in G0 . Blood 90:658-668. Portela, R.A. and Stockert, J.C. 1979. Chromatin fluorescence by pyronine staining. Experientia 35:1663-1665. Scott, J.E. 1967. On the mechanism of the methyl green–pyronin stain for nucleic acids. Histochemie 9:30-47.
Traganos, F., Crissman, H.A., and Darzynkiewicz, Z. 1988. Staining with pyronin Y detects changes in conformation of RNA during mitosis and hyperthermia of CHO cells. Exp. Cell Res. 179:535-544. Wilson, W.D. and Jones, R.L. 1982. Intercalation in biological systems. In Intercalation Chemistry (M.S. Whittingham and A.J. Jacobson, eds.) pp. 445-501. Academic Press, New York. Wolf, N.S., Kone, A., Priestley, G.V., and Bartelmez, S.H. 1993. In vivo and in vitro characterization of long-term repopulating primitive hematopoietic cells isolated by sequential Hoechst 33342-rhodamine 123 FACS selection. Exp. Hemat. 21:614-622.
Contributed by Zbigniew Darzynkiewicz and Gloria Juan New York Medical College Valhalla, New York Edward F. Srour Indiana University School of Medicine Indianapolis, Indiana
Differential Staining of DNA and RNA
7.3.16 Supplement 30
Current Protocols in Cytometry
Analysis of DNA Content and DNA Strand Breaks for Detection of Apoptotic Cells
UNIT 7.4
This unit describes a method for identifying apoptotic cells based on in situ detection of DNA strand breaks. This method is of special value in analysis of the cell cycle phase specificity of apoptosis. One of the most characteristic features of apoptosis is activation of an endonuclease that degrades nuclear DNA (Arends et al., 1990; Compton, 1992; Wyllie et al., 1992). DNA cleavage generates large numbers of DNA strand breaks. Their presence in the cell may thus serve as a marker of apoptosis. The 3′-OH termini of the strand breaks can be detected by attaching a fluorochrome to them via “tail labeling” with deoxynucleotides in a reaction catalyzed preferentially by exogenous terminal deoxynucleotidyltransferase (TdT) enzyme (Gorczyca et al., 1992, 1993a,b; Gold et al., 1994; Li and Darzynkiewicz, 1995). A variety of deoxynucleotides labeled directly or indirectly (e.g., via digoxigenin or biotin) with fluorochromes are commercially available and can be used for DNA strand break labeling. Of all the labeled deoxynucleotides tested, 5-bromodeoxyuridine triphosphate (BrdUTP) is the most advantageous with respect to sensitivity, low cost, and simplicity of the reaction (Li and Darzynkiewicz, 1995). This deoxynucleotide, once incorporated into DNA breaks, is detected by an FITC-conjugated antibody raised against 5-bromodeoxyuridine (BrdU). Unlike the detection of BrdU incorporated during DNA replication (Dolbeare and Selden, 1994), detection of BrdUTP attached to DNA strand breaks by TdT does not require prior DNA denaturation to make the precursor accessible to antibody. It should be stressed, however, that detection of DNA strand breaks requires prefixation of cells with a cross-linking agent such as formaldehyde, which, unlike ethanol used in other methods of detection of apoptosis (e.g., on the basis of DNA content; UNIT 7.5), prevents the extraction of degraded DNA. Thus, despite the sequential cell washings during the procedure, the DNA content of early apoptotic cells (and with it the number of DNA strand breaks and 3′-OH termini) is not markedly diminished compared to unfixed cells. Counterstaining of DNA with a fluorochrome of another color, distinct from the one used to detect DNA strand breaks, followed by bivariate analysis of the data, makes it possible to simultaneously measure DNA content and detect DNA strand breaks. Such an analysis makes it possible to detect apoptotic cells in relation to their position in the cell cycle (Gorczyca et al., 1993b). DNA STRAND BREAK LABELING WITH BrdUTP Cells subjected to DNA strand break labeling have to be prefixed in formaldehyde, a step that prevents extraction of the degraded, low-molecular-weight DNA from the cell, which otherwise occurs after cell permeabilization during the labeling and staining procedure. Short prefixation in 1% formaldehyde in this protocol is adequate; longer fixation times or higher formaldehyde concentrations impair subsequent DNA stainability with PI. After fixation in formaldehyde the cells are postfixed in 70% ethanol. They can be stored indefinitely in ethanol at −20°C. Exposure of the cells to ethanol results in their permeabilization to the reagents used for DNA strand break labeling. After removal of ethanol, DNA strand breaks are labeled with BrdUTP during incubation in the presence of terminal deoxynucleotidyltransferase (TdT). The 3′-OH ends in DNA strand breaks serve as primers, and the enzyme covalently attaches BrdU to each primer. Up to 20 to 30 molecules of BrdU can be incorporated per single break until the reaction rate plateaus. The presence of cobalt ions in the reaction makes the 3′-OH in the recessed ends of DNA breaks accessible to the enzyme (Tabor, 1995). Following incorporation of Contributed by Zbigniew Darzynkiewicz and Gloria Juan Current Protocols in Cytometry (1997) 7.4.1-7.4.8 Copyright © 1997 by John Wiley & Sons, Inc.
BASIC PROTOCOL
Nucleic Acid Analysis
7.4.1
BrdU into strand breaks the cells are incubated with FITC-conjugated BrdU monoclonal antibody (MAb). This results in strong green fluorescence labeling at the sites of DNA breaks. Cellular DNA is subsequently counterstained with PI in the presence of RNase. Finally, the cellular fluorescence, green from FITC (representing DNA strand break labeling) and red from PI (representing DNA stainability), is measured by flow cytometry. Materials Cells to be analyzed Phosphate-buffered saline (PBS; APPENDIX 2A), pH 7.4 1% (w/v) methanol-free formaldehyde (Polysciences) in PBS, pH 7.4 70% ethanol, ice cold TdT reaction buffer (see recipe) BrdUTP stock solution: 2 mM BrdUTP (Sigma; 100 nmol in 50 µl) in 50 mM Tris⋅Cl (APPENDIX 2A), pH 7.5, made freshly and protected from light 25 U/µl TdT in storage buffer (Boehringer Mannheim) 10 mM CoCl2 (Boehringer Mannheim) Rinsing buffer: 0.1% (v/v) Triton X-100 and 5 mg/ml BSA in PBS, pH 7.4 (can be stored at 4°C) FITC-conjugated anti-BrdU MAb solution (see recipe) PI staining buffer: 5 µg/ml propidium iodide (PI) and 200 µg/ml DNase-free RNase A (APPENDIX 2A) in PBS, pH 7.4, made freshly Silanized or polypropylene 15-ml conical tube Flow cytometer equipped with 488-nm argon laser or mercury arc lamp, with blue (BG 12) excitation filter (∼50% cut off at 470 nm) Additional reagents and equipment for trypsinizing cells (APPENDIX 3B) or dissociating cells from tissues (UNIT 5.2) Prepare cell suspension for fixation 1a. For cells in suspension cultures or hematologic samples: Rinse cells once with PBS and suspend in PBS to ∼106 cells/ml. 1b. For cells attached to tissue culture dishes: Collect cells from flasks or petri plates by trypsinization (APPENDIX 3B) and pool the trypsinized cells with cells floating in the medium. Rinse once with medium containing serum and suspend cells in PBS to ∼106 cells/ml. Cells floating in the medium consist mostly of detached mitotic and apoptotic cells. Serum is present in the medium during rinsing to inactivate the trypsin; other means of trypsin inactivation such as addition of protease inhibitors may also be used.
1c. For cells isolated from solid tumors (UNIT 5.2): Rinse free of any enzyme used for cell dissociation and suspend in PBS to ∼106 cells/ml. If the sample prepared in step 1 contains very few cells, which may be lost during repeated centrifugations (steps 3 to 8), carrier cells (e.g., chick red blood cells) may be included.
Prefix and fix samples 2. Fix 0.5 ml cells in suspension 15 min in 5 ml 1% methanol-free formaldehyde in PBS on ice.
Analysis of DNA Content and DNA Strand Breaks for Detection of Apoptotic Cells
CAUTION: Formaldehyde is a toxic reagent, and care should be exercised in its handling. Breathing formaldehyde vapors is harmful.
3. Centrifuge 5 min at 300 × g. Suspend cell pellet in 5 ml PBS, centrifuge as before, and resuspend cells (∼106 cells) in 0.5 ml PBS.
7.4.2 Current Protocols in Cytometry
4. Add the 0.5-ml aliquot of cell suspension to 5 ml ice-cold 70% ethanol in a silanized or polypropylene 15-ml conical tube. Fix cells ≥4 hr. The cells can be stored several weeks in ethanol at −20°C. To minimize cell loss, all following steps should be done in the same tube.
Label DNA strand breaks and stain samples 5. Centrifuge 5 min at 300 × g, remove ethanol, suspend cells in 5 ml PBS, and centrifuge as before. 6. Resuspend the pellet (≤106 cells) in 50 µl of a solution containing: 10 µl TdT reaction buffer 2.0 µl BrdUTP stock solution 0.5 µl (12.5 U) TdT in storage buffer 5 µl 10 mM CoCl2 33.5 µl distilled H2O. The terminal deoxynucleotidyltransferase from Boehringer Mannheim is supplied in a storage buffer and should be stored at −20°C. To maintain enzyme activity, repeated freezing and thawing of TdT should be avoided.
7. Incubate cells in the reaction mixture 40 min at 37°C, with occasional gentle shaking (e.g., once every ∼10 min). Alternatively and for best results, incubation can be carried out at 22° to 24°C overnight.
8. Add 1.5 ml rinsing buffer and centrifuge 5 min at 300 × g. 9. Add 100 µl FITC-conjugated anti-BrdU MAb solution. Incubate 1 hr at room temperature, occasionally shaking gently. 10. Add 1 ml PI staining solution containing RNase. Incubate 30 min at room temperature in the dark. Perform flow cytometry 11. Set up and adjust the flow cytometer for excitation with blue light (488-nm laser line or BG 12 excitation filter) and for detection of the green fluorescence of FITC-conjugated anti-BrdU MAb at 530 ± 20 nm and the red fluorescence of PI at >600 nm. 12. Run the sample; measure and record cell fluorescence. The data may be analyzed as illustrated in Figure 7.4.2.
OTHER FLUOROCHROMES AND STRATEGIES FOR LABELING DNA STRAND BREAKS DNA strand breaks can be labeled with a variety of deoxynucleotides tagged with other markers besides BrdUTP. For example, seven types of dUTP conjugates, including three BODIPY dyes, fluorescein, Cascade blue, Texas red, and dinitrophenol, are available from Molecular Probes. Cyanine dye conjugates (e.g., CY-3-dCTP) are available from Biological Detection Systems. These are direct deoxynucleotide-fluorochrome conjugates, differing in spectral properties of the fluorochromes. Likewise, indirect labeling, through the use of biotinylated or digoxigenin-conjugated deoxynucleotides, also offers a multiplicity of commercially available fluorochromes (fluorochrome-conjugated avidin or streptavidin, as well as digoxigenin antibodies) with different excitation and emission characteristics. It is possible, therefore, to label DNA strand breaks with a dye of any desired fluorescence color and excitation wavelength.
ALTERNATE PROTOCOL
Nucleic Acid Analysis
7.4.3 Current Protocols in Cytometry
The procedure described in the Basic Protocol is easily adapted to utilize other fluorochromes. For direct labeling, the fluorochrome-conjugated deoxynucleotide is included in the reaction solution (0.25 to 0.5 nmol per 50 µl) as described in step 6 instead of BrdUTP. Step 9 is then omitted, so that after the incubation and rinse (steps 7 and 8) the cells are stained directly with PI (step 10). For indirect labeling, instead of BrdUTP, digoxigenin- or biotin-conjugated deoxynucleotides are included in the reaction buffer (0.25 to 0.5 nmol per 50 µl) during step 6. The cells are then incubated either with fluorochrome-conjugated digoxigenin antibody (0.2 to 0.5 µg per 100 µl PBS containing Triton X-100 and BSA made as in the Basic Protocol) or with fluorochrome-conjugated avidin or streptavidin (0.2 to 0.5 µg per 100 µl, as above) during step 9 and then processed through step 10 as described in the Basic Protocol. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
FITC-conjugated anti-BrdU MAb solution 0.3 µg anti-BrdU fluorescein isothiocyanate (FITC)-conjugated MAb (Becton Dickinson) 0.3% (v/v) Triton X-100 1% (w/v) BSA PBS (APPENDIX 2A) to 100 µl Store ≤1 month at 4°C TdT reaction buffer, 5× 1 M potassium or sodium cacodylate 125 mM Tris⋅Cl, pH 6.6 (APPENDIX 2A) 1.25 mg/ml BSA Store ≤1 year at −20°C TdT reaction buffer contributes 10 ìl to each 50 ìl of reaction solution. The TdT reaction buffer is also available from Boehringer Mannheim as part of a kit containing TdT reaction buffer, TdT enzyme in storage buffer, and CoCl2.
COMMENTARY Background Information
Analysis of DNA Content and DNA Strand Breaks for Detection of Apoptotic Cells
Cleavage of DNA in the course of apoptosis occurs stepwise. Initially, DNA is cleaved at sites that are relatively distant from one another, at the points of attachment of DNA loops to nuclear matrix. This generates 50- to 300-kb DNA fragments in the cell, which, after DNA isolation, can be detected by pulsed-field gel electrophoresis (Oberhammer et al., 1993). Subsequently, DNA is cleaved at internucleosomal (linker) sections, which leads to the appearance of DNA fragments of nucleosomal (~180 bp) and oligonucleosomal size (Arends et al., 1990; Compton, 1992). Thus, for a relatively long period during apoptosis, there is heterogeneity of DNA, in terms of molecular weight, within the cell, because together with still undegraded DNA there are DNA fragments 50 to 300 kb in size as well as ~180 bp and multiples of 180 bp.
The low-molecular-weight DNA is extractable from the cells, e.g., after cell lysis or extensive washing even following fixation in ethanol. Extraction of low-molecular-weight DNA from the cell results in a loss of DNA strand breaks. To retain the maximum amount of low-molecular-weight DNA within apoptotic cells, and with it a large number of 3′-OH ends in DNA breaks, which serve as primers in the TdT reaction, the cells in the Basic Protocol are briefly prefixed in 1% formaldehyde, which cross-links DNA to other cell constituents. Postfixation in ethanol increases cell permeability to the reagents. Identification of apoptotic cells via DNA strand break labeling offers certain advantages over other approaches that rely on different principles (e.g., Nicoletti et al., 1991; Dive et al., 1992; Hotz et al., 1994; Koopman et al., 1994; reviewed by Darzynkiewicz et al., 1997).
7.4.4 Current Protocols in Cytometry
First of all the presence of numerous DNA strand breaks appears to be a very specific marker of apoptosis: the number of DNA strand breaks per genome in necrotic cells or in cells exposed to as much as 20 Gy radiation is lower by at least 1 order of magnitude than in apoptotic cells (Gorczyca et al., 1992). This uniquely high intensity of cell labeling in DNA strand break assays is a very specific marker in most cases of apoptosis. Another advantage, especially over methods that rely on analysis of plasma membrane transport, stems from the fact that cells to be assayed for DNA strand breaks can be fixed and stored prior to analysis. It extends applicability of this methodology to instances when cell analysis cannot be immediately done on live cells (e.g., for clinical or archival samples). Also, unlike analysis of plasma membrane transport, DNA strand break assays can be used for detection of apoptosis in tissue sections. The major advantage of this method, however, is the possibility of correlating the presence of DNA strand breaks with cellular DNA content in both apoptotic and nonapoptotic cell populations. By offering a unique opportunity to analyze the cell cycle position (and/or DNA ploidy) of cells undergoing apoptosis, the DNA strand break assay has found applications in studies of the cell cycle specificity of cytotoxic drugs (Gorczyca et al., 1993b). DNA strand break labeling may also be applied to identify DNA-replicating cells following BrdU incorporation into DNA and selective DNA photolysis by the SBIP (DNA strand break induction by photolysis) procedure (Li and Darzynkiewicz, 1995; Li et al., 1996).
Critical Parameters The technical problems of the procedure presented in this unit stem from the fact that DNA strand breaks are labeled in an enzymatic reaction. This requires the presence of active TdT enzyme and stringent ionic conditions during incubation. Repeated freezing and thawing of the enzyme stock solution or prolonged storage of BrdUTP is expected to result in loss of reagent utility. Positive and negative controls (see below) should be included to reveal whether the apparent absence of DNA strand break labeling in the studied material is due to methodological problems. Control cells, included with kits for DNA strand break labeling, are available from some vendors (e.g., Phoenix Flow Systems). Cell loss during repeated centrifugations is often a problem when the sample initially con-
tains an inadequate number of cells. To minimize cell loss, polypropylene or silanized glass tubes should be used, and all steps of the procedure (including fixation) should be done in the same tube. Transferring cells from one tube to another results in irreversible attachment of a large fraction of cells to the surface of each new tube. Addition of 1% (w/v) BSA to rinsing solutions also decreases cell loss. When the sample contains very few cells, carrier cells, which can later be recognized on the basis of differences in DNA content (e.g., chick erythrocytes), may be included. Controls and standards The procedure of DNA strand break labeling is rather complex and involves many reagents. Negative results, therefore, may not necessarily mean the absence of DNA strand breaks but may be a result of methodological problems, such as loss of TdT activity, degradation of BrdUTP, etc. It is necessary, therefore, to include positive and negative controls. An excellent control is a sample of HL-60 cells treated (during their exponential growth) for 3 to 4 hr with 0.2 µM of the DNA topoisomerase I inhibitor camptothecin (CPT). Because CPT induces apoptosis selectively during S phase, cells in G1 and G2/M may serve as negative control populations, while the S phase cells in the same sample represent the positive control (Fig. 7.4.1). Data interpretation The most common problem with data interpretation relates to the fact that in some cell types, or with particular inducers of apoptosis, the cells may lack one or more typical apoptotic features. In some instances DNA degradation may be incomplete, resulting in 50- to 300-kb fragments rather than progressing to internucleosomal fragmentation. The number of DNA strand breaks is then at least an order of magnitude lower compared with apoptosis in which DNA cleavage progresses to internucleosomal (linker) sections. Intensity of cell labeling in such a case is low, and the presence of DNA strand breaks may be overlooked if an exponential rather than linear scale of coordinates is used for data acquisition or display. It should be stressed, however, that the apoptotic mode of cell death should be confirmed (or excluded) on the basis of cell morphology. Morphological criteria for identifying apoptotic cells, by light or UV light microscopy, are presented in detail elsewhere (Darzynkiewicz et al., 1997).
Nucleic Acid Analysis
7.4.5 Current Protocols in Cytometry
DNA strand breaks
UV
Control
104
400 103
γ RAD
CPT
FST
G1 G2
G1
G1 S G2M
S
102 G2M
101
0
400 200
G1 S
G1
800 600
G2M
S
S
G1 S 0
10
G2M
G2M
0
400 0
1000 200
400 0 200
400 0 200
400 200
DNA content
Figure 7.4.1 Measurement of DNA content combined with detection of apoptotic cells identified by the presence of DNA strand breaks, demonstrating the analysis of cell sensitivity to apoptosis in relation to the cell cycle phase. Apoptosis of HL-60 cells was induced by exposure to UV light (UV), treatment with the DNA topoisomerase I inhibitor camptothecin (CPT), γ irradiation (γ RAD), or treatment with the DNA topoisomerase II inhibitor fostriecin (FST) as described (Gorczyca et al., 1993b; Darzynkiewicz et al., 1997). Populations of apoptotic cells are distinguished by the presence of DNA strand breaks. Note that UV light preferentially induces apoptosis of G1 cells, CPT of S phase cells, and γ irradiation of G2/M cells, whereas fostriecin is less selective, killing cells in all phases of the cycle. The left-hand panel shows untreated control cells, including their DNA content frequency histogram (inset). Separation of apoptotic and nonapoptotic cell populations by gating allows one to analyze the cell cyle distribution within each of these populations using software for deconvoluting DNA content frequency histograms.
Analysis of DNA Content and DNA Strand Breaks for Detection of Apoptotic Cells
Another common problem with data interpretation stems from the often-held assumption that the measured percentage of apoptotic cells (apoptotic index) represents the rate of cells dying by apoptosis. Apoptosis is of short and variable duration. The time interval during which apoptotic cells can be recognized varies depending on the method used, cell type, and nature of the inducer of apoptosis. Some inducers may either slow down or accelerate the apoptotic process by affecting the rate of formation and shedding of apoptotic bodies, the rate of endonucleolysis or proteolysis, as well as the activity of other apoptotic effectors. No method exists that allows one to estimate the rate of apoptosis, as there is in the case of mitosis (e.g., mitosis can be arrested by microtubule poisons, and the cumulative increase in mitotic index can be measured in the stathmokinetic experiment). The percentage of apoptotic cells in a cell population estimated by a given method, therefore, may not necessarily represent the rate of cells dying by apoptosis. However, the kinetics of cell death can be estimated when (in parallel to measuring the apoptotic index) an absolute number of live cells is measured in the control and in the drug-treated
culture, together with the rate of cell proliferation. The latter may be obtained from the rate of cell entrance to mitosis (“cell birth rate” estimated, e.g., in the stathmokinetic experiment; Darzynkiewicz, 1993). The observed difference between the actual number of live cells and the expected number of cells estimated based on the cell birth rate provides a measure of cumulative cell loss (death) over a given interval. Apoptotic cells detach from the surface of culture flasks and float in the medium. The standard procedure of discarding the medium and treating the attached cells with trypsin or EDTA, therefore, results in selective loss of apoptotic cells and cannot be used for quantitative analysis of apoptosis. To estimate the apoptotic index in cultures of adherent cells, floating cells have to be pooled with the trypsinized ones and measured together. Likewise, density separation methods (e.g., FicollHypaque or Percoll gradients) also may result in selective loss or enrichment of apoptotic cells (Gorman et al., 1996).
Anticipated Results Identification of apoptotic cells is rather simple due to their intense labeling with FITC-
7.4.6 Current Protocols in Cytometry
DNA strand breaks
1000
A
B
C
800 600
Ap
Ap
Ap
400 G1 G2M S
200
L
L
L
0 0
400 200
800 600
0
400
1000 200
800 600
0
1000 200
400
800 600
1000
DNA content
Figure 7.4.2 Detection of apoptosis during treatment of leukemia. Apoptotic cells (Ap) characterized by the presence of DNA strand breaks were detected in the peripheral blood of a patient with acute myelogenous leukemia (AML), prior to (A) and 2 days (B) and 4 days (C) after treatment with the DNA topoisomerase I inhibitor topotecan. Populations of apoptotic cells are very heterogeneous with respect to the number of DNA strand breaks. Cells with low FITC (green) fluorescence (L; within the trapezoid window) represent nonapoptotic cell populations. Note that most apoptotic cells have a DNA content equivalent to that of G1 cells.
conjugated anti-BrdU monoclonal antibody, which frequently requires use of the exponential scale (logarithmic amplifiers of the flow cytometer) for data acquisition and display (Fig. 7.4.1). Because the cellular DNA content of both apoptotic and nonapoptotic cell populations is measured, the Basic Protocol offers the unique possibility of analyzing the cell cycle position and/or DNA ploidy of both populations. The method is useful for clinical material, for example, in leukemias, lymphomas, and solid tumors (Fig. 7.4.2; Gorczyca et al., 1993a), and can be combined with surface immunophenotyping. In the latter case, the cells are first immunophenotyped, then fixed with formaldehyde (which stabilizes bound antibody on the cell surface), and subsequently subjected to the DNA strand break detection assay using differently colored fluorochromes thaw those used for immunophenotyping.
Time Considerations The cell fixation procedure (formaldehyde followed by ethanol) takes ∼25 min, but cells have to be stored in fixative (ethanol) for at least 4 hr. The DNA strand break labeling and DNA staining procedure takes at least 3 hr. Overnight incubation with anti-BrdU antibody, however, gives more satisfactory results.
Literature Cited Arends, M.J., Morris, R.G., and Wyllie, A.H. 1990. Apoptosis: The role of endonuclease. Am. J. Pathol. 136:593-608. Compton, M.M. 1992. A biochemical hallmark of apoptosis: Internucleosomal degradation of the genome. Cancer Metast. Rev. 11:105-119. Darzynkiewicz, Z. 1993. Mammalian cell-cycle analysis. In The Cell Cycle: A Practical Approach (P. Fantes and R. Brooks, eds.) pp. 45-68. IRL Press, Oxford. Darzynkiewicz, Z., Juan, G., Li, X., Gorczyca, W., Murakami, T., and Traganos, F. 1997. Cytometry in cell necrobiology. Analysis of apoptosis and accidental cell death (necrosis). Cytometry 27:120. Dive, C., Gregory, C.D., Phipps, D.J., Evans, D.L., Milner, A.E., and Wyllie, A.H. 1992. Analysis and discrimination of necrosis and apoptosis (programmed cell death) by multiparameter flow cytometry. Biochim. Biophys. Acta 1133:275285. Dolbeare, F. and Selden, J.R. 1994. Immunochemical quantitation of bromodeoxyuridine: Application to cell kinetics. Methods Cell Biol. 41:297316. Gold, R., Schmied, M., Giegerich, G., Breitschopf, H., Hartung, H.P., Toyka, K.V., and Lassman, H. 1994. Differentiation between cellular apoptosis and necrosis by the combined use of in situ tailing and nick translation techniques. Lab. Invest. 71:219-225. Gorczyca, W., Bruno, S., Darzynkiewicz, R.J., Gong, J., and Darzynkiewicz, Z. 1992. DNA strand breaks occurring during apoptosis: Their early in situ detection by the terminal deoxynu-
Nucleic Acid Analysis
7.4.7 Current Protocols in Cytometry
cleotidyl transferase and nick translation assays and prevention by serine protease inhibitors. Int. J. Oncol. 1:639-648. Gorczyca, W., Gong, J., Ardelt, B., Traganos, F., and Darzynkiewicz, Z. 1993b. The cell cycle related differences in susceptibility of HL-60 cells to apoptosis induced by various antitumor agents. Cancer Res. 53:3186-3192. Gorczyca, W., Bigman, K., Mittelman, A., Ahmed, T., Gong, J., Melamed, M.R., and Darzynkiewicz, Z. 1993a. Induction of DNA strand breaks associated with apoptosis during treatment of leukemias. Leukemia 7:659-670. Gorman, A., McCarthy, J., Finucane, D., Reville, W., and Cotter, T. 1996. Morphological assessment of apoptosis. In Techniques in Apoptosis: A User’s Guide (T.G. Cotter and S.J. Martin, eds.) pp. 3-21. Portland Press, London. Hotz, M.A., Gong, J., Traganos, F., and Darzynkiewicz, Z. 1994. Flow cytometric detection of apoptosis. Comparison of the assays of in situ DNA degradation and chromatin changes. Cytometry 15:237-244.
Li, X., Melamed, M.R., and Darzynkiewicz, Z. 1996. Detection of apoptosis and DNA replication by differential labeling of DNA strand breaks with fluorochromes of different color. Exp. Cell Res. 222:28-37. Nicoletti, I., Migliorati, G., Pagliacci, M.C., Grignani, F., and Riccardi, C. 1991. A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry. J. Immunol. Methods 139:271-280. Oberhammer, F., Wilson, J.M., Dive, C., Morris, I.D., Hickman, J.A., Wakeling, A.E., Walker, P.R., and Sikorska, M. 1993. Apoptotic death in epithelial cells: Cleavage of DNA to 300 and/or 50 kb fragments prior to or in the absence of internucleosomal fragmentation. EMBO J. 12:3679-3684. Tabor, S. Template-independent DNA polymerases. 1995. In Short Protocols in Molecular Biology, 3rd ed. (F.A. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 3.23-3.24. John Wiley & Sons, New York.
Koopman, G., Reutelingsperger, C.P.M., Kuijten, G.A.M., Keehnen, R.M.J., Pals, S.T., and van Oers, M.H.J. 1994. Annexin V for flow cytometric detection of phosphatidylserine expression of B cells undergoing apoptosis. Blood 84:14151420.
Wyllie, A.H., Arends, M.J., Morris, R.G., Walker, S.W., and Evan, G. 1992. The apoptosis endonuclease and its regulation. Semin. Immunol. 4:389-398.
Li, X. and Darzynkiewicz, Z. 1995. Labelling DNA strand breaks with BrdUTP: Detection of apoptosis and cell proliferation. Cell. Prolif. 28:571579.
Contributed by Zbigniew Darzynkiewicz and Gloria Juan New York Medical College Valhalla, New York
Analysis of DNA Content and DNA Strand Breaks for Detection of Apoptotic Cells
7.4.8 Current Protocols in Cytometry
DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
UNIT 7.5
In flow cytometry, analysis of DNA ploidy (DNA index or DI) and/or discrimination of cells in G0/G1 versus S versus G2/M phases of the cell cycle is generally done by measuring cellular DNA content alone. Indeed, univariate DNA content analysis is an established clinical assay in oncology and is also widely used for research in cell and molecular biology (see UNIT 7.1 for an overview of nucleic acid analysis). A large number of DNA fluorochromes can be used for this purpose, and the binding characteristics and spectral properties of most nucleic acid probes are described in UNIT 4.2. A great variety of techniques for measuring DNA utilizing these fluorochromes have been developed since the mid 1970s (Darzynkiewicz et al., 1994). The techniques differ primarily in the mode of cell permeabilization (detergent versus prefixation with alcohols) and composition of the stain solution. The most commonly used procedures are described in this unit. Relatively simple and universally applicable methods for staining ethanol-fixed cells are presented in Basic Protocol 1 and Alternate Protocol 1. Because cells may be stored in fixative for extended periods and may be transported while in the fixative, this method allows one to prepare and collect cells independently of the timing of their analysis. The methods presented utilize the two most widely used DNA fluorochromes, propidium iodide (PI; see Basic Protocol 1) and 4′,6-diamidino-2-phenylindole (DAPI; see Alternate Protocol 1). The second set of methods presented (see Basic Protocol 2 and Alternate Protocol 2) utilize detergents and/or hypotonic solutions to permeabilize cells; these methods generally provide more accurate estimates of DNA content compared to measurement of fixed cells. The approach presented in Basic Protocol 2 combines detergent treatment with use of proteolytic enzymes; it is widely used for clinical material, especially for DNA analysis in samples of solid tumors. Alternate Protocol 2 is a simpler method designed for uniform populations (e.g., tissue culture cells). A third approach is required for DNA content measurements in live cells (see Basic Protocol 3). The primary application of this method is for cell sorting, where cells selected on the basis of differences in DNA content can be subcultured for the purpose of analyzing their growth characteristics, testing their sensitivity to drugs, cloning, or expanding their number. On the other hand, archival samples of paraffin-embedded tissues can be analyzed by flow cytometry following nuclear isolation (see Basic Protocol 4). This methodology is widely used in retrospective studies probing the prognostic value of DNA content in tumors. Univariate DNA content measurement can also discriminate apoptotic cells, which are characterized by fractional DNA content due to DNA degradation by the apoptosis-associated endonuclease(s) (Wyllie, 1992; Darzynkiewicz et al., 1997). Apoptotic cells can therefore be identified within a population as the cells that evidence fractional DNA content following extraction of the degraded DNA and subsequent cell staining with PI or DAPI (see Basic Protocol 5). This approach is combined with analysis of DNA degradation by gel electrophoresis (see Support Protocol). The characteristic pattern of DNA degradation, preferentially producing internucleosomal DNA sections and generating so-called DNA laddering on the gels, is considered to be a hallmark of apoptosis. An alternative approach to detecting apoptotic cells by simultaneously measuring DNA content and DNA strand breaks, in which formaldehyde-fixed cells are enzymatically labeled with 5-bromodeoxyuridine triphosphate (BrdUTP) and then costained with Contributed by Zbigniew Darzynkiewicz and Gloria Juan Current Protocols in Cytometry (1997) 7.5.1-7.5.24 Copyright © 1997 by John Wiley & Sons, Inc.
Nucleic Acid Analysis
7.5.1
FITC-conjugated antibody to BrdU (to detect DNA strand breaks) and with PI (to analyze DNA content), is presented in UNIT 7.4. Discrimination of cells in particular phases of the cell cycle on the basis of differences in DNA content is helped by computer analysis. The software used to deconvolute the histograms often allows one to measure the cell cycle distribution of both diploid normal cells (host infiltrating and stromal cells) and aneuploid cell populations in aneuploid tumors. The protocols presented in this unit can easily be modified for use of other DNA fluorochromes. For example, PI, which is used in Basic Protocols 1, 2, and 5, can be replaced by DAPI (e.g., see Alternate Protocols 1 and 2; see Basic Protocol 4), which then is applied at lower concentration and without the need to incubate cells with RNase. BASIC PROTOCOL 1
DNA CONTENT ANALYSIS OF FIXED CELLS WITH PROPIDIUM IODIDE This protocol uses ethanol to fix and permeabilize cells, aiding access of dye to DNA in intact cells and allowing DNA content analysis of stained cells by flow cytometry. The use of a fixation step makes this protocol applicable in situations when samples have to be stored before the analysis. The fixed cells are rinsed with PBS and then stained with the DNA fluorochrome PI in a solution containing Triton X-100 as well as RNase A. Flow cytometry requires excitation with blue light and detection of PI emission at red wavelengths. Alternate Protocol 1 uses DAPI (which requires UV excitation) instead of PI. Materials 70% ethanol Cells to be stained Phosphate-buffered saline (PBS; APPENDIX 2A) Propidium iodide (PI)/Triton X-100 staining solution with RNase A (see recipe), freshly made 12 × 75–mm centrifuge tubes, preferably polypropylene or silanized Beckman TJ rotor or equivalent Flow cytometer with 488-nm argon ion laser or mercury arc lamp as fluorescence excitation source Additional reagents and equipment for preparing cell suspensions for fixation (UNIT 7.4) Fix cells with ethanol 1. Prepare the fixative by filling 12 × 75 mm–centrifuge tubes with 4.5 ml of 70% ethanol. Keep tubes on ice. 2. Collect cells (UNIT 7.4) and suspend 106 to 107 cells in 5 ml PBS in a centrifuge tube. 3. Centrifuge cells 6 min at ∼200 × g (e.g., 1000 rpm in Beckman TJ rotor). 4. Using a Pasteur pipette thoroughly resuspend cells in 0.5 ml PBS. It is important to achieve a single-cell suspension. Fixation of cells that are in aggregates while suspended in PBS stabilizes the aggregates, which are then impossible to disperse. It is essential, therefore, to have a monodisperse cell suspension at the time of mixing cells with ethanol.
DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
5. Transfer the cell suspension into the tubes containing 70% ethanol. Keep cells in fixative ≥2 hr. Cells suspended in 70% ethanol can be stored at 0° to −40°C for several months if not years.
7.5.2 Current Protocols in Cytometry
Stain cells with PI 6. Centrifuge the ethanol-suspended cells 5 min at 200 × g. Decant ethanol thoroughly. 7. Suspend the cell pellet in 5 ml PBS, wait 60 sec, and centrifuge 5 min at 200 × g. 8. Suspend cell pellet in 1 ml PI/Triton X-100 staining solution with RNase A. Keep either 15 min at 37°C or 30 min at room temperature. Perform flow cytometry 9. Set up and adjust the flow cytometer for excitation with blue light and detection of PI emission at red wavelengths. For excitation, the 488-nm argon ion laser line may be used. Alternatively, use a BG 12 optical filter when the source of illumination is mercury arc or xenon lamp. A long-pass (>600 nm) filter is recommended for detecting PI emission.
10. Measure cell fluorescence in the flow cytometer. Use the pulse width–pulse area signal to discriminate between G2 cells and cell doublets and gate out the latter. Analyze the data (Fig. 7.5.1) using DNA content frequency histogram deconvolution software. DNA CONTENT ANALYSIS OF FIXED CELLS WITH DAPI This protocol is similar to Basic Protocol 1, except the cells are stained with DAPI rather than PI. Thus, there is no need for cell treatment with RNase. Excitation of DAPI, however, requires a UV light source, which is not universally available. Emission of DAPI is measured at blue wavelengths.
ALTERNATE PROTOCOL 1
Additional Materials (also see Basic Protocol 1) Cells to be stained DAPI/Triton X-100 staining solution (see recipe), freshly made Flow cytometer with UV light illumination source (e.g., mercury arc lamp, laser tuned to 340 to 380 nm) Fix cells with ethanol 1. Fix cells in 70% ethanol (see Basic Protocol 1, steps 1 to 5). Stain cells with DAPI 2. Centrifuge the ethanol-suspended cells 5 min at 200 × g. Decant ethanol thoroughly. 3. Suspend the cell pellet in 5 ml PBS, wait 60 sec, and centrifuge again, 5 min at 200 × g. 4. Suspend cell pellet in 1 ml DAPI/Triton X-100 staining solution. Keep 30 min in the dark. Perform flow cytometry 5. Set up and adjust flow cytometer for UV excitation at 340 to 380 nm and detection of DAPI emission at blue wavelengths. For excitation, an UG 1 optical filter (short-pass, 390 nm) may be used when the light source is a mercury arc or xenon lamp. For detecting DAPI, a band-pass filter at 470 ± 20 nm is recommended.
6. Measure cell fluorescence in the flow cytometer. Use the pulse width–pulse area signal to discriminate between G2 cells and cell doublets and gate out the latter. Analyze the data (Fig. 7.5.1) using software that deconvolutes DNA content frequency histograms.
Nucleic Acid Analysis
7.5.3 Current Protocols in Cytometry
BASIC PROTOCOL 2
DNA CONTENT ANALYSIS OF SAMPLES UTILIZING DETERGENTS AND TRYPSIN This protocol uses detergent-based lysis and staining solutions, which improve DNA content analysis by flow cytometry compared to staining of intact cells as in Basic Protocol 1 and Alternate Protocol 1. This protocol is suitable for tissue samples, whereas Alternate Protocol 2 provides a simplified detergent-based method designed for fixed cells or cells from suspension cultures. Cells are collected via aspiration from tissue samples into a sucrose-citrate buffer that contains DMSO, which allows for long-term storage of samples if needed. After samples are supplemented with an internal DNA standard (a mixture of chicken and trout erythrocytes), cells are lysed, digested with trypsin, and stained with PI. The stained nuclei are then subjected to flow cytometry. Materials Tissue sample with cells to be stained (e.g., resected tumor) Citrate/DMSO buffer (see recipe) Internal DNA content standard (e.g., chicken or trout erythrocytes; see recipe) Cell lysis solution with trypsin (see recipe) Trypsin-inactivating solution (see recipe), ice cold Propidium iodide (PI)/spermine staining solution (see recipe), ice cold 0.5 × 25–mm needles (25 G × 1 in.) 10-ml disposable syringes 38 × 12.5–mm polypropylene tubes with caps 24- to 34-µm nylon mesh Flow cytometer with 488-nm argon ion laser fluorescence excitation source Collect cells from tissue samples 1. Insert the needle into the tissue sample and apply suction by syringe. Move the needle several times forward and back in different directions. Do not draw the aspirate into the syringe: keep the material within the needle. The 1-in. needle should be adequate for collecting cells in vitro from resected tumors. Longer needles may be needed for aspirating cells in vivo. Alternatively, specimens may be frozen in dry ice, stored at −60° to −80°C, thawed to room temperature, and then aspirated.
2. Flush the aspirate plug with 200 µl citrate/DMSO buffer into a 38 × 12.5–mm polypropylene tube. The tubes may be capped and mechanically shaken to further disperse the cells.
3. Count the cells in a hemacytometer. Repeat aspirations (steps 1 and 2) until 106 cells per sample are obtained. The samples may be analyzed the same day (steps 5 to 11) or stored at −40° to −80°C for up to several years. The DMSO in the citrate cell collection buffer protects cells during prolonged storage.
Lyse cells and stain with PI 4. Thaw stored samples in a 37°C water bath and transfer to room temperature. DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
5. Add the internal DNA content standard. The mixture of chicken and trout erythrocytes used as the internal DNA content standard (chicken-to-trout cell ratio 3:5), which may be thawed to room temperature in a 37°C water
7.5.4 Current Protocols in Cytometry
bath after retrieval from −40° to −80°C storage, if necessary, is added to the 200 ìl sample of tumor cells in a proportion of ∼1:5 with respect to the number of studied cells.
6. Add 1.8 ml cell lysis solution with trypsin to 200 µl of tumor aspirate cells in citrate/DMSO buffer. Cap the tubes and incubate 10 min at room temperature, mixing gently by inverting the tubes five to ten times. 7. Add 1.5 ml ice-cold trypsin-inactivating solution. Mix gently 10 min at room temperature. 8. Add 1.5 ml ice-cold PI/spermine staining solution. Wrap tubes in aluminum foil or keep them in the dark, on ice. Mix gently and filter the sample through 24 to 34-µm nylon mesh to remove debris and cell aggregates. 9. Store cells on ice. Measure cell fluorescence between 20 min and 4 hr after adding the staining solution. Perform flow cytometry 10. Set up and adjust flow cytometer to provide excitation with blue light and detection of PI emission at red wavelengths. For excitation, the 488-nm argon ion laser line can be used. Use a BG 12 optical filter (short-pass, 470 nm) when the source of illumination is mercury arc or xenon lamp. For detecting emission of PI, a long-pass (>600 nm) filter is recommended.
11. Measure cell fluorescence in a flow cytometer. Use the pulse width–pulse area signal to discriminate between G2 cells and cell doublets and gate out the latter. Analyze the data (Fig. 7.5.1) using software that deconvolutes DNA content frequency histograms. DNA CONTENT ANALYSIS OF SAMPLES UTILIZING DETERGENTS This protocol uses detergent only, rather than detergent with trypsin as in Basic Protocol 2, to lyse cells and aid staining of DNA for flow cytometric analysis. In this simplified method, cells in suspension are mixed with staining solution that contains DAPI, buffers, and Triton X-100; the DNA content of the stained nuclei is then measured by flow cytometry, using UV excitation. This method also allows simultaneous analysis of DNA and protein if the protein-specific dye sulforhodamine 101 is included in the staining solution.
ALTERNATE PROTOCOL 2
Additional Materials (see also Basic Protocol 2) Cells to be stained DAPI/PIPES staining solution (see recipe) Flow cytometer with UV light illumination source (e.g., mercury arc lamp, laser tuned to 340 to 380 nm) Additional reagents and equipment for preparing cell suspensions from tissue cultures (UNIT 7.4) Stain cells with DAPI 1. Mix 0.2 ml cell suspension (105 to 106 cells) with 2 ml DAPI/PIPES staining solution. Keep the sample ≥10 min on ice. Cells used in this protocol may be collected directly from tissue culture flasks or plates (UNITS 7.3 & 7.4) and suspended in PBS to ∼5 × 106 cells/ml. In addition, cells may be fixed ≥2 hr in 70% ethanol (see Basic Protocol 1), then rinsed and resuspended in 0.2 ml PBS. The advantage of ethanol fixation is that it offers the possibility of sample storage or transport prior to analysis.
Nucleic Acid Analysis
7.5.5 Current Protocols in Cytometry
The DAPI/detergent staining solution may be supplemented with sulforhodamine 101 (0.02 mg/ml final concentration in the staining solution) to allow simultaneous measurement of DNA and protein. Fixed cells should be used for analysis of protein content. Regardless of the dye(s) used, cell fluorescence should be measured within 10 to 60 min after staining.
Perform flow cytometry 2. Set up the flow cytometer for UV excitation at 340 to 380 nm and detection of DAPI fluorescence at blue wavelengths. For UV excitation, use an UG 1 optical filter when the source of excitation is mercury arc or xenon lamp. For detecting DAPI emission, a band-pass filter at 470 ± 20 nm is recommended. Fluorescence of sulforhodamine (which like DAPI also is excited with UV light) is at red wavelengths > 600 nm.
3. Measure cell fluorescence in a flow cytometer within 60 min of staining. Use the pulse width–pulse area signal to discriminate between G2 cells and cell doublets and gate out the latter. Analyze the data (Fig 7.5.1) using software which deconvolutes DNA content frequency histograms (see Chapter 10). BASIC PROTOCOL 3
SUPRAVITAL STAINING OF DNA Supravital staining of DNA offers the possibility of sorting of live cells on the basis of differences in their DNA content. This protocol uses Hoechst 33342 to measure DNA by flow cytometry in live cells. The actual procedure of cell staining is simple. Cells suspended in culture medium are incubated in the presence of 2.0 to 5.0 µg/ml Hoechst 33342 for 20 to 90 min. Cell fluorescence is then measured directly, without any additional treatments, centrifugations, etc. Materials Cells to be stained, 106 cells/ml suspended in tissue culture medium 1 mg/ml Hoechst 33342 staining solution (see recipe) Flow cytometer with UV light illumination source Stain cells with Hoechst 33342 1. Add Hoechst 33342 staining solution to cells suspended in tissue culture medium (106 cells/ml) to obtain a final dye concentration of 2 µg/ml. Incubate 20 min at 37°C. Perform flow cytometry 2. Set up and adjust flow cytometer for UV excitation at 340 to 380 nm and detection of Hoechst 33342 at blue wavelengths. An UG 1 optical filter may be used when the source of excitation is a mercury arc or xenon lamp. For detecting the blue fluorescence of Hoechst 33342, a band-pass filter at 470 ± 20 nm is recommended.
3. Measure cell fluorescence in the flow cytometer. Use the pulse width–pulse area signal to discriminate between G2 cells and cell doublets and gate out the latter. When intensity of cell fluorescence or resolution of cells in the cell cycle phases is inadequate, prolong the staining time (up to 90 min) and/or increase the dye concentration in the medium (up to 5 ìg/ml). The same sample may be reanalyzed after prolonged incubation and/or addition of more staining solution. DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
This protocol is predominantly used for sorting live cells. However, because sensitivity of cells to Hoechst 33342 varies depending on the cell type, it is possible that viability and cell cycle sorted progression of cells may be affected by the staining procedure.
7.5.6 Current Protocols in Cytometry
DNA CONTENT ANALYSIS OF PARAFFIN-EMBEDDED SAMPLES This protocol describes DNA content analysis of archival samples embedded in paraffin blocks. The technique is based on preparation of thick microtome sections of the paraffin-embedded material, solubilization and extraction of paraffin from the sections, tissue rehydration in graded ethanols, and isolation of nuclei by proteolytic digestion of the tissue. Samples are then stained with DAPI and subjected to flow cytometry.
BASIC PROTOCOL 4
Materials Paraffin-embedded tissue blocks Xylene or xylene substitute (e.g., Histo-Clear, National Diagnostics) 100%, 95%, 80%, and 50% ethanol Protease solution (see recipe), freshly made 0.15 M NaCl, for diluting nuclei if needed DAPI/phosphate staining solution (see recipe), freshly made Microtome 57-µm nylon mesh bags, 1 × 1 cm 1- to 2-mm-diameter glass beads Phase-contrast or differential interference–contrast (Nomarski optics) microscope Flow cytometer with UV light illumination source Prepare paraffin sections 1. Cut a standard thin section (5 to 10 µm from the paraffin-embedded tissue block), adjacent to the subsequent section that will be subjected to nuclear isolation. Process by routine hematoxylin and eosin (HE) staining. 2. Examine the thin section by light microscopy and select the area (e.g., tumor site) to be processed by flow cytometry. On the basis of examination of the thin section, with a scalpel trim the block from the undesired tissue. 3. Mount the paraffin block on a microtome. Cut sections 50 to 100 µm thick. The sections may curl up as they come from the microtome knife. Depending on the size (area) of the section and the cell density in the tissue, one to four thick sections are generally adequate for DNA analysis.
Isolate cell nuclei from paraffin sections 4. With forceps transfer the tissue sections into 1 × 1–cm fine mesh (57 µm) nylon bags. Add one or two 1- to 2-mm glass beads to prevent floating of bags on the surface of solutions used in subsequent steps. 5. Immerse bags in 20 ml xylene or xylene substitute and mix 60 min on a slowly rotating shaker at room temperature. CAUTION: Xylene is toxic. Wear gloves and keep lids on jars. When possible, xylene should be substituted by less toxic reagents such as Histo-Clear. Keep xylene and ethanol solutions in aliquots of 20 ml in closed glass or plastic, xylene-resistant containers (e.g., Coplin jars or Erlenmeyer flasks).
6. Drain xylene and transfer the bags with sections to 20 ml of 100% ethanol. Keep 10 min at room temperature. Successively transfer the bags to 95%, 80%, and 50% ethanol, keeping the bags 20 min in each solution. 7. Transfer each bag to a separate 15-ml tube containing distilled water. Keep 30 min at room temperature. Repeat rinse with water. Separation of bags from one another at this stage is necessary because with rehydration the tissue becomes soft and breaks up, which may cause cross-contamination of samples. Cross-contamination may be avoided by using bags made of fine nylon mesh (<4 ìm).
Nucleic Acid Analysis
7.5.7 Current Protocols in Cytometry
8. Transfer the bags into small (5 ml) tubes containing 1 ml protease solution. With scissors, cut through the nylon mesh and release the contents of the bag into the protease solution. Leave the opened bag in the tube to release nuclei from pieces of tissue that may remain attached to the bag. Incubate 30 min in water bath at 37°C on rotary shaker. 9. After incubation examine the nuclear suspension with a phase-contrast or differential interference–contrast (Nomarski optics) microscope. If isolation of nuclei is inadequate, prolong the incubation in protease solution up to 2 hr. Count the number of isolated nuclei with a hemacytometer. If the number of isolated nuclei per 1-ml sample exceeds 106, adjust the number by removing a portion of the sample and diluting the remainder to 1 ml with 0.15 M NaCl. Stain cells with DAPI 10. Add 1 ml of DAPI/phosphate staining solution. Store ≥60 min on ice prior to the measurement; the nuclei can be stored overnight at 0° to 4°C. Filter through 57-µm nylon mesh before analyzing by flow cytometry. The final DAPI concentration is 2 ìg/ml. Although DAPI is recommended for staining DNA in nuclei isolated from paraffin blocks, PI can be used instead. To stain with PI, add RNase A (100 ìg/ml) at step 7 and add 1 ml PI solution (10 ìg/ml in PBS), instead of DAPI, at step 10.
Perform flow cytometry 11. Set up and adjust flow cytometer for UV excitation at 340 to 380 nm and for detection of DAPI emission at blue wavelengths. For excitation, an UG 1 optical filter may be used when the source of excitation is a mercury arc or xenon lamp. For detecting the emission of DAPI, a band-pass filter at 470 ± 20 nm is recommended.
12. Measure cell fluorescence with the flow cytometer. Use the pulse width–pulse area signal to discriminate between G2 cells and cell doublets and gate out the latter. Analyze the data using software which deconvolutes DNA content frequency histograms (Chapter 10). Note that because cell fixation and paraffin embedding after DNA stainability, the external standard of DNA content (DI), e.g., normal lymphocytes or chick erthrocytes, cannot be used. A subpopulation of cells with distinctly lower fluorescence is generally considered to be representative of diploid cells (DI 1.0). BASIC PROTOCOL 5
DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
DNA CONTENT ANALYSIS FOR DETECTION OF APOPTOTIC CELLS DNA content measurement with flow cytometry can be used to detect apoptotic cells, which have diminished DNA content. In this protocol, cells are fixed in ethanol before being subjected to mild extraction of low-molecular-weight DNA that leaks from the cells. Samples are then stained with PI in the presence of RNase A and analyzed by flow cytometry. Analysis of apoptosis as presented in this protocol can be combined with electrophoretic analysis of low-molecular-weight DNA extracted from fixed cells by the phosphate-citrate buffer; for description of such a procedure for agarose gel electrophoresis, see Support Protocol. Combining the methods presented in this protocol and the Support Protocol enables one to simultaneously analyze the molecular weight of DNA extracted from the very same cells that are subjected to flow cytometry.
7.5.8 Current Protocols in Cytometry
Materials 70% ethanol Cells to be analyzed Phosphate-buffered saline (PBS; APPENDIX 2A) DNA extraction buffer: 0.2 M phosphate-citrate buffer, pH 7.8 (see recipe) Propidium iodide (PI) Triton X-100 staining solution with RNase A (see recipe), freshly made 15-ml polypropylene centrifuge tubes and caps Flow cytometer with 488-nm argon ion laser fluorescence excitation source Fix cells in ethanol 1. Distribute 10-ml aliquots of 70% ethanol into 15-ml polypropylene centrifuge tubes. Keep tubes in ice. 2. Suspend 1–5 × 106 cells in 1 ml PBS. Fix cells in suspension by rapidly admixing, with a Pasteur pipette, 1 ml cell suspension into 10 ml of 70% ethanol in centrifuge tubes on ice. Fix cells ≥2 hr. Cells can be stored in fixative at −20°C for several weeks.
Extract low-molecular-weight DNA from cells 3. Centrifuge cells 5 min at 200 × g. Thoroughly decant ethanol. Add 50 µl DNA extraction buffer. Transfer tubes to a 37°C water bath, cap, and incubate 30 min on the shaker. The volume of DNA extraction buffer may vary. If the extracted DNA will be subjected to gel electrophoresis (see Support Protocol), small volumes (∼50 ìl) should be used. This simplifies subsequent steps in the gel electrophoresis procedure: such small volumes may be directly incubated with RNase and proteinase K and loaded on the gel without the necessity of concentrating the DNA. The efficiency of DNA extraction in step 3 should be controlled for optimal separation of apoptotic cells. If DNA degradation within apoptotic cells is extensive or if the cells have already shed apoptotic bodies, there is no need to extract low-molecular-weight DNA in step 3, as the apoptotic cells will already have a significantly reduced DNA content and will be well resolved on DNA content frequency histograms. On the other hand, if DNA degradation is incomplete and sub-G1 and G1 peaks are not separated, try extending the rinsing times (e.g., up to 2 hr) and using greater volumes of DNA extraction buffer (e.g., up to 500 ìl).
4. Centrifuge cells 10 min at 1500 × g. Reserve supernatant for analysis of low-molecular-weight DNA by agarose gel electrophoresis (see Support Protocol), if desired. Stain cells with PI 5. Resuspend cells in 1 ml PI staining solution with RNase A. Keep 30 min at room temperature, protected from light. Perform flow cytometry 6. Set up and adjust flow cytometer for excitation with blue light and detection of PI fluorescence at red wavelengths. For excitation, the 488-nm argon ion laser line may be used; a BG 12 optical filter is recommended when the source of illumination is a mercury arc or xenon lamp. For detecting PI emission, a long-pass (>600 nm) filter is recommended.
7. Measure cell fluorescence in a flow cytometer. Use the pulse width–pulse area signal to discriminate between G2 cells and cell doublets and gate out the latter. Analyze the data as shown in Fig. 7.5.4.
Nucleic Acid Analysis
7.5.9 Current Protocols in Cytometry
Supplement 11
SUPPORT PROTOCOL
AGAROSE GEL ELECTROPHORESIS OF DNA EXTRACTED FROM APOPTOTIC CELLS In this protocol, low-molecular-weight DNA, extracted from the same cells that are subjected to flow cytometry (see Basic Protocol 5), is subsequently analyzed by agarose gel electrophoresis (Gong et al., 1994). Cell extraction is described in Basic Protocol 5 (i.e., cells are prefixed in 70% ethanol and, after removal of ethanol, DNA is extracted with a small volume of 0.2 M phosphate-citrate buffer, pH 7.8). In this protocol, the extract is sequentially treated with RNase A and proteinase K and then directly subjected to electrophoresis. Additional Materials (also see Basic Protocol 5) Cells to be studied 2 mg/ml DNase-free RNase A stock solution (APPENDIX 2A) 1 mg/ml proteinase K (Sigma) 6× gel loading buffer (APPENDIX 2A) 0.8% agarose gel (see recipe) DNA molecular weight standards, 100 to 1000 bp Electrophoresis buffer: 10× TBE buffer (APPENDIX 2A) Ethidium bromide staining solution (APPENDIX 2A) Extract low-molecular-weight DNA from cells 1. Fix cells in ethanol and extract low-molecular-weight DNA in DNA extraction buffer (see Basic Protocol 5, steps 1 to 3). 2. Centrifuge cells 10 min at 1500 × g. Withdraw 40 µl of supernatant and transfer to a 0.5-ml microcentrifuge tube. Remove proteins and RNA from cell extract 3. Add 5 µl of 2 mg/ml DNase-free RNase A stock solution and incubate 30 min at 37°C. Cap tube to prevent evaporation. 4. Add 5 µl of 1 mg/ml proteinase K and incubate 30 min at 37°C. Perform electrophoresis 5. Add 5 µl of 6× gel loading buffer and transfer the entire tube contents to one well of a 0.8% agarose horizontal gel. 6. Prepare and load a sample of DNA molecular weight standards in a total of 55 µl in 1× gel loading buffer. 7. Assemble gel electrophoresis apparatus, using electrophoresis buffer to fill the reservoir. Run electrophoresis 16 to 20 hr at 2 V/cm. Turn off the power when the bromphenol blue from the loading buffer migrates a distance sufficient for separation of DNA fragments. 8. To visualize the bands, stain the gel 20 to 30 min with ethidium bromide staining solution. CAUTION: Ethidium bromide is a potential carcinogen. Wear gloves when handling.
9. Transfer the gel onto a UV transilluminator. Observe after illumination. DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
CAUTION: Ultraviolet light is dangerous to eyes and exposed skin. Wear protective eyewear and facewear.
10. Photograph the gel using a red or orange (e.g., Kodak Wratten no. 23A) emission filter and a clear UV light blocking filter (e.g., Kodak Wratten no. 2B).
7.5.10 Supplement 11
Current Protocols in Cytometry
REAGENTS AND SOLUTIONS Use distilled, deionized water for the preparation of all buffers. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Agarose gel, 0.8% Dissolve 1.6 g agarose in 200 ml hot (boiling) electrophoresis buffer (10× TBE). Cool to 55°C and pour solution onto a 15 × 15–cm sealed gel-casting platform. Insert the gel comb. Cool to room temperature. After the gel has hardened, remove the seal from the gel-casting platform and remove the gel comb. Place into an electrophoresis tank containing sufficient electrophoresis buffer to cover gel to a depth of ∼1 mm. Cell lysis solution with trypsin Dissolve 3 mg trypsin (Sigma type IX from porcine pancreas) in 100 ml detergent stock solution (see recipe) and adjust to pH 7.6. Store ≤1 year at −40° to −80°C in aliquots of 5 to 10 ml in tubes. Before use, bring to room temperature in a 37°C water bath. Avoid repeated thawings. Citrate/DMSO buffer 85.50 g sucrose (0.25 M final) 11.76 g trisodium citrate dihydrate (40 mM final) 5 ml DMSO (Sigma; 0.5% final) H2O to 1000 ml Adjust to pH 7.6 Store 1 month at 4°C To prepare the buffer, dissolve the dry ingredients and DMSO in ~800 ml water and then dilute to 1000 ml.
DAPI/phosphate staining solution, for nuclei isolated from paraffin tissue blocks Add 4 µl of 1 mg/ml DAPI to 80 ml water. Add 11.4 mg anhydrous Na2HPO4 (0.8 M final) and 0.82 g citric acid monohydrate (40 mM final), and mix to dissolve. The solution should be pH 7.4. Add water to 100 ml. Prepare freshly. The final DAPI concentration in this staining solution is 2 ìg/ml. A DAPI stock solution, made by dissolving 1 mg DAPI in 1 ml water, may be stored several months in dark or foil-wrapped bottles at or below −20°C.
DAPI/PIPES staining solution, for detergent-lysed cells Add 100 µl of 1 mg/ml DAPI to 100 ml PIPES/Triton X-100 buffer (see recipe). Store several weeks in dark or foil-wrapped bottles at 0° to 4°C. A DAPI stock solution, made by dissolving 1 mg DAPI in 1 ml water, may be stored several weeks in dark or foil-wrapped bottles at or below −20°C. For simultaneous staining of DNA and protein, add 100 ìl of 1 mg/ml DAPI to 100 ml PIPES/Triton X-100 buffer (see recipe) and then add 2 mg sulforhodamine 101 (Molecular Probes); stir to dissolve. Store several weeks in dark or foil-wrapped bottles at 0° to 4°C.
DAPI/Triton X-100 staining solution, for ethanol-fixed cells To 10 ml of 0.1% Triton X-100 in PBS add 10 µl of 1 mg/ml DAPI (Molecular Probes). Prepare freshly. A DAPI stock solution, made by dissolving 1 mg DAPI in 1 ml water, can be stored several months in dark or foil-wrapped bottles at −20°C.
Nucleic Acid Analysis
7.5.11 Current Protocols in Cytometry
Detergent stock solution 1 g trisodium citrate dihydrate (3.4 mM final) 1 ml Nonidet P-40 (Sigma; 0.1% final) 522 mg spermine tetrahydrochloride (1.5 mM final) 61 mg Tris (Sigma 7-9; 0.5 mM final) H2O to 1000 ml Store ≤1 year at 0° to 4°C Dissolve the dry ingredients and NP-40 in ~800 ml water and then dilute to 1000 ml.
DNA extraction buffer (0.2 M phosphate citrate buffer, pH 7.8) 192 ml 0.2 M Na2HPO4 8 ml 0.1 M citric acid Store several months at 4°C Hoechst 33342 staining solution, 1 mg/ml Dissolve 1 mg Hoechst 33342 (Molecular Probes) in 1 ml water. Store in dark or foil-wrapped bottles several months at 0° to 4°C. Internal DNA Content Standard Chicken and trout erythrocytes are convenient internal DNA content standards. Chicken blood is acquired by heart puncture and collected with heparin (50 U/ml blood). The blood is diluted with citrate/DMSO buffer (see recipe) to obtain 1.5 × 106 cells/ml, as counted with a hemacytometer. Rainbow trout blood is obtained by caudal vein puncture of an anesthetized fish, immediately mixed with citrate/DMSO buffer, and adjusted to 2.5 × 106 cells/ml. These solutions are then mixed 1:1 (v/v) to obtain 2 × 106 cells/ml. The proportion of chicken to trout erythrocytes in such a mixture provides approximately similar height peaks on DNA frequency histograms. The standard cells can be kept frozen, in small aliquots, at −40° to −80°C. PIPES/Triton X-100 buffer 3.02 g PIPES (Calbiochem; 10 mM final) 5.84 g NaCl (0.1 M final) 406 mg MgCl2⋅6H2O (2 mM final) 1 ml Triton X-100 (Sigma; 0.1% final) H2O to 1000 ml Adjust to pH 6.8 Store ≤1 year at 0° to 4°C Dissolve the dry ingredients and Triton X-100 in ~800 ml water, adjust pH with NaOH or HCl, and then dilute to 1000 ml.
Propidium iodide (PI)/spermine staining solution, for detergent-lysed cells Dissolve 20 mg PI (Molecular Probes) and 116 mg spermine tetrahydrochloride (Sigma) in 100 ml detergent stock solution (see recipe) and adjust to pH 7.6. Store ≤1 year at −40° to −80°C in small aliquots in 5- to 10-ml foil-wrapped tubes. Before use, thaw in a 37°C water bath and then keep on ice, protected from light. Propidium iodide (PI)/Triton X-100 staining solution with RNase A, for ethanolfixed cells To 10 ml of 0.1% (v/v) Triton X-100 (Sigma) in PBS add 2 mg DNase-free RNase A (Sigma) and 200 µl of 1 mg/ml PI (e.g., Molecular Probes). Prepare freshly. DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
A stock solution of PI, made by dissolving 1 mg PI in 1 ml water, can be stored several months at 0° to 4°C. If the RNase is not DNase-free, boil a solution of 2 mg RNase A in 1 ml water for 5 min.
7.5.12 Current Protocols in Cytometry
Protease solution Dissolve 100 mg Sigma XXIV bacterial protease in 80 ml water. Add 1.58 g Tris⋅Cl (Sigma; 0.1 M final) and 0.41 g NaCl (0.7 M final) and dissolve. Adjust to pH 7.2 and add water to 100 ml. Prepare freshly. The protease solution makes use of the “Carlsberg subtilisin.”
Trypsin-inactivating solution Dissolve 50 mg chicken egg white trypsin inhibitor (Sigma) and 10 mg RNase A (Sigma) in 100 ml detergent stock solution (see recipe) and adjust to pH 7.6. Store ≤1 year at −40° to −80°C in small aliquots in 5- to 10-ml tubes. Before use, bring to 0° to 4°C in a 37°C. COMMENTARY Background Information Choosing a particular protocol among those presented in this unit depends primarily on the sample type (unfixed or fixed cells, paraffinembedded tissue blocks) and the necessity for sample storage (or transport) between cell collection and analysis. The discussion below describes characteristics of each of the methods and its applicability to different material. Analysis of fixed samples In Basic Protocol 1 DNA content is measured in prefixed cell samples. The preference for analysis of fixed cells often is dictated by the need to store or transport samples (e.g., clinical samples of solid or hematologic tumors). Extended storage of unfixed cells, unless done at low temperatures following cell suspension in cryopreservative media, leads to cell deterioration and DNA degradation. Fixed cells, on the other hand, often can be stored for months if not years without much deterioration. The fixative essentially has two functions: (1) it preserves the cells by preventing their lysis and autolytic degradation, and (2) it makes the cells permeable and their DNA accessible to the fluorochrome. Precipitating fixatives (ethanol, methanol, acetone) are preferred over cross-linking agents (formaldehyde, glutaraldehyde). This is because cross-linking of chromatin has deleterious effects on the stoichiometry of DNA staining. Precipitating fixatives, though inferior in terms of stabilization and preservation of the low-molecular-weight constituents within the cell, adequately stabilize undamaged DNA. It should be stressed, however, that damaged DNA, especially DNA having large numbers of double-strand breaks (e.g., as present in apoptotic cells, see Basic Protocol 5), leaks from the ethanol-fixed cells during their hydration and subsequent staining.
Although absolute alcohols or acetone, or a mixture of absolute ethanol and acetone (1:1), can be used and may be preferred for some applications (e.g., to obtain better stabilization and retention of particular proteins for their immunocytochemical detection), they induce more extensive cell aggregation, and the aggregates cannot be easily dissociated after hydration of the fixed cells. Fixation of cells in 70% to 80% ethanol (at 0° to 4°C), on the other hand, results in less cell clumping and is generally preferred in situations when the analysis is limited to DNA content alone. Sample storage at −20° to −40°C, especially when prolonged (months), appears to be more advantageous compared to storage at room temperature. A variety of DNA fluorochromes (UNIT 4.2) can be used to stain DNA in the fixed cells. Staining with dyes that react with both DNA and RNA, such as PI used in Basic Protocol 1, requires preincubation of cells with RNase. For cytometry, PI requires blue light as the fluorescence excitation source, which is conveniently provided by the 488-nm line of the argon ion laser available on most flow cytometers. Alternate Protocol 1 employs DAPI instead of PI for staining DNA in fixed cells. The advantage of DAPI is its greater specificity toward DNA, which often is reflected by lower coefficient of variation (CV) values of the mean DNA content of G1 cell populations. A disadvantage of DAPI is the requirement for UV excitation, which may not be possible in all flow cytometers. Analysis of detergent-lysed samples The major advantage of detergent-based methods is better accuracy in DNA content estimates. Exposure of live cells to detergents results in rupture of the plasma membrane and leakage of cytoplasmic constituents. Thus, iso-
Nucleic Acid Analysis
7.5.13 Current Protocols in Cytometry
DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
lated nuclei, rather than whole cells, are stained. Because several cytoplasmic constituents either are autofluorescent or nonspecifically interact with DNA fluorochromes, the specificity of DNA staining by methods based on cell permeabilization by detergents is superior compared to that of methods based on cell fixation. This is reflected by the high accuracy of DNA content estimates, which is represented by low values of the CV of the mean DNA content of cells having uniform DNA content, such as the G1 cell population. It has to be taken into account, however, that lysis of mitotic cells, which lack a nuclear envelope, leads to dispersion of individual chromosomes. Thus, mitotic cells may not be detected by methods utilizing detergents or hypotonic solutions. Furthermore, the presence of isolated chromosomes or chromosome aggregates may contribute to an increased frequency of detection of objects with low fluorescence values, generally classified as debris or apoptotic cells. Likewise, the lysis of apoptotic cells, which have fragmented nuclei, releases numerous nuclear fragments from a single cell. This generally precludes application of detergent-based methods for analysis of the frequency of apoptotic cells on the basis of fractional DNA content, and an alternative method must be sought (see Basic Protocol 5). Further improvement in the accuracy of DNA content analysis is seen after mild and controlled proteolysis of detergent-lysed cells. It is likely that the proteolytic step removes nuclear proteins known to restrict the accessibility of DNA to many fluorochromes, resulting in improved stoichiometry of DNA staining (Darzynkiewicz et al., 1984). This approach was perfected by Vindeløv (1983a,b,c,d; Vindeløv and Christiansen, 1994), who developed a highly accurate method of cellular DNA content analysis. These authors also pioneered in introducing internal DNA content standards as an intrinsic part of the staining protocol. Their methodology, presented in Basic Protocol 2, is now widely used, especially in clinical settings for DNA content analysis of tumor samples. Alternate Protocol 2 is a simplified detergent method that is more applicable to uniform cell populations such as tissue culture cells. Vindeløv’s procedure was designed for fine needle aspiration of normal tissue and tumor biopsies. The aspiration has two functions: collection of cells from the tissue and cell disaggregation. The aspiration can be done either in vivo (needles longer than 1 in. may be needed) or in vitro, from the resected tumor. In addition,
the specimen may initially be frozen at dry ice temperature, stored at −60° to −80°C, thawed to room temperature, and then aspirated. The presence of the cryopreservative dimethylsulfoxide (DMSO) in the citrate buffer used for collecting cells in Basic Protocol 2 protects cells from damage if the samples are stored at low temperatures prior to staining. Alternate Protocol 2 employs only detergent and is simpler than the detergent/proteolytic enzyme procedure of Basic Protocol 2. It also offers excellent resolution for uniform cell populations such as tissue culture cells (the CV of the mean fluorescence of G1 populations is typically <2%). Its resolution for mixed cell samples (human tumors), however, is below that of the method employing a combination of detergent and proteolytic enzymes (see Basic Protocol 2). Alternate Protocol 2 uses either unfixed or ethanol-fixed cells. Although the cells can be stained with DAPI alone, inclusion of sulforhodamine 101, which counterstains proteins (Stöhr et al., 1978), provides an attractive combination that allows for simultaneous bivariate analysis of DNA (DAPI, blue fluorescence) and protein (sulforhodamine 101, red fluorescence after UV excitation). Alternate Protocol 2, therefore, can be used for either univariate DNA content or two-parameter (DNA and protein content) analysis. Analysis of live cells Staining of DNA in live cells (see Basic Protocol 3) is generally attempted for identification of DNA ploidy or cell cycle position of cells to be sorted on the basis of DNA content. Ideally, the fluorochrome used for supravital cell staining should be nontoxic and should not alter cell metabolism. Such a probe has yet to be developed. Although most DNA fluorochromes are charged molecules, which do not adequately penetrate the plasma membrane, some uncharged Hoechst dyes can pass through the membrane at limited rates. The most frequently used supravital DNA fluorochrome is Hoechst 33342. The procedure of staining cells with Hoechst 33342, followed by sorting of the stained cells, does not appear to induce immediate cytotoxicity to the sorted cells (Loken et al., 1980). Delayed toxicity attributed to Hoechst 33342, however, especially when stained cells were subsequently treated with antitumor drugs or radiation, has been observed. It also should be mentioned that Hoechst dyes photosensitize cells that have incorporated 5-bromodeoxyuridine (BrdU) into
7.5.14 Current Protocols in Cytometry
their DNA, in particular to UV light at ∼300 nm. Thus, the viability of sorted BrdU-labeled cells, counterstained with Hoechst dyes and illuminated with UV light, is expected to be impaired. The intensity of cell staining and resolution of DNA content (i.e., the possibility of discriminating cells in different cell cycle phases) vary among different cell types. This variability, to a large degree, is due to a rapid efflux of Hoechst 33342 from the cell generated by the P glycoprotein pump. Cells characterized by rapid efflux mechanisms (e.g., multidrug resistant tumor cells or stem cells) stain poorly with Hoechst 33342. It has been observed, however, that agents which may impair the efflux function (e.g., calcium channel blocking drugs such as verapamil) improve stainability of some cell types with Hoechst 33342 (Krishan, 1987). Analysis of paraffin-embedded samples The method of isolating cell nuclei from paraffin-embedded tissues to retrieve archival samples for flow cytometric analysis (see Basic Protocol 4) was developed by Hedley et al. (1983). This methodology opened new possibilities for retrospective studies to determine the prognostic significance of DNA ploidy or cell cycle distribution. The method also can be used for prospective studies in situations when fresh material is not available or when the tumor sample is so small that the entire sample is required for histopathological examination. A variety of modifications of the original method involve different types of tissues, different fixatives, and other variables. One of the advantages of this methodology is that the tissue fragments to be processed by flow cytometry can be selected on the basis of microscopic examination of a parallel tissue section. The blocks may then be trimmed to exclude, e.g., areas of noninvolved tissue (to diminish the proportion of stromal cells) or necrotic and hemorrhagic areas (to decrease the quantity of debris). The accuracy of DNA content analysis of nuclei from paraffin blocks, as reflected by the CV of the mean DNA content of the G1 population, is generally inferior compared to the methods that rely on either ethanol fixation (see Basic Protocol 1 and Alternate Protocol 1) or detergent treatment of fresh tissue (see Basic Protocol 2 and Alternate Protocol 2). This is due to the fact that tissues to be embedded in paraffin blocks frequently are prefixed in formaldehyde. Fixation with formaldehyde, by cross-linking DNA and proteins, impairs
stoichiometric DNA stainability with most dyes. Compared with other fluorochromes, however, DAPI is the least affected by intercellular variability in chromatin structure (Darzynkiewicz et al., 1984) and therefore is preferred for staining nuclei isolated from paraffin blocks. Another factor that may decrease accuracy of identification of aneuploid cells or discrimination of cells in different phases of the cell cycle is the presence of debris, most of which may be due to the presence of transected nuclei with incomplete DNA content. Because probability of transecting a nucleus is directly proportional to thickness of the sections and to nuclear size, preparation of thicker sections (≥50 µm) may be advisable, especially for tumors with large nuclei (e.g., tetraploid and higher-ploidy stemlines). Basic Protocol 4 presents the procedure as developed by Hedley et al. (1983) and modified by Heiden et al. (1991), which is applicable to most material. Further modifications, however, may be required for certain tissues or for different methods of cell fixation. Factors influencing the stainability of DNA in nuclei from paraffin blocks and approaches to optimize staining are discussed in detail by Hedley (1994) and Hitchcock and Ensley (1993). Analysis of apoptotic cells Apoptosis, frequently referred to as programmed cell death, is an active and physiological mode of cell death, in which the cell executes the program of its own demise and subsequent body disposal via shedding of “apoptotic bodies,” which are engulfed by neighboring cells (reviewed by Wyllie, 1992; Darzynkiewicz et al., 1997). A complex multistep mechanism regulates the propensity of cells to respond to various stimuli by apoptosis. The possibility of intervention in regulatory mechanisms raised wide interest in apoptosis in many disciplines, particularly in oncology. Because of this wide interest, methods for identifying apoptotic cells are being routinely used in numerous laboratories. Most methods employ fluorochromes and are based on flow cytometric cell analysis. One of the early and very characteristic events of apoptosis is activation of an endonuclease, which initially cleaves nuclear DNA at the sites of attachment to nuclear matrix to generate DNA fragments 50 to 300 kb in size, and subsequently at internucleosomal (spacer) sections to generate DNA fragments of 180 to 200 bp and multiples of such fragments (Oberhammer et al., 1993). The latter products are
Nucleic Acid Analysis
7.5.15 Current Protocols in Cytometry
DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
visualized after agarose gel electrophoresis as a discontinuity (DNA “laddering”) and are considered a hallmark of apoptosis. Extensive DNA cleavage provided foundations for the development of flow cytometric assays to identify apoptotic cells. Two different approaches are frequently used. One is based on extracting low-molecular-weight DNA prior to cell staining, as described in Basic Protocol 5. It is also possible to analyze the extracted DNA by agarose gel electrophoresis (see Support Protocol). The other approach relies on fluorochrome labeling of DNA strand breaks in situ and is described in UNIT 7.4. In the approach presented in Basic Protocol 5, cellular DNA is stained and measured following cell fixation in ethanol. Cell fixation in ethanol (or other precipitating fixatives, such as methanol or acetone) does not fully preserve the degraded DNA within apoptotic cells; this fraction of DNA leaks out during subsequent cell rinsing and staining. As a consequence, apoptotic cells have reduced DNA content and therefore can be recognized, following staining of cellular DNA, as cells with low DNA stainability. This peak (the “sub-G1” peak) is located to the left of the peak representing G1 cells on DNA content frequency histograms. Some cells represented by the sub-G1 cell population may also have already lost their DNA by shedding apoptotic bodies—cellular fragments containing pieces of chromatin, mitochondria, and other organelles wrapped in the plasma membrane of the disintegrating apoptotic cell (Wyllie, 1992). Loss of DNA by such a mechanism characterizes cells at late stages of apoptosis. Regardless of the mechanism of DNA loss, apoptotic cells have generally lower DNA stainability. This is in contrast to cells dying by the alternative mode, necrosis, whose DNA content are little changed compared to live cells. The degree of DNA degradation varies depending on the stage of apoptosis, cell type, and often the nature of the apoptosis-inducing agent. The extractability of DNA during the staining procedure (and thus separation of apoptotic from live cells) also varies. However, using a high-molarity phosphate-citrate buffer as the rinsing solution, as presented in Basic Protocol 5, enhances extraction of the degraded DNA. This approach can be used to control the extent of DNA extraction from apoptotic cells to the desired level to obtain maximal separation of apoptotic cells by flow cytometry (see Fig. 7.5.4). Because measurement of DNA content provides information about the cell
cycle position of the nonapoptotic cell population, this protocol can be applied to investigate the cell cycle specificity of apoptosis. The pattern of DNA cleavage during apoptosis, arising from preferential cleavage of internucleosomal (“spacer”) sections, is reflected by the characteristic DNA “laddering” during gel electrophoresis. For many years, before flow cytometric methods became common in analysis of apoptosis, DNA gel electrophoresis was one of the most widely used assays (after microscopy) to detect apoptosis, and DNA laddering was considered to be a hallmark of the apoptotic mode of cell death (Compton, 1992; Wyllie, 1992). This methodology, which is still widely used, is presented in the Support Protocol. The method is rapid and, unlike traditional biochemical techniques, uses nontoxic reagents to extract DNA (phenol, chloroform, etc., are avoided). The high molarity of the DNA extraction buffer (0.2 M Na2HPO4) weakens the electrostatic interaction between DNA and proteins in the cell, thereby allowing low-molecular-weight DNA to be released into the solution. In addition, the cells may be stored in 70% ethanol for months, prior to analysis, without any significant DNA degradation. Treatment with ethanol also inactivates many pathogens, thereby increasing the safety of sample handling. The method is applicable to clinical samples; cells can be fixed in ethanol and then stored and/or safely transported prior to analysis.
Critical Parameters and Troubleshooting Fluorochrome concentration and cell density in samples Several features of interactions between fluorochromes and nucleic acids are of importance in practical applications of DNA dyes in flow cytometry. First, as mentioned in the overview (UNIT 7.1) and discussed in more detail in UNIT 4.2 on nucleic acid probes, the interactions, whether intercalative or involving DNA helix minor groove binding, are not covalent and thereby are reversible. The number of sites along a DNA molecule that interact with a particular fluorochrome, corresponding to the intensity of DNA staining, therefore, is expected to vary depending on the dye concentration. Variability is expected at low concentrations of the dye, up to a concentration when all potential binding sites are saturated. Stable levels of fluorescence, proportional to DNA con-
7.5.16 Current Protocols in Cytometry
tent, thus require an equilibrium in which there is a significant excess of dye over number of binding sites. Under these conditions minor intersample variation in cell number does not significantly affect the concentration of the fluorochrome in the unbound, free form. One also has to remember, however, that even with the same number of cells, the number of binding sites may vary, for example, when the cells become hyperdiploid or arrested in G2/M. According to the law of mass action, significant intersample variation in free (unbound) dye concentration and, consequently, in DNA stainability is expected when no excess of fluorochrome over number of binding sites exists. Given the above, how does the mass action law translate into practical conditions of equilibrium staining in flow cytometry? An individual human cell in the G1 phase of the cycle contains 6 × 109 base pairs (bp) of DNA, and the minimal size of the intercalative binding site of free DNA is 6 bp (Le-Pecq, 1971). One expects, therefore, in a typical sample containing 106 diploid cells in a volume of 1 ml, that at 1 µM fluorochrome concentration (1 nmol/ml) there is an inadequate number of fluorochrome molecules to saturate all potential DNA binding sites (6 × 1014 molecules per 1015 binding sites). It should be remembered, however, that, due to DNA interactions with chromatin constituents in the cell nucleus, only a fraction of potential binding sites is accessible to the dye. The size of the inaccessible fraction is unknown. It has been observed, however, that exposure of cell nuclei to 0.1 M HCl, which dissociates histones from DNA, results in a manyfold increase in binding of most fluorochromes to nuclear DNA. The lowest increase (<50%) was seen in the case of DAPI, whereas the highest increase (>13-fold) was for 7-aminoactinomycin D (7-AAD; Darzynkiewicz et al., 1984). This indicates that DNA-histone interactions (i.e., nucleosomal chromatin structures) restrict the accessibility of DNA to different degrees, depending on chemical structure of the fluorochrome. Another level of restriction, especially for intercalating dyes, is the presence of closed loops of DNA in chromatin (van Holde, 1988). Because accommodation of each intercalator molecule involves extension and unwinding of the DNA helix, the closed circular structures provide a topological restraint and prevent full unwinding. It is likely, therefore, that DNA topology in chromatin provides an additional, at least 2-fold restriction for binding the dyes that unwind DNA. It is not known to what
extent direct DNA-nuclear matrix interactions (unrelated to stabilization of chromatin DNA loops) and other nuclear structures may additionally restrict DNA accessibility. Nonetheless, it is safe to assume that fewer than 10% of the potential DNA binding sites are actually accessible to most DNA fluorochromes in the nucleus. Thus, under typical staining conditions of 106 cells in 1 ml of 1 µM fluorochrome solution, there is an approximately 5-fold excess of dye molecules per accessible binding site. The above calculations indicate the need to maintain a relatively constant number of cells per sample, if the intensity of cell fluorescence between samples is compared. When large numbers of cells (especially hyperdiploid cells) are present in samples and the dye is used at low concentrations, significant variation in DNA stainability is expected due to the intersample cell density variation. The choice of optimal dye concentration is a compromise between the signal-to-noise ratio during the cell fluorescence measurement and the stoichiometric relations between dye and binding sites. Although the fluorescence of most DNA fluorochromes is increased manyfold on binding to DNA, a large excess of the dye (high concentration) in solution makes the solution itself fluorescent (noise), thereby decreasing the signal-to-noise ratio during measurement of the cell. Furthermore, at high concentrations of even the most specific DNA dyes (e.g., DAPI, Hoechst 33342), the nonspecific component of dye fluorescence, resulting from binding of dye to cell constituents other than DNA, is elevated. Optimal staining conditions, therefore, should rely on a minimal but adequate dye concentration, at which intersample variation in cell number does not significantly alter the free (unbound) dye concentration. Stoichiometry of DNA staining The accessibility of DNA to various fluorochromes also is modulated by tissue-specific differences in chromatin structure. The extreme case are cells undergoing erythropoiesis or spermatogenesis, in which DNA accessibility to most fluorochromes is markedly reduced during differentiation (Darzynkiewicz et al., 1984). The degree of reduction varies for individual fluorochromes. For example, binding of DAPI is the least influenced by chromatin structure, whereas binding of 7-AAD, an intercalating fluorochrome with a bulky substituent which locates at the minor grove of DNA helix, is affected to a large degree. Accessibility of the
Nucleic Acid Analysis
7.5.17 Current Protocols in Cytometry
DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
dimeric cyanine fluorochromes such as TO-TO or YOYO also is significantly affected by differences in chromatin structure. In practical terms, therefore, one may expect intercellular variation in DNA stainability when mixed cell types are measured in the same sample. This can be manifested on DNA content frequency histograms as the presence of pseudo-aneuploid populations, or widening of the G1 peak (increased CV value). For example, monocytes, showing higher DNA stainability with PI compared to lymphocytes or granulocytes, often form a typical pseudo-hyperdiploid peak on DNA frequency histograms. However, subjecting cells to the detergent methods, and in particular the combination of detergent and proteolytic treatment (as in the Vindeløv procedure; Basic Protocol 2), appears to diminish the effect of chromatin structure on DNA stainability, thereby improving the stoichiometric relationship between DNA content and cell fluorescence. Cell fixation with a cross-linking reagent such as formaldehyde or glutaraldehyde stabilizes DNA-protein interactions in chromatin and therefore amplifies the chromatin structure–related differences in the accessibility of DNA to fluorochromes. This is manifested by a high CV of the mean DNA content of the G1 population. Cell fixation in formaldehyde is thus the primary reason why nuclei isolated from paraffin blocks (Basic Protocol 4) generally yield rather DNA poor resolution (high CV) compared to cells permeabilized with detergents (Basic Protocol 2 and Alternate Protocol 2) or fixed in ethanol (Basic Protocol 1 and Alternate Protocol 1). One has to remember, however, that most cross-linking reactions are reversible. Therefore, if cross-linking itself is the major culprit for poor DNA resolution, prolonged incubations of deparaffinized and rehydrated cells in aqueous media are expected to improve DNA stainability. Furthermore, DAPI, being the least sensitive to differences in chromatin structure, is preferred for stoichiometric staining of DNA. There are several ways to estimate the stoichiometry of DNA staining. For uniform cell populations such as cell lines in tissue culture, a comparison of the intensity of staining of the cell populations represented by the G2/M versus G1 peaks, which is expected to approach 2.0, provides a crude practical measure of linearity (Crissman and Steinkamp, 1990). Normal hepatocytes grow at different DNA ploidy levels and therefore may also serve as markers of linearity in DNA measure-
ments. Likewise, incubation of cells in culture with cytochalasin B (5 µg/ml) induces polyploidization, making such cells useful as linearity standards. Inclusion of internal standards such as chicken or trout erythrocytes (Basic Protocol 2) provides still another marker of the stoichiometry of DNA measurement (Vindeløv et al., 1983d). These standards are discussed in more detail in UNIT 7.2. There are situations, however, even under optimal conditions of cell staining, when stoichiometry in DNA analysis cannot be attained. This is often the case if cells have been pretreated with antitumor drugs that modify DNA and chromatin structure. Intercalating drugs, drugs with chromophores able to interact with the dyes via fluorescence resonance energy transfer, and drugs damaging DNA structure can all alter DNA stainability, often in unpredictable ways. Base specificity of DNA fluorochromes The base specificity of the fluorochrome plays a major role in analysis of individual chromosomes, which differ in GC/AT ratios (Gray and Cram, 1990). Thus a combination of chromomycin, which has preference for CG base pairs, and Hoechst 33342, which reacts with AT, makes it possible to discriminate nearly all human chromosomes, which otherwise cannot be distinguished on the basis of DNA content alone. Base specificity is also of importance in analysis of cells that differ in base composition, such as microorganisms of different species or plant cells. Ratiometric analysis of fluorescence intensities of fluorochromes reacting with GC and AT base pairs, respectively, can be used to estimate the relative ratios of these bases (e.g., for identification of particular bacterial species). Still another situation where the base composition of DNA plays a role in DNA stainability is the S phase of the cell cycle. Namely, replication of the DNA associated with heterochromatin which is rich in AT occurs at the end of S phase. This should be taken into account when analyzing DNA replication rates in relation to the degree of cell progression through S phase (cellular DNA content) based on the pattern of the S phase cell labeling with base-specific fluorochromes. Accuracy of DNA content measurements The most critical issue in DNA content analysis is the accuracy of the DNA content measurement. The accuracy, as mentioned, is reflected by the extent of variation in cellular
7.5.18 Current Protocols in Cytometry
fluorescence between cells with identical DNA content, such as G0/G/1 cells. The CV of the DNA-associated mean fluorescence of G1 cells, therefore, is considered an index of the accuracy of DNA content measurements. High accuracy is required, in particular, in studies of DNA ploidy, to distinguish between DNA diploid and aneuploid cells, which may differ minimally in DNA content. The possibility of discriminating aneuploid cell populations is directly related to the CV values of DNA-associated fluorescence of the measured diploid and aneuploid cell populations. Highly accurate DNA content measurement is also critical in analysis of cell cycle distributions. Regardless of the type of software used to deconvolute DNA content, frequency histograms (see Chapter 10 for further information on data analysis), the accuracy of estimates of cell proportions in respective phases of the cell cycle directly correlates with the accuracy of DNA content measurements. There is no formal consensus regarding the acceptable maximal CV value of the mean DNA content of the G0/G1 cell population (i.e., acceptable maximal error in cellular DNA content estimates). Most researchers, however, would consider the accuracy to be poor and results unacceptable if CV values of G0/G1 populations of normal, nontumor cells exceed 6%. An exception is analysis of the DNA content of cell nuclei isolated from paraffin blocks (Basic Protocol 4), where by the nature of the sample, which generally is prefixed in formaldehyde and contains nuclei with fractional DNA content (partially cut during sectioning), good accuracy is difficult, if not impossible, to achieve. A variety of factors can contribute to poor accuracy in DNA content analysis. The most common is suboptimal optical and sample flow adjustment of the flow cytometer. Proper maintenance of the instrument and its careful adjustment prior to analysis of the experimental samples, to maximize the electronic signal intensity an d m inimize variability of the measurement of the uniformly stained cells or fluorescent beads, are required to achieve accurate DNA measurements (see UNIT 1.3 for discussion of standardization, calibration, and control in flow cytometry). Problems in sample preparation, either resulting in mechanical damage to the cells or involving incorrect composition of buffers and staining solutions, represent another common cause of poor resolution in DNA analysis. As discussed above, an excessively large number of cells (DNA) in the sample, which leads to significant depletion of
the free, unbound fluorochrome in the solution and changes the conditions of equilibrium staining, may be still another source of the problems that prevent accurate DNA content analysis. Diluting samples to achieve a lower cell concentration may improve results. DNA content analysis in living cells (Basic Protocol 3) is expected to produce variable results, depending on the ability of cells to remove the Hoechst 33342 dye by efflux mechanisms. Depending on the cell type, the CV of the mean DNA content of the DNA-uniform cell populations (G0/G1, G2/M) may vary from 5% to 10%, or even more. If the intensity of cell fluorescence is low, or if resolution of cells in various phases of the cell cycle is inadequate, staining times can be prolonged and/or the dye concentration can be increased. To improve staining (i.e., fluorescence intensity and stoichiometry) one may also attempt to use inhibitors of the P glycoprotein efflux pump (Krishan, 1987). It should be stressed that there are situations when, despite good accuracy of DNA content measurements (in terms of proper instrument adjustments and sample staining), the CV of the mean DNA content of G1 cell populations may be relatively large. This may be the case when significant numbers of dead or dying cells are present in the sample or when the cells were treated with drugs that interact with DNA. Furthermore, because of the nature of the tumor, which may either be polyclonal or have developed drug resistance by gene amplification mechanisms (e.g., as reflected by the presence minute chromosomes), tumor cell populations may have variable DNA content and therefore intrinsically high CV values for the G0/G1 cell populations. Detection of apoptotic cells For optimal detection of apoptotic cells (Basic Protocol 5), the sub-G1 peak should be positioned at the midpoint between the G1 peak and the channel zero. It is also recommended to gate out objects with fluorescence values <10% of that of the G1 peak (e.g., by positioning the G1 peak at 100 channels and gating out all objects with fluorescence intensity lower than ten channels). In this way objects with minimal fluorescence, and therefore less likely to be apoptotic cells, will be excluded from the counts. Although very late apoptotic cells may be gated out by such a strategy, the consistent underestimate of apoptosis is of lesser significance than erroneous classification of the nonapoptotic events as apoptotic cells.
Nucleic Acid Analysis
7.5.19 Current Protocols in Cytometry
Extraction of the degraded low-molecularweight DNA by 0.2 M phosphate-citrate buffer (DNA extraction buffer) is a result of weakening of electrostatic interactions between DNA and cellular proteins at the high ionic strength of this buffer; such interactions restrict elution of DNA from ethanol-prefixed cells (Gong et al., 1994). Thus, optimal positioning of apoptotic cells on DNA frequency histograms can be done by controlled DNA extraction with this buffer. In other words, if DNA degradation within apoptotic cells is extensive or if the cells have already lost a fraction of their DNA by shedding apoptotic bodies, there is no need for step 3 in Basic Protocol 5. Conversely, extended rinsing with higher volumes of DNA extraction buffer is needed in situations when DNA degradation is incomplete and the sub-G1 and G1 peaks are not separated. It should be stressed that detection of apoptotic cells by DNA content analysis is incompatible with methods which rely on cell lysis (i.e., employing detergents and proteolytic enzymes). The nucleus of an apoptotic cell is fragmented, and therefore a single cell may contain numerous nuclear fragments separated from one another. Lysis of such a cell releases these fragments, which following DNA content measurement will be located at the sub-G1 peak position. Individual nuclear fragments, therefore, may be mistakenly identified as single apoptotic cells. Other critical factors in detection of apoptotic cells are presented in UNIT 7.4 and discussed elsewhere (Darzynkiewicz et al., 1997). Critical factors pertaining to DNA content analysis which are more specific to particular methods and are discussed above (see Background Information).
Anticipated Results
DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
DNA content analysis of fixed cells Figure 7.5.1 presents DNA frequency histograms of HL-60 cells stained with PI (panel A) and DAPI (panel B) according to Basic Protocol 1 and Alternate Protocol 1, respectively. On the basis of differences in DNA content, one can identify a population of G0/G1 cells with uniform, low DNA content values, G2/M cells with DNA content twice that of G0/G1 cells, and S phase cells with intermediate DNA content. The DNA frequency histograms were deconvoluted using Multicycle software (Phoenix Flow Systems).
DNA content analysis utilizing trypsin and/or detergent Figure 7.5.2 presents DNA content distribution histograms for cells obtained from a fineneedle aspirate of a surgical biopsy of human breast cancer and stained with PI according to Basic Protocol 2. The peaks from left to right represent chicken (C) and trout (T) erythrocytes, diploid normal nuclei (D), and a hyperdiploid (DI > 1.0) population of tumor cell nuclei. Under conditions of stoichiometry of DNA stainability, the ratio of the mean DNA content of diploid human cells to chicken erythrocytes is 2.857, the ratio to trout erythrocytes is 1.258, and the ratio of mean DNA in trout versus chicken erythrocytes is 2.28 (Vindeløv et al., 1983d). Another landmark of linearity in DNA content analysis is the ratio of G2 to G1 peaks, which is expected to approach 2.0. Figure 7.5.3 presents DNA histograms of HL-60 cells stained with DAPI according to Alternate Protocol 2. DNA content analysis of paraffin-embedded samples Accuracy of DNA content analysis in cell nuclei isolated from paraffin blocks (Basic Protocol 4)—especially of tissues that were fixed in formaldehyde—is generally inferior to that of cells permeabilized with detergents or fixed by suspension in ethanol or methanol. Thus, the CV of the mean DNA content of the DNA-uniform cell populations (e.g., G0/G1 cells) rarely can be <4% and is generally within the range of 5% to 8%. Furthermore, there is often a high frequency of events with fractional DNA content (sub-G1 peak) on the DNA content frequency histogram. In contrast to intact cells that are fixed in suspension, where such events are recognized as apoptotic cells (see Basic Protocol 5), in the case of nuclei isolated from paraffin blocks these events represent predominantly nuclear fragments resulting from sectioning through the nucleus. The frequency of such mechanically fragmented nuclei is inversely proportional to the thickness of the section and to the nuclear diameter. It is difficult to find appropriate external or internal standards for DNA content (DI) for nuclei isolated from paraffin blocks. One has to be cautious, therefore, in interpreting differences in DNA stainability between cells of different lineages (or normal versus tumor cells) in different samples or even within the same sample. Chemical reactions of formaldehyde with nuclear constituents during fixation
7.5.20 Current Protocols in Cytometry
A
B
500
1200
G1
G1 1000 Cell number
800 600 400
S
G2M
200
0
S
G2M
0 0
200
0
400
32
64
DNA content
96 128 160 192 224 256 DNA content
Figure 7.5.1 Flow cytometric DNA content analysis of HL-60 cells stained with PI (A) and DAPI (B) following ethanol fixation according to Basic Protocol 1 and Alternate Protocol 1, respectively. The DNA frequency histogram (B) was deconvoluted using Multicycle software (Phoenix Flow Systems).
3
Cell number (× 103)
A 2 C T D
1
0 0
50
100
150
200
250
DNA content
Figure 7.5.2 Flow cytometric DNA content analysis of biopsied human breast cancer cells stained with PI following detergent/trypsin treatment according to Basic Protocol 2. DNA analysis revealed chicken erythrocytes (C), trout erythrocytes (T), diploid normal nuclei (D), and hyperdiploid (aneuploid) tumor cell nuclei (A). Reprinted from Vindeløv and Christensen (1990).
Nucleic Acid Analysis
7.5.21 Current Protocols in Cytometry
1200 G1 1000
Cell number
800 600 400 200 S
G2/M
0 0
32
64
96
128 160 192 224 256
DNA content
Figure 7.5.3 Flow cytometric DNA content analysis of detergent-permeabilized HL-60 cells stained with DAPI according to Alternate Protocol 2.
affect subsequent in situ DNA stainability in proportion to the time of cell exposure to formaldehyde, and temperature. The extent of chemical modification is also related to size of the fixed tissue (i.e., rate of formaldehyde penetration), intracellular ionic conditions, and the chemical nature (tissue specificity) of nuclear constituents. Isolation of nuclei from paraffin blocks and their rehydration cannot fully restore the original DNA stainability. Thus, the stoichiometry of DNA detection may be different in normal cell nuclei (used as DI standard) and tumor cell nuclei, even if the tissues were fixed identically.
DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
DNA content analysis for detection of apoptotic cells Populations of apoptotic cells are discriminated by Basic Protocol 5 on the basis of reduced DNA content (stainability) in these cells. On DNA content frequency histograms, apoptotic cells are thus represented by the sub-G1 peak, which should be separated from the G1 peak (Fig. 7.5.4). Depending on the degree of DNA degradation, however, or the prior loss of cellular DNA via shedding apoptotic bodies, the position of the sub-G1 peak may vary, from near overlap with the G1 peak to very low channels. When apoptotic cells are measured at very low fluorescence channels, it is difficult to distinguish them from individual apoptotic
bodies, fragments of chromatin, broken nuclei, chromosomes, cell debris, etc. The procedure for detecting apoptotic cells presented in Basic Protocol 5 is simple and inexpensive. It allows one to enumerate apoptotic cells and, in addition, to reveal DNA ploidy and/or the cell cycle distribution of the nonapoptotic cell population. Another advantage of this method is its applicability to any DNA fluorochrome (see Background Information) or instrument. Degraded DNA extracted from ethanol-fixed apoptotic cells can be directly analyzed by gel electrophoresis. Additional issues pertaining to the identification of apoptotic cells are discussed in UNIT 7.4. In the Support Protocol, the presence of apoptotic cells in the suspension of cells extracted with the phosphate-citrate buffer is manifested as a DNA “ladder” on the gels (Fig. 7.5.4C). The sensitivity of the method (i.e., the minimal number of apoptotic cells that can be detected per sample) varies depending on the degree of DNA degradation in individual cells (advancement of apoptosis) and efficiency of DNA extraction. The method, however, is more sensitive than traditional DNA extraction using phenol. This is because the degraded, low-molecular-weight DNA (i.e., DNA which is responsible for generating the ladder) is selectively extracted from the sample, whereas highmolecular-weight DNA remains in the cells.
7.5.22 Current Protocols in Cytometry
A
C
B
markers control CPT
1600 Control 1400
G1 Ap
1200 1000 800
1444
600
955 736 585 476 341 258
G2M
400 200
G2M
S
Size (bp)
Cell number
CPT
G1
S
0 0
32
64
96
128 0 32 DNA content
64
CPT CON
96 128 160 M
Figure 7.5.4 Flow cytometric DNA content analysis of control (A) and camptothecin-treated cells (B) subjected to extraction of low-molecular-weight DNA and stained with PI according to Basic Protocol 5, and electrophoretic analysis of the low-molecular-weight DNA (C) according to the Support Protocol. Cells treated with camptothecin (CPT) show the sub-G1 population of apoptotic cells (AP). The gel shows the DNA laddering characteristic of apoptotic cells (lane CPT), whereas control cells (lane CON) show no loss of low-molecular-weight DNA. Lane M shows DNA markers.
Thus, relatively small fractions of apoptotic cells (e.g., 1% to 2%) can be detected if DNA is extracted from a large number (>106) of cells. It should be stressed that not always is DNA cleared to fragments of 180 to 200 bp during apoptosis (Oberhammer et al., 1993). The methods described in this chapter (see Basic Protocol 5 and UNIT 7.4) cannot defect such apoptosis, and other means should then be used (Darzynkiewicz et al., 1997).
Time Considerations
For Basic Protocol 1, cell fixation takes ∼10 min, but cells have to be kept in the fixative ≥2 hr; the cell staining procedure takes up to ∼45 min. Alternate Protocol 1 requires identical times for cell fixation and staining. The detergent-based cell staining procedure of Basic Protocol 2 takes ∼60 min, but collecting cells from tissues may require up to 1 hr. For Alternate Protocol 2, on the other hand, which is much simpler, cell staining takes ∼15 min. Supravital staining of cells (Basic Protocol 3) requires 20 min for staining, although extended staining times (up to 90 min) may be needed.
For Basic Protocol 4, the whole staining procedure (sectioning and deparaffinizing tissue, isolating nuclei, and staining) takes ∼7 hr. Cell staining to detect apoptotic cells (Basic Protocol 5) takes ∼1 hr. Ethanol prefixation of cells, however, adds ≥2 hr to the time requirements. In the Support Protocol, preparing DNA for gel electrophoresis takes ∼1.5 hr. Gel electrophoresis takes up to 20 hr. Staining and visualizing the DNA on gels takes ~30 min. The time required for flow cytometric measurements of cell fluorescence may vary from 1 to 10 min per sample, depending on cell concentration, flow rate, number of cells to be measured, etc. The time for data analysis may also vary, from 1 to 10 min per sample, depending on sample complexity (e.g., presence of aneuploid populations, debris, etc.; see Chapter 10).
Literature Cited Compton, M.M. 1992. A biochemical hallmark of apoptosis: Internucleosomal degradation of the genome. Cancer Metastat. Rev. 11:105-119. Crissman, H.A. and Steinkamp, J.A. 1990. Cytochemical techniques for multivariate analysis of DNA and other cell constituents. In Flow Cytometry and Cell Sorting, 2nd ed. (M.R.
Nucleic Acid Analysis
7.5.23 Current Protocols in Cytometry
Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 227-247. Wiley-Liss, New York. Darzynkiewicz, Z., Juan, G., Li, X., Gorczyca, W., Murakami, T., and Traganos, F. 1997. Cytometry in cell necrobiology. Analysis of apoptosis and accidental cell death (necrosis). Cytometry 27:120. Darzynkiewicz, Z., Robinson, J.P., and Crissman, H.A. (eds.) 1994. Flow Cytometry, 2nd ed., Part A. Academic Press, San Diego. Darzynkiewicz, Z., Traganos, F., Kapuscinski, J., Staiano-Coico, L., and Melamed, M.R. 1984. Accessibility of DNA in situ to various fluorochromes: Relationship to chromatin changes during erythroid differentiation of Friend leukemia cells. Cytometry 5:355-363. Gong, J., Traganos, F., and Darzynkiewicz, Z. 1994. A selective procedure for DNA extraction from apoptotic cells applicable for gel electrophoresis and flow cytometry. Anal. Biochem. 218:314319. Gray, J.W. and Cram, L.S. 1990. Flow karyotyping and chromosome sorting. In Flow Cytometry and Cell Sorting, 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 503529. Wiley-Liss, New York. Hedley, D.W. 1994. DNA analysis from paraffinembedded blocks. In Flow Cytometry, 2nd ed., Part A (Z. Darzynkiewicz, J.P. Robinson, and H.A. Crissman, eds.) pp. 231-240. Academic Press, San Diego. Hedley, D.W., Friedlander, M.I., Taylor, I.W., Rugg, C.A., and Musgrove, E. 1983. Method for analysis of cellular DNA content of paraffin-embedded pathological material using flow cytometry. J. Histochem. Cytochem. 31:1333-1335. Heiden, T., Wang, N., and Tribukait, B. 1991. An improved Hedley method for preparation of paraffin-embedded tissues for flow cytometric analysis of ploidy and S-phase. Cytometry 12:614-621. Hitchcock, C.L. and Ensley, J.E. 1993. Technical considerations for dissociation of fresh and archival tumors. In Clinical Flow Cytometry: Principles and Applications (K.B. Bauer, R.E. Duque, and T.V. Shankey, eds.) pp. 93-109. Williams & Wilkins, Baltimore. Krishan, A. 1987. Effect of drug efflux blockers on vital staining of cellular DNA with Hoechst 33342. Cytometry 8:642-645. Le-Pecq, J.B. 1971. Use of ethidium bromide for separation and determination of nucleic acids of various conformational forms and measurement of their associated enzymes. In Methods of Biochemical Analysis (D. Glick, ed.) pp. 41-86. Wiley, New York.
Loken, M.M. 1980. Simultaneous quantitation of Hoechst 33342 and immunofluorescence on viable cells using a fluorescence activated cell sorter. Cytometry 1:136-142. Oberhammer, F., Wilson, J.M., Dive, C., Morris, I.D., Hickman, J.A., Wakeling, A.E., Walker, P.R., and Sikorska, M. 1993. Apoptotic death in epithelial cells. Cleavage of DNA to 300 and/or 50 kb fragments prior to or in the absence of internucleosomal degradation. EMBO J. 12:3679-3684. Stöhr, M., Vogt-Schaden, M., Knobloch M., Vogel, R., and Futterman, G. 1978. Evaluation of eight fluorochrome combinations for simultaneous DNA-protein flow analysis. Stain Technol. 53:205-212 van Holde, K.E. 1988. Chromatin. Springer-Verlag, New York. Vindeløv, L.L. and Christensen, I.J. 1990. A review of technqiues and results obtained in one laboratory by an integrated system of methods designed for routine clinical flow cytometric analysis. Cytometry 11:753-770. Vindeløv, L.L. and Christensen, I.J. 1994. Detergent and proteolytic enzyme-based techniques for nuclear isolation and DNA content analysis.In Flow Cytometry, 2nd ed., Part A (Z. Darzynkiewicz, J.P. Robinson, and H.A. Crissman, eds.) pp. 219-230. Academic Press, San Diego. Vindeløv, L.L., Christensen, I.J., Jensen, G., and Nissen, N.I. 1983a. Limits of detection of nuclear DNA abnormalities by flow cytometric DNA analysis. Results obtained by a set of methods for sample storage, staining and internal standardization. Cytometry 3:332-339. Vindeløv, L.L., Christensen, I.J., Keiding, N., Spang-Thomsen, M., and Nissen, N.I. 1983b. Long-term storage of samples for flow cytometric DNA analysis. Cytometry 3:317-322. Vindeløv, L.L., Christensen, I.J., and Nissen, N.I. 1983c. A detergent-trypsin method for the preparation of nuclei for flow cytometric DNA analysis. Cytometry 3:323-327. Vindeløv, L.L., Christensen, I.J., and Nissen, N.I. 1983d. Standardization of high-resolution flow cytometric analysis by simultaneous use of chicken and trout red blood cells as internal standards. Cytometry 3:283-331. Wyllie, A.H. 1992. Apoptosis and the regulation of cell numbers in normal and neoplastic tissues. An overview. Cancer Metast. Rev. 11:95-103.
Contributed by Zbigniew Darzynkiewicz and Gloria Juan New York Medical College Valhalla, New York
DNA Content Measurement for DNA Ploidy and Cell Cycle Analysis
7.5.24 Current Protocols in Cytometry
Analysis of Nuclear DNA Content and Ploidy in Higher Plants
UNIT 7.6
Flow cytometry is increasingly employed as the method of choice for determination of nuclear DNA content and ploidy level in plants. It thereby supplants previous methods that involved either fluorescence and absorption microspectrophotometry or chromosome counting, primarily because flow cytometry provides exceptional rapidity, convenience, and accuracy. Methods for flow cytometric measurement of DNA content have been developed for individual plant cells, protoplasts, and intact plant tissues. These methods can be employed with all commercial flow cytometers, including portable models, and hence can expect to find further applications in field biology, ecology and taxonomy, and agriculture, as well as in the systematic classification of plant-genome sizes (Galbraith et al., 1983; Galbraith, 1990; Arumuganathan and Earle, 1991; Marie and Brown, 1993; Bharathan et al., 1994; Bennett and Leitch, 1995; also see Internet Resources). A flow cytometer is equipped with a powerful source of illumination, provided either by a laser or an arc lamp. For the use of a typical laser-based flow cytometer in estimation of nuclear DNA content with the fluorochromes propidium iodide (PI) and mithramycin (MI), see Basic Protocol. Corresponding procedures are also described for an arc lamp–based cytometer (see Alternate Protocols 1 and 2). Both types of cytometer can be employed for analysis of DNA content using fluorochromes other than PI. The emission spectrum of a high-pressure mercury-arc lamp contains lines in both UV and visible wavelengths. Thus, arc lamp–based instruments are suitable for analysis of samples stained by other fluorochromes, such as mithramycin, DAPI, and Hoechst 33258. Large water-cooled lasers can be adjusted to produce UV light; however, the high capital cost of these lasers and the cost of running them represent disadvantages. Intact plant tissues used in the protocols below should be disease- and stress-free. For leaves, it is important to choose the appropriate age. Young, rapidly-growing leaves usually give the best results. However, developmental variations in cell-cycle behavior have been noted (see for example, Galbraith et al., 1983, 1991). Leaves may be transported or sent by post wrapped in moistened paper tissue and enclosed in a plastic bag. High temperatures should be avoided during transportation. Protocols are also included for preparation of fixed plant materials for those instances in which analysis cannot be performed immediately after collection and processing—including fixed protoplasts (see Alternate Protocol 3), formaldehyde-fixed tissues and cells (see Alternate Protocol 4), and ethanol/acetic acid–fixed intact cells (see Alternate Protocol 5). Bulk ploidy screening—e.g., for determining ploidy in large populations of plants—is covered as well (see Alternate Protocol 6). Protocols are included for preparation of chicken red blood cells for use as internal standards (see Support Protocol 1) and for instrument alignment (see Support Protocol 2). ANALYSIS OF SOMATIC DNA CONTENT, PLOIDY, AND CELL-CYCLE STATUS OF PLANT TISSUES USING A LASER-BASED FLOW CYTOMETER
BASIC PROTOCOL
The flexibility of laser-based flow cytometers depends on the number of lasers attached to the instrument and whether these lasers are tunable or produce light of a fixed wavelength. The following protocol describes use of the simplest instrumentation configuration—a single argon laser producing either 488-nm or 457-nm light—and two different fluorochromes, propidium iodide (PI) and mithramycin (MI). Nuclei are obNucleic Acid Analysis Contributed by David W. Galbraith, Georgina M. Lambert, Jiri Macas, and Jaroslav Dolezel Current Protocols in Cytometry (1997) 7.6.1-7.6.22 Copyright © 1997 by John Wiley & Sons, Inc.
7.6.1 Supplement 2
tained from chopped plant material or cell cultures, then stained with PI or MI and run on the flow cytometer. The flow cytometer should be set up according to the manufacturer’s instructions for analysis of PI- or MI-derived fluorescence. The steps described below for preparing the flow cytometer are for the Coulter Elite, although the instructions are in general valid for other laser-based flow cytometers that have four photomultiplier tubes. Materials Plant material for analysis: intact plant tissues, plant tissue culture or callus, or plant protoplasts Internal standards: e.g., chicken red blood cells (unfixed CRBCs; see Support Protocol 1) or plants with known nuclear DNA content (see Critical Parameters) Homogenization buffer (see recipe) 1 mg/ml propidium iodide (PI) stock solution (see recipe) 1 mg/ml RNase stock solution (see recipe) 0.1 mg/ml mithramycin (MI) stock solution (see recipe) Appropriate sheath fluid Fluorescent microspheres (DNACheck, Coulter) 5.5-cm plastic petri dishes New single-edged razor blades 15-µm pore-size nylon mesh Flow cytometer with 488 nm (PI) or 457 nm (MI) light source 0.22-µm Millipore GSWP 047 filters Cell-cycle analysis software (optional; Phoenix Flow Systems, Verity Software, or Eric Martz at http://www.bio.umass.edu/mcbfacs/flowcat.html) NOTE: All steps must be carried out on ice; it is also recommended that procedures be performed in a walk-in cold room. Prepare and stain suspensions of nuclei 1a. For intact plant tissue, cell cultures, or callus: Weigh plant materials and place in plastic petri dishes. Add 0.3 ml homogenization buffer for every 100 mg fresh weight of tissue. Chop tissues using a new single-edged razor blade, to homogenize the tissues and release the nuclei. If plants with known nuclear DNA content are to be used as internal standards, chop these simultaneously in the same petri dish with the plant tissue of interest. Plant organs should be disease- and stress-free and ideally should comprise young, growing tissues. Cell cultures must first be centrifuged at low speed (5 min at 50 × g) to remove the growth medium. The packed cells are then transferred to the petri dish using a spatula and homogenized after addition of homogenization buffer. Callus materials can be transferred directly using a spatula.
Analysis of Nuclear DNA Content and Ploidy in Higher Plants
Time of homogenization and number of “chops” required are determined empirically. Intact plant tissues typically take 2 to 3 min at ∼5 chops/sec. Suspension cultures take longer to homogenize, since the cells are less readily disrupted by razor blades. The process can be monitored through epifluorescence microscopy following staining with PI. For a standard Zeiss microscope equipped with a 50-W mercury arc source, PI-stained nuclei can be observed via excitation at 510 to 560 nm, beam splitting at 575 nm, and emission detection at >590 nm. For MI-stained nuclei, the corresponding settings are: excitation 400 to 450 nm, beam splitting at 490 nm, and emission detection at >510 nm. The released nuclei should appear intact and homogeneously stained.
7.6.2 Supplement 2
Current Protocols in Cytometry
1b. For protoplasts: Resuspend protoplasts in homogenization buffer at a concentration of 106 protoplasts/ml. Include reference protoplasts of known DNA content as needed. Harkins et al. (1990) provide information about protoplast preparation. The protoplasts break as a consequence of the detergent present in the homogenization buffer. Protoplast viability is an important consideration. Protoplasts should not be employed without gradient purification and typically should be 90% to 100% viable as determined via staining with fluorescein diacetate (Harkins et al., 1990).
2. Filter nuclear suspension through 15-µm nylon mesh. This step is critical for removing large debris, which otherwise might block the flow cell. For nuclei of plants having genome sizes much larger than tobacco (i.e., >4.8 × 109 base pairs/haploid genome or >10 pg DNA/2C nucleus), larger mesh sizes (40 to 60-ìm) are appropriate. The nylon mesh can be conveniently placed over the tip of a standard 5-ml disposable syringe and held in place using the plastic cover provided with disposable syringes, after cutting the tip off this cover to produce a ring. Alternatively, filter units can be made using ordinary 1-ml pipet tips. Cut off the tapering tip end and gently press the cut edge on a preheated hot plate until the plastic softens (since this method produces plastic fumes, work in a fume hood). Immediately press the hot edge onto the middle of a mesh square cut a bit larger than the pipet tip. Check that the seal is fully round and complete.
3a. To stain with propidium iodide: Add 1 mg/ml PI stock to a final concentration of 50 to 200 µg/ml and 1 mg/ml RNase stock to a final concentration of 10 µg/ml. Let sample stand 5 min prior to flow cytometric analysis. If unfixed CRBCs (see Support Protocol 1) are to be used as internal standards, they should be mixed with the cells prior to the addition of PI (or MI, see below). RNase must be added to avoid binding of PI to RNA. Since mithramycin binds specifically to DNA, RNase is not used in step 3b.
3b. To stain with mithramycin: Add 0.1 mg/ml MI stock to a final concentration of 10 µg/ml. Let the sample stand for 5 min prior to flow cytometric analysis. The precise concentrations of PI or MI to be employed, relative to the mass of tissue homogenized, should be determined through saturation analysis (see Critical Parameters). The time of incubation prior to flow analysis should be sufficient to permit the fluorescence signal to become invariant with respect to time.
Set up and align the flow cytometer 4. Power up the computer, cytometer, and laser. Adjust laser-emission wavelength to 488 nm (for PI) or 457 nm (for MI). For air-cooled lasers, 20-mW power output at 488 nm is sufficient. For water-cooled lasers, 100-mW output at 457 nm is employed.
5. Prepare an adequate amount of sheath fluid that closely matches the sample fluid in ionic strength and filter through an 0.22-µm Millipore GSWP 047 filter. Empty the waste tank and fill the sheath tank with the filtered sheath fluid. The sheath fluid should have the same ionic composition as the sample fluid (in this case the homogenization buffer) but detergents should be omitted.
6. Design and load a protocol for alignment using fluorescent microspheres. A Coulter Elite is normally equipped with four photomultiplier tubes (PMTs). In the standard filter configuration, 90° light scatter is assigned to PMT1, green (505- to 545-nm) fluorescence to PMT2, orange (555- to 595-nm) fluorescence to PMT3, and red (670- to 680-nm) fluorescence to PMT4. The flow cell has an orifice of 100-ìm diameter. For PI,
Nucleic Acid Analysis
7.6.3 Current Protocols in Cytometry
Supplement 2
the protocol should include acquisition of single-parameter histograms of forward-angle light scatter (FS), a single fluorescence channel (PMT3; peak and integral signal, 555- to 595-nm), and biparametric histograms of PMT3 fluorescence versus time. For MI, the protocol should substitute PMT2 (integral signal, 505- to 545-nm) for PMT3. Further blocking filters (BG38) can be employed to eliminate chlorophyll autofluorescence.
7. Align the cytometer using fluorescent microspheres diluted 1⁄10 with deionized water. Collect uniparametric histograms of fluorescence emission (peak and integral signal) and FS at a sample flow rate of 70 particles/sec, with the FS discriminator set to 100 and all other discriminators switched off. Adjust the optics until population coefficients of variation (CVs) for pulse integral and FS are minimized (typically <2% for fluorescence and 2% to 3% for FS). 8. Rinse the calibration beads out of the sample lines with water or buffer.. The instrument and sample are now ready for DNA content analysis (steps 9 to 14). After initial analyses, final adjustments to the beam and flow cell positions should be made so as to obtain the lowest possible CV.
Analyze the fluorochrome-stained nuclear suspensions 9. Trigger on peak fluorescence signal. Although triggering is conventionally done on FS, in this case the homogenate contains many more light-scattering particles than it does nuclei. If triggering is done using FS, the desired signals from the minor population of nuclei will be obscured by the large background of particles that do not correspond to nuclei.
10. Run samples, including internal standards, at a data rate of 100 to 150 events/sec. Higher data rates lead to peaks that have larger CVs. A minimum of 5,000 total events should be acquired; 20,000 is recommended. The authors recommend that the stability of fluorescence emission be monitored over time using the internal standards. This is done via biparametric analysis of fluorescence emission versus time. The DNA content of CRBCs appears to be strain- and sex-specific (Nakamura et al., 1990). It is recommended that independent measurements of CRBC DNA content be done using non-flow-based procedures (Burton, 1968) to establish this value. For some plant species, the presence of secondary products (e.g., phenolic compounds) can impede quantitative analysis. This is seen as an absence of discrete peaks on the DNA histograms, or as peaks having unacceptably large CVs (>5% to 7%). Various supplements to the homogenization buffer can be helpful in suppressing these interferences, including (final concentrations in buffer): 15 mM 2-mercaptoethanol (Dolezel et al., 1994), 5% (v/v) polyvinylpyrrolidone (MW 40,000; PVP 40), 10 mM ascorbic acid, or 10 mM dithiothreitol (Bharathan et al., 1994). These supplements may be added individually or in combination; the ideal combination is determined empirically.
Analyze flow cytometric data 11. Determine the channel number of the sample G1 nuclear DNA peak and that of the internal standard. The peaks are usually symmetrical, so either the mode position or calculated mean can be employed.
12. Calculate the somatic (2C) DNA content of the sample according to the equation: sample 2C DNA content (pg DNA) = Analysis of Nuclear DNA Content and Ploidy in Higher Plants
sample G1 peak mean × standard 2C DNA content standard G peak mean 1
7.6.4 Supplement 2
Current Protocols in Cytometry
If necessary, DNA content values can be expressed in base pairs, using the following formula: 1 pg DNA = 0.965 × 109 bp (see Internet Resources). Whereas genome sizes in base pairs are conventionally reported in terms of the haploid size (1C) of the genome, mass values for genome size are typically reported per 2C value or even per 4C value. Care should therefore be taken in converting these values. Because of base-pair specificity effects, the fluorochrome employed for the DNA content measurement should always be identified.
13. Carry out ploidy analysis if desired. Ploidy analysis represents a special case in which the DNA contents of the unknown plant nuclei are compared to those from the same species having known ploidy.This comparison can be made either between two analyses performed under identical conditions, or, in some cases, through mixing the two (or more) samples for simultaneous measurement.
14. Carry out cell-cycle analysis, if desired, using cell-cycle analysis software. Cell-cycle analysis can be conveniently done on uniparametric flow histograms of plant nuclear DNA contents using commercially available software. This assumes an absence of endoreduplication. Alternatively, the proportions of cells within the various phases of the cell cycle can be determined by using a scissor to cut out from the appropriate plots the various areas corresponding to G1, S, and G2 cells, and weighing the paper. This approach can also be used for analysis of endoreduplicated cells, in which case the graphing of logarithmic uniparametric plots is required.
ANALYSIS OF SOMATIC DNA CONTENT IN PLANT TISSUES USING AN ARC LAMP–BASED FLOW CYTOMETER The following alternate protocols are all for use with arc lamp–based flow cytometers and involve staining with 4′,6-diamidino-2-phenylindole (DAPI) or propidium iodide (PI). Alternate Protocol 1 is for isolation of nuclei using lysis buffer LB01 (Dolezel et al., 1989), which works well with most plant species and tissues. For a very few species, resolution of DNA histograms is not satisfactory using LB01. In those cases, Alternate Protocol 2, which is a modification of a procedure originally developed by Otto (1990), can provide improved resolution (also see Dolezel and Göhde, 1995). The disadvantages of Alternate Protocol 2 are the it is quite laborious and that it is not universally applicable (i.e., some species give histograms with increased levels of background and high CVs). If none of these procedures works, the authors recommend the use of Partec buffer (de Laat et al., 1987). This buffer gives very good resolution with Arabidopsis thaliana tissues. The procedure for analysis using Partec buffer is identical to that for LB01 buffer (see Alternate Protocol 1). The flow cytometer should be set up according to the manufacturer’s instructions for analysis of PI- or DAPI-derived fluorescence. The steps described below for preparing the flow cytometer are for Partec flow cytometers. Nuclear DNA Content Analysis Using LB01 Buffer or Partec Buffer and Arc Lamp–Based Flow Cytometer
ALTERNATE PROTOCOL 1
The following instructions are for preparation of cells with either LB01 (Dolezel et al., 1989) or Partec buffer (de Laat et al., 1987). Additional Materials (also see Basic Protocol) Lysis buffer LB01 or Partec buffer (see recipes) with fluorochrome, ice-cold 1 mg/ml propidium iodide (PI) stock solution (see recipe) or 0.1 mg/ml 4′,6-diamidino-2-phenylindole (DAPI) stock solution (see recipe) and 1 mg/ml ribonuclease (RNase) stock solution (see recipe)
Nucleic Acid Analysis
7.6.5 Current Protocols in Cytometry
Supplement 2
Particles for instrument alignment: e.g., microspheres or fixed chicken red blood cells (CRBCs; see Support Protocol 2) stained with the fluorochrome used (also see UNIT 1.3) Glass petri dish 42-µm pore-size nylon mesh Flow cytometer equipped with high-pressure mercury-arc lamp Filters and dichroic mirror appropriate for fluorochrome used Prepare suspensions of nuclei 1a. For intact plant tissue, cell cultures, or callus: Weigh a small amount of plant material (typically 20 mg) and chop with a new razor blade or a sharp scalpel in 1 ml of ice-cold fluorochrome-containing LB01 lysis buffer or Partec buffer in a glass petri dish (see Basic Protocol, step 1a, for chopping technique). It is preferable to include the DNA fluorochrome (DAPI or PI) in the buffer. Alternatively, the stain may be added immediately after filtration (step 2). Note that commercially available Partec buffer contains 4 ìg/ml DAPI. The actual quantity of plant material to be used for nuclei isolation depends both on the type of tissue and on the species, and must be determined experimentally (larger quantities are usually needed for callus or cultured cells).
1b. For protoplasts: Resuspend protoplasts in ice-cold fluorochrome-containing LB01 lysis buffer or Partec buffer to a concentration of 1 × 105 to 1 × 106 protoplasts/ml. For protoplasts the concentration of detergent (Triton X-100) in LB01 buffer should be increased to 0.5% (v/v); this improves the release of the nuclei from the protoplasts. Nuclei cannot be released from “collapsed” protoplasts; hence protoplast viability is an important consideration. Typically the protoplasts should be 90% to 100% viable as determined using FDA (Harkins et al., 1990).
2. Filter the suspension through a 42-µm nylon mesh. If the DAPI or PI was not initially added to the LB01 or Partec buffer, add 0.1 mg/ml 4′,6-diamidino-2-phenylindole (DAPI) stock solution (see recipe) to a final concentration of 2 ìg/ml or 1 mg/ml propidium iodide (PI) stock solution (see recipe) to a final concentration of 50 ìg/ml along with 1 mg/ml RNase stock solution (see recipe) to a final concentration of 50 ìg/ml.
3. Store on ice prior to analysis (a few minutes to 1 hr). Set up and align the flow cytometer 4. Empty the waste container of the flow cytometer and fill the sheath-fluid container with distilled water. Older Partec cytometers (e.g., PAS II) ran under vacuum and sheath fluid had to be deaerated. This is not necessary with the newer models.
5. Choose the optical filter set that corresponds to the fluorochrome used to stain nuclear DNA. For DAPI-stained samples, use UG1 as the excitation filter, TK420 as the dichroic mirror, and GG435 as a barrier filter (a dry 20× objective is sufficient for most applications). For PI-stained samples, use BP520 as an excitation filter, TK575 as the dichroic mirror, and RG590 as a barrier filter; the use of a 40×/1.25 glycerol-immersion objective is recommended. Analysis of Nuclear DNA Content and Ploidy in Higher Plants
6. Set the amplification to linear mode. 7. Prepare particles suitable for instrument alignment (e.g., microspheres or fixed CRBCs stained with appropriate fluorochrome; see Support Protocol 2 and UNIT 1.3).
7.6.6 Supplement 2
Current Protocols in Cytometry
8. Using the particles prepared in step 7 and a low data rate (20 to 50 events/sec), adjust the gain of the instrument so that the peak corresponding to the G1 nuclei appears approximately at one-fifth of the distance across the x axis (e.g., channel 50 on a 256 scale). It is important to run the analyses at a low data rate to achieve the highest resolution.
9. Check the resolution (the population coefficients of variation of the fluorescent peaks) and the linearity of the instrument. For CRBC nuclei, the coefficient of variation should be <2% (1% after DAPI staining). Because of the occurrence of clumped nuclei (e.g., doublets and triplets), fixed CRBC nuclei are very useful for checking instrument linearity.
10. If necessary, adjust the objective focus, the flow cell position, and/or the lamp focus and position to achieve the highest resolution. 11. Rinse the calibration particles from the sample line with water to avoid contamination of subsequent samples. The new generation of lamp-based Partec cytometers are very stable after initial alignment, and usually no further adjustment is needed.
Analyze the suspension of stained nuclei 12. Begin analyzing the nuclei (from step 3) at a rate of 20 to 50 nuclei/sec using linear amplification. Linear is preferred over logarithmic amplification for most applications. However, logarithmic amplification may be advantageous for the analysis of phenomena where a broad range DNA contents is expected (e.g., polysomaty or endoreduplication).
13. Adjust the gain of the instrument so that the peak corresponding to the G1 nuclei appears approximately at one-fifth of the distance across the x axis (e.g., channel 50 on a 256 scale). Depending on the requirements of different types of experiment, the G1 peak may be moved to different channel positions.
14. Acquire data for a total of 5,000 to 20,000 nuclei. 15. Analyze data (see Basic Protocol, steps 11 to 14). Two-Step Nuclear DNA Content Analysis Using Arc Lamp-Based Flow Cytometer The following is a two-step procedure involving successive use of two different buffers (Otto, 1990; Dolezel and Göhde, 1995). Large numbers of samples can be prepared and simultaneously centrifuged (step 3). If necessary, the protocol can be interrupted after step 4 and the samples kept at room temperature for prolonged periods of time before continuing with addition of Otto II buffer and analysis.
ALTERNATE PROTOCOL 2
Additional Materials (also see Basic Protocol) Otto I buffer (see recipe) and Otto II buffer (with fluorochrome; see recipe) 1 mg/ml propidium iodide (PI) stock solution (see recipe) or 0.1 mg/ml 4′,6-diamidino-2-phenylindole (DAPI) stock solution (see recipe) and 1 mg/ml ribonuclease (RNase) stock solution (see recipe) Glass petri dish 42-µm pore-size nylon mesh Tabletop centrifuge Flow cytometer equipped with high-pressure mercury-arc lamp
Nucleic Acid Analysis
7.6.7 Current Protocols in Cytometry
Supplement 2
Additional reagents and equipment for analysis with an arc lamp–based flow cytometer (see Alternate Protocol 1, steps 4 to 15) 1a. For intact plant tissue, cell cultures, or callus: Weigh a small amount of plant material (typically 20 mg) and chop with a new razor blade or a sharp scalpel in 1 ml of ice-cold Otto I buffer in a glass petri dish (see Basic Protocol, step 1a, for chopping technique). 1b. For protoplasts: Resuspend protoplasts in ice-cold Otto I buffer to a concentration of 1 × 105 to 1 × 106 protoplasts/ml. 2. Filter the suspension through a 42-µm nylon mesh. 3. Centrifuge 5 min at 150 × g, 4°C. Remove supernatant, leaving ∼100 µl of liquid above the pellet, which contains the nuclei. 4. Resuspend the nuclei by gentle shaking, then add 100 µl of fresh Otto I buffer. 5. Incubate 10 to 60 min at room temperature with occasional shaking. The optimal incubation period depends on the plant species. Select the incubation period that gives the lowest background and CV.
6. Add 1 ml of Otto II buffer with fluorochrome. Hold sample at room temperature. It is preferable to include the fluorochrome (DAPI or PI) in the Otto II buffer at this step. Alternatively, these compounds can be added to the sample as the DAPI or PI/RNA stock solutions (see recipes) after the Otto II buffer has been added.
7. Analyze by flow cytometry within 5 to 15 min (see Alternate Protocol 1, steps 4 to 15). ALTERNATE PROTOCOL 3
DNA CONTENT ANALYSIS OF FIXED PROTOPLASTS In some cases, it is not possible to analyze plant material immediately after collection; hence a procedure is necessary that permits storage of material for analysis at later date. This is especially critical in experiments involving the measurement of cell-cycle kinetics, where samples have to be collected and analyzed at specific time intervals. This protocol describes preparation of fixed protoplasts. Because of changes in chromatin structure, nuclei isolated from fixed tissues are not recommended for determination of DNA content in absolute units (genome size). Additional Materials (also see Basic Protocol) AES fixative (see recipe), ice-cold 70% ethanol Homogenization buffer (see recipe) Tabletop centrifuge 60-µm pore-size nylon mesh 1. Prepare protoplasts according to methods specific for the tissue and species of interest. Purify the protoplasts using gradient centrifugation. Centrifuge 5 min at 50 × g, 4°C, to obtain a pellet containing 1 × 105 to 1 × 106 protoplasts. Harkins et al. (1990) provides information about protoplast preparation and gradient purification.
Analysis of Nuclear DNA Content and Ploidy in Higher Plants
2. Remove supernatant, then fix protoplasts by gently resuspending in 2 ml ice-cold AES fixative and incubating 5 min on ice. Centrifuge 5 min at 50 × g, 4°C, to recover protoplasts.
7.6.8 Supplement 2
Current Protocols in Cytometry
3. Repeat step 2, then wash twice, each time by removing the supernatant, adding 5 to 10 ml ice-cold 70% ethanol, then centrifuging 5 min at 50 × g, 4°C. Fixed protoplasts can be stored in 70% ethanol for up to a week at 4°C.
4. Remove supernatant and resuspend pellet in 2 ml homogenization buffer. 5. Filter the suspension through a 60-µm nylon mesh. 6. Stain and analyze the fixed protoplasts (see Basic Protocol or Alternate Protocol 1). DNA CONTENT ANALYSIS OF FORMALDEHYDE-FIXED TISSUES AND CELLS
ALTERNATE PROTOCOL 4
This procedure for fixation and staining of plant tissues and cells is from Sgorbati et al. (1986). Additional Materials (also see Basic Protocol and Alternate Protocol 1) Formaldehyde fixative (see recipe), 4°C Tris buffer (see recipe), 4°C Lysis buffer LB01 (see recipe) without fluorochrome, 4°C 0.1 mg/ml DAPI stock solution (see recipe). Tabletop centrifuge Glass petri dish 42-µm pore size nylon mesh 1. Add 20 ml formaldehyde fixative to 10 to 100 mg plant tissue and incubate 10 min at 4°C. The optimal concentration of formaldehyde and the duration of fixation should be determined empirically for any given material (to achieve the highest possible resolution of peaks in the DNA content histograms).
2. Decant formaldehyde fixative and remove residual fixative with a Pasteur pipet. Wash out formaldehyde fixative by incubating in three changes of Tris buffer, each time for 10 min at 4°C. The fixed tissues can be stored at 4°C for up to several days. For prolonged storage, it is preferable that the material be stored as fixed nuclei rather than fixed tissue (see step 5).
3. Homogenize the tissues by crushing with a glass rod in 1 ml ice-cold Tris buffer or lysis buffer LB01 in a glass petri dish. Alternatively, isolate nuclei by chopping the tissues with a new razor blade and/or scalpel (see Basic Protocol). It is also possible to use a motorized homogenizer (e.g., Polytron PT 1200 from Brinkmann). In this case, transfer fixed tissues to a 12 × 75–mm polystyrene tube containing ice-cold LB01 buffer. This approach is especially convenient for isolation of nuclei from very small root tips and/or small amounts of cultured cells.
4. Filter the suspension through 42-µm nylon mesh to isolate nuclei. 5. Store the nuclei at 4°C prior to analysis. Fixed nuclei can be stored for more than a week.
6. Add DAPI stock solution to a final concentration of 2 µg/ml. Binding of propidium iodide to DNA in formaldehyde-fixed chromatin is impaired and the use of DAPI for DNA staining is recommended. Alternatively, the negative effect of the fixation may be reversed by heating (Overton and McCoy, 1994) and the nuclei stained with PI.
7. Analyze relative DNA content of isolated nuclei (see Basic Protocol, steps 4 to 14 or Alternate Protocol 1, steps 4 to 15).
Nucleic Acid Analysis
7.6.9 Current Protocols in Cytometry
Supplement 2
ALTERNATE PROTOCOL 5
DNA CONTENT ANALYSIS OF FIXED, INTACT CELLS This sample preparation technique may be used with plant tissues or isolated plant cells (Pfosser, 1989). Additional Materials (also see Basic Protocol or Alternate Protocol 1) Ethanol/acetic acid fixative (see recipe), ice-cold 70% ethanol, ice-cold Citrate buffer (see recipe), 4°C Enzyme solution (see recipe) 0.1 mg/ml DAPI stock solution (see recipe). Tabletop centrifuge 1. Transfer the plant material (tissues or cells) into an excess of ice-cold ethanol-acetic acid fixative. 2. Centrifuge cells 5 min at 100 × g, 4°C. After centrifugation, replace the fixative and continue fixing for at least 8 hr at 4°C. 3. Wash cells twice, each time by centrifuging at 100 × g, 4°C, removing the supernatant, adding ice-cold 70% ethanol, then centrifuging again and removing the supernatant. Store fixed cells in 70% ethanol at −20°C. 4. Centrifuge cells 5 min at 100 × g, 4°C. Wash cells three times, each time by removing the supernatant, adding 4°C citrate buffer, then centrifuging 5 min at 100 × g, 4°C, and removing the supernatant. 5. Resuspend cell pellet in enzyme solution and incubate 2 hr at 37°C to release the nuclei. The concentration of enzymes and the duration of incubation should be determined empirically to achieve the highest resolution of DNA content histograms.
6. Centrifuge 5 min at 100 × g, 4° C, and remove enzyme solution. 7. Add DAPI stock solution to citrate buffer for a final concentration of 1 µg/ml. Using this solution, resuspend the cell pellet at a concentration of 1 × 106 nuclei/ml. Let sample stand 5 min prior to analysis. 8. Analyze relative fluorescence intensity via flow cytometry (see Basic Protocol, steps 4 to 14, or Alternate Protocol 1, step 4 to 15). ALTERNATE PROTOCOL 6
Analysis of Nuclear DNA Content and Ploidy in Higher Plants
BULK PLOIDY SCREENING Some applications of flow cytometric ploidy determination involve large populations of plants. In order to certify the ploidy homogeneity of seed lots, random seed samples are taken and germinated, and the ploidy status of the plants determined. It is inefficient to measure each plant individually, and for the purposes of certification, a pooling strategy can be employed. Various numbers of the germinated plants are homogenized together and the mixed populations of nuclei are analyzed. Since nuclei of different ploidies (e.g., diploid and triploid) are well separated on one-dimensional DNA histograms, as long as sufficient nuclei are analyzed one can confidently determine the relative proportions of the two ploidy classes in the homogenates. 1. Germinate seeds under appropriate conditions. If required, the seeds can be surface-sterilized by immersion for 15 min in diluted (30% v/v) commercial bleach solution. The residual bleach is removed by washing the seeds several times in sterile water.
7.6.10 Supplement 2
Current Protocols in Cytometry
2. Select seedlings at the appropriate growth stage and excise tissue samples. Pool equal fresh weights of the sampled tissues. The numbers of plants that can be pooled should be determined empirically, using individuals of known ploidies. Thus, one triploid plant tissue sample should be mixed with various numbers of diploid plant samples, or vice versa, prior to homogenization.
3. Subject the plant samples to homogenization and flow analysis (see Basic Protocol or Alternate Protocol 1). Examine the resultant DNA histograms to determine the numbers of nuclei (hence time of analysis) required for accurate detection of the minor ploidy class. From these measurements, the most efficient strategy for ploidy determination can be devised.
PREPARATION OF UNFIXED CHICKEN RED BLOOD CELLS FOR USE AS INTERNAL STANDARDS IN GENOME-SIZE MEASUREMENTS
SUPPORT PROTOCOL 1
Materials Live chicken Acid citrate dextrose (ACD) buffer (see recipe) Dimethylsulfoxide (DMSO) 21-G needle Heparinized blood collection tube Tabletop centrifuge Polypropylene cryotubes (e.g., Nunc) Liquid nitrogen 1. Using a 21-G needle, collect 10 ml of venous blood in a heparinized tube from a blood vessel under the wing of a healthy chicken. Mix the tube by gently inverting and place on ice. 2. Centrifuge 10 min at 500 × g, 4°C. 3. Aspirate plasma and gently resuspend pellet in 10 ml ACD buffer. 4. Centrifuge 10 min at 500 × g, 4°C. Aspirate supernatant and resuspend pellet in 10 ml ACD buffer. Repeat centrifugation and resuspend pellet again in ACD buffer. 5. Add 0.8 ml DMSO to resuspended cells and mix gently. 6. Pipet 100-µl aliquots into cryotubes and freeze in liquid nitrogen. Store at –80°C up to one year. 7. When ready to use, thaw aliquot on ice and add directly to sample as a reference standard. CRBCs may be diluted in sample buffer if necessary.
Nucleic Acid Analysis
7.6.11 Current Protocols in Cytometry
Supplement 2
SUPPORT PROTOCOL 2
PREPARATION OF FIXED CHICKEN RED BLOOD CELL NUCLEI FOR INSTRUMENT ALIGNMENT This procedure was developed by J. Dolezel (unpub. observ.). For additional discussion of the use of particles for instrument alignment, see UNIT 1.3. Materials Fresh chicken blood (collected in heparinized tube to prevent coagulation) CRBC buffers I, II, and III (see recipes) Ethanol/acetic acid fixative (see recipe), ice-cold 70% ethanol, ice-cold Tabletop centrifuge 15-ml polypropylene centrifuge tubes 30-G needle and syringe 42-µm pore-size nylon mesh 1. Mix 1 ml fresh chicken blood with 3 ml CRBC buffer I in a 15-ml tube. Centrifuge at 5 min at 50 × g, 4°C. 2. Discard the supernatant, add 3 ml CRBC buffer I, and gently mix. Centrifuge 5 min at 50 × g, 4°C. 3. Discard the supernatant, resuspend the pellet in 2 ml CRBC buffer II, and vortex briefly. 4. Immediately add 2 ml CRBC buffer III and vortex briefly. 5. Centrifuge 5 min at 250 × g, 4°C. Discard the supernatant, add 2 ml CRBC buffer III, and mix gently. 6. Centrifuge 5 min at 120 × g, 4°C. Discard the supernatant. Transfer the pellet to a clean 15-ml tube, add 2 ml CRBC buffer III, and mix gently. 7. Centrifuge 5 min at 90 × g, 4°C. Discard the supernatant. 8. Resuspend the pellet in the residual supernatant, add 2 ml ice-cold freshly prepared ethanol/acetic acid fixative, and vortex briefly. 9. Leave overnight at 4°C. Do not shake. 10. Gently remove the fixative and resuspend the pellet of nuclei in residual fixative. 11. Add 6 ml ice-cold 70% ethanol, vortex briefly, and pass three times through a 30-G needle using a syringe. 12. Filter the nuclear suspension through a 42-µm nylon mesh to remove large clumps. 13. Store in 2-ml aliquots in 70% ethanol at −20°C. If the concentration of nuclei is too high, dilute with ice-cold 70% ethanol before storage. Isolated nuclei can be stored for several years without any sign of deterioration. IMPORTANT NOTE: Fixed nuclei are not suitable as a standard for estimation of DNA content in absolute units (genome size).
Analysis of Nuclear DNA Content and Ploidy in Higher Plants
7.6.12 Supplement 2
Current Protocols in Cytometry
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Acid citrate dextrose (ACD) buffer 136 mM glucose 75 mM trisodium citrate 4.2 mM citric acid monohydrate Filter through 0.22-µm filter Prepare fresh AES (acetic acid/ethanol/sorbitol) fixative Prepare a solution of 18 parts ethanol:3 parts acetic acid:1 part water. Dissolve sorbitol in this solution to a 1% (w/v) final concentration. Prepare fresh. Recipe from Galbraith and Shields (1982).
Citrate buffer 10 mM disodium EDTA 10 mM sodium citrate Adjust pH to 4.8 Prepare fresh CRBC buffer I 140 mM NaCl 10 mM sodium citrate 1 mM Tris⋅Cl, pH 7.1 (APPENDIX 2A) Prepare fresh CRBC buffer II 140 mM NaCl 5% (v/v) Triton X-100 Prepare fresh CRBC buffer III 320 mM sucrose 15 mM MgSO4⋅7 H2O 15 mM 2-mercaptoethanol 1 mM Tris⋅Cl, pH 7.1 (APPENDIX 2A) Prepare fresh DAPI stock solution Prepare a 0.1 mg/ml solution of 4′,6-diamidino-2-phenylindole (DAPI; Molecular Probes). Filter through a 0.22-µm filter to remove small particles. Store up to 6 months at −20°C in 1-ml aliquots. Enzyme solution Citrate buffer (see recipe) containing: 0.5% (v/v) cellulase Onozuka R10 (Serva) 5% (v/v) pectinase from Aspergillus niger (Sigma) 1 mg/ml 4′,6-diamidino-2-phenylindole (DAPI; Molecular Probes) Prepare fresh Ethanol/acetic acid fixative Mix 3 parts 96% ethanol with 1 part glacial acetic acid. Prepare fresh. Nucleic Acid Analysis
7.6.13 Current Protocols in Cytometry
Supplement 2
Formaldehyde fixative Add formaldehyde to freshly prepared Tris buffer (see recipe) to a final concentration of 4% (v/v). Prepare fresh. Homogenization buffer 45 mM MgCl2 20 mM 3-(N-morpholino) propane sulfonate (MOPS) 30 mM sodium citrate 0.1% (v/v) Triton X-100 Adjust pH to 7.0 with NaOH Filter through 0.22-µm filter Store up to 6 months frozen at −20°C This recipe is from Galbraith et al. (1983). Certain additives may be helpful in suppressing interference from the presence of secondary products in some plant species; see Basic Protocol, step 10 annotation.
Lysis buffer LB01 15 mM Tris⋅Cl, pH 7.5 (APPENDIX 2A) 2 mM disodium EDTA 0.5 mM spermine tetrahydrochloride 80 mM KCl 20 mM NaCl 0.1% (v/v) Triton X-100 (0.5% for protoplasts) Adjust to pH 7.5 with 1 M NaOH Filter through 0.22-µm filter Add 2-mercaptoethanol to 15 mM Store up to 6 months at in 10-ml aliquots at −20°C Optional: Prior to use, add 0.1 mg/ml DAPI stock solution (see recipe) to a final concentration of 2 µg/ml or 1 mg/ml PI stock solution (see recipe) to a final concentration of 50 µg/ml along with 1 mg/ml RNase stock solution (see recipe) to a final concentration of 50 µg/ml. Recipe is from Dolezel et al. (1989). Alternatively, the DAPI or PI/RNase may be added to the cell suspension after it has been filtered.
Mithramycin (MI) stock solution Prepare 0.1 mg/ml mithramycin (Sigma) in homogenization buffer (see recipe). Filter through a 0.22-µm filter. Store up to 6 months in 0.5-ml aliquots at −20°C. Different samples of commercial mithramycin vary in purity. The optical density of the solution at 420 nm should be checked and adjusted to 0.6 with additional mithramyicin if necessary.
Otto I buffer 0.1 M citric acid 0.5% (v/v) Tween 20 Filter through a 0.22-µm filter Prepare fresh
Analysis of Nuclear DNA Content and Ploidy in Higher Plants
Otto II buffer Prepare 0.4 M Na2HPO4⋅12H2O. Filter through a 0.22-µm filter; store up to 6 months at room temperature. Prior to use (if stain is to be incorporated directly in the buffer), add 0.1 mg/ml DAPI stock solution (see recipe) to a final concentration of 4 µg/ml or 1 mg/ ml PI stock solution to a final concentration of 50 µg/ml along with 1 mg/ml RNase stock solution (see recipe) to a final concentration of 50 µg/ml. RNase is needed to prevent PI from binding to RNA. Since DAPI specificially stains DNA RNase is not needed with that dye (see Background Information).
7.6.14 Supplement 2
Current Protocols in Cytometry
Partec buffer 0.2 M Tris⋅Cl, pH 7.5 (APPENDIX 2A) 4 mM MgCl2 0.5% (v/v) Triton X-100 Prepare fresh Optional: Prior to use, add 0.1 mg/ml DAPI stock solution (see recipe) to a final concentration of 2 µg/ml or 1 mg/ml propidium iodide PI stock solution (see recipe) to a final concentration of 50 µg/ml along with 1 mg/ml RNase stock solution (see recipe) to a final concentration of 50 µg/ml. RNase is needed to prevent PI from binding to RNA. Since DAPI specificially stains DNA RNase is not needed with that dye (see Background Information). Recipe is from de Laat et al. (1987). Note that the commercially available Partec buffer contains 4 ìg/ml DAPI.
Propidium iodide (PI) stock solution, 1 mg/ml Prepare 1 mg/ml propidium iodide (PI; Molecular Probes). Filter through a 0.22-µm filter. Store up to 6 months in 0.5-ml aliquots at −20°C. RNase stock solution, 1 mg/ml Prepare a 1 mg/ml solution of RNase (type IIA, Sigma). Heat 15 min at 90°C to inactivate DNases. Filter through a 0.22-µm filter. Store up to 6 months in 0.5-ml aliquots at −20°C. Tris buffer 100 mM NaCl 10 mM disodium EDTA 0.1% (v/v) Triton X-100 10 mM Tris⋅Cl, pH 7.5 (APPENDIX 2A) Prepare fresh COMMENTARY Background Information Eukaryotic organisms are characterized by the envelopment of the cellular genetic material within organelles. In interphase cells, the majority of this material is found in the nucleus in the form of linear double-stranded DNA molecules, organized within hereditary units termed chromosomes. With the development of methods for the measurement of nuclear DNA content, it has long been recognized that these values are tightly controlled and are characteristic of the species in which they are measured. In most cases, different somatic cell types have a constant value for nuclear DNA content, which is twice that of the gametes. From this has come the definition of the somatic DNA content (2C). The between-species variation in C value is particularly striking in the higher plants—from a low of ∼0.15 pg in Arabidopsis thaliana to as much as 127.4 pg in Fritillaria assyriaca—and may have evolutionary significance (Bharathan et al., 1994).
If it is assumed that the gametes are haploid (for higher plants this may not always be the case—e.g., as a result of polyploidization during speciation), then diploid somatic cells have a 2C DNA content. Measurement of DNA content values and comparison to known standards can therefore provide an assessment of ploidy. Ploidy analysis turns out to have considerable importance in agriculture. Particularly in Europe, commercial seed production requires certification of the presence of a single ploidy class in the seed. For some species (e.g., soybeans and hybrid maize) this requirement is superfluous, either because of the breeding system or the reproductive biology of the plant. For other species, the breeding strategy leading to seed production may require that seed producers plant only seeds of a specific ploidy class (triploid seedless watermelon and sugar beet are two such examples). Certification of ploidy homogeneity in these cases is critical. A new application of ploidy analysis has recently emerged in the field of plant genetic engineer-
Nucleic Acid Analysis
7.6.15 Current Protocols in Cytometry
Supplement 2
Analysis of Nuclear DNA Content and Ploidy in Higher Plants
ing. The regeneration of transformed crop plants, or of chemically induced doubled haploids, frequently provides populations many of which are non-euploid off-types. Maintenance of these undesirable plants in greenhouses constitutes a considerable economic sink. The availability of flow cytometric techniques to efficiently and accurately rogue out non-diploids at an early stage is of considerable benefit. Finally, the accuracy with which nuclear DNA contents can be measured has led to use of flow cytometry to identify aneuploidy (Pfosser et al., 1995). This should be important in a variety of molecular breeding strategies, as well as for the isolation of individual chromosomes through flow sorting. Flow cytometric analysis can also be extremely useful for analysis of the cell-division cycle in plants, which is thought to operate in a manner similar to that seen in other eukaryotes. Somatic cells having a 2C DNA content are in either G0 phase (noncycling) or G1 phase (cycling). Cycling cells subsequently enter S phase, wherein the nuclear DNA content is doubled, and then G2 phase. This is followed by mitosis and cell division. Analysis of the nuclear DNA content via flow cytometry provides an accurate estimate of the proportions of the cells within the various phases of the celldivision cycle, although it can be difficult to distinguish cells in G0 from those in G1. In general, the flow histograms of plant tissues show a major peak of G0/G1 cells, a well defined G2 peak, and few cells in S phase, even for rapidly growing cell cultures. An additional complication in the analysis of flow histograms from plant tissues is the occurrence of somatic endoreduplication (DeRocher et al., 1990; Galbraith et al., 1991). This involves multiple occurrences of S phase without intervening mitoses, and results in the presence of discrete populations of cells having DNA contents in the series 2C, 4C, 8C, 16C, and so on. Somatic endoreduplication is found in specialized situations (e.g., the maize endosperm), but can also be found in “normal” tissues, such as leaf epidermis and mesophyll in Arabidopsis thaliana. Analysis of endoreduplication requires the use of logarithmic abscissas, and current software programs do not allow calculation of the proportions of cells within the various endoreduplicative cycles. In any case, distinguishing 4C cells in G2 of the diploid cell cycle from G1 cells in the first endoreduplicative cycle is problematic. The knowledge of DNA content in absolute units (genome size) is important in many areas
of research ranging from evolutionary studies to genome mapping. Flow cytometry is a very convenient approach to determining genome size in plants. However, some precautions should be taken in this application. It has already been mentioned that DAPI and mithramycin exhibit a pronounced base-pair specificity in DNA binding. In calculation of absolute DNA contents in comparison to internal standards such as CRBCs, it is clear that use of these fluorochromes will introduce systematic errors that in some cases can be quite large. Godelle et al. (1993) have described methods whereby this feature can be employed for the analysis of genome base-pair ratios. The various fluorochromes differ in the manner in which they bind to nucleic acids. Ethidium bromide and PI bind through intercalation. In order to measure DNA contents, treatment with RNase is required to eliminate fluorescence contributed by binding to RNA. Hoechst 33258, DAPI, and MI specifically bind DNA and not RNA; hence RNase treatment is not required. There is a disadvantage, however, in using these dyes. Unlike ethidium bromide and propidium iodide, they exhibit base-pair specificity in binding to DNA, and this must be remembered if they are to be used for estimation of genome size. On the other hand, the results obtained with different fluorochromes can be used to estimate AT/GC ratios. DAPI typically provides DNA histograms having G1 peaks with the lowest coefficients of variation, and is a preferred fluorochrome for ploidy analysis, including the detection of aneuploidy. It should also be noted that PI, being an intercalating dye, is sensitive to the degree of chromatin condensation. For some species, this property has been employed to explore the proportions of euchromatin and heterochromatin present in the genome (Rayburn et al., 1992). In all cases, it should be remembered that alterations in chromatin condensation as a function of growth state or tissue type might well affect total fluorescence emission (O’Brien et al., 1996). Thus it is recommended that comparative analysis between different samples employ tissues of similar metabolic and developmental state.
Critical Parameters and Troubleshooting Sample Preparation If the acceptance criterion is defined as a CV of <5%, many plant species provide acceptable DNA histograms. If unacceptably high CVs are
7.6.16 Supplement 2
Current Protocols in Cytometry
encountered, supplementing the homogenization medium with various components can be helpful. In particular, the authors have found that inclusion of antioxidants, bovine serum albumin (BSA), and polyvinylpyrrolidone (PVP) can be critical for obtaining adequate flow histograms from some species (Bharathan et al., 1994; Dolezel et al., 1994). Empirical analysis of the effects of varying the level of Triton X-100 in the homogenization medium over the range of 0.1% to 1% or even higher has also been recommended (S.C. Brown, pers. comm.). It is critical that sample-staining conditions be uniform and that the fluorochrome saturate all available binding sites. Investigators should perform saturation binding curve measurements using the system of interest (Galbraith et al., 1983), and then adhere to these conditions in subsequent experiments. PI is known to bind to plastic surfaces, including the sample-introduction tubing. It is recommended that the flow cytometer be preequilibrated by passing an appropriate concentration of PI through the system before analyzing the samples of interest. Conversely, prior to further experiments, traces of PI can be removed from the flow cytometer by passing a 10% solution of commercial bleach through the system, followed by deionized water. Standards Use of standard fluorescent microspheres for setting up and aligning flow cytometers is critical and should be routinely employed. Furthermore, the authors consider essential the use of internal standards for applications involving comparative analysis of plant nuclear DNA contents. For large-scale screening of ploidy levels, internal standards may be less important. Sequential analysis of unknowns followed by plants of known ploidy status is usually sufficient to affirm cytometer stability. For measurements requiring greater precision— e.g., the detection of aneuploidy, an internal standard is essential. The use of internal standards is also important in analysis of genome size. Various standards can be employed, including CRBCs, trout red blood cells, or plant homogenates having nuclei of known DNA content (Dolezel et al., 1992; Bharathan et al., 1994). For greatest consistency, it is recommended that the DNA contents of these standards be determined using some method that is independent of fluorescence (e.g., the diphenylamine procedure; Burton, 1968; Galbraith et al., 1983). Intraspecies
variation in DNA content has been noted for some plant species (Rayburn et al., 1992) but not others (Baranyi and Greilhuber, 1996; Greilhuber and Obermayer, 1997). Flow cytometric analysis is usually characterized by a high degree of resolution, with CVs ranging from 1% to 3%. It should be noted that this precision is related to the accuracy of individual measurements and does not directly report the reproducibility of DNA-content estimation. It is thus important to perform sufficient independent experiments both to provide statistical significance and to uncover any unexpected variation. Flow cytometry In all cases, it should be remembered that the flow cytometric measurements aim to identify a minor subpopulation of fluorescent nuclei within a cellular homogenate that contains a large population of different particles, all of which scatter light and some of which are autofluorescent (e.g., chloroplasts). For laserbased instruments, it has already been noted that triggering based on fluorescence and not light scatter is required (see Basic Protocol). In some cases (e.g., tobacco), the peak of fluorescent nuclei is well separated from that of other fluorescent particles. This corresponds to the situation where the nuclear DNA content is quite large. In other species of interest, the nuclear DNA content can be much smaller (e.g., Arabidopsis thaliana). In this latter case, the contribution from chlorophyll autofluorescence can overlap PI-induced nuclear fluorescence on one-dimensional histograms. The use of biparametric (orange versus red) displays can be helpful, since the endoreduplicated series of nuclei can be readily identified and gates placed around these nuclei (Galbraith and Lambert, 1995). It is recommended that flow histograms of nuclear fluorescence emission versus time be routinely collected. The major G1 peak of fluorescence should be stable in intensity as a function of time. Aberrant events, discontinuities, and so on can provide an early warning of problems in staining stoichiometries or kinetics, or in flow cytometer function (e.g., a partially blocked flow cell tip) that might otherwise lead to unacceptably high CVs. As a general rule, sufficient events should always be collected to provide unambiguous and statistically significant data. Typically, 5,000 to 20,000 nuclei should be analyzed for each sample.
Nucleic Acid Analysis
7.6.17 Current Protocols in Cytometry
Supplement 2
For determination of the G1 DNA content value, chicken red blood cells are included as an internal standard (Fig. 7.6.1B). In this case, an additional peak is observed (at about channel 100). Since the abscissa is linear, simple division of the peak positions allows calculation of the DNA content of the “unknown” by reference to the known DNA content of CRBCs. Use of N. tabacum as an internal standard is shown in Figure 7.6.2A. Here, the 2C peak for the unknown (Alstroemeria caryophyllacea) has a DNA content larger than that of N. tabacum and considerably larger than that of CRBCs
Anticipated Results A typical uniparametric display of the population fluorescence emission of PI-stained nuclei from a tobacco leaf homogenate is given in Figure 7.6.1A (Bharathan et al., 1994). The major peak is from G1 nuclei having a 2C DNA content. A minor peak is observed at twice the fluorescence-intensity value (in this case, at about channel 800), corresponding to the G2 nuclei. S-phase nuclei are found between these two peaks. Integration of the number of nuclei beneath each area of the histogram provides a description of the cell cycle status of the tissue.
A G1
Number of nuclei
1000
750
500
250
S
G2
0
B 750
Number of nuclei
CRBC 500
2C 250
4C 0 0
Analysis of Nuclear DNA Content and Ploidy in Higher Plants
200
400 600 800 Fluorescence (channel number)
1000
Figure 7.6.1 (A) Flow cytometric analysis of tobacco leaf homogenates stained with propidium iodide. (B) Same analysis with inclusion of chicken red blood cells (CRBC) as an internal standard for genome size measurement. In leaf tissue the G1 nuclei, which have a 2C DNA content, greatly outnumber those in S phase and in G2, which have a 4C DNA content.
7.6.18 Supplement 2
Current Protocols in Cytometry
(Bharathan et al., 1994). For accurate estimation of larger genome sizes, use of standards that are close to that of the unknown is advantageous. For situations in which large dynamic ranges are unavoidable, logarithmic amplification and display are helpful (Fig. 7.6.2B). DNA histograms from samples having large genome sizes generally have little background, and the various peaks are discrete and well separated because the intensity of fluorescence,
even employing PI, greatly exceeds that of any other type of particle in the homogenate. For species having much smaller genomes—e.g., Arabidopsis thaliana—the contribution of fluorescent particles can cause problems. Thus, the endogenous orange-red autofluorescence of pigments within Arabidopsis chloroplasts contributes a significant amount of background to the analysis of PI-stained nuclei. This can be ameliorated though use of gating on biparamet-
A N. tabacum 2C
Number of nuclei
600
A. caryophyllacea 2C
400
200 4C 0 0
B
2500
200
400 600 800 Fluorescence (channel number)
1000
A. caryophyllacea 2C
Number of nuclei
2000
1500
N. tabacum 2C
1000
500 4C
4C
0 0
200 400 600 800 Log fluorescence (channel number)
1000
Figure 7.6.2 Analysis of the genome size of Alstroemeria caryophyllacea using N. tabacum as the internal standard and PI as the fluorochrome. (A) Employing linear amplification, only the G1 (2C) peak for A. caryophyllacea is on scale. (B) Employing logarithmic amplification, G1 and G2 nuclei for both species are observed (from Bharathan et al., 1994).
Nucleic Acid Analysis
7.6.19 Current Protocols in Cytometry
Supplement 2
A
600
2C 2C
600 Number of nuclei
400
CRBC
4C
200
8C 0
400
100
150
200
250
4C 200 CRBC
8C 0 0
50
100
150
250
200
B 500
4C Number of nuclei
400
2C
300
200
8C
100
16C
0 0
100
200 300 400 Fluorescence (channel number)
500
Figure 7.6.3 Flow cytometric analysis of nuclear DNA contents in young Arabidopsis thaliana leaf tissues. (A) Use of mithramycin as the fluorochrome and CRBCs as the internal standard to provide a measure of the genome size (insert represents the same sample under logarithmic amplification); from Galbraith et al., 1991. (B) Use of fluorochrome DAPI, Partec buffer, and the Partec PAS II flow cytometer. This gives particularly well separated and discrete peaks of fluorescence; the extensive somatic endoreduplication is readily observed and accurately measured. Analysis of Nuclear DNA Content and Ploidy in Higher Plants
7.6.20 Supplement 2
Current Protocols in Cytometry
400 2C
Number of nuclei
6C 300
3C 4C
200
12C
100
24C 48C 0 0
200
400 600 800 Log fluorescence (channel number)
1000
Figure 7.6.4 Flow cytometric analysis of nuclear DNA content distribution in maize endosperm (28 days after pollination) using PI as the fluorochrome (from Galbraith and Lambert, 1995).
ric displays, as previously described (see Critical Parameters). Alternatively, the emission spectrum of MI or DAPI is sufficiently removed from that of chlorophyll to largely eliminate this background (Fig. 7.6.3). Estimation of the genome size of Arabidopsis requires correction for the differences in AT/GC composition between Arabidopsis and CRBC nuclei (Galbraith et al., 1991). An interesting observation within Arabidopsis is the occurrence of systemic endoreduplication (Galbraith et al., 1991), which is also observed in succulents having small genomes (DeRocher et al., 1990). Endoreduplication is a conspicuous feature of specialized storage tissues, notably the developing maize endosperm (Fig. 7.6.4), and results in nuclei with extraordinarily high C values (Kowles et al., 1990).
Time Considerations Routine DNA analysis on intact plant tissues can be done by one person at a rate of 5 to 6 samples per hr. If the procedures are pipelined appropriately and an autosampling flow cytometer is employed, this can be increased to a maximum of ∼20 samples/hr, since each sample requires ∼2 min of flow cytometric data acquisition.
Literature Cited Arumuganathan, K., and Earle, E.D. 1991. Nuclear DNA content of some important plant species. Plant Mol. Biol. Rep. 9:208-218. Baranyi, M., and Greilhuber, J. 1996. Flow cytometric and Feulgen densitometric analysis of genome size variation in Pisum. Theor. Appl. Genet. 92: 297-307. Bennett, M.D., and Leitch, I.J. 1995. Nuclear DNA amounts in angiosperms. Ann. Bot. 76:113-176. Bharathan, G., Lambert, G.M., and Galbraith, D.W. 1994. Nuclear DNA contents of monocotyledons and related taxa. Am. J. Bot. 81:381-386. Burton, K. 1968. Determination of DNA concentration with diphenylamine. Methods Enzymol. 12:163-168. de Laat, A.M.M., Göhde, W., and Vogelzang, J.D.C. 1987. Determination of ploidy of single plants and plant populations by flow cytometry. Plant Breed. 99:303-307. DeRocher, E.J., Harkins, K.R., Galbraith, D.W., and Bohnert, H.J. 1990. Developmentally-regulated systemic endopolyploidy in succulents with small genomes. Science 250:99-101. Dolezel, J., and Göhde, W. 1995. Sex determination in dioecious plants Melandrium album and M. rubrum using high-resolution flow cytometry. Cytometry 19:103-106. Dolezel, J., Binarova, P., and Lucretti, S. 1989. Analysis of nuclear DNA content in plant cells by flow cytometry. Biol. Plant. 31:113-120. Nucleic Acid Analysis
7.6.21 Current Protocols in Cytometry
Supplement 2
Dolezel, J., Sgorbati, S., and Lucretti, S. 1992. Comparison of three DNA fluorochromes for flow cytometric estimation of nuclear DNA content in plants. Physiol. Plant. 85:625-631. Dolezel, J., Dolezelova, M., and Novak, F.J. 1994. Flow cytometric estimation of nuclear DNA amount in diploid bananas (Musa acuminata and M. balbisiana). Biol. Plant. 36: 351-357. Galbraith, D.W. 1990. Flow cytometric analysis of plant genomes. Methods Cell Biol. 33:549-562. Galbraith, D. W. and Lambert, G. M. 1995. Advances in the flow cytometric characterization of plant cells and tissues. In Flow Cytometric Applications in Cell Culture (M. Al-Rubeai and A. N. Emery, eds.) pp. 311-326. Marcel Dekker, New York. Galbraith, D.W., and Shields, B.A. 1982. The effects of inhibitors of cell wall synthesis on tobacco protoplast development. Physiol. Plant. 55:2530. Galbraith, D. W., Harkins, K. R., Maddox, J. R., Ayres, N. M., Sharma, D. P., and Firoozabady, E. 1983. Rapid flow cytometric analysis of the cell cycle in intact plant tissues. Science 220:10491051. Galbraith, D.W., Harkins, K.R., and Knapp, S. 1991. Systemic endopolyploidy in Arabidopsis thaliana. Plant Physiol. 96:985-989. Godelle, B., Cartier, D., Marie, D., Brown, S.C., and Siljak-Yakovlev, S. 1993. A heterochromatin study demonstrating the non-linearity of fluorometry useful for calculating genomic base composition. Cytometry 14:618-626. Greilhuber, J., and Obermayer, R. 1997. Genome size and maturity group in Glycine max. Heredity 78:547-551. Harkins, K.R., Jefferson, R.A., Kavanaugh, T.A., Bevan, M.W., and Galbraith, D.W. 1990. Expression of photosynthetic-related gene fusions is restricted by cell type in transgenic plants and in transfected protoplasts. Proc. Natl. Acad. Sci. U.S.A. 87:816-820. Kowles, R.V., Srienc, F., and Phillips, R.L. 1990. Endoreduplication of nuclear DNA in the development of maize endosperm. Dev. Genet. 11:125-132. Marie, D., and Brown, S.C. 1993. A cytometric exercise in plant DNA histograms, with 2C values for 70 species. Biol. Cell 78:41-51. Nakamura, D., Tiersch, T.R., Douglass, M., and Chandler, R.W. 1990. Rapid identification of sex in birds by flow cytometry. Cytogenet. Cell Genet. 53:201-205. O’Brien, I. E. W., Baguley, B. C., Morris, B. A. M., Murray, B. G., and Ferguson, I. B. 1996. Chromatin condensation followed by nDNA frag-
mentation indicative of Programmed Cell Death occurs in plants. Cytometry I.S.A.C. Congress XVIII Abstract BD21. Otto, F. 1990. DAPI staining of fixed cells for highresolution flow cytometry of nuclear DNA. In Methods in Cell Biology, Vol. 33. (Z. Darzynkiewicz and H.A. Crissman eds.) pp. 105110, Academic Press, New York. Overton, W.R., and McCoy, J.P. 1994. Reversing the effect of formalin on the binding of propidium iodide to DNA. Cytometry 16:351-356. Pfosser, M. 1989 Improved method for critical comparison of cell cycle data of asynchronously dividing and synchronized cell cultures of Nicotiana tabacum. J. Plant Physiol. 134:741-745. Pfosser M., Amon, A., Lelley, T., and Heberle-Bors, E. 1995. Evaluation of sensitivity of flow cytometry in detecting aneuploidy in wheat using disomic and ditelosomic wheat-rye addition lines. Cytometry 21:387-393. Rayburn, A.L., Auger, J.A., and McMurphy, L.M. 1992. Estimating percentage constitutive heterochromatin by flow cytometry. Exp. Cell . Res. 198:175-178. Sgorbati, S., Levi, M., Sparvoli, E., Trezzi, F., and Lucchini, G. 1986 Cytometry and flow cytometry of 4′,6-diamidino-2-phenylindole (DAPI)-stained suspensions of nuclei released from fresh and fixed tissues of plants. Physiol. Plant. 68:471-476.
Key Reference Galbraith, D. W., Harkins, K. R., Maddox, J. R., Ayres, N. M., Sharma, D. P., and Firoozabady, E. 1983. Rapid flow cytometric analysis of the cell cycle in intact plant tissues. Science 220:10491051. Outlines the basic procedures for rapid analysis of plant nuclei using flow cytometry which, with various modifications, are now employed worldwide for a variety of different applications.
Internet Resources Bennett, M.D., Cox, A.V., and Leitch, I.J. 1997. An giosp erm DNA C-valu es d atab ase. http://www.rbgkew.org.uk/cval/database1.html.
Contributed by David W. Galbraith, Georgina M. Lambert, and Jiri Macas University of Arizona Tucson, Arizona Jaroslav Dolezel De Montfort University Norman Borlaug Centre for Plant Science Olomouc, Czech Republic
Analysis of Nuclear DNA Content and Ploidy in Higher Plants
7.6.22 Supplement 2
Current Protocols in Cytometry
Analysis of DNA Content and BrdU Incorporation
UNIT 7.7
Incubation of cells in medium containing halogenated DNA precursors such as 5-bromodeoxyuridine (BrdU) or 5-iododeoxyuridine (IdU) results in incorporation of these thymidine analogs into strands of DNA during DNA replication (S phase). The incorporated precursors can be detected either cytochemically—e.g., by virtue of a propensity of BrdU to quench the fluorescence of several DNA fluorochromes such as Hoechst 33358 or acridine orange (Latt, 1973, 1977)—immunocytochemically using polyclonal or monoclonal antibodies developed against this precursor (Gratzner, 1982; Dolbeare et al., 1983; Dolbeare and Selden, 1994)—or by selective photolysis of BrdU-containing DNA followed by fluorochrome labeling of the photolysis-generated DNA strand breaks (Li and Darzynkiewicz, 1995). During the past two decades, techniques based on detection of BrdU incorporation by flow cytometry have replaced the cumbersome autoradiographic methodology utilizing tritiated thymidine for analysis of cell-cycle kinetics. Immunocytochemical detection of BrdU, combined with simultaneous measurement of cellular DNA content followed by bivariate data analysis (Dolbeare et al., 1983), is the most common technique of cell-cycle analysis based on BrdU incorporation. It can be used in the clinic to evaluate parameters of cell proliferation in tumors—e.g., labeling index (LI), S-phase duration (Ts), or tumor potential doubling time (Tpot; Begg et al., 1985). Continuous exposure to BrdU results in labeling of all proliferating cells and thus enables one to estimate the tumor growth fraction (or conversely, a fraction of noncycling cells), while pulse-chase experiments can be used to measure cell-cycle kinetics (for review see Dolbeare and Selden, 1994). BrdU incorporated into double-stranded DNA in nuclear chromatin is inaccessible to anti-BrdU antibody. Immunocytochemical detection of BrdU, therefore, requires partial DNA denaturation and strand separation to make the incorporated precursor accessible. In practice, DNA can be denatured by exposure of cells to heat or strong acid. Because the intercalating dye propidium iodide (PI) is generally used to counterstain DNA, a fraction of cellular DNA should remain nondenatured to be stainable with PI (see Background Information). Furthermore, dissociation of histones, which can be accomplished by pretreating cells with 0.1 M HCl and suspending cells in low-ionic-strength solutions during thermal DNA denaturation, enhances both accessibility of BrdU to antibody and denaturability of DNA (Moran et al., 1985). The original (heat-denaturation) method of Dolbeare et al., (1983)—with modifications by Moran, et al. (1985), Beisker et al. (1987), and Gray et al. (1990)—is presented below (see Basic Protocol 1). A variant procedure using acid denaturation of DNA in place of heat denaturation is also presented (see Basic Protocol 2). Basic Protocols 1 and 2 use FITC-conjugated anti-BrdU antibody; for situations where directly conjugated anti-BrdU antibody is not available, an indirect labeling procedure is also presented (see Alternate Protocol 1). Finally, a procedure is outlined that utilizes heat denaturation as in Basic Protocol 1, but without the repeated centrifugations (see Alternate Protocol 2). CAUTION: BrdU is potentially mutagenic and can be absorbed through skin or by inhalation. Therefore, when handling BrdU, wear suitable protective clothing, including gloves and eye and face protection. NOTE: BrdU solutions are light-sensitive and should be protected from light during storage and handling. Nucleic Acid Analysis Contributed by Zbigniew Darzynkiewicz and Gloria Juan Current Protocols in Cytometry (1997) 7.7.1-7.7.9 Copyright © 1997 by John Wiley & Sons, Inc.
7.7.1 Supplement 2
BASIC PROTOCOL 1
ANALYSIS OF DNA CONTENT AND BrdU INCORPORATION FOLLOWING THERMAL DENATURATION OF DNA The critical point in analysis of BrdU incorporation is the DNA-denaturation step. Optimal conditions for DNA denaturation vary between cell types, depending on differences in nuclear chromatin structure. Materials 1 mg/ml BrdU stock solution in distilled water (prepare fresh) Cells to be labeled and appropriate serum-containing tissue culture medium (APPENDIX 3B) Phosphate-buffered saline (PBS; APPENDIX 2A) 70% ethanol, ice-cold 0.1% Triton X-100/0.1 M HCl, ice-cold (prepare fresh) DNA denaturation buffer (see recipe) Antibody diluting buffer (see recipe) 0.2 to 0.5 µg/100 µl FITC-conjugated anti-BrdU antibody (Becton Dickinson Immunocytometry) in antibody diluting buffer (see recipe for buffer) Propidium iodide (PI) staining solution (see recipe) Tabletop centrifuge 15-ml tubes (preferably glass) 85° to 95°C water bath 25-G needle Flow cytometer with 488-nm laser line filter or BC 12 blue filter as well as 530 ± 20–nm bandpass filter and 620-nm long-pass filter Additional reagents and equipment for trypsinizing tissue culture cells (APPENDIX 3B) Label cells with BrdU and prepare cell suspension for fixation 1. Add 1 mg/ml BrdU stock to cell suspension for a final concentration of 10 to 30 µg/ml. Incubate for the appropriate period of time under light-proof conditions (e.g., by wrapping the tissue culture flasks in aluminum foil). Duration of the labeling may vary depending on the design of the experiment (pulse, pulse-chase, or prolonged continuous labeling). Pulse incubations are generally of 30 or 60 min duration. If the cells are labeled in vivo, remove the tissue, dissociate the cells (e.g., by enzymatic or mechanical dissociation), and suspend cells in PBS. For discussion of necessary controls, see Anticipated Results.
2a. For cells growing in suspension or hematologic samples: Centrifuge cells 5 min at 300 × g, room temperature, and remove the supernatant. Rinse cells by resuspending in room temperature PBS and centrifuging again for 5 min at 300 × g. Remove the supernatant and resuspend the cells in room temperature PBS at ∼1 × 106 cells/ml. 2b. For cells growing attached to petri dishes or flasks: Collect cells from dishes or flasks by trypsinization (APPENDIX 3B) and pool the trypsinized cells with the cells floating in the medium (the latter consisting of detached mitotic, apoptotic, and dead cells). Centrifuge cells 5 min at 300 × g, room temperature, and remove the supernatant. Rinse cells by resuspending in serum-containing medium and centrifuging again for 5 min at 300 × g. Remove the supernatant and resuspend the cells in room temperature PBS at ∼1 ×106 cells/ml. Analysis of DNA Content and BrdU Incorporation
Medium containing serum should be used to inactivate trypsin; other means of trypsin inactivation such as addition of protease inhibitors may be used instead.
7.7.2 Supplement 2
Current Protocols in Cytometry
2c. For cells isolated from solid tumors: Rinse cells free of any enzyme used for cell dissociation by centrifuging 5 min at 300 × g, room temperature, removing the supernatant, adding PBS, then centrifuging again at 300 × g and removing the supernatant. Resuspend cells in room temperature PBS at ∼1 ×106 cells/ml. In the final PBS suspension the cells should be well dispersed (not in aggregates).
Fix cells and pretreat with HCl/Triton X-100 3. With a Pasteur pipet, transfer 1-ml aliquots of the cell suspension (containing 1–2 × 106 cells) into 15 ml-tubes (preferably glass), each containing 10 ml of ice-cold 70% ethanol. Fix by incubating for the appropriate time period at 4°C. Optimal time of fixation (storage) in ethanol at 4°C may vary from 4 hr to several months (usually 1 to 2 days). Rapid injection of the cell suspension into the cold fixative minimizes cell clumping. Glass tubes are preferred because not all plastic tubes withstand the heating at 85° to 90°C required for thermal DNA denaturation.
4. Centrifuge cells 5 min at 300 × g, 4°C, aspirate supernatant, then resuspend cell pellet in 1 ml ice-cold 0.1% Triton X-100/0.1 M HCl. Incubate 1 min on ice, then centrifuge cells again for 5 min at 300 × g, 4°C. 5. Aspirate supernatant, then drain pellet thoroughly and resuspend in 5 ml DNA denaturation buffer. Centrifuge 5 min at 300 × g, room temperature, then resuspend pellet in 1 ml DNA denaturation buffer. Denature DNA 6. Heat cells 5 min at 85° to 95°C, then place on ice for 5 min. Optimal denaturation temperature may vary for different cell types (see Critical Parameters).
7. Add 5 ml of antibody diluting buffer, then centrifuge 5 min at 300 × g, room temperature. 8. Aspirate supernatant, then vortex gently to loosen the cell pellet. If clumping occurs, disperse clumps by pipetting through a 25-G needle. Label cells with FITC-conjugated anti-BrdU antibody, stain with PI, and measure fluorescence 9. Add 100 µl of 0.2 µg/100 µl FITC-conjugated anti-BrdU antibody in antibody diluting buffer to the cell pellet. Vortex gently, then incubate 30 min at room temperature. This step may vary depending on the antibody. Follow the instructions provided by the supplier regarding the dilution, time, and temperature of incubation with anti-BrdU antibody.
10. Add 5 ml antibody diluting buffer. Centrifuge 5 min at 300 × g, room temperature, and aspirate the supernatant. 11. Resuspend cells in 2 ml PI staining solution. Leave cells in this solution for 30 min at room temperature in the dark prior to measuring fluorescence. 12. Set up the flow cytometer for excitation of PI and FITC with blue light, using either a 488-nm laser line filter or a BC 12 blue filter when the source of illumination is a mercury arc or xenon lamp. Collect green emission (FITC) with a 530 ± 20–nm bandpass filter and red emission (PI) with a 620-nm long-pass filter. Measure the BrdU-associated green fluorescence of FITC and DNA-associated red fluorescence of PI.
Nucleic Acid Analysis
7.7.3 Current Protocols in Cytometry
Supplement 2
BASIC PROTOCOL 2
ANALYSIS OF DNA CONTENT AND BrdU INCORPORATION FOLLOWING ACID DENATURATION OF DNA While heat-induced DNA denaturation is most effective for some cell types, in others, optimal denaturation is achieved after the cells are exposed to acid (HCl). Materials 1 mg/ml BrdU stock solution in distilled water (prepare fresh) Cells to be labeled and appropriate serum-containing tissue culture medium (APPENDIX 3B) Phosphate-buffered saline (PBS; APPENDIX 2A) 70% ethanol, ice-cold 2 M HCl Phosphate/citric acid buffer, pH 7.4 (see recipe) 0.2 to 0.5 µg/100 µl FITC-conjugated anti-BrdU antibody (Becton Dickinson Immunocytometry) in antibody diluting buffer (see recipe for buffer) Antibody diluting buffer (see recipe) Propidium iodide (PI) staining solution (see recipe) Flow cytometer with 488-nm laser line filter or BC 12 blue filter as well as 530 ± 20–nm bandpass filter and 620-nm long-pass filter Label cells with BrdU, fix cells, and denature DNA with acid 1. Label cells with 10 to 30 µg/ml BrdU (added as 1 mg/ml stock), rinse with PBS, and fix with 70% ethanol (see Basic Protocol 1, steps 1 to 3). For discussion of necessary controls, see Anticipated Results.
2. Centrifuge cells 5 min at 300 × g, room temperature, and resuspend cell pellet in 0.25 ml of 2 M HCl. Let stand 20 min at room temperature. Optimal acid concentration for denaturation may vary for different cell types (see Critical Parameters).
3. Add 5 ml of phosphate/citric acid buffer, pH 7.4, centrifuge again for 5 min at 300 × g, room temperature, and drain off the supernatant. Resuspend cells in 5 ml of phosphate/citric acid buffer, pH 7.4. Centrifuge as before and remove the supernatant. Label cells with FITC-conjugated anti-BrdU antibody, stain with PI, and measure fluorescence 4. Add 100 µl of 0.2 µg/100 µl FITC-conjugated anti-BrdU antibody in antibody diluting buffer to the cell pellet. Vortex gently, then incubate 30 min at room temperature. This step may vary depending on the antibody. Follow the instructions provided by the supplier regarding the dilution, time, and temperature of incubation with anti-BrdU antibody.
5. Add 5 ml antibody diluting buffer. Centrifuge 5 min at 300 × g, room temperature, and aspirate the supernatant. 6. Resuspend cells in 2 ml PI staining solution. Leave cells in this solution for 30 min at room temperature in the dark prior to measuring fluorescence.
Analysis of DNA Content and BrdU Incorporation
7. Set up the flow cytometer for excitation of PI and FITC with blue light, using either a 488-nm laser line filter or a BC 12 blue filter when the source of illumination is a mercury arc or xenon lamp. Collect green emission (FITC) with a 530 ± 20–nm bandpass filter and red emission (PI) with a 620-nm long-pass filter. Measure the BrdU-associated green fluorescence of FITC and DNA-associated red fluorescence of PI.
7.7.4 Supplement 2
Current Protocols in Cytometry
INDIRECT IMMUNOCYTOCHEMICAL METHOD FOR BrdU DETECTION If anti-BrdU antibody that is directly conjugated to FITC is unavailable, this indirect immunocytochemical method, which involves use of an unlabeled mouse anti-BrdU antibody which is in turn detected with a FITC-conjugated goat anti-mouse antibody, may be used.
ALTERNATE PROTOCOL 1
Additional Materials (also see Basic Protocols 1 and 2) 0.2 to 0.5 µg/100 µl mouse anti-BrdU antibody (unlabeled) in antibody diluting buffer (see recipe for buffer) 0.5 to 1 µg/ml FITC-conjugated goat anti-mouse antibody in antibody diluting buffer 1. Label cells with BrdU, fix cells, and denature DNA (see Basic Protocol 1, steps 1 to 8, or see Basic Protocol 2, steps 1 to 3). 2. Add 100 µl of antibody diluting buffer containing 0.2 to 0.5 µg unlabeled mouse anti-BrdU antibody to the pellet of ∼1 × 106 cells. Vortex gently, then incubate 30 min at room temperature. This step may vary depending on the antibody. Follow the instructions provided by the supplier regarding the dilution, time, and temperature of incubation with anti-BrdU antibody.
3. Add 5 ml antibody diluting buffer, then centrifuge 5 min at 300 × g, room temperature. Remove supernatant. 4. Resuspend cell pellet (∼1 × 106 cells) in 100 µl of antibody diluting buffer containing 0.5 to 1 µg FITC-conjugated goat anti-mouse antibody. Vortex gently, then incubate 30 to 60 min at room temperature. Place on ice for 5 min to cool. 5. Add 5 ml antibody diluting buffer, then centrifuge 5 min at 300 × g, room temperature. Remove supernatant. 6. Stain cells with PI , set up flow cytometer, and measure cell fluorescence (see Basic Protocol 1, steps 11 and 12). “WASHLESS” TECHNIQUE FOR BrdU DETECTION The “washless” procedure (Larsen et al., 1991; Larsen, 1994) combines BrdU labeling with an immunocytochemical BrdU-detection protocol that does not involve repeated centrifugations. Advantages of this approach are its rapidity and the prevention of cell loss that otherwise occurs during centrifugations. The original versions of this procedure, which include washless staining of unfixed nuclei and DNA denaturation by acid, are presented in Larsen et al. (1991). The method below is a modification of these methods, combining thermal (instead of acid) denaturation with washless immunocytochemical detection of BrdU. For materials, see Basic Protocol 1.
ALTERNATE PROTOCOL 2
1. Label cells with BrdU and fix with ethanol (see Basic Protocol 1, steps 1 to 3). 2. Centrifuge ethanol-fixed cells 5 min at 300 × g, room temperature. Drain alcohol thoroughly. 3. Add 100 µl distilled water to the cell pellet. Vortex gently. 4. Heat cells 5 min at 85° to 95°C. Optimal denaturation temperature may vary for different cell types (see Critical Parameters).
5. Add 100 µl of 0.2 µg/100 µl FITC-conjugated anti-BrdU antibody in antibody diluting buffer. Vortex gently.
Nucleic Acid Analysis
7.7.5 Current Protocols in Cytometry
Supplement 4
6. Incubate 30 min at room temperature. 7. Resuspend cells in 2 ml PI staining solution. Leave cells in this solution for 30 min at room temperature in the dark with occasional gentle mixing. 8. Set up flow cytometer and measure fluorescence (see Basic Protocol 1, step 12). REAGENTS AND SOLUTIONS Use distilled, deionized water for the preparation of all buffers. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Antibody diluting buffer PBS (APPENDIX 2A) containing: 0.1% (v/v) Triton X-100 1% (w/v) bovine serum albumin (BSA) Prepare fresh DNA denaturation buffer 8.7 mg NaCl (0.15 mM final) 4.4 mg trisodium citrate dihydrate (15 µM final) 1000 ml H2O Store up to 2 months at 4°C Phosphate/citric acid buffer, pH 7.4 Mix 182 ml of 0.2 M Na2HPO4 with 18 ml of 0.1 M citric acid. Check pH, which should be 7.4. If not, adjust with 1 M HCl or 1 M NaOH. Store up to 2 months at 4°C. Propidium iodide (PI) staining solution 100 µg PI 1 mg RNase A 10 ml PBS (APPENDIX 2A) Prepare fresh Keep in dark in aluminum-foil-wrapped tubes or bottles during use COMMENTARY Background Information
Analysis of DNA Content and BrdU Incorporation
Simultaneous staining of DNA with the intercalating dye PI and detection of incorporated BrdU by antibody requires the presence of both double-stranded DNA (to be stainable with PI) and denatured DNA (to have incorporated BrdU accessible to antibody). A variety of methods and modifications have been proposed to ensure that the degree of DNA denaturation is optimal for simultaneous staining with PI and detection of BrdU (for reviews see Dolbeare and Selden, 1994, and Gray et al., 1990). Alternatives to heat- or acid-induced DNA denaturation as presented in this unit include partial DNA cleavage with exonuclease III or EcoRI, which expose BrdU to the antibody and still preserve sections of DNA in double-stranded conformation, reactive with PI (Dolbeare and Gray, 1988). This procedure, therefore, although more complex and expensive, can be
used in situations when it is necessary to preserve (e.g., for immunocytochemical detection) other antigens that are degraded, extracted, or denatured by heat or acid treatment. Cytochemical methods of BrdU detection (Latt, 1973, 1977; Darzynkiewicz et al., 1978) are not discussed here. These methods are reviewed by Crissman and Steinkamp (1990). Several of these techniques, including simultaneous differential detection of BrdU and iododeoxyuridine (IdU), are presented elsewhere (Darzynkiewicz et al., 1994). Still another approach for BrdU detection is based on selective photolysis of DNA containing the incorporated BrdU. The photolytically generated DNA strand breaks are subsequently labeled with fluoresceinated nucleotides in a reaction catalyzed by exogenous terminal transferase, referred to as DNA strand break induction by photolysis (SBIP; Li and Dar-
7.7.6 Supplement 4
Current Protocols in Cytometry
zynkiewicz, 1995; Li et al., 1996). Because no DNA-denaturation step is required in SBIP, the procedure is applicable in combination with immunocytochemical analysis—e.g., for cell immunophenotyping combined with analysis of DNA replication. This method can also be combined with simultaneous detection of apoptotic cells (Li et al., 1996).
Critical Parameters and Troubleshooting The critical step in this procedure is induction of partial DNA denaturation. This step often results in cell damage, aggregation (clumping), and significant cell loss—predominantly resulting from cell attachment to surface of the tubes. Furthermore, there are significant differences between cell types in respect to sensitivity of their DNA to denaturation, depending on differences in structure of their nuclear chromatin (Darzynkiewicz, 1990). Thus, while induction of DNA denaturation by acid may prove to be satisfactory with one cell type, it may fail with another. For optimal results some cell types require higher acid concentration (4 M HCl ) than that indi-
cated in Basic Protocol 2 (2 M HCl). Likewise, in the case of thermal DNA denaturation, for optimal results some cell types require heating at 85°C, others at 90°C or 95°C, and still others at 100°C. Generally, when DNA denaturation is too extensive, the detection of BrdU is satisfactory but the stainability of the cells with PI (red fluorescence intensity) is markedly decreased. This coincides with poor resolution of DNA-content frequency histograms. Conversely, when DNA denaturation is inadequate, DNA-content frequency histograms are satisfactory but the detection of the incorporated BrdU is poor. Cell adherence to tube surfaces, especially during the DNA-denaturation step (heat or acid treatment), is another critical factor. Using polypropylene tubes or siliconizing glass tubes minimizes cell attachment. Likewise, smaller microcentrifuge tubes with reduced surface area may be preferred, to reduce cell loss.
Anticipated Results Figure 7.7.1 illustrates a typical bivariate histogram of DNA content versus BrdU incorporation based on immunocytochemical detec-
103
S
BrdU incorporation
102
G2 101
G1 100 0
200
400 DNA content
600
Figure 7.7.1 Typical bivariate distribution of cells showing incorporation of BrdU versus DNA content. Exponentially growing HL-60 cells were incubated in the presence of BrdU for 30 min and then subjected to the procedure described in Basic Protocol 1.
Nucleic Acid Analysis
7.7.7 Current Protocols in Cytometry
Supplement 2
tion of BrdU. DNA-content analysis allows one to position the cell in the respective phase of the cell cycle (G0/1 versus S versus G2/M) while detection of BrdU allows identification of cells in which DNA is replicating. The presence of cells with a DNA content typical of S-phase cells, namely with DNA index between 1.0 and 2.0, but which do not incorporate BrdU, generally indicates that such cells were dead during the incubation with BrdU. Situations exist, however, in which cells may be arrested in S phase, either as a result of drug treatment or poor exposure to oxygen (or growth factors in some tumors), and thus may be alive but remain unlabeled after incubation with anti-BrdU antibody. Progression of the labeled cells through the cell cycle after the pulse exposure to the precursor allows one to measure the kinetic parameters of the cell populations (Begg et al., 1985; Crissman and Steinkamp, 1990; Gray et al., 1990). In the case of pulse exposure to BrdU (30 to 60 min), the unlabeled G1 cells serve as a negative control. G1 cells are expected to have minimal green fluorescence; elevated staining indicates cell autofluorescence or nonspecific binding of anti-BrdU antibody. Another negative control is provided by cells that were not incubated with BrdU. The latter control is necessary when incubations with BrdU were relatively long and cells labeled with the precursor are expected not only in S, but also in other phases of the cell cycle. In the methods utilizing direct immunostaining, a FITC-conjugated irrelevant isotype antibody (e.g., anti-thyroglobulin) can also be used as an additional negative control (Larsen et al., 1991). It is a good practice to have a positive standard as well. A large batch of exponentially growing cells incubated with BrdU—e.g., for 1 hr and tested for BrdU incorporation—can be stored in fixative and used every time a new cell type or different conditions of cell labeling are used.
Time Considerations
Cell fixation requires ∼15 min for Basic Protocols 1 and 2. Cells must remain in fixative at least 4 hr. Cell staining takes ∼2 hr. Overall time for cell preparation and staining is ∼1.5 hr using the washless procedure (see Alternate Protocol 2).
Literature Cited Begg, A.C., McNally, N.J., Shrieve, D.C., and Karcher, H. 1985. A method to measure the duration of DNA synthesis and the potential doubling time from a single sample. Cytometry 6:620- 626. Beisker, W., Dolbeare, F., and Gray, J.W. 1987. An improved immunocytochemical procedure for high sensitivity detection of incorporated bromodeoxyuridine. Cytometry 8:235-239. Crissman, H.A. and Steinkamp, J.A. 1994. Cytochemical techniques for multivariate analysis of DNA and other cellular constituents. In Flow Cytometry and Sorting (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 227247. Wiley-Liss, New York. Darzynkiewicz, Z. 1990. Probing nuclear chromatin by flow cytometry. In Flow Cytometry and Sorting (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 315-340. Wiley-Liss, New York. Darzynkiewicz, Z., Andreeff, M., Traganos, F., and Melamed M.R. 1978. Discrimination of cycling and noncycling lymphocytes by BudR-suppressed acridine orange fluorescence in a flow cytometric system. Exp. Cell Res. 115:31-35. Darzynkiewicz, Z., Robinson, J.P., and Crissman, J.P. (eds.) 1994. Flow Cytometry, 2nd ed, part A. Academic Press, San Diego. Dolbeare, F. and Gray, J.W. 1988. Use of restriction endonuclease and exonuclease III to expose halogenated pyrimidines to immunocytochemical staining. Cytometry 9:631-635. Dolbeare, F. and Selden, J.R. 1994. Immunocytochemical quantitation of bromodeoxyuridine: Application to cell-cycle kinetics. Methods Cell Biol. 41:298-316. Dolbeare, F., Gratzner, H., Pallavicini, M., and Gray, J.W. 1983. Flow cytometric measurement of total DNA content and incorporated bromodeoxyuridine. Proc. Natl. Acad. Sci. U.S.A. 80:5573-5577. Gratzner, H.G. 1982. Monoclonal antibody to 5bromo-2-deoxyuridine: A new reagent for detection of DNA replication. Science 218:747-748. Gray, J.W., Dolbeare, F., and Pallavicini, M.G. 1990. Quantitative cell cycle analysis. In Flow Cytometry and Sorting (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 445-467. WileyLiss, New York. Larsen, J.R. 1994. Washless procedures for nuclear antigen detection. Methods Cell Biol. 41:377388. Larsen, J.K., Christensen, I.J., Christiansen, J., and Mortensen, T. 1991. Washless double staining of unfixed nuclei for flow cytometric analysis of DNA and a nuclear antigen (Ki-67 or bromodeoxyuridine). Cytometry 12:429-437.
Analysis of DNA Content and BrdU Incorporation
7.7.8 Supplement 2
Current Protocols in Cytometry
Latt, S.A. 1973. Microfluorimetric detection of deoxyribonucleic acid replication in human metaphase chromosomes. Proc. Natl. Acad. Sci. U.S.A. 70:3395-3399. Latt, S.A. 1977. Fluorometric detection of deoxyribonucleic acid synthesis: Possibilities for interfacing bromodeoxyuridine dye techniques with flow cytometry. J. Histochem. Cytochem. 25:913-926. Li, X. and Darzynkiewicz, Z. 1995. Labelling DNA strand breaks with BrdUTP. Detection of apoptosis and cell proliferation. Cell Prolif. 28:571579.
Moran, R., Darzynkiewicz, Z., Staiano-Coico, L., and Melamed, M.R. 1985. Detection of 5-bromodeoxyuridine (BrdUrd) incorporation by monoclonal antibodies: Role of DNA denaturation step. J. Histochem. Cytochem. 33:821-827.
Contributed by Zbigniew Darzynkiewicz and Gloria Juan New York Medical College Valhalla, New York
Li, X., Melamed, M.R., and Darzynkiewicz, Z. 1996. Detection of apoptosis and DNA replication by differential labeling of DNA strand breaks with fluorochromes of different color. Exp. Cell Res. 222:28-37.
Nucleic Acid Analysis
7.7.9 Current Protocols in Cytometry
Supplement 2
Analysis of DNA Denaturation
UNIT 7.8
Sensitivity of DNA to denaturation (i.e., the tendency of double-stranded DNA to unwind into the single-stranded conformation) is altered in cell differentiation and apoptosis, as a cytotoxic effect of chemotherapeutic drugs (intercalators) on target tumor cells, and between different stages of the cell cycle. Measurement of DNA sensitivity to denaturation in situ can therefore be used to obtain information about changes in nuclear chromatin or to identify cells with different chromatin structure. This unit presents a flow cytometric procedure for measuring DNA denaturation that utilizes the metachromatic property of acridine orange (AO). It circumvents the limitations of classical biochemical methods of studying DNA denaturation, which destroy higher orders of chromatin structure and, since they are applied to bulk cell populations, cannot detect variations between cells. DIFFERENTIAL STAINING OF SINGLE- VERSUS DOUBLE-STRANDED DNA WITH ACRIDINE ORANGE
BASIC PROTOCOL
In this protocol cells are fixed in 70% ethanol (apoptotic cells are briefly prefixed in 1% formaldehyde). In addition to fixation, ethanol permeabilizes cells, making DNA accessible to the dye. The cells are then centrifuged to remove the fixative, treated with RNase, and then with 0.1 M HCl to denature DNA, and stained with acridine orange (AO) at pH 2.6. The low pH of the staining solution precludes DNA renaturation, which would otherwise occur. AO differentially stains the nondenatured (double-stranded; green fluorescence) versus denatured (single-stranded; red fluorescence) sections of DNA. Because DNA denaturation is impeded at low temperatures, both the HCl and staining solutions must be at room temperature rather than ice cold when added to the cells. Materials Cells to be stained PBS, pH 7.4 (APPENDIX 2A) 70% ethanol, 0° to 4°C 1% formaldehyde in PBS, pH 7.4 AO stock solution (see recipe) AO staining solution (see recipe) RNase A solution (see recipe) 0.1 M HCl (preferably stored in an automatic dispensing pipettor bottle set to dispense 0.5 ml) 5-ml glass or plastic test tubes 15-ml glass or plastic (preferably polypropylene) centrifuge tubes Clinical centrifuge Ice-bath hemacytometer Flow cytometer equipped either with 488- or 457-nm argon ion laser or with mercury arc or xenon lamp and BC12 blue filter as fluorescence excitation source, and with appropriate filters (530 ± 30–nm bandpass filter and 640-nm long-pass filter for green and red fluorescence, respectively) Additional reagents and equipment for cell culture and trypsinization (APPENDIX 3B) Prepare cell suspension 1a. For cells growing in suspension, hematologic samples, or cells isolated from solid tumors: Pellet the cells by centrifuging 5 min at 300 × g, 0° to 4°C. Rinse once by resuspending in PBS and centrifuging 5 min at 300 × g, room temperature; perform Nucleic Acid Analysis Contributed by Zbigniew Darzynkiewicz and Gloria Juan Current Protocols in Cytometry (1998) 7.8.1-7.8.9 Copyright © 1998 by John Wiley & Sons, Inc.
7.8.1 Supplement 3
a cell count on an aliquot before the spin (APPENDIX 3A). Resuspend in PBS at ∼106 cells/ml. When working with tumor cells, in the final suspension in PBS the cells must be well dispersed (not in aggregates) and their density should not exceed 5 × 106 cells/ml. In this case the rinse serves to remove any enzyme used for cell dissociation.
1b. For adherent cells: Collect cells from flasks or petri dishes by trypsinization (APPENDIX 3B) and pool the trypsinized cells with cells floating in the medium by centrifuging 5 min at 300 × g, room temperature (the latter consist mostly of detached mitotic and dead cells). Cells can also be detached using a rubber policeman. Rinse once by resuspending in medium containing serum and centrifuging again as before; perform a cell count on an aliquot before the spin. Resuspend cells in PBS at ∼106 cells/ml. Serum inactivates the trypsin. Another means of accomplishing this is addition of protease inhibitors (e.g., soybean trypsin inhibitor).
Fix cells 2. If preservation and identification of apoptotic cells is desired: On ice, mix 1 ml of cell suspension from step 1 (∼106 cells/ml) with 10 ml of 1% formaldehyde in PBS at pH 7.4, and let sit 15 min. Pellet the cells and rinse with PBS as in step 1, then proceed to step 3. This brief prefixing in formaldehyde cross-links the degraded DNA with other cell constituents, thereby preventing the extraction that would otherwise occur during the subsequent acid treatment.
3. With a Pasteur pipet transfer 1 ml cell suspension into 15-ml glass tubes containing 10 ml of 70% ethanol at 0° to 4°C, injecting the cells below the fluid surface. Rapid injection of the cell suspension into the cold fixative minimizes cell clumping; do not layer cells on the surface and mix. Addition in reversed order (ethanol into cell suspension) results in extensive cell loss from adherence to the glass surface and aggregation. Cells may be stored in ethanol at 4°C anywhere from 2 hr to several months.
Stain cells 4. Centrifuge cells in fixative 5 min at 300 × g, room temperature. Decant the supernatant and resuspend the cell pellet in 1 ml PBS. Add 100 µl RNase A solution (100 µg). Incubate 20 min at 37°C. 5. Add 5 ml of PBS, centrifuge 5 min at 300 × g, room temperature, and resuspend cell pellet in 0.5 ml of PBS. Transfer a 0.2-ml aliquot of cell suspension to a small (e.g., 5-ml) tube. Add 0.5 ml of 0.1 M HCl to this cell suspension. After 30 sec add 2.0 ml AO staining solution. Treat cells with 0.1 M HCl and stain with AO at room temperature. At lower temperature DNA denaturation is incomplete, so the HCl and AO solutions should be brought to room temperature before use.
Perform flow cytometry and data analysis 6. Set up and adjust flow cytometer for excitation with blue light using the 488 or 457 nm argon-ion laser line, or using a BC12 blue optical filter if the illumination source is a mercury arc or xenon lamp. 7. Collect green emission with a 530 ± 30 nm bandpass filter and red emission with a 640 nm long-pass filter. Analysis of DNA Denaturation
Acquire either as red versus green fluorescence intensities of individual cells, or as total fluorescence (sum of red plus green) versus the ratio of red to total fluorescence (Dar-
7.8.2 Supplement 3
Current Protocols in Cytometry
A
B 80
Total fluorescence
Green fluorescence
80
0
0
G1 S G2
M
80
G2, M S G1 A
B 0
0
αt
0.8
Red fluorescence
Figure 7.8.1 Characteristic AO staining pattern of exponentially growing Friend erythroleukemia cells after incubation with RNase and partial DNA denaturation by HCl. The cells were processed and stained as described in Basic Protocol 1. The left panel shows cell distribution with respect to red versus green fluorescence; the right panel represents the total (red plus green) fluorescence versus αt, where αt represents the ratio of red to total fluorescence. G1 cells may divide into A and B subpopulations, as shown (dashed skewed line). The DNA of early G1 (G1A) postmitotic cells is more denatured (higher αt) than that of late G1 (G1B) cells. Mitotic (M) cells have the highest αt and can easily be distinguished from G2 cells.
zynkiewicz, 1994). The latter ratio, called αt, permits the analog signals from green and red photomultipliers to be added and then respectively divided, through use of a specially designed electronic circuit board (e.g., such as are used in some flow cytometers for compensation of the signals when the measured emission spectra overlap); the total fluorescence and the αt ratio are then digitized and recorded in listmode. Alternatively, software that transforms the red and green fluorescence intensities of individual cells to the total fluorescence and αt values of the same cells can be used (Figs. 7.8.1, 7.8.2, and 7.8.3). Such software is commercially available from Phoenix Flow Systems.
8. Measure cell fluorescence in flow cytometer. The fluorescence pattern of the cells will remain stable for several hours as long as the sample is kept at room temperature, protected from light.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
AO staining solution Mix 90 ml of 0.1 M citric acid with 10 ml of 0.2 M Na2HPO4. (Final pH should be 2.6; if not, adjust with 1 M HCl or 1 M NaOH.) Add 0.6 ml AO stock solution (see recipe) to 100 ml of this buffer (6 µg/ml AO final concentration). Store in the dark at 4°C (stable several months). For convenience, store the AO staining solution in a dark automatic-dispensing pipettor bottle set to dispense at 2.0 ml.
AO stock solution Dissolve acridine orange (high purity, chromatographically tested; available from Molecular Probes) in water to a final dye concentration of 1 mg/ml. Store in the dark at 4°C (stable several months).
Nucleic Acid Analysis
7.8.3 Current Protocols in Cytometry
Supplement 3
A
B
G0
T n fluo
resce
nce
Re
d
flu
or
es
ce
nc
e
Gree
G1
C
G1
D G0
G1 S
G2
M
Figure 7.8.2 Changes in stainability with AO of human peripheral blood lymphocytes mitogenically stimulated with phytohemagglutinin (PHA). (A) Unstimulated cells. (B) Cells stimulated with PHA for 18 hr. (C) Cells stimulated for 3 days. (D) Cells stimulated for 3 days, with vinblastine included in the culture for the final 6 hr to cause mitotic arrest. Cells were stained as described in Basic Protocol 1. Transition of cells from G0 to G1 is associated with an increase in green fluorescence and a decrease in red—i.e., a decrease in DNA denaturability (cells in transition T, panel B). A population of mitotic cells (M) becomes apparent in the vinblastine-treated culture.
RNase A solution Dissolve 1 mg RNase A (Sigma; 50 to 100 Kunitz units/mg) in 1 ml of H2O. Boil 5 min to inactivate any possible contaminating DNase activity. IMPORTANT NOTE: The robust nature of RNase makes it extremely troublesome in molecular biology laboratories. Utmost caution should be exercised to prevent the powder from becoming airborne or the solution from contaminating shared counters or equipment.
COMMENTARY Background Information
Analysis of DNA Denaturation
In aqueous medium at physiological pH and ionic strength, DNA isolated from cell nuclei has double-stranded (ds) conformation. Exposure to heat, acid, or base causes the two strands to unwind and separate into a single-stranded (ss) conformation. This process, known as DNA denaturation or melting, results from destruction of the hydrogen bonding between the paired bases of the opposite strands. Sensitivity of free DNA in solution to denaturation depends on its GC/AT ratio, because the GC pair, with its additional hydrogen bond, confers greater stability than does AT. In nuclear chro-
matin, DNA is further stabilized by interactions with histones and other proteins (van Holde, 1989). Studies on the stability of DNA in chromatin therefore provide insight into chromatin structure, making it possible to discern DNA/protein interactions that stabilize the double helix (Rigler, 1966; Subirana, 1973; Darzynkiewicz and Kapuscinski, 1990). Classical biochemical methods for the study of DNA denaturation in chromatin are based on measurement of changes in UV light absorption (hypochromicity) as the sample is heated. Such measurements involve chromatin isolation, shearing, and solubilization, which destroy
7.8.4 Supplement 3
Current Protocols in Cytometry
Total fluorescence
A
60
G2
40
S 20
G1 M
G1A
Total fluorescence
B
0 60
40
AP
20
0
Total fluorescence
C
60
40
AP
20
0 0
20
40
60
80
αt
Figure 7.8.3 Detection of apoptotic cells based on differences in DNA sensitivity to denaturation. Total fluorescence is on the y axis. (A) Exponentially growing HL-60 promyelocytic leukemia cells (control). (B,C) Cells treated in vitro with 0.15 µM camptothecin (CPT) for 2 (B) and 3 hr (C) to induce apoptosis. Cells were stained as described in Basic Protocol 1. Note that in the control only mitotic (M) cells have high αt. The S-phase cells undergoing apoptosis (Ap) in the presence of CPT are characterized by increased DNA denaturability. Nucleic Acid Analysis
7.8.5 Current Protocols in Cytometry
Supplement 3
Analysis of DNA Denaturation
higher orders of chromatin structure (Subirana, 1973). Furthermore, biochemical methods are applied to whole-cell populations in bulk and therefore cannot measure intercellular variability or detect cell subpopulations with different properties. Flow cytometric measurement of DNA sensitivity to denaturation in situ, which makes use of the metachromatic property of AO whereby the dye differentially stains ds versus ss nucleic acids, has no such limitations and can be applied to individual cells. In situ sensitivity of DNA to denaturation is assayed by subjecting cells that have first been permeabilized (ethanol-fixed) and treated with RNase (to remove RNA, which also stains with AO) to heat or acid, and then staining them with AO (Darzynkiewicz et al., 1977; Darzynkiewicz, 1994). Therefore, the relative proportions of red and green fluorescence of cells stained this way represent the proportions of the denatured and native, nondenatured DNA, respectively. The metachromatic property of AO results from the two different modes whereby the fluorochrome binds nucleic acids. When dsDNA is exposed to AO, the dye intercalates into the DNA and emits green fluorescence when excited by blue light; maximum emission varies between 520 and 525 nm, depending on the nucleic acid base composition. Interactions of AO with ss nucleic acids, on the other hand, result in condensation and precipitation of the dye on the chromatin, which then emits red fluorescence maximally at 640 nm. The maximum excitation of the red fluorescence of AO– nucleic acid complexes occurs at 426 to 457 nm, depending on the base composition, while maximum excitation of green fluorescence occurs at 492 to 504 nm. The red fluorescence, characterized by a large Stokes shift and a duration severalfold longer than the green fluorescence, is likely the result of intersystem crossing occurring in the condensed (solidstate) products formed by interaction of AO with nucleic acids and other biopolymers (Darzynkiewicz and Kapuscinski, 1990). In situ DNA stability varies significantly between cell types, cells in different phases of the cell cycle, differentiated versus nondifferentiated cells, and live versus apoptotic cells, and even within individual chromosomes. In the latter case the differences are reflected as chromosome banding. Except in cells undergoing spermatogenesis, where the late stages of chromatin condensation in spermatozoa are paralleled by an increase rather than a decrease in DNA stability (Evenson et al., 1980), in situ
DNA sensitivity to denaturation correlates with the degree of chromatin condensation: the more condensed the chromatin, the more unstable the DNA. Therefore, this method can be applied in diverse studies designed to provide information about changes in nuclear chromatin or to identify cells with different chromatin structure. Altered sensitivity of DNA to denaturation accompanies cell differentiation and apoptosis as well as paralleling cytotoxic effects of chemotherapeutic drugs (DNA intercalators) on target tumor cells. The most common application of the DNA denaturability assay, however, stems from the variation in DNA stability during the cell cycle. DNA in mitotic cells, and in quiescent cells characterized by condensed chromatin, is very sensitive to denaturation (Darzynkiewicz et al., 1977; Darzynkiewicz, 1994). Conversely, DNA in late G1 (G1B) and early S is the most resistant to denaturation. The method thus distinguishes cells in traditional phases of the cell cycle, including mitotic cells, and in some situations can identify quiescent cells arrested in G1 (G1Q), S (SQ), or G2 (G2Q) that are characterized by condensed chromatin. The possibility of rapid quantification of mitotic cells contributed to the application of this technique to score mitotic indices, especially in stathmokinetic experiments in which cells are arrested in mitosis (Darzynkiewicz et al., 1987).
Critical Parameters and Troubleshooting The use of AO for analyzing DNA denaturation is restricted by the fact that it is not completely specific to nucleic acids, but stains other polyanions as well. Cells containing large amounts of glycosaminoglycans or proteoglycans, such as normal fibroblasts, mast cells, chondrocytes, and differentiated keratinocytes, all have unacceptably high AO fluorescence unrelated to DNA. The present method is therefore not suitable for such cells. On the other hand, most cultured cell lines, especially of tumor origin, as well as lymphocytes, monocytes, leukemias, and lymphoma, and cells isolated from most solid tumors, exhibit good or at least adequate specificity of DNA staining with AO. The degree of nonspecific fluorescence can be estimated by incubating control cells with DNase prior to AO staining. The most critical points of this procedure relate to selecting an appropriate AO concentration and ensuring adequate separation of the green and red emission components.
7.8.6 Supplement 3
Current Protocols in Cytometry
Concentration of free dye To be at equilibrium, differential staining of dsDNA versus ssDNA requires an appropriate concentration (∼6 µg/ml) of free (unbound) dye in the final staining solution and during the actual measurement: that is, at the moment the cell in the flow intersects with the laser beam. Certain problems common to all methods employing AO, relating to variability of AO concentration due to excess of cell number or different geometry of flow channels (dye diffusion), are pertinent to this method; these are discussed in detail in UNIT 7.3. Other problems are associated with the need to maintain the proper dye concentration in the solution, and include the following. 1. When the cell number in the original suspension exceeds 5 × 106 cells/ml (or even less if the cells are hyperploid), the amount of AO bound to DNA is proportionally high, and the free dye concentration may be significantly reduced, as a straightforward consequence of the law of mass action. Denatured DNA will then stain green rather than red, and no clear distinction will be visible in the staining patterns of denatured and nondenatured DNA. This problem can be avoided by diluting the original cell suspension to have fewer cells in the sample. 2. In cell sorters, where cell measurements take place after a pressure drop outside the nozzle, significant diffusion of dye from the sample to the sheath fluid occurs prior to the stream’s intersection with the laser beam. This upsets the equilibrium and lowers the actual AO concentration in the sample at the time of cell measurement. Dye diffusion is also a problem in some instrument designs that have a narrow sample stream and long flow channel (e.g., Cytofluorograf 50 made by Ortho Diagnostics). This problem can be resolved by increasing the AO concentration in the staining solution (up to 20 µg/ml) and by increasing the sample flow rate to compensate for the diffusion. Wherever possible, channels with favorable geometry should be used (i.e., those with wider sample stream and/or shorter distance between the nozzle and intersection with the laser beam). Optimal dye concentration for a particular instrument can be established by preparing serial dilutions of AO (e.g., from 5 to 20 µg/ml), and determining the concentration where the separation of G2 and M cells is as shown in Fig. 7.8.1 (see below, How to get started with this method).
Adsorption to tubing Like rhodamine 123 and other strongly fluorescing cationic dyes, AO adsorbs to the surface of sample tubing. Release of dye from the previously used tubing may interfere with measurement of subsequent samples, especially when the cells have low fluorescence as in weak immunofluorescence. Prevent this by rinsing the sample flow line with 10% bleach, then with 50% ethanol, then with PBS, for 10 min each. Alternatively, keep a spare set of sample flow tubing for use with AO only. How to get started with this method Because mitotic cells are most readily distinguished by this method (see Anticipated Results), to test whether the methodology performs according to expectation, begin by staining a cell population enriched in mitotic cells. Obtain such a population by incubating cells during exponential growth for 3 to 4 hr with a mitotic inhibitor such as colcemid or vinblastine (Darzynkiewicz et al., 1987). Under proper cell staining and fluorescence measurement conditions the results should be similar to those shown in Figure 7.8.1. Thus, on the green versus red fluorescence bivariate display, the cluster of mitotic cells with higher red and lower green fluorescence than G2 cells is diagonal to the G2 cell cluster (Fig. 7.8.1A). On the bivariate Ltot versus αt index display, mitotic cells and G2 cells have similar Ltot values but much higher αt values (Fig. 7.8.1B). If denaturation of DNA is inadequate or AO concentration is too low (as when the dye diffuses extensively from the cells to sheath flow in some sorting channels), the separation will be less marked than shown in this figure. Increased AO concentration in the staining solution can compensate for the diffusion effects. On the other hand, excessive DNA denaturation gives mitotic cells disproportionately high red and extremely low green fluorescence. Controls and standards Standardizing the staining and measurement procedure requires that the numerical increase in red fluorescence intensity resulting from DNA denaturation should be proportional to the decrease in green fluorescence. Thus, for cells with the same DNA content but different DNA sensitivity to denaturation (e.g., G2 versus M), the sum of red and green fluorescence intensities (Ltot in Fig. 7.8.1) should be the same. Obtaining this optimal staining pattern requires proper selection of (1) emission filters,
Nucleic Acid Analysis
7.8.7 Current Protocols in Cytometry
Supplement 6
(2) AO concentration, and (3) red and green photomultiplier sensitivity (voltage) settings. Nonstimulated human peripheral blood lymphocytes are a convenient standard against which to compare the measured cells, for instance in tumor cell analysis. A batch of fixed lymphocytes can be stored at 4°C for several weeks without any change in DNA denaturability, and may be used as an external and/or internal reference standard to be compared with the αt value (the ratio of red to total fluorescence) of the measured cell population. Under proper conditions of cell staining and filter selection, the sensitivities of the red and green channel photomultipliers should then be routinely set to give equal mean numerical values of red and green fluorescence of the lymphocyte population, so that the mean αt value of lymphocytes will be 0.5.
Anticipated Results
Analysis of DNA Denaturation
Determination of cell cycle distribution of exponentially growing cells Figure 7.8.1 shows the typical distribution of a population of exponentially growing cells at different phases of the cell cycle. Their green and red fluorescence after partial denaturation of DNA and staining with AO are recorded either as bivariate distribution of green and red fluorescence values (Fig. 7.8.1A) or total cell fluorescence (green plus red; Ltot) and αt value (red fluorescence/Ltot; Fig. 7.8.1B). Ltot represents total cellular DNA, while αt is a measure of a portion of denatured DNA. Theoretically, αt can vary from 0 (all DNA stains green, no red fluorescence) to 1.0 (all DNA stains red, no green fluorescence; Darzynkiewicz et al., 1977). In practice, because of overlap of the green and red AO spectrum, αt usually varies from 0.3 to 0.8. The relative intensities of red and green fluorescence for each cell correlate with the extent of DNA denaturation, which may reflect the degree of chromatin condensation. Mitotic cells (M) exhibit maximal denaturation of DNA, and are easily distinguished from interphase cells by their high red and decreased (compared to G2) green fluorescence (Fig. 7.8.1A). Adjusting staining conditions for optimal dye concentration, optical filters, and photomultiplier sensitivity settings should ensure that the theoretical line connecting M and G2 cell clusters is diagonal (i.e., at a 45° angle to red and green fluorescence coordinates), and M cells have green fluorescence similar to that of G1 cells.
The following populations can be identified by gating analysis based on differences in total fluorescence and/or αt (Fig. 7.8.1B). All of these populations must be identified in performing detailed analysis of stathmokinetic experiments that track cell progression through different phases of the cell cycle (Darzynkiewicz et al., 1987). 1. Cells in mitosis (M) have the highest αt. 2. Cells in G2 and M form a typical G2/M peak in Ltot histograms. The fraction of cells in G2/M, S, and in G1 can be estimated by histogram deconvolution (see UNIT 7.3, Basic Protocol 6). Subtracting M cells (distinguished by αt) from the G2/M peak on Ltot histograms reveals the proportion of G2 cells. 3. G1A cells have chromatin significantly different (αt) from cells in early S. To classify them, the gating window is at first located at the lowest quartile of the S population, and the mean αt and standard deviation (SD) from the mean values established. The threshold dividing G1A from G1B is then plotted on the αt coordinate at the α value 2 SD above the mean αt of these early S cells. G1A cells are early postmitotic cells characterized by condensed chromatin; cell residence times in G1A have stochastic distribution. Cells in G1B compartment are late G1 cells with decondensed chromatin. 4. Gating windows can be located along the S-phase cluster (total fluorescence; DNA content) to identify cells in early, mid, and late S. Identification of quiescent cells Nonstimulated and mitogen-stimulated peripheral blood lymphocytes are examples of quiescent and cycling cells, respectively (Darzynkiewicz et al., 1976). Distribution of these cells with respect to their red and green fluorescence after partial DNA denaturation and staining with AO is shown in Fig. 7.8.2. Changes in AO staining are a reflection of chromatin decondensation (αt changes) and progression through the cell cycle (DNA content increase; total fluorescence changes) following mitogenic stimulation. Quiescent, noncycling lymphocytes (G0 cells) having more condensed chromatin are distinguished from their cycling counterparts (G1 cells) by higher αt values. The αt index of G0 cells, however, is lower than that in mitotic cells. Detection of apoptotic cells In apoptotic cells some sections of DNA are single-stranded (Frankfurt et al., 1993), and DNA in situ in chromatin, like that in mitotic
7.8.8 Supplement 6
Current Protocols in Cytometry
cells, is also very sensitive to denaturation (Darzynkiewicz, 1994; Darzynkiewicz et al., 1997). The method presented here can therefore be used to identify apoptotic cells (Fig. 7.8.3). Furthermore, while apoptotic cells are discriminated based on changes in αt index, the changes in distribution of cell populations with respect to Ltot reveal the cell cycle phase specificity of apoptosis, as shown in Figure 7.8.3 (Hotz et al., 1992). It should be stressed that because mitotic and apoptotic cells have similar DNA in situ sensitivity to denaturation, they cannot be distinguished from each other based on differences in αt index.
Time Considerations Preparation and fixation of cells in suspension takes 15 to 20 min. Cells must remain in fixative at least 2 hr at 4°C. Preparation for staining requires ∼30 min, and the staining itself only 30 sec.
Literature Cited Darzynkiewicz, Z. 1994. Acid-induced denaturation of DNA in situ as a probe of chromatin structure. Methods Cell Biol. 41:527-541. Darzynkiewicz, Z. and Kapuscinski, J. 1990. Acridine orange: A versatile probe of nucleic acids and other cell constituents. In Flow Cytometry and Sorting (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 291-314. Wiley-Liss, New York. Darzynkiewicz, Z., Traganos, F., Sharpless, T., and Melamed, M.R. 1977. Different sensitivity of DNA in situ in interphase and metaphase chromatin to heat denaturation. J. Cell Biol. 73:128138. Darzynkiewicz, Z., Traganos, F., Sharpless, T., and Melamed, M.R. 1976. Lymphocyte stimulation: A rapid multiparameter analysis. Proc. Natl. Acad. Sci. U.S.A. 73:2881-2884.
Darzynkiewicz, Z., Traganos, F., and Kimmel, M. 1987. Assay of cell cycle kinetics by multivariate flow cytometry using the principle of stathmokinesis. In Techniques in Cell Cycle Analysis (J.W. Gray and Z. Darzynkiewicz, eds.) pp. 291336. Humana Press, Clifton, N.J. Darzynkiewicz, Z., Juan, G., Li, X., Gorczyca, W., Murakami, T., and Traganos, F. 1997. Cytometry in cell necrobiology: Analysis of apoptosis and accidental cell death (necrosis). Cytometry 27:120. Evenson, D.P., Darzynkiewicz, Z., and Melamed, M.R. 1980. Relation of mammalian sperm chromatin heterogeneity to fertility. Science 210:1131-1133. Frankfurt, O.S., Byrnes, J.J., Seckinger, D., and Sugarbaker, E.V. 1993. Apoptosis (programmed cell death) and the evaluation of chemosensitivity in chronic lymphocytic leukemia and lymphoma. Onc. Res. 5:37-42. Hotz, M.A., Traganos, F., and Darzynkiewicz, Z. 1992. Changes in nuclear chromatin related to apoptosis or necrosis induced by the DNA topoisomerase II inhibitor fostriecin in MOLT-4 and HL-60 cells are revealed by altered DNA sensitivity to denaturation. Exp. Cell Res. 201:184192. Rigler, R., Jr. 1966. Microfluorometric characterization of intranuclear nucleic acids and nucleoproteins by acridine orange. Acta Physiol. Scand. 67(Suppl. 267):1-122. Subirana, J.A. 1973. Studies on the thermal denaturation of nucleohistone. J. Mol. Biol. 74:363385. van Holde, K.E. 1989. Chromatin. Springer-Verlag, New York.
Contributed by Zbigniew Darzynkiewicz and Gloria Juan New York Medical College Elmsford, New York
Nucleic Acid Analysis
7.8.9 Current Protocols in Cytometry
Supplement 3
Bivariate Analysis of DNA Content and Expression of Cyclin Proteins
UNIT 7.9
Several units in this chapter describe different methods for cell cycle analysis. Results obtained with each of these methods yield different and often complementary information regarding the distribution of cells in particular phases of the cycle, kinetics of the cell cycle progression, and proliferative potential of the cell populations under study. Nucleic acid analysis—whether it is univariate, limited to DNA content (UNITS 7.5 & 7.6), or bivariate, revealing DNA sensitivity to denaturation (UNIT 7.8), combining measurement of DNA with RNA (UNIT 7.3), or including DNA replication (UNIT 7.7)—is an essential aspect of all studies of the cell cycle. A special category of methods for cell cycle analysis combines measurement of DNA content and expression of proliferation-associated proteins. The latter are detected immunocytochemically, using specific antibodies that are fluorochrome labeled either directly or indirectly (via secondary antibody). This unit is devoted to cyclins, whose analysis provides an insight into the actual components of the cell cycle progression machinery (see Background Information). Because some cyclins are expressed transiently at very specific time intervals during the cell cycle, their presence in a cell can be considered to be a marker of this particular portion of the cycle. BIVARIATE ANALYSIS OF DNA CONTENT AND CYCLINS This protocol combines measurement of DNA content with expression of either one of the D-type cyclins or cyclins E, A, or B1. Cells are fixed and labeled with cyclin antibody followed by fluorescein isothiocyanate (FITC)-conjugated secondary antibody, and finally stained with propidium iodide (PI) for DNA measurement. If directly conjugated FITC-anti-cyclin antibody is available, secondary antibody labeling (steps 6 and 7) can be omitted. This analysis is adapted to the most commonly used instruments, such as those equipped with a single-laser (488 nm) illumination source.
BASIC PROTOCOL
Materials Cells to be analyzed Phosphate-buffered saline (PBS; APPENDIX 2A), pH 7.4 Fixative: 80% ethanol or 100% methanol, −20°C Permeabilization solution: 0.25% (v/v) Triton X-100 in PBS, pH 7.4 (store at 4°C) Rinsing buffer: 1% bovine serum albumin (BSA) in PBS, pH 7.4 (store at 4°C) Cyclin antibodies: e.g., mouse monoclonal antibodies to cyclin B1 (clone GNS-1), cyclin A (clone BF-683), cyclin D1 (clone G124-326), cyclin D3 (clone G107-565), and cyclin E (clone HE12; all provided by PharMingen; cyclin D1 may also be obtained from Immunotech) Isotypic control: mouse IgG1 1 g/liter FITC-conjugated goat anti–mouse IgG antibody PI staining buffer: 5 µg/ml PI and 200 µg/ml DNase-free RNase A (APPENDIX 2A) in PBS, pH 7.4, made fresh Silanized or polypropylene 15-ml conical tube Flow cytometer equipped with 488-nm argon laser or mercury arc lamp with blue (BG12) excitation filter (∼50% cutoff at 470 nm) Additional reagents and equipment for trypsinizing cells (APPENDIX 3B) or dissociating cells from tissues (UNIT 5.2) Nucleic Acid Analysis Contributed by Gloria Juan and Zbigniew Darzynkiewicz Current Protocols in Cytometry (1998) 7.9.1-7.9.6 Copyright © 1998 by John Wiley & Sons, Inc.
7.9.1 Supplement 4
Prepare cell suspension for fixation 1a. For cells growing in suspension or hematologic samples: Rinse once with PBS and suspend in PBS at ∼106 cells/ml. 1b. For cells growing attached to tissue culture dishes: Collect cells from flasks or petri dishes by trypsinization and pool the trypsinized cells with the cells floating in the medium (mostly detached mitotic, apoptotic, and dead cells) by centrifuging 5 min at 300 × g, room temperature. Rinse once with medium containing serum (to inactivate the trypsin) and centrifuge again as before. Suspend cells in PBS at ∼106 cells/ml. Other means of trypsin inactivation, such as addition of protease inhibitors, may also be used.
1c. For cells isolated from solid tumors: Rinse free of any enzyme used for cell dissociation and resuspend in PBS as above. The final cell suspension should be well dispersed (not in aggregates) with a density no higher than 5 × 106 cells/ml.
Fix cells 2. With a Pasteur pipet transfer 1 ml of the cell suspension to a 15-ml tube containing 10 ml of 80% ethanol or 100% methanol at 0° to 4°C. For best results, cells should be rapidly added to the fixative, rather than the reverse. Time of fixation (storage) at 4°C may vary from 4 hr to several weeks. IMPORTANT NOTE: To minimize cell loss, all the subsequent steps should be performed in the same tube.
Label cells with anticyclin antibody 3. Centrifuge 5 min at 300 × g, room temperature, remove supernatant, suspend cells in 5 ml PBS, and centrifuge as before. 4. Remove supernatant and resuspend the cell pellet (≤106 cells) in 1 ml permeabilization solution. Keep on ice for 5 min, add 5 ml PBS, and centrifuge 5 min at 300 × g, room temperature. 5. Suspend the cell pellet in 100 µl rinsing buffer containing the cyclin antibody at the appropriate dilution to obtain 0.5 µg of antibody per sample. Incubate 60 min at room temperature with gentle agitation or at 4°C overnight. Treat control cells identically, but incubate with the isotypic antibody at the same titer, instead of cyclin antibody. Antibodies to cyclin proteins are offered by a variety of sources, but only a few are satisfactory for flow cytometric detection of cyclin proteins (e.g., PharMingen, Immunotech).
Label cells with FITC-conjugated antibody 6. Add 5 ml rinsing buffer and centrifuge 5 min at 300 × g, room temperature. 7. Suspend the cell pellet in 100 µl of 1 g/liter FITC-conjugated goat anti–mouse IgG antibody diluted 1:30 in rinsing buffer. Incubate 30 min in the dark at room temperature with gentle agitation.
Bivariate Analysis of DNA Content and Expression of Cyclin Proteins
Stain cells with PI 8. Add 5 ml rinsing buffer and centrifuge 5 min at 300 × g, room temperature. 9. Suspend the cell pellet in PI staining buffer. Incubate 20 min at room temperature in the dark before measurement.
7.9.2 Supplement 4
Current Protocols in Cytometry
Perform flow cytometery 10. Set up and adjust the flow cytometer for excitation with blue light (488-nm laser line or BG12 blue filter). Use a 530 ± 20-nm bandpass filter for detection of FITC emission and a 620-nm long-pass filter for PI emission. 11. Measure the cyclin-associated green fluorescence of FITC and DNA-associated red fluorescence of PI. COMMENTARY Background Information Cyclins are the key elements of the cell cycle progression machinery. They combine with particular cyclin-dependent protein kinases (CDKs) to form the holoenzymes that phosphorylate different sets of proteins at consecutive stages of the cell cycle, thereby driving the cell through the cycle (Pines and Hunter, 1991; Norbury and Nurse, 1992; Draetta, 1994; Hartwell and Kastan, 1994; Sherr, 1994; CardonCardo, 1995; Morgan, 1995). The function of cyclins in these holoenzymes is to activate their partner CDKs and to target them to specific protein substrates whose phosphorylation is essential for the traverse of a particular section of the cell cycle. Several cyclins, notably D-type cyclins and cyclins E, A, and B, are expressed transiently during the cell cycle. Cyclins D and E belong to the family of G1 cyclins, whereas cyclins A and B are G2 cyclins. During unperturbed growth of normal, nontumor cells, the synthesis and degradation of cyclins take place at very specific points of the cycle. The periods of expression of these cyclins by the cell are therefore landmarks of the cell cycle. These landmarks complement the traditional mileposts of the cycle detected by DNA content measurement, namely mitosis and DNA replication. The development of antibodies to cyclins made it possible to detect them immunocytochemically and investigate their expression in individual cells by cytometry (Gong et al., 1994a,b, 1995a,b; Sherwood et al., 1994; Lukas et al., 1995; Urbani et al., 1995; Juan et al., 1996, 1997). Bivariate analysis of DNA content and cyclin expression provides a framework for subdividing the cell cycle into several subcompartments and defining the cell cycle point of arrest of some antitumor drugs with greater precision than before (Darzynkiewicz et al., 1996). Furthermore, the differences in cyclin expression make it possible to discriminate between cells having the same DNA content but residing at different phases, such as G2 versus M cells, or G2/M cells of a lower DNA
ploidy versus G1 cells of a higher DNA ploidy (Gong et al., 1993). Unlike any other approach, the bivariate analysis of cyclin expression versus DNA content detects the unscheduled expression of cyclins—namely, the presentation of G1 cyclins by cells in G2/M and of G2/M cyclins by G1 cells—without the need for cell synchronization. Such unscheduled (“ectopic”) expression of cyclins B1 and A was seen when cell cycle progression was halted, such as after synchronization at the G1/S boundary by inhibitors of DNA replication (Gong et al., 1995b; Urbani et al., 1995). Unscheduled expression of cyclins B1 or E, representing a characteristic feature of a particular tumor phenotype, was also observed in some tumor cell lines when their growth was unperturbed (Gong et al., 1994a). Likewise, while the expression of cyclins D1 or D3 in nontumor cells was restricted to an early section of G1, the presentation of these proteins in many tumor cell lines was also seen during S and G2/M (Juan et al., 1996). As specific markers of cell proliferation, cyclins are expected to reflect the proliferative potential of tumors and therefore be the key prognostic markers in neoplasia.
Critical Parameters and Troubleshooting Cell fixation and permeabilization are critical steps for immunocytochemical detection of intracellular proteins and often must be customized for particular antigens. The fixative is expected to stabilize the antigen in situ and preserve its epitope in a state where it continues to remain reactive with the available antibody. The cell must be permeable to allow access of the antibody to the epitope. General strategies of cell fixation and permeabilization have been described by Bauer and Jacobberger (1994). Most studies on cyclins have employed precipitating fixatives such as 70% to 80% ethanol, 100% methanol, or a 1:1 mixture of methanol and acetone cooled to −20° to −40°C. Brief (15 to 30 min) treatment with 1% paraformaldehyde followed by 70% cold ethanol has been
Nucleic Acid Analysis
7.9.3 Current Protocols in Cytometry
Supplement 4
The relative cellular content of a particular cyclin plays a role in its detection. The signalto-noise ratio (ratio of fluorescence intensity of the cyclin-positive cells to the control cells, stained with the isotype immunoglobin), for example, is higher in the case of cyclin B1 than in the case of cyclins E or A, most likely because the absolute level of cyclin B1 is higher than the absolute level of cyclins E or A. The level of expression of D-type cyclins varies markedly depending on the cell type and the phase of cell growth. High instrument sensitivity and a low level of cell autofluorescence are therefore of greater importance for the detection of cyclins E or A than of cyclin B1 or D-type cyclins. While isotypic immunoglobin or irrelevant isotypic antibody is generally accepted as an appropriate control, neither is perfect, as their fluorescence may also vary depending on the source (vendor) and may not be representative of the actual background. An ideal control consists of cells of the same type and species but with a deletion of the gene that codes for the detected protein. Such control cells should be subjected to the same immunocytochemical procedure as the studied cells. Unfortunately, few cell lines with deleted cyclin genes are currently available.
used for fixation of D-type cyclins, although this cyclin can also be detected following fixation with cold methanol. The choice of fixative, then, appears not to be a critical factor for cyclin detection and, although the absolute level of the immunofluorescence may vary, various fixation protocols yield essentially similar cyclin distributions with respect to the cell cycle position. Each fixative has some undesirable effects (e.g., increased cell clumping in the case of ethanol/acetone mixture, or cell autofluorescence and poor DNA stainability when formaldehyde is used), and one often must compromise between these effects and the optimal detection of a particular cyclin. Fixation in cold 80% ethanol, as presented in this protocol, offers such a compromise. Much more critical for the detection of cyclins is the proper choice of antibody. Very often an antibody applicable to immunoblotting fails in immunocytochemical applications, and vice versa, possibly because of differences in the accessibility of the epitope or in the degree of denaturation of the antigen on the immunoblots compared with its in situ location. Some epitopes may not be accessible in situ at all. Because there is strong homology between different cyclin types, cross-reactivity may also be a problem. Commercially available monoclonal antibodies may differ in specificity, degree of cross-reactivity, and so on, so it is important to use reagents that have already been tested and referenced in published papers. Authors should also provide information (the vendor and hybridoma clone number) for the reagents used in their studies.
Cyclin expression
1000
IgG 400
Anticipated Results As mentioned above, the scheduled timing of expression of cyclins B1, A, E, and D1 in relation to the major phases of the cell cycle is reflected by a very characteristic pattern of the bivariate cyclin versus cellular DNA content distributions. These distributions are as shown
D1
E
A
B1
800
600 400 G2M 0 G1 0 S
200 0 0
400
400
M 800
0
400
0
400
0
400
0
400
DNA content
Bivariate Analysis of DNA Content and Expression of Cyclin Proteins
Figure 7.9.1 Typical bivariate cyclin versus DNA content distributions (scatter plots) showing expression of cyclin D1 in human normal fibroblasts and cyclins E, A, and B1 in PHA-stimulated human lymphocytes. The IgG control (representing the level of fluorescence of control cells stained with the isotype IgG rather than the respective cyclin monoclonal antibody, prior to fluoresceinated secondary antibody) is included in the left panel. Abbreviations: PHA, phytohemagglutinin.
7.9.4 Supplement 4
Current Protocols in Cytometry
in Figure 7.9.1 for normal human proliferating lymphocytes (cyclins B1, A, and E) and fibroblasts (cyclin D1). As is evident from the cytograms, the expression of cyclin B1 is limited to late-S-phase cells and cells with a G2/M content of DNA, although early- and mid-Sphase cells show a very low level of this protein. Cells in G1 phase are essentially cyclin B1 negative. Like cyclin B1, cyclin A is expressed minimally, if at all, in G1 cells. Expression of cyclin A becomes pronounced during S phase, however, and its level progressively increases as the cells advance towards G2. Maximal expression of cyclin A is seen in cells with a G2/M DNA content (Fig. 7.9.1). It should be mentioned, however, that because cyclin A is abruptly degraded during prometaphase (Pines and Hunter, 1991), mitotic cells that have advanced past prometaphase are essentially cyclin A negative (not shown). Expression of cyclin E can be summarized as follows: (1) the maximal level of this protein is detected in cells undergoing the transition from G1 to S; (2) its level continuously decreases during cell progression through S, so that most G2/M cells are cyclin E negative; and (3) a distinct threshold in its expression is apparent at the G1/S transition. As is evident from the continuity of the cell clusters on scatter plots (Fig. 7.9.1), cells must accumulate cyclin E above the threshold level to enter S phase. The presence of cyclin D1 in exponentially growing normal fibroblasts is limited to cells in G0/G1 (Fig. 7.9.1). Most cells in S and G2/M are cyclin D1 negative, with the exception of a very few cells with a G2/M DNA content. The latter may be G1 cell doublets, since not all doublets can be identified by analysis of the shape (pulse width) of the electronic signal. It should be stressed that, as mentioned above (see Background Information), the cyclin distributions as shown in Figure 7.9.1 characterize only normal, exponentially, and asynchronously growing cells. The distributions are very much different when the cell cycle progression is perturbed, or in the case of tumor cell lines, which display unscheduled expression of these cyclins.
Time Considerations Cell fixation can range from 4 hr to several weeks. The remaining labeling, staining, and flow cytometry requires 4 to 5 hr.
Literature Cited Bauer, K.D. and Jacobberger, J.W. 1994. Analysis of intracellular proteins. Methods Cell Biol. 41:352-373. Cardon-Cardo, C. 1995. Mutations of cell cycle regulators. Biological and clinical implications for human neoplasis. Am. J. Pathol. 147:545560. Darzynkiewicz, Z., Gong, J., Juan, G., Ardelt, B., and Traganos, F. 1996. Cytometry of cyclin proteins. Cytometry 25:1-13. Draetta, F.-G. 1994. Mammalian G1 cyclins. Curr. Opin. Cell Biol. 6:842-846. Gong, J., Traganos, F., and Darzynkiewicz, Z. 1993. Simultaneous analysis of cell cycle kinetics at two different DNA ploidy levels based on DNA content and cyclin B measurements. Cancer Res. 53:5096-5099. Gong, J., Ardelt, B., Traganos, F., and Darzynkiewicz, Z. 1994a. Unscheduled expression of cyclin B1 and cyclin E in several leukemic and solid tumor cell lines. Cancer Res. 54:42854288. Gong, J., Li, X., Traganos, F., and Darzynkiewicz, Z. 1994b. Expression of G1 and G2 cyclins measured in individual cells by multiparameter flow cytometry: A new tool in the analysis of the cell cycle. Cell Prolif. 27:357-371. Gong, J., Bhatia, U., Traganos, F., and Darzynkiewicz, Z. 1995a. Expression of cyclins A, D2, and D3 in individual normal mitogen stimulated lymphocytes and in MOLT-4 leukemic cells analyzed by multiparameter flow cytometry. Leukemia 9:983-899. Gong, J., Traganos, F., and Darzynkiewicz, Z. 1995b. Growth imbalance and altered expression of cyclins B1, A, E, and D3 in MOLT-4 cells synchronized in the cell cycle by inhibitors of DNA replication. Cell Growth Differ. 6:14851493. Hartwell, L.H. and Kastan, M.B. 1994. Cell cycle control and cancer. Science 266:1821-1823. Juan, G., Gong, J., Traganos, F., and Darzynkiewicz, Z. 1996. Unscheduled expression of cyclins D1 and D3 in human tumor cell lines. Cell Prolif. 29:259-266 Juan, G., Li, X., and Darzynkiewicz, Z. 1997. Correlation between DNA replication and expression of cyclins A and B1 in individual MOLT-4 cells. Cancer Res. 57:803-807. Lukas, J., Bartkova, J., Welcker, M., Petersen, O.W., Peters, G., Strauss, M., and Bartek, J. 1995. Cyclin D2 is a moderately oscillating nucleoprotein required for G1 phase progression in specific cell types. Oncogene 10:2125-2134. Morgan, D.O. 1995. Principles of CDK regulation. Nature 374:131-134. Norbury, C. and Nurse, P. 1992. Animal cell cycles and their control. Annu. Rev. Biochem. 61:441470. Nucleic Acid Analysis
7.9.5 Current Protocols in Cytometry
Supplement 4
Pines, J. and Hunter, T. 1991. Human cyclin A and cyclin B are differentially located in the cell and undergo cell cycle–dependent nuclear transport. J. Cell Biol. 115:1-17.
Urbani, L., Sherwood, S.W., and Schimke, R.T. 1995. Dissociation of nuclear and cytoplasmic cell cycle progression by drugs employed in cell synchronization. Exp.Cell Res. 219:159-168.
Sherr, C.J. 1994. G1 phase progression. Cycling on cue. Cell 79:551-555. Sherwood, S.W., Rush, D.P., Kung, A.L., and Schimke, R.T. 1994. Cyclin B1 expression in HeLa cells studied by flow cytometry. Exp. Cell Res. 211:275-281
Contributed by Gloria Juan and Zbigniew Darzynkiewicz New York Medical College Valhalla, New York
Bivariate Analysis of DNA Content and Expression of Cyclin Proteins
7.9.6 Supplement 4
Current Protocols in Cytometry
Flow Cytometric Analysis of Reticulated Platelets Reticulated platelets, by analogy with reticulocytes as their erythrocyte counterparts, comprise the youngest platelet population. These cells can be distinguished from mature platelets by flow cytometry following fluorescent derivatization of their RNA content. In this context, the extensively investigated dye thiazole orange (TO) and the less frequently used dye auramine O express high quantum yields, and in the case of TO, a 3000-fold fluorescence enhancement, upon binding to RNA. Excitation for both dyes can be achieved at the argon laser wavelength of 488 nm, available in most benchtop flow cytometers. The dyes are lipophilic and readily permeate cell membranes; consequently, live platelets rapidly accumulate TO fluorescence. Nonsaturable staining kinetics in viable cells, however, are a major problem. Formaldehyde-fixed cells, in contrast, are characterized by a saturable number of binding sites. Intravital staining followed by formaldehyde fixation, therefore, is a method that combines rapid staining kinetics and stability of cellular fluorescence.
UNIT 7.10
BASIC PROTOCOL
The precise identification of platelets in unseparated whole blood is an another critical aspect of the procedure. This can be achieved by the counterstaining of abundantly expressed platelet-specific glycoproteins such as GPIIb/IIIa (CD41/CD61) or GPIb (CD42b) with phycoerythrin-labeled monoclonal antibodies. Such dual-color staining of unseparated whole blood samples avoids selective cell losses that may occur during the enrichment of platelets by centrifugation, and also serves as a basis for reliable platelet identification both in thrombocytopenia and in samples with abnormal platelet light-scatter characteristics. Materials Normal and patient whole blood samples anticoagulated with di- or tripotassium EDTA (1.5 mg/ml final EDTA concentration) Phycoerythrin-conjugated platelet-specific monoclonal antibody—e.g., anti-GPIb (CD42b) or anti-GPIIb/GPIIIa (CD41/CD61)—at concentration such that required quantity (determined by titration) can be delivered in 5 µl 1 µg/ml thiazole orange (TO) working solution (see recipe) Formaldehyde solution (see recipe) 12 × 75–mm polypropylene or polystyrene tubes as required by flow cytometer Flow cytometer with 488-nm argon laser and band-pass filters centered around 520 nm and 585 nm Stain whole blood 1. For both normal and patient blood samples, gently mix 5 ml of undiluted EDTA-anticoagulated blood with 5 µl of phycoerythrin-conjugated CD42b monoclonal antibody and 50 µl of 1 µg/ml TO working solution in a 12 × 75–mm tube appropriate for the flow cytometer. The exact amount of the monoclonal antibody to be used depends on the particular antibody and must be determined by titration. Typically, low amounts of antibodies (below the saturation level) are sufficient for a reliable discrimination of platelets from other cells in this assay. TO levels for optimal resolution of immature platelets generally range from 100 ng/ml to 10 ìg/ml.
2. Incubate 15 min at room temperature. 3. Add 1 ml formaldehyde solution to fix sample. Acquire data within 45 min. TO fluorescence will drop slightly upon prolonged storage. Contributed by Goran B. Matic, Gregor Rothe, and Gerd Schmitz Current Protocols in Cytometry (1998) 7.10.1-7.10.6 Copyright © 1998 by John Wiley & Sons, Inc.
Nucleic Acid Analysis
7.10.1 Supplement 17
104
R1 103
platelets
Anti-GPlb-phycoerythrin
102
leukocytes 101
erythrocytes 100 100 101 Thiazole orange
102
103
104
Figure 7.10.1 Dot plot showing TO green fluorescence (520 nm) versus phycoerythrin orange fluorescence (585 nm). Platelets are discriminated from erythrocytes and leukocytes based on the phycoerythrin fluorescence following labeling of surface GPIb with an anti-CD42b monoclonal antibody. Leukocytes are characterized by their bright nuclear TO fluorescence. An analysis region around the platelet population (R1) is drawn for further analysis.
Acquire data on flow cytometer 4. Set up flow cytometer with logarithmic amplification for scatter and fluorescence signals. Collect emission fluorescence at 520 nm for TO and 585 nm for phycoerythrin, with no compensation. Using the normal blood sample, set PMT voltages to achieve a platelet distribution similar to Figure 7.10.1 and Figure 7.10.2, with most platelets in the first or second decade for TO, in the second and third decade for phycoerythrin, and in the second decade for side scatter. Finally, define a threshold that discriminates platelets from erythrocytes on the basis of phycoerythrin intensity for data acquisition. 5. Acquire 5000 phycoerythrin-positive events.
Flow Cytometric Analysis of Reticulated Platelets
Analyze listmode files 6. Set an analysis region (R1) that excludes TO-bright leukocytes as shown in Figure 7.10.1. Following gating on platelets in R1, define an analysis region in a side scatter versus TO fluorescence dot plot along a virtual line formed by the upper edge of the platelet population, which contains 1% of platelets in a normal control sample. A diagonal gate on TO-positive platelets reduces interferences based on platelet size or aggregate formation.
7.10.2 Supplement 17
Current Protocols in Cytometry
104
A
B
C
D
103
102
101
100 104
103
TO fluorescence
102
101
100 100
101
102
103
104
100
101
102
103
104
Side scatter
Figure 7.10.2 Enumeration of reticulated platelets. An analysis gate is defined on the upper edge of the normal platelet population in a TO fluorescence versus side scatter dot plot. (A) Healthy volunteer, normal platelet count. Gate is adjusted to include 1% of platelets. (B) Platelet aggregates as defined by typical appearance along the lower border of the gate do not lead to false-positive results. (C) Patient after chemotherapy, platelet count 11 × 109/liter. Beginning thrombopoietic response. (D) Same patient, 1 day later, platelet count 10 × 109/liter. Enhanced thrombopoietic response preceding rise of platelet count.
7. Evaluate patient samples using the same analysis regions as defined for the normal sample. Calculate the percentage of TO-positive platelets for each sample. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Formaldehyde solution Dilute 1 part ultrapure, methanol-free 10% (w/v) formaldehyde (Polysciences) in 9 parts calcium- and magnesium-free PBS (APPENDIX 2A) to make a 1% (w/v) working solution. Thiazole orange (TO) stock and working solutions Stock solution: Dissolve thiazole orange powder (Molecular Probes) in methanol at 1 mg/ml. Store up to several months protected from light at −20°C. Working solution: Just before use, dilute stock solution in calcium- and magnesiumfree PBS (APPENDIX 2A) to a final concentration of 1 µg/ml. Discard unused working solution.
Nucleic Acid Analysis
7.10.3 Current Protocols in Cytometry
Supplement 6
COMMENTARY Background Information General considerations Reticulated platelets, first detected in 1969 by light microscopy (Ingram and Coopersmith, 1969), are characterized through their nucleic acid content (predominantly RNA) and were suspected in that study to be the youngest platelets in circulation, analogous to the erythroid reticulocyte. The fluorescent dye thiazole orange (TO; Lee et al., 1986), which showed a high fluorescence enhancement upon binding to RNA, significantly improved the flow cytometric analysis of RNA content in anucleate cells. Kienast and Schmitz (1990) published the first report on the diagnostic value of reticulated platelets in thrombocytopenia, with and without enhanced megakaryocytic activity, analyzed by flow cytometry based on the use of TO. Their studies were followed by others, published by a number of authors using modified test protocols. Reticulated platelets were shown to increase in conditions characterized by high platelet turnover—e.g., in idiopathic thrombocytopenic purpura (ITP)—and increased reticulated platelet counts preceded by several days the rise in total platelet counts in cancer patients after chemotherapy (Ault et al., 1992; Dale et al., 1995; Richards and Baglin, 1995; Rinder et al., 1993; and Romp et al., 1994). Experiments by Ault and Knowles (1995) and Dale et al. (1995) finally demonstrated that platelets staining positive for RNA constitute the youngest platelet population. Published test protocols differ in incubation times, concentration of dyes, and the additional use of a secondary antibody to distinguish platelets from other cells and debris (Matic et al., 1998; Schmitz et al., 1998). Many of these methods for enumeration of reticulated platelets have been shown to be valid in various clinical settings. The whole-blood protocol described in this unit was optimized for stability of staining characteristics, ease of use, and good resolution of reticulated platelets in patients with low platelet counts or platelet subpopulations with unusual scatter properties (Matic et al., 1998).
Flow Cytometric Analysis of Reticulated Platelets
Choice of method for the definition of TO-positive platelets Thresholds to define the reticulated platelet population that is not clearly resolved from normal platelets based on TO staining are prob-
lematic. Essentially two strategies have been proposed. The first is to define a gate individually for each test sample by plotting a TO fluorescence histogram from an unstained control, processed in parallel, with the gate beginning towards the end of the right histogram slope including 1% of unstained platelets (Kienast and Schmitz, 1990). Another method, with a similar strategy regarding individual gate settings, was suggested by Bonan et al. (1993) and uses autologous erythrocyte fluorescence as an internal standard. Published normal values, depending on which of these methods was used, vary between 3% and 20% reticulated platelets. The second possibility is to draw a gate along a virtual line formed by the upper edge of the platelet population in a TO-fluorescence versus side scatter dot plot, so as to contain 1% of platelets in TO-stained blood samples from healthy subjects (Fig. 7.10.2; Ault et al., 1992; Watanabe et al., 1995). This approach reduces interferences that are due to platelet size and platelet aggregation, and therefore is recommended in this unit (Matic et al., 1998). Furthermore, these predefined gate settings are easier to automate using commercial software than the sample-dependent individual gate settings described above. Platelet analysis in unseparated whole blood Analysis of platelets in unseparated whole blood involves less work than analysis in isolated platelet suspensions, which require multiple centrifugation steps to prepare. Furthermore, whole-blood methods are more rapid, especially the automated auramine O method presented by Watanabe et al. (1995), and an uncontrolled loss of platelet subpopulations is avoided. In thrombocytopenia or in abnormal samples, however, identification of platelets in whole blood by scatter characteristics alone is often unreliable (Matic et al., 1998). Therefore, whole-blood analysis is typically based on counterstaining of abundantly expressed platelet glycoproteins, allowing a precise identification of these cells (Schmitz et al., 1998). Owing to the high and constitutive expression of GPIIb/IIIa (CD41/CD61) and GPIb (CD42b) on platelets, antibodies against these glycoproteins are usually preferred for counterstaining. Phycoerythrin conjugates in combination with TO have the advantages of high fluorescence and lack of significant overlap with the TO channel independent of compensation.
7.10.4 Supplement 6
Current Protocols in Cytometry
Nonspecific staining by TO Nonspecific staining of platelets that is not dependent on RNA content has often been discussed as a problem in TO staining. Furthermore, this may be associated with nonsaturable staining characteristics of cells during intravital staining. Therefore, Robinson et al. (1998) have recently suggested that there is potential benefit in inducing platelet degranulation to abolish nonspecific TO fluorescence. In normal samples, which contain only a low number of RNAexpressing platelets, these authors demonstrated a decrease of TO-positive events upon platelet activation with the thrombin receptor ligand TRAP. The protocol described in this unit uses a high TO concentration of 1 µg/ml, whereas Robinson et al. (1998) applied only 25 ng/ml. This may explain the high specificity for RNA and lack of demonstrated effect produced by thrombin degranulation with this method in samples with a high number of reticulated platelets (Matic et al., 1998).
Critical Parameters and Troubleshooting Incubation time, concentration of thiazole orange, and the dilution that accompanies formaldehyde fixation are critical. Regarding the incubation time, after fixation TO fluorescence slowly decreases over time. A plateau, however, allows reproducible analysis within at least the initial 45 min. Therefore, data acquisition should be performed shortly after fixation. Regarding TO concentration, at low dye concentrations the effect of counterstaining with the relatively bright phycoerythrin on the dim TO fluorescence shows spectral interference, which may or may not be correctly controlled by color compensation. Furthermore, a relatively high amount of nonspecific staining with TO has been reported under such staining conditions (Robinson et al., 1998). In summary, every single TO concentration appears to have unique characteristics considering kinetics of staining and staining specificity that need to be evaluated. As a problem in this respect, the ReticCount solution, which is a commercially distributed TO solution from Becton Dickinson, seems to have only a low TO concentration, ∼25 ng/ml, making it difficult to achieve high levels of TO staining. Based on the short lifetime of reticulated platelets (Ault and Knowles, 1995; Dale et al., 1995) a rapid kinetic degradation of RNA is to be expected for these cells, similar to that ob-
served in reticulocytes. Therefore, the time elapsed after the blood sample is drawn may be critical.
Anticipated Results Depending on the extent of detection of the most immature fraction of platelets, the fraction of reticulated platelets is expected to correspond to the degree of bone marrow thrombopoietic activity in comparison to peripheral platelet counts. Thus, at a normal platelet count, an increased fraction of reticulated platelets may indicate increased platelet turnover compensated by increased production—e.g., due to platelet activation and sequestration in preeclampsia. At decreased total platelet counts, a low reticulated platelet fraction will help to distinguish thrombocytopenia due to bone marrow failure from uncompensated platelet loss in immune thrombocytopenia. The reticulated platelet fraction, however, is of special interest during therapy. An increase in this fraction should be an early indicator of the effectiveness of thrombopoietin therapy, just as an increase in the reticulated platelet fraction is an early indicator of bone marrow regeneration following chemotherapy.
Time Considerations Staining with the Basic Protocol requires 15 min. Depending on platelet counts, which will affect analysis time, up to 25 samples can be processed in parallel.
Literature Cited Ault, K.A. and Knowles, C. 1995. In vivo biotinylation demonstrates that reticulated platelets are the youngest platelets in circulation. Exp. Hematol. 23:996-1001. Ault, K.A., Rinder, H.M., Mitchell, J., Carmody, M.B., Vary, C.P., and Hillman, R.S. 1992. The significance of platelets with increased RNA content (reticulated platelets): A measure of the rate of thrombopoiesis. Am. J. Clin. Pathol. 98:637-646. Bonan, J.L., Rinder, H.M., and Smith, B.R. 1993. Determination of the percentage of thiazole orange (TO)-positive, “reticulated” platelets using autologous erythrocyte TO fluorescence as an internal standard. Cytometry 14:690-694. Dale, G.L., Friese, P., Hynes, L.A., and Burstein, S.A. 1995. Demonstration that thiazole-orangepositive platelets in the dog are less than 24 hours old. Blood 85:1822-1825. Ingram, M. and Coopersmith, S. 1969. Reticulated platelets following acute blood loss. Br. J. Haematol. 17:225-228. Nucleic Acid Analysis
7.10.5 Current Protocols in Cytometry
Supplement 6
Kienast, J. and Schmitz, G. 1990. Flow cytometric analysis of thiazole orange uptake by platelets: A diagnostic aid in the evaluation of thrombocytopenic disorders. Blood 75:116-121. Lee, L.G., Chen, C.H., and Chiu, L.A. 1986. Thiazole orange: A new dye for reticulocyte analysis. Cytometry 7:508-517. Matic, G.B., Chapman, E.S., Zaiss, M., Rothe, G., and Schmitz, G. 1998. Whole blood analysis of reticulated platelets: Improvements of detection and assay stability. Cytometry In press. Richards, E.M. and Baglin, T.P. 1995. Quantitation of reticulated platelets: Methodology and clinical application. Br. J. Haematol. 91:445-451. Rinder, H.M., Munz, U.J., Ault, K.A., Bonan, J.L., and Smith, B.R. 1993. Reticulated platelets in the evaluation of thrombopoietic disorders. Arch. Pathol. Lab. Med. 117:606-610. Robinson, M.C., Mackie, I.J., Khair, K., Liesner, R., Goodall, A.H., Savidge, G.F., Machin, S.J., and Harrisson, P. 1998. Flow cytometric analysis of reticulated platelets: Evidence for a large proportion of non-specific labelling of dense granules by fluorescent dyes. Br. J. Haematol. 100:351357.
Romp, K.G., Peters, W.P., and Hoffman, M. 1994. Reticulated platelet counts in patients undergoing autologous bone marrow transplantation: An aid in assessing marrow recovery. Am. J. Hematol. 46:319-324. Schmitz, G., Rothe, G., Ruf, A., Barlage, S., Tschöpe, D., Clemetson, K.J., Goodall, A.H., Michelson A.D., Nurden, A.T., and Shankey T.V. 1998. European Working Group on Clinical Cell Analysis: Consensus protocol for the flow cytometric characterisation of platelet function. Thromb. Haemost. 79:885-896. Watanabe, K., Takeuchi, K., Kawai, Y., Ikeda, Y., Kubota, F., and Nakamoto, H. 1995. Automated measurement of reticulated platelets in estimating thrombopoiesis. Eur. J. Haematol. 54:163171.
Goran B. Matic, Gregor Rothe, and Gerd Schmitz University of Regensburg Regensburg, Germany
Flow Cytometric Analysis of Reticulated Platelets
7.10.6 Supplement 6
Current Protocols in Cytometry
Assessment of Viability, Immunofluorescence, and DNA Content
UNIT 7.11
Nonviable (dead) cells that are present in considerable numbers in cell preparations to be analyzed on a flow cytometer can interfere with accurate data analysis. Problems arise from the increased autofluorescence, altered levels of cell-surface antigen expression, nonspecific antibody binding, and DNA fragmentation that can all be observed in dying and dead cells. Although nonviable cells can often be distinguished from viable cells on the flow cytometer by their light-scatter profiles (i.e., higher side-scatter and lower forward-scatter signals), a more reliable method is to identify cells with compromised membranes by their uptake of nonvital DNA dyes. Like trypan blue, these dyes are added directly to the cell preparation, where they enter and label dying and dead cells. Fluorescent indicator dyes for measurement of membrane integrity are widely used for dead-cell discrimination on the flow cytometer (UNIT 9.2) in unfixed and unpermeabilized cell preparations. However, for assays such as DNA-content analysis where cells need to be permeabilized, nonvital DNA dyes such as propidium iodide or ethidium bromide leak out of the stained dead cells into viable cells after permeabilization, causing the fluorescence distinction to disappear. In contrast, the nonvital dye 7-aminoactinomycin D (7-AAD) can be used in fixed and permeabilized cell preparations, because its diffusion from stained to unstained cells can be prevented by blocking DNA binding with its nonfluorescent analog actinomycin D (AD). The Basic Protocol describes a method for DNA-content analysis combined with deadcell discrimination using 7-AAD. Pyronin Y (PY; also known as pyronin G) is used for staining DNA (after removal of RNA by RNase digestion) because its fairly narrow emission in the orange range of the spectrum can be effectively separated on the flow cytometer from the far red emission of 7-AAD. Emissions from both PY and 7-AAD can be measured separately from the green emission of fluorescein isothiocyanate (FITC), permitting simultaneous staining with FITC-labeled probes. Alternate Protocol 1 describes staining for cell-surface antigen expression and DNA content combined with viability assessment using 7-AAD. Alternate Protocol 2 describes staining for intracellular antigen expression and DNA content combined with viability assessment using 7-AAD. NOTE: These protocols are of intermediate complexity and require basic understanding of cellular staining principles and flow cytometry. In addition, previous experience in setting up the flow cytometer for multicolor experiments will facilitate success. NOTE: The protocols are sensitive to differences in reagent concentrations and quality because of the need to balance the maintenance of 7-AAD staining with adequate cell permeabilization and with PY staining for DNA content. DNA CONTENT ANALYSIS IN COMBINATION WITH ASSESSMENT OF CELL VIABILITY
BASIC PROTOCOL
In this protocol, cells are first stained with 7-AAD for discrimination of intact cells from cells that have lost membrane integrity. The cells are then fixed and permeabilized, and stained with PY for DNA content.
Nucleic Acid Analysis Contributed by Ingrid Schmid Current Protocols in Cytometry (1999) 7.11.1-7.11.9 Copyright © 1999 by John Wiley & Sons, Inc.
7.11.1 Supplement 10
Materials Cell suspension (e.g., cells from suspension cultures; Ficoll-Hypaque-purified cells, either mononuclear cell fraction or suspension of cells isolated from tissue) PBS (APPENDIX 2A) 1 mg/ml 7-aminoactinomycin D (7-AAD) stock solution (see recipe) Staining buffer (see recipe) 1 mg/ml actinomycin D (AD) stock solution (see recipe) Fixation solution (see recipe) Permeabilization solution (see recipe), 37°C DNase-free RNase A (Sigma) 100 µg/ml pyronin Y (PY; also called pyronin G; Polysciences) stock solution in deionized, distilled H2O 12 × 75–mm culture tubes 37°C water bath Flow cytometer with 488-nm excitation and appropriate collection filters for PY (585-nm band-pass) and 7-AAD (650- or 670-nm long-pass) Additional reagents and equipment for counting cells (APPENDIX 3A) Stain dead cells 1. Count cells (APPENDIX 3A). 2. Wash an amount of cell suspension containing 106 cells by centrifuging 5 min at 250 × g, 4°C and removing the supernatant. Vortex briefly, resuspend in 2 ml PBS, repeat centrifugation, and discard supernatant. 3. Vortex briefly and resuspend in 250 µl staining buffer in a 12 × 75–mm culture tube. Add 1 µl of 1 mg/ml 7-AAD and mix well. For optimization of 7-AAD staining see Critical Parameters.
4. Incubate cells protected from light 30 min at 20° to 25°C. 5. Wash once with 2 ml PBS by centrifuging 5 min at 250 × g, 4°C. Remove the supernatant completely and vortex briefly. 6. Immediately add 2 ml 4°C PBS containing 2 µg/ml AD to the cell pellet and wash again by centrifuging 5 min at 250 × g, 4°C. Remove and discard supernatant. Complete removal of unbound 7-AAD is critical because residual 7-AAD dye in the solution will bind to the DNA of unstained live cells when cells are permeabilized. Therefore, it is important to remove as much liquid as possible after each wash without disturbing the cell pellet.
Fix and permeabilize cells and stain DNA 7. Resuspend the cell pellet in 875 µl 4°C PBS containing 2 µg/ml AD and vortex the mixture immediately. 8. Add 125 µl 4°C fixation solution and vortex immediately. Incubate 30 min at 4°C. Resuspend cells quickly throughout the procedure, because if the cells remain in close proximity for a prolonged time, transfer of 7-AAD between cells may occur.
9. Centrifuge the cell suspension 5 min at 250 × g, 4°C. Remove and discard supernatant. Assessment of Viability, Immunofluorescence, and DNA Content
A cell pellet may not be visible after the fixation step because fixed cells aggregate less well and therefore tend to spread out at the bottom of the tube.
7.11.2 Supplement 10
Current Protocols in Cytometry
10. Gently resuspended the cell pellet in 0.5 ml of 37°C permeabilization solution containing 2 µg/ml AD, 200 µg/ml DNase-free RNase A, and 2 µg/ml PY. Incubate 15 min at 37°C in a water bath. 11. Centrifuge cell suspension 5 min at 250 × g, 4°C. Remove and discard. 12. Resuspend cell pellet in 0.5 ml staining buffer containing 2 µg/ml AD, 1 µg/ml PY, and 200 µg/ml DNase-free RNase A. 13. Analyze on the flow cytometer with excitation at 488 nm. Collect PY fluorescence with a 585 ± 20–nm band-pass filter and 7-AAD fluorescence with either a 650-nm or a 670-nm long-pass filter. Set up photomultiplier tube voltage and compensation of spectral overlap by first using cells single-stained, with PY alone and with 7-AAD alone; then optimize flow cytometer settings further by instrument setting adjustments on dual-color stained samples to compensate for dye-dye interactions; also see Critical Parameters and see Anticipated Results.
ANALYSIS OF DNA CONTENT AND CELL-SURFACE ANTIGEN EXPRESSION COMBINATION WITH ASSESSMENT OF CELL VIABILITY
ALTERNATE PROTOCOL 1
Cells are simultaneously stained with FITC-labeled antibody to detect cell-surface antigen expression and with 7-AAD to discriminate dead cells. The cells are then fixed and permeabilized before staining with PY for DNA content. Additional Materials (also see Basic Protocol) FITC-labeled antibody to antigen of interest FITC-labeled isotypic control antibody 1. Add the appropriate amount of FITC-labeled antibody and 1 µl of 1 mg/ml 7-AAD stock solution to 1 × 106 PBS-washed cells resuspended in 250 µl staining buffer in a 12 × 75–mm culture tube, and mix well (see Basic Protocol, steps 1 to 3). Follow manufacturer’s recommendations for the appropriate antibody concentration. For antibodies obtained from non-commercial sources, titer the reagent on the same cell type prior to performing Alternate Protocol 1. Always set up a sample stained with the appropriate amount of FITC-labeled isotypic control antibody to determine background staining. For optimization of 7-AAD staining see Critical Parameters.
2. Stain, fix, and permeabilize cells as outlined in the previous procedure (see Basic Protocol, steps 4 to 12). 3. Analyze on the flow cytometer with excitation at 488 nm. Collect 7-AAD fluorescence using either a 650-nm or a 670-nm long-pass filter and FITC fluorescence using a 530 ± 15–nm band-pass filter. Set up photomultiplier tube voltage and compensation of spectral overlap by first using cells single stained with PY alone and with 7-AAD alone; then optimize flow cytometer settings further by instrument setting adjustments on dual-color stained samples to compensate for dye-dye interactions; also see Critical Parameters and see Anticipated Results.
ANALYSIS OF DNA CONTENT AND INTRACELLULAR ANTIGEN EXPRESSION IN COMBINATION WITH ASSESSMENT OF CELL VIABILITY Cells are stained with 7-AAD for discrimination of dead cells, then fixed and permeabilized. The cells are then stained with PY for DNA content and with FITC-labeled antibody for intracellular-antigen expression.
ALTERNATE PROTOCOL 2
Nucleic Acid Analysis
7.11.3 Current Protocols in Cytometry
Supplement 10
Additional Materials (also see Basic Protocol) FITC-labeled antibody to antigen of interest FITC-labeled isotypic control antibody Wash solution (see recipe) 1. Stain, fix, and permeabilize cells as outlined for the first procedure in this unit (see Basic Protocol, steps 1 to 11). 2. Resuspend cell pellet in 100 µl staining buffer containing 2 µg/ml AD, 1 µg/ml PY, and 200 µg/ml DNase-free RNase A. Add the appropriate amount of FITC-labeled antibody for staining of intracellular antigens. If the antibody has been specifically produced for flow cytometric intracellular staining, follow manufacturer’s recommendations for the appropriate antibody concentration. For other antibodies and antibodies obtained from non-commercial sources, titer the reagent on the same cell type prior to performing Alternate Protocol 2. Always set up a control sample with a FITC-labeled isotypic control antibody at the same protein concentration as the relevant antibody to determine background staining.
3. Incubate 30 min at 4°C. Numerous factors influence the formation of an intracellular antibody-antigen complex. Furthermore, antibodies for staining of intracellular antigens can differ dramatically in their reactivity with a given intracellular antigen. Therefore, the optimal antibody concentration and staining conditions have to be determined empirically and may require a different temperature, e.g., 20° to 25°C, and longer incubation times; also see Critical Parameters.
4. Add 1 ml 20% to 25% wash solution containing 2 µg/ml AD. Wash by centrifuging 5 min at 250 × g, 4°C. Remove and discard supernatant. Repeat. 5. Resuspend cell pellet in 0.5 ml staining buffer containing 2 µg/ml AD, 1 µg/ml PY, and 200 µg/ml DNase-free RNase A. 6. Analyze on the flow cytometer with excitation at 488 nm. Collect FITC fluorescence with a 530 ± 15–nm band-pass filter, PY fluorescence with a 585 ± 20–nm band-pass filter, and 7-AAD fluorescence with either a 650-nm or a 670-nm long-pass filter. Set up photomultiplier tube voltage and compensation of spectral overlap by first using cells single-stained with FITC-labeled antibody alone, with PY alone, and with 7-AAD alone; then optimize flow cytometer settings further by instrument setting adjustments on triple-color-stained samples to compensate for dye-dye and dye-fluorochrome interactions; also see Critical Parameters and see Anticipated Results.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Actinomycin D (AD) stock solution, 1 mg/ml To 1 mg actinomycin D (actinomycin C1; Roche Molecular Biochemicals) add 50 µl absolute ethanol. Mix well. Add 950 µl PBS (APPENDIX 2A) and mix again. Sonicate 10 min at 4°C, and keep at 4°C overnight before use. Store <1 month at 4°C. Do not exceed the sonication time, or the temperature in the solution will become too high and will detrimentally affect AD.
Assessment of Viability, Immunofluorescence, and DNA Content
For optimal preservation of 7-AAD staining of dead cells, a fresh AD stock solution is prepared weekly. Working solutions (prepared immediately before use) contain 2 ìg/ml AD. PBS containing 2 ìg/ml of AD is used at 4°C; permeabilization solution containing 2 ìg/ml of AD is used at 37°C; wash solution containing 2 ìg/ml of AD is used at 20° to 25°C.
7.11.4 Supplement 10
Current Protocols in Cytometry
7-Aminoactinomycin D (7-AAD) stock solution, 1 mg/ml To 1 mg 7-AAD (Calbiochem or Sigma) add 50 µl absolute methanol and mix well. Add 950 µl PBS (APPENDIX 2A) and mix again. Store protected from light <1 month at 4°C. Fixation solution Dissolve 2 g paraformaldehyde (Sigma) in 100 ml PBS (APPENDIX 2A); heat solution to 70°C in a fume hood for ∼1 hr. Cool to room temperature. Adjust pH to 7.2 with 0.1 M NaOH or 0.1 M HCl. Check pH periodically. Store <1 month protected from light at 4°C. For best results use only very pure preparations of paraformaldehyde and discard solutions that have become acidic during storage. Methanol-free formaldehyde solution is also available from Polysciences, Inc.
Permeabilization solution Mix 200 µl Tween 20 (Sigma) with 100 ml PBS (APPENDIX 2A). Store <1 month in an amber container at 4°C. Warm to 37°C before use. Staining buffer PBS without Ca2+ and Mg2+ (APPENDIX 2A) 0.1% (w/v) sodium azide 2% (v/v) newborn calf serum (e.g., Omega Scientific) Adjust pH to 7.2 with 0.1 M NaOH or 0.1 M HCl Store up to 6 months at 20° to 25°C. Wash solution Mix 100 µl Tween 20 (Sigma) with 100 ml PBS (APPENDIX 2A). Store <1 month in an amber container at 4°C. Warm to 20° to 25°C before use. COMMENTARY Background Information Identification of nonviable cells by means of fluorescent dyes that mark cells that have lost membrane integrity is commonly used for flow cytometric analysis of unfixed cells and cell preparations that have not been permeabilized. However, combining dead-cell discrimination with DNA-content measurements—where dyes have to pass through membranes to gain access to the DNA—is difficult, because after membrane permeabilization, fluorescent indicator dyes leak out of stained dead cells into unstained live cells. Muirhead et al. (1985) have circumvented this problem by preincubating cells with DNase before DNA staining. Because nonviable cells are permeable to DNase, partial degradation of their DNA occurs; dead cells can then be distinguished from live cells by their decreased DNA content (Muirhead et al., 1985). Another approach was described by Boltz et al. (1994), who used Hoechst 33342 for DNA staining on unfixed cells in combination with viability assessment by ethidium bromide staining; for this technique both dyes were excited by an ultra-
violet light source. A third method, outlined in the Basic Protocol and first described by Schmid et al. (1999), uses 7-AAD as a dead-cell discriminator in fixed and permeabilized cell preparations. This dye forms very stable complexes with DNA that have a slow dissociation rate. Furthermore, the transfer of 7-AAD from stained dead cells into live cells after cell permeabilization can be prevented by the addition of nonfluorescent actinomycin D, which blocks access of 7-AAD to the DNA of unstained cells (Fetterhoff et al., 1993; Schmid and Giorgi, 1995). For DNA-content measurement by flow cytometry, propidium iodide (PI) is the most widely used DNA dye. However, because the emission spectra of propidium iodide and 7AAD overlap, these two DNA dyes cannot be combined for DNA staining and viability assessment. In contrast, pyronin Y (PY) has an emission peak with a maximum at 569 nm that can be separated from the 7-AAD emission (emission maximum at 655 nm) on the flow cytometer. PY is the xanthene homologue of acridine orange, which intercalates into dou-
Nucleic Acid Analysis
7.11.5 Current Protocols in Cytometry
Supplement 10
ble-stranded DNA and RNA with a preference for G-C pairs (Müller and Crothers, 1975). The formation of its fluorescent complexes with nucleic acids is concentration dependent and has been proven quite complex (Darzynkiewicz et al., 1987; Kapuscinski and Darzynkiewicz, 1987). In flow cytometry, PY has commonly been used for estimation of RNA content in single cells (UNIT 7.3) following a preliminary step in which its access to DNA is blocked by preincubation with DNA-specific dyes (Tanke et al., 1980; Shapiro, 1981; Toba et al., 1995). By contrast, in the procedure described in the Basic Protocol, PY is used to label DNA while cellular RNA is degraded by ribonuclease A (RNase A) treatment. Furthermore, staining with PY for DNA content and 7-AAD for viability assessment can be applied to cell preparations that are simultaneously stained with FITC-labeled probes for either cell surface or intracellular immunofluorescence. Thus, the procedures described in Alternate Protocols 1 and 2 provide the option of collecting relevant data on cell proliferation by permitting accurate assessment of phenotype and DNA content in cell populations with low viability.
Critical Parameters and Troubleshooting 7-AAD staining 7-AAD fluorescence intensity in stained dead cells becomes dimmer after fixation and permeabilization; therefore, finding the optimal dye concentration for maximizing the separation between viable 7-AAD-negative and nonviable 7-AAD-positive cells is critical. Moreover, 7-AAD dye batches and 7-AAD obtained from different vendors can vary in staining intensity. Therefore, if sufficient separation between 7-AAD-negative and 7-AADpositive cells is not achieved, the optimal dye concentration must be determined by titration. However, higher concentrations of 7-AAD, (i.e., >10 µg/ml) may lead to increased background staining in live cells and require increased compensation (UNIT 1.14) between the 7-AAD channel and the PY channel when 7AAD and PY staining are combined.
Assessment of Viability, Immunofluorescence, and DNA Content
PY staining In the Basic Protocol fixed and permeabilized cells are stained with PY at concentrations of 1 to 2 µg/ml (3.3 to 6.6 µM) in the presence of RNase. Under these circumstances (after RNA digestion) it is possible to stain DNA with micromolar concentrations of PY,
although it has been shown previously that PY concentrations in the millimolar range are needed to form a fluorescent pyronin-DNA complex when RNA is present (Darzynkiewicz et al., 1987). Another important consideration is that many commercially available PY preparations have low purity and, because PY is hygroscopic, can contain variable amounts of water. Consequently, PY preparations obtained from different vendors and dye batches may vary in their nucleic acid staining. For consistent results always use PY of high purity, e.g., from Polysciences. Immunofluorescent staining While antibody staining for cell-surface antigen expression has become routine, intracellular staining experiments remain complicated (Schmid and Giorgi, 1995), because the numerous factors that influence the formation of a given antigen-antibody complex (e.g., antibody mobility, non-specific staining) are not completely understood. Intracellular-antigen staining depends on the ability of antibodies to cross cellular membranes. Various methods for cell permeabilization have been developed; they generally rely either on alcohol fixation/permeabilization or on formaldehyde fixation followed by detergent treatment. Optimal cell fixation depends on the cell type, on the susceptibility of the intracellular antigen to denaturation, on the intracellular antigen location, and on the reactivity of the antibody used for staining. When intracellular staining is combined with DNA staining, it is important to keep the concentration of the formaldehyde fixative low, i.e., between 0.25% and 0.5%. If too much fixative is used, the coefficients of variation of the DNA distributions will become unacceptably large due to increased histone and DNA cross-linking (Schmid et al., 1991). Furthermore, the reactivity of different antibodies with any given antigen can vary. Therefore, the optimal antibody concentration, incubation time, and incubation temperature must be determined empirically for each individual staining experiment. Flow cytometer setup The emission spectra of FITC, PY, and 7AAD differ enough that these fluorochromes can be combined for flow cytometric analysis using appropriate instrument settings for compensation of spectral-emission overlap. However, the standard technique of setting photomultiplier-tube voltages and spectral compensation using single-color control tubes will not
7.11.6 Supplement 10
Current Protocols in Cytometry
101
live 74%
100
100 80 60 40 20 0
120 80 40 0
104 103 102 101 100
F
200 160
Cell number
E
120 80 40
H 7-AAD fluorescence
1000 800 600 400 200 0
0
200 400 600 800 1000 Forward scatter
200 160 120 80 40 0
0
G Side scatter
120
160
Cell number
102
Cell number
103
200
I
104 103 102
Cell number
7-AAD fluorescence
D
C
B
104
Cell number
7-AAD fluorescence
A
R1
101
100 100
101
102
103
104
70 60 50 40 30 20 10 0 0
200 400 600 800 1000
CD8-FITC
PY (DNA)
Figure 7.11.1 Combination of cell-surface antigen and DNA content analysis with viability assessment. Human peripheral blood mononuclear cells (PBMC) were cultured for 7 days in the presence of 100 ng/ml surface-bound OKT-3 (Ortho Diagnostics), then stained with 7-AAD, CD8-FITC (Becton Dickinson Immunocytometry Systems; BDIS), and PY as described in Alternate Protocol 1. (A) Forward scatter (FSC) vs. 7-AAD fluorescence dot plot of cells stained with 7-AAD alone, ungated. (B) Histogram of CD8 fluorescence of cells stained with CD8-FITC alone, ungated. (C) Histogram of PY (DNA) fluorescence of cells stained with PY alone, gated on singlets using a dot plot of PY fluorescence (FL2)-width (W) vs. FL2-area (A). (D) FSC vs. 7-AAD fluorescence dot plot of cells triple-stained with 7-AAD, CD8, and PY run with flow cytometer settings optimized for single-color stained samples, ungated. (E) Histogram of CD8 fluorescence on triple-stained cells run with settings optimized for single-color-stained samples, ungated. (F) Histogram of PY (DNA) fluorescence of triple-stained cells run with settings optimized for single-color stained samples, gated on singlets using a dot plot of FL2-W vs. FL2-A. (G) FSC vs. side scatter dot plot of triple-stained cells, ungated. (H) Dot plot of CD8 fluorescence vs. 7-AAD fluorescence of triple-stained cells, ungated. (I) Histogram of PY (DNA) fluorescence of CD8-positive, 7-AAD-negative, live cells, gated on R1 as shown in (H) and on singlets using a dot plot of FL2-W vs. FL2-A. Fluorescence distributions shown in (H) and (I) were obtained by adjusting instrument settings on triple-stained cells for dye-dye and dye-fluorochrome interactions.
produce an optimal instrument setup for samples stained simultaneously with 7-AAD, FITC, and PY. For instance, in Figure 7.11.1A, 7.11.1B, and 7.11.1C flow cytometer settings have been optimized for single-color control tubes. When these settings were used to run a sample that was triple-stained with 7-AAD, CD8-FITC, and PY (Figure 7.11.1D, E, and F, respectively), 7-AAD fluorescence was markedly changed while CD8 and PY fluorescence distributions remained on scale. The differ-
ences in staining patterns between three-colorstained samples and those stained individually may be caused by the ability of PY to quench the fluorescence intensity of other fluorochromes when it is added to the cell suspension. Another contributing factor may be energy transfer due to fluorochrome proximity between 7-AAD and PY and between FITC and PY when DNA staining is combined with internal staining using FITC-labeled antibodies. Therefore, to compensate for these interac-
Nucleic Acid Analysis
7.11.7 Current Protocols in Cytometry
Supplement 10
B
A
104
7-AAD fluorescence
Side scatter
1000 800 600 400 200
102 R1 live 64%
101 100
0 0
200
C
600 800 900 Forward scatter
0
400
D
200
400
600
800 1000
Forward scatter
150 Cell number
180 Cell number
103
150
120
120 90 60
90 60 30
30
0
0 0
200
400 600 PY (DNA)
800 1000
100
101
103 102 CD3-FITC
104
Figure 7.11.2 Combination of DNA content and intracellular antigen analysis with viability assessment. Molt-4f cells were stained with 7-AAD, PY, and CD3-FITC (BDIS) as described in Alternate Protocol 2. (A) FSC vs. side scatter dot plot, ungated. (B) FSC vs. 7-AAD fluorescence dot plot, ungated. (C) Histogram overlay of PY (DNA) fluorescence gated on singlets using a dot plot of PY fluorescence (FL2)-W vs. FL2-A (thin solid line) and singlet PY fluorescence of 7-AAD-negative, live cells (within R1 as shown in B; bold solid line). (D) Histogram overlay of CD3-FITC fluorescence ungated (thin solid line) and gated on 7-AAD-negative, live cells (within R1 as shown in B; bold solid line).
tions, make instrument-setting adjustments on triple-stained samples and move the cell clusters back on scale until the pattern matches the one observed on single-stained tubes. As demonstrated in Figure 7.11.1H and I, this strategy permits clear discrimination of stained cell clusters for analysis.
Anticipated Results
Assessment of Viability, Immunofluorescence, and DNA Content
The first example shows the application of the method described in Alternate Protocol 1 to the analysis of CD8-bearing, proliferating cells in a culture of human peripheral blood mononuclear cells. The forward versus side scatter dot plot of this fixed and permeabilized cell preparation (Fig. 7.11.1G) does not indicate the presence of any dead cells, while 26% 7-AAD-positive (dead) cells appear in the Figure 7.11.1A dot plot. As expected, after 7 days of culture most of the dim CD8-positive cells have died;
furthermore, as compared to all cells (Fig. 7.11.1C and 7.11.1F), CD8-positive, live cells contained fewer cells in G1 and in S phase and more cells in the G2/M phase of the cell cycle (Fig. 7.11.1I). The second example shows the application of the method described in Alternate Protocol 2 to the analysis of a T-cell leukemia cell line that has been stained for intracellular expression of CD3 (Fig. 7.11.2). Again, as with Figure 7.11.1, it is impossible to discriminate dead cells on the basis of light-scatter differences because of the overlap of live with dead cells in the area of low forward and low side scatter (Fig. 7.11.2A). In contrast, 7-AAD-positive cells can be clearly discriminated from the 7-AAD-negative cells on the forward scatter versus 7-AAD fluorescence dot plot (Fig. 7.11.2B). DNA histograms that either include or exclude 7-AAD-positive, dead cells differ
7.11.8 Supplement 10
Current Protocols in Cytometry
(Fig. 7.11.1C); there are more live cells in the G1 phase of the cell cycle and fewer in S and G2/M. In addition, the coefficient of variation of the G1 peak decreased when only live cells were included in the gate. Furthermore, live Molt-4f cells expressed higher levels of intracellular CD3 than did dead cells (Fig. 7.11.1D). These data emphasize the need for dead cell exclusion in cell preparations that contain considerable numbers of dead or dying cells.
Time Considerations The procedure outlined in the Basic Protocol will take ∼2 hr. When cell-surface antigen staining is also performed as described in Alternate Protocol 1, no additional time is required because the incubation with the antibody is done during 7-AAD staining. When intracellular staining is performed as outlined in Alternate Protocol 2, additional time is needed for the antibody incubation (e.g., 30 min) and the two washing steps.
Literature Cited Boltz, R.C., Fischer, P.A., Wicker, L.S., and Peterson, L.B. 1994. Single UV excitation of Hoechst 33342 and ethidium bromide for simultaneous cell cycle analysis and viability determinations on in vitro cultures of murine B lymphocytes. Cytometry 15:28-34. Darzynkiewicz, Z., Kapuscinski, J., Traganos, F., and Crissman, H.A. 1987. Application of pyronin Y(G) in cytochemistry of nucleic acids. Cytometry 8:138-145.
Müller, W. and Crothers, D.M. 1975. Interactions of heteroaromatic compounds with nucleic acids. Eur. J. Biochem. 54:267-277. Schmid, I. and Giorgi, J.V. 1995. Section on Intracellular Staining in Holmes, K., Foulkes, B.J., Schmid, I., and Giorgi, J.V. Preparation of cells and reagents for flow cytometry. In Current Protocols in Immunology (J.E. Coligan, A.M. Kruisbeek, D.H. Margulies, E.M. Shevach, and W. Strober, eds.) pp. 5.3.1-5.3.23. John Wiley & Sons, New York. Schmid, I., Uittenbogaart, C.H., and Giorgi, J.V., 1991. A gentle fixation and permeabilization method for combined cell surface and intracellular staining with improved precision in DNA quantification. Cytometry 12:279-285. Schmid, I., Ferbas, J., Uittenbogaart, C.H., and Giorgi, J.V. 1999. Flow cytometric analysis of live cell proliferation and phenotype in populations with low viability. Cytometry 35:64-74. Shapiro, H.M. 1981. Flow cytometric estimation of DNA and RNA content in intact cells stained with Hoechst 33342 and Pyronin Y. Cytometry 2:143-150. Tanke, H.J., Nieuwenhuis, I.A.B., Koper, G.J.M., Slats, J.C.M., and Ploem, J.S. 1980. Flow cytometry of human reticulocytes based on RNA fluorescence. Cytometry 1:313-320. Toba, K., Winton, E.F., Koike, T., and Shibata, A. 1995. Simultaneous three-color analysis of the surface phenotype and DNA-RNA quantitation using 7-aminoactinomycin D and pyronin Y. J. Immunol. Methods 182:193-207.
Key Reference Schmid et al., 1999. See above.
Fetterhoff, T.J., Holland, S.P., and Wile, K.J. 1993. Fluorescent detection of non-viable cells in fixed cell preparations [Abstract]. Cytometry Supplement 6:27.
Describes the procedures presented in the Basic Protocol and Alternate Protocols 1 and 2.
Kapuscinski, J. and Darzynkiewicz, Z. 1987. Interactions of pyronin Y(G) with nucleic acids. Cytometry 8:129-137.
Contributed by Ingrid Schmid UCLA School of Medicine Los Angeles, California
Muirhead, K.A., Kloszewski, E.D., Antell, L.A., and Griswold, D.E. 1985. Identification of live cells for flow cytometric analysis of lymphoid subset proliferation in low viability populations. J. Immunol. Methods 77:77-86.
Nucleic Acid Analysis
7.11.9 Current Protocols in Cytometry
Supplement 10
Flow Cytometric Analysis of RNA Synthesis by Detection of Bromouridine Incorporation
UNIT 7.12
The cellular synthesis of RNA can be analyzed by flow or image cytometry, based on in-vitro or in-vivo incorporation of the RNA precursor 5′-bromouridine (BrU), followed by its immunocytochemical detection with a suitable cross-reacting anti-5′-bromo-2′-deoxyuridine (BrdU) antibody. This dynamic approach, which may be valuable in studies of cellular activation and gene expression, yields information that is complementary to that provided by analysis of the cellular RNA content at a given time point (see UNIT 7.3). The method enables measurement of the amount of BrU incorporated during a preset time interval into the cells of a particular subset, which is detected by additional cytometric markers, such as DNA content (see Basic Protocol and Alternate Protocols 1 and 2) or cell surface antigen expression (see Alternate Protocol 3). It is necessary to permeabilize the cells to enable the immunochemical staining of the intracellular BrU-substituted RNA, and to fix the cellular or nuclear suspension in order to avoid breakdown of RNA by cellular RNases. However, results from the various methods used in cell preparation for flow cytometry differ depending on whether the particular subtypes of BrU-substituted RNA that are localized to the nuclear matrix, nucleoli, and cytoplasm, respectively, are included in the detected immunofluorescence signal. The qualities of the particular method in this respect may be examined by fluorescence microscopy of the BrU-labeled cells or nuclei (see Support Protocol). One method for preparation of nuclear suspensions (see Basic Protocol) uses detergent at physiological pH and results in a diffuse pattern of extranucleolar staining with occasional dim staining of nucleoli. Another method for preparation of nuclear suspensions, by treatment at pH 2, reveals intense staining of BrU in both nucleoli and extranucleolar foci (see Alternate Protocol 1). With a method for preparation of entire cells (see Alternate Protocol 2), moderate cytoplasmic and intense nuclear staining, but without staining of the nucleoli, may be obtained. BIVARIATE ANALYSIS OF DNA CONTENT AND BrU INCORPORATION IN NUCLEI
BASIC PROTOCOL
The method of cell preparation here is according to Landberg and Roos (1991), and that method was adapted to this protocol by Jensen et al. (1993a). An advantage of this method is a high recovery of fixed nuclei; a disadvantage is that no distinct staining of nucleoli is obtained (Fig. 7.12.1A). Materials Cells to be labeled and appropriate serum-containing tissue culture medium (APPENDIX 3B) 100 mM BrU stock solution (see recipe) Phosphate-buffered saline (PBS), pH 7.2 (APPENDIX 2A; store up to 2 months at 4°C) Landberg’s lysis buffer (see recipe) 100% methanol, –20°C PBS (APPENDIX 2A) containing 0.01% (v/v) Nonidet P-40 (NP-40) Anti-BrdU antibody (see recipe) Polyclonal FITC-conjugated rabbit anti-mouse immunoglobulins (F[ab′]2 fragment, Dako; protect from light during storage and handling) PBS (APPENDIX 2A) containing 5% normal rabbit serum (Dako) Nucleic Acid Analysis Contributed by Jørgen K. Larsen, Peter Østrup Jensen, and Jacob Larsen Current Protocols in Cytometry (2000) 7.12.1-7.12.11 Copyright © 2000 by John Wiley & Sons, Inc.
7.12.1 Supplement 12
50 µg/ml propidium iodide (PI) in PBS (store up to several months at 4°C; protect from light during storage and handling) 20 mg/ml RNase (ribonuclease I-A, Sigma) in PBS (store up to 2 weeks at 4°C; optional) Green fluorescence standard microspheres (e.g., 1.7-µm microspheres, Fluoresbrite 17687, Polysciences) Unfixed chicken or trout erythrocytes stained with PI (see UNITS 7.5 & 7.6) Sample tubes as required for flow cytometer Refrigerated centrifuge, 4°C
Cytometric Analysis of RNA Synthesis
Figure 7.12.1 Fluorescence photomicrographs showing the subcellular distribution of BrUsubstituted RNA in HL-60 cells prepared by different methods, labeled for 1 hr with 1 mM BrU (see Basic Protocol and Alternate Protocols 1 to 3). All samples were stained indirectly with anti-BrdU antibody (ABDM, Partec), but staining with propidium iodide or phycoerythrin-conjugated antibody was omitted (see Support Protocol). The FITC staining pattern of incorporated BrU (left panels) may be compared with the localization of nuclei and nucleoli according to simultaneous staining of DNA with DAPI (right panels). (A and B) Lysis of cells with Triton X-100, followed by fixation of the nuclei with methanol (Basic Protocol). (C and D) Lysis of cells with citric acid and Tween 20 at pH 2, followed by fixation of the nuclei with methanol (Alternate Protocol 1). (E and F) Fixation of cells with formaldehyde followed by permeabilization with ethanol (Alternate Protocol 2). (G and H) Fixation of cells with a mixture of formaldehyde and Nonidet P-40 (Alternate Protocol 3). All methods resulted in FITC staining of the nuclei. However, the nucleoli became brightly stained with FITC only with Alternate Protocol 1 (C). Cytoplasmic FITC staining was observed with Alternate Protocol 2 (E) and Alternate Protocol 3 (G). The photomicrographs are shown with equal magnification (100× objective, bar size 10 µm).
7.12.2 Supplement 12
Current Protocols in Cytometry
Flow cytometer with 488-nm argon-laser excitation and emission detectors for FITC fluorescence (525 ± 20 nm band-pass) and PI fluorescence (630 ± 20 nm band-pass) Additional reagents and equipment for counting cells (APPENDIX 3A) Label cells with BrU 1. Add 1/100 vol of 100 mM BrU stock solution to the cell culture (1 mM final concentration). Include one cell culture without BrU as an unlabeled control. Incubate for a pulse period of 30 to 60 min under culture conditions (protect BrU solutions and labeled cells from intense light exposure), then transfer the cultures to an ice bath to arrest BrU labeling. This default procedure is for in-vitro experiments using BrU pulse-labeling of cells growing in suspension. For adherent cell cultures, after labeling as above, harvest by resuspending the cells according to the standard treatment with trypsin/EDTA (APPENDIX 2A) performed at culture temperature, then place the harvested cell suspension on an ice bath to arrest further labeling. Wash the cells once in PBS (see step 2), add a small amount of serum to inactivate residual trypsin, then wash again once in PBS as in step 2. For pulse-chase experiments in studies of the processing of labeled RNA, replace the medium with fresh BrU-free, prewarmed medium after the pulse period, and subsequently harvest cells at different time intervals.
2. Estimate the number of cells harvested from BrU-labeled and unlabeled cultures, respectively, by counting in a hemacytometer (APPENDIX 3A). Wash the cells by adding 5 to 10 ml PBS, centrifuging 5 min at 300 × g, 4°C, and aspirating the supernatant, leaving the cell pellet suspended in a residual volume of ∼100 µl. Prepare suspensions of fixed nuclei 3. Add 0.5 ml Landberg’s lysis buffer and incubate 15 min by agitating the tubes slowly in an ice bath. 4. Without intermediate washing, fix the nuclei by adding 3 ml methanol precooled to −20°C. Immediately close the tubes and mix the cells by turning the tubes upside down. Store at −20°C. Fixation at higher temperature increases the risk for cell (nuclei) aggregation. CAUTION: Methanol and methanol-containing reagents should be handled in a fume hood.
Stain nuclear suspensions 5. Transfer samples of 2–5 × 105 fixed cell nuclei from the batches of harvested BrU-labeled and unlabeled cells, respectively, to sample tubes suitable for the flow cytometer. Resuspend nuclei in 3 ml PBS, and centrifuge 5 min at 300 × g. 6. Remove supernatant and resuspend cell pellet in 50 µl monoclonal anti-BrdU antibody diluted 1:10 in PBS containing 0.01% NP-40. Agitate the samples slowly in an ice bath for 60 min. Wash once in PBS by centrifuging 5 min at 300 × g, 4°C, and aspirating the supernatant. 7. Add 50 µl of FITC-conjugated polyclonal rabbit anti-mouse immunoglobulins, diluted 1:10 in PBS containing 5% normal rabbit serum, to each pellet. Agitate the samples slowly in an ice bath for 60 min. Wash once in PBS by centrifuging 5 min at 300 × g, 4°C, and aspirating the supernatant. 8. Add 100 µl of 50 µg/ml PI to each sample. 9. Optional: Add 25 µl of 20 mg/ml RNase to each sample. Nucleic Acid Analysis
7.12.3 Current Protocols in Cytometry
Supplement 12
Treatment with RNase improves the coefficient of variation (CV) of the DNA measurement and usually does not remove the immunochemically stained BrU. However, if RNase is added before the immunochemical staining, the BrU signal will decrease to background level.
10. Agitate samples slowly in an ice bath for 15 min. Keep the samples in the ice bath until ready for analysis by flow cytometry. Acquire fluorescence data on flow cytometer 11. Set up flow cytometer. Collect forward and side light scatter, FITC fluorescence (log scale), PI fluorescence pulse height and pulse area (linear scale), and elapsed time. Use PI fluorescence as threshold trigger for the elimination of noise signal from debris particles. Compensation for emission overlap is generally not necessary.
12. Optimize the DNA measurement resolution using suitable reference cells, e.g., unfixed chicken or trout erythrocytes stained with detergent and PI, to ensure minimum CV. Use a relatively low sample flow rate to decrease the core stream width, thus increasing the measurement precision. If chicken or trout erythrocytes are to be used as internal DNA content references, they must be prepared and stained by exactly the same procedure as the test cells.
13. Calibrate the FITC fluorescence scale using green fluorescent standard microspheres. The microspheres may be added to the samples as an internal reference. In that case it may be necessary to use light scatter or a supplementary fluorescence measurement with a broader bandwidth (e.g., transmitting at >515 nm) as the threshold trigger.
14. Run the test samples and acquire listmode files of 1–5 × 104 total counts. Analyze listmode data files 15. When relevant, gate the bivariate distribution of log FITC fluorescence versus PI fluorescence area by the following parameters. a. Forward or side scatter versus PI fluorescence, to exclude nonlysed cells or cell aggregates. b. PI fluorescence versus elapsed time, to ensure that the sample count rate and the PI staining conditions during measurement have been stable. c. PI fluorescence area versus PI fluorescence height or width (pulse-shape analysis), to eliminate doublets or coincident measurements. d. Intensity of the PI fluorescence, to distinguish in histograms and correlate cell cycle phase or DNA ploidy with BrU labeling. 16. Read the BrU labeling index from the gated FITC versus PI fluorescence distribution, by comparing the test sample with the nonlabeled control sample. 17. Read the BrU labeling intensity by comparing with green fluorescent standard microspheres, and with the nonlabeled control sample.
Cytometric Analysis of RNA Synthesis
7.12.4 Supplement 12
Current Protocols in Cytometry
BIVARIATE ANALYSIS OF DNA CONTENT AND BrU INCORPORATION IN NUCLEI, INCLUDING NUCLEOLI
ALTERNATE PROTOCOL 1
The method of cell preparation is modified after Otto (1990). An advantage of this method is a distinct staining of nucleoli (Fig. 7.12.1C). Disadvantages are a lower recovery of fixed nuclei and a higher risk of cell aggregation and clogging because of swelling of the nuclei. Additional Materials (also see Basic Protocol) Otto’s cell pretreatment buffer (see recipe) 1. Label cells with BrU, count, and wash (see Basic Protocol, steps 1 and 2). 2. Add 1 ml of Otto’s pretreatment buffer to cells. Close the tubes and mix the cells by turning tubes upside down; then agitate the tubes slowly for 15 min in an ice bath. Wash the nuclei in 5 to 10 ml PBS by centrifuging 5 min at 300 × g, 4°C, and aspirating the supernatant. Nuclei are released from adherent cultures by this treatment, so prior treatment with trypsin/EDTA is not necessary.
3. Fix the nuclei by addition of 1 ml methanol at −20°C. Immediately close the tubes and mix the cells by turning tubes upside down. Store the samples at −20°C. 4. Stain the cells (see Basic Protocol, steps 5 to 8 and 10) and perform flow cytometry (see Basic Protocol, steps 11 to 17). With this protocol, avoid final treatment with RNase (see Basic Protocol, step 9), which will result in loss of nuclear integrity and further decrease the recovery of nuclei. However, a good CV of the DNA measurement is obtained without RNase treatment.
BIVARIATE ANALYSIS OF DNA CONTENT AND BrU INCORPORATION IN CELLS
ALTERNATE PROTOCOL 2
The method of cell preparation is according to Li et al. (1994). An advantage of this method is that the processing of labeled RNA to cytoplasmic RNA can be monitored. Additional Materials (also see Basic Protocol) 1% paraformaldehyde (see recipe), cold 70% ethanol 1. Label cells with BrU, count, and wash (see Basic Protocol, steps 1 and 2). 2. Fix the cells in suspension by addition of 1 ml cold 1% paraformaldehyde. Immediately close the tubes and mix the cells by turning the tubes upside down. Agitate the tubes slowly for 15 min in an ice bath. Wash the fixed samples in 5 to 10 ml cold PBS by centrifuging 5 min at 300 × g, 4°C, and aspirating the supernatant. 3. Permeabilize the cells by addition of 1 ml 70% ethanol at −20°C. Store the fixed cells at −20°C. 4. Stain cells and perform flow cytometry (see Basic Protocol, steps 5 to 17).
Nucleic Acid Analysis
7.12.5 Current Protocols in Cytometry
Supplement 12
ALTERNATE PROTOCOL 3
BIVARIATE ANALYSIS OF CELL SURFACE ANTIGEN EXPRESSION AND BrU INCORPORATION The method of cell preparation is modified after Carayon and Bord (1992), and used for this purpose by Jensen et al. (1993b). The method has been applied to simultaneous staining of a cell surface antigen, but may also be used for staining of an intracellular antigen; in that case the cells must be permeabilized (steps 4 and 5) before staining of the intracellular antigen (as in step 3). Additional Materials (also see Basic Protocol 1) R-phycoerythrin-conjugated monoclonal antibody, specific for cell surface antigen of choice (protect from light during storage and handling) R-phycoerythrin-conjugated negative control antibody of same isotype Phosphate-buffered saline (PBS; APPENDIX 2A) PBS (APPENDIX 2A) containing 0.1% (w/v) bovine serum albumin (BSA) PBS (APPENDIX 2A) containing 1% (w/v) paraformaldehyde and 0.05% (v/v) Nonidet P-40 (NP-40) PBS (APPENDIX 2A) containing 1% (w/v) glycine FITC-conjugated anti-BrU antibody (e.g., Becton Dickinson 7583) PBS (APPENDIX 2A) containing 0.1% (w/v) BSA and 0.1% (v/v) NP-40 Standard microspheres (FITC, R-phycoerythrin, and blank) for compensation of spectral emission overlap (e.g., Calibrite Beads, Becton Dickinson) 1. Label cells with BrU, count, and wash (see Basic Protocol, steps 1 and 2). 2. Transfer two samples of ∼2 × 105 unfixed cells from each of the batches of harvested BrU-labeled and unlabeled cells, respectively, to sample tubes suitable for the flow cytometer. This enables staining of cell surface antigen antibody (in one sample) as well as control antibody (in another).
3. Dilute R-phycoerythrin-conjugated antibody directed against cell surface antigen, in PBS containing 0.1% BSA, to the final concentration specified by the manufacturer. Add 100 µl to the one set of BrU-labeled and unlabeled cells, respectively. Also dilute R-phycoerythrin-conjugated negative control antibody of the same isotype and add equal amounts to the other parallel set. Stain for 30 min at room temperature. Wash once in 3 ml PBS by centrifuging 5 min at 300 × g, 4°C, and aspirating the supernatant. 4. Fix the cells in suspension by adding 1 ml of PBS containing 1% paraformaldehyde and 0.05% (v/v) NP-40 at room temperature. Immediately close the tubes and mix the cells by turning the tubes upside down. Agitate slowly 15 min at room temperature. Store the fixed nuclei cells in the paraformaldehyde solution at 4°C at least overnight before staining. Fixed cells can be kept for several days in the refrigerator until further staining.
5. Wash cells once in cold 1% glycine in PBS, and then once in PBS, by centrifuging 5 min at 300 × g, 4°C, and aspirating the supernatant. 6. Dilute FITC-conjugated anti-BdU antibody 25× in PBS containing 0.1% BSA and 0.1% (v/v) NP-40. Add 100 µl of the diluted antibody to each sample tube and stain for 45 min. Wash once in 3 ml PBS with 0.1% BSA by centrifuging 5 min at 300 × g, 4°C, and aspirating the supernatant. Cytometric Analysis of RNA Synthesis
7. Resuspend with 200 µl PBS containing 0.1% BSA and 0.1% (v/v) Nonidet P40.
7.12.6 Supplement 12
Current Protocols in Cytometry
8. Set up flow cytometer. Collect forward and side light scatter, FITC fluorescence (530 ± 20 nm band-pass, log scale), and R-phycoerythrin fluorescence (575 ± 20 nm band-pass, log scale). Use forward scatter as threshold trigger for the elimination of noise signal from debris particles. 9. Compensate for spectral emission overlap using, e.g., Becton Dickinson Calibrite beads. 10. Run the test samples and acquire listmode files of 1–2 × 104 total counts. 11. Gate the bivariate distribution of FITC fluorescence versus R-phycoerythrin fluorescence by the forward and side scatter to exclude debris and aggregates. 12. Read the BrU labeling index for the subpopulations of surface antigen-expressing and nonexpressing cells, using a quadrant region setting. FLUORESCENCE MICROSCOPY OF SUBCELLULAR DISTRIBUTION OF BrU Fluorescence microscopy of cells stained for flow cytometric analysis may serve as a quality control with respect to cell morphology and proper intracellular localization of the BrU staining, and is particularly useful in optimizing flow cytometric protocols for new cell lines or tissues.
SUPPORT PROTOCOL
Additional Materials (also see Basic Protocol and Alternate Protocols 1 to 3) Embedding medium (e.g., Fluoromount-G, Southern Biotechnology Associates) containing 0.5 µg/ml 4′,6-diamidino-2-phenylindole (DAPI; e.g., Serva) Cytocentrifuge with containers of, e.g., 1 ml, 30 mm2 size (Zytokammer, Hettich-Zentrifugen) Fluorescence microscope with filter sets for FITC and DAPI fluorescence 1. Label cells with BrU and stain with anti-BrdU antibody (see Basic Protocol and Alternate Protocols 1 to 3), but omit the final staining with PI. 2. Transfer aliquots of the stained suspension of cells or nuclei to cytocentrifuge containers and centrifuge 5 min at 800 × g. Nuclear suspensions prepared according to Alternate Protocol 1 are not suitable for cytocentrifugation; therefore, pipet aliquots of nuclear suspension onto microscope slides and air dry for 30 min.
3. Mount the slides with embedding medium containing DAPI for staining DNA. 4. Examine the subcellular distribution of the BrU according to colocalization of the FITC and DAPI fluorescence images (also see Anticipated Results). REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Anti-BrdU antibodies For the Basic Protocol, dilute ABDM mouse anti-BrdU monoclonal antibody (Partec) 1:10 in PBS containing 0.01% (v/v) Nonidet P-40 (NP-40). In addition to this antibody, several anti-BrdU antibodies exist that will bind to BrU in RNA: clone B-44 (Becton-Dickinson), and clone BMC9318 (Boehringer Mannheim), both from mouse, and clone BU1/75 (Sera-Lab) derived from rats. However, the clone BU20a antiBrdU antibody (Dako), which is derived from mice immunized with BrdU, shows no cross-reactivity to BrU.
Nucleic Acid Analysis
7.12.7 Current Protocols in Cytometry
Supplement 12
BrU stock solution, 100 mM Dissolve 323 mg BrU (5-bromouridine; Aldrich) in 10 ml PBS (APPENDIX 2A). Store this stock solution in aliquots up to several weeks at −80°C. Protect from light during storage and handling. CAUTION: When handling BrU, wear suitable protective clothing, as it is toxic.
Landberg’s lysis buffer PBS (APPENDIX 2A) containing: 0.5% (v/v) Triton X-100 1% bovine serum albumin (BSA; e.g., Behringwerke AG Diagnostica) 0.2 µg/ml EDTA Stir to dissolve the BSA Store up to 2 weeks at 4°C. Otto’s cell pretreatment buffer Distilled H2O containing: 2.1% (w/v) citric acid 0.5% (v/v) Tween 20 Final pH ∼2 Store up to 2 weeks at 4°C Paraformaldehyde, 1% Dissolve 1 g paraformaldehyde in 100 ml PBS (APPENDIX 2A) to prepare a methanolfree formaldehyde solution. (Paraformaldehyde will dissolve quickly when the mixture is heated to ∼70°C.) Adjust the pH to 7.4 by addition of HCl or NaOH if needed. Store up to 2 weeks at 4°C. CAUTION: Use a safety hood for handling formaldehyde-containing reagents and samples, because of the toxic fumes. Methanol-free paraformaldehyde is also available from Polysciences.
COMMENTARY Background Information
Cytometric Analysis of RNA Synthesis
For dynamic measurements of RNA synthesis in living cells, RNA in cell cultures or living organisms must be labeled in vitro or in vivo with a detectable precursor during the time period of interest. RNA synthesis has traditionally been investigated by incorporating [3H]uridine and counting the silver grains developed in an autoradiographic film over the individual cells. As an alternative to this timeconsuming and laborious method, a faster, nonradioactive technology has emerged based on immunocytochemical detection. This method utilizes the incorporation of the brominated RNA precursor BrU (Jensen et al., 1993a) followed by immunofluorescent staining using anti-bromodeoxyuridine (BrdU) antibodies. BrU enters the live cell and phosphate groups are attached by the cell’s own anabolic machinery to produce BrUTP, which is suitable for incorporation into nascent RNA by the RNA polymerases during transcription. In contrast to BrU, the incorporation of other brominated RNA precursors such as BrUTP, biotin-UTP, or
fluorescein-UTP into nascent RNA depends on methods for facilitating the access of the reagent to the transcription sites in the cells, e.g., by stripping or permeabilization of the plasma membrane, microinjection, or fusion with precursor-containing liposomes. The labeling efficiency of the various RNA precursors, analogs of uridine, into the various types of RNA has not been completely investigated. Similar to [3H]uridine incorporation, exposure to BrU has little effect initially on transcription rates (Jackson et al., 1998), but the chemically modified uridines are known to inhibit the subsequent processing of RNA transcripts (Wansink et al., 1994). The BrU-substituted RNA may be detected by permeabilizing the cells and staining with certain anti-BrdU antibodies. Indeed, several of the antibodies that are marketed as anti-BrdU antibodies for use in detection of DNA-replicating cells labeled with BrdU (see UNIT 7.7) are produced by hybridomas derived from rodents that have been immunized with halogenated uridine, and not deoxyuridine (Dolbeare,
7.12.8 Supplement 12
Current Protocols in Cytometry
1995). It is therefore not surprising that these antibodies also bind to BrU incorporated into RNA (for examples of suitable antibodies, see Reagents and Solutions). A FITC-conjugated anti-BrdU antibody may be used for visualization of the BrU-substituted RNA (see Alternate Protocol 3), or this may be indirectly stained with a fluorochrome-conjugated secondary antibody (see Basic Protocol and Alternate Protocols 1 and 2). The secondary structure of RNA may influence the accessibility of the antibodies to the BrU-labeled RNA, in analogy with the well known experience from staining of BrdU-labeled DNA (see UNIT 7.7), where staining with anti-BrdU antibodies requires that the BrdU be exposed in a denatured, single-stranded form (Dolbeare, 1995). Thus, it is to be expected that BrU-substituted singlestranded RNA is far more accessible to the anti-BrdU antibodies than is BrU-substituted double-stranded RNA, and furthermore that a partial removal of proteins enhances the accessibility of the antigen (Dolbeare, 1995). Since RNA is heterogeneous with respect to its configuration, location, and stability, the nature of the binding of anti-BrdU antibody may influence the staining pattern of BrU-labeled cells differently with regard to the various types of RNA. The measurement of incorporated BrU by flow cytometry has provided a basis for correlating the overall transcriptional activity to the cell cycle by simultaneous measurement of DNA content (Jensen et al., 1993a; Li et al., 1994; Haider et al., 1997), as well as to the phenotype by simultaneous measurement of a cell surface marker (Jensen et al., 1993b). In toxicological studies, it was shown that the RNA synthesis according to BrU incorporation was affected at a lower concentration of 5azacytidine than necessary for affecting the DNA synthesis (Murakami et al., 1995). Immunoseparation of the RNA labeled with brominated precursors during a short labeling period and conditioned by specific external stimuli, followed by RT-PCR and Northern blotting or sequencing for relating the signal to specific genes, seems to open new perspectives in transcriptional research (Haider et al., 1997).
Critical Parameters and Troubleshooting For control of the specificity of the immunofluorescence signal, comparison with a BrU-unlabeled sample as well as a BrU-labeled and RNase-treated sample is recommended. For this purpose, it is important to apply the RNase treatment before immunochemical
staining of the BrU. The FITC fluorescence is then reduced to the level of BrU-unlabeled cells (Jensen et al., 1993a). Further indication of specificity is that addition to the cell culture of the RNA polymerase inhibitor actinomycin D prior to BrU results in decreased FITC fluorescence. Moreover, DNA that might be labeled due to ribonucleotide reductase activity, which could convert BrU into BrdU and lead to labeling of DNA during replication, is not detectable by the Basic Protocol (Jensen et al., 1993a). Because the BrU signal is quite stable against RNase treatment after immunostaining (Basic Protocol and Alternate Protocols 2 and 3), final addition of RNase to propidium iodide–stained samples is advantageous for increasing the resolution of the DNA measurement (Jensen et al., 1993a). The flow cytometric detection level for BrU incorporation is at ∼15 min incubation with 100 µM BrU (Jensen et al., 1993a). Application of BrU over extended periods of time (≥50 µM for ≥24 hr) may induce cell cycle perturbation and apoptotic cell death, as shown for HL-60 and MOLT-4 cells by Li et al. (1994). Steric hindrances due to masking of the incorporated BrU by proteins as a result of fixation with formaldehyde, or incorporation of the BrU into double-stranded RNA, may result in decreased immunofluorescence. Regarding Alternate Protocol 2, the use of citric acid buffer is expected to ensure a high retention of total nuclear RNA. The strong BrU staining of nucleoli (Fig. 7.12.1C) is an interesting feature of this method, and it is likely to represent nascent ribosomal RNA, thus indicating RNA polymerase I activity. The ABDM antibody from Partec, like most of the currently available anti-bromodeoxyuridine antibodies, binds to bromodeoxyuridine-substituted DNA only after its denaturation into single-stranded DNA. In this context, moderate acid treatment with citric acid at low ionic strength is thought to improve the access of the anti-bromodeoxyuridine antibody to BrU-substituted RNA, due to denaturation of double-stranded RNA in the nucleoli and removal of RNA-associated proteins. Comparison with green fluorescence standard microspheres generally indicates a lower background of unspecific FITC fluorescence for the nuclear suspensions (see Basic Protocol and Alternate Protocol 1) than for cell suspensions (see Alternate Protocols 2 and 3). The coefficient of variation of DNA histograms is also lower for the nuclear suspensions than for the cell suspensions.
Nucleic Acid Analysis
7.12.9 Current Protocols in Cytometry
Supplement 12
Figure 7.12.2 Flow cytometric dual-parameter dot plots of log FITC fluorescence (BrU-labeling of RNA) versus propidium iodide fluorescence (DNA content) from a pulse-chase experiment. Exponentially growing HL-60 cells were incubated with 1 mM BrU for 1 hr before replacement with BrU-free medium. Nuclear suspensions were prepared according to Alternate Protocol 1. (A to F) Cells were harvested at 30, 60, 90, 120, 150, and 270 min after addition of BrU. (G) Control sample not exposed for BrU. All samples were stained indirectly with anti-BrdU antibody (ABDM, Partec)
Anticipated Results
Cytometric Analysis of RNA Synthesis
Bivariate analysis of BrU incorporation and DNA content Dual-parameter analysis of FITC (BrU) and propidium iodide (DNA) fluorescence enables the correlation of transcriptional activity and cell cycle phase distribution. Figure 7.12.2 shows as an example HL-60 cells prepared according to Alternate Protocol 1. Nuclei with S-phase DNA content show a high level of BrU
incorporation, whereas those with G1 and G2/M-phase DNA content show a low or intermediate level. The horseshoe-shaped distribution suggests that the major part of RNA in this exponentially growing cell line is synthesized at the time of DNA replication. In pulse-chase experiments such as the one shown in Figure 7.12.2, the stability of the BrU-labeling intensity during the first hour after BrU removal suggests a balance between continued incorporation of BrU from intracel-
7.12.10 Supplement 12
Current Protocols in Cytometry
lular pools and processing degradation of labeled RNA. This processing of nascent RNA is expected to be inhibited by BrU incorporation, since splicing of RNA transcripts is inhibited in vitro in a BrU-dose-dependent manner (Wansink et al., 1994). However, it is unlikely that the processing of nascent RNA is completely blocked in such an experiment. A processing of the RNA must occur, since the intensity of the total BrU-specific fluorescence decreases after replacement with BrU-free culture medium, and FITC staining can be found in the cytoplasm (using Alternate Protocol 2). No gross perturbation of the cell cycle distribution during the pulse-chase experiment is indicated. Subcellular distribution of BrU-substituted RNA Fluorescence microscopy confirms that cell preparation according to Basic Protocol and Alternate Protocol 1 results in suspensions of pure nuclei without any remaining cytoplasm. Alternate Protocol 1 induces a swelling of the nuclei. The cytoplasm is morphologically fairly well conserved with Alternate Protocols 2 and 3. Figure 7.12.1 shows the subcellular distribution of BrU-substituted RNA, as revealed by fluorescence microscopy of HL-60 cells that have been labeled for 1 hr with 1 mM BrU. When nuclei are prepared according to Alternate Protocol 1, they show intense staining of the nucleoli as well as several smaller extranucleolar foci in a darker nuclear matrix (Fig. 7.12.1C). This staining pattern is similar to what is known from labeling with [3H]uridine and FITC-UTP. In contrast, nuclei prepared with the milder extraction according to the Basic Protocol show only extranucleolar staining (Fig. 7.12.1B). With Alternate Protocols 2 and 3 the FITC staining in the nucleus is similarly confined to the extranucleolar portion, and, in addition, the cytoplasm contains FITCstained foci.
Time Considerations The method may be performed stepwise, as labeled and fixed cells or nuclei may be stored until staining and flow cytometric analysis are convenient. Pulse-labeling with BrU and harvest of cells from cell cultures requires 1 to 3 hr. Cell lysis and fixation procedures require 1 hr for Basic Protocol and Alternate Protocols 1 and 2, and overnight fixation is required for Alternate Protocol 3. Staining requires 3 hr for Basic Protocol and Alternate Protocols 1 and 2, whereas for Alternate Protocol 3 it requires 1
hr before as well as 1 hr after the fixation procedure.
Literature Cited Carayon, P. and Bord, A. 1992. Identification of DNA-replicating lymphocyte subsets using a new method to label the bromodeoxyuridine incorporated into the DNA. J. Immunol. Methods 147:225-230. Dolbeare, F. 1995. Bromodeoxyuridine: A diagnostic tool in biology and medicine. Part 1: Historical perspectives, histochemical methods and cell kinetics. Histochem. J. 27:339-369. Haider, S.R., Juan, G., Traganos, F., and Darzynkiewicz, Z. 1997. Immunoseparation and immunodetection of nucleic acids labeled with halogenated nucleotides. Exp. Cell Res. 234:498506. Jackson, D.A., Iborra, F.J., Manders, E.M.M., and Cook, P.R. 1998. Numbers and organization of RNA polymerases, nascent transcripts, and transcription units in HeLa nuclei. Mol. Biol. Cell 9:1523-1536. Jensen, P.Ø., Larsen, J., Christiansen, J., and Larsen, J.K. 1993a. Flow cytometric measurement of RNA synthesis using bromouridine labelling and bromodeoxyuridine antibodies. Cytometry 14:455-458. Jensen, P.Ø., Larsen, J., and Larsen, J.K. 1993b. Flow cytometric measurement of RNA synthesis based on bromouridine labelling and combined with measurement of DNA content or cell surface antigen. Acta Oncol. 32:521-524. Landberg, G. and Roos, G. 1991. Antibodies to proliferating cell nuclear antigen (PCNA) as Sphase specific probes in flow cytometric cell cycle analysis. Cancer Res. 51:4570-4575. Li, X., Patel, R., Melamed, M.R., and Darzynkiewicz, Z. 1994.The cell cycle effects and induction of apoptosis by 5-bromouridine in cultures of human leukemic MOLT-4 and HL-60 cell lines and mitogen stimulated normal lymphocytes. Cell Prolif. 27:307-320. Murakami, T., Li, X., Gong, J., Bhatia, U., Traganos, F., and Darzynkiewicz, Z. 1995. Induction of apoptosis by 5-azacytidine: Drug concentrationdependent differences in cell cycle specificity. Cancer Res. 55:3093-3098. Otto, F. 1990. DAPI staining of fixed cells for highresolution flow cytometry of nuclear DNA. Methods Cell Biol. 33:105-110. Wansink, D.G., Nelissen, R.L.H., and de Jong, L. 1994. In vitro splicing of pre-mRNA containing bromouridine. Mol. Biol. Rep. 19:109-113.
Contributed by Jørgen K. Larsen, Peter Østrup Jensen, and Jacob Larsen The Finsen Laboratory, Rigshospitalet Copenhagen, Denmark Nucleic Acid Analysis
7.12.11 Current Protocols in Cytometry
Supplement 12
Sperm Chromatin Structure Assay for Fertility Assessment
UNIT 7.13
The integrity of mammalian sperm DNA is of prime importance for the paternal genetic contribution to normal offspring. Damaged DNA in the single sperm that fertilizes a female oocyte can have a dramatic negative impact on fetal development and health of the offspring throughout adult life. Animal and human fertility clinics typically assess semen quality by measuring sperm density, total count, motility, and morphology. Clinics rarely measure sperm DNA integrity, perhaps because they are unaware of the availability of, or lack the instrumentation for, a rapid, reliable, and practical test. The sperm chromatin structure assay (SCSA; see Basic Protocol) is one such test whose importance is becoming increasingly recognized. Whereas light microscope measurements are commonly made on 100 to 200 sperm per sample, the SCSA can utilize a fresh (or frozen-thawed) semen sample and collect data on 5000 or more cells in just a few minutes. Figure 7.13.1 shows two examples of a SCSA clinical report that can be generated and sent to a clinician by FAX or e-mail within 15 to 30 min of measurement. A protocol for SCSA is outlined in this unit (see Basic Protocol). Support Protocol 1 describes flow cytometer alignment and setup. Support Protocol 2 outlines sample collection and storage conditions so that all experimental data is related back to the standard; the need for a reference sample is highlighted. Support Protocol 3 details the very important analysis of SCSA data; means to generate the calculated parameter of αt [αt = red/(red + green)] both on and off line are discussed. Support Protocols 4 and 5 detail methods to obtain purified sperm nuclei that are needed for analysis of nuclear DNA (eliminating mitochondrial DNA) or nuclear -SH groups (eliminating the large number of -SH groups in the sperm tail). Sonication is sometimes used to verify that red fluorescence is derived from denatured DNA rather than residual cytoplasmic RNA. Rat sperm are always sonicated since the very large tails tend to plug the flow cell. Support Protocol 6 details fixation of sperm in ethanol. SCSA data on thousands of semen samples from humans (Evenson 1997, 1999a; Evenson et al., 1991, 1999; Fossa et al., 1997; Larson et al., 1999, 2000; Grajewski et al., 2000), bulls (Ballachey et al., 1987, 1988; Evenson, 1999b), stallions (Evenson et al., 2000b), boars (Evenson et al., 1994), and exotic cats (unpublished) show the clinical value of this assay for human/animal fertility assessment. SPERM CHROMATIN STRUCTURE ASSAY The SCSA measures the susceptibility of sperm chromatin to DNA denaturation in situ following low-pH treatment, which potentially induces this DNA denaturation. This assay is an adaptation of the “two-step AO” procedure originally designed by Darzynkiewicz et al. (1975) for simultaneous measurements of DNA and RNA content in somatic cells (UNIT 7.8). Whatever minute amounts of RNA may be present in mature sperm do not interfere with SCSA data. Many papers contain details about the critical nature of AO staining for DNA or differential DNA versus RNA measurements (Darzynkiewicz et al., 1975; Evenson and Darzynkiewicz, 1990; Evenson and Jost, 1994; UNIT 7.3 and UNIT 7.8). Of interest, this procedure denatures protamine-associated DNA in sperm but does not readily denature somatic-cell DNA associated with histones (Evenson et al., 1985).
BASIC PROTOCOL
Nucleic Acid Analysis Contributed by Donald Evenson and Lorna Jost Current Protocols in Cytometry (2000) 7.13.1-7.13.27 Copyright © 2000 by John Wiley & Sons, Inc.
7.13.1 Supplement 13
100
test: SCSA
High green
A
75 50
COMP
25 0 0
25
50
75
Total Fluorescence
Green Fluorescence
clinical report
125
B
100 75
75
50
50
25
25 0
25
50
2
75
100
0
0.5
Alpha t
Date
Measurement
Xαt
SDαt
COMPαt
% HGRN
A2
12/5/91
1 2
213.5 221.2
111.6 118.1
6.8 8.3
5.0 5.4
217.4 5.4
114.8 4.6
7.5 1.1
5.2 0.2
clinical report
100
test: SCSA
High green
D
pregnancy outcome: not pregnant
E
100
125
75
50
50
50
25
25
25
75
0
0 25
50
75
100
F
100
75
COMP
1.0
Alpha t
Patient
mean sd
0
1
0
0
100
C
100
Red Fluorescence
Green Fluorescence
pregnancy outcome: pregnant
1 2
0 0
25
50
75
100
0
0.5
Alpha t
Red Fluorescence Patient
Date
Measurement
Xαt
SDαt
COMPαt
% HGRN
D1
12/6/91
1 2
563.7 561.4
307.0 304.8
64.9 64.9
6.4 7.2
562.6 1.2
305.9 1.1
64.9 0.0
6.8 0.4
mean sd
1.0
Alpha t
Figure 7.13.1 Two standard clinical reports from the authors’ laboratory, showing SCSA data and calculated parameters of human sperm. The upper report (A-C) contains SCSA clinical data on a fertile donor whose wife conceived several days from the time of sample collection and is classical for a high-quality semen sample. The lower report (D-F) contains SCSA clinical data on the male partner of a couple that did not conceive over the course of one year.
Sperm Chromatin Structure Assay for Fertility Assessment
Red and green fluorescence from 5000 individual sperm per sample are collected through red (630-nm long-pass) and green (515 to 530-nm band-pass) filters. Computer software is used to calculate the ratio of red to total fluorescence, known as alpha t (αt; Darzynkiewicz et al., 1975; UNIT 7.8). For fertility assessment, the most important variables of the αt distribution have been found to be the mean (Xαt), the variation (SDαt), and the cells outside the main population (COMPαt, i.e., the percent of denatured sperm cells).
7.13.2 Supplement 13
Current Protocols in Cytometry
The AO staining technique was designed by Darzynkiewicz et al. (1975) to have a flow rate of ~200 cells/sec when the sample contains ~1 × 106 cells/ml and the AO staining solution contains 6 µg AO/ml. This protocol provides ~2 AO molecules/DNA phosphate group. The details in this protocol may give the impression that the AO-based SCSA is difficult, but once set up and put into practice, it is very simple to perform. Materials AO equilibration buffer (see recipe) Reference sample (see Support Protocol 2) Semen or caudal epididymal or testicular aspirate sample(s) 1× TNE buffer (see recipe) Acid detergent solution (see recipe) AO staining solution (see recipe) Ethanol/bleach tubing cleanser solution (see recipe) Household bleach tubing cleanser solution (see recipe) Flow cytometer (see Table 7.13.1 and Fig. 7.13.2) with 488-nm excitation appropriate filters for collection of green and red fluorescence, and ≥15 to 35 mW power, interfaced to a computer with appropriate software (Table 7.13.2) for calculating SCSA parameters Nonadjustable 200 µl pipettor and appropriate tips 0.20 to 0.80-ml and 0.80 to 3.0-ml Oxford adjustable dispensers with amber glass bottles (Fisher Scientific) Polystyrene 12 × 75–mm (4.5-ml) conical tubes (Sarstedt) Additional reagents and solutions for setting up and aligning the flow cytometer (Support Protocol 1), employing a reference sample (Support Protocol 2), counting sperm using a hemacytometer (APPENDIX 3A), and data analysis (Support Protocol 3) CAUTION: For personnel safety against potential infectious agents (e.g., hepatitis and HIV) handle human samples using disposable gloves in a biological safety cabinet. IMPORTANT NOTE: The SCSA procedure requires that samples be thawed and processed in the immediate vicinity of the flow cytometer. Human samples are thawed and prepared in a biological safety hood near the flow cytometer. All necessary equipment (e.g., 2 to 3 ice buckets containing wet ice with reagent bottles and TNE container deeply imbedded in the ice, sample tubes, disposable gloves, automatic pipetters and tips, stopwatch, 37°C water bath, marker) should be easily accessible in the safety hood or on a nearby laboratory cart.
Table 7.13.1 Flow Cytometers That Have Successfully Been Used To Run the SCSAa
Flow cytometer
Configuration
Company
Cytofluorograf Elite FACScan ICP22A Skatron
Orthogonal Orthogonal Orthogonal Epiillumination Epiillumination
Ortho Coulter Becton Dickinson Ortho Bio-Rad
aData taken from Evenson et al. (1995a), also see Fig. 7.13.2
Nucleic Acid Analysis
7.13.3 Current Protocols in Cytometry
Supplement 13
Green fluorescence
Human
Bull
A
D
B
E
C
F
Red fluorescence
Cytogram
Xαt
SDαt
COMPαt
Mean Red
Mean Green
A
252.0
144.0
13.0
128.0
387.0
B
241.0
153.0
11.0
123.0
381.0
C
260.0
125.0
11.0
134.0
393.0
D
230.0
74.0
26.0
143.0
488.0
E
235.0
91.0
25.0
120.0
397.0
F*
250.0
75.0
11.0
123.0
376.0
* Different bull sample
Figure 7.13.2 Comparative SCSA cytograms from different flow cytometers (Evenson et al., 1995a). Cytograms of human sperm generated on three orthogonal instruments: (A) Ortho Cytofluorograf, (B) Becton-Dickinson FACScan, and (C) Coulter Elite, and of bull semen measured on the (D) Ortho Cytofluorograf and two epiillumination instruments: (E) Ortho ICP22A and (F) Skatron. Data in F cannot be directly compared with those of D and E, as data in F are derived from a different bull.
Sperm Chromatin Structure Assay for Fertility Assessment
7.13.4 Supplement 13
Current Protocols in Cytometry
Table 7.13.2 PC Hardware/Software Options for Collecting and Analyzing SCSA Data
Hardware CICERO SYSTEMa
Supplier Cytomation
Software LISTVIEW WINLISTb FCS Express
Phoenix Flow Software Verity Software House De Novo Software
aThis packages allows for viewing the calculated α parameters in live time t
that is very helpful when setting up the instrument and checking/maintaining system stability and alignment. The Cicero System can be interfaced with all known commercial flow cytometers. bThe macro for analyzing SCSA data using WINLIST software is not provided by Verity House
NOTE: Although quick indications of the sperm with denatured DNA can be calculated from the red versus green fluorescence cytogram in “live-time” or on a stand-alone computer, the authors’ laboratory calculates the SCSA data after all the samples have been measured in an experiment. The actual αt values are taken from αt frequency histogram statistics. NOTE: For this protocol, all reagents, buffers, and samples must be kept on ice (i.e., 4°C). 1. Set up the flow cytometer with excitation at 488 nm and dichroic filters to collect red (≥630 nm) and green (515 to 530 nm) fluorescence in the peak (or height) mode (see Support Protocol 1). IMPORTANT NOTE: Do not use hardware configurations that define αt as red (>630 nm)/total (515 long-pass) fluorescence; this adds the unknown component of 530- to 630-nm data into the equation.
2. Place a tube of AO equilibration buffer in the sample chamber and allow ~1 to 2 ml (~15 min) of this buffer to flow through the system while preparing the sample. AO equilibration buffer is run through the instrument fluidic lines for ∼15 min prior to sample measurement to insure that AO is equilibrated with the sample tubing. It is also run through the instrument between different samples to maintain the necessary AO equilibrium conditions and help flush the prior sample out of the lines.
3. If running human semen samples, turn on the biological safety hood. CAUTION: Prepared samples are preferably run on a flow cytometer with a closed flow cell as opposed to a jet-in-air system (UNIT 1.2), although biological hazard containment is available for most systems. The HIV virus may be inactivated by the low-pH, acid-detergent treatment; this has not been tested in the authors’ laboratory. Always wear disposable gloves when working with potentially infectious samples.
4. Check and adjust the instrument settings using the reference sample (see Support Protocol 2). Utilizing a reference sample ensures that instrument fluidic and photomultiplier tube (PMT) settings should remain fairly consistent from day to day and throughout the measurement period, depending on slight alignment differences during setup between days and/or measurement periods.
5a. For fresh/raw semen samples: Place on ice in the biological safety hood and check concentration of the semen sample using a hemacytometer (APPENDIX 3A).
Nucleic Acid Analysis
7.13.5 Current Protocols in Cytometry
Supplement 13
If samples need to be measured for a quick clinical decision, estimate concentration and dilute using 1× TNE. Check the flow rate and if necessary, resample and restain with the proper dilution to attain the required flow rate of ≤300 cells per second.
5b. For frozen semen samples: Quickly immerse the sample tube in a 37°C water bath, just until the last remnant of ice disappears. Do not allow water into the vial. When analyzing a series of human samples, it is more efficient to obtain the sperm concentration prior to freezing the samples so that costly flow cytometer time is not wasted in determining the proper dilution for sample measurement (see Support Protocol 2). When a number of samples are to be analyzed by flow cytometry, embed the samples in dry ice in an ~18-in deep styrofoam box with a good cover (e.g., FreezSafe Insulated Containers; Curtin Matheson), and place it near the workbench. Remove, thaw, stain, and immediately measure the individual samples one by one.
6. Using adjustable pipettors, dilute an aliquot of the sample to 1–2 × 106 sperm/ml with 1× TNE buffer. 7. Using a nonadjustable 200-µl pipettor, place a 200-µl aliquot of the diluted sperm suspension in a sampling tube suitable for the flow cytometer. The dispenser must be accurate. Do not allow the sample to sit in the tube but proceed directly to the next step.
8. Add 400 µl acid detergent solution from the Oxford adjustable dispenser and start the stopwatch immediately. At the beginning of sample measurement and after a long time interval between measurements, dispense several volumes of acid detergent solution and AO staining solutions from the dispensers before starting with the samples; AO in the plastic delivery tube is sensitive to light and solutions in the delivery tubes may be warmer than 4°C. The Triton X-100 in the acid detergent solution punches holes in the cell membranes, allowing the AO dye molecules to enter the sperm nucleus and access the DNA. The low pH potentially denatures DNA in situ.
9. Gently mix by swirling in hand for 30 sec. Hold by the top of the tube so as not to warm the sample.
10. Add 1.20 ml AO staining solution from the Oxford adjustable dispenser to the sample at exactly 30 sec on the stopwatch. Acridine Orange (AO) stains double-stranded DNA with green fluorescence and singlestranded DNA with red fluorescence (UNIT 7.8).
11. Place the stained sample on the flow cytometer and start sample flow immediately. Check the sperm flow rate after ∼2 min on the instrument. If it is too rapid (i.e., >300 cells/sec) make a new sample at the appropriate dilution with 1× TNE. If the sample flow rate is too high, this same sample cannot be diluted with AO buffer to lower the concentration. The sample and sheath-flow valve settings of the instrument are never changed during these measurements, so the instrument flow rate is constant. Any change in sperm count rate becomes a function of sperm-cell concentration only. It is preferable to keep the sample in an ice bath, although the configuration of some instrument sample chambers may not permit this. For most samples this is not a problem; however, more dilute samples run longer and can denature if not on ice.
12. At 3 min on the stopwatch, begin data acquisition. Collect green and red fluorescence from 5000 cells/sample in listmode. Sperm Chromatin Structure Assay for Fertility Assessment
The 3 min allows time for AO equilibration in the sample and hydrodynamic stabilization of the sample within the flow stream.
7.13.6 Supplement 13
Current Protocols in Cytometry
13. Allow the first sample to continue running through the instrument while preparing a second one (steps 5 to 12) from the same thawed aliquot. Measure each aliquot in duplicate for statistical considerations. There is nearly always an exact cytogram dot pattern between the duplicate measurements (Evenson et al., 1991). The correlations of SCSA values between these samples are almost always near 0.99 (Evenson et al., 1999).
14. After measurement of the second sample, place a tube of AO equilibration buffer on the sample port of the instrument and then thaw, dilute, and stain the next sample for measurement. The AO equilibration buffer should run ~1 to 2 minutes. Running the buffer between measurements maintains AO equilibration conditions in the instrument sample lines and helps flush the previous sample out of the tubing. There is no need to run this buffer between the duplicate measurements of the same sample.
15. Run a freshly thawed and prepared reference sample after about every 5 samples (10 measurements). This insures stability of the instrument and quality control over measurements (Evenson and Jost, 1994). The authors’ instrument has been very stable over dozens of successive measurements during a single day of experimentation; however, green fluorescence tends to rise slightly throughout the day and a minor adjustment needs to be made to bring the reference signal back to the established value.
16. Shut down the instrument at the end of measurements. Once a week, thoroughly clean the tubing by running ethanol/bleach tubing cleanser solution for ∼15 min to insure the removal of all cell particles and AO adhering to the sample lines. If a different dye is to be used on the instrument next, and a thorough cleaning is not needed, it is appropriate to use the household bleach tubing cleanser solution. Backflush for 10 min after using either cleansing solution. This removes all AO from the system and has no negative effects on the instrument or on subsequent experiments using different fluorochromes.
17. Analyze samples according to SCSA parameters and constraints (see Support Protocol 3). FLOW CYTOMETER ALIGNMENT AND SETUP Alignment and setup are essentially the same for all flow cytometry studies requiring the best possible resolution, and are described in this protocol. Unique to the SCSA is the need to flush an AO solution through the instrument tubing before starting the sample measurements. Also, the flow rate of sample has to be adjusted to ~250 cells/sec with a cell concentration of ~1.5 × 106 cells/ml.
SUPPORT PROTOCOL 1
Additional Materials (also see Basic Protocol) Sheath fluid (see recipe) Standard fluorescent beads Sperm sample, 1–2 × 106/ml Reference sample (see Support Protocol 2) 1. Turn the flow cytometer on, check that sheath fluid levels are sufficient for the experiment (~2.5 liter/8 hr), and allow the specified warm up time for the instrument according to manufacturer’s instructions. 2. Check the instrument alignment using standard fluorescent beads. Nucleic Acid Analysis
7.13.7 Current Protocols in Cytometry
Supplement 13
3. Place a tube of AO equilibration buffer in the instrument sample chamber and let the buffer flow through the system ∼15 min prior to establishing instrument settings. This insures that AO is fully equilibrated with the sample tubing, as the dye transiently adheres to the sample tubing. This buffer should be running through the instrument during warm up, prior to alignment, and again just before measuring samples. After SCSA measurements, AO can easily be cleansed from the lines by rinsing the system ∼10 min with household bleach tubing cleanser solution followed by 10 min with filtered water. The authors have utilized many fluorescent dyes and sample types without any associated problems (UNIT 7.3) after measuring AO-stained sperm. Using AO in a flow cytometer does not ruin it for other purposes. The sample lines do not need to be replaced, although some researchers prefer to have a separate set of tubing for use with AO only.
4. Set the instrument for measuring green and red fluorescence signals processed and displayed as “peak” (or “height”) signals rather than area signals. 5. Set up the acquisition protocol to display the following (see Support Protocol 3 for details on gating): Two-parameter cytograms (1000 channels in both X and Y axes) of: Peak red fluorescence (x axis) versus peak green fluorescence (y axis) Alpha t (x-axis) versus total fluorescence (y-axis) One-parameter histograms (displaying 1000 channels) of: Red fluorescence Green fluorescence Total fluorescence αt IMPORTANT NOTE: Most SCSA figures have been published with the green fluorescence (FL1 on some instruments) data on the y axis and the red fluorescence (FL3) data on the x axis. The authors highly encourage the continuation of this orientation so that readers can make easy comparisons between data in manuscripts.
6. For initial SCSA measurements, set up the proper hydrodynamic conditions by measuring several sperm samples with a known cell count of ~1.5 × 106 sperm/ml as determined using a hemacytometer (APPENDIX 3A) and adjust the flow rate settings (if possible) to ~200 cells/sec. Record these conditions. A known concentration of fluorescent beads can also be used for this adjustment. The low flow rate setting on a FACScan delivers the correct flow rate.
7. Measure the reference sample by SCSA (see Basic Protocol, step 5b) to establish daily instrument settings. Record these settings each day. The reference samples are used to set the red and green photomultiplier tube (PMT) voltage gains to yield the same mean red and green fluorescence levels and αt values (Xαt, SDαt, and COMPαt) from day to day. The mean red and green fluorescence values are set at ~130/1000 and ~500/1000 channels, respectively, for all species (e.g., the values used for human semen in the authors’ laboratory are mean red and green fluorescence of 125 ± 5 and 475 ± 5 channels, respectively). The values established by a laboratory (preferably the same as above) should be used consistently thereafter. All reference-sample mean red and green fluorescence values should fall within ±5 channels of this established laboratory standard. The reference values should be strictly maintained in this range throughout the experiment. Sperm Chromatin Structure Assay for Fertility Assessment
8. Begin sample measurement (see Basic Protocol, step 5).
7.13.8 Supplement 13
Current Protocols in Cytometry
IDENTIFYING, COLLECTING, AND FREEZING A REFERENCE SAMPLE Very few flow cytometery protocols are as demanding as the SCSA in their requirement for the use of a reference sample. Because SCSA variables are very sensitive to small changes in chromatin structure, studies on sperm using this protocol require very precise, reproducible instrument settings for all comparative measurements, whether done at the same or different times. These settings are established by using aliquots of a single semen sample called the reference sample. These aliquots are used only once and are not to be saved for use as little as an hour later, even if kept at 4°C. The authors do not refer to this as a “control sample” since that terminology implies normal fertility and the reference sample may or may not be of normal fertility.
SUPPORT PROTOCOL 2
The following procedures for collecting, preparing, and freezing a reference sample apply to all semen samples. Additional Materials (also see Basic Protocol) Semen donor Clinical specimen jars, polystyrene, sterile (VWR Scientific) 0.5- to 1.5-ml polypropylene microcentrifuge tubes (Sarstedt) 1.2-ml cryogenic vials (Fisher) NOTE: All studies with human subjects must be approved by the Institutional Review Board (IRB), which must adhere to the Office for the Protection from Research Risk (OPRR) guidelines or other applicable governmental regulations for using human subjects. NOTE: It is important to work quickly and efficiently with samples for freezing. NOTE: Quick-frozen and cryoprotectant-frozen sperm provide equivalent SCSA data (Evenson et al., 1994), which is a unique feature of mammalian sperm cells due to the highly condensed crystalline nature of the nuclei. 1. Identify a donor whose semen sample demonstrates heterogeneity of αt (e.g., 15% to 20% COMPαt). Before running a major experiment and after initially measuring several samples and optimizing the SCSA protocol for one’s laboratory and instrument, a reference sample must be located to insure day-to-day standardization of the protocol.
2. Using standard laboratory procedures and practices and in compliance with the rules and regulations of the institution’s Review Board, have the donor provide a sample. Human semen samples are typically obtained by masturbation into plastic clinical specimen jars, preferably after 2 to 3 days’ abstinence.
3. Allow 30 min for semen liquefaction at room temperature. Freshly ejaculated semen consists of a large amount of a gel-like substance within which the sperm are lodged. Normally this material “liquefies” ∼30 min after ejaculation and frees the sperm, thereby providing a homogenous suspension of sperm that insures a random sampling of all sperm that are in the sample.
4. Place the sample on ice. Semen can be kept for up to 5 hr at room temperature prior to measuring/freezing without loss of quality, allowing for collection within a medical institution and transportation to the andrology unit; however, this time should be minimized as much as possible. If transport is required outside a building complex, the sample may be conveyed in an insulated box such as a FreezSafe Insulated Container, Polyfoam Packer (Curtin Matheson), or jacket pocket to keep it from freezing, or on liquid ice if the ambient temperature is high.
Nucleic Acid Analysis
7.13.9 Current Protocols in Cytometry
Supplement 13
5. Determine sperm concentration using a hemacytometer (APPENDIX 3A). It is important to work quickly and efficiently with samples for freezing
6. Dilute the sample with 1× TNE buffer, 4°C, to a working concentration of 1–2 ×106 cells/ml. Once a sample has been diluted in TNE buffer, it is recommended that it immediately be either frozen or measured by the SCSA.
7. Quickly (within 15 min) divide the diluted sample into several hundred ~300-ml aliquots in 0.5- to 1.5-ml polypropylene microcentrifuge tubes and immediately place in a −70° to −110°C freezer or aliquot into 1.2-ml cryogenic vials for storage in a liquid nitrogen tank. Samples should be frozen in vials that are ∼ 1⁄4 larger than the semen volume in order to reduce the air-to-surface interface and thereby minimize reactive oxygen–related damage. Keep the tubes vertical when freezing, as samples frozen at the bottom of a tube can be easily and safely thawed in a water bath. Raw semen can be frozen directly without cryoprotectants in an ultracold freezer (−70° to −110°C) or liquid nitrogen tank.
8. Measure at least one reference aliquot in duplicate by SCSA (see Basic Protocol, step 5b) each day at the beginning of sample measurement, and at least once following every five samples thereafter. 9. When the supply of reference samples runs low, prepare a new batch of reference samples, preferably from the same individual donor. Consistency between reference samples and measurements made over time can be maintained by enrolling a new donor and comparing his sample with that of the previous reference donor.
ACQUISITION AND ANALYSIS PROTOCOL OF SCSA LISTMODE MEASUREMENTS Figures 7.13.1 (actual data) and 7.13.3 (schematic representation) illustrate cytograms, histograms, gates, and regions necessary for acquiring and analyzing SCSA listmode data and are referred to extensively in this support protocol. The arrows in Figure 7.13.3 indicate which signals are gated into other cytograms and histograms. (For an overview of gates, see UNIT 1.8.)
A
B Total fluorescence
Green fluorescence
SUPPORT PROTOCOL 3
region 1
region 1
region 2
region 1
region 1
Alpha t
Red fluorescence
C
G
D region 1
E
F region 1
region 1
region 2 Green fluorescence
Sperm Chromatin Structure Assay for Fertility Assessment
7.13.10 Supplement 13
Red fluorescence
Alpha t
Total fluorescence
Figure 7.13.3 Schematic of cytograms, histograms, and regions (both gating and statistical) that are needed for the SCSA. (A-F) illustrate the standard SCSA flow cytometric protocol constraints and (G) shows the regions for calculating percent high green fluorescence (%HIGRN), where region 2 is a subset of region 1. Current Protocols in Cytometry
All sperm-cell fluorescence signals of the red versus green fluorescence cytogram, excluding cellular debris signals (i.e., region 1 of Fig. 7.13.3, panel A), are gated into an αt versus total fluorescence cytogram, and into green and red fluorescence histograms. All sperm-cell fluorescence signals of the total fluorescence versus αt cytogram (i.e., region 1 of Fig. 7.13.3, panel B), are gated into the αt and total fluorescence histograms. A region enclosing the entire histogram is set in each of the four gated histograms (green, red, αt, and total) and the mean and standard deviation of each histogram are recorded from these regions (i.e., includes mean and SD of red, green, total fluorescence, and αt). Besides providing the Xαt and SDαt, the αt histogram is also used for calculating COMPαt. A second region (region 2) is set in the αt histogram that starts at the right hand side of the main peak of αt and goes to the upper channel. The reference sample main population serves as the defining population for the experiment as to where this region 2 is set. The two SCSA parameters of greatest practical value for a clinician are the percentage of sperm with denatured DNA (COMPαt) and the percentage of sperm with increased DNA stainability (% HIGRN). % HIGRN is a ratio and is the only data taken from a 2 parameter cytogram (Fig. 7.13.3, panel G). In toxicology studies (Evenson et al., 1993a,b,c), region 2 of the αt histogram (i.e., defines COMPαt; Fig. 7.13.3, panel E) was defined by setting the region in the control samples and all cells to the right of that line were defined as COMPαt cells. Therefore, in some cases, where the entire population shifts (Evenson et al., 1993b) COMPαt cannot be defined as “the cells to the right of the main population,” but is actually to the right of the main, normal population as defined in the control samples. At times COMPαt may equal 100% when the entire population shifts. The standard deviation of αt (SDαt) is important for toxicology studies. The mean channel of the αt population (Xαt) has been found useful in some studies. Figure 7.13.4 diagrams some of the most common SCSA populations that will be encountered. For extensive discussion of this figure, see Anticipated Results. Additional Materials (also see Basic Protocol 1) Offline software capable of generating calculated parameters (see Table 7.13.2) 1. Generate the necessary, calculated parameters of alpha t (αt) in the software package according to manufacturer’s instructions. For most packages, total fluorescence = 0.5 (red + green) fluorescence and αt = 0.5 red/total fluorescence.
2. Set up the analysis protocol to mimic the acquisition protocol and display the following (see Fig. 7.13.3): Two-parameter cytograms (1000 display channels in both axes) of: a. Peak red versus peak green fluorescence b. αt versus total fluorescence (gated from peak red versus peak green fluorescence). One-parameter histograms (1000 display channels) of: c. αt d. Green fluorescence e. Red fluorescence f. Total fluorescence.
Nucleic Acid Analysis
7.13.11 Current Protocols in Cytometry
Supplement 13
Gating region
5 6
Green fluorescence (peak or height)
4
1
2
A
B
3
6
7
C
D
Red fluorescence (peak or height) Figure 7.13.4 Cartoon illustrations of normal fluorescence populations and common variations when sperm are measured by the SCSA. See Anticipated Results for extensive discussion. (A) Classical SCSA cytogram. The numbered areas are explained in Anticipated Results. (B) Panel B is the same as panel A except that area (6) represents a semen sample with excessive bacterial contamination.(C) Panel C is also the same as A except that area (7) is an approximation of a sperm sample that has been compromised by excessive freeze/thaw cycles or left in a thawed state for an extended time period.(D) This panel is a composite of panels A to C.
3. Draw a statistical region on the peak red versus peak green fluorescence cytogram to be used as a gate including all sperm signals, but excluding cellular debris signals (i.e., signals located at the origin in this cytogram) from the analysis (this region corresponds to region 1 of Fig. 7.13.3, panel A). Draw the region very near the perimeters of the cytogram boundaries, but exclude those events that are beyond the full channel limits (i.e., >100% of scale). Complete the region by drawing a descending ∼45° line just below the bottom of the sperm signal as in Fig. 7.13.2. A descending ∼45° line is appropriate only for samples having very little debris falling near the origin. Human SCSA data are often more complicated, and in some cases elliptical curves (Fig. 7.13.1, panels A and D) have been used to exclude the debris signal from the data (Evenson et al., 1991). After inspection of the whole sample set, a decision is made on what best fits the experiment. Whatever method is chosen, it is important to be consistent with analysis configurations, and preferably between previous, current, and future experiments. Sperm Chromatin Structure Assay for Fertility Assessment
NOTE: Do not include signals built up in the highest green fluorescence channel (Fig. 7.13.1).
7.13.12 Supplement 13
Current Protocols in Cytometry
4. Gate this region of peak green versus peak red fluorescence to the total fluorescence versus αt cytogram, and to histograms of green fluorescence and red fluorescence. This conversion allows a better visual interpretation of the different types of cells present as can be seen in Figure 7.13.1, panels B and E.
5. Draw a gating region in the total fluorescence versus αt cytogram, which sends signals to the αt and total fluorescence histograms. 6. Draw a region in all four histograms covering channels 1 to 1000, inclusive (i.e., Fig. 7.13.3, region 1 of C, D, E, and F). 7. Place a second region in the αt histogram that denotes the denatured sperm cells (COMPαt), using the reference sample as a guide to where the main population leaves off (see Support Protocol 2). This region corresponds to region 2 of Fig. 7.13.3, panel E. COMPαt is the most important variable of this assay for fertility assessment.
8. Gather the following statistics into a data set for analysis: a. Mean and standard deviation from the region covering channels 1 to 1000, inclusive, from all four histograms. b. % COMPαt from the region in the αt histogram that denotes the denatured sperm cells. 9. Draw a region in the peak red versus peak green fluorescence cytogram similar to the one used in step 3, except this time include the upper green fluorescence channel data (Fig. 7.13.3, panel G shaded areas). % HIGRN is calculated separately, and different regions must be set in the red versus green cytogram.
10. Draw a second region that contains everything above the main population, including the upper green fluorescence channel. % HIGRN equals the number of cells in region 2/number of cells in region 1 × 100. NOTE: The HIGRN parameter is not as precise as the other SCSA parameters, since measurement is potentially compromised by cell doublets. However, doublets generally comprise a very small percent of the total population. Interestingly, inclusion of doublets in the αt parameters is not a problem since αt is a ratio measurement.
SONICATION OF SPERM CELLS FOR SCSA OF SPERM NUCLEI SCSA data from sonicated versus unsonicated semen have been essentially identical (Evenson et al., 1991). Sonication has been used to verify that potential RNA in cytoplasmic droplets is not contributing to the red fluorescence signal of denatured, single-stranded DNA. Sonication is preferred over RNase incubation, which has the potential of introducing incubation-related changes in chromatin structure from protease digestion of chromatin proteins (Evenson et al., 1985). Sonication is rarely used anymore because it has been shown repeatedly that there are no significant differences between whole cells and their sonicated counterparts when measured by SCSA. Figure 7.13.5 provides an illustration of exactly how the sonication apparatus should be designed.
SUPPORT PROTOCOL 4
Nucleic Acid Analysis
7.13.13 Current Protocols in Cytometry
Supplement 13
rubber stopper cryovial with sample cup-horn sonicator
peristaltic pump sonifier
bucket filled with wet ice
copper coil immersed in wet ice
Figure 7.13.5 Schematic of a cup-horn sonicator arrangement used in sonicating human semen samples (Support Protocol 5). A peristaltic pump is used to drive water at a flow rate of ~21 ml/min through ∼3 feet of copper tube coil (1/4-in. i.d.) set in a 4-liter flask filled with an ice-water slurry. The cup horn sonicator is preferred for ease of use, uniformity between samples, and safety considerations.
Additional Materials (also see Basic Protocol) 1–2 × 106 sperm cells in 1× TNE buffer 2-ml screw-cap cryogenic vial (Corning) No. 11 rubber stopper with 12-mm hole Branson Sonifier II, Model 450, coupled to a Branson Cup Horn (VWR Scientific) and linked to a Masterflex peristaltic pump (Cole-Parmer Instrument) or equivalent CAUTION: Because human semen samples potentially contain infectious agents such as hepatitis or HIV, sonication must be done only in a closed tube using a cup horn sonicator. IMPORTANT NOTE: Tail/cytoplasm separation varies between species and needs to be tested for each sonicator to achieve a ≥95% head/tail separation. Using a microscope, check the separation by scoring >100 sperm cells on 3 different slides (for each time/setting tested). Test the sonication unit (probe or cup horn) using several different semen samples to determine optimal time and power required for ≥95% head/tail separation. Because of the size of the tails on rat sperm used in toxicology studies, these samples must always be sonicated or the whole sperm will clog the sample filter. 1. Place 0.5 ml 1–2 × 106 sperm cells diluted in 1× TNE buffer in a 2-ml screw-cap cryogenic vial. 2. Insert the top end of this capped vial into the bottom of a no. 11 rubber stopper which has a 12-mm hole drilled into it that will hold the vial securely.
Sperm Chromatin Structure Assay for Fertility Assessment
3. Place the rubber stopper with the vial on top of the cup horn so that the vial protrudes down into the cup horn and its bottom is just off the base of the cup.
7.13.14 Supplement 13
Current Protocols in Cytometry
4. Sonicate human samples for 40 sec using 70% of 1-sec cycle bursts at a setting of 3.0 output power. Keep the sample cold by flowing water at 4°C through the cup horn (Fig. 7.13.5). A peristaltic pump is used to drive water (21 ml/min) through ∼3 ft of copper tube coil (1⁄4 in id) set in a 4-liter flask filled with an ice-water slurry. The cup horn sonicator is preferred for ease of use, uniformity between samples, and safety precautions.
PREPARING CYTOPLASM-FREE SPERM NUCLEI Cytoplasm-free sperm nuclei are prepared to verify that the red fluorescence signal is not derived from cytoplasmic droplets when seen in a light microscope (Evenson et al., 1985). Purified nuclei are also used for other studies related to sperm chromatin structure (Evenson et al., 1989a, 2000b).
SUPPORT PROTOCOL 5
Additional Materials (also see Basic Protocol) 60% sucrose solution, pH 7.5 (see recipe) 1× TNE buffer (see recipe) Beckman T-J6 benchtop clinical ultracentrifuge or equivalent Additional reagents and equipment for sonication of sperm cells (Support Protocol 4) 1. Sonicate fresh or frozen-thawed sperm (see Support Protocol 4). 2. Mix 800 µl sonicate with 200 µl 60% sucrose solution, pH 7.5. 3. Carefully, layer this over 900 µl 60% sucrose solution. 4. Centrifuge 7 min at 214,000 × g, 4°C. 5. Discard supernatant and suspend pellet in 100 to 500 µl 1× TNE buffer 4°C. The amount of TNE required for resuspension depends on the original sample concentration and the estimated loss of cells from centrifugation.
ETHANOL FIXATION OF SAMPLES SCSA data are best when using fresh or frozen samples. In cases where this is not possible or impractical, this protocol details fixation of sperm in ethanol. Data from ethanol-fixed sperm are similar to but never quite as good as those obtained on fresh/frozen-thawed sperm material (Evenson et al., 1986a).
SUPPORT PROTOCOL 6
Additional Materials (also see Basic Protocol) Semen sample Hank’s balanced salt solution (HBSS; Life Technologies, also see APPENDIX 2A) 80% ethanol, −20°C 1. Microcentrifuge sample 10 min at 300 × g, 4°C, to pellet sperm out of semen. 2. Resuspend sperm in Hank’s balanced salt solution (HBSS) to ∼107 sperm/ml. 3. Forcefully pipet 1 ml of this suspension into 10 ml 80% ethanol at −20°C in tubes with good screw caps. Cap the samples and wrap the cap with Parafilm to insure a tight seal. 4. Store samples at −20°C until needed. 5. Prior to analysis, centrifuge the sperm again as in step 1, wash once in 1× TNE buffer, and then process for SCSA, always keeping the sample at 4°C.
Nucleic Acid Analysis
7.13.15 Current Protocols in Cytometry
Supplement 13
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Acid detergent solution 20 ml 2.0 N HCl (0.08 N final; Sigma) 4.39 g NaCl (0.15 M final) 0.5 ml Triton X-100 (0.1% final) H2O to 500 ml Adjust pH to 1.2 with 5 N HCl (Ricca Chemical) Store up to several months at 4°C IMPORTANT NOTE: Use purchased standardized 2.0 N HCl (Sigma). Do not prepare from a more concentrated HCl solution that is likely less pure and may be of questionable strength. Triton X-100 is very viscous. The authors use a wide mouth pipet to carefully measure the exact amount. Wipe the outside of the pipet free of Triton X-100, and then expel with force. Rinse remaining Triton X-100 from the pipet by drawing solution in and out until it is all dispensed.
Acridine orange (AO) equilibration buffer 400 µl acid detergent solution (see recipe) 1.20 ml AO staining solution (see recipe) Keep on ice (4°C) in between use. Do not store more than a couple of hours.
Acridine orange (AO) staining solution Add 600 µl AO stock solution to each 100 ml of staining buffer (see recipe). Rinse the pipet tip several times in the staining buffer. Store up to 2 weeks in an amber glass container at 4°C. Acridine orange (AO) stock solution, 1 mg/ml Tare a 15-ml, flat-bottom scintillation vial on an electronic balance. Using a microspatula, carefully remove and transfer 3 to 6 mg chromatographically purified AO powder (Polysciences) from the stock bottle into the vial. Add an equivalent number of ml of water (i.e., 1 ml water/mg AO powder). Wrap the capped vial in aluminum foil to protect from light. Store up to several months at 4°C CAUTION: AO is a toxic chemical and precautions should be taken when handling it. Flush away any remnants of AO adhering to the spatula under running tap water, dispose of as dictated by each individual institution’s biosafety board. IMPORTANT NOTE: The authors’ laboratory has used AO obtained exclusively from Polysciences and thus have full confidence in this source. Do not use a more crude preparation of AO; failure will result.
Ethanol/bleach tubing cleanser solution 50 ml ethanol 50 ml household bleach (contains ∼5% sodium hypochlorite) 2.92g NaCl Filter through a 0.45-µm filter Store up to 2 months at room temperature This solution is used for a thorough cleaning or flushing of flow cytometer sample lines, including removal of AO. Use at least 1 time per week.
Sperm Chromatin Structure Assay for Fertility Assessment
7.13.16 Supplement 13
Current Protocols in Cytometry
Household bleach tubing cleanser solution 50 ml H2O 50 ml household bleach (contains ∼5% sodium hypochlorite) Filter through a 0.45-µm filter Store at room temperature This solution is used to remove adherent AO from flow cytometer fluidic components in contact with AO, especially when changing experimental dyes or probes.
Na2 PO4 buffer, 0.2 M 28.4 g sodium phosphate, dibasic H2O to 1 liter Store up to several months at 4°C NOTE: When 0.2 M Na2PO4 buffer is removed from the refrigerator, salt crystals will be present. Heat in 37°C water bath until the salts are fully dissolved.
Sheath fluid 0.5 ml Triton X-100 (0.05% v/v final) 1 liter H2O passed through a 0.22-µm filter Triton X-100 minimizes surface tension in the flow cytometer fluidic lines, thereby reducing bubble retention.
Staining buffer, pH 6.0 370 ml 0.1 M citric acid buffer (21.01 g citric acid monohydrate in 1 liter H2O; store up to several months at 4°C) 630 ml 0.2 M Na2PO4 buffer (see recipe) 372 mg disodium EDTA (formula wt., 372.24; 1 mM final) 8.77 g NaCl (0.15 M final) Mix overnight on a stir plate to insure that the EDTA is entirely in solution Adjust pH to 6.0 with concentrated NaOH pellets Store up to several months at 4°C IMPORTANT NOTE: Slowly and carefully adjust the pH using very small pieces of concentrated NaOH pellets (cut with a scalpel and handled with forceps).
Sucrose buffer, 10×, pH 7.5 (for isolated and purified nuclei) 9.48 g Tris⋅Cl (0.1 M final) 4.467 g disodium EDTA (20 mM final) H2O to 600 ml pH to 7.5 with 2 N NaOH Store up to 1 year at 4°C To make 1× buffer (0.01 M Tris⋅Cl, 2 mM EDTA), dilute 1:10 in H2O and check pH Sucrose solution, 60% (w/w), pH 7.5 240 ml 1× sucrose buffer, pH 7.5 (see recipe) 360 g sucrose, ultrapure Adjust pH to 7.5 with 2 N NaOH Store up to 1 year at 4°C This solution has to be stirred for hours, usually overnight, to get all the sucrose into solution.
Nucleic Acid Analysis
7.13.17 Current Protocols in Cytometry
Supplement 14
TNE buffer, 10×, pH 7.4 9.48 g Tris⋅Cl (0.1 M final) 52.6 g NaCl (1.50 M final) 2.23 g disodium EDTA (10 mM final) H2O to 600 ml Adjust pH to 7.4 with 2 N NaOH Store up to 1 year at 4°C To make 1× buffer (0.01 M Tris⋅Cl, 0.15 NaCl, 1 mM EDTA), dilute 1:10 in H2O and check pH COMMENTARY Background Information In recent years the sperm chromatin structure assay (SCSA) has been presented by various symposium speakers at fertility and toxicology congresses as a promising assay for use in animal and human reproductive clinics. The recent advent of the intracytoplasmic sperm injection (ICSI) procedure for human fertilization has focused increased attention on the quality of the paternal genome (Evenson, 1997, 1999a). ICSI bypasses virtually all Mother Nature’s checks for insuring that a high-quality sperm reaches the oocyte in vivo. With ICSI, an individual sperm with the best looking morphology and motility, as seen and judged by a technician using light microscopy, is picked up with a micromanipulator and injected into an oocyte. If fertilization occurs, the ICSI procedure is considered successful. However, the only real consideration for success by the paternal contribution should be whether or not the integrity of the paternal genome is sufficient for proper embryo development, resulting in the birth of a healthy child.
From the authors’ experience with sperm DNA integrity assays, including dUTP-biotin nick end-labeling (TUNEL; Martin and Lenardo, 2000), single-cell microgel electrophoresis assay (COMET; Sutherland and Costa, 1999), and nick translation (Gorzcyca et al., 1993; Sailer et al., 1995a; Aravindan et al., 1997), the SCSA is the most practical and statistically significant for assessing DNA integrity in a semen sample. COMPαt (cells outside the main population, i.e., percentage of sperm with denatured DNA) thresholds of ~0 to 15%, 16% to 29%, and ≥30% relate to high, moderate, and very low human fertility potential respectively (see Table 7.13.3). Data taken from Evenson et al. (1999) showed that of 144 couples, 73 became pregnant during the first three months. SCSA data from these men served as the “gold standard” for semen with high fertility potential. Note that the semen from the couples that became pregnant during months 4 to 12 had significantly (p < 0.01) poorer semen quality. Semen from those couples that did not achieve pregnancy after 12 months on the study were of even poorer quality
Table 7.13.3 Least Squares Means and Standard Errors (SE) by Pregnancy Outcome Group for Selected SCSA Parametersa
Pregnancy outcome group
Sperm Chromatin Structure Assay for Fertility Assessment
7.13.18 Supplement 14
Pregnancy in 3 months SE Pregnancy in 4-12 months SE % increase no pregnancy SE % increase
n
Xαt
SDαt
COMPαt
73
234.6
137.9
11.2
40
4.6 255.1b
4.6 157.9b
0.85 15.5b
5.9 8.7 270.3c 6.7 15.2
6 14.7 173.7c 6.8 30.0
1.09 38.3 17.2c 1.24 53.6
31
aData taken from Evenson et al. (1999). bp < 0.01. cp < 0.001.
Current Protocols in Cytometry
(p < 0.001). Notice the percent of increase in SCSA parameters for the 4 to 12 month and no pregnancy by 12 month categories. No couple achieved pregnancy if the COMPαt was ≥30%. These human SCSA data, along with those of bull, stallion, and other mammalian sperm, have established a threshold that is predicative of sub/infertility; if ≥30% of the sperm in a sample have SCSA-detected DNA denaturation (COMPαt), fertilization attempts by in vivo or in vitro methods are likely to fail (Evenson et al., 1980, 1999). Unusual cases of sterility range up to as high as 80 to 90% COMPαt. There is evidence for a correlation between the susceptibility of sperm-cell DNA denaturation and the presence of nuclear DNA strand breaks (Gorzcyca et al., 1993; Sailer et al., 1995a; Aravindan et al., 1997). The origin of DNA strand breaks has sometimes been attributed to an aborted type of sperm-cell apoptosis (Gorzcyca et al., 1993). These same threshold values seem to apply to assisted reproductive techniques (ART) samples. Even when sperm are purified by either swim-up techniques to increase percent motility or density gradients to remove dead cells, when semen samples contain ≥27% sperm with denatured DNA, no pregnancies occurred with either in vitro fertilization (IVF) or ICSI techniques (Larson et al., 2000). If ~30% of the sperm are determined to be abnormal by the SCSA, then why do the other 70% not contribute to a normal fertilization and pregnancy? The authors suggest a “tip of the iceberg” phenomenon, i.e., that the physical/chemical stresses placed on the sperm are sufficient to draw out 30% of the sperm as a discrete, abnormal population. The same type of damage likely exists throughout the whole sperm population to a lesser extent, but sufficient to cause subfertility. A couple attending an ART fertility clinic could benefit medically, emotionally, and financially by having an aliquot of the man’s semen analyzed by SCSA prior to any proposed IVF/ICSI procedure before making decision on whether to proceed with the clinical work. A young male patient just diagnosed with cancer could use SCSA data to decide whether to freeze and store his semen for future use. The chromatin quality of sperm in cancer patients prior to treatment may range from excellent to very poor (Evenson and Melamed, 1983; Evenson et al., 1984; Fossa et al., 1997), and those at the poor-quality level should seriously consider the possible futility and unnecessary expense of storing the semen.
Although the emphasis in this unit is on SCSA data as related to fertility potential, the SCSA is also an excellent technique for assessing toxicant-induced sperm chromatin damage (Evenson et al., 1989b, c, d, 1993a, b, c, 1995b; Evenson and Jost, 1993), especially in well controlled rodent studies. The SCSA parameter SDαt (variation) was the most sensitive parameter for detecting abnormal chromatin in mouse epididymal sperm 40 days after the testes were exposed to low level X-irradiation, when the sperm cells were at the stem cell stage (Sailer et al., 1995b). The assay can detect within hours molecular-level defects that may not be observed until 1 to 3 weeks later by light microscopy as metaphase chromosome breaks or embryonically as dead fetuses (Estop et al., 1993; Evenson et al., 1993b). A man’s age appears to have some influence on SCSA data (Spanö et al., 1998) as do long periods of abstinence. Very short abstinence times (e.g., several hours), do not affect SCSA data significantly (Evenson et al., 1991). Limited evidence suggests that long-term environmental pollution may have a negative effect. Few SCSA toxicology studies have been done on human sperm, but data have been presented showing that exposure to environmental stresses and pollutants, as well as cigarette smoking, may alter chromatin structure (Wyrobek et al., 1997; Rubes et al., 1998; Lemasters et al., 1999; Potts et al., 1999; Evenson et al., 2000a; Grajewski et al., 2000; Selevan et al., 2000; Vine et al., 2000).
Critical Parameters and Troubleshooting Cytometer brand The highly condensed sperm nucleus has a much higher index of refraction than the sample sheath (water) in a flow cytometer. This differential, coupled with the typical nonspherical shape of sperm nuclei and their orientation in the flow channel, produces an optical artifact consisting of an asymmetric, bimodal emission of DNA dye fluorescence (Gledhill et al., 1976) when measured in flow cytometers with orthogonal configuration, which have 90° angles between laser light beam, axis of flow, and axis of fluorescence collection. Because αt is a computer-calculated ratio, this optical artifact of DNA-stained sperm does not significantly interfere with SCSA data (Evenson and Jost, 1994). Although each orthogonal flow cytometer with different configurations of lens and fluidics produces different cytogram patterns,
Nucleic Acid Analysis
7.13.19 Current Protocols in Cytometry
Supplement 13
Sperm Chromatin Structure Assay for Fertility Assessment
αt data are essentially the same (Evenson et al., 1995a) as shown in Fig. 7.13.2. On the Ortho ICP22A (Fig. 7.13.2, panel E), the cells are confined in the sheath and measured as they emerge from a tube oriented along the axis of a fluorescent microscope objective and make a 90° turn into a flowing stream (Shapiro, 1985, 1988). Cells on the Skatron (Fig. 7.13.2, panel F) are confined in sheath, being ejected from a nozzle onto a slide surface; this arrangement is called a JOOS, or “jet on open surface,” flow chamber. As the cells flow along this surface the signal is observed using a fluorescence microscope objective (Shapiro, 1985, 1988). Epiillumination instruments use the same optical lens to both focus the excitation light and collect the fluorescence of the sperm nuclei as they flow in the optical axis of the instrument. The long axes of the nuclei are oriented in the direction of the flow by hydrodynamic forces. Orientation of the plane of each nucleus in the flow stream and the end of the nucleus that leads into the flow channel are random. Because flow direction and the optical axis are parallel, emitted fluorescence signals are insensitive to the rotational orientation of the sperm and no artifact is produced. The Partec flow cytometer is a currently available epiillumination instrument which has not been used with the SCSA, but data should resemble that from the Ortho ICP22A (Fig. 7.13.2, panel E). Fig. 7.13.2 shows SCSA data collected on several different cytometer brands. Notice that all statistical data are essentially equivalent. There are obvious differences in COMPαt appearance between cytograms from the Ortho Cytofluorograph, Becton Dickinson FACScan, and the Coulter Elite because more dots are shown on the Coulter Elite cytogram due to computer display options. In an Ortho Cytofluorograf, three events must occur within the same coordinate before being recorded as a “visible” dot, while the Coulter Elite records a dot with only 1 or 2 events per coordinate. Even with these different systems the COMPαt values are 13.0, 11.0, and 11.0, respectively, for measurements A, B, and C. Comparing cytograms of bull semen, there are dramatic appearance differences between the (D) Ortho Cytofluorograf (orthogonal) and (E) Ortho ICP22A (epiillumination) that removes the “artifact” seen in D (refer to Table 7.13.1), but COMPαt values are exactly the same (26.0%). The other parameters are not expected to be the same due to the different types of signals. The epiillumination configu-
ration of the Skatron (F) results in cytograms with different patterns because the optics and fluidics differ considerably. Sampling order If the intent of an experiment is to determine the smallest amount of change measurable over time or toxicant dosages, then ideally one should measure all experimental samples of a particular set in the chronological order of sample collection, or, if in a dose-response toxicology study, start with the sample at the lowest dose and move in sequence to the highest dose. Statisticians prefer randomly measuring coded samples. In a dose-response study, low doses usually have few altered cells while high doses may cause up to 100% altered cells. There is always the possibility of some abnormal cells from a high-dose sample sticking to the sample tubing, being released with time, and being measured with a low-dose sample. Recognizing that a flow cytometer randomly and objectively measures 5000 sperm from each sample, it is recommended that a series or set of samples be measured in a single time frame. If totally random measurements are necessary, excellent results can be obtained with careful use of the reference samples for repeat instrument settings and using the AO solution between measurements to purge the system of as many “stuck” cells as possible. Gating and debris exclusion Resolution of debris and sperm signal is partly instrument dependent, with the best, in our experience, being the Ortho Cytofluorograf, Becton Dickinson FACScan, and Coulter Elite, and a more difficult being the Becton Dickinson “jet-in-air” instrument. SCSA data are based on fluorescence signals values of sperm only and not from any extraneous debris giving off fluorescence signals (i.e., free cellular components, contamination, instrument noise). These debris signals can sometimes be eliminated by washing the sperm or processing through gradients. However, these procedures increase the risk of losing cell types and compromise the advantage of using whole semen measurements. Light scatter is often used to resolve somatic cells from debris, but the high variations in forward and right-angle light scatter from sperm do not effectively resolve sperm cells from debris. This debris problem is accentuated in samples with low sperm concentration and from patients undergoing chemotherapy that can cause extensive debris from killed cells.
7.13.20 Supplement 13
Current Protocols in Cytometry
In difficult cases of resolving sperm signal from debris, regular green and red fluorescence peak mode signals may not be sufficient. Sometimes, debris can initially be gated out using a green peak versus green area cytogram that results in resolving “true” sperm fluorescence from debris signals. This then gates into a relatively “debris-free” and analyzable red versus green peak mode cytogram. These are worst-case scenarios, as most samples can be gated with a simple 45° line as stated earlier (see Support Protocol 3) and shown in Figures 7.13.1-7.13.4. Signal resolution Resolution between debris and sperm signal is partly dependent on the physical parameters of the flow cell. Since AO is an equilibrium dye, a long distance between the sample injection point and the laser intersection point can cause a dilution of the effective AO stainability. For example, the Ortho Cytofluorograf sorting cell did not work satisfactorily for AO-stained cells for this reason. Freezing and thawing Repeated evidence from various species shows that freezing and thawing and then immediately refreezing a semen sample once does not cause significantly altered SCSA data relative to fresh samples. However, samples once frozen must stay at a constant temperature in an ultracold freezer or in dry ice (−70°C or colder). The authors have measured samples that were measured fresh, then frozen, thawed, and refrozen, and measured up to three years later (Evenson et al., 1986a, 1989b, 1993b) with equivalent results. However, samples frozen at −70°C for up to a decade or more may produce some difference in stainability, likely due to slow but continual nuclear-SH oxidation. Avoid using refrigerators and freezers with automatic defrosters because of the continual rise and fall of temperature. Likewise, removal of samples from an ultracold freezer into ambient air may cause the sample to undergo a ~20° to 70°C temperature change, not apparent to the eye as it may not reach the liquid state. The strong physical forces of temperature changes may damage the chromatin structure and cause an artifact in the SCSA data. Of importance, Figure 7.13.4, panel C, area 7, shows a tell-tale signature pattern that is consistent with a sample that has been thawed (e.g., because of freezer failure) for a significant period of time and refrozen, or repeatedly thawed and refrozen (see Anticipated Results
for extensive discussion of Fig. 7.13.4). For less than precise data collection, a sample may be thawed, an aliquot removed, and the sample immediately refrozen several times. However, the authors’ established laboratory rule is that for publication of critical slight differences, samples frozen and thawed no more than once must be used. When handling any sample (e.g., moving from box to box in the freezer), do not pick up the sample by the body of the tube, as a warm human hand will produce immediate microthawing of the sample. Use gloves or forceps, or grasp only the top of the tube. When sorting or rearranging samples, place them in a deep ice chest containing dry ice that will keep the box and the samples at least dry-ice cold (~70°C). Place sample boxes in an ultracold freezer in a rack that others are instructed not to remove. Placement of sample boxes near the bottom of the freezer is best, so that they are not subjected to temperature changes if the door is left open for extended periods of time. Shipping of samples Semen samples are shipped only by Federal Express (or other reputable overnight carrier) in an insulated, commercial shipping container containing 10 pounds of dry ice, broken up into pieces, satisfactory for Priority One overnight shipments from any point in the United States during any season. This amount of ice will keep for at least 2 days. A layer of dry ice chunks is placed on the bottom of the shipping container, the sample box is placed near the center of the shipping box, and then more dry ice is used to fill the box. It is best to ship only Monday through Wednesday, but not later in the week, in order to avoid the weekend. Liquid nitrogen dry shippers are an alternative option. A great advantage of the SCSA for male fertility studies is that the raw semen can be collected at home or work sites, transported in a well-insulated, temperature-equalizing container, frozen as raw semen without cryoprotectants on dry ice or plunged into liquid nitrogen and stored until ready for SCSA measurements. Thus, laboratories can accumulate a number of samples and measure them in one time period. On a typical day, a single technician can measure ~40 samples, in duplicate, plus the reference samples. Also, laboratories around the world can ship semen samples of interest by express-air courier to a laboratory with expertise in the SCSA. Samples collected and frozen on dry ice overseas should be carried as hand luggage to
Nucleic Acid Analysis
7.13.21 Current Protocols in Cytometry
Supplement 13
avoid delay in Customs and the risk of warming as the dry ice sublimes. Some shipments have been made from overseas with the samples in 70% ethanol. Regulations differ for each species (e.g., receiving bull semen samples from countries that have hoof-and-mouth disease). In one case the Customs officer agreed to accept the shipment provided that the sperm were sonicated, and the sucrose gradient-purified nuclei sent in well-sealed vials containing 70% ethanol. Sample volume The SCSA requires an extremely small fraction of a total ejaculate with a reasonable sperm count. As an example, if an ejaculate of 2.5 ml had an abnormally low sperm count of 10 ×106/ml, the sample would have 25 million cells. A 12.5-µl sample, diluted with 225 µl TNE buffer, contains enough sample for one or two measurements. This represents only 12.5/2500 µl or 0.5% of the original amount. In studies conducted by multiple investigators, it is recommended that this SCSA aliquot be removed first, diluted (if required) in 1× TNE buffer, and placed directly into a box with dry ice or an ultracold freezer unless the raw aliquot is measured immediately.
Sperm Chromatin Structure Assay for Fertility Assessment
Counseling physician/patients or managers of animal breeding stations Semen quality data on any individual are only a “snapshot” of sperm that have just undergone ∼2 months of maturation in the reproductive tract. These data do not indicate the quality of the sperm prior to or after the current sample. However, an excellent-quality sample is more likely to represent previous and future quality than a poor-quality sample. A variety of conditions—high fever, medications, diseases, and mental and physical stress—can profoundly affect sperm chromatin quality. If a sample is of poor chromatin quality (≥30% COMPαt) and the outcome is of great relevance (e.g., the sale of a high-priced stallion or bull or the question of alternative methods for human conception), then the counselor/consultant should carefully inquire about any previous stressful situations. A high fever (39°C) within the past 2 months represents transient damage to developing sperm. Data in the authors’ laboratory show that the quality comes back to normal after a full spermatogenic cycle (Evenson et al., 2000a). Such an individual should have another semen sample measured by the SCSA in 2 months
time. If the quality is still poor, a third sample can be measured in another month; however, provided no environmental/drug/chemical exposure is known, the condition is perhaps inherent and the likelihood of fathering a child is very poor. The SCSA predicts subfertility and infertility, but not fertility, because infertility may be due to many factors (e.g., poor motility or acrosome quality), even while the chromatin quality may be excellent. In controlled fertility experiments, the authors have yet to find an instance of a sample having ≥30% COMPαt and pregnancy occurring within a short time of the SCSA measurement. The authors have measured samples where all the classical parameters are normal, but the chromatin quality measured by the SCSA is very poor and therefore the likely cause of infertility. The clinical reports note only whether the sperm sample is compatible or incompatible with a good or poor fertility potential, based mostly on the percent of cells with DNA denaturation and percent with high DNA stainability. Unusual/interesting sample cases 1. Low concentration and/or “azoospermic samples.” A near-azoospermic sample will produce a very low flow rate (e.g., ~10 sperm/sec). These measurements are still valid, but if the rate is even lower, it may be best to count only a thousand cells or to concentrate the sperm. Remember that 103 sperm is still 5 to 10 times the number analyzed for classical parameters with a light microscope. Also, a sample with a 10 cells/sec flow rate may well have been diagnosed in an andrology laboratory as azoospermic. Thus, the SCSA is better suited to detect and evaluate the quality of near-azoospermic samples than classical light microscopic methods. When measuring a lowconcentration sample, do not increase the flow rate by increasing the sample pressure, which increases the diameter of the sample stream, decreases the sensitivity of the focus, and changes the AO staining characteristics from the regular sample stream size. 2. Bacterially infected samples. Significant bacterial infection can easily be detected by the SCSA when this population is detected above the 45° debris line (Evenson et al., 1991). The bacterial population is seen as a long thin dot pattern to the left of and parallel to the main SCSA sperm population as shown in Fig. 7.13.4, panel B, area 6 (also see Anticipated Results).
7.13.22 Supplement 13
Current Protocols in Cytometry
3. If samples have been frozen and thawed several times, or long periods of time have elapsed between thaw and refreeze procedures, an atypical fluorescence pattern appears (see Fig. 7.13.4C, area 7; also see Anticipated Results). If a batch of samples that have been processed and stored together all show this phenomenon, one must suspect that the samples have been compromised by a freezer failure or the human error of failing to return the samples to the freezer immediately. 4. Crystalline material. Some rather rare semen samples contain crystalline material that may significantly interfere with preparation and measurement. 5. Viscosity. In those cases where the sample does not undergo normal liquefaction and the sample remains gel-like, it is difficult to retrieve sperm from the sample, and the measured sperm might not be representative of the whole population. In near-azoospermic samples, the semen is not diluted and the sample is viscous by definition. This viscosity is greatly reduced, however, by the initial treatment with acid detergent solution, and the sample can then be handled like any other. Light microscopic AO method Several laboratories have tried to adapt the SCSA to light microscopic evaluation of AOstained sperm on glass slides (Tejada et al., 1984). Because AO equilibrium staining conditions are so strict, the method can give only a crude estimate of sperm chromatin integrity at best. Furthermore, the authors have found that the repeatability is very poor and the data do not agree well with the flow cytometer SCSA data (unpublished data). The use of this adaptation is not recommended.
Anticipated Results Each sample measured by the SCSA should yield data and results similar to Figure 7.13.1, the layout depending upon the instrument and software package used. Notice that there are dramatically increased SCSA values from the infertile (Fig. 7.13.1; panels D–F) male partner of a couple that did not conceive over the course of one year. Compared to the fertile sample, the Xαt, SDαt, and COMPαt values are more than doubled, with the latter increased by 10-fold. However, the % HIGRN is about the same. Note that there is little variation between the duplicate sample measurements, an indication of the high precision and repeatability of the assay. No other measure of semen quality has this precision of measurement. COMPαt values are
derived from the αt frequency histogram data (Fig. 7.13.1, panels C and F) and are the percentage of sperm with native DNA (area 1) and denatured DNA (area 2). Note how the conversion of the green versus red fluorescence cytogram to the total versus αt cytogram improves the visualization of the cell populations into normal and abnormal. The sperm signal is well resolved from the debris in both examples (Fig. 7.13.1, panels A and D). A gating region is drawn in these cytograms to include “true” sperm fluorescence signals for data calculations. Here, an elliptically shaped gate was used to exclude the debris signal, while often an ~45° angle is used for this portion of the region. For a poor-fertility sample, the patient in Fig. 7.13.1, panel D, did not have a significant amount of seminal debris. The shift in sperm from normal chromatin (Fig. 7.13.1, panel A; the largest, oval-shaped cluster of cells with little red fluorescence, i.e., the main population) to denatured DNA (anything to the right of the normal chromatin) is at a descending 45° angle. These cells outside the main population (COMPαt) represent the percentage of cells with denatured DNA. Note that in the sample from the infertile donor (Fig. 7.13.1, panel D), 65% of the sperm demonstrated DNA denaturation, well beyond our current threshold of 30% for predicted infertility (see Commentary). The pattern consists of a proportionally increasing population of abnormal sperm rather than two or more discrete clusters. A small percentage of the sperm have increased DNA stainability (high green fluorescence; % HIGRN) and this includes all sperm above the indicated threshold (dashed line at ~75% of scale) that starts at the top edge of the main population of sperm. All sperm cells falling above this dashed line, including those falling above and outside the region line, are characterized by increased DNA stainability and are included in the final percent high green calculation as shown in Figure 7.13.1, but are excluded from αt parameters. The mean green fluorescence is likely related to the condensation level of the sperm chromatin and extent of restricted access by DNA dyes. The sperm nuclear condensation process for mammals normally produces a 5fold reduction of DNA stainability with intercalating dyes such as AO relative to round spermatids (Evenson et al., 1986b). Lack of appropriate sperm maturation results in increased DNA stainability. Studies have shown that patients attending an infertility clinic often
Nucleic Acid Analysis
7.13.23 Current Protocols in Cytometry
Supplement 13
Sperm Chromatin Structure Assay for Fertility Assessment
have an increased DNA stainability (Evenson and Melamed, 1983; Engh et al., 1992 ), which can be visualized by univariate analysis (Engh et al., 1992) as well as by the SCSA bivariate analysis. In most cases, the defects of high DNA stainability and DNA denaturation are mutually exclusive, and any single cell rarely has both defects. However, the authors have observed a limited number of human clinic cases where both defects occur within the same sample. Figure 7.13.1 illustrates how to resolve normal versus high green fluorescing cells. % HIGRN is calculated from the red versus green fluorescence cytogram, and has been discussed in a recent paper (Evenson et al., 1999). The authors do not yet have enough experience with this parameter to be clear-cut as to what percent indicates the likelihood of infertility. COMPαt is the most important variable for fertility counselling. A recent study (Evenson et al., 1999) correlated SCSA data with accumulated fertility data. Due to high correlations and the establishment of thresholds of sperm with denaturable DNA for compatibility with fertility, the authors anticipate that the SCSA will be of most use prospectively in the human and animal clinics. According to this study, men with ≤15%, 15% to 30%, and ≥30% COMPαt would be predicted to have high fertility potential (pregnancy within 3 months), medium fertility potential (pregnancy within 4 to 12 months) and very low fertility potential (pregnancy in >1 year, defined as infertile), respectively (Table 7.13.3). A couple requesting counselling on their chances of pregnancy may be advised that their chances of pregnancy are greatly reduced if the percent sperm with abnormal chromatin is above 30%. Figure 7.13.4 consists of cartoon illustrations of normal fluorescence populations and common variations when sperm are measured by the SCSA. Panel A is a classical SCSA cytogram with the following numbered areas: (1) The main population of cells in a semen sample that, under the conditions imposed, remain as a coherent population. Although the DNA content in these cells is the same haploid amount, the cytogram cluster is elliptical in shape due to the optical artifact discussed in the text. This artifact is of little to no consequence to the outcome and interpretation of SCSA data. (2) This area contains the cells outside the main population, or COMPαt. These cells typically move out and downward from the main population at a ~45° angle, showing the increase of red fluorescence at the expense of green fluorescence. Some samples show a continuous
cluster ranging from just outside the main population to ones with very high red and very low green values. Alternatively, there are several discrete clusters in the COMPαt population with two shown in this illustration. (3) This area represents seminal debris consisting of broken cellular components and other particulate matter stained with AO. For proper SCSA analysis, the debris signals must be resolved from the sperm signals. Since the COMPαt population forms a ~45° slope, downward and to the right, an effective means to delineate between these cell clusters is to draw a 45° computer gate between them at the bottom edge of the main population. Note that this is also an active gate during acquisition so that 5000 or more sperm signals are accumulated and any debris signals excluded. This is most important in samples with a high ratio of debris to sperm so that the same number of sperm cells are statistically analyzed per sample. (4) The sperm signals appearing in areas 4 and 5 have high AO stainability and are termed sperm with high green stainability (% HIGRN). Chromatin in these sperm is probably not fully condensed, thus allowing a greater accessibility by intercalating DNA dyes. The percentage of HIGRN sperm is a ratio of the number of cells in areas 4 and 5 divided by the total number of cells contained in the gating region (includes sperm populations 1, 2, and 4) plus population 5 × 100. (Note that for most SCSA parameters the cytogram gating region contains the total number of sperm cells in areas 1, 2, and 4). The % HIGRN region starts at ~70%-75% of the green fluorescence scale. (5) This area contains sperm that also have higher DNA stainability, aggregates of sperm, early sperm forms, and possible somatic cells. The authors have not sorted this population for light microscopic identification; thus, a conclusion on how much of the region to include in this calculation has not been finalized. Panel B of Figure 7.13.5 is the same as panel A except that area (6) represents a semen sample with excessive bacterial contamination. Due to random size clumping of the bacteria, a straight line of signal is seen to the left of the sperm population. When present, this population is gated out during acquisition in order to accumulate 5000 sperm signals. Note that this tends to be confounded with cells in the upper high green stainability area also. Panel C of Figure 7.13.5 is also the same as A except that area (7) is an approximation of a sperm sample that has been compromised by excessive freeze/thaw cycles or left in a thawed state for an extended time period. This popula-
7.13.24 Supplement 13
Current Protocols in Cytometry
tion has an increased DNA stainability and shifts a bit to the right, as discussed in the text, but this compromised state is also revealed by a telltale hook usually present at the top of the population cluster. Samples showing this artifact are removed from the database. When all the samples in a box have this telltale artifact, inquiry usually confirms that a freezer failure or similar event occurred. Panel D of Figure 7.13.5 is a composite of possible sperm populations.
Time Considerations For fresh sperm samples, following collection of the sample, and a 30-min period for semen liquefaction, an aliquot of the sample can be measured by the SCSA, or be placed into a cryotube and plunged directly into liquid nitrogen or an ultracold freezer for storage. Since cryoprotectants are not needed, this is accomplished within a couple of minutes. Longitudinal studies of changes in sperm parameters are possible with the SCSA, as the sperm can be stored frozen until all samples are collected. The time of course depends on the length of the experimental study. The SCSA requires paying very strict attention to time. Once a sample that has been frozen is warmed for thawing, the process is committed. As mentioned in the text, a sample may be refrozen once or more for a nonexacting measurement of SCSA parameters. Thawing the sample takes ~30 to 60 sec and preparing the sample for flow cytometry measurement and equilibration in the flow cytometer takes 3 min. Data on 5000 cells can be collected in less than 1 min. When a flow operator becomes proficient, 5 to 6 samples measured in duplicate can be run per hour. Data analysis on each sample takes ∼10 min.
Literature Cited Aravindan, G.R., Bjordahl, J., Jost, L.K., and Evenson, D.P. 1997. Susceptibility of human sperm to in situ DNA denaturation is strongly correlated with DNA strand breaks identified by single-cell electrophoresis. Exp. Cell Res. 236:231-237. Ballachey, B.E., Hohenboken, W.D., and Evenson, D.P. 1987. Heterogeneity of sperm nuclear chromatin structure and its relationship to fertility of bulls. Biol. Reprod. 36:915-925 Ballachey, B.E., Saacke, R.G., and Evenson, D.P. 1988. The sperm chromatin structure assay: Relationship with alternate tests of sperm quality and heterospermic performance of bulls. J. Androl. 9:109-115.
Darzynkiewicz, Z., Traganos, F., Sharpless, T., and Melamed, M.R. 1975. Thermal denaturation of DNA in situ as studied by Acridine Orange staining and automated cytofluorometry. Cell Res. 90:411-428. Engh, E., Clausen, O.P.F., Scholberg, A., Tollefsrud, A., and Purvis, K. 1992. Relationship between sperm quality and chromatin condensation measured by sperm DNA fluorescence using flow cytometry. Int. J. Androl. 15:407-415. Estop, A.M., Munne, S., Jost, L.K., and Evenson, D.P. 1993. Alterations in sperm chromatin structure correlates with cytogenetic damage of mouse sperm following in vitro incubation. J. Androl. 14:282-288. Evenson, D.P. 1997. Sperm nuclear DNA strand breaks and altered chromatin structure: Are there concerns for natural fertility and assisted fertility in the andrology lab? Moving Beyond Boundaries: Clinical Andrology in the 21st Century. Andrology Laboratory Workshop, Postgraduate Course, Baltimore, Md. Evenson, D.P. 1999a. Alterations and damage of sperm chromatin structure and early embryonic failure. In Towards Reproductive Certainty: Fertility and Genetics Beyond 1999. Proceedings of the 11th World Congress on In Vitro Fertilization and Human Reproductive Genetics (R. Jannsen and D. Mortimer, eds.) pp. 313-329. Parthenon Publishing Group, New York. Evenson, D.P. 1999b. Loss of livestock breeding efficiency due to uncompensable sperm nuclear defects. Reprod. Fertility Dev. 11:1-15. Evenson, D.P. and Darzynkiewicz, Z. 1990. Acridine orange induced precipitation of mouse testicular sperm cell DNA reveals new patterns of chromatin structure. Exp. Cell Res. 187:328-334. Evenson, D.P. and Jost, L.K. 1993. Hydroxyurea exposure alters mouse testicular kinetics and sperm chromatin structure. Cell Prolif. 26:147159. Evenson, D.P. and Jost, L.K. 1994. Sperm chromatin structure assay: DNA denaturability. In Methods in Cell Biology, Vol. 42: Flow Cytometry (Z. Darzynkiewicz, J.P. Robinson, and H.A. Crissman, eds.) pp. 159-176. Academic Press, Orlando, Fla. Evenson, D.P. and Melamed, M.R. 1983. Rapid analysis of normal and abnormal cell types in human semen and testis biopsies by flow cytometry. J. Histochem. Cytochem. 31:248-253. Evenson, D.P., Darzynkiewicz, Z., and Melamed, M.R. 1980. Relation of mammalian sperm chromatin heterogeneity to fertility. Science 240:1131-1133. Evenson, D.P., Klein, F.A., Whitmore, W.F., and Melamed, M.R. 1984. Flow cytometric evaluation of sperm from patients with testicular carcinoma. J. Urol. 132:1220-1225.
Nucleic Acid Analysis
7.13.25 Current Protocols in Cytometry
Supplement 13
Evenson, D.P., Higgins, P.H., Grueneberg, D., and Ballachey, B. 1985. Flow cytometric analysis of mouse spermatogenic function following exposure to ethylnitrosourea. Cytometry 6:238-253. Evenson, D.P., Baer, R.K., Jost, L.K., and Gesch, R.W. 1986a. Toxicity of thiotepa on mouse spermatogenesis as determined by dual parameter flow cytometry. Toxicol. Appl. Pharmacol. 82:151-163. Evenson, D.P., Darzynkiewicz, Z., Jost, L., Janca, F., and Ballachey, B. 1986b. Changes in accessibility of DNA to various fluorochromes during spermatogenesis. Cytometry 7:45-53. Evenson, D.P., Baer, R.K., and Jost, L.K. 1989a. Flow cytometric analysis of rodent epididymal spermatozoal chromatin condensation and loss of free sulfhydryl groups. Mol. Reprod. Dev. 1:283-288. Evenson, D.P., Baer, R.K., and Jost, L.K. 1989b. Long term effects of triethylenemelamine exposure on mouse testis cells and sperm chromatin structure assayed by flow cytometry. Environ. Mol. Mutagen. 14:79-89.
Evenson, D., Jost, L., and Sailer, B. 1995b. Flow cytometry of sperm chromatin structure as related to toxicology and fertility. Proceedings of the Seventh International Spermatology Symposium, Cairns, Australia. Evenson, D.P., Jost, L.K., Zinaman, M.J., Clegg, E., Purvis, K., de Angelis, P., and Clausen, O.P. 1999. Utility of the sperm chromatin structure assay (SCSA) as a diagnostic and prognostic tool in the human fertility clinic. Hum. Reprod. 14:1039-1049. Evenson, D.P., Jost, L.K., Corzett, M., and Balhorn, R. 2000a. Effect of elevated body temperature on human sperm chromatin structure. J. Andrology. Submitted for publication. Evenson, D.P., Jost, L.K., and Varner, D.D. 2000b. Stallion sperm nuclear protamine-SH status and susceptibility to DNA denaturation are not strongly correlated. J. Fertility Reprod. Suppl. In press.
Evenson, D.P., Janca, F.C., Jost, L.K., Baer, R.K., and Karabinus, D.S. 1989c. Flow cytometric analysis of effects of l,3-dinitrobenzene on rat spermatogenesis. J. Toxicol. Environ. Health 28:81-98.
Fossa, S.D., De Angelis, P., Kraggerud, S.M., Evenson, D., Theodorsen, L., and Claussen, O.P. 1997. Prediction of post-treatment spermatogenesis in patients with testicular cancer by flow cytometric sperm chromatin structure assay. Commun. Clin. Cytometry 30:192-196.
Evenson, D.P., Janca, F.C., Baer, R.K., Jost, L.K., and Karabinus, D.S. 1989d. Effect of l,3-dinitrobenzene on prepubertal, pubertal and adult mouse spermatogenesis. J. Toxicol. Environ. Health 28:67-80.
Gledhill, B.L., Lake, S., Steinmetz, L.L., Gray, J.W., Crawford, J.R., Dean, P.N., and VanDilla, M.A. 1976. Flow microfluorometric analysis of sperm DNA content: Effect of cell shape on the fluorescence distribution. J. Cell Physiol. 87:367-376.
Evenson, D.P., Jost, L., Baer, R., Turner, T., and Schrader, S. 1991. Individuality of DNA denaturation patterns in human sperm as measured by the sperm chromatin structure assay. Reprod. Toxicol. 5:115-125.
Gorzcyca, W., Gong, J., and Darzynkiewicz, Z. 1993. Detection of DNA strand breaks in individual apoptotic cells by the in situ terminal deoxynucleotidyl transferase and nick translation assays. Cancer Res. 53:945-951.
Evenson, D.P., Emerick, R.J., Jost, L.K., KayongoMale, H., and Stewart, S.R. 1993a. Zinc-silicon interactions influencing sperm chromatin integrity and testicular cell development in the rat as measured by flow cytometry. J. Anim. Sci. 71:955-962.
Grajewski, B., Cox, C., Schrader, S.M., Murray, W.E., Edwards, R.M., Turner, T., Smith, J.M., Shekar, S., Evenson, D., Simon, S.W., and Conover, D.L. 2000. Semen quality and hormone levels among radiofrequency heat sealer operators. J. Occupational Environ. Med. In press.
Evenson, D.P., Jost, L.K., and Baer, R.K. 1993b. Effects of methyl methanesulfonate on mouse sperm chromatin structure and testicular cell kinetics. Environ. Mol. Mutagen. 21:144-153. Evenson, D.P., Jost, L.K., and Gandy, J.G. 1993c. Glutathione depletion potentiates ethyl methanesulfonate-induced susceptibility of rat sperm DNA denaturation in situ. Reprod. Toxicol. 7:297-304. Evenson, D.P., Thompson, L., and Jost, L. 1994. Flow cytometric evaluation of boar semen by the sperm chromatin structure assay as related to cryopreservation and fertility. Theriogenology 41:637-651. Sperm Chromatin Structure Assay for Fertility Assessment
nal axes flow cytometers. Cytometry 19:295303.
Evenson, D., Jost, L., Gandour, D., Rhodes, L., Stanton, B. Clausen, O.P., De Angelis, P., Coico, R., Daley, A., Becker, K., and Yopp, T. 1995a. Comparative sperm chromatin structure assay measurements on epiillumination and orthogo-
Larson, K.L., Brannian, J.D., Timm, B.K., Jost, L.K., and Evenson, D.P. 1999. Density gradient centrifugation and glass wool filtration of semen remove sperm with damaged chromatin. Hum. Reprod. 14:2015-2019. Larson, K., DeJonge, C., Barnes, A., Jost, L., and Evenson, D. 2000. Relationship between assisted reproductive techniques (ART) outcome and status of chromatin integrity as measured by the sperm chromatin structure assay (SCSA). Hum. Reprod. In press. Lemasters, G.K., Olsen, D.M., Yiin, J.H., Lockey, J.E., Shukla, R., Selevan, S.G., Schrader, S.M., Toth, G.P., Evenson, D.P., and Huszar, G.B. 1999. Male reproductive effects of solvent and fuel exposure during aircraft maintenance. Reprod. Toxicol. 13:155-166.
7.13.26 Supplement 13
Current Protocols in Cytometry
Martin, D. and Lenardo, M. 2000. Morphological, biochemical, and flow cytometric assays of apoptosis. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 14.13.1-14.13.21. John Wiley & Sons, New York. Potts, R.J., Newbury, C.J., Smith, G., Notarianni, L.J., and Jefferies, T.M. 1999. Sperm chromatin damage associated with male smoking. Mutat. Res. 423:103-111. Rubes, J., Lowe, X., Moore, D., Perreault, S., Slott, V., Evenson, D., Selevan, S., and Wyrobek, A.J. 1998. Cigarette-smoking lifestyle is associated with increased sperm disomy in teenage men. Fertility Sterility 70:715-723. Sailer, B.L., Jost, L.K., Erickson, K.R., Tajiran, M.A,. and Evenson, D.P. 1995a. Effects of X-ray irradiation on mouse testicular cells and sperm chromatin structure. Environ. Mol. Mutagen. 25:23-30.
Sassa, and I.G. Sipes, eds.) pp. 3.5.1-3.5.6. John Wiley & Sons, New York. Tejada, R.I., Mitchell, J.C., Norman, A., Marik, J.J., and Friedman, S. 1984. A test for the practical evaluation of male fertility by acridine orange (AO) fluorescence. Fertil. Steril. 42:87-91. Vine, M.F., Hulka, B.S., Everson, R.B., and Evenson, D. 2000. An assessment of DNA damage in the sperm of smokers and nonsmokers using the sperm chromatin structure assay. Cancer Epidemiol. Biomarkers Prevention. In press. Wyrobek, A.J., Schrader, S.M., Perrault, S.D., Fenster, L., Huszar, G., Katz, D.F., Osorio, A.M., Sublet, V., and Evenson, D. 1997. Assessment of reproductive disorders and birth defects in communities near hazardous chemical sites. III. Guidelines for field studies of male reproductive disorders. Reprod. Toxicol. 11:243-259.
Key References Evenson et al., 1980. See above.
Sailer, B.L., Jost, L.K., and Evenson, D.P. 1995b. Mammalian sperm DNA susceptibility to in situ denaturation associated with the presence of DNA strand breaks as measured by the terminal deoxynucleotidyl transferase assay. J. Androl. 16:80-87.
The first article correlating SCSA data and fertility.
Selevan, S.G., Borkovec, L., Slott, V.L., Zudova, Z., Rubes, J., Evenson, D.P., and Perreault, S.D. 2000. Semen quality and reproductive health of young Czech men exposed to seasonal air pollution. Submitted for publication.
Evenson et al., 1991. See above.
Evenson et al., 1999. See above. The most definitive paper on the SCSA and its application to human fertility clinics.
First longitudinal study utilizing the SCSA and showing the within-donor repeatability of samples over time and the repeatability of the SCSA when measuring the same sample several times.
Shapiro, H.M. 1985. Practical Flow Cytometry. Alan R. Liss, New York, p. 36.
Evenson et al., 1995a. See above.
Shapiro, H.M. 1988. Practical Flow Cytometry, 2nd ed. Alan R. Liss, New York, pp. 67-68, 205-207.
For more details on instrumentation already used with the SCSA.
Spanò, M., Kolstad, H., Larsen, S.B., Cordelli, E., Leter, G., Giwercman, A., and Bonde, J.P. 1998. ASCLEPIOS Study Group: The applicability of the flow cytometric sperm chromatin structure assay in epidemiological studies. Hum. Reprod. 13:2495-2505.
Contributed by Donald Evenson and Lorna Jost South Dakota State University Brookings, South Dakota
Sutherland, J.E and Costa, M. 1999. Assays for DNA damage. In Current Protocols in Toxicology (M.D. Maines, L.G. Costa, D.J. Reed, S.
This research is based upon work supported by Environmental Protection Agency Grant No. R827019, United States Department of Agriculture Sabbatical Grant No. 9803911, National Science Foundation EPSCoR Grant OSR 9452894, and South Dakota Futures Funds. This is South Dakota Agricultural Experiment Station Publication No. 3168 of the journal series.
Nucleic Acid Analysis
7.13.27 Current Protocols in Cytometry
Supplement 13
Analysis of Cell Proliferation and Cell Survival by Continuous BrdU Labeling and Multivariate Flow Cytometry
UNIT 7.14
A key parameter of cell culture is the quantitation of cell proliferation and cell survival. This can be determined by allowing proliferating cells to incorporate halogenated DNA precursors, such as 5-bromodeoxyuridine (BrdU) and 5-iododeoxyuridine, into DNA during passage through the S phase of the cell cycle. In the methods described in this unit, the halogenated DNA precursors can be detected by virtue of their property of quenching Hoechst dye fluorescence (Latt, 1973). After staining with Hoechst 33258 or 33342, cells can be analyzed with a flow cytometer (Kubbies and Rabinovitch, 1983; Rabinovitch, 1983). This way, cells that have undergone one, two, or three cell cycles during the period of labeling with halogenated DNA precursors (i.e., incorporation of halogenated DNA precursors during one, two, or three S phases) can be resolved. Counterstaining cells with a dye that exhibits fluorescence proportional to DNA content and in a different range of the emission spectrum than that of the Hoechst dyes (i.e., green, yellow, or red fluorescence emission), allows resolution of cells in the G1, S, and G2 phases of the cell cycle. Combining the information from the two DNA dyes allows one to distinguish cells in the G1, S, and G2 phases of three consecutive cell cycles (Rabinovitch et al., 1988). Recently, protocols have been developed that allow this type of information to be obtained from cells distinguished simultaneously on the basis of differential expression of up to two surface antigens or green fluorescent protein (GFP). In addition, calibration for the sample volume analyzed allows determination of the absolute number of proliferating and nonproliferating cells in a sample. With this technique, the proliferative survival of cells (i.e., the number of surviving and proliferating cells) can be determined. Presented in this unit are four protocols for analysis of cell proliferation and cell survival by flow cytometry: (1) a generic protocol to analyze the proliferative history of cells after continuous labeling with BrdU based on cell permeabilization and staining with Hoechst 33258 and ethidium bromide (see Basic Protocol); (2) a protocol to analyze the proliferative history of GFP-expressing versus nonexpressing cells in a cell culture (see Alternate Protocol 1); (3) a protocol to analyze the proliferative history of cell surface antigen–expressing versus non-expressing cells in a cell culture (see Alternate Protocol 2); and (4) a protocol to determine the absolute number of proliferating and nonproliferating cells in a sample by calibration for the sample volume analyzed (see Alternate Protocol 3). NOTE: BrdU is a photosensitizing drug. To avoid photochemical damage to the stained cells, all incubations are performed in subdued light. NOTE: All protocols described have been performed on cultured animal cells; limited data exist regarding the use of these methods for plant cells and in yeast. Expertise is assumed for basic techniques in flow cytometry as well as in cell culture and harvesting (of both suspension cultures and adherent cells; APPENDIX 3B), immunocytochemistry (Watkins, 1989), and fluorescence microscopy (UNIT 2.4). CAUTION: DMSO and dye solutions are potentially toxic to humans. Use nitrile gloves and wear eye protection at all stages of handling. Seek medical advice if dye or dye solutions are ingested or inhaled. The dyes mentioned are for in vitro use only; do not administer either externally or internally. All staining solutions should be poured through a funnel with a filter containing activated charcoal in a fume hood. The nonfluorescent filtrate can be poured down the sink. When the passing solution becomes fluorescent, the Contributed by Martin Poot, M. Rosato, and Peter S. Rabinovitch Current Protocols in Cytometry (2001) 7.14.1-7.14.9 Copyright © 2001 by John Wiley & Sons, Inc.
Nucleic Acid Analysis
7.14.1 Supplement 15
filter should be incinerated or disposed of according to applicable rules for environmental hygiene and a fresh filter should be installed. BASIC PROTOCOL
ANALYSIS OF CELL PROLIFERATION BY CONTINUOUS 5-BrdU INCORPORATION FOLLOWED BY HOECHST 33258 AND ETHIDIUM BROMIDE FLOW CYTOMETRY Incorporation of BrdU into DNA leads to quenching of the Hoechst dyes. After excitation with UV light (∼360 nm), the Hoechst dye emits blue fluorescence (∼450 nm) and ethidium bromide emits red fluorescence (590 to 630 nm). Ethidium bromide can also be excited at 488 nm with an argon laser. This protocol allows determining the distribution of cells among the G1, S, and G2 compartments of three successive cell cycles. Materials 10 mM 5-bromodeoxyuridine (BrdU) in distilled water Cells in appropriate culture medium containing 10% FBS (APPENDIX 2A) Generic Hoechst staining buffer (see recipe) 10% (v/v) Nonidet P-40 (or IGEPAL, Sigma) 1 mg/ml ethidium bromide (Sigma) in distilled water 15-ml screw-capped centrifuge tubes Flow cytometer with either a mercury arc lamp or an argon laser (tuned to 360 nm) as excitation source; alternatively, two time-resolved argon lasers (tuned to 360and 488-nm excitation wavelengths, respectively) can be used 12 × 75–mm polypropylene flow cytometer sample tubes Computer and appropriate software for data collection and processing Label cells with BrdU 1. Add 10 mM BrdU to a final concentration of 100 µM to the cell culture medium of each culture of interest, wrap the plates or flasks with aluminum foil, and continue cell culture for the desired time period. It is important that the BrdU not be depleted during cell growth, and for this reason the starting cell concentration should not exceed 100,000 cells/ml or 2,500 cells/cm2. Since BrdU is a photosensitizer, all staining steps have to be performed in subdued light. The duration of BrdU labeling depends on the cell cycle time of the cells under study and on the kind of information required (see Commentary). For instance, if detailed information regarding cell cycle compartment transition times is required, several samples spanning at least two cell cycle periods have to be taken. In some cell types it may be necessary to include an equimolar amount of deoxycytidine (see Commentary).
Prepare cells 2. Harvest BrdU-labeled cells by standard procedures in 15-ml screw-capped centrifuge tubes and centrifuge 5 min at 200 × g, room temperature. Proceed to cell staining. Pellets may be resuspended in cell culture medium supplemented with 10% FBS and 10% DMSO and stored at −20°C for up to 6 months without detectable degradation.
3. Resuspend cell pellets at 0.5-1.0 × 106 cells/ml in generic Hoechst staining buffer, add an aliquot of 10% NP-40 solution to a final concentration of 0.1%, and incubate ≥15 min at room temperature in the dark. Since BrdU is a photosensitizer all staining steps have to be performed in subdued light. Analysis of Cell Proliferation and Cell Survival by Continuous BrdU Labeling
4. After 15 min staining in the dark, add 1 mg/ml ethidium bromide to a final concentration of 2.0 µg/ml and continue staining for another 15 min, room temperature, in the dark.
7.14.2 Supplement 15
Current Protocols in Cytometry
Set up flow cytometer 5. Set up and optimize the flow cytometer. Collect ultraviolet (360 nm)-excited blue fluorescence (Hoechst fluorescence) and 488 nm-excited fluorescence between 590 and 630 nm (ethidium bromide). Carefully resuspend cell sample and disrupt cell clumps by gently pipetting up and down a few times in 12 × 75–mm polypropylene flow cytometer sample tubes immediately before analysis. To avoid “bleeding” of Hoechst fluorescence into the ethidium bromide channel, time-resolved sample excitation (with spatially separated UV and 488-nm laser beams) can be used. Ethidium bromide can be excited by both 360- and 488-nm light. In addition, ethidium bromide will be excited by blue Hoechst 33342 fluorescence via fluorescence energy transfer. Thus, excitation can be made with UV light only, but the visual representation is clearer with dual-wavelength excitation. During harvesting, cells tend to clump; to obtain meaningful data on a per-cell basis it is essential to resuspend cells immediately before analysis. If this procedure is not sufficient to disrupt cell clumps, it is advised that they be broken up by aspirating the samples through a 0.5-mm internal diameter 21-G needle.
6. Collect ≥10,000 signals for each sample on a computer and appropriate software. If detailed information regarding cell cycle distributions among two or three successive cell cycles is required, one should acquire 2 or 3 times 10,000 signals.
ANALYSIS OF CELL PROLIFERATION OF GFP-EXPRESSING VERSUS NONEXPRESSING CELLS IN A SINGLE-CELL CULTURE Upon excitation with 488-nm laser light, GFP emits green fluorescence (∼510 nm). This fluorescence can be detected in samples that are stained with Hoechst 33258 and 7-AAD dyes. With this protocol, the distribution of GFP-expressing and nonexpressing cells among the G1, S, and G2 compartment of three successive cell cycles can be determined.
ALTERNATE PROTOCOL 1
Additional Materials (also see Basic Protocol) GFP-containing vector/virus cells 2% (w/v) paraformaldehyde solution (see recipe) Phosphate buffered saline (PBS; APPENDIX 2A) 1% (w/v) saponin (Sigma) in PBS 1 mg/ml 7-aminoactinomycin D (7-AAD; Calbiochem) in DMSO Label cells with BrdU 1. After transfection/transduction with a GFP-containing vector/virus, label cells with 100 µM BrdU as described in the Basic Protocol, step 1. Prepare and fix cells 2. Harvest BrdU-labeled cells by standard procedures in 15-ml screw-capped centrifuge tubes and centrifuge 5 min at 200 × g, room temperature. 3. Pipet 7 ml of 2% paraformaldehyde into a second 15-ml screw-capped cell culture tube. 4. Resuspend the cell pellet in 7 ml PBS and add cell suspension dropwise to the tube with 1 ml of 2% paraformaldehyde while vortexing at maximum speed. The final paraformaldehyde concentration is 0.25%. Let the tube sit 15 min at room temperature to fix cells. This procedure is intended to minimize the formation of cell clumps during the fixation step.
5. Centrifuge samples 5 min at 200 × g, room temperature, decant supernatant, and resuspend cell pellets in PBS.
Nucleic Acid Analysis
7.14.3 Current Protocols in Cytometry
Supplement 15
At this stage samples can be stored overnight or several days at 4°C.
6. Centrifuge cells, resuspend in generic Hoechst buffer, and add an aliquot of 1% saponin solution to a final concentration of 0.05%. Saponin is a mild detergent that dissolves cholesterol molecules in the cell membranes; this makes the nuclear membranes permeable to 7-AAD.
7. Incubate ≥1 hr at room temperature in the dark, add an aliquot of 1 mg/ml 7-AAD stock solution to a final concentration of 20 µg/ml, and continue staining 1 hr more. Set up flow cytometer 8. Set up and optimize the flow cytometer. Collect ultraviolet (360 nm)-excited blue fluorescence (Hoechst fluorescence), 488 nm-excited green GFP fluorescence (∼510 nm), yellow autofluorescence (∼575 nm), and 488 nm-excited red 7-AAD fluorescence (>630 nm). To avoid “bleeding” of Hoechst fluorescence into the GFP and the 7-AAD channels, time-resolved sample excitation (with spatially separated UV and 488-nm laser beams) must be used. The fluorescence emission maximum of GFP is 508 nm. Cellular autofluorescence covers the region between 500 and 600 nm, but is usually stronger >550 nm. Thus, the yellow autofluorescence can be used to compensate for the influence of green autofluorescence on GFP detection. This allows improved detection of weak GFP fluorescence (see Commentary). During harvesting, cells tend to clump; to obtain meaningful data on a per-cell basis it is essential to resuspend cells immediately before analysis. If this procedure is not sufficient to disrupt cell clumps, it is advised that the samples be passed at least ten times through a 0.5-mm internal diameter 21-G needle. ALTERNATE PROTOCOL 2
ANALYSIS OF CELL PROLIFERATION OF ANTIGEN-EXPRESSING VERSUS NONEXPRESSING CELLS This protocol enables analysis of the proliferative history of cell surface antigen-expressing versus nonexpressing cells. The method involves cell surface antigen labeling with antibodies, followed by cell fixation, detergent treatment, and staining with Hoechst 33258 and 7-AAD. Additional Materials (also see Basic Protocol) FBS-PBS: 2% fetal bovine serum and 0.1 % NaN2 in PBS Primary and secondary antibodies for desired antigens 2% (w/v) paraformaldehyde (see recipe) Phosphate buffered saline (PBS; APPENDIX 2A) 1% (w/v) saponin (Sigma) in PBS 1 mg/ml 7-aminoactinomycin D (7-AAD; Calbiochem) in DMSO Label cells with BrdU 1. After isolation, label cells in culture with 100 µM BrdU as described in the Basic Protocol, step 1. Prepare and fix cells 2. Harvest BrdU-labeled cells by standard procedures in 15-ml screw-capped centrifuge tubes and centrifuge 5 min at 200 × g, room temperature.
Analysis of Cell Proliferation and Cell Survival by Continuous BrdU Labeling
3. Resuspend cell pellets in FBS-PBS, centrifuge again, and resuspend again in CFMPBS at a concentration of 4 × 106 cells/ml. 4. Stain cells with monoclonal antibodies for desired cell surface antigens, such as CD3, CD4, CD8, CD14, CD19, and CCR5, either directly conjugated with fluorescein
7.14.4 Supplement 15
Current Protocols in Cytometry
isothiocyanate (FITC) and phycoerythrin (PE) or with fluorescently labeled secondary antibodies (see UNIT 6.2). 5. Pipet 1 ml 2% paraformaldehyde into a new 15-ml screw-capped cell culture tube. 6. After immunostaining, resuspend cells in 7 ml PBS and add cell suspension dropwise to the tube with 1 ml 2% paraformaldehyde while vortexing at maximum speed (final paraformaldehyde concentration, 0.25%). Let the tubes sit 15 min at room temperature to fix cells. This procedure is intended to minimize the formation of cell clumps during the fixation step.
7. Centrifuge samples 5 min at 200 × g, room temperature, decant supernatant, and resuspend cell pellets in PBS. Wash cells two times with ice-cold FBS-PBS. At this stage samples can be stored overnight or several days at 4°C.
8. Centrifuge cells, resuspend in generic Hoechst buffer, and add an aliquot of 1% saponin solution to a final concentration of 0.05%. Saponin is a mild detergent that dissolves cholesterol molecules in the cell membranes; this makes the nuclear membranes permeable to 7-AAD.
9. Incubate ≥1 hr at room temperature in the dark, add an aliquot of 1 mg/ml 7-AAD stock solution to a final concentration of 20 µg/ml, and continue staining 1 hr more. Set up flow cytometer 10. Set up and optimize the flow cytometer. Collect ultraviolet (360 nm)-excited blue fluorescence (Hoechst fluorescence), 488 nm-excited green (∼525 nm) and yellow fluorescence (∼575 nm) from the FITC- and PE-labeled antibodies, respectively, and 488 nm-excited red fluorescence (>630 nm) from the 7-AAD. To avoid “bleeding” of Hoechst fluorescence into the green (FITC), yellow (PE), and red (7-AAD) channels, time-resolved sample excitation (with the UV laser beam set as the second excitation source) must be used. If only FITC labeling is used (no PE), then 488 nm-excited yellow autofluorescence can be collected and compensation used to improve the detection of low-level FITC staining, as described for GPF detection in Alternate Protocol 1. During harvesting, cells tend to clump; to obtain meaningful data on a per-cell basis, it is essential to resuspend cells immediately before analysis. If this procedure is not sufficient to disrupt cell clumps, it is advised that they be broken up by aspirating the samples using a 0.5-mm internal diameter 21-G needle.
ANALYSIS OF PROLIFERATIVE SURVIVAL In this protocol, the absolute number of proliferating and nonproliferating cells in a sample is determined by calibration for the sample volume analyzed. This calibration is achieved by adding a known number of standard particles (e.g., chicken erythrocyte nuclei; CEN) to the buffer in which all cell pellets are resuspended. The procedure can be used in combination with the generic protocol (see Basic Protocol) or with the protocols for the analysis of GFP-expressing or surface-marker-expressing samples (see Alternate Protocols 1 and 2).
ALTERNATE PROTOCOL 3
Nucleic Acid Analysis
7.14.5 Current Protocols in Cytometry
Supplement 15
Additional Materials (also see Basic Protocol) Chicken erythrocyte nuclei (CEN; BioSure Controls) Label cells with BrdU 1. Add 100 µM BrdU to the cell culture medium of each culture of interest, wrap the plates or flasks with aluminum foil, and continue cell culture for the desired time period. Since BrdU is a photosensitizer, all staining steps have to be performed in subdued light. The duration of BrdU labeling depends on the cell cycle time of the cells under study and on the kind of information required (see Commentary). For instance, if detailed information regarding cell cycle compartment transition times is required, several samples spanning at least two cell cycle periods have to be taken.
Prepare cells 2. Harvest BrdU-labeled cells by standard procedures in 15-ml screw-capped centrifuge tubes and centrifuge 5 min at 200 × g, room temperature. Proceed to cell staining. Pellets may be resuspended in cell culture medium supplemented with 10% FBS and 10% DMSO and stored at −20°C for up to 6 months without detectable degradation.
3. Resuspend CEN in provided storage buffer by vigorously vortexing the dropper bottle with CEN. Add 1 µl CEN per ml of Hoechst buffer. 4. Resuspend cell pellets at 0.5-1.0 × 106 cells per ml in CEN-supplemented generic Hoechst staining buffer and incubate ≥15 min, room temperature, in the dark. Since BrdU is a photosensitizer, all staining steps have to be performed in subdued light.
5. After 15 min of staining in the dark, add an aliquot of 1 mg/ml ethidium bromide stock solution to a final concentration of 2.0 µg/ml, and continue staining 15 min longer in the dark. Set up flow cytometer 6. Set up and optimize the flow cytometer. Collect ultraviolet (360 nm)-excited blue fluorescence (Hoechst fluorescence) and 488 nm-excited fluorescence between 590 and 630 nm (ethidium bromide). Carefully resuspend the cell sample and disrupt cell clumps by gently pipetting up and down a few times immediately before analysis. To avoid “bleeding” of Hoechst fluorescence into the ethidium bromide channel, time-resolved sample excitation (with the UV laser beam set as the second excitation source) can be used, as described in the Basic Protocol. During harvesting, cells tend to clump; to obtain meaningful data on a per-cell basis, it is essential to resuspend cells immediately before analysis. If this procedure is not sufficient to disrupt cell clumps, it is advised that they be broken up by aspirating the samples at least ten times through a 0.5-mm internal diameter 21-G needle.
7. Collect at least 10,000 cell signals for each sample on a computer with the appropriate software. If detailed information regarding cell cycle distributions among two or three successive cell cycles is required, one should acquire 2 or 3 times 10,000 signals. The number of CEN signals is usually between 5% and 10% of total signals, such that between 500 and 3,000 CEN signals will be acquired. If fewer than 500 CEN signals are acquired, a larger total number of signals should be collected to allow for accurate enumeration of CEN. Analysis of Cell Proliferation and Cell Survival by Continuous BrdU Labeling
7.14.6 Supplement 15
Current Protocols in Cytometry
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Generic Hoechst buffer 100 mM Tris⋅Cl, pH 7.4 154 mM NaCl 1 mM CaCl2 0.5 mM MgCl2 0.2% (w/v) bovine serum albumin (BSA) 1.2 µg/ml Hoechst 33258 (Sigma) Store up to 1 month at 4°C in the dark Paraformaldehyde solution, 2% Weigh out 10 g of paraformaldehyde (J.T. Baker). Measure 500 ml of PBS (APPENDIX 2A) and adjust pH to 12.0 with NaOH or KOH. Heat PBS, pH 12, to 60°C in fume hood, slowly add the paraformaldehyde powder, and stir for 5 to 10 min. Allow to cool to room temperature in the fume hood. At room temperature adjust pH to 7.4. Store up to 1 month at 4°C in the dark. COMMENTARY Background Information Halogenated analogs of deoxyuridine (such as BrdU) are incorporated into DNA in place of thymidine. This incorporation into DNA leads to a specific distortion of the geometry of the DNA helix (Loontiens et al., 1990) such that the fluorescence of Hoechst dyes is quenched (Latt, 1973). This quenching of Hoechst fluorescence allows cells that have incorporated BrdU to be distinguished from those that have not. Analysis of Hoechst 33258-stained cells by flow cytometry has become a simple way to reliably enumerate cells that have incorporated BrdU (Kubbies and Rabinovitch, 1983; Rabinovitch et al., 1988). Counterstaining cells with a dye that exhibits fluorescence proportional to DNA content and fluoresces in a different range of the emission spectrum than that of the Hoechst dyes (i.e., green, yellow, or red fluorescence emission) allows resolution of cells in the G1, S, and G2 phases of the cell cycle. This methodology resolves cells that have undergone one, two, or three cell cycles during the period of labeling with BrdU (i.e., have incorporated BrdU during one, two, or three S phases) and simultaneously distinguishes cells in the G1, S, and G2 phases of these three consecutive cell cycles (Rabinovitch et al., 1988).
Critical Parameters and Troubleshooting Incorporation of BrdU into DNA may lead to an enhanced sensitivity of cells towards cer-
tain drugs (Poot et al., 1991) and to UV light. In this way, BrdU may elicit a cytotoxic response that may lead to a high level of G2 cells. It should be noted that G2 arrest will not destroy the utility of this assay for detecting the proportion of noncycling (G0/G1) cells, but it will prevent the assessment of the progression of cells from one cycle to the next. If the latter is important in the experimental design, it is recommended that each new cell type and drug to be investigated be first tested with a range of BrdU concentrations for BrdU-dependent G2 arrest before the BrdU method is used on a routine basis. If a high level of G2-phase cells is observed, the BrdU concentration in the cell culture medium should be reduced. It has been suggested that BrdU may perturb cellular nucleotide pools in some cells, such as human fibroblasts, which then may affect cell cycling (Poot et al., 1994). To avoid this, addition of 65 µM deoxycytidine to the cell labeling medium has been recommended (Poot et al., 1994). A too-low concentration of ethidium bromide (see Basic Protocol) or 7-AAD (see Alternate Protocols 1 and 2) may lead to a poor coefficient of variation (CV) in this fluorescence parameter. This can easily be remedied by adding more ethidium bromide or 7-AAD and reanalyzing the sample. In addition, increasing the staining time (in particular with paraformaldehyde-fixed samples) may improve CVs for both Hoechst and for ethidium bromide (see Basic Protocol) or 7-AAD (see Alternate Protocols 1 and 2) fluorescence.
Nucleic Acid Analysis
7.14.7 Current Protocols in Cytometry
Supplement 15
Poor quenching of Hoechst fluorescence may be due to poor BrdU incorporation or to insufficient Hoechst staining (Kubbies and Rabinovitch, 1983). The first may result from unusually high levels of thymidine in the serum used to culture the cells. In this case a higher BrdU concentration has to be used. Insufficient Hoechst staining may result from too high a cell density. Diluting the sample with Hoechst staining buffer and adding a compensating amount of ethidium bromide or 7-AAD may improve the cytogram. When samples show excessive clumping it is recommended that samples be passed several times through a syringe with a 0.5-mm internal diameter 21-G needle. If difficulties in distinguishing GFP-expressing and nonexpressing cells are experienced, the experimenter should consult the unit on reporters of gene expression (UNIT 9.12).
Anticipated Results With optimal BrdU labeling and after appropriate staining, a cytogram similar to Figure
7.14.1 should be obtained. It should be noted that the signal clusters representing the G0/G1 and the G1 cells of the second and third cell cycle are on a horizontal line, since they all have identical EB-fluorescence intensity. This result allows for easier analysis of the distribution of signals among the G1, S, and G2 compartments of each cell cycle (see section on data processing below) and is obtained only when two spatially separated laser beams (i.e., UV and 488 nm) are used. If single (UV) excitation is used, the G0/G1 and the G1 clusters appear on a line that passes through the origin of the Hoechst/ethidium bromide cytogram. If the cytograms show distorted images, the experimenter should consult the troubleshooting guide. Processing of data The cytogram shown in Figure 7.14.1 represents the result of an experiment in which cells have progressed through three consecutive cycles. After harvesting, the sample was processed according to Alternate Protocol 3.
128 1st cycle
112 6
Ethidium bromide fluorescence
7 G2/M
96
8
G2/M
3rd cycle S
80 S 64
48
5
G1 2nd cycle
G0/G1
32
16 CEN 0 0
Analysis of Cell Proliferation and Cell Survival by Continuous BrdU Labeling
16
32
80 48 64 Hoechst fluorescence
96
112
128
Figure 7.14.1 Bivariate cytogram of cultured human cells stained with Hoechst 33258 dye (x-axis) and ethidium bromide (y-axis) with chicken erythrocyte standard added (see Alternate Protocol 3). The signal cluster corresponding to resting cells is labeled G0/G1, the cluster of first-cycle cells (S + G2/M) is labeled as such, as are the G1, S, and G2/M cells in the 2nd and 3rd cell cycles, and the cluster of chicken erythrocyte nuclei is labeled CEN.
7.14.8 Supplement 15
Current Protocols in Cytometry
Analysis software was used to obtain the numbers of cells in the following cell cycle compartments: (1) cells that have not incorporated BrdU (G0/G1); (2) cells that have incorporated BrdU, but not divided (1st cycle); (3) cells with incorporated BrdU that have divided once (2nd cycle); (4) cells that have divided twice (3rd cycle); and (5) chicken erythrocyte nuclei (CEN). From these data the following parameters can be derived. The percentage of proliferating cells. This number is most meaningful if expressed as the percentage of cells originally in the culture that proliferated. This can be calculated by adding together the number of cells that have incorporated BrdU, but have not divided (1st cycle S plus G2/M), plus the number of cells that have divided once (2nd cycle cells) divided by two (since two cells in the second cycle have resulted from one cell in the first cycle), plus the number of cells that have divided twice (3rd cycle cells) divided by four (since four cells in the third cycle have resulted from one cell in the first cycle), divided by the total number of cells originally plated (the above number plus the G0/G1 cells) multiplied by 100%. Proliferative survival. This is the number of proliferating cells (see above) normalized to the number of chicken erythrocyte nuclei (CEN). In this way the number of proliferating (and surviving) cells can be determined. If the actual number of CENs added to the culture is known, the absolute number of proliferating and surviving cells can be calculated; however it is often easiest to express this number relative to the proliferative survival of untreated control cells. This parameter is particularly informative in systems in which cell death by apoptosis or necrosis takes place. The impact of cell death on the cell total number in a culture is thus assessed. The CEN population is shown “circled” in Figure 7.14.1. Identification of CEN against a background of debris from the larger cells is facilitated by simultaneous analysis of forward and right-angle scatter signals; gating on these two scatter signals plus the two fluorescence parameters usually allows unambiguous identification of CEN (not shown).
Time Considerations For staining and analyzing 12 samples, 1 hr for sample staining and 1 hr for flow cytometric analysis should be allocated. Staining of paraformaldehyde-fixed samples should be ≥2 hr before analysis by flow cytometry.
Literature Cited Kubbies, M. and Rabinovitch, P.S. 1983. Flow cytometric analysis of factors which influence the BrdUrd-Hoechst quenching effect in cultivated human fibroblasts and lymphocytes. Cytometry 3:276-281. Latt, S.A. 1973. Microfluorometric detection of deoxyribonucleic acid replication in human metaphase chromosomes. Proc. Natl. Acad. Sci. U.S.A. 70:3395-3399. Loontiens, F.G., Regenfuss, P., Zechel, A., Dumortier, L., and Clegg, R.M. 1990. Binding characteristics of Hoechst 33258 with calf thymus DNA, poly[d(A-T)], and d(CCGGAATTCCGG): Multiple stoichiometries and determination of tight binding with a wide spectrum of site affinities. Biochemistry 29:9029-9039. Poot, M., Schuster, A., and Hoehn, H. 1991. Cytostatic synergism between bromodeoxyuridine, bleomycin, cisplatin and chlorambucil demonstrated by a sensitive cell kinetic assay. Biochem. Pharmacol. 41:1903-1909. Poot, M., Hoehn, H., Kubbies, M., Grossmann, A., Chen, Y., and Rabinovitch, P.S. 1994. Cell-cycle analysis using continuous bromodeoxyuridine labeling and Hoechst 33358-ethidium bromide bivariate flow cytometry. Methods Cell Biol. 41:327-340. Rabinovitch, P.S. 1983. Regulation of human fibroblast growth rate by both noncycling cell fraction transition probability is shown by growth in 5-bromodeoxyuridine followed by Hoechst 33258 flow cytometry. Proc. Natl. Acad. Sci. U.S.A. 80:2951-2955. Rabinovitch, P.S., Kubbies, M., Chen, Y.C., Schindler, D., and Hoehn, H. 1988. BrdU-Hoechst flow cytometry: A unique tool for quantitative cell cycle analysis. Exp. Cell Res. 174:309-318. Watkins, S. 1989. Immunohistochemistry. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 14.6.1-14.6.13. John Wiley & Sons, New York.
Contributed by Martin Poot, M. Rosato, and Peter S. Rabinovitch University of Washington Seattle, Washington
Nucleic Acid Analysis
7.14.9 Current Protocols in Cytometry
Supplement 15
Ultraviolet-Induced Detection of Halogenated Pyrimidines (UVID)
UNIT 7.15
Halogenated pyrimidines such as bromodeoxyuridine (BrdU) are used to identify DNAsynthesizing cells. Immunocytochemical detection of these DNA precursors when combined with counterstaining of DNA allows one to distinguish S-phase cells from cells in G0/G1 and G2/M. Their detection with monoclonal antibodies, however, requires the denaturation of DNA to make them accessible to the antibody (UNIT 7.7). The harsh conditions needed to denature DNA (acid or heat treatment) can lead to cell loss and abrogation of other cellular markers and antigens. Partial photolysis of BrdU with ultraviolet-B light (UV-B) induces extensive damage to nuclei which incorporated the halogenated pyrimidine. Under hypotonic conditions certain monoclonal antibodies can detect BrdU in the unfolding chromatin. In contrast to other methods, such as treatment with DNase or SBIP (Li et al., 1994a), this approach requires no DNA denaturation step nor any form of enzymatic treatment and allows for the detection of other cellular markers. The Basic Protocol can be used with either coagulative fixation (ethanol) or fixation with cross-linking agents (paraformaldehyde), and allows the simultaneous detection of a wide range of antigens. In the authors’ experience, when using the UVID method, paraformaldehyde offers some advantages over ethanol for certain surface markers and often improves the retention of intracellular antigens. ULTRAVIOLET-INDUCED DETECTION OF BrdU IN ETHANOL- OR PARAFORMALDEHYDE-FIXED CELLS
BASIC PROTOCOL
This protocol describes the ultraviolet-induced detection (UVID) of BrdU in ethanol- or paraformaldehyde-fixed cells. Exponentially growing cells are incubated with BrdU, harvested, and irradiated with UV-B light to partially photolyze the DNA sections that contain incorporated BrdU and to induce nuclear damage. Cells are then fixed in cold ethanol. Hypotonic conditions subsequently allow the detection of BrdU with monoclonal antibodies. The simultaneous detection of other cellular markers is optional. For surface antigens, cells should be stained after UV irradiation and prior to fixation. Intracellular antigens can be detected by adding the antibody to the hypotonic buffer. The counterstaining of DNA with 7-aminoactinomycin D (7-AAD) is optional and is performed simultaneously with the hypotonic treatment. Materials 10 mM BrdU stock solution (see recipe) Exponentially growing cells Phosphate buffered saline (PBS; APPENDIX 2A) with and without 0.1% BSA 70% (v/v) high purity ethanol, −20°C 0.2% (w/v) paraformaldehyde in PBS, 20°C (see recipe) Sodium tetraborate solution (see recipe) Anti-BrdU FITC antibody 7-AAD solution (optional; see recipe) 37°C, 5% CO2 incubator Tabletop centrifuge and swinging bucket rotor Hand-held 8-W UV-B lamp with an intensity of ∼22 W/m2 measured 0.5 cm above the filter surface 14-ml clear polypropylene tubes (e.g., Sarstedt) Contributed by Hans-Joerg Hammers and Peter Schlenke Current Protocols in Cytometry (2001) 7.15.1-7.15.6 Copyright © 2001 by John Wiley & Sons, Inc.
Nucleic Acid Analysis
7.15.1 Supplement 16
Clear tape or rubber band Microcentrifuge Vortex mixer Tubes suitable for use with flow cytometer Distilled water Flow cytometer equipped with a 488-nm argon laser and filters for detection of FITC (525 nm), phycoerythrin (PE; 575 nm), and 7-AAD (650-nm long-pass or 675-nm band-pass) NOTE: Not every clone can be used with the UVID approach. The authors recommend the Bu20a clone (Dako), but the 3D4 clone (PharMingen) and the B44 clone (Becton Dickinson) are acceptable as well. Label DNA-synthesizing cells with BrdU 1. Add 30 µl 10 mM BrdU stock solution per 10 ml exponentially growing cells to obtain 30 µM BrdU final concentration in the culture. Mix well. Optimal incorporation of BrdU requires culture conditions that facilitate exponential growth; these conditions are cell-type dependent. As a rule of thumb, density in suspension cultures should not exceed 1 × 106 cells/ml and cell confluence in adherent cell cultures should be avoided. Up to 4 × 106 cells per sample can be processed; however, for optimal results, 2 × 106 cells per sample are recommended. Avoid any subsequent light exposure of the cell culture.
2. Incubate another 60 min in a 37°C, 5% CO2 incubator. 3. Harvest cells and centrifuge in a tabletop centrifuge with a swinging bucket rotor at 20°C, selecting time and g force appropriate for the cell line used (e.g., 40 min at 300 × g). Partially photolyze BrdU with UV-B CAUTION: Take precautions when working with UV-B. Do not look into light source and avoid irradiation of skin. Work under a hood and use maximal protection from UV light if possible. 4. Turn on hand-held 8-W UV-B lamp and warm up for 5 min. 5. Remove supernatant and resuspend cells in 12 ml PBS/0.1% BSA per sample. 6. Transfer cells into 14-ml clear polypropylene tubes. 7. Place tubes directly on the filter surface of the upright standing UV-B lamp. Fix tube with clear tape or rubber band. Take care that any labeling or frosting on the tube is facing away from the lamp surface. Irradiate 5 min (approximate dose of 6.6 kJ/m2). Under these conditions, with the use of a hand-held 8-W UV-B lamp, cells are irradiated with an average intensity of ∼22 W/m2 (∼2.2 mW/cm2). Cells are irradiated in an upright standing position to avoid partial shielding by the tube cap.
8. Centrifuge cells 10 min at 300 × g, room temperature. After UV irradiation and before fixation, the cells are more sensitive and require gentle handling and pipetting. Optimally, centrifugation conditions should be the same as step 3.
Ultraviolet-Induced Detection of Halogenated Pyrimidines (UVID)
Fix cells 9. Remove supernatant. Carefully resuspend cells in 1000 µl PBS and transfer into microcentrifuge tubes. 10. Microcentrifuge 3 min at 3000 rpm. Remove supernatant and resuspend in 100 µl PBS.
7.15.2 Supplement 16
Current Protocols in Cytometry
For ethanol fixation 11a. Place tube containing the cell pellet on vortex mixer and add dropwise 700 µl 70% (v/v) high purity ethanol, −20°C. 12a. Place tube into a −20°C freezer and fix 60 min. For paraformaldehyde fixation 11b. To the tube containing the cell pellet, add 1000 µl 0.2% paraformaldehyde in PBS. 12b. Fix cells 15 min at 20°C. The temperature and time play a crucial role in avoiding overfixation of the cells, especially in laboratories without air conditioning. Alternatively, higher concentrations at lower temperatures can be used (e.g., 0.5% paraformaldehyde on ice for 10 min). Overfixation may result in decreasing sensitivities or abrogation of the BrdU detection.
13. Microcentrifuge 3 min at 3000 rpm. Remove supernatant. 14. Wash with 1000 µl PBS. 15. Microcentrifuge 3 min at 3000 rpm. Remove supernatant and resuspend in 200 µl PBS. NOTE: After this step, cells can be stored overnight at 4°C.
16. Transfer cells to a flow cytometer tube. Stain with anti-BrdU under hypotonic conditions 17. Add 1000 µl distilled water and 50 µl sodium tetraborate solution. 18. Add appropriate amount of anti-BrdU FITC antibody (10 µl for the Bu20a clone and 20 µl for the 3D4 and B44 clones). 19. (Optional) Add 20 µl 7-AAD stock solution. 20. Incubate at room temperature at least 30 min for ethanol-fixed cells or at least 60 min for paraformaldehyde-fixed cells. Cells fixed with paraformaldehyde are stable for at least 2 to 3 hr, while ethanol-fixed cells should be stable overnight.
21. Resuspend by vortexing. Analyze cells on the flow cytometer 22. Set up the flow cytometer with excitation at 488 nm, using dichroic, band-pass, and/or long-pass filters to measure green fluorescence (525 nm) and deep-red fluorescence (>650 nm or 675-nm band-pass). If additional cellular markers are detected with conjugated antibodies, discriminate those signals in the appropriate range (∼575 nm for PE). Compensate for spectral overlap with FITC and 7-AAD (UNIT 1.14). 23. Set the discriminator on 7-AAD fluorescence. 24. If possible use doublet discrimination features (e.g., integral versus peak 7-AAD fluorescence). 25. Measure at low sample flow pressure and acquire data for more than 10,000 cells if possible.
Nucleic Acid Analysis
7.15.3 Current Protocols in Cytometry
Supplement 16
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
7-AAD stock solution, 0.5 mg/ml Allow 1 mg 7-AAD (Sigma-Aldrich) to dissolve completely overnight in 2 ml PBS (APPENDIX 2A). Store at 4°C and keep protected from light. BrdU stock solution, 10 mM Dissolve 31 mg BrdU (Sigma-Aldrich) in 10 ml PBS (APPENDIX 2A). Store up to 3 months at 4°C. Keep protected from light by either using a brown bottle or wrapping in aluminum foil. Paraformaldehyde solution, 0.2% Dissolve 100 mg paraformaldehyde (Serva Biochemicals) in 50 ml PBS (APPENDIX 2A). Heat to 50°C until dissolved. Prepare fresh and store at 20°C until use. Sodium tetraborate solution Dissolve 1.9 g sodium tetraborate decahydrate (Na2B4O7⋅10H2O; Sigma-Aldrich) in 50 ml distilled water. Adjust pH with HCl to 8.0. COMMENTARY Background Information
Ultraviolet-Induced Detection of Halogenated Pyrimidines (UVID)
The use of halogenated pyrimidines such as BrdU is a valuable tool to discriminate S-phase cells from other phases of the cell cycle and to provide information about the cell cycle kinetics (Dolbeare, 1995a,b, 1996). Its detection, however, requires denaturation of DNA with harsh physical measures, such as heat or acid treatment, that often abrogate the simultaneous detection of other cellular markers. Alternative strategies employ enzymatic treatment, but suffer from specific disadvantages. Digestion of DNA with DNase I can lead to over- and underdigestion (Carayon and Bord, 1992; Penit and Vasseur, 1993; Takagi et al., 1993) and strand-break labeling of photolyzed BrdU (SBIP) can be confounded with apoptotic events (Li et al., 1994a,b, 1996). The ultraviolet-induced detection (UVID) of BrdU is the first nonenzymatic approach that is able to preserve cellular markers and to avoid enzymespecific artifacts (Hammers et al., 2000). The principle used by this approach is the selective induction of severe oxidative damage to the nucleus and chromatin, which in turn allows the binding of some anti-BrdU clones to their epitope under hypotonic conditions (with these protocols 68 mM NaCl). The mechanism is not fully understood at this time, but data suggest that the unfolding of the chromatin fibers in the hypotonic environment permits the effect. The detection is reversible as the further addition of sodium chloride or other salts will
decrease the signal. The composition of the hypotonic buffer and the pH play an important role as well and were chosen to maximize the signal-to-noise (S/N) ratio. After UV irradiation, cells are fixed with either a coagulative or a cross-linking fixative. Ethanol is the classic coagulative fixative and provides the highest S/N ratio since coagulative fixation does not interfere with chromatin decondensation. It also allows the extraction of low-molecular-weight (LMW) DNA from apoptotic nuclei, identifying them as a subG0/G1 peak. In the authors’ experience, ethanol-fixed cells seem to be more stable under the hypotonic conditions as well and might be advantageous for some antigens when longer incubation periods are necessary. Paraformaldehyde, on the other hand, is currently the most commonly used cross-linking fixative in flow cytometry. It usually provides excellent preservation of cellular markers, but higher concentrations or prolonged incubation times lead to rigid cross-linking of chromatin and inhibit the necessary structural changes that allow the detection of BrdU. For this reason the S/N ratio of paraformaldehyde-fixed cells tends to be lower than with ethanol fixation and therefore strict adherence to the protocol is essential; however, good preservation of surface antigens, low frequency of doublet formation, and a potential to retain intracellular antigens make this approach preferred for many applications.
7.15.4 Supplement 16
Current Protocols in Cytometry
104
1023
G1 mid-S
Events
DNA peak
BrdU FITC
103 102 101
G2M
doublets
S
0 0
1023 DNA integral
100
G1 S G2M 0
1023 DNA content
1023
0 DNA content
Figure 7.15.1 Left: Exclusion of doublets in the DNA peak versus DNA integral histogram. Middle: Typical bivariate cell-cycle analysis with anti-BrdU FITC and DNA counterstaining. The signal intensity of mid-S phase cells and of the G1 cells can used to calculate a signal-to-noise (S/N) ratio. Right: The corresponding DNA histogram.
Critical Parameters and Troubleshooting BrdU incorporation into DNA-synthesizing cells: The most important single parameter is the condition of the cells to be labeled. Optimal cell culture conditions and appropriate cell densities are imperative to ensure exponential growth and the optimal incorporation of BrdU analog. If cell lines are used, make sure that optimal conditions are applied at least 24 hr before the addition of the precursor. Partial photolysis of BrdU with UV-B: In the authors’ experience the photolysis of BrdU with UV is surprisingly robust as long as frequencies in the UV-B range (280 to 320 nm) are used. UV-C (254 nm) induces the detection as well, but tends to be more destructive and the S/N ratios are lower. UV-A alone cannot replace UV-B; only in combination with Hoechst dyes does UV-A sufficiently photolyze BrdU. The authors found that irradiation for 5 min with a simple 8-W UV-B hand lamp produces reliable and reproducible results. If other UV-B sources are employed, experimentally determine the irradiation time that produces the highest S/N. When the irradiation is prolonged, a decrease of the S/N ratio and damage of the cells is observed. Irradiated cells are generally more sensitive to shear stress and should be treated gently until fixation. Fixation of UV-treated cells: Fixation with cross-linking agents such as paraformaldehyde
requires well-defined conditions. The optimal margin for fixation (time and paraformaldehyde concentration) is narrow, and too extensive fixation results in a decrease in sensitivity of BrdU detection. Hypotonic treatment: The hypotonic treatment described here was successful for a wide variety of different cell types. It should be stressed, however, that cells once exposed to the hypotonic buffer are more sensitive to shear stress than are the unexposed cells. Their resuspension after centrifugation should be carried out with a vortex mixer rather than a pipetting device. In most cases, cells remain stable in the hypotonic buffer for ∼24 hr, but analysis after one hour is preferred. Ethanol fixation seems to be more stabilizing than the mild paraformaldehyde fixation. Keep in mind that the required incubation with BrdU antibody for paraformaldehyde-fixed cells is twice as long as for ethanol-fixed cells. DNA counterstaining: 7-AAD, a GC-specific intercalator, can be used to counterstain the DNA and does not interfere with the BrdU detection. Surprisingly and in contrast to aciddenatured DNA, the nonspecific intercalating dye propidium iodide (PI) strongly suppresses the anti-BrdU FITC signal and cannot be used. The use of 7-AAD has certain advantages, since this dye has a very large Stokes’ shift which allows the simultaneous measurement of other
Nucleic Acid Analysis
7.15.5 Current Protocols in Cytometry
Supplement 16
cellular markers with PE-conjugated antibodies. Anti-BrdU FITC clones: It should be noted that only a handful of anti-BrdU clones can be used with the UVID procedure. This indicates that the BrdU epitope detected by UVID is different from the one that becomes accessible after DNA denaturation. The authors recommend the Bu20a clone from Dako. Other clones such as 3D4 (PharMingen) and B44 (Becton Dickinson) can be used as an alternative.
Anticipated Results Pulse-labeling of cells (incubation with BrdU for 60 min) leads to the detection of S-phase cells with a characteristic DNA content that places them on the DNA content histograms between G1 and G1/M cells. Successful experiments then depict the typical horseshoelike pattern of the BrdU-FITC versus DNA content scatterplots or contour maps (Fig. 7.15.1).
Time Considerations BrdU incorporation requires 60 min, but 40 min can be sufficient. The UV treatment should take no longer than 5 min. Fixation with ethanol calls for a 40 to 60 min fixation period with 30 min hypotonic treatment. Fixation with paraformaldehyde takes only 15 min, but the hypotonic treatment must be increased to 60 min. Overall time, including the BrdU labeling period should be around 2.5 hr, with the paraformaldehyde protocol being slightly faster (∼15 min) than the ethanol protocol. Hands-on time is considerably shorter, because most of the time required is for incubation purposes.
Literature Cited Carayon, P. and Bord, A. 1992. Identification of DNA replicating lymphocyte subsets using a new method to label the bromo-deoxyuridine incorporated into the DNA. J. Immunol. Methods 147:225-230. Dolbeare, F. 1995a. Bromodeoxyuridine: A diagnostic tool in biology and medicine. Part I: Historical perspectives, histochemical methods and cell kinetics. Histochem. J. 27:339-369.
Dolbeare, F. 1995b. Bromodeoxyuridine: A diagnostic tool in biology and medicine. Part II: Oncology, chemotherapy and carcinogenesis. Histochem. J. 27:923-964. Dolbeare, F. 1996. Bromodeoxyuridine: A diagnostic tool in biology and medicine. Part III: Proliferation in normal, injured and diseased tissue, growth factors, differentiation, DNA replication sites and in situ hybridization. Histochem. J. 28:531-575. Hammers, H.J., Kirchner, H., and Schlenke, P. 2000. Ultraviolet-induced detection of halogenated pyrimidines: Simultaneous analysis of DNA replication and cellular markers. Cytometry 40:327335. Li, X., Traganos, F., Melamed, M.R., and Darzynkiewicz, Z. 1994a. Detection of 5-bromo-2deoxyuridine incorporated into DNA by labeling strand breaks induced by photolysis (SBIP). Int. J. Oncol. 4:1157-1161. Li, X., Traganos, F., and Darzynkiewicz, Z. 1994b. Simultaneous analysis of DNA replication and apoptosis during treatment of HL-60 cells with camptothecin and hyperthermia and mitogen stimulation of human lymphocytes. Cancer Res. 54:4289-4293. Li, X., Melamed, M.R., and Darzynkiewicz, Z. 1996. Detection of apoptosis and DNA replication by differential labeling of DNA strand breaks with fluorochromes of different color. Exp. Cell Res. 222:28-37. Penit, C. and Vasseur, F. 1993. Phenotype analysis of cycling and postcycling thymocytes: Evaluation of detection methods for BrdUrd and surface proteins. Cytometry 14:757-763. Takagi, S., McFadden, M.L., Humphreys, R.E., Woda, B.A., and Sairenji, T. 1993. Detection of 5-bromo-2-deoxyuridine (BrdUrd) incorporation with monoclonal anti-BrdUrd antibody after d eox yribo nu clease treatment. Cytometry 14:640-648.
Key Reference Hammers et al., 2000. See above. This article describes the original work and gives examples of how to combine the method with the simultaneous detection of surface markers.
Contributed by Hans-Joerg Hammers and Peter Schlenke University of Luebeck Luebeck, Germany
Ultraviolet-Induced Detection of Halogenated Pyrimidines (UVID)
7.15.6 Supplement 16
Current Protocols in Cytometry
Analysis of DNA Content and Green Fluorescent Protein Expression
UNIT 7.16
Analysis of DNA content by flow cytometry is a powerful and frequently used method to investigate cell-cycle progression. Cell-cycle effects of a specific gene can be studied by measuring cellular DNA content in cells transfected with a vector that drives expression of that gene. To distinguish between transfected and nontransfected cells on the flow cytometer, the gene of interest is generally co-transfected with a reporter gene such as green fluorescent protein (GFP), an intracellular reporter molecule widely utilized for assessment of gene transfer and expression. While wild-type GFP is not well suited for the 488-nm argon-ion laser excitation of routine flow cytometry, “red-shifted” variants of GFP (e.g., EGFP from Clontech Laboratories and Green Lantern from Life Technologies—see also UNIT 9.12) have excitation maxima around 490 nm and emit fluorescence in the green range of the spectrum. Thus, cells carrying the gene of interest that co-express red-shifted GFP can be readily identified on standard flow cytometers equipped with fluorescein filter sets and can consequently be selected by fluorescence-activated cell sorting for further analysis. As the fluorescence intensity of red-shifted GFP expression correlates with gene expression, low-expressing cells can be discriminated on the fluorescent scale from cells with high gene expression and both these subpopulations can be analyzed and sorted differentially. Simultaneous measurement of red-shifted GFP and propidium iodide (PI) fluorescence proves difficult, because PI, the most commonly used dye for cell-cycle analysis by flow cytometry, requires that cells be permeabilized for DNA staining; however, red-shifted GFP accumulates in the cytoplasm and then readily leaks out of permeabilized cells. In contrast, fixation of cells with formaldehyde in high concentration for red-shifted GFP retention leads to DNA histograms with unacceptably high coefficients of variations (CV) for the G0/1 peaks. Thus, an appropriate cell preparation technique for detection of red-shifted GFP expression and cell-cycle progression requires a delicate balance between retaining GFP fluorescence and obtaining adequate DNA histogram resolution. Alternate use of a vital DNA dye (e.g., Hoechst 33342) that enters live cells can prevent problems associated with cell membrane permeabilization. The first protocol (see Basic Protocol) describes a method for cell fixation/permeabilization for combined measurement of red-shifted GFP expression and PI DNA content. Cells are first fixed with formaldehyde followed by treatment with ethanol for cell membrane solubilization, and then stained with PI for DNA content. The second protocol (see Alternate Protocol) describes staining of unpermeabilized cells with the vital dye Hoechst 33342 for combined red-shifted GFP fluorescence and cell-cycle analysis. It is important that this unit be read in conjunction with UNIT 7.5, UNIT 9.5, and UNIT 9.12, which provide further background on nucleic acid analysis and on reporter genes.
Nucleic Acid Analysis Contributed by Ingrid Schmid and Kathleen M. Sakamoto Current Protocols in Cytometry (2001) 7.16.1-7.16.10 Copyright © 2001 by John Wiley & Sons, Inc.
7.16.1 Supplement 16
BASIC PROTOCOL
DNA CONTENT ANALYSIS IN COMBINATION WITH ASSESSMENT OF RED-SHIFTED GREEN FLUORESCENT PROTEIN EXPRESSION IN FIXED/PERMEABILIZED CELLS In this protocol, cells are fixed with 1% formaldehyde, permeabilized with 70% ethanol, and then stained with PI in the presence of ribonuclease A for DNA content. Materials Cells Phosphate buffered saline (PBS; APPENDIX 2A), ∼4°C Fixation solution, ∼4°C (see recipe) 70% ethanol, −20°C Propidium iodide (PI) working solution (see recipe) Centrifuge, 2° to 8°C 37°C water bath 70-µm nylon mesh (e.g., Fisher Scientific’s Spectra/Mesh N, Becton Dickinson’s Falcon Nylon Cell Strainers) Flow cytometer with 488-nm excitation and 530/30 and 585/42 nm band-pass filters or equivalent Additional reagents and equipment for counting cells (APPENDIX 2A) and retroviral transduction of red-shifted GFP into cells (UNIT 9.12). Fix cells with formaldehyde 1. Count cells (APPENDIX 3A). See UNIT 9.12 for retroviral transduction of red-shifted GFP into cells and see Kain (2000) for strategies for cell transfection.
2. Place ∼106 cells into a 12 × 75-mm test tube and wash once with phosphate buffered saline (PBS) by centrifuging 5 min at 300 × g, 2° to 8°C. Preferably, the PBS should be cold (4°C); however, this is not absolutely necessary as the cells will stop growing when exposed to UV light.
3. Remove supernatant by aspiration or rapid decanting and add 500 µl cold (∼4°C) PBS to the cell pellet. Mix gently. Add 500 µl cold (∼4°C) fixation solution and mix again. Incubate 1 hr at 2° to 8°C. Different concentrations of formaldehyde fixative may be needed for optimal retention of red-shifted GFP in various cell types, and for obtaining acceptable coefficients of variations on DNA histograms as discussed below (see Commentary).
Permeabilize cells with ethanol 4. Centrifuge cells 5 min at 300 × g, 2° to 8°C. Remove supernatant by aspiration or rapid decanting and wash once with cold PBS. Add 1 ml 70% ethanol at −20°C dropwise to the cell pellet with the tube sitting on a vortex mixer. Incubate cell suspension overnight at 2° to 8°C. A cell pellet may not be visible after the fixation step with formaldehyde, because fixed cells aggregate less well and therefore tend to spread out at the bottom of the tube. For this reason, careful aspiration is preferable to decanting. Vortexing cells gently during the addition of ethanol can reduce the formation of cell clumps; however, avoid extensive vortexing, because it can lead to cell disruption. Incubation times with 70% ethanol can be shorter than overnight, but should not be less than 2 hr. Analysis of DNA Content and Green Fluorescent Protein Expression
5. Centrifuge cells 5 min at 300 × g, 2° to 8°C. Remove supernatant by aspiration or rapid decanting and add 1 ml propidium iodide (PI) working solution. Incubate cell suspension 30 min in a 37°C water bath in the dark. If needed, filter samples through a 70-µm nylon mesh to remove clumps.
7.16.2 Supplement 16
Current Protocols in Cytometry
Acquire fluorescence data on the flow cytometer 6. Set up flow cytometer with 488-nm excitation. Use forward versus side scatter to look at cell size versus cell granularity. Collect red-shifted GFP fluorescence in log amplification with a 530/30 nm band-pass or similar filter, and PI fluorescence in linear amplification with a 585/42 nm band-pass or a similar filter that collects fluorescence above 590 nm. Cells that have been fixed and permeabilized will have altered light scattering properties. Usually, they have less forward-scatter and more side-scatter signal on the flow cytometer compared to untreated cells. Use orange (PI) fluorescence area and the pulse width in addition to height for discrimination of doublets. Alternatively, current DNA analysis software can provide aggregate modeling algorithms to compensate for cell clumps during DNA histogram deconvolution. Always use nonGFP-expressing cells as controls for background fluorescence. To set up photomultiplier tube voltages and compensation of spectral emission overlap between red-shifted GFP and PI, use GFP-expressing cells (i.e., GFP+) that were not stained with PI, and cells that do not express red-shifted GFP that were stained with PI alone. Accurate setup of fluorescence compensation is critical for correct correlation of GFP expression with cell-cycle status (also see UNIT 9.12). Acquire samples on the cytometer at a low flow rate for improved coefficients of variation on DNA histograms. For samples that contain few GFP+ cells, set acquisition counters to collect a minimum of 10,000 GFP+ events. In most cases, at least 10,000 events are needed to obtain accurate estimation of cell frequencies within different cell cycle phases after mathematical modeling for DNA histogram deconvolution (Ott, 1993; Shankey et al., 1993). For best results, analyze the cells on the flow cytometer as soon as possible. Samples can be held longer (e.g., for up to one week at 2° to 8°C); however, the time period will depend on the cell type and the retention of red-shifted GFP.
DNA CONTENT ANALYSIS IN COMBINATION WITH ASSESSMENT OF RED-SHIFTED GREEN FLUORESCENT PROTEIN EXPRESSION IN UNFIXED CELLS
ALTERNATE PROTOCOL
In this protocol cells are stained at 37°C with the cell-permeant DNA dye Hoechst 33342 for combined GFP and DNA content analysis. A cytometer with UV excitation is required. Additional Materials (see also Basic Protocol) Culture medium, 37°C 1 mg/ml Hoechst 33342 stock solution (see recipe) Flow cytometer with 488-nm and 325-nm excitation and 530/30 and 424/44 nm band-pass filters or equivalent Additional reagents and equipment for concurrent dead cell discrimination (UNIT 9.2; optional). Stain cells for DNA content 1. Count cells (APPENDIX 3A). See UNIT 9.12 for retroviral transduction of red-shifted GFP into cells and see Kain (2000) for strategies for cell transfection.
2. Place ∼106 cells into a 12 × 75-mm test tube and centrifuge 5 min at 300 × g, room temperature. 3. Remove supernatant by aspiration or rapid decanting and add to the cell pellet 500 µl of the same medium that was used for growing the cells, prewarmed to 37°C. Mix gently. Add 5 µl 1 mg/ml Hoechst 33342 stock solution and mix again. Incubate 45 min at 37°C. The optimal Hoechst 33342 dye concentration and staining time for different cell types vary, as dye uptake depends on cellular metabolic rates; therefore, both have to be
Nucleic Acid Analysis
7.16.3 Current Protocols in Cytometry
Supplement 16
determined empirically. In general, dye concentrations between 1 and 10 ìg/ml, and incubation times between 20 and 90 min, will produce DNA histograms with acceptable coefficients of variations (see Commentary and UNIT 7.5 for further details). Because Hoechst DNA staining is performed on unfixed cells, it is possible to use other nonvital DNA dyes (e.g., PI or 7-aminoactinomycin D), for concurrent dead cell discrimination (UNIT 9.2). Both PI and 7-aminoactinomycin D (7-AAD) are excited by the 488-nm line of an argon-ion laser and have emission spectra that can be effectively separated from the emissions of red-shifted GFP and Hoechst 33342, respectively.
Acquire fluorescence data on the flow cytometer 4. Set up flow cytometer with 488-nm excitation and with 325-nm ultraviolet excitation. Use forward versus side scatter to look at cell size versus cell granularity. Collect red-shifted GFP fluorescence in log amplification with a 530/30 nm band-pass or similar filter and Hoechst fluorescence in linear amplification with a 424/44 nm band-pass or similar filter that collects fluorescence between 400 nm and 500 nm. Always use nonGFP-expressing cells as controls for background fluorescence. Use blue (Hoechst) fluorescence width versus area in addition to height for discrimination of doublets. Alternatively, current DNA analysis software can provide aggregate modeling algorithms to compensate for cell clumps during DNA histogram deconvolution. Fluorescence emissions of red-shifted GFP and Hoechst 33342 do not overlap; therefore, compensation between their fluorescence channels is not needed. It is advisable to use cells that were not stained with Hoechst to control for effects that Hoechst dyes can have on cellular background fluorescence. Acquire samples on the cytometer at a low flow rate for improved coefficients of variation on DNA histograms. For samples that contain few GFP+ cells, set acquisition counters to collect a minimum of 10,000 GFP+ events. In most cases, at least 10,000 events are needed to obtain accurate estimation of cell frequencies within different cell cycle phases after mathematical modeling for DNA histogram deconvolution (Ott, 1993; Shankey et al., 1993).
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Fixation solution Add 2 g high-purity paraformaldehyde (Polysciences; 2% w/v final) to 100 ml PBS (APPENDIX 2A). Heat to 70°C in a fume hood until the paraformaldehyde completely dissolves (∼1 hr). Cool to room temperature and adjust pH to 7.2 with 0.1 M NaOH or 0.1 M HCl. Store at 2° to 8°C protected from light. The solution is stable for at least 1 month. Check pH periodically. Do not heat the solution above 70°C. For best results, use only very pure preparations of paraformaldehyde (i.e., electron microscopy grade from Polysciences). See Commentary for further details.
Hoechst 33342 stock solution, 1 mg/ml Dissolve 1 mg Hoechst 33342 powder (Molecular Probes) in 1 ml distilled water. Store at 2° to 8°C protected from light for up to 1 month.
Analysis of DNA Content and Green Fluorescent Protein Expression
Propidium iodide (PI) stock and working solutions Stock solution: Dissolve 1 mg PI in 1 ml PBS (APPENDIX 2A). Add 2.5 mg DNase-free ribonuclease A (Sigma-Aldrich) to the solution. Store at 2° to 8°C protected from light for up to 1 month. Working solution: Dilute PI stock solution 1:25 with PBS (APPENDIX 2A). Final concentration of PI is 40 µg/ml. Final concentration of ribonuclease A is 100 µg/ml. Make fresh before staining. Protect from light.
7.16.4 Supplement 16
Current Protocols in Cytometry
COMMENTARY Background Information Verification on the flow cytometer of cellular uptake and expression of a candidate gene after transfection requires co-transfection with a marker gene (see Kain, 2000, and UNIT 9.12). One method commonly used for assessment of the frequency of transfected cells is co-transfection with the E. coli lacZ reporter gene that encodes β-galactosidase (Fiering et al., 1991; UNIT 9.5); however, this technique requires the addition of a fluorogenic substrate and can lead to considerable cell death in the cell preparation (Klein et al., 1997). A second method involves co-transfection with genes for the expression of cell surface antigens such as CD19 (Adams et al., 1997) or CD20 (van den Heuvel and Harlow, 1993). Nevertheless, for antigen detection, the cells need to be stained with fluorescently tagged monoclonal antibodies, and since CD20 is a calcium channel (Bubien et al., 1993), its expression can lead to undesirable side effects such as alterations of cell-cycle progression (Kanzaki et al., 1995). More recently, green fluorescent protein (GFP), which shows bright, light-stimulated fluorescence independent from other co-factors and substrates, has been isolated from the jellyfish Aequorea victoria (Chalfie et al., 1994); however, wildtype GFP is excited maximally at 396 nm and has only a minor excitation peak around 475 nm. Thus, it is not well suited for routine 488nm flow cytometry excitation (Ropp et al., 1995). Generation of variants of GFP that have red-shifted excitation maxima around 490 nm and show enhanced fluorescence intensity as compared to wild-type GFP has overcome these limitations (Heim et al., 1995; Ropp et al.,1995; Cormack et al., 1996; Stauber et al., 1998). Cloning of red-shifted variants of GFP into various expression vectors suited for many different cell types has led to the use of this novel genetic reporter system with species-independent expression in a multitude of applications (UNIT 9.12). The development of variants of wild-type GFP that either have altered excitation maxima (e.g., optimized ultraviolet excitation), or show different emission characteristics (e.g., blue emission or yellow-green emission that can be separated effectively from the emission of red-shifted GFP), provides the possibility of simultaneous flow cytometric analysis of the expression of multiple genes in single cells (Ropp et al., 1996; Lybarger et al., 1998).
In order to study the effect of a gene of interest or its expression level on cell-cycle progression, it is desirable to measure redshifted GFP expression in combination with DNA content; however, the unique nature of GFP makes this approach difficult. GFP is expressed in live cells, and due to its small size of 26.9 kD, will leak out of the cytosol when cells are permeabilized with ethanol (Kalejta et al., 1997; Chu et al., 1999). In contrast, most fluorescent DNA dyes suited for high-resolution cell-cycle analysis by flow cytometry cannot pass through intact cell membranes and require permeabilization for cellular entry. To address these problems, researchers have developed membrane-localized green fluorescent protein variants that remain detectable after cell permeabilization with ethanol for combined GFP and cell-cycle analysis (Kalejta et al., 1997, 1999; Jiang and Hunter, 1998). Similarly, it has been shown that red-shifted GFP targeted to the endoplasmic reticulum can be measured in combination with DNA content in unfixed cells that are permeabilized by digitonin (Pestov et al., 1999). In contrast, Chu et al. (1999) demonstrated that it is possible to retain cytoplasmic red-shifted GFP with low-concentration formaldehyde fixation followed by alcohol treatment to achieve adequate cell-cycle profiles. Furthermore, the use of vital DNA dyes that enter unpermeabilized cells, such as the Hoechst compounds (Arndt-Jovin and Jovin, 1977) and the newly developed, deep-red fluorescing anthraquinone number 5 (DRAQ5; Smith et al., 2000), can avoid compromising GFP analysis.
Critical Parameters and Troubleshooting Two critical parameters define any protocol that can be used successfully for simultaneous measurement of cytoplasmic GFP expression and DNA content by staining with nonvital dyes. The first parameter is the need for cell fixation to retain GFP. The second is the requirement for sufficient cell membrane permeabilization to allow nonvital DNA dyes like PI access to the DNA for adequate nucleic acid staining. Conversely, intrinsic cellular factors of dye uptake and efflux are the main determinants for a successful outcome of experiments of combined assessment of GFP expression and cell cycle using vital DNA staining. Nucleic Acid Analysis
7.16.5 Current Protocols in Cytometry
Supplement 16
Analysis of DNA Content and Green Fluorescent Protein Expression
Reagents for cell fixation and permeabilization The standard fixative for flow cytometric applications is formaldehyde. It cross-links proteins and when used without detergents or alcohols for membrane solubilization does not permit sufficient dye access to DNA. The preferred reagent for formaldehyde cell fixation is a buffered solution made from paraformaldehyde powder, because some commercial formalin preparations may contain formic acid generated either during manufacturing or by exposure to excessive heat or light. Unbuffered solutions of formalin can contain methanol as a stabilizing agent. Both substances can have detrimental effects on the detection of redshifted GFP fluorescence, which, although it is stable up to pH 11.5, decreases markedly below pH 7.0 (see Clontech Laboratories manual), and which may be altered by methanol denaturation. For laboratories without access to a fume hood for the preparation of formaldehyde fixative from paraformaldehyde, high-quality electron microscopy formalin sealed into ampules is available from Polysciences. GFP is not affected by formaldehyde treatment (Chalfie et al., 1994). The main protocol of this unit (see Basic Protocol), first described by Chu et al. (1999), uses a 1-hr fixation with buffered 1% formaldehyde solution. One percent formaldehyde fixative maintains redshifted GFP fluorescence after ethanol cellmembrane solubilization. Nevertheless, its concentration is low enough to avoid excessive cross-linking of chromosomal proteins and DNA, which has been shown to adversely affect the resolution of DNA histograms (Schmid et al., 1991). For this reason, some commercial fixation and permeabilization reagent systems (e.g., Permeafix from Ortho Diagnostics Systems and Fix and Perm from Caltag Laboratories) are unsuitable for combined analysis of red-shifted GFP expression and cell cycle because their solutions contain cross-linking fixatives at high concentrations. These systems may be able to retain red-shifted GFP fluorescence, but are not optimized for DNA staining and consequently will lead to unacceptably broad peaks on DNA histograms (Kalejta et al., 1997; Schmid, 2000). The procedure outlined in the Basic Protocol uses cell membrane solubilization with 70% ethanol instead of detergent treatment. Because alcohols coagulate proteins and solubilize lipids, they simultaneously fix and permeabilize cellular membranes, as opposed to detergents,
which cannot provide the added fixation that may aid in GFP retention. Vital DNA staining In contrast to nonvital nucleic acid stains such as PI, the benzimidazole compounds Hoechst 33258 and Hoechst 33342 are DNAspecific dyes that are able to cross intact cell membranes (Arndt-Jovin and Jovin, 1977; UNIT 7.5). Of these two dyes, Hoechst 33342 has been used most commonly for determination of DNA content in living cells by flow cytometry (Loken, 1980; Shapiro, 1981; Crissman et al., 1990). Hoechst 33342 is more hydrophobic than 33258, and therefore crosses cell membranes more easily. Moreover, it tends to exert less toxicity than Hoechst 33358, although its general toxicity is highly cell-type dependent (Fried et al., 1982). Hoechst dyes require ultraviolet (UV) excitation, which until recently was available only from large argon-ion lasers supplied on cell sorters. The commercial availability of benchtop flow cytometers equipped with air-cooled helium-cadmium UV-emitting lasers has made their use more practical. Stoichiometric vital Hoechst 33342 DNA staining generally requires dye concentrations between 5 and 10 µM, with the optimal concentration dependent on differences in the balance between cellular dye uptake and/or efflux. As optimal Hoechst staining concentrations are cytotoxic for most cell types (Loken, 1980; Fried et al., 1982), selection and recovery of cells with intact reproductive capacity after cell sorting can be difficult (Loken, 1980; Shapiro, 1995). Because staining with Hoechst 33342 can be performed without cell fixation and permeabilization, this dye readily permits concurrent staining of dead cells which have lost membrane integrity with nonvital DNA dyes such as PI or 7-AAD (for further details see UNIT 9.2) and makes the Alternate Protocol particularly suited for use with GFP-containing cells that show loss of fluorescence of the marker gene after cell membrane permeabilization; however, the precision of DNA histograms from intact cells is not as good as the high-resolution cell-cycle profiles that can be achieved with cell preparations that have been permeabilized. In addition, some cell types have been shown to be refractive to vital DNA staining with Hoechst dyes (Crissman et al., 1990). Moreover, it has been reported that live staining with Hoechst dyes can result in redistribution of DNA-binding proteins, possibly due to alterations in chromatin structure caused by their
7.16.6 Supplement 16
Current Protocols in Cytometry
interactions with DNA (Baumann and Reyes, 1999). Thus, experiments that use Hoechst compounds to study the expression of proteins related to cell cycle by combined GFP and DNA content analysis must be carefully monitored by fluorescence microscopy; otherwise, incorrect protein localization may occur.
cellular GFP fluorescence. Thus, it is advisable to confirm the location of GFP fluorescence with fluorescence microscopy. This is an important confirmation step for proteins with known localization, as flow cytometry cannot offer this information.
Anticipated Results Troubleshooting The Basic Protocol has successfully been used on murine leukemic cells, but may not be appropriate for all cell types and for all GFP applications. To determine the effect of cell fixation and permeabilization on red-shifted GFP expression levels, use as controls untreated cells transfected with a red-shifted GFPcontaining plasmid. If red-shifted GFP fluorescence is lost or diminished extensively after cell membrane permeabilization, try to increase the concentration of formaldehyde in the fixation solution gradually. If this is unsuccessful or leads to inadequate PI DNA histograms, try to apply the protocol as outlined in UNIT 9.12 to the cell preparation. To avoid problems that can occur with fixation/permeabilization of GFP-containing cells, use vital DNA dyes such as the Hoechst dyes (Arndt-Jovin and Jovin, 1977; Loken, 1980; Shapiro, 1981) for simultaneous detection of red-shifted GFP and DNA content as described in the Alternate Protocol; however, Hoechst dyes require UV excitation, and as dye uptake and retention vary among cell types, staining conditions have to be optimized individually. Even with individual optimization, certain cell types will yield unsatisfactory DNA histograms, or vital Hoechst DNA staining may not be attainable (Crissman et al., 1990), although on occasion, improvement in the resolution of Hoechst 33342 DNA histograms through addition of membrane potential–modifying compounds such as DiOC5(3) has been accomplished (Crissman et al., 1990; UNIT 7.5). In these difficult cases, the new vital DNA dye DRAQ5 (Smith et al., 2000) may offer a valuable alternative, particularly because DRAQ5 can be excited with 488-nm argon-ion lasers available on standard flow cytometers. Finally, it may be necessary to modify the transfection/transduction procedure and utilize GFP variants that are targeted to cellular membranes (Kalejta et al., 1997, 1999; Jiang and Hunter, 1998; Pestov et al., 1999). Furthermore, any fixation/permeabilization procedure as well as overexpression of the GFP fusion protein (Baumann and Reyes, 1999) may lead to a redistribution of the localization of intra-
Figure 7.16.1 shows the application of the Alternate Protocol to the combined analysis of EGFP and DNA content in 293 T cells. Figure 7.16.1A shows the background fluorescence of nontransduced cells and Figure 7.16.1B shows EGFP expression in cells that were retrovirally transduced with an EGFP-containing vector. Dead cells which have lost membrane integrity are stained with PI. It is critical to set up the flow cytometer with appropriate compensation between green (FL-1) EGFP fluorescence and orange (FL-2) PI fluorescence; for instance, in the example shown here, compensation of FL-2 out of FL-1 is set to 0.4% and compensation of FL-1 out of FL-2 to 18.8%, respectively. Note that regions instead of quadrant markers are used to accurately determine the frequency of live EGFP+ cells. Markers as shown in Figure 7.16.1A and B, set according to the autofluorescence levels in the control sample, lead to underestimation of EGFP+ cells because they do not include EGFPdim+ cells (i.e., with lowlevel EGFP expression) in the positive quadrant. As a result, in the sample shown in panel B, the frequency of 39% EGFP+ cells falling within R1 would be reduced to 29% of cells contained in the lower right quadrant. After the addition of Hoechst 33342 and incubation at 37°C for 45 min, there are slightly more dead cells in the nonEGFP-containing and EGFP-containing sample and both show a small increase in background fluorescence (Fig. 7.16.1C and D) compared to cells that were not stained with Hoechst 33342 (Fig. 7.16.1A and B); however, the frequency and the relative mean fluorescence intensity (RFL) of EGFP+ cells as shown in panel D (%EGFP+ = 42, RFL = 207) compare well to cells displayed in panel B (%EGFP+ = 39, RFL = 201). By setting a gate on EGFP-negative cells versus EGFP+ cells (Fig. 7.16.1E) their DNA content can then be compared (Fig. 7.16.1F: %G1 = 52, %S = 36, %G2+M = 12; Fig. 7.16.1G: %G1 = 49, %S = 39, %G2+M = 12). Results obtained from other experiments may differ markedly from results shown here, as the frequency of GFP+ cells and their expression level vary considerably depending on cell type, experimental conditions, GFP variant, transfection plasmid,
Nucleic Acid Analysis
7.16.7 Current Protocols in Cytometry
Supplement 16
104
104
A
B 103
102
PI
PI
103 R1
102
R1
101
101 100 100
101
102
103
100 100
104
101
EGFP 104
102 EGFP
103
104
C
D 103
102
PI
R1
R1
101
101
102 EGFP
103
100 100
104
150
101
1000
E
600 R2
0
R3 150
400
103
104
F
101
102 EGFP
103
104
200 400 600 800 1000 Hoechst 33342 EGFP-positive
Counts
200
0
Analysis of DNA Content and Green Fluorescent Protein Expression
102 EGFP
0
800
101
EGFP-negative
Counts
100 100
Hoechst 33342
102
G
0
PI
103
0 100
104
200 400 600 800 1000 Hoechst 33342
Figure 7.16.1 Combined analysis of EGFP expression and DNA content in 293 T cells (kindly provided by Dr. K. Morizono) using vital DNA staining with Hoechst 33342 as described (see Alternate Protocol) with 2 µg/ml PI added for dead cell discrimination. EGFP fluorescence versus PI fluorescence of nontransduced cells, not stained with Hoechst 33342 (A) or stained with Hoechst 33342 (C). EGFP fluorescence versus PI fluorescence of cells transduced with an EGFP-containing vector, not stained with Hoechst 33342 (B) or stained with Hoechst 33342 (D). Dot plot (E) of EGFP fluorescence versus Hoechst 33342 fluorescence area displays the same data file as shown in (D) and is gated on PI-negative live cells and on Hoechst fluorescence width versus area for doublet discrimination. Histogram of Hoechst fluorescence of EGFP-negative live cells contained within gate R2 (F) as shown in (E). Histogram of Hoechst fluorescence of EGFP-positive live cells contained within gate R3 (G) as shown in (E).
7.16.8 Supplement 16
Current Protocols in Cytometry
promoter sequences, and flow cytometry instrumentation.
Heim, R., Cubitt, A.B., and Tsien, R.Y. 1995. Improved green fluorescence. Nature 373:663-664.
Time Considerations
Jiang, W. and Hunter, T. 1998. Analysis of cell-cycle profiles in transfected cells using a membrane targeted GFP. BioTechniques 24:348-354.
The Basic Protocol will take ∼2 hr on the first day, and 1 hr the following day for sample preparation. If the incubation with 70% ethanol is shortened to 2 hr instead of overnight, the procedure can be completed in ∼5 hr on the same day. Preparation of samples for the Alternate Protocol will take ∼20 min up to 90 min, based on the optimal incubation time for Hoechst 33342 staining. Flow cytometer setup time will differ depending on the instrumentation used; furthermore, the acquisition time per sample can vary widely, because it is contingent on the frequency of GFP+ cells in the cell preparation.
Literature Cited Adams, P.D., Lopez, P., Sellers, W.R., and Kaelin, W.G. Jr. 1997. Fluorescent-activated cell sorting of transfected cells. Methods Enzymol. 283:5972. Arndt-Jovin, D.J. and Jovin, T.M. 1977. Analysis and sorting of living cells according to deoxyribonucleic acid content. J. Histochem. Cytochem. 25:585-589. Baumann, C.T. and Reyes, J.C. 1999. Tracking components of the transcription apparatus in living cells. Methods 19:353-361. Bubien, J.K., Zhou, L.-J., Bell, P.D., Frizzell, R.A., and Tedder, T.F. 1993. Transfection of the CD20 cell surface molecule into ectopic cell types generates a Ca2+ conductance found constitutively in B lymphocytes. J. Cell Biol. 121:1121-1132. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W.W., and Prasher, D.C. 1994. Green fluorescent protein as a marker for gene expression. Science 263:802805. Chu, Y.-W., Wang, R., Schmid, I., and Sakamoto, K.M. 1999. Analysis with flow cytometry of green fluorescent protein expression in leukemic cells. Cytometry 36:333-339. Cormack, B.P., Valdivia, R.H., and Falkow, S. 1996. FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173:33-38. Crissman, H.A., Hofland, M.H., Stevenson, A.P., Wilder, M.E., and Tobey, R.A. 1990. Supravital cell staining with Hoechst 33342 and DiOC5(3). Methods Cell Biol. 33:89-95.
Kain, S.R. 2000. Flow cytometric analysis of GFP expression in mammalian cells. In In Living Color: Protocols in Flow Cytometry and Cell Sorting (R.A. Diamond and S. DeMaggio, eds.) pp. 199-226. Springer-Verlag, Berlin. Kalejta, R.F., Shenk, T., and Beavis, A.J. 1997. Use of a membrane-localized green fluorescent protein allows simultaneous identification of transfected cells and cell cycle analysis by flow cytometry. Cytometry 29:286-291. Kalejta, R.F., Brideau, A.D., Banfield, B.W., and Beavis, A.J. 1999. An integral membrane green fluorescent protein marker, Us9-GFP, is quantitatively retained in cells during propidium iodide-based cell cycle analysis by flow cytometry. Exp. Cell Res. 248:322-328. Kanzaki, M., Shibata, H., Mogami, H., and Kojima, I. 1995. Expression of calcium-permeable cation channel CD20 accelerates progression through the G1 phase in Balb/c 3T3 cells. J. Biol. Chem. 270:13099-13104. Klein, D., Indraccolo, S., von Rombs, K., Amadori, A., Salmons, B., and Günzburg, W.H. 1997. Rapid indentification of viable retrovirustransduced cells using the green fluorescent protein as a marker. Gene Ther. 4:1256-1260. Loken, M.R. 1980. Simultaneous quantitation of Hoechst 33342 and immunofluorescence on viable cells using a fluorescence activated cell sorter. Cytometry 1:136-142. Lybarger, L., Dempsey, D., Patterson, G.H., Piston, D.W., Kain, S.R., and Chervenak, R. 1998. Dualcolor flow cytometric detection of fluorescent proteins using single-laser (488-nm) excitation. Cytometry 31:147-152. Ott, L. 1993. An Introduction to Statistical Methods and Data Analysis, 4th Edition. Duxbury Press, Belmont, Ca. Pestov, D.G., Polonskaia, M., and Lau, L.F. 1999. Flow cytometric analysis of the cell cycle in transfected cells without cell fixation. BioTechniques 26:102-106. Ropp, J.D., Donahue, C.J., Wolfgang-Kimball, D., Hooley, J.J., Chin, J.Y.W., Hoffman, R.A., Cuthbertson, R.A., and Bauer, K.D. 1995. Aequorea green fluorescent protein analysis by flow cytometry. Cytometry 21:309-317.
Fiering, S.N., Roederer, M., Nolan, G.P., Micklem, D.R., Parks, D.R., and Herzenberg, L.A. 1991. Improved FACS-Gal: Flow cytometric analysis and sorting of viable eukaryotic cells expressing reporter gene constructs. Cytometry 12:291-301.
Ropp, J.D., Donahue, C.J., Wolfgang-Kimball, D., Hooley, J.J., Chin, J.Y.W., Cuthbertson, R.A., and Bauer, K.D. 1996. Aequorea green fluorescent protein: Simultaneous analysis of wild-type and blue-fluorescing mutant by flow cytometry. Cytometry 24:284-288.
Fried, J., Doblin, J., Takamoto, S., Perez, A., Hansen, H., and Clarkson, B. 1982. Effects of Hoechst 33342 on survival and growth of two tumor cell lines and on hematopoietically normal bone marrow cells. Cytometry 3:42-47.
Schmid, I. 2000. Intracellular antigen detection by flow cytometry. In In Living Color: Protocols in Flow Cytometry and Cell Sorting (R.A. Diamond and S. DeMaggio, eds.) pp. 524-531. Springer-Verlag, Berlin.
Nucleic Acid Analysis
7.16.9 Current Protocols in Cytometry
Supplement 16
Schmid, I., Uittenbogaart, C.H., and Giorgi, J.V. 1991. A gentle fixation and permeabilization method for combined cell surface and intracellular staining with improved precision in DNA quantification. Cytometry 12:279-285.
Stauber, R.H., Horie, K., Carney, P., Hudson, E.A., Tarasova, N.I., Gaitanaris, G.A., and Pavlakis, G.N. 1998. Development and applications of enhanced green fluorescent protein mutants. BioTechniques 24:462-471.
Shankey, V.T., Rabinovitch, P.S., Bagwell, B., Bauer, K.D., Duque, R.E., Hedley, D.W., Mayall, B.H., and Wheeless, L. 1993. Guidelines for implementation of clinical DNA cytometry. Cytometry 14:472-477.
van den Heuvel, S. and Harlow, E. 1993. Distinct roles for cyclin-dependent kinases in cell cycle control. Science 262:2050-2054.
Shapiro, H.M. 1981. Flow cytometric estimation of DNA and RNA content in intact cells stained with Hoechst 33342 and Pyronin Y. Cytometry 2:143-150.
Chu et al., 1999. See above.
Shapiro, H.M. 1995. Parameters and probes. In Practical Flow Cytometry, pp. 254-255. John Wiley & Sons, New York. Smith, P.J., Blunt, N., Wiltshire, M., Hoy, T., Teesdale-Spittle, P., Craven, M.R., Watson, J.V., Amos, W.B., Errington, R.J., and Patterson, L.H. 2000. Characteristics of a novel deep red/infrared fluorescent cell-permeant DNA probe, DRAQ5, in intact human cells analyzed by flow cytometry, confocal and multiphoton microscopy. Cytometry 40:280-291.
Key Reference Describes the procedure presented in the Basic Protocol.
Contributed by Ingrid Schmid and Kathleen M. Sakamoto UCLA School of Medicine Los Angeles, California
Analysis of DNA Content and Green Fluorescent Protein Expression
7.16.10 Supplement 16
Current Protocols in Cytometry
Analysis of Viral Infection and Viral and Cellular DNA and Proteins by Flow Cytometry Viruses are obligate intracellular parasites that require the host cell replication, transcription, and translation machinery for the production of new viral progeny. They come in many sizes and shapes, and have a wide range of hosts, which include bacteria, plants, and animals. Each viral group provides a unique series of viral-cellular interactions, leading to unique cellular pathologies and/or replication of the virus. As knowledge of viruses has increased so has understanding of the diversity and types of virus-cell interactions that exist. Studies have provided insight not only into virus replication and control of host functions but also into cellular regulatory events, since the virus must utilize the host metabolic systems. A considerable amount of information has been gathered regarding cellular functions such as eukaryotic replication, transcription, translation, and regulation of these events. The ability to assay viral events in each cell in a population provides information that is useful in developing a knowledge base to understand viral replication and pathologies. Flow and static cytometry coupled with other technologies provides the opportunity to monitor the infection process on a single-cell basis.
UNIT 7.17
BASIC PROTOCOL
Flow and static cytometry monitor the infection and quantitate the viral-cellular events, on intact, single, fixed cells, for DNA, RNA, and proteins. This protocol was developed for the study of mammalian viruses, utilizing SV40 as the prototype (Jacobberger et al., 1986; Lehman et al., 1988; Laffin and Lehman, 1994), and to correlate viral proteins (T antigen and the VPs), cellular and viral DNA, and cell cycle proteins, such as the cyclins, PCNA, Cdks, Cdk inhibitors, pRb and p53, as infection proceeded (Friedrich et al., 1993; Whalen et al., 1999; Lehman et al., 2000). Cells are cultured, harvested using trypsinEDTA to obtain a single-cell suspension, and fixed. The cells are stained with a suitable monoclonal or polyclonal primary antibody, then labeled with a FITC-conjugated secondary antibody to the protein of interest and with propidium iodide (PI) for DNA. Positive and negative controls should always be included. A wide range of cells have been analyzed (see below). CAUTION: When working with human blood, cells, or infectious agents, appropriate biosafety practices must be followed. Materials Cells of interest, monolayer-grown or suspension-grown PBS (see recipe) 10× trypsin-EDTA (Life Technologies), 37°C Wash solution (see recipe) Methanol, ice cold Primary antibody (see recipe) Secondary antibody (see recipe) RNase A solution (see recipe) Propidium iodide (PI; see recipe) Inverted microscope to view monolayer cells and the trypsin-EDTA dispersion 37°C, 5% CO2 incubator 1.0- and 1.5-ml microcentrifuge tubes Variable microcentrifuge (e.g., Fisher Model 95A) Contributed by John M. Lehman, Thomas D. Friedrich, and Judith Laffin Current Protocols in Cytometry (2001) 7.17.1-7.17.9 Copyright © 2001 by John Wiley & Sons, Inc.
Nucleic Acid Analysis
7.17.1 Supplement 17
∼50-µm pore size nylon mesh (Nytex) Flow cytometer with 488-nm excitation and filters for collection of green (535-nm band-pass filter) and red (640-nm long-pass filter) fluorescence Harvest cells 1. Remove growth medium from cells and wash cells with PBS. A wide range of cells have been analyzed, including monkey kidney cells (CV-1), human diploid fibroblasts (IMR-90, WI38, etc.), mouse fibroblasts, Syrian hamster fibroblasts, Chinese hamster fibroblasts, and numerous human and animal tumor cell lines. These cells have mostly been monolayer-grown cells; however, suspension-grown cells have also been utilized.
2. Add 1 to 5 ml prewarmed 1× trypsin-EDTA in PBS to cells. Incubate 30 to 60 sec and decant solution. To prevent cell loss and clumping, do not leave cells in trypsin-EDTA any longer than necessary.
3. Place cells in the 37°C, 5% CO2 incubator until they detach; observe with an inverted microscope. Add 1.0 ml wash solution and shake gently to resuspend cells. 4. Transfer cells to a 1.5-ml microcentrifuge tube. Centrifuge in a variable microcentrifuge 15 sec at 5000 rpm, room temperature. Remove supernatant and resuspend the pellet in 1 ml PBS. Repeat centrifugation. Fix cells 5. Resuspend pellet in 20 to 100 µl PBS and add 900 µl ice-cold (−20°C) methanol. Different fixatives can be utilized. Methanol was chosen for the SV40 large T antigen because it maintains antigenicity, results in minimal clumping of the cells during the staining protocol, and provides a strong fluorescence signal from the stained cells.
6. Count cells and adjust concentration to 1 × 106 cells/ml with PBS. Store at −20°C until ready to stain. The length of storage at −20°C may be variable depending on the viral antigen studied. The SV40 large T antigen maintains antigenicity for up to 1 year, and the fluorescence signals generated are comparable over this period.
Stain with primary antibody 7. Microcentrifuge the stored cells in 1.5-ml tubes 15 sec at 5000 rpm, room temperature, to remove the fixative and wash once with 1 ml PBS. Repeat centrifugation. All centrifugation steps are similar for this protocol.
8. Add 0.5 ml primary antibody and suspend cells by gently vortexing. Vortexing is an important step and may be variable depending upon cell type. Gentle vortexing provides a single-cell suspension with minimal clumping and loss of cells.
9. Incubate ≥30 min at 37°C. The time can vary from as short as 15 min to as long as overnight at 4°C. Each antigen and antibody may have different requirements, so staining time and antibody concentration must be experimentally determined. With the SV40 T antigen, the antibody was incubated 30 min at room temperature in most experiments.
Analysis of Viral Infection by Flow Cytometry
10. Wash in PBS to remove unbound antibody. Microcentrifuge 15 sec at 5000 rpm, room temperature, and resuspend pellet in wash solution with gentle vortexing. Repeat centrifugation and resuspension step at least two times.
7.17.2 Supplement 17
Current Protocols in Cytometry
Stain with secondary antibody 11. Add 0.5 ml secondary antibody to each tube and vortex gently. 12. Incubate 30 to 60 min at 37°C, or until a detectable signal is obtained. Depending on the amount of antigen per cell and concentration of antibody, the time required to obtain a detectable fluorescence signal may vary.
13. Wash to remove excess antibody. Repeat centrifugation, wash, and vortex as described in step 10. Numerous secondary antibodies from a number of companies have been utilized with good to excellent results. Selection of secondary antibody is dependent upon sensitivity, specificity, stability, and signal intensity detected.
Stain DNA 14. Add 0.5 ml RNase A solution and 0.5 ml PI solution. Vortex gently. Keep cells at 4°C in the dark and analyze within 4 hr. Just before running on the flow cytometer, filter cells through 50-µm nylon mesh to minimize clumps and provide a single-cell population. Cells are stained with PI to detect the quantity of DNA per cell for characterization of cell-cycle changes occurring with infection. This also provides an opportunity to measure the increase in viral DNA as infection proceeds, since both cellular and viral DNA will intercalate the dye between the bases of DNA.
Analyze cells 15. Collect listmode data for green fluorescence (FITC) with a 535-nm band-pass filter and for red fluorescence (PI) with a 640-nm long-pass filter. Gate on light scatter versus red (DNA) fluorescence to eliminate noncellular debris. Then analyze these cells using red area versus red peak to select single cells. Display these gated cells in a third dual histogram of red area versus green area (T antigen; see Fig. 7.17.1). Collect 10,000 to 50,000 cells for each data point. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Phosphate buffered saline without Ca2+ and Mg2+ (PBS) 8.0 g NaCl 0.2 g KCl 1.2 g Na2HPO4 0.2 g KH2PO4 Dissolve in 1 liter distilled water, adjust pH to 7.4, filter sterilize through a 0.22-µm filter, and store ≤1 year at room temperature. Primary antibody (1° Ab) Many of the primary antibodies whether monoclonal or polyclonal may be obtained from the American Type Culture Collection, Oncogene Research Products, and Santa Cruz Biotechnology, Inc. A large number of antibodies have been prepared to many viruses and may be obtained from the above listed companies or directly from the investigators that have published on the antibody. The antibodies used were stable 4 to 5 years stored in small aliquots at −20°C. Nucleic Acid Analysis
7.17.3 Current Protocols in Cytometry
Supplement 17
Propidium iodide (PI) solution 10.0 mg propidium iodide (Sigma) 100 ml PBS (see recipe) 20 µl Triton X-100 0.1 g sodium azide This solution has been stored 2 to 3 years at 4°C. Ribonuclease A (RNase A) solution 0.1 g RNase A (Sigma) 100 ml PBS (see recipe) 20 µl Triton X-100 0.1 g sodium azide Boil 1 hr and store ≤1 year at 4°C. Secondary antibody (2° Ab) These antibodies may be purchased from a number of companies and were generally fluorescein isothiocyanate (FITC) labeled. Recently, Alexa Fluor 488 (Molecular Probes) conjugated to secondary antibodies has been used with excellent results. Wash solution 100 ml normal goat serum 900 ml PBS (see recipe) 200 µl Triton X-100 1.0 g sodium azide Heat-inactivate goat serum 60 min at 56°C, filter sterilize through a 0.22-µm filter, and store 6 months at 4°C. COMMENTARY Background Information
Analysis of Viral Infection by Flow Cytometry
The papovaviruses, which include the papillomaviruses and polyomaviruses (Cole, 1996), are small DNA viruses. The polyomavirus virion, SV40, is surrounded by a protein coat composed of the viral proteins VP1, VP2, and VP3, which cover the double-stranded (ds) closed circular DNA (5243 bp) associated with the cellular histones H2A, H2B, H3, and H4 in a chromatin complex. The SV40 genome codes for at least four other proteins. Two major proteins are the T antigens, small t antigen and large T antigen, which are important in virus replication and transformation. Recently, another T antigen, a tiny t, has been described. A fourth late protein, agnoprotein, is involved in the spread of the virus between cells. These viruses have two major options when they infect a cell: (1) cell lysis with production of progeny viruses (permissive infection) and (2) transformation (nonpermissive infection). Because the majority of viral events occur in the nucleus, they are classified as nuclear viruses. The replication cycle (permissive or lytic infection) involves the following steps: (1) adsorption of the virus to the cell; (2) virion entry
and uncoating (nucleus); (3) transcription and translation of early viral mRNA (T antigens); (4) replication of viral DNA; (5) transcription and translation of late mRNA (coat proteins); and (6) assembly of the viral DNA and late proteins into the virion. Transformation (nonpermissive) includes: (1) steps 1 to 3 of the replication cycle; (2) absence of viral DNA replication; (3) integration of all or a portion of the viral genome into the host DNA; (4) failure to transcribe and translate the late mRNA and proteins; and (5) expression of the transformed phenotype by the infected cells. The transformed cell phenotype displays the following characteristics: increased saturation density, growth in soft agar, increased cloning efficiency, loss of contact inhibition, reduced growth factor requirements, cell membrane changes, chromosomal changes, immortality, and tumorigenicity. A considerable amount of the information regarding the transformed cell was first identified with this viral-transforming model and has provided the basis for defining and providing assay systems to study various models of neoplasia.
7.17.4 Supplement 17
Current Protocols in Cytometry
The viral components and products may be studied directly after disruption of the virion and/or infected cell to allow the assay and characterization of the viral DNA, RNA, and proteins. A number of procedures have been utilized to study viruses and their effects on the host cell. These include Southern, northern, and western gel analysis, immunoprecipitation, and analytical ultracentrifugation, to name just a few. However, these procedures require disruption of the cell, precluding the analysis of viral events at the single-cell level as the infection proceeds. An area of particular interest in the viral lytic cycle is the production of specific viral proteins and DNA content changes as infection proceeds (Lehman et al., 1988, 1993; Friedrich et al., 1994). The availability of monoclonal and/or polyclonal antibodies directed to both viral and cellular proteins offers the ability to assay a specific molecule in the intact (fixed) cell and the opportunity to correlate viral events with cellular events such as progression through the cell cycle. These studies have provided insight into how the infectious agent utilizes its genetic information and products (protein and RNA) in conjunction with hostcell proteins to initiate and complete viral replication or initiate and maintain the transformed cell. Interest centered on the viral proteins synthesized during the infection and the DNA content changes (both cellular and viral) that occurred as infection proceeds. Cells for these studies were derived from monolayer cell cultures of infected, mock-infected, and transformed populations. The protocol has been used for many other viral agents, such as polyoma virus, cytomegalovirus (CMV), and human immunodeficiency virus (HIV), as well as any viral protein to which a monoclonal or polyclonal antibody is available (Elmendorf et al., 1988; McSharry et al., 1990). Both the lytic and the transforming cell models of SV40 infection have been studied with this technique. Knowledge of the cell model system and how to obtain a single cell suspension is critical. Another important aspect of the procedure is the availability of specific antibody to the viral proteins. SV40 virus has available numerous polyclonal and monoclonal antibodies made to the epitopes of the viral proteins.
Critical Parameters and Troubleshooting This procedure has been utilized in experimental cell-virus model systems, both mono-
layer and suspension culture. Clinical samples—including blood, and bladder and lung washes—have been assayed for specific viral proteins utilizing this technique. Archival material (paraffin-embedded) may be utilized if the antigen (viral) is stable after fixation, paraffin embedding, and removal of paraffin. Solid tissue may also be utilized if a single-cell suspension can be prepared. Clumping is a major concern in all flow cytometry assays and considerable effort may be necessary to obtain a viable single-cell preparation for a particular virus-cell system. The major events leading to clumping are associated with the initial trypsin-EDTA treatment to detach the cells from the substrate. Therefore, careful attention should be addressed to the length of trypsin treatment, temperature, and concentration of the trypsinEDTA. Addition of serum to inactivate the trypsin-EDTA and immediate fixation of the cells will minimize clumping. Once fixed, cells may be retained as a single-cell suspension by paying particular attention to resuspension of the pellet and the centrifugation step. The investigator who has the most experience with the particular virus-cell system under study best determines these steps. Each cell type may present a new set of problems that may need to be addressed with an experiment to determine the optimal conditions. To define the fixation and antibody dilutions required, the assay is performed on cells grown on coverslips, which allows monitoring of the staining by immunofluorescence microscopy. This also provides an opportunity to determine the specificity, location of antigen, and optimal dilution of antibody, both primary and secondary, for the antigen of interest. The use of coverslips and fluorescent microscopy provides an initial dilution, which is used for the flow cytometry assay. Generally, for flow cytometry, the antibodies can be diluted further (10- to 100-fold), suggesting that the assay is more sensitive. This protocol requires that the antibodies be specific and saturating. Therefore, the cell number (antigen) must be constant so that if most cells contain the antigen, antibody is available to react with all epitopes within or on the cells. This may not be an issue if only a small number of cells in the population are antigen positive. The authors have always kept the cell number constant; usually the ideal concentration is 1-1.5 × 106 cells/ml. This concentration allows for cell loss during the staining protocol and for collection of 104 to 105 cells for flow cytometry analysis. The primary
Nucleic Acid Analysis
7.17.5 Current Protocols in Cytometry
Supplement 17
12 hr
36 hr
16 hr
52 hr
T antigen
1000
1 0
127
0
127
0
127
0
127
DNA content Figure 7.17.1 DNA content (linear scale) and T antigen (log scale) of CV-1 cells infected with SV40. Cells were harvested 12, 16, 36, and 52 hr postinfection. Cells above the line drawn across the histograms are infected cells expressing T antigen; they were stained with monoclonal antibody to T antigen (PAb101) followed by Alexa Fluor 488 conjugated to secondary antibody. Cells below the line are uninfected cells. These data demonstrate the progress of infection with (1) the increase in the number of T antigen–positive cells; (2) the increase in quantity of T antigen per cell; and (3) the increase in DNA content per cell. At 12 hr postinfection (pi), the majority of the population is G1 and uninfected. By 16 hr, the cells are in G1, but the majority are T-antigen positive. As the infection proceeds, the DNA reaches a total of 10 to 12C with accumulation of replicating viral DNA and increase in cellular DNA. This protocol allows the detection of infected and uninfected cells, and the determination of quantity of T antigen/cell, DNA content/cell, and progress of the infection on a single-cell basis. Multiple parameters would allow characterization of the replication, transcription, and translation of virus per cell.
Analysis of Viral Infection by Flow Cytometry
antibodies are generally monoclonal and usually prepared from either cell-culture fluid of hybridomas or mouse ascites fluid. Following a 50% cut with ammonium sulfate, the antibodies are desalted and/or purified on a protein A Sepharose column and stored at −20°C in the presence of 0.1% sodium azide. This preparation should not be contaminated with proteases and/or DNases, which will lower the antibody titer and decrease or destroy the DNA staining. A number of secondary antibodies from different companies have been utilized, requiring some experimentation to determine the optimal antibody and concentration. These studies have primarily utilized indirect immunofluorescence, but the authors have also in some studies used a direct immunofluorescence assay (primary antibody conjugated to a fluorochrome). The utilization of the direct procedure will (1) shorten the procedure without loss of specificity; (2) minimize the number of steps in the protocol; and (3) reduce cell loss since the procedure requires fewer steps. The majority of the 2° antibodies have been labeled with fluorescein isothiocyanate (FITC). However, recently, Molecular Probes developed secondary antibody conjugated to Alexa Fluor 488 dye, which provides an increased and stable signal compared to FITC and therefore requires lower concentration of antibody. Furthermore, it is also important to include a positive and negative cell population (control)
in the assay to allow monitoring of the staining protocol of each experiment. This provides a control for the specificity, saturation of the antibody, and reproducibility of average fluorescence value of the control cells. Inclusion of these controls (positive and negative cells) provides a standard to which many experiments, over time, may be compared. If the values of these cells vary, this allows the determination of which step in the protocol is responsible. The authors have always included positive and negative cells as a control for each experiment run on the flow cytometer. The size of the cells should be considered since the size may change during the progress of the infection. Size can be discerned from the scatter parameter, which is routinely included with the saved listmode data. This is important information to know so that size and background fluorescence may be taken into consideration when calculating the results of the experiment. CV-1 cells infected with SV40 increase in DNA, protein, size, etc., as the infection proceeds. It is important to determine the amount of autofluorescence and nonspecific binding of the antibodies by staining both positive and negative cells. The following is performed on positive/negative cells: (1) one series stained with 1° antibody and PI, but no secondary antibody; (2) one series with secondary antibody and PI, but no 1°; (3) one series with PI
7.17.6 Supplement 17
Current Protocols in Cytometry
A
1000
1000
T antigen
mock
54 hr
100
100
10
10
1
1 0
B
32
96
127
0
450
32
64
96
127
450 mock
405
Cell number
64
54 hr
405
360
360
315
315
270
270
225
225
180
180
135
135
90
90
45
45
0
0 0
256
512
768 1023
0
256
512
768 1023
DNA
Figure 7.17.2 (A) T antigen and DNA content of mock- and SV40-infected CV-1 cells. (B) DNA content profiles. By 54 hr pi, the majority of the cells are infected and expressing T antigen. The lower histograms demonstrate DNA content increases obtained in an SV40-permissive infection.
only; and (4) one series with an unrelated antibody of the same isotype as the 1°, with the secondary and PI. These data then allow selection of proper controls for each antibody and cell-virus system once the specificity and background have been determined. It is important to know the level of background fluorescence in order to calculate specific fluorescence/cell versus background/nonspecific fluorescence per cell. Once this measurement is performed, one can calculate the fluorescence attributable to the antigen of interest. The T antigen (and VPs) was an excellent choice as the initial viral protein assayed by this procedure, since it provided a log signal ∼40 to 60 channels above the control (uninfected cell) signal. If the signal for the antigen is in the linear range, careful attention must be focused on the background/nonspecific fluorescence. Therefore, the above experiment to determine the background/nonspecific staining is critical and may require
background subtraction for each channel (data point) collected. Proper safety precautions should be taken, since a virus is an infectious agent. The laboratory and investigators should maintain the proper conditions in handling an infectious agent. The following standards and guidelines are recommended by NIH for use in developing and implementing health and safety operating procedures and practices for both personnel and facilities, and they serve to supplement prevailing federal, state, and local laws and regulations: Biosafety in Microbiological and Biomedical Laboratories, U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, and the National Institutes of Health. HHS Publication No. (CDC) 93-8395. Investigators should follow these procedures and be knowledgeable about handling the agent of interest. For this protocol the cells and virus are fixed with methanol (or
Nucleic Acid Analysis
7.17.7 Current Protocols in Cytometry
Supplement 17
another fixative), thus eliminating the infectious nature of the agent and its products. However, investigators must maintain proper conditions prior to fixation. If the material is not fixed, the investigator must adhere to the biosafety standards for that agent and be concerned with aerosol generation during flow cytometry analysis.
Anticipated Results The results shown in Figures 7.17.1 through 7.17.3 are examples of the types of data that may be obtained. Figure 7.17.1 provides twocolor data of DNA content and T antigen as the infection proceeds. These histograms allow assessment of the efficiency of the infection by gating the infected versus the uninfected population, which provides the number of infected cells and quantity of viral antigen (T antigen) made per cell and population. Figure 7.17.2 shows the results for DNA versus T antigen of control and infected cells. The changes in DNA content between these two populations are
shown in the bottom panel. These results provide data on the progress of the infection and DNA content changes while monitoring the T antigen-positive/negative cells. Figure 7.17.3 shows the DNA content shifts of the infected population compared to a control CV-1 population. The ability to perform this assay enabled the conclusion that multiple S phases occurred during the SV40 permissive infection of CV-1 cells.
Time Considerations
Fixation will require ∼15 to 30 min depending on how much time the trypsin-EDTA needs to disperse the monolayer into a single suspension. The primary and secondary antibody stainings are usually 30 min each with 15 to 30 min for washes to remove the antibodies. The PI staining requires 30 min prior to filtration and running on the flow cytometer. Therefore the procedures for one sample may be completed within 2.5 to 3 hr. However, staining multiple samples, up to 30 to 60 per experiment,
Cell number
24 hr
48 hr
72 hr
DNA
Analysis of Viral Infection by Flow Cytometry
Figure 7.17.3 DNA profiles of uninfected cells (gray) and SV40-infected cells (black) at 24, 48, and 72 hr pi. The infected population has a total increase in DNA content of 10-12C: ∼2C from replication of SV40 DNA (1 × 106 copies) and an additional 6 to 8C of cellular DNA (2C from the first S phase and 4 to 6C from a second S phase resulting from a block to mitosis by the virus replication process).
7.17.8 Supplement 17
Current Protocols in Cytometry
takes considerably longer, and some experiments can extend over 2 days. Usually, the primary antibody is left to stain at 4°C overnight with the rest of the procedure performed the next day at room temperature, except for the 2° antibody, which is incubated at 37°C.
Literature Cited Cole, C. 1996. Polyomavirinae: The viruses and their replication. In Fundamental Virology, 3rd ed. (B.W. Fields, D.M. Knipe, and P.M. Howley, eds.) pp. 917-945. Lippencott-Raven Press, Philadelphia.
McSharry, J.J., Costantino, R., Robbiano, E., Echols, R.M., Stevens, R., and Lehman, J.M. 1990. Detection and quantitation of human immunodeficiency virus-infected peripheral blood mononuclear cells by flow cytometry. J. Clin. Microbiol. 28:724-733. Whalen, B., Laffin, J., Friedrich, T.D., and Lehman, J.M. 1999. SV40 small t antigen enhances rereplication and alters expression of G1 regulatory proteins in permissive CV1 cells. Exp. Cell Res. 251:121-127.
Key References Jacobberger et al., 1986. See above.
Elmendorf, S., McSharry, J., Laffin, J., Fogleman, D., and Lehman, J.M. 1988. Detection of an early cytomegalovirus antigen with two-color quantitative flow cytometry. Cytometry 9:254-260.
Initial description of the protocol for viral protein and DNA detection.
Friedrich, T.D., Laffin, J., and Lehman, J.M. 1993. Hypophosphorylated retinoblastoma gene product accumulates in SV40 infected CV-1 cells acquiring a tetraploid DNA content. Oncogene 8:1673-1677.
A review of the protocol with specific comments addressing the steps and potential problems.
Friedrich, T.D., Laffin, J., and Lehman, J.M. 1994. Induction of tetraploid DNA content by simian virus 40 is dependent on a T antigen function in G2 phase of the cell cycle. J. Virol. 68:40284030. Jacobberger, J.W., Fogleman, D., and Lehman, J.M. 1986. Analysis of intracellular antigens by flow cytometry. Cytometry 7:356-364. Laffin, J. and Lehman, J.M. 1994. Detection of intracellular virus and viral product. Methods Cell Biol. 41:543-557. Lehman, J.M., Laffin, J., Jacobberger, J., and Fogleman, D. 1988. Analysis of simian virus 40 infection of permissive CV-1 cells by quantitative two-color fluorescence with flow cytometry. Cytometry 9:52-59. Lehman, J.M., Friedrich, T.D., and Laffin, J. 1993. Quantitation of simian virus 40 T antigen correlated with the cell cycle of permissive and nonpermissive cells. Cytometry 14:401-410. Lehman, J.M., Laffin, J., and Friedrich T.D. 2000. Simian virus 40 induces multiple S phases with the majority of viral DNA replication in the G2 and second S phase in CV-1 cells. Exp. Cell Res. 258:215-222.
Laffin and Lehman, 1994. See above.
Internet Resources http://library.thinkquest.org/23054/basics/ index.html Information on general virology. http://www.asmusa.org/ Web page for The American Society for Microbiology (Virology). http://www.isac-net.org/ Web site for the International Society for Analytical Cytology. http://www.atcc.org/ Web site for the American Type Culture Collection: cells, viruses, and hybridomas. http://www.orcbs.msu.edu/biological/BMBL/ BMBL-1.htm Site providing Biosafety in Microbiological and Biomedical Laboratories publication.
Contributed by John M. Lehman, Thomas D. Friedrich, and Judith Laffin Albany Medical College Albany, New York
Nucleic Acid Analysis
7.17.9 Current Protocols in Cytometry
Supplement 17
Apoptosis Signaling Pathways In the past few years much has been learned about the molecular signals governing apoptosis, or programmed cell death, in lymphocytes and other cells of the immune system (for reviews see van Parijs and Abbas, 1996; Lenardo et al., 1998). Two major pathways, active and passive apoptosis, have been identified. Active apoptosis, also termed propriocidal cell death or antigen-induced cell death, occurs when cells are stimulated through a family of TNF-related receptors termed death receptors. These receptors are up-regulated in activated lymphocytes and trigger apoptosis chiefly in effector cells that have been recently stimulated through the antigen receptor. Passive or lymphokine withdrawal apoptosis occurs when activated lymphocytes are deprived of essential growth cytokines and does not require death receptors. Cell death initiated by either pathway is carried out by a unique family of intracellular cysteine proteases, the caspases. Defects in each of these pathways produce distinct pathologies and can synergize to allow accumulation of excess and autoimmune lymphocytes (Reap et al., 1995).
ACTIVE APOPTOSIS Death receptor–induced apoptosis serves to eliminate potentially harmful lymphocytes during responses against chronically expressed antigens. To allow the initial expansion of cells against these antigens, death receptor–induced apoptosis is regulated at a number of points during lymphocyte activation. Expression of Fas (CD95) and tumor necrosis factor (TNF) receptors is up-regulated during initial lymphocyte activation, but sensitivity to apoptosis lags behind this step, indicating that other mechanisms must be operative. The state of T cells in the cell cycle may be important, as progression through G1 and into S phase has been shown to be necessary for T cell receptor (TCR)–induced apoptosis (Boehme and Lenardo, 1993; Lissy et al., 1998). In activated T cells, Fas-ligand (FasL) and TNF expression are up-regulated after reengagement of antigen receptors, which allows cell “suicide” through Fas/FasL interactions, even on a single cell (Brunner et al., 1995; Dhein et al., 1995; Ju et al., 1995). However, mixing experiments have shown that only those T cells that are restimulated actually undergo apoptosis, indicating that a “competency-todie” signal that is separate from FasL up-regu-
lation emanates from the TCR (Hornung et al., 1997). B cells are also sensitized to undergo apoptosis when stimulated through surface Ig, but do not express FasL, and are thus dependent on other cell types to mediate Fas-induced cell death (Rathmell et al., 1995; Foote et al., 1996). These mechanisms cooperate to allow elimination of chronically stimulated B and T cells during viral infections and also participate in the elimination of self-reactive activated peripheral lymphocytes, while bystander cells activated by other stimuli are relatively preserved. In this way, death receptor signaling restrains the proliferation of antigen-specific effector cells during an immune response. The importance of death receptor signaling is illustrated by the phenotype of lpr/lpr or gld/gld mice, which are deficient in Fas and FasL, respectively (Watanabe-Fukunaga et al., 1992; Lynch et al., 1994). These mice have massive accumulation of abnormal T cells and autoantibody-mediated disease. A similar syndrome has been described in humans resulting from dominant negative mutations in Fas (Fisher et al., 1995). There are now five described death receptors in humans (Table 7.18.1), and their relative roles in apoptosis in different cell types is just beginning to be studied. Death receptors, like other TNF family members, recruit intracellular signaling molecules after oligomerization at the plasma membrane in response to soluble or membranebound ligands. In the case of Fas, as illustrated in Figure 7.18.1, this involves recruitment of the adapter molecule FADD and caspase-8 through protein-protein interaction domains. Fas recruits FADD via the death domain, and FADD recruits the pro-caspase-8 or FLICE molecule via death effector domains. Recent structural studies have shown that the death domain and death effector domains share the same basic structure containing six α-helices, but heterotypic interaction between proteins with these domains has not been found. Viral and cellular proteins containing death effector domains have recently been identified (FLICE inhibitory proteins, or FLIPs). These proteins can interfere with the FADD/caspase-8 interaction and block apoptosis induced by a number of death receptors (Bertin et al., 1997; Han et al., 1997; Hu et al., 1997; Irmler et al., 1997; Shu et al., 1997; Srinivasula et al., 1997; Thome
Contributed by Richard M. Siegel and Michael J. Lenardo Current Protocols in Cytometry (2002) 7.18.1-7.18.10 Copyright © 2002 by John Wiley & Sons, Inc.
UNIT 7.18
Nucleic Acid Analysis
7.18.1 Supplement 21
Supplement 21
TRAMP, WSL, APO-2, LARD
TRAIL-R1
TRAIL-R2
TRAIL-R3, DcR1
TNFR2
DR3
DR4
DR5
TRID
aND, not defined.
TRAIL- DcR2 R4
CD120b
TNFR1
42 (predicted)
27 (predicted)
45 (predicted)
50 (predicted)
45 (predicted)
75-80
55-60
45-50
Fas
DR1, APO-1, CD95 DR2, CD120a
Mol. wt. (kDa)
TRAIL
TRAIL
TRAIL
TRAIL
TWEAK (APO-3)
TNF
TNF
FasL
Ligand
Can heterodimerize with DR4, induced by DNA damage GPI-linked extracellular receptor; blocks TRAIL-induced apoptosis
NDa
NDa
Incomplete death domain; blocks TRAIL-induced apoptosis
May bind FADD, but no requirement for FADD in DR4-induced death found in FADD knockout mice
NDa
In mature CD8 T cells, appears to mediate most TCR-induced apoptosis Expression restricted to lymphocytes; induced after DNA damage via p53-dependent mechanism
No intracellular domain
References
Pan et al. (1997); Walczak et al. (1997); Wu et al. (1997) Degli-Esposti et al. (1997b); Pan et al. (1997); Sheridan et al. (1997) Degli-Esposti et al. (1997a); Marsters et al. (1997)
Schneider et al. (1997a,b); Yeh et al. (1998)
Chinnaiyan et al. (1996); Kitson et al. (1996); Screaton et al. (1997)
Zheng et al. (1995)
Trauth et al. (1989); Itoh et al. (1991); Oehm et al. (1992) Cytotoxicity can be modulated by NF-κB Loetscher et al. (1990); Schall et al. (1990)
Most consistently pro-apoptotic
Comments
TRADD, FADD (indirect), TNFR1, caspase-10
TRADD, RIP (indirect), FADD (indirect), TRAF2 (indirect) TRAF-1/2
FADD, caspase-8
Interacting molecules
Death Receptors and Their Signaling Pathways
Molecule Other names
Table 7.18.1
Apoptosis Signaling Pathways
7.18.2
Current Protocols in Cytometry
death ligand + receptor
ALPS mutants/decoy receptors
death receptor
death receptors (e.g., Fas/CD95)
pro-caspase-8
FADD active caspase-8 (activator caspases)
FLIP lymphokine withdrawal
14-3-3 BAD
BAD-P
Akt*
BID
signal integration
Bcl-2 Bcl-x
mitochondrial permeability transition
pro-caspase-9 (APAF-3) active caspase-9
p35, IAPs caspase-3,-6,-7 (effector caspases)
dATP
cytochrome c (APAF-2)
Bcl-2 APAF-1 Bcl-x
cleavage of cellular substrates (e.g., ICAD, PARP, DNAPK, actin)
cytoskeletal reorganization/cellular shrinkage
execution
plasma membrane phosphaditylserine exposure DNA cleavage
Figure 7.18.1 Induction of apoptosis by death receptors and lymphokine withdrawal. Four major steps in apoptosis signaling are depicted. The Fas/CD95 receptor is shown as an example of deathreceptor signaling. Ligand binding triggers trimerization and recruitment of FADD through the death domain (grey rectangle). Pro-caspase-8 is then recruited through death effector domain interactions (hexagons). Oligomerization of caspase-8 triggers its proteolytic cleavage into the active p17 and p11 subunits. Apoptosis can then proceed via a direct pathway (right arrow), involving cleavage of effector caspases, or an indirect pathway, requiring release of cytochrome c and activation of caspase-3 via APAF-1. The indirect pathway of death-receptor signaling, possibly mediated by BID, is shown by the left arrow. One possible pathway by which lymphokine withdrawal might trigger apoptosis is also depicted. Other mechanisms, such as a change in the Bcl-2/Bax ratio, have also been described (Broome et al., 1995). Once effector caspases are activated, the execution phase of apoptosis inevitably proceeds. Abbreviations: Akt*, activated Akt; ALPS, autoimmune lymphoproliferative syndrome; DNA-PK, DNA protein kinase; ICAD, inhibitor of caspase-activated DNase; PARP, poly(ADP-ribose) polymerase.
Nucleic Acid Analysis
7.18.3 Current Protocols in Cytometry
Supplement 21
Apoptosis Signaling Pathways
et al., 1997; Rasper et al., 1998). Clustering of caspase-8 in the Fas signaling complex triggers proteolytic autoactivation of the protease encoded by this protein, which initiates a cascade of caspase activation to produce apoptosis. Other death receptors can activate caspases but produce distinct responses, probably because they associate with different adapter molecules (Table 7.18.2). The exact connections between specific death receptors and the intracellular signaling machinery is currently under study. The execution phase of apoptosis is associated with activation of the caspase family of cysteine proteases (for reviews see Cohen, 1997; Nicholson and Thornberry, 1997). The caspases are a distinct family of cysteine proteases that cleave substrates at characteristic sequences containing aspartate in the P1 position (Table 7.18.3). Each caspase is produced as a pro-form that encodes a prodomain of variable length attached to peptides encoding the large and small catalytic subunits of the enzyme. These segments are separated by spacer segments and caspase cleavage sites. Biochemical and genetic studies have enabled the death-associated caspases to be divided into a number of subgroups. Activator caspases (caspase-8, -9, and -10) appear to function primarily by activating other downstream caspases, as their cleavage specificities closely match the cleavage sites of other caspases. This is understood most clearly in the Fas pathway, in which experimental activation of caspase-8 through oligomerization can carry out the function of the Fas signaling complex, trigger downstream caspases, and produce apoptosis (Martin et al., 1998; Muzio et al., 1998; Yang et al., 1998). Effector caspases (caspase-3, -6, and -7) are the major active caspases present in apoptotic cells, and have specificities matching cellular substrates whose cleavage is known to be important in generating the apoptotic phenotype. Some specific features of apoptosis, such as DNA cleavage, have been linked to caspases (Enari et al., 1998; Sakahira et al., 1998). Most likely, the cleavage sites and short prodomains of effector caspases prohibit autoprocessing, making their activity dependent on upstream events (Thornberry et al., 1997). Caspase-2 has substrate preferences resembling the effector caspases but has a long prodomain that can bind the adapter molecule RAIDD through a caspase recruitment domain (CARD), a protein–protein interaction domain similar to the death effector domain (Duan and Dixit, 1997). Thus, this caspase may be a more upstream effector
caspase. Studies with caspase-deficient mice created through gene targeting studies suggest that there is significant redundancy between caspases that varies between cell types. Caspase-3-deficient mice have impaired neuronal apoptosis but normal lymphocyte apoptosis (Kuida et al., 1996), whereas caspase-2-deficient animals seem to have a specific defect in germ-cell apoptosis (Bergeron et al., 1998).
PASSIVE APOPTOSIS Although also dependent on caspases, passive apoptosis in lymphocytes does not occur through death-receptor signaling. Instead, this type of cell death occurs in response to the withdrawal of trophic cytokines. During an immune response, much of the cell loss that occurs after peak expansion may be caused by lymphokine withdrawal. Artificially maintaining high interleukin 2 (IL-2) concentrations can prolong the lifespan of T cells responding to a superantigen challenge (Kuroda et al., 1996). The Bcl-2 oncoprotein and related molecules are known to suppress this form of apoptosis. Overexpression of Bcl-2 can prolong B cell survival as well as memory B cell responses, indicating the physiological importance of this death pathway (McDonnell et al., 1990; Nunez et al., 1990, 1991). Although it has been known for many years that anti-apoptotic proteins of the Bcl-2 family could block passive apoptosis, the signals by which lymphokine withdrawal initiates apoptosis are unclear. One possible mechanism was found in IL-3-dependent cell lines, in which IL-3 maintains phosphorylation of the pro-apoptotic Bcl-2 family member BAD through the phosphatidylinositol-3′-kinase (PI3K)/Akt pathway. 14-3-3 proteins sequester phosphorylated BAD, preventing interaction with other Bcl-2 family members on intracellular membranes. After IL-3 withdrawal, BAD is dephosphorylated and released from the 143-3 protein, and can then form inhibitory dimers with other Bcl-2 family members to block the anti-apoptotic function of these proteins (del Peso et al., 1997). In other cell types, levels of Bcl-2 have been shown to drop after lymphokine withdrawal, which may also predispose these cells to apoptosis (Broome et al., 1995). Whereas death receptors can activate caspases directly, Bcl-2 family proteins control a separate pathway associated with the mitochondrial release of cytochrome c and possibly other pro-apoptotic factors. Although cytochrome c is normally an essential component of mitochondrial respiration at the mitochon-
7.18.4 Supplement 21
Current Protocols in Cytometry
Nucleic Acid Analysis
7.18.5
Current Protocols in Cytometry
Supplement 21
27
74
34
23 (predicted)
FADD
RIP
TRADD
RAIDD (CRADD) DAXX
120
Mol. wt. (kDa)
TNF-R1 via death domain RIP via death domain Fas or TNF-R1 death domains
TNF-R1 via death domain and TRADD
Fas via C-terminal death domain
Upstream molecule Death domain protects from Fas-mediated apoptosis; death effector domain induces apoptosis without Fas cross-linking Induces cell death via death domain
Caspase-8 (and caspase-10) via death effector domain
No direct downstream molecule identified
References
Hsu et al. (1995)
Stanger et al. (1995); Impaired NF-kB Ting et al. (1996); activation and hypersensitivity to death Kelliher et al. (1998) following TNF treatment
Boldin et al. (1995); Impaired Fas-mediated apoptosis; block in T cell Chinnaiyan et al. (1995) mitogenesis
Effects of blocking or deficiency
Ahmad et al. (1997); Duan and Dixit (1997) Activates JNK; synergizes with Inhibition of Fas-induced Yang et al. (1997) Fas for induction of apoptosis apoptosis
(1) TRAF-2 via intermediate domain; substrates not clear (IKK is not a direct substrate); (2) RAIDD via death domain TRAF-2, RIP, FADD Causes cell death through recruitment of FADD ICH-1 via CARD domain Induces apoptosis
Effects of overexpression
Downstream molecule
Adapter Proteins in the Death Pathway
Name
Table 7.18.2
7.18.6
Supplement 21
Current Protocols in Cytometry
DXXD, DEXD
Effector
Activator
Caspase-3 (CPP32, Yama, apopain), caspase-6 (Mch2), caspase-7 (Mch3, ICE-LAP3, CMH-1)
Caspase-8 (FLICE, MACH, Mch5), caspase-10 (Mch4, FLICE-2)
Activator
DEHD
Effector/ activator
Caspase-2 (ICH-1, NEDD-2)
Caspase-9 (ICE-LAP-6, Mch6)
(W/L)EHD
ICE-related
Caspase-1 (ICE), caspase-4 (TX, ICE-REL II, ICH-2), caspase-5 (ICE-REL III) caspase-11, caspase-12
Caspase-3, PARP
LETD Caspases 1-10 (caspase-8), IEAD (caspase-10) LEHD
Inhibitors
RAIDD?, caspase-3
YVAD-CHO Caspase-11 required for caspase-1 activation; growth-factor withdrawal and ischemia can activate ICE subfamily proteases
Activators
APAF-1/ cytochrome c
Oligomerization, aggregation in complexes with FADD
Caspase-3 and -6 may be major active species present in apoptotic cells; caspase-3 deficiency disturbs neuronal apoptosis during development
Essential in germ cell apoptosis; may inhibit neuronal apoptosis
Secreted IL-1 probably required for apoptosis induction
Comments
Forms a complex with Bcl-2 family members and APAF-1 (ced-4 homologue), which cleaves caspase-3 when active
zVAD-FMK, Crm Important for death A (caspase-8 >> receptor–activated caspase-10), FLIPs apoptosis (role in other forms of apoptosis not clear)
Upstream caspases DEVD-CHO Many: PARP, (e.g., caspase-8, -9) c-IAP family S-REBP, lamin (caspase-6 only), DNA-PK, actin, PKC, D4-GDI, ICAD, Rb, caspase-9
Pro-IL-1β, γ-interferon inducing factor
Ideal Protein targets cleavage site
Subfamily
Caspase Subfamilies
Caspase(s)
Table 7.18.3
Li et al. (1997); Zou et al. (1997); Pan et al. (1998)
Boldin et al. (1996); Muzio et al. (1996); Scaffidi et al. (1997)
Nicholson et al. (1995); Kuida et al. (1996); Takahashi et al. (1996); Faleiro et al. (1997); Rosen and Casciola-Rosen (1997); Sakahira et al. (1998)
Bergeron et al. (1998)
Friedlander et al. (1996); Wang et al. (1998)
References
drial inner membrane, a number of apoptotic stimuli (e.g., staurosporine, Bcl-2 antagonists, and glucocorticoids) can trigger a mitochondrial permeability transition characterized by loss of membrane potential and efflux of macromolecules such as cytochrome c to the cytoplasm. Agents that block the permeability transition (e.g., bongkrekic acid) can prevent many of the phenotypic features of apoptosis (Marchetti et al., 1996). In the presence of dATP, cytoplasmic cytochrome c binds and activates APAF-1, a 120-kD protein with homology to the C. elegans death-promoting ced4 gene product. Activated APAF-1 recruits and activates caspase-9, which in turn activates caspase-3 through proteolytic processing (Li et al., 1997; Zou et al., 1997). The sensitivity of different cell types to apoptosis inhibition by Bcl-2 may correlate with the requirement for caspase activation through the cytochrome c/APAF pathway. In some cell types, including primary lymphocytes, death-receptor signaling may directly activate downstream caspases without a requirement for mitochondrial changes (Scaffidi et al., 1998). In other cells, such as fibroblasts, death-receptor signaling can be blocked by Bcl-2. Thus, cross-talk between these two cell death pathways can differ dramatically between cell lineages. In cell types in which death-receptor signaling is dependent on the mitochondria, it is not clear how death-receptor signaling proteins interact with the Bcl-2 family of proteins. Recent data suggests that the pro-apoptotic molecule BID may mediate the signal from caspase-8 to mitochondria by binding to Bcl-2 or Bcl-xL at the outer mitochondrial membrane. Cleavage of BID by caspase-8 has been shown to increase its ability to bind to anti-apoptotic Bcl-2 family members (Li et al., 1998; Luo et al., 1998).
LITERATURE CITED Ahmad, M., Srinivasula, S.M., Wang, L., Talanian, R.V., Litwack, G., Fernandes-Alnemri, T., and Alnemri, E.S. 1997. CRADD, a novel human apoptotic adaptor molecule for caspase-2, and FasL/tumor necrosis factor receptor-interacting protein RIP. Cancer Res. 57:615-619. Bergeron, L., Perez, G.I., Macdonald, G., Shi, L., Sun, Y., Jurisicova, A., Varmuza, S., Latham, K.E., Flaws, J.A., Salter, J.C., Hara, H., Moskowitz, M.A., Li, E., Greenberg, A., Tilly, J.L., and Yuan, J. 1998. Defects in regulation of apoptosis in caspase-2-deficient mice. Genes & Dev. 12:1304-1314.
Bertin, J., Armstrong, R.C., Ottilie, S., Martin, D.A., Wang, Y., Banks, S., Wang, G.H., Senkevich, T.G., Alnemri, E.S., Moss, B., Lenardo, M.J., Tomaselli, K.J., and Cohen, J.I. 1997. Death effector domain-containing herpesvirus and poxvirus proteins inhibit both Fas- and TNFR1induced apoptosis. Proc. Natl. Acad. Sci. U.S.A. 94:1172-1176. Boehme, S.A. and Lenardo, M.J. 1993. Propriocidal apoptosis of mature T lymphocytes occurs at S phase of the cell cycle. Eur. J. Immunol. 23:15521560. Boldin, M.P., Varfolomeev, E.E., Pancer, Z., Mett, I.L., Camonis, J.H., and Wallach, D. 1995. A novel protein that interacts with the death domain of Fas-Apo-1 contains a sequence motif related to the death domain. J. Biol. Chem. 270:7795-7798. Boldin, M.P., Goncharov, T.M., Goltsev, Y.V., and Wallach, D. 1996. Involvement of MACH, a novel MORT1/FADD-interacting protease, in Fas/APO-1- and TNF receptor-induced cell death. Cell 85:803-815. Broome, H.E., Dargan, C.M., Krajewski, S., and Reed, J.C. 1995. Expression of Bcl-2, Bcl-x, and Bax after T cell activation and IL-2 withdrawal. J. Immunol. 155:2311-2317. Brunner, T., Mogil, R.J., LaFace, D., Yoo, N.J., Mahboubi, A., Echeverri, F., Martin, S.J., Force, W.R., Lynch, D.H., Ware, C.F., et al. 1995. Cellautonomous Fas (CD95)/Fas-ligand interaction mediates activation-induced apoptosis in T-cell hybridomas. Nature 373:441-444. Chinnaiyan, A.M., O’Rourke, K., Tewari, M., and Dixit, V.M. 1995. FADD, a novel death domaincontaining protein, interacts with the death domain of Fas and initiates apoptosis. Cell 81:505512. Chinnaiyan, A.M., O’Rourke, K., Yu, G.L., Lyons, R.H., Garg, M., Duan, D.R., Xing, L., Gentz, R., Ni, J., and Dixit, V.M. 1996. Signal transduction by DR3, a death domain-containing receptor related to TNFR-1 and CD95. Science 274:990992. Cohen, G.M. 1997. Caspases: The executioners of apoptosis. Biochem. J. 326:1-16. Degli-Esposti, M.A., Dougall, W.C., Smolak, P.J., Waugh, J.Y., Smith, C.A., and Goodwin, R.G. 1997a. The novel receptor TRAIL-R4 induces NF-kappaB and protects against TRAIL-mediated apoptosis, yet retains an incomplete death domain. Immunity 7:813-820. Degli-Esposti, M.A., Smolak, P.J., Walczak, H., Waugh, J., Huang, C.P., DuBose, R.F., Goodwin, R.G., and Smith, C.A. 1997b. Cloning and characterization of TRAIL-R3, a novel member of the emerging TRAIL receptor family. J. Exp. Med. 186:1165-1170. del Peso, L., Gonzalez-Garcia, M., Page, C., Herrera, R., and Nunez, G. 1997. Interleukin-3-induced phosphorylation of BAD through the protein kinase Akt. Science 278:687-689. Nucleic Acid Analysis
7.18.7 Current Protocols in Cytometry
Supplement 21
Dhein, J., Walczak, H., Baumler, C., Debatin, K.M., and Krammer, P.H. 1995. Autocrine T-cell suicide mediated by APO-1/(Fas/CD95). Nature 373:438-441. Duan, H. and Dixit, V.M. 1997. RAIDD is a new ‘death’ adaptor molecule. Nature 385:86-89.
Kelliher, M.A., Grimm, S., Ishida, Y., Kuo, F., Stanger, B.Z., and Leder, P. 1998. The death domain kinase RIP mediates the TNF-induced NF-kappaB signal. Immunity 8:297-303.
Enari, M., Sakahira, H., Yokoyama, H., Okawa, K., Iwamatsu, A., and Nagata, S. 1998. A caspaseactivated DNase that degrades DNA during apoptosis, and its inhibitor ICAD. Nature 391:43-50.
Kitson, J., Raven, T., Jiang, Y.P., Goeddel, D.V., Giles, K.M., Pun, K.T., Grinham, C.J., Brown, R., and Farrow, S.N. 1996. A death-domain-containing receptor that mediates apoptosis. Nature 384:372-375.
Faleiro, L., Kobayashi, R., Fearnhead, H., and Lazebnik, Y. 1997. Multiple species of CPP32 and Mch2 are the major active caspases present in apoptotic cells. EMBO J. 16:2271-2281.
Kuida, K., Zheng, T.S., Na, S., Kuan, C., Yang, D., Karasuyama, H., Rakic, P., and Flavell, R.A. 1996. Decreased apoptosis in the brain and premature lethality in CPP32-deficient mice. Nature 384:368-372.
Fisher, G.H., Rosenberg, F.J., Straus, S.E., Dale, J.K., Middleton, L.A., Lin, A.Y., Strober, W., Lenardo, M.J., and Puck, J.M. 1995. Dominant interfering Fas gene mutations impair apoptosis in a human autoimmune lymphoproliferative syndrome. Cell 81:935-946. Foote, L.C., Schneider, T.J., Fischer, G.M., Wang, J.K., Rasmussen, B., Campbell, K.A., Lynch, D.H., Ju, S.T., Marshak-Rothstein, A., and Rothstein, T.L. 1996. Intracellular signaling for inducible antigen receptor–mediated Fas resistance in B cells. J. Immunol. 157:1878-1885. Friedlander, R.M., Gagliardini, V., Rotello, R.J., and Yuan, J. 1996. Functional role of interleukin 1 beta (IL-1 beta) in IL-1 beta- converting enzymemediated apoptosis. J. Exp. Med. 184:717-724. Han, D., Chaudhary, P.M., Wright, M.E., Friedman, C., Trask, B.J., Riedel, R., Baskin, D., Schwartz, S.M., and Hood, L. 1997. MRIT, a novel deatheffector domain-containing protein, interacts with caspases and BclXl and initiates cell death. Proc. Natl. Acad. Sci. U.S.A. 94:11333-11338. Hornung, F., Zheng, L., and Lenardo, M.J. 1997. Maintenance of clonotype specificity in CD95/Apo-1/Fas-mediated apoptosis of mature T lymphocytes. J. Immunol. 159:3816-3822. Hsu, H., Xiong, J., and Goeddel, D.V. 1995. The TNF receptor 1-associated protein TRADD signals cell death and NF-kappaB activation. Cell 81:495-504. Hu, S., Vincenz, C., Ni, J., Gentz, R., and Dixit, V.M. 1997. I-FLICE, a novel inhibitor of tumor necrosis factor receptor-1- and CD-95-induced apoptosis. J. Biol. Chem. 272:17255-17257. Irmler, M., Thome, M., Hahne, M., Schneider, P., Hofmann, K., Steiner, V., Bodmer, J.L., Schroter, M., Burns, K., Mattmann, C., Rimoldi, D., French, L.E., and Tschopp, J. 1997. Inhibition of death receptor signals by cellular FLIP. Nature 388:190-195. Itoh, N., Yonehara, S., Ishii, A., Yonehara, M., Mizushima, S., Sameshima, M., Hase, A., Seto, Y., and Nagata, S. 1991. The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis. Cell 66:233-243. Apoptosis Signaling Pathways
tions required for programmed cell death after T-cell activation . Nature 373:444-448.
Ju, S.T., Panka, D.J., Cui, H., Ettinger, R., el-Khatib, M., Sherr, D.H., Stanger, B.Z., and MarshakRothstein, A. 1995. Fas(CD95)/FasL interac-
Kuroda, K., Yagi, J., Imanishi, K., Yan, X.J., Li, X.Y., Fujimaki, W., Kato, H., Miyoshi-Akiyama, T., Kumazawa, Y., Abe, H., and Uchiyama, T. 1996. Implantation of IL-2-containing osmotic pump prolongs the survival of superantigen-reactive T cells expanded in mice injected with bacterial superantigen. J. Immunol. 157:1422-1431. Lenardo, M.J., Chan, F., Hornung, F., McFarland, H., Siegel, R., Wang, J., and Zheng, L. 1998. Mature T lymphocyte apoptosis—immune regulation in a dynamic and unpredictable antigenic environment. Annu. Rev. Immunol. In Press. Li, H., Zhu, H., Xu, C.J., and Yuan, J. 1998. Cleavage of BID by caspase 8 mediates the mitochondrial damage in the Fas pathway of apoptosis. Cell 94:491-501. Li, P., Nijhawan, D., Buduhardjo, I., Srinivasula, S., Ahmad, M., Alnemri, E., and Wang, X. 1997. Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91:479-489. Lissy, N.A., Van Dyk, L.F., Becker-Hapak, M., Vocero-Akbani, A., Mendler, J.H., and Dowdy, S.F. 1998. TCR antigen-induced cell death occurs from a late G1 phase cell cycle check point. Immunity 8:57-65. Loetscher, H., Pan, Y.C., Lahm, H.W., Gentz, R., Brockhaus, M., Tabuchi, H., and Lesslauer, W. 1990. Molecular cloning and expression of the human 55 kd tumor necrosis factor receptor. Cell 61:351-359. Luo, X., Budihardjo, I., Zou, H., Slaughter, C., and Wang, X. 1998. BID, a Bcl-2 interacting protein, mediates cytochrome c release from mitochondria in response to activation of cell surface death receptors. Cell 94:481-490. Lynch, D.H., Watson, M.L., Alderson, M.R., Baum, P.R., Miller, R.E., Tough, T., Gibson, M., DavisSmith, T., Smith, C.A., Hunter, K., et al. 1994. The mouse Fas-ligand gene is mutated in gld mice and is part of a TNF family gene cluster. Immunity 1:131-136. Marchetti, P., Castedo, M., Susin, S.A., Zamzami, N., Hirsch, T., Macho, A., Haeffner, A., Hirsch, F., Geuskens, M., and Kroemer, G. 1996. Mitochondrial permeability transition is a central co-
7.18.8 Supplement 21
Current Protocols in Cytometry
ordinating event of apoptosis. J. Exp. Med. 184:1155-1160. Marsters, S.A., Sheridan, J.P., Pitti, R.M., Huang, A., Skubatch, M., Baldwin, D., Yuan, J., Gurney, A., Go dd ard, A.D., Godowski, P., and Ashkenazi, A. 1997. A novel receptor for Apo2L/TRAIL contains a truncated death domain. Curr. Biol. 7:1003-1006. Martin, D.A., Siegel, R.M., Zheng, L., and Lenardo, M.J. 1998. Membrane oligomerization and release activates the caspase 8 (FLICE/MACHα1) death signal. J. Biol. Chem. 273:4345-4349. McDonnell, T.J., Nunez, G., Platt, F.M., Hockenberry, D., London, L., McKearn, J.P., and Korsmeyer, S.J. 1990. Deregulated Bcl-2-immunoglobulin transgene expands a resting but responsive immunoglobulin M and D-expressing B-cell population. Mol. Cell. Biol. 10:1901-1907. Muzio, M., Chinnaiyan, A.M., Kischkel, F.C., O’Rourke, K., Shevchenko, A., Ni, J., Scaffidi, C., Bretz, J.D., Zhang, M., Gentz, R., Mann, M., Krammer, P.H., Peter, M.E., and Dixit, V.M. 1996. FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death-inducing signaling complex. Cell 85:817-827. Muzio, M., Stockwell, B.R., Stennicke, H.R., Salvesen, G.S., and Dixit, V.M. 1998. An induced proximity model for caspase-8 activation. J. Biol. Chem. 273:2926-2930. Nicholson, D. and Thornberry, N. 1997. Caspases: Killer proteases. Trends Biochem. Sci. 22:299306. Nicholson, D.W., Ali, A., Thornberry, N.A., Vaillancourt, J.P., Ding, C.K., Gallant, M., Gareau, Y., Griffin, P.R., Labelle, M., Lazebnik, Y.A., et al. 1995. Identification and inhibition of the ICE/CED-3 protease necessary for mammalian apoptosis. Nature 376:37-43. Nunez, G., London, L., Hockenberry, D., Alexander, M., McKearn, J.P., and Korsmeyer, S.J. 1990. Deregulated Bcl-2 gene expression selectively prolongs survival of growth factor-deprived hemopoietic cell lines. J. Immunol. 144:36023610. Nunez, G., Hockenberry, D., McDonnell, T.J., Sorensen, C.M., and Korsmeyer, S.J. 1991. Bcl2 maintains B cell memory. Nature 353:71-73. Oehm, A., Behrmann, I., Falk, W., Pawlita, M., Maier, G., Klas, C., Li-Weber, M., Richards, S., Dhein, J., Trauth, B.C., et al. 1992. Purification and molecular cloning of the APO-1 cell surface antigen, a member of the tumor necrosis factor/nerve growth factor receptor superfamily. Sequence identity with the Fas antigen. J. Biol. Chem. 267:10709-10715. Pan, G., Ni, J., Wei, Y.F., Yu, G., Gentz, R., and Dixit, V.M. 1997. An antagonist decoy receptor and a death domain-containing receptor for TRAIL. Science 277:815-818. Pan, G., O’Rourke, K., and Dixit, V.M. 1998. Caspase-9, Bcl-XL, and Apaf-1 form a ternary complex. J. Biol. Chem. 273:5841-5845.
Rasper, D.M., Vaillancourt, J.P., Hadano, S., Houtzager, V.M., Seiden, I., Keen, S.L.C., Tawa, P., Xanthoudakis, S., Nasir, J., Martindale, D., Koop, B.F., Peterson, E.P., Thornberry, N.A., Huang, J-Q., Black, S.C., Hornung, F., Lenardo, M.J., Hayden, M.R., Roy, S., and Nicholson, D.W 1998. Cell death attenuation by ‘Usurpin’, a mammalian DED-caspase homologue that precludes caspase-8 recruitment and activation by the CD-95 (Fas, APO-1) receptor complex. Cell Death Differ. 5:271-288. Rathmell, J.C., Cooke, M.P., Ho, W.Y., Grein, J., Townsend, S.E., Davis, M.M., and Goodnow, C.C. 1995. CD95 (Fas)-dependent elimination of self-reactive B cells upon interaction with CD4+ T cells. Nature 376:181-184. Reap, E.A., Felix, N.J., Wolthusen, P.A., Kotzin, B.L., Cohen, P.L., and Eisenberg, R.A. 1995. bcl-2 transgenic Lpr mice show profound enhancement of lymphadenopathy. J. Immunol. 155:5455-5462. Rosen, A. and Casciola-Rosen, L. 1997. Macromolecular substrates for the ICE-like proteases during apoptosis. J. Cell. Biochem. 64:50-54. Sakahira, H., Enari, M., and Nagata, S. 1998. Cleavage of CAD inhibitor in CAD activation and DNA degradation during apoptosis. Nature 391:96-99. Scaffidi, C., Medema, J.P., Kramer, P.H., Peter, M.E. 1997. FLICE is predominantly expressed as two functionally active isoforms, caspase-8/a and caspase-8/b. J. Biol. Chem. 272:26953-26958. Scaffidi, C., Fulda, S., Srinivasan, A., Friesen, C., Li, F., Tomaselli, K.J., Debatin, K.M., Krammer, P.H., and Peter, M.E. 1998. Two CD95 (APO1/Fas) signaling pathways. EMBO J. 17:16751687. Schall, T.J., Lewis, M., Koller, K.J., Lee, A., Rice, G.C., Wong, G.H., Gatanaga, T., Granger, G.A., Lentz, R., Raab, H., et al. 1990. Molecular cloning and expression of a receptor for human tumor necrosis factor. Cell 61:361-70. Schneider, P., Bodmer, J.L., Thome, M., Hofmann, K., Holler, N., and Tschopp, J. 1997a. Characterization of two receptors for TRAIL. FEBS Lett. 416:329-334. Schneider, P., Thome, M., Burns, K., Bodmer, J.L., Hofmann, K., Kataoka, T., Holler, N., and Tschopp, J. 1997b. TRAIL receptors 1 (DR4) and 2 (DR5) signal FADD-dependent apoptosis and activate NF-kappaB. Immunity 7:831-836. Screaton, G.R., Xu, X.N., Olsen, A.L., Cowper, A.E., Tan, R., McMichael, A.J., and Bell, J.I. 1997. LARD: A new lymphoid-specific death domain containing receptor regulated by alternative pre-mRNA splicing. Proc. Natl. Acad. Sci. U.S.A. 94:4615-4619. Sheridan, J.P., Marsters, S.A., Pitti, R.M., Gurney, A., Skubatch, M., Baldwin, D., Ramakrishnan, L., Gray, C.L., Baker, K., Wood, W.I., Goddard, A.D., Godowski, P., and Ashkenazi, A. 1997. Control of TRAIL-induced apoptosis by a family of signaling and decoy receptors. Science 277:818-821.
Nucleic Acid Analysis
7.18.9 Current Protocols in Cytometry
Supplement 21
Shu, H.B., Halpin, D.R., and Goeddel, D.V. 1997. Casper is a FADD- and caspase-related inducer of apoptosis. Immunity 6:751-763. Srinivasula, S.M., Ahmad, M., Ottilie, S., Bullrich, F., Banks, S., Wang, Y., Fernandes-Alnemri, T., Croce, C.M., Litwack, G., Tomaselli, K.J., Armstrong, R.C., and Alnemri, E.S. 1997. FLAME-1, a novel FADD-like anti-apoptotic molecule that regulates Fas/TNFR1-induced apoptosis. J. Biol. Chem. 272:18542-18545. Stanger, B.Z., Leder, P., Lee, T.H., Kim, E., and Seed, B. 1995. RIP: A novel protein containing a death domain that interacts with Fas/APO-1 (CD95) in yeast and causes cell death. Cell 81:513-523. Takahashi, A., Alnemri, E.S., Lazebnik, Y.A., Fernandes-Alnemri, T., Litwack, G., Moir, R.D., Goldman, R.D., Poirier, G.G., Kaufmann, S.H., and Earnshaw, W.C. 1996. Cleavage of lamin A by Mch2 alpha but not CPP32: Multiple interleukin 1 beta-converting enzyme-related proteases with distinct substrate recognition properties are active in apoptosis. Proc. Acad. Natl. Sci. U.S.A. 93:8395-8400. Thome, M., Schneider, P., Hofmann, K., Fickenscher, H., Meinl, E., Neipel, F., Mattmann, C., Burns, K., Bodmer, J.L., Schroter, M., Scaffidi, C., Krammer, P.H., Peter, M.E., and Tschopp, J. 1997. Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors. Nature 386:517-521. Thornberry, N.A., Rano, T.A., Peterson, E.P., Rasper, D.M., Timkey, T., Garcia-Calvo, M., Houtzager, V.M., Nordstrom, P.A., Roy, S., Vaillancourt, J.P., Chapman, K.T., and Nicholson, D.W. 1997. A combinatorial approach defines specificities of members of the caspase family and granzyme B. Functional relationships established for key mediators of apoptosis. J. Biol. Chem. 272:17907-17911. Ting, A.T., Pimentel-Muinos, F.X., and Seed, B. 1996. RIP mediates tumor necrosis factor receptor 1 activation of NF-kappaB but not Fas/APO1-initiated apoptosis. EMBO J. 15:6189-6196. Trauth, B.C., Klas, C., Peters, A.M., Matzku, S., Moller, P., Falk, W., Debatin, K.M., and Krammer, P.H. 1989. Monoclonal antibody–mediated tumor regression by induction of apoptosis. Science 245:301-305. van Parijs, L. and Abbas, A.K. 1996. Role of Fasmediated cell death in the regulation of immune responses. Curr. Opin. Immunol. 8:355-361. Walczak, H., Degli-Esposti, M.A., Johnson, R.S., Smolak, P.J., Waugh, J.Y., Boiani, N., Timour, M.S., Gerhart, M.J., Schooley, K.A., Smith, C.A., Goodwin, R.G., and Rauch, C.T. 1997. TRAIL-R2: A novel apoptosis-mediating receptor for TRAIL. EMBO J. 16:5386-5397.
Wang, S., Miura, M., Jung, Y.K., Zhu, H., Li, E., and Yuan, J. 1998. Murine caspase-11, an ICE-interacting protease, is essential for the activation of ICE. Cell 92:501-509. Watanabe-Fukunaga, R., Brannan, C.I., Copeland, N.G., Jenkins, N.A., and Nagata, S. 1992. Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 356:314-317. Wu, G.S., Burns, T.F., McDonald, E.R. III, Jiang, W., Meng, R., Krantz, I.D., Kao, G., Gan, D.D., Zhou, J.Y., Muschel, R., Hamilton, S.R., Spinner, N.B., Markowitz, S., Wu, G., and El-Deiry, W.S. 1997. KILLER/DR5 is a DNA damage-inducible p53-regulated death receptor gene. Nature Genet. 17:141-143. Yang, X., Korshravi-Far, R., Chang, H.Y., and Baltimore, D. 1997. Daxx, a novel Fas-binding protein activates Jnk and apoptosis. Cell 89:10671076. Yang, X., Chang, H.Y., and Baltimore, D. 1998. Autoproteolytic activation of pro-caspases by oligomerization. Mol. Cell 1:319-325. Yeh, W.C., Pompa, J.L., McCurrach, M.E., Shu, H.B., Elia, A.J., Shahinian, A., Ng, M., Wakeham, A., Khoo, W., Mitchell, K., El-Deiry, W.S., Lowe, S.W., Goeddel, D.V., and Mak, T. W. 1998. FADD: Essential for embryo development and signaling from some, but not all, inducers of apoptosis. Science 279:1954-1958. Zheng, L., Fisher, G., Miller, R.E., Peschon, J., Lynch, D.H., and Lenardo, M.J. 1995. Induction of apoptosis in mature T cells by tumour necrosis factor. Nature 377:348-351. Zou, H., Henzel, W.J., Liu, X., Lutschg, A., and Wang, X. 1997. Apaf-1, a human protein homologous to C. elegans CED-4, participates in cytochrome c–dependent activation of caspase3. Cell 90:405-413.
INTERNET RESOURCES http://www.apopnet.com A commercial site with links to tutorials, images, and reagent resources. http://www.access.digex.net/∼regulate/apolist.html A listing of apoptosis sites maintained by Trevigen, a biotechnology company.
Contributed by Richard M. Siegel and Michael J. Lenardo National Institute of Allergy & Infectious Diseases Bethesda, Maryland
Apoptosis Signaling Pathways
7.18.10 Supplement 21
Current Protocols in Cytometry
Flow Cytometry of Apoptosis
UNIT 7.19
This unit describes the most common methods applicable to flow cytometry that make it possible to: (1) identify and quantify dead or dying cells, (2) reveal a mode of cell death (apoptosis or necrosis), and (3) study mechanisms involved in cell death. Gross changes in cell morphology and chromatin condensation, which occur during apoptosis, can be detected by analysis with laser light beam scattering. An early event of apoptosis, dissipation of the mitochondrial transmembrane potential, can be measured using a number of fluorochromes that are sensitive to the electrochemical potential within this organelle (see Basic Protocol 1). Another early event of apoptosis, caspase activation, can be measured either directly, by immunocytochemical detection of the epitope that characterizes activated caspase (see Basic Protocol 2), or indirectly by immunocytochemical detection of the caspase-3 cleavage product, the p85 fragment of poly(ADP-ribose) polymerase (see Basic Protocol 4). Exposure of phosphatidylserine on the exterior surface of the plasma membrane can be detected by the binding of fluoresceinated annexin V (annexin V-FITC); this assay is combined with analysis of the exclusion of the plasma membrane integrity probe propidium iodide (PI; see Basic Protocol 5). Also described are methods of analysis of DNA fragmentation based either on DNA content of cells with fractional (“sub-G1”) DNA content (see Basic Protocol 6 and Alternate Protocol 1) or by DNA strand-break labeling (Terminal deoxynucleotidyltransferase–mediated dUTP Nick End Labeling, TUNEL; or In Situ End Labeling, ISEL; see Basic Protocol 7). Still another hallmark of apoptosis is the activation of tissue transglutaminase (TGase), the enzyme that crosslinks protein and thereby makes them less immunogenic. Methods for analyzing TGase activation are presented in Basic Protocol 8 and Alternate Protocol 2. STRATEGIC PLANNING The choice of a particular method often depends on the cell type, the nature of the inducer of apoptosis, the desired information (e.g., specificity of apoptosis with respect to the cell cycle phase or DNA ploidy), and technical restrictions. For example, sample transportation or prolonged storage before the measurement requires prior cell fixation, thereby eliminating the use of “supravital” methods that rely on analysis of freshly collected live cells. Positive identification of apoptotic cells is not always simple. Apoptosis was recently defined as a caspase-mediated cell death (Blagosklonny, 2000). Activation of caspases, therefore, appears to be the most specific marker of apoptosis (Shi, 2002). The detection of caspase activation, either directly (e.g., by antibody that is reactive with the activated enzyme; see Basic Protocol 2) or indirectly by the presence of poly(ADP-ribose) polymerase (PARP) cleavage product (PARP p85; see Basic Protocol 4), provides the most definitive evidence of apoptosis. Extensive DNA fragmentation is also considered as a specific marker of apoptosis. The number of DNA strand breaks in apoptotic cells is so large that intensity of their labeling in the TUNEL reaction (see Basic Protocol 7) ensures their positive identification and discriminates them from cells that have undergone primary necrosis (Gorczyca et al., 1992). However, in the instances of apoptosis when internucleosomal DNA degradation does not occur (Collins et al., 1992; Catchpoole and Stewart, 1993; Ormerod et al., 1994; Knapp et al., 1999), the number of DNA strand breaks may be inadequate to distinguish apoptotic cells by the TUNEL method. Likewise, in some instances of apoptosis, DNA fragmentation stops after the initial DNA cleavage to fragments of 50 to 300 kb (Collins et al., 1992, Oberhammer et al., 1993). The frequency of DNA strand breaks in nuclei of these cells is low, and therefore, they may not be easily detected by the TUNEL method. Contributed by Piotr Pozarowski, Jerzy Grabarek, and Zbigniew Darzynkiewicz Current Protocols in Cytometry (2003) 7.19.1-7.19.33 Copyright © 2003 by John Wiley & Sons, Inc.
Nucleic Acid Analysis
7.19.1 Supplement 25
The ability of cells to bind annexin V is still another marker considered to be specific to apoptosis. One should keep in mind, however, that use of the annexin V binding assay is hindered in some instances, e.g., when the plasma membrane is damaged during cell preparation or storage, leading to the loss of asymmetry in distribution of phosphatidylserine across the membrane. Furthermore, macrophages and other cells engulfing apoptotic bodies may also be positive in the annexin V assay (Marguet et al., 1999). Apoptosis can be recognized with greater certainty when the cells are subjected to several assays probing different apoptotic attributes (Hotz et al., 1994). For example, the assay of plasma membrane integrity (exclusion of PI) and annexin V binding combined with analysis of PARP cleavage or DNA fragmentation may provide a more definitive assessment of the mode of cell death than can be determined by each of these methods used alone. A plethora of kits designed to label DNA strand breaks and applicable to flow cytometry are available from different vendors. Most of these kits were designed by the authors (Gorczyca et al., 1992; Li and Darzynkiewicz, 1995). For example, Phoenix Flow Systems, BD PharMingen, and Alexis Biochemicals provide kits to identify apoptotic cells based on a single-step procedure utilizing either TdT and FITC-conjugated dUTP (APO-DIRECT; Li et al., 1995) or TdT and BrdUTP, as described in Basic Protocol 7 (APO-BRDU; Li and Darzynkiewicz, 1995). A description of the method, which is nearly identical to the one presented in this unit, is included with the kit. Another kit (ApopTag), based on a two-step DNA strand-break labeling with digoxygenin-16-dUTP by TdT, also designed by the authors (Gorczyca et al., 1992), was initially offered by ONCOR, later by Intergen, and most recently by Serologicals. BASIC PROTOCOL 1
MITOCHONDRIAL TRANSMEMBRANE POTENTIAL (∆ψm) MEASURED BY RHODAMINE 123 OR DiOC6(3) FLUORESCENCE The critical role of mitochondria during apoptosis is associated with the release of two intermembrane proteins, cytochrome c and apoptosis-inducing factor (AIF), that are essential for sequential activation of pro-caspase 9 and pro-caspase 3 (Liu et al., 1996; Yang et al., 1997). AIF is also involved in proteolytic activation of apoptosis-associated endonuclease (Susin et al., 1997). Still another protein, Smac/Diablo, that interacts with the inhibitors of caspases, thereby promoting apoptosis, is released from mitochondria (Deng et al., 2002). Dissipation (collapse) of mitochondrial transmembrane potential (∆ψm), also called the permeability transition (PT), likewise occurs early during apoptosis (Cossarizza et al., 1994; Kroemer, 1998; Zamzani et al., 1998). However, a growing body of evidence suggests that this event may be transient when associated with the release of cytochrome c or AIF, and mitochondrial potential may be restored for some time in the cells with activated caspases (Finucane et al., 1999; Scorrano et al., 1999; Li et al., 2000).
Flow Cytometry of Apoptosis
The membrane-permeable lipophilic cationic fluorochromes such as rhodamine 123 (R123) or 3,3′-dihexyloxacarbocyanine iodide [DiOC6(3)] can serve as probes of ∆ψm (Darzynkiewicz et al., 1981, 1982; Johnson et al., 1980). When live cells are incubated in their presence, the probes accumulate in mitochondria, and the extent of their uptake, as measured by intensity of cellular fluorescence, reflects ∆ψm. A combination of R123 and PI discriminates among live cells that stain only with R123, early apoptotic cells that have lost the ability to accumulate R123, and late apoptotic/necrotic cells whose plasma membrane integrity is compromised and that stain only with PI (Darzynkiewicz et al., 1982; Darzynkiewicz and Gong, 1994). The specificity of R123 and DiOC6(3) as ∆ψm probes is increased when they are used at low concentrations (<0.5 µg/ml). Still another probe of ∆ψm is the J-aggregate-forming lipophilic cationic fluorochrome 5,5′,6,6′tetrachloro-1,1′,3,3′-tetraethylbenzimidazolylcarbocyanine iodide (JC-1). Its uptake by
7.19.2 Supplement 25
Current Protocols in Cytometry
charged mitochondria driven by the transmembrane potential is detected by the shift in color of fluorescence from green, which is characteristic of its monomeric form, to orange, which reflects its aggregation in mitochondria (Cossarizza and Salvioli, 2001; UNIT 9.14). In light of the recent evidence that the collapse of ∆ψm may not be a prerequisite for release of cytochrome c, AIF, and other apoptotic events (Finucane et al., 1999; Scorrano et al., 1999; Li et al., 2000), one should be cautious in interpreting the lack of collapse of ∆ψm as a marker of non-apoptotic cells. Materials Cells of interest in appropriate complete culture medium 10 µM rhodamine 123 (R123; see recipe) or 10 µM DiOC6(3) (see recipe for 0.1 mM stock solution) or 0.2 mM JC-1 stock solution (see recipe) Phosphate-buffered saline (PBS; APPENDIX 2A) 1 mg/ml propidium iodide (PI; Molecular Probes) in distilled water; store at 4°C in the dark 12 × 75–mm tubes suitable for flow cytometer Flow cytometer with 488-nm excitation and filters for collection of green, orange, and red fluorescence Stain with R123 or DiOC6(3) and PI 1a. Add either 20 µl of 10 µM R123 (200 nM final) or 5 µl of 10 µM DiOC6(3) (50 nM final) to ∼106 cells suspended in 1 ml complete tissue culture medium (with 10% serum), and incubate 20 min at 37°C in the dark. 2a. Centrifuge cells 5 min at 300 × g, room temperature. Resuspend cell pellet in 1 ml PBS. 3a. Add 10 µl PI solution and incubate 5 min at room temperature in the dark. 4a. Analyze cell fluorescence on the flow cytometer. Excite fluorescence with blue (488-nm) laser. Set the signal-triggering threshold on forward- and side-scatter signals. Collect green fluorescence [R123 or DiOC6(3)] at 530 ± 20 nm and red fluorescence (PI) above 600 nm. Stain with JC-1 1b. Suspend cell pellet (∼106 cells) in 1 ml complete tissue culture medium with 10% serum. 2b. Add 10 µl of 0.2 mM JC-1 stock solution. Vortex cells intensely during addition and for the next 20 sec. Wash cells two times with PBS; centrifuge each time 5 min at 200 × g, room temperature. Addition of JC-1 to the cell suspension without vortexing may lead to formation of precipitate. Vortexing too vigorously, on the other hand, may cause cell damage.
3b. Incubate cells 15 min at room temperature in the dark. 4b. Analyze cell fluorescence on the flow cytometer, using 488-nm excitation. Collect green fluorescence at 530 ± 20 nm and orange fluorescence at 570 ± 20 nm with a band-pass filter or above 570 nm with a long-pass filter.
Nucleic Acid Analysis
7.19.3 Current Protocols in Cytometry
Supplement 25
BASIC PROTOCOL 2
IMMUNOCYTOCHEMICAL DETECTION OF ACTIVATED CASPASES BY ZENON TECHNOLOGY Caspases are activated by transcatalytic cleavage of their zymogen procaspase molecules into large and small subunits. The subunits then assemble to form the heterotetramer consisting of two small and two large subunits, which is the active caspase (Budihardjo et al., 1999; Earnshaw et al., 1999). Antibodies that are specific to activated caspase-3, caspase-8, and caspase-9 are now commercially available and one expects that antibodies reactive with other active caspases will soon be developed as well. It is possible, therefore, to detect caspase activation by immunocytochemical means. This protocol combines the use of activated caspase-specific antibody with staining of cellular DNA by propidium iodide (PI) to concurrently detect cells with activated caspases and relate caspase activation to the cell-cycle position. The immunocytochemical detection of caspase-3 in this protocol makes use of Zenon technology (Haugland, 2002). Zenon technology consists of a labeling complex that is formed by a fluorochrome-labeled Fab fragment (Zenon Alexa Fluor 488) of an anti-IgG antibody that is directed against the Fc portion of a mouse (or rabbit) IgG1 antibody. Mixing of the labeled Fab fragment with the primary antibody forms the labeling complex. Excess unbound labeled Fab fragments is removed by admixture of nonspecific mouse (or rabbit) IgG. The labeling complex is then used to stain cells in the same manner as a covalently labeled primary antibody (Haugland, 2002). Materials Cells of interest (see APPENDIX 3B for culture techniques), both untreated (control) and induced to apoptosis (e.g., exponentially growing HL-60 cells incubated 2 to 4 hr with 0.15 µM camptothecin) Phosphate-buffered saline (PBS; APPENDIX 2A) Fixatives: 1% (v/v) methanol-free formaldehyde (Polysciences) in PBS, 0° to 5°C 4% (v/v) methanol-free formaldehyde (Polysciences) in PBS, room temperature 70% (v/v) ethanol, –20°C Rinse solution (see recipe) Primary antibody: cleaved (activated) caspase-3 antibody (Cell Signaling Technology, cat. no. 9661) Zenon Alexa Fluor 488 rabbit IgG labeling kit (Molecular Probes, cat. no. Z-25302) 10% (v/v) Triton X-100 in PBS DNA staining solution with PI (see recipe) 12 × 75–ml tubes suitable for use on flow cytometer Flow cytometer with 488-nm excitation and filters for collection of green and red fluorescence 1. Suspend ∼106 cells in 0.5 ml PBS. 2. Fix cells by transferring the above cell suspension into tubes containing 4.5 ml of 1% methanol-free formaldehyde in PBS at 0° to 5°C. Let stand 15 min at 0° to 5°C. 3. Centrifuge 5 min at 300 × g, room temperature. Decant supernatant. 4. Resuspend cell pellet in 3 ml of 70% ethanol at –20°C. Allow to sit at least 2 hr (cells can be stored several days in 70% ethanol at –20°C). 5. Bring the cell suspension in 70% ethanol to room temperature, add 2 ml PBS to this suspension, and let sit 5 min at room temperature.
Flow Cytometry of Apoptosis
6. Centrifuge 5 min at 300 × g, room temperature. Decant supernatant.
7.19.4 Supplement 25
Current Protocols in Cytometry
7. Resuspend cell pellet in 5 ml PBS and let sit 5 min at room temperature. 8. Centrifuge 5 min at 300 × g, room temperature. Decant supernatant. 9. Resuspend cell pellet in rinse solution. Let stand 30 min at room temperature. 10. Prepare the staining solution as follows. a. Mix 4 µl primary antibody (anti-caspase-3) with 16 µl rinse solution and with 5 µl solution A Zenon (from kit) in a 1.5-ml microcentrifuge tube. b. Keep 5 min in the dark at room temperature. c. Add 5 µl solution B Zenon (from kit). d. Keep 5 min in the dark at room temperature. e. Add 0.3 µl of 10% Triton X-100 in PBS. 11. Centrifuge the cell suspension (from step 9) 5 min at 300 × g, room temperature. Thoroughly drain the rinse solution by blotting on filter paper. Add 15 µl of the staining solution prepared in step 10, and 85 µl rinse solution, for a final volume of 100 µl. Resuspend the cell pellet. 12. Incubate cells with the staining solution 1 hr in the dark at room temperature. 13. Add 5 ml PBS, centrifuge 5 min at 300 × g, room temperature, and decant supernatant. 14. Resuspend cell pellet in 1 ml of 4% methanol-free formaldehyde in PBS and let stand 5 min at room temperature. 15. Centrifuge cells 5 min at 300 × g, room temperature. Decant supernatant. 16. Resuspend cell pellet in 1 ml DNA staining solution with PI. 17. Analyze cell fluorescence on the flow cytometer, using 488-nm excitation (or a mercury arc lamp with a BG12 filter). Collect green Alexa 488 fluorescence at 530 ±20 nm and red PI fluorescence above 600 nm. DETECTION OF APOPTOTIC CELLS USING FLUOROCHROME-LABELED INHIBITORS OF CASPASES (FLICAs)
BASIC PROTOCOL 3
Exposure of live cells to fluorochrome-labeled inhibitors of caspases (FLICAs) results in uptake of these reagents by apoptotic cells (Smolewski et al., 2001). Unbound FLICAs are removed from the nonapoptotic cells by rinsing the cells with wash buffer. The cells may also be fixed with formaldehyde; after fixation only apoptotic cells retain the label. Cells labeled with FLICAs can be examined by fluorescence microscopy, or their fluorescence can be measured by flow cytometry. FLICAs are convenient markers of apoptotic cells, and when used in combination with PI as described in the protocol below, they reveal three sequential stages of apoptosis. FAM-VAD-FMK, the inhibitor designed to react with all caspases, except perhaps caspase-2, is used in this protocol. It should be stressed, however, that FLICAs appear to react in apoptotic cells also with targets other than activated caspases. Cell labeling with FLICAs, therefore, although perhaps reflecting caspase activity, although reflecting caspase activation, cannot be interpreted as indicating reactivity with active enzyme centers of caspases only.
Nucleic Acid Analysis
7.19.5 Current Protocols in Cytometry
Supplement 25
Materials Cells of interest (see APPENDIX 3B for culture techniques) Medium supplemented with 10% (v/v) serum or 1% (w/v) serum albumin FLICA kit (Immunochemistry Technologies) containing: FAM-VAD-FMK reagent (see recipe) Fixative Hoechst stain Rinse solution: 1% (w/v) BSA in PBS (APPENDIX 2A) 1 mg/ml propidium iodide (PI; Molecular Probes) in distilled water; store at 4°C in the dark 12 × 75–ml tubes suitable for use on flow cytometer Flow cytometer with 488-nm excitation and filters for collection of green and red fluorescence 1. Suspend ∼106 cells in 0.3 ml medium containing 10% serum or 1% serum albumin. 2. Add 10 µl FAM-VAD-FMK working solution to this cell suspension (final concentration 10 µM). Mix gently and incubate 1 hr at 37°C. Sulforhodamine-labeled FLICA (SR-VAD-FMK) may be used instead of FAM-VAD-FMK to make apoptotic cells fluoresce in the red.
3. Add 2 ml rinse solution, mix gently, and centrifuge 5 min at 200 × g, room temperature. 4. Resuspend cell pellet in 2 ml rinse solution and centrifuge as in step 3. Cells may be fixed 15 min in 1% formaldehyde in PBS, then suspended in 70% ethanol and stored for several days. A fluorochrome of a different color than FLICA may be used to counterstain other cellular components (e.g., DNA) or other markers of apoptosis (e.g., DNA strand breaks).
5. Resuspend cell pellet in 1 ml rinse solution. Add 1.0 µl of 1 mg/ml PI stock solution. Keep 5 min at room temperature. Protect samples from light at all times. Staining with PI is optional. It allows one to distinguish the cells that have compromised plasma-membrane integrity (e.g., necrotic and late apoptotic cells, cells with mechanically damaged membranes, or isolated cell nuclei) to the extent that they cannot exclude PI.
6. Analyze cell fluorescence on the flow cytometer, using 488-nm excitation (or a mercury arc lamp with a BG12 filter). Collect green FAM-VAD-FMK fluorescence at 530 ± 20 nm and red PI fluorescence above 600 nm. BASIC PROTOCOL 4
DETERMINATION OF POLY(ADP-RIBOSE) POLYMERASE (PARP) CLEAVAGE PARP is a nuclear enzyme that is involved in DNA repair and activated in response to DNA damage (de Murcia and Menissier-de Murcia, 1994). Early in apoptosis, PARP is cleaved by caspases, primarily by caspase-3 (Kaufmann et al., 1993; Lazebnik et al., 1994; Alnemri et al., 1996). The specific cleavage of this protein results in distinct 85-kDa and 24-kDa fragments, usually detected electrophoretically, and is considered to be a hallmark of the apoptotic mode of cell death.
Flow Cytometry of Apoptosis
The development of antibodies that recognize the cleaved PARP products prompted their use as immunocytochemical markers of apoptotic cells. The antibody that recognizes the 85-kDa fragment (PARP p85) was initially used to score the frequency of apoptosis in tissue sections (Sallman et al., 1997; Kockx et al., 1998). Recently, this antibody has been adapted to label apoptotic cells for detection by flow cytometry and laser scanning
7.19.6 Supplement 25
Current Protocols in Cytometry
cytometry (LSC; Li and Darzynkiewicz, 2000; Li et al., 2000). A good correlation was observed between the frequency of apoptosis detected cytometrically with PARP p85 Ab and that detected by the DNA strand-break (TUNEL) assay. However, at least in some cell systems, the cleavage of PARP occurs prior to the onset of DNA fragmentation (Li and Darzynkiewicz, 2000). In these instances, the correlation may not be apparent at early stages of apoptosis because the apoptotic index estimate based on PARP cleavage may exceed the estimate based on the TUNEL reaction. Cytometric analysis of cells differentially stained for PARP p85 and DNA, similar to the TUNEL assay, makes it possible not only to identify and score apoptotic cell populations, but also to correlate apoptosis with the cell cycle position or DNA ploidy. The classic immunocytochemical indirect (two-step) method to detect the 85-kDa PARP fragment is presented below. Alternatively, however, one can use the Zenon technology as described above (see Basic Protocol 2) for detection of activated caspases. Materials Cells of interest Phosphate-buffered saline (PBS; APPENDIX 2A) 1% methanol-free formaldehyde (Polysciences) in PBS (APPENDIX 2A) 70% ethanol 0.25% (v/v) Triton X-100 (Sigma) in PBS (APPENDIX 2A); store at 4°C PBS/BSA solution: 1% (w/v) bovine serum albumin (Sigma) in PBS; store at 4°C Anti-PARP p85 antibody (Promega anti-PARP-85 fragment, rabbit polyclonal) Fluorescein-conjugated anti-rabbit immunoglobulin antibody (Dako) 1 mg/ml propidium iodide (PI; Molecular Probes) in distilled water; store at 4°C in the dark RNase solution (APPENDIX 2A) 12 × 75–mm centrifuge tubes suitable for use on the flow cytometer Pasteur pipets Flow cytometer with 488-nm excitation and filters for collection of green and red fluorescence 1. Suspend ∼106 cells in 0.5 ml PBS. Transfer this suspension to a 12 × 75–mm (5-ml) tube containing 4.5 ml of 1% methanol-free formaldehyde and fix cells 15 min on ice. Centrifuge the cells 5 min at 300 × g, 4°C, wash once with 5 ml PBS, centrifuge 5 min at 300 × g, and resuspend the cell pellet in 0.5 ml PBS. With a Pasteur pipet, transfer this cell suspension into a new 12 × 75–mm centrifuge tube containing 4.5 ml of ice-cold 70% ethanol. The cells may be stored several days in ethanol at −20°C.
2. Centrifuge cells 5 min at 200 × g, room temperature, and resuspend the cell pellet in 5 ml PBS; repeat centrifugation. 3. Resuspend cells in 5 ml 0.25% Triton X-100/PBS for 10 min. 4. Centrifuge cells 5 min at 300 × g, room temperature, and resuspend in 2 ml BSA/PBS solution for 10 min. 5. Centrifuge cells 5 min at 300 × g, room temperature, and resuspend in 100 µl BSA/PBS containing anti-PARP p85 Ab diluted 1:200. Incubate 2 hr at room temperature, or overnight at 4°C. 6. Add 5 ml BSA/PBS solution, let sit 5 min, and centrifuge 5 min at 300 × g, room temperature.
Nucleic Acid Analysis
7.19.7 Current Protocols in Cytometry
Supplement 25
7. Resuspend cell pellet in 100 µl PBS/BSA containing fluorescein-conjugated secondary Ab [F(ab′)2 fragment, swine anti-rabbit immunoglobulin] diluted 1:30. Incubate 1 hr in the dark at room temperature. 8. Add 5 ml BSA/PBS, centrifuge 5 min at 200 × g, room temperature, and resuspend cell pellet in 1 ml PBS. Add 20 µl of 1 mg/ml PI and 50 µl RNase stock solution. Incubate 20 min in the dark at room temperature. 9. Analyze cell fluorescence on the flow cytometer, using 488-nm excitation (or a mercury arc lamp with a BG12 filter). Collect green FITC-anti PARP p85 fluorescence at 530 ± 20 nm and red PI fluorescence above 600 nm. BASIC PROTOCOL 5
ANNEXIN V BINDING Phospholipids of the plasma membrane are asymmetrically distributed between the inner and outer leaflets of the membrane. Phosphatidylcholine and sphingomyelin are exposed on the external leaflet of the lipid bilayer, while phosphatidylserine is located on the inner surface. During apoptosis, this asymmetry is disrupted and phosphatidylserine becomes exposed on the outside surface of the plasma membrane (Fadok et al., 1992; Koopman et al., 1994; van Engeland et al., 1998). Because the anticoagulant protein annexin V binds with high affinity to phosphatidylserine, fluorochrome-conjugated annexin V has found an application as a marker of apoptotic cells, in particular for their detection by flow cytometry (van Engeland et al., 1998). The cells become reactive with annexin V prior to their loss of plasma membrane ability to exclude cationic dyes such as PI. Therefore, by staining cells with a combination of annexin V-FITC and PI, it is possible to detect unaffected, non-apoptotic cells (annexin V-FITC negative/PI negative), early apoptotic cells (annexin V-FITC positive/PI negative), and late apoptotic (“necrotic stage” of apoptosis) as well as necrotic cells (PI positive). Materials Cells of interest Fluorescein-conjugated annexin V (see recipe) in binding buffer (see recipe) 1 mg/ml propidium iodide (PI; Molecular Probes) in distilled water; store at 4°C in the dark Flow cytometer with 488-nm excitation and filters for collection of green and red fluorescence 1. Suspend 105 to 106 cells in 1 ml fluorescein-conjugated annexin V in binding buffer for 5 min at room temperature in the dark. 2. Prior to analysis, add an appropriate volume of 1 mg/ml PI solution to the cell suspension to have a final PI concentration of 1.0 µg/ml. Incubate 5 min at room temperature in the dark. 3. Analyze cells on the flow cytometer, using 488-nm excitation. Set gates based on light scatter. Collect green annexin V fluorescence at 530 ± 20 nm and red PI fluorescence above 600 nm.
Flow Cytometry of Apoptosis
7.19.8 Supplement 25
Current Protocols in Cytometry
DNA FRAGMENTATION: DETECTION OF CELLS WITH FRACTIONAL (SUB-G1) DNA CONTENT USING PI
BASIC PROTOCOL 6
Endonucleases activated during apoptosis target internucleosomal DNA sections and cause extensive DNA fragmentation (Kerr et al., 1972; Arends et al., 1990; Nagata, 2000). The fragmented, low-molecular-weight DNA can be extracted from the cells following their fixation in precipitating fixatives such as ethanol. Conversely, fixation with crosslinking fixatives such as formaldehyde results in the retention of low-molecularweight DNA in the cell and therefore should be avoided. Generally, the extraction occurs during the process of cell staining in aqueous solutions after transfer from the fixative. As a result, apoptotic cells often end up with deficient DNA content, and when stained with a DNA-specific fluorochrome, they can be recognized by cytometry as cells having less DNA than G1 cells. On the DNA content frequency histograms, they form a characteristic “sub-G1 ” peak (Umansky et al., 1981; Nicoletti et al., 1991; Gong et al., 1994). It should be noted that loss of DNA may also occur as a result of the shedding of apoptotic bodies containing fragments of nuclear chromatin. The degree of DNA degradation varies depending on the stage of apoptosis, cell type, and often the nature of the apoptosis-inducing agent. Hence, the extractability of DNA during the staining procedure also varies. A high-molarity phosphate-citrate buffer enhances extraction of the fragmented DNA (Gong et al., 1994). With some limitations, this approach can be used to extract DNA from apoptotic cells to the desired level in order to achieve their optimal separation by flow cytometry. Materials Cells of interest PBS (APPENDIX 2A) 70% ethanol DNA extraction buffer (see recipe) DNA staining solution with PI (see recipe) 12 × 75–mm tubes suitable for use on the flow cytometer Flow cytometer with 488-nm excitation and filter for collection of red fluorescence 1. Suspend 1-2 × 106 cells in 0.5 ml PBS and fix cells by adding suspension to 4.5 ml of 70% ethanol in a 12 × 75–mm tube on ice. Cells may be stored several weeks in fixative at −20°C.
2. Centrifuge cells 5 min at 200 × g, decant ethanol, suspend the cell pellet in 5 ml PBS, and centrifuge 5 min at 300 × g, room temperature. 3. Suspend cell pellet in 0.25 ml PBS. To facilitate extraction of low-molecular-weight DNA, add 0.2 to 1.0 ml DNA extraction buffer. 4. Incubate 5 min at room temperature, then centrifuge 5 min at 300 × g, room temperature. 5. Suspend cell pellet in 1 ml DNA staining solution with PI. Incubate cells 30 min at room temperature in the dark. 6. Analyze cells on the flow cytometer, using 488-nm excitation (or a mercury arc lamp with a BG12 filter). Collect forward light scatter and red fluorescence above 600 nm.
Nucleic Acid Analysis
7.19.9 Current Protocols in Cytometry
Supplement 25
ALTERNATE PROTOCOL 1
DNA FRAGMENTATION: DETECTION OF CELLS WITH FRACTIONAL (“SUB-G1”) DNA CONTENT USING DAPI Cellular DNA may be stained with other fluorochromes instead of PI, and other cell constituents may be counterstained in addition to DNA. The following is the procedure used to stain DNA with DAPI. This protocol requires a UV laser. Additional Materials (also see Basic Protocol 6) DNA staining solution with DAPI (see recipe) Flow cytometer equipped with UV excitation and filter for collection of blue fluorescence 1. Follow Basic Protocol 6, steps 1 to 4. Then, suspend the cell pellet in 1 ml DNA staining solution containing DAPI. Keep on ice 20 min. 2. Analyze cells on the flow cytometer, using UV excitation (e.g., 351-nm line from an argon-ion laser, or mercury lamp with a UG1 filter). Collect blue DAPI fluorescence in a band from 460 to 500 nm.
BASIC PROTOCOL 7
DNA FRAGMENTATION: DETECTION OF DNA STRAND BREAKS (TUNEL ASSAY) DNA fragmentation during apoptosis, particularly when it progresses to internucleosomal regions (Arends et al., 1990; Oberhammer et al., 1993), generates a multitude of DNA strand breaks in the nucleus. The 3′-OH ends of the breaks can be detected by attaching a fluorochrome. This is generally done directly or indirectly (e.g., via biotin or digoxygenin) using fluorochrome-labeled deoxynucleotides in a reaction catalyzed preferably by exogenous terminal deoxynucleotidyltransferase (TdT; Gorczyca et al., 1992, 1993; Li and Darzynkiewicz, 1995; Li et al., 1995). The reaction is commonly known as TUNEL, from TdT-mediated dUTP-biotin nick-end labeling (Gavrieli et al., 1992). This acronym is actually a misnomer, since double-stranded DNA breaks are labeled, rather than single-stranded nicks. Of all the markers used to label DNA breaks, BrdUTP appears to be the most advantageous with respect to sensitivity, low cost, and simplicity of reaction (Li and Darzynkiewicz, 1995). When attached to DNA strand breaks in the form of poly-BrdU, this deoxynucleotide can be detected with a FITC-conjugated anti-BrdU Ab, the same Ab commonly used to detect BrdU incorporated during DNA replication (UNIT 7.4). Poly-BrdU attached to DNA strand breaks by TdT, however, is accessible to the Ab without the need for DNA denaturation, which otherwise is required to detect the precursor incorporated during DNA replication (UNIT 7.7). It should be stressed that the detection of DNA strand breaks by this method requires pre-fixation of cells with a crosslinking agent such as formaldehyde. Unlike ethanol, formaldehyde prevents the extraction of small pieces of fragmented DNA. Thus, despite cell permeabilization and the subsequent cell washings required, the DNA content of early apoptotic cells (and the number of DNA strand breaks) is not markedly diminished through extraction. Labeling of DNA strand breaks in this procedure, which utilizes fluorescein-conjugated anti-BrdU Ab, can be combined with staining of DNA by a fluorochrome of another color (PI, red fluorescence). Cytometry of cells that are differentially stained for DNA strand breaks and for DNA allows one to distinguish apoptotic from non-apoptotic cell subpopulations and reveals the cell cycle distribution in these subpopulations (Gorczyca et al., 1992, 1993)
Flow Cytometry of Apoptosis
7.19.10 Supplement 25
Current Protocols in Cytometry
Materials Cells of interest 1% (v/v) methanol-free formaldehyde (Polysciences) in PBS (APPENDIX 2A), pH 7.4 (primary fixative) PBS (APPENDIX 2A) 70% ethanol (secondary fixative) 5× TdT reaction buffer (see recipe) 2 mM BrdUTP (Sigma) in 50 mM Tris⋅Cl, pH 7.5 TdT in storage buffer (both from Roche Diagnostics), 25 U in 1 µl 10 mM cobalt chloride (CoCl2; Roche Diagnostics) Rinsing buffer: PBS with 0.1% (v/v) Triton X-100 and 0.5% (w/v) BSA FITC-conjugated anti-BrdU MAb in PBS (APPENDIX 2A; see recipe) PI staining buffer: PBS with 5 µg/ml PI and 200 µg/ml DNase-free RNase Flow cytometer with 488-nm excitation and filters for collection of green and red fluorescence Fix cells 1. Fix 1-5 × 106 cells in suspension 15 min in 1% methanol-free formaldehyde in PBS, pH 7.4, on ice. 2. Centrifuge 5 min at 200 × g, 4°C, resuspend cell pellet (∼2 × 106 cells) in 5 ml PBS, centrifuge 5 min at 200 × g, 4°C, and resuspend cells in 0.5 ml PBS. 3. Add the 0.5-ml cell suspension from step 2 to 5 ml ice-cold 70% ethanol. The cells can be stored several weeks in ethanol at −20°C.
4. Centrifuge 5 min at 200 × g, 4°C, remove ethanol, and resuspend cells in 5 ml PBS. Repeat centrifugation. Stain cells 5. Resuspend the pellet (not more than 106 cells) in 50 µl of a solution that contains: 10 µl 5× TdT reaction buffer 2.0 µl 2 mM BrdUTP stock solution 0.5 µl TdT in storage buffer (12.5 U final) 5 µl 10 mM CoCl2 solution 32.5 µl dH2O. 6. Incubate cells in this solution 40 min at 37°C. Alternatively, incubation can be carried out overnight at 22° to 24°C.
7. Add 1.5 ml rinsing buffer and centrifuge 5 min at 200 × g, room temperature. 8. Resuspend cells in 100 µl FITC-conjugated anti-BrdU MAb solution. 9. Incubate 1 hr at room temperature or overnight at 4°C. Add 2 ml rinsing buffer and centrifuge 5 min at 200 × g, room temperature. 10. Resuspend cell pellet in 1 ml PI staining solution containing RNase. Incubate 30 min at room temperature in the dark. 11. Measure cell fluorescence on the flow cytometer, using 488-nm excitation (or a mercury arc lamp with a BG12 filter). Collect green FITC-anti BrdU MAb fluorescence at 530 ± 20 nm and red PI fluorescence above 600 nm. Nucleic Acid Analysis
7.19.11 Current Protocols in Cytometry
Supplement 25
BASIC PROTOCOL 8
DETECTION OF TISSUE TRANSGLUTAMINASE ACTIVATION BY CELL RESISTANCE TO DETERGENTS Extensive protein crosslinking takes place during apoptosis. The ubiquitous transglutaminase TGase 2 (also called “tissue transglutaminase”; tTGase) was identified as the enzyme responsible for this reaction (Fesus et al., 1987; Melino and Piacentini, 1998). It is presumed that activation of TGase 2 during apoptosis prevents release of soluble and immunogenic proteins from dying cells because protein crosslinking makes these proteins less soluble, and thereby decreases a possibility of induction of autoimmune reaction. Furthermore, protein packaging into apoptotic bodies may be facilitated when proteins remain in solid state rather than in solution. The additional role of TGase 2 as one of the “executor enzymes” during apoptosis is still being debated. This protocol is a simple and rapid approach to identify apoptotic cells with activated TGase 2. The method is based on the propensity of crosslinked protein to withstand treatment with detergents. The authors have noticed that when live, nonapoptotic cells are subjected to treatment with solutions of nonionic detergents, lysis of their plasma membrane and release of the content of cytoplasm is complete, resulting in preparation of isolated nuclei. In contrast, apoptotic cells resist the detergent treatment; their cytoplasmic protein remains insoluble, attached to the nucleus in the form of a shell-like cover (Grabarek et al., 2002). It is possible, therefore, by flow or laser scanning cytometry to distinguish apoptotic cells from the nuclei isolated from nonapoptotic cells, by means of the abundance of protein in the former. In addition, bivariate gating analysis of cellular DNA and protein content makes it possible to reveal the cell cycle distribution separately for the population of cells with crosslinked protein (activated TGase 2) and for the population of cells that did not show protein crosslinking (Grabarek et al., 2002). Alternate Protocol 2 combines the detection of TGase 2 activity by binding of fluoresceinated cadaverine (F-CDV) with analysis of the cell cycle. Materials Cells of interest DAPI/sulforhodamine 101/detergent solution (see recipe) Flow cytometer equipped with UV excitation and filters for collection of blue and red fluorescence 1. Collect ∼106 cells from the culture and centrifuge 5 min at 300 × g, room temperature. 2. Suspend the cell pellet in 1 ml DAPI/sulforhodamine 101/detergent solution and vortex 20 sec. 3. Analyze cells on the flow cytometer, using UV excitation (e.g., 351-nm line from an argon-ion laser, or mercury lamp with a UG1 filter). Collect blue DAPI fluorescence in a band from 460 to 500 nm and red fluorescence of sulforhodamine 101 above 600 nm.
ALTERNATE PROTOCOL 2
DETECTION OF TGase 2 ACTIVATION BY FLUORESCEINATED CADAVERINE (F-CDV) BINDING This alternate protocol is based on the covalent attachment, by the activated TGase 2, of the fluorescein-tagged cadaverine to the respective protein substrates within the cell (Lajemi et al., 1998). This assay was adapted to flow cytometry and combined with concurrent analysis of cellular DNA content (Grabarek et al., 2002). Like the detergentbased assay, this method is simple and also offers good distinction between apoptotic cells with activated versus nonactivated TGase 2.
Flow Cytometry of Apoptosis
7.19.12 Supplement 25
Current Protocols in Cytometry
It should be noted that when the cost of the reagents for the procedure is taken into account, the detergent-based assay (see Basic Protocol 8) is less expensive by several orders of magnitude than this F-CDV assay. Materials Fluoresceinated cadaverine solution (F-CDV; see recipe) Cells of interest 100% methanol DNA staining solution with PI (see recipe) Flow cytometer with 488-nm excitation and filters for collection of green and red fluorescence 1. Add aliquots of F-CDV stock solution (1 part per 50) directly to cell cultures (106 to 107 cells) to obtain 50 µM final F-CDV concentration in the culture. 2. Incubate cells in the presence of F-CDV for the desired time interval during which activity of TGase 2 has to be detected (e.g., one to several hours). Because crosslinking is a cumulative process, intensity of cell labeling increases with time of incubation.
3. Harvest the culture by centrifuging 5 min at 200 × g, room temperature. 4. Suspend the cells in 0.5 ml PBS and fix in 5 ml of 100% methanol on ice. Keep in methanol ≥2 hr at 0° to 4°C; cells may be stored in the fixative for several days. 5. Centrifuge cells 5 min at 200 × g, room temperature, decant the fixative thoroughly, and suspend cell pellet in 2 ml DNA staining solution with PI. 6. Keep ≥30 min at room temperature. 7. Measure cell fluorescence on the flow cytometer, using 488-nm excitation (or a mercury arc lamp with a BG12 filter). Collect green FITC-CDV fluorescence at 530 ± 20 nm and red PI fluorescence above 600 nm. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Binding buffer 10 mM HEPES-NaOH, pH 7.4 140 mM NaCl 5 mM CaCl2 May be stored several weeks at 4°C DAPI/sulforhodamine 101/detergent solution Dissolve 100 µg DAPI, 2 mg sulforhodamine 101 (Molecular Probes), and 0.1 ml Triton X-100 in 100 ml PBS. Triton X-100 may be replaced by 0.1 ml Nonidet NP-40. This solution may be stored several weeks at 0° to 4°C. DiOC6 (3) stock solution Prepare a 0.1 mM solution of DiOC6 (3) (Molecular Probes) by dissolving 5.7 mg dye in 10 ml dimethyl sulfoxide (DMSO). May be stored for weeks in small (e.g., 0.5- or 1-ml) aliquots protected from light at −20°C. Prior to use, dilute ten-fold with PBS to obtain 10 µM concentration. Nucleic Acid Analysis
7.19.13 Current Protocols in Cytometry
Supplement 25
DNA extraction buffer 192 ml 0.2 M Na2HPO4 8 ml 0.1 M citric acid pH 7.8 May be stored for months at 4°C DNA staining solution with DAPI Dissolve 100 µg DAPI in 100 ml PBS. Can be stored at 0° to 4°C for several weeks. DNA staining solution with PI 200 µg propidium iodide (PI) 2 mg DNase-free RNase 10 ml PBS Prepare fresh before use FAM-VAD-FMK Stock solution: following kit directions, dissolve FAM-VAD-FMK in dimethyl sulfoxide (DMSO) to obtain 150× solution. Store aliquots protected from light ≤3 months at −20°C. Working solution: Just prior to use, prepare a 30× solution by diluting FAM-VADFMK stock solution 1:5 in PBS. Mix the vial until the solution becomes transparent and homogenous. Protect all FAM-VAD-FMK solutions from light. Discard unused portions. Do not store. FITC-conjugated anti-BrdU MAb solution 100 µl PBS (APPENDIX 2A) 0.3 µg FITC-conjugated anti-BrdU MAb (Becton Dickinson) 0.3% (v/v) Triton X-100 1% (w/v) BSA Prepare fresh before use Fluorescein-conjugated annexin V Dissolve FITC-conjugated annexin V (1:1 stoichiometric complex; e.g., from BRAND Applications) in binding buffer (see recipe) at a concentration of 1.0 µg/ml. This solution must be prepared fresh just prior to use. Fluoresceinated cadaverine solution Dissolve 5-[(5-aminopentyl)thioureidyl]fluorescein dihydrobromide (F-CDV; Molecular Probes) in distilled water to obtain a 2.5 mM stock solution. Aliquots (0.2 to 0.5 ml) of this solution may be stored several weeks at −20°C. JC-1 stock solution Prepare a 0.2 mM solution of JC-1 (Molecular Probes) by dissolving 1.3 mg dye in 10 ml N,N-dimethylformamide (Sigma). May be stored for weeks in small (e.g., 0.5or 1.0-ml) aliquots protected from light at −20°C. Use glass containers; N,N-dimethylformamide will dissolve plastics. Rhodamine 123 (R123) stock solution Prepare a 0.1 mM solution of R123 (Molecular Probes) by dissolving 0.38 mg dye in 10 ml methanol. May be stored for months in small aliquots protected from light at −20°C. Prior to use, dilute ten-fold with PBS to obtain 10 µM concentration.
Flow Cytometry of Apoptosis
7.19.14 Supplement 25
Current Protocols in Cytometry
Rinse solution 1 g BSA 0.1 ml Triton X-100 100 ml PBS (APPENDIX 2A) Store up to 1 week at 4°C TdT reaction buffer, 5× 1 M potassium (or sodium) cacodylate 125 mM Tris⋅Cl, pH 6.6 (APPENDIX 2A) 0.125% bovine serum albumin (BSA) May be stored for weeks at 4°C COMMENTARY Background Information Applications of flow cytometry in cell necrobiology have two distinct goals (for reviews, see Darzynkiewicz et al., 1992, 1997; Ormerod, 1998; van Engeland et al., 1998; Vermes et al., 2000). One goal is to elucidate molecular and functional mechanisms associated with cell death. For this purpose, flow cytometry is used to measure cellular levels of components involved in the regulation and/or execution of apoptosis or cell necrosis. The most frequently studied are pro- and anti-apoptotic members of the Bcl-2 protein family, caspases, proto-oncogenes (e.g., c-myc or ras), or tumor suppressor genes (e.g., p53 or pRB). Flow cytometry is also widely used to study functional attributes of the cell such as mitochondrial metabolism, oxidative stress, intracellular pH, or ionized calcium, all closely associated with mechanisms of apoptosis. The major advantage of flow cytometry in these applications is that it provides the possibility of multiparametric measurements of cell attributes. Multivariate analysis of such data reveals the correlation between the measured cell constituents. For example, when one of the measured attributes is cellular DNA content (the parameter that reports the cell cycle position or DNA ploidy), an expression of the other measured attribute can then be directly related to the cell cycle position (or cell ploidy) without a need for cell synchronization. Since individual cells are measured, intercellular variability can be assessed, cell subpopulations identified, and rare cells detected. The second goal of cytometry applications is to identify and quantify dead cells and discriminate between apoptotic and necrotic modes of death. Dead-cell recognition is generally based on the presence of a particular biochemical or molecular marker that is characteristic for apoptosis, necrosis, or both. Nu-
merous methods have been developed, especially for the identification of apoptotic cells. Apoptosis-associated changes in the gross physical attributes of cells, such as cell size and granularity, can be detected by analysis of laser light scattered by the cell in forward and side directions (Swat et al., 1981; Ormerod et al., 1995). Some methods rely on apoptosis-associated changes in the distribution of plasma membrane phospholipids (Fadok et al., 1992; Koopman et al., 1994). Others detect the loss of active transport function of the plasma membrane. Still other methods probe the mitochondrial transmembrane potential (Cossarizza et al., 1994; Kroemer, 1998; Zamzani et al., 1998). The detection of DNA fragmentation provides another convenient marker of apoptosis; apoptotic cells are then recognized either by their fractional (subdiploid, sub-G1) DNA content due to extraction of low-molecularweight DNA from the cell (Umansky et al., 1981; Nicoletti et al., 1991), or by the presence of DNA strand breaks, which can be detected by labeling their 3′-OH termini with fluorochrome-conjugated nucleotides in a reaction utilizing exogenous terminal deoxynucleotidyl transferase (TdT; Gorczyca et al., 1992, 1993; Li and Darzynkiewicz, 1995; Li et al., 1996). More recently, flow cytometric methods have been developed to detect activation of caspases and tissue transglutaminase (TGase 2; Grabarek et al., 2002). It should be noted, however, that the fluorochrome-labeled inhibitors of caspases (FLICAs), initially described as markers of caspase activation (Smolewski et al., 2001), or serine proteases, although convenient markers of apoptotic cells and most likely detecting activation of these proteases, do not have sufficient specificity to be used as specific probes of their active enzymatic centers (Pozarowski et al., in press).
Nucleic Acid Analysis
7.19.15 Current Protocols in Cytometry
Supplement 25
A variety of kits are commercially available to identify apoptotic cells using the methods presented in this unit. Since the reagents are already prepackaged and the procedures are described in cookbook format, the kits offer the advantage of simplicity. Their cost, however, is many-fold higher than that of the individual reagents. Furthermore, the kits do not allow one the flexibility that is often required to optimize procedures for a particular cell system. In many situations, therefore, the preparation of samples for analysis by flow cytometry as described herein may be preferred. Light-scattering changes during apoptosis Intersection of cells with the laser light beam in a flow cytometer results in light scattering. Analysis of light scattered in different directions reveals information about cell size and structure. The intensity of the forward lightscatter signal correlates with cell size. Side scatter, on the other hand, yields information on light-refractive and light-reflective properties of the cell and reveals optical inhomogeneity of the cell structure resulting from condensation of cytoplasm or nucleus, granularity, and so on. As a consequence of cell shrinkage, a decrease in forward light scatter is observed at a relatively early stage of apoptosis (Swat et al., 1981; Ormerod et al., 1995). Initially, there is
little change in side scatter during apoptosis. In fact, in some cell systems, an increase in intensity of side-scatter signal may be seen, reflecting perhaps chromatin and cytoplasm condensation and nuclear fragmentation, the events that may lead to an increase in the light-refractive and light-reflective properties of the cell. When apoptosis is more advanced and the cells shrink in size, the intensity of side scatter also decreases (Fig. 7.19.1). Late apoptotic cells, therefore, are characterized by markedly diminished intensity of both forward- and sidescatter signals. In contrast to apoptosis, the early stages of cell necrosis are marked by cell swelling, which is detected by a transient increase in forward light scatter. Rupture of the plasma membrane and leakage of the cytosol during subsequent steps of necrosis correlate with a marked decrease in intensity of both forward- and side-scatter signals. Analysis of light scatter is often combined with other assays, most frequently surface immunofluorescence (to identify the phenotype of the dying cell), or another marker of apoptosis. It should be stressed, however, that the change in light scatter alone is not a specific marker of apoptosis or necrosis. Mechanically broken cells, isolated nuclei, cell debris, and individual apoptotic bodies may also display diminished light-scatter properties. Therefore, the analysis of light scatter must be combined
control 1000
Side scatter
1000
TNF
B
A
C 0
Flow Cytometry of Apoptosis
0
400 Forward scatter
400
Figure 7.19.1 Changes in light scattering properties of cells undergoing apoptosis. HL-60 cells were untreated (left panel) or treated 3 hr with TNF-α and cycloheximide (CHX) to induce apoptosis (right panel). Cell population A in the treated culture (right panel) represents cells that have light scattering properties similar to those of untreated cells. Early apoptotic cells (B) have diminished forward scatter and are very heterogenous with respect to side scatter. Late apoptotic cells (C) have both forward and side scatter diminished.
7.19.16 Supplement 25
Current Protocols in Cytometry
with measurements that can provide a more definite identification of apoptotic or necrotic cells. Mitochondrial potential A point to consider in measuring ∆ψm is that the mitochondrial potential probes lack absolute specificity and also accumulate in the cytosol. Probe distribution in mitochondria versus cytosol follows the Nernst equation, according to which the ratio of mitochondrial to cytosolic free cation concentration should be 100:1 at 120 mV mitochondrial transmembrane potential (Waggoner, 1979). However, the specificity of particular mitochondrial probes towards mitochondria is higher at low probe concentrations. It is advisable, therefore, to use these probes at the lowest possible concentration. The limit for the minimal dye concentration that still provides an adequate signal-to-noise ratio during the measurement is dictated by sensitivity of the instrument (laser power, optics, photomultiplier sensitivity) and by the mitochondrial mass per cell; the latter varies depending on the cell type or upon mitogenic stimulation (Darzynkiewicz et al., 1981). A series of MitoTracker dyes (chloromethyltetramethylrosamine analogs) of different colors was introduced by Molecular Probes as new mitochondrial ∆ψm-sensitive probes (UNIT 4.4). Some of these dyes remain attached to mitochondria following cell fixation using crosslinking agents (Poot et al., 1997; Haugland, 2002). It should be noted, however, that because these dyes bind to thiol moieties within mitochondria, their retention after fixation may not be a reliable marker of the transmembrane potential (Ferlini et al., 1998; Gilmore and Wilson, 1999). Furthermore, they are potent inhibitors of respiratory chain I and may themselves induce dissipation of ∆ψm (Scorrano et al., 1999). Because it is likely that other ∆ψm probes may predispose the cells to the permeability transition, one has to be cautious in interpreting the data on their use in analysis of apoptosis. It has been reported that other mitochondrial probes, 10-nonyl acridine orange, MitoFluor Green, and MitoTracker Green, are markers of mitochondrial mass and are not sensitive to ∆ψm (Ratinaud et al., 1988; Poot et al., 1997). It was proposed, therefore, to measure both ∆ψm and mitochondrial mass by using a combination of ∆ψm-sensitive and ∆ψm-insensitive probes (Petit et al., 1995). Recent observations, however, indicate that 10-nonyl acridine orange, MitoFluor Green, and MitoTracker
Green are quite sensitive to changes in ∆ψm and therefore, either alone or in combination with ∆ψm-sensitive probes, cannot be used as markers of mitochondrial mass (Keiji et al., 2000). Caspases Caspases (cysteine-aspartic acid specific proteases) are activated in response to different inducers of apoptosis (Kaufmann et al., 1993; Lazebnik et al., 1994; Alnemri et al., 1996). The process of their activation is considered to be the key event of apoptosis, a marker of cell commitment to disassemble the machinery that supports cell life (for reviews, see Budihardjo et al., 1999; Earnshaw et al., 1999; Shi, 2002). Caspases recognize a four-amino-acid sequence on their substrate proteins; within this sequence, the carboxyl end of aspartic acid is the target for cleavage. Detection of caspase activation is of primary interest in oncology as well as in other disciplines of medicine and biology, and several methods have been developed to accomplish this. One approach that is potentially useful for cytometry utilizes fluorogenic caspase substrates. The peptide substrates are colorless or nonfluorescent, but upon caspase-induced cleavage, they generate colored or fluorescing products (Gorman et al., 1999; Hug et al., 1999; Liu et al., 1999; Komoriya et al., 2000; Telford et al., 2002). Many kits designed to measure activity of caspases using fluorometric or colorimetric assays are commercially available (e.g., from Biomol Research Laboratories or Calbiochem). Some kits can be used to detect activation of multiple caspases, while other are based on the substrates that are specific for caspase-1, caspase-3, or caspase-8. The second approach in studies of caspase activation applicable to cytometry is based on immunocytochemical detection of the epitope of these enzymes that is characteristic of their active form. This epitope appears as a result of conformational changes that occur during activation of caspases, such as those associated with the transcatalytic cleavage of the zymogen pro-caspases (for reviews, see Budihardjo et al., 1999; Earnshaw et al., 1999). Antibodies developed to react only with the activated caspases have recently become commercially available (e.g., from Promega). These antibodies can be used in standard immunocytochemical assays. Basic Protocol 3 describes the methodology of detection of caspase-3 activation based on this principle, combined with the convenient immunocytochemical Zenon technology (Haugland, 2002). Another approach that was proposed to
Nucleic Acid Analysis
7.19.17 Current Protocols in Cytometry
Supplement 25
probe caspase activation utilizes fluorochromelabeled inhibitors of caspases (FLICAs; e.g., FAM-VAD-FMK), which are reagents designed as affinity labels to the active enzyme center of these enzymes (Smolewski et al., 2001). It was recently found, however, that these reagents lack the desired specificity with respect to the active enzyme center of caspases (Pozarowski et al., in press). However, they are convenient and specific markers of apoptotic cells and can be used as such, particularly when there is a need to distinguish between apoptotic and necrotic modes of cell death. Basic Protocol 3 describes application of FAM-VAD-FMK to identify three sequential stages of apoptosis. Finally, caspase activation can be detected indirectly, by identifying specific protein fragments which are the products of cleavage by particular caspases. For example, immunocytochemical detection of the poly(ADP-ribose) polymerase 85-kDa fragment (PARP p85 fragment), which results from cleavage of the PARP substrate by caspase-3, is described in Basic Protocol 4. Concurrent analysis of cellular DNA content allows one to correlate activation of caspases with the cell cycle position or DNA ploidy. Using immunocytochemical methods, however, one has to remember that, often, antibodies that are applicable to immunoblotting or immunoprecipitation may not always be useful in immunocytochemical assays, and vice versa.
Critical Parameters and Troubleshooting
Flow Cytometry of Apoptosis
Positive and negative controls There are instances when cells die by a process of atypical apoptosis that lacks one or more characteristic apoptotic features. Obviously, apoptosis cannot be detected if the feature serves as a marker. It is also possible that the assay (kit) used to identify apoptotic cells malfunctions for technical reasons. For example, the enzyme TdT used in the TUNEL assay may be inactive due to improper storage. A mistake might be made during the staining procedure. It is essential, therefore, to distinguish between the genuine lack of apoptosis and the inability to detect it due to technical causes. The distinction can be made using a positive control consisting of cells known to be apoptotic (confirmed by a standard method and inspection of cell morphology). Such control cells have to be processed in parallel with the investigated sample through all the prescribed protocol steps. Some vendors provide positive
and negative control cells with their kits (e.g., APO-BRDU from Phoenix Flow Systems). The positive control cells can be prepared in large quantity and stored in aliquots to be used during each experiment. Such a convenient control may be, to give an example, exponentially growing HL-60 or U-937 leukemic cells treated 3 to 6 hr in culture with 0.2 µM camptothecin (CPT) to induce apoptosis. The cells so treated consist of subpopulations of apoptotic (S-phase) and non-apoptotic (G1-phase) cells, present in the same sample. However, to induce apoptosis of S-phase cells with CPT, it is critical to have the cultures in the exponential phase of growth, at relatively low cell density (<800,000 cells/ml); subconfluent cultures are quite resistant to CPT. A large batch of cells treated in such a manner can be appropriately fixed in 70% ethanol and then stored in aliquots at −20°C to be used as a positive and negative control for each assay that utilizes fixed cells. For the assays that require live cells, controls should be freshly made and must not be fixed. Cells from healthy, untreated cultures may also serve as negative controls. False-positive apoptosis The exposure of phosphatidylserine on cell surfaces that occurs during apoptosis (Fadok et al., 1992; Koopman et al., 1994) makes apoptotic cells and apoptotic bodies attractive to neighboring cells, which phagocytize them. The ability to engulf apoptotic bodies is not solely the property of professional phagocytes, but is shared by cells from fibroblast or epithelial lineages. It is frequently observed, especially in solid tumors, that the cytoplasm of both nontumor and tumor cells located in the neighborhood of apoptotic cells contains inclusions typical of apoptotic bodies. The remains of apoptotic cells engulfed by neighboring cells contain altered plasma membrane, fragmented DNA, and other constituents with attributes characteristic of apoptosis. Thus, if the distinction is based on any of these attributes, the live, nonapoptotic cells that phagocytized apoptotic bodies cannot be distinguished from genuine apoptotic cells by flow cytometry. For example, nonapoptotic cells that engulf apoptotic bodies become reactive with annexin V (Marguet et al., 1999). Most likely this is due to the fact that during engulfment, the plasma membrane of apoptotic bodies fuses with the plasma membrane of the phagocytizing cell. It has also been shown that nonapoptotic cells (primarily monocytes and macrophages) in bone marrow of patients undergoing chemotherapy have
7.19.18 Supplement 25
Current Protocols in Cytometry
large quantities of apoptotic bodies in their cytoplasm and are strongly positive in the TUNEL reaction (Bedner et al., 1999). Even after relative mild treatment such as trypsinization, mechanical agitation, detachment with rubber policeman, or electroporation, the plasma membrane of live nonapoptotic cells may have phosphatidylserine, reactive with annexin V, exposed on the surface. Such cells may also be erroneously identified as apoptotic cells. Distinction between apoptosis, necrosis, and the “necrotic stage” of apoptosis There are several differences between typical apoptotic and necrotic cells (Kerr et al. 1972; Arends et al., 1990; Majno and Joris, 1995) that provided a basis for development of numerous markers and methods that can discriminate between these two modes of cell death (Darzynkiewicz et al., 1992, 1997). The major difference stems from the early loss of integrity of the plasma membrane during necrosis. This event results in a loss of the cell’s ability to exclude charged fluorochromes such as trypan blue or PI. In contrast, the plasma membrane and membrane transport functions remain, to a large extent, preserved during the early stages of apoptosis. A cell’s permeability to PI or its ability to retain some fluorescent probes, such as products of enzyme activity (e.g., fluorescein diacetate hydrolyzed by esterases), is the most common marker distinguishing apoptosis from necrosis (Darzynkiewicz et al., 1994). A combination of fluorochrome-conjugated annexin V with PI distinguishes live cells (unstainable with both dyes) from apoptotic cells (stainable with annexin V but unstainable with PI) from necrotic cells (stainable with both dyes; Koopman et al., 1994). The same holds true for a combination of PI with FLICA (Smolewski et al., 2001). However, while this approach works well in many instances, it has limitations and possible pitfalls. (1) Late-stage apoptotic cells resemble necrotic cells to such an extent that the term “apoptotic necrosis” was proposed to define them (Majno and Joris, 1995). This is a consequence of the fact that the integrity of the plasma membrane of late apoptotic cells is compromised, which makes the membrane leaky and permeable to charged cationic dyes such as PI. Since the ability of such cells to exclude these dyes is lost, the discrimination between late apoptosis and necrosis cannot be accomplished by methods based on the use of
dye exclusion (PI, trypan blue) or annexin V binding. (2) The permeability and asymmetry of plasma membrane phospholipids (accessibility of phosphatidylserine) may change, as a result of prolonged treatment with proteolytic enzymes (trypsinization), mechanical damage (e.g., cell removal from flasks by rubber policeman, cell isolation from solid tumors, or even repeated centrifugations), electroporation, or treatment with some drugs. (3) Many flow cytometric methods designed to quantify the frequency of apoptotic or necrotic cells are based on the differences between live, apoptotic, and necrotic cells in the permeability of plasma membrane to different fluorochromes such as PI, 7-AAD, or Hoechst dyes. It should be stressed, however, that plasma membrane permeability probed by dye accumulation in the cell may vary depending on the cell type and other factors unrelated to apoptosis or necrosis (e.g., very active efflux mechanism that rapidly pumps dye out of the cell). The assumptions, therefore, that live cells maximally exclude a particular dye, while early apoptotic cells are somewhat leaky and late apoptotic or necrotic cells are fully permeable to the dye, and that these differences are large enough to identify these cells, are not universally applicable. It is particularly difficult to discriminate between apoptotic and necrotic cells in suspensions from solid tumors. Necrotic areas form in tumors as a result of massive local cell death due to poor accessibility to oxygen and growth factors when tumors grow in size and their local vascularization becomes inadequate. Needleaspirated samples or cell suspensions from the resected tumors may contain many cells from the necrotic areas. Such cells are indistinguishable from late apoptotic cells by many markers. Because the AI in solid tumors, representing spontaneous or treatment-induced apoptosis, should not include cells from the necrotic areas, one has to eliminate such cells from analysis. Because incubation of cells with trypsin and DNase selectively digests all cells whose plasma membrane integrity is compromised, i.e., primarily necrotic cells (Darzynkiewicz et al., 1994), such a procedure may be used to remove necrotic cells from suspensions. It should be noted, however, that late apoptotic cells have partially leaky plasma membrane and are also expected to be sensitive to this treatment. In conclusion, examination of cells by microscopy may be the only way to distinguish
Nucleic Acid Analysis
7.19.19 Current Protocols in Cytometry
Supplement 25
between apoptosis and necrosis, based on their characteristic differences in morphology.
Flow Cytometry of Apoptosis
Preferential loss of apoptotic cells during sample preparation Cell detachment in culture. Early during apoptosis, cells detach from the surface of culture flasks and float in the medium. The standard procedure of discarding the medium, trypsinizing the attached cells or treating them with EDTA, and collecting the detached cells results in selective loss of those apoptotic cells that are discarded with the medium. Such loss may vary from flask to flask depending on how the culture is handled, e.g., the degree of mixing or shaking, efficiency in discarding the old medium, and so on. Surprisingly, some authors still occasionally report discarding the medium and trypsinizing the cells. Needless to say, such an approach cannot be used for quantitative analysis of apoptosis. To estimate the frequency of apoptotic cells in adherent cultures, it is essential to collect floating cells, pool them with the trypsinized ones, and measure them as a single sample. It should be stressed that trypsinization, especially if prolonged, results in digestion of cells with a compromised plasma membrane. Thus, collection of cells from cultures by trypsinization is expected to cause selective loss of late apoptotic and necrotic cells. Density-gradient centrifugation. Similarly, density-gradient separation of cells (e.g., using ficoll-hypaque or percoll solutions) may result in selective loss of dying and dead cells, because early during apoptosis the cells become dehydrated, have condensed nuclei and cytoplasm, and therefore have a higher density than nonapoptotic cells. Knowledge of any selective loss of dead cells in cell populations purified by such an approach is essential when one is studying apoptosis. Centrifugations, mechanical agitations. Repeated centrifugations lead to cell loss by at least two mechanisms. One involves electrostatic cell attachment to the tubes and may be selective for a particular cell type. For example, preferential loss of monocytes and granulocytes is observed during repeated centrifugation of white blood cells, while lymphocytes remain in suspension (Bedner et al., 1997). Cell loss is of particular concern when hypocellular samples (<5 × 104) are processed. In such a situation, carrier cells in excess (e.g., chick erythrocytes) may be added to preclude disappearance of the cells of interest through centrifugation. The second mechanism of cell loss
involves preferential disintegration of fragile cells. Because apoptotic cells are very fragile, especially at late stages of apoptosis, they may selectively be lost from samples that require centrifugation or are repeatedly vortexed or pipetted. Addition of serum or bovine serum albumin to cell suspensions, shortened centrifugation time, and decreased gravity force all may have a protective effect against cell breakage by mechanical factors. Apoptotic cells may also preferentially disintegrate in biomass cultures that require constant cell mixing. It should be noted that sensitivity of apoptotic cells to mechanical factors depends on activation of TGase 2. Cells with activated TGase 2 have their cytoplasmic protein crosslinked and are resistant to mechanical stress. They also resist treatment with detergents. In contrast, apoptotic cells that do not activate TGase 2 are overly sensitive and easily undergo disintegration. It was observed that cells may often undergo apoptosis without evidence of TGase 2 activation (Grabarek et al., 2002). Abundance of extractable (fragmented) DNA is not a quantitative measure of apoptosis A common misconception in analysis of apoptosis is that the amount of fragmented (low-molecular-weight, “extractable”) DNA detected in cultures, tissue or cell extracts, or other samples reflects incidence of apoptosis. Many methods have been developed to estimate the abundance of fragmented DNA and numerous reagent kits are being sold for that purpose. They include direct quantitative colorimetric analysis of “soluble” DNA, densitometry of “DNA ladders” on gels, and immunochemical assessment of nucleosomes. These approaches and the related kits are advertised as quantitative, in that they provide information regarding the incidence of apoptosis in cell populations. Such claims are grossly incorrect for the reason that the amount of fragmented (low-molecularweight) DNA that can be extracted from a single apoptotic cell varies over a wide range depending on the stage of apoptosis. Although at an early stage of apoptosis only a small fraction of DNA is fragmented and extractable, nearly all DNA can be extracted from the cell that is more advanced in apoptosis. Thus, the total content of low-molecular-weight DNA extracted from the cell population, or the ratio of low- to high-molecular-weight fraction, does not provide information about the frequency of apoptotic cells (AI), even in relative terms, e.g.,
7.19.20 Supplement 25
Current Protocols in Cytometry
for comparison of cell populations. For this simple reason, biochemical methods based on analysis of fragmented DNA can be used to quantitatively estimate the frequency of apoptosis only when comparing cell populations that have identical distribution of cells across the stages of apoptosis. DNA “laddering” observed during electrophoresis provides evidence of internucleosomal DNA cleavage, which is considered one of the hallmarks of apoptosis (Arends et al., 1990). Analysis of DNA fragmentation by gel electrophoresis to detect such laddering is thus a valuable method to demonstrate the apoptotic mode of cell death; however, it should not be used as a means to quantify the frequency of apoptosis. It should be noted that in some cell types, particularly of fibroblast and epithelial lineages, apoptosis may occur without internucleosomal DNA cleavage. The products of DNA fragmentation in these cells are large (50to 300-kb) DNA sections that cannot be extracted from the cell (Oberhammer et al., 1993).
104
Obviously, in these cases, apoptosis cannot be revealed by the presence of DNA laddering on gels or by analysis of low-molecular-weight products. These large DNA fragments, however, can be identified by pulse-field gel electrophoresis. Changes in cell morphology, the “gold standard” for identification of apoptosis Apoptosis was originally defined as a specific mode of cell death based on very characteristic changes in cell morphology (Kerr et al., 1972; Fig. 7.19.2). These changes are still considered the “gold standard” for identification of apoptotic cells. Although particular markers may be used in conjunction with flow cytometry for detection and quantitative assessment of apoptosis in cell populations, the mode of cell death should always be confirmed by inspection of cells by light or electron microscopy. If there is any ambiguity regarding the mechanism of cell death, the morphological
104 TNF
Rhodamine 123
control
A C B
104
104
PI fluorescence
Figure 7.19.2 Detection of the collapse of mitochondrial electrochemical potential (ψm) by rhodamine 123 (R123). HL-60 cells were untreated (control; left panel) or treated 3 hr with TNF-α and CHX (right panel) to induce apoptosis. The cells were then incubated with R123 and PI according to Basic Protocol 1. The early apoptotic cells have diminished green fluorescence of R123 but exclude PI (cell population B). The late apoptotic (also necrotic) cells are stained strongly by PI (population C).
Nucleic Acid Analysis
7.19.21 Current Protocols in Cytometry
Supplement 27
Table 7.19.1
Morphological Criteria for Identification of Apoptosis or Necrosis
Apoptosis
Necrosis
Reduced cell size, convoluted shape Plasma membrane undulations (“blebbing,” “budding”) Chromatin condensation (DNA hyperchromicity) Loss of the structural features of the nucleus (smooth, planate appearance) Nuclear fragmentation (karyorrhexis)
Cell and nuclear swelling Patchy chromatin condensation
Presence of apoptotic bodies Dilatation of the endoplasmic reticulum Relatively unchanged cell organelles Shedding of apoptotic bodies Phagocytosis of the cell remnants Cell detachment from tissue culture flasks
Flow Cytometry of Apoptosis
changes should be the deciding factor in resolving the uncertainty. The characteristic morphological features of apoptosis and necrosis are listed in Table 7.19.1. The most specific (and generic to apoptosis) of these changes is chromatin condensation; the chromatin of apoptotic cells is very “smooth” (structureless) in appearance, and the structural framework that otherwise characterizes the cell nucleus is entirely lost. Because of the condensation, chromatin shows strong hyperchromicity with any of the DNA-specific dyes (Hotz et al., 1992). Apart from chromatin condensation, however, other changes are less generic to apoptosis, and may not always be apparent. For example, nuclear fragmentation, although commonly observed during apoptosis of hematopoietic-lineage cells, may not occur during apoptosis of some epithelial- or fibroblast-lineage cells. Likewise, cell shrinkage, at least early during apoptosis, is not a universal marker of the apoptotic mode of cell death. It should be stressed that optimal preparations for light microscopy require cytospinning of live cells followed by fixation and staining on slides. The cells become flat, facilitating assessment of their morphology. On the other hand, when cells are initially fixed and stained in suspension, transferred to slides, and analyzed under the microscope, their morphology is obscured by unfavorable geometry; the cells are spherical and thick, and require confocal
Swelling of mitochondria Vacuolization in cytoplasm Plasma membrane rupture (“ghost-like” appearance of lysed cells) Dissolution of nuclear chromatin (karyolysis) Attraction of inflammatory cells
microscopy to reveal details such as early signs of apoptotic chromatin condensation. Differential staining of cellular DNA and protein of cells on slides with DAPI and sulforhodamine 101, respectively, is rapid and simple and provides very good morphological resolution of apoptosis and necrosis (Darzynkiewicz et al., 1997). A combined cell illumination with UV light (to excite the DAPI or other DNA fluorochrome) and light-transmission microscopy utilizing interference contrast (Nomarski illumination) is the authors’ favorite method of cell visualization to identify apoptotic cells. Other DNA fluorochromes, such as PI, 7-aminoactinomycin D, or acridine orange, can be used as well. In conclusion, regardless of which cytometric assay has been used to identify apoptosis, the mode of cell death should be confirmed by inspection of cells by light or electron microscopy. Morphological changes during apoptosis have a very characteristic pattern (Kerr et al., 1972; Majno and Joris, 1995) and should be the deciding factor in situations where any ambiguity arises regarding the mode of cell death. It should be noted, however, that the laser scanning cytometer (LSC), offering many attributes of flow cytometry and simultaneously providing the means to examine morphology of the measured cells (Darzynkiewicz et al., 1999; Kamentsky, 2001), appears to be the instrument of choice in analysis of apoptosis (Bedner et al., 1999).
7.19.22 Supplement 27
Current Protocols in Cytometry
Membrane potential It should be stressed that ∆ψm, like other functional markers, is sensitive to minor changes in cell environment. The samples to be compared, therefore, should be incubated and measured under identical conditions, taking into an account temperature, pH, time elapsed between the onset of incubation and actual fluorescence measurement, and other potential variables. Annexin V Interpretation of the results may be complicated by the presence of non-apoptotic cells with damaged membranes. Such cells may have phosphatidylserine exposed on the plasma membrane and therefore, like apoptotic cells, bind annexin V. Mechanical disaggregation of tissues to isolate individual cells, extensive use of proteolytic enzymes to disrupt cell aggregates, to remove adherent cells from cultures, or to isolate cells from tissue, mechanical removal of the cells from tissue-culture flasks (e.g., with a rubber policeman), or cell electroporation all affect the binding of annexin V. Such treatments, therefore, may introduce experimental bias in the subsequent analysis of apoptosis by this method. Even intact and live cells take up PI upon prolonged incubation. Therefore, fluorescence measurement should be performed rather shortly after addition of the dye. DNA fragmentation It should be stressed that the degree of extraction of low-molecular-weight DNA and consequently the content of DNA remaining in apoptotic cells for flow cytometric analysis varies markedly depending on the extent of DNA degradation (duration of apoptosis), the number of cell washings, and the pH and molarity of the washing and staining buffers. DNA fragmentation is often so extensive that most DNA is removed during the post-fixation rinse with PBS and in the staining solution, and a DNA extraction step is therefore unnecessary. Conversely, when DNA degradation does not proceed to internucleosomal regions but stops after generating 50- to 300-kb fragments (Oberhammer et al., 1993), little DNA can be extracted, and this method may fail to detect such apoptotic cells. It also should be noted that if G2, M, or even late S-phase cells undergo apoptosis, the loss of DNA from these cells may not be adequate to place them at the sub-G1 peak, as they may end up with DNA content equivalent of that of G1 or early S-phase cells
and therefore be indistinguishable from the latter. It is a common practice to use detergents or hypotonic solutions instead of fixation in the process of DNA staining for flow cytometry (Nicoletti et al., 1991). Such treatments cause lysis of plasma membrane and release of the nucleus. Although this approach is simple and yields excellent resolution for DNA-content analysis, when used to quantify apoptotic cells it introduces bias owing to the fact that nuclei of apoptotic cells are often fragmented. Lysis of cells with fragmented nuclei releases nuclear fragments rather than individual nuclei, and consequently several fragments can be released from a single cell. Likewise, lysis of mitotic cells that happen to be in the specimen releases individual chromosomes or chromosome aggregates. In the case of micronucleation (e.g., after cell irradiation), the micronuclei are released upon cell lysis. Each nuclear fragment, chromosome, or micronucleus is then recorded by the flow cytometer as an individual object with a sub-G1 DNA content. Such objects are then erroneously classified as individual apoptotic cells. This bias is increased if DNA content is displayed on a logarithmic scale. Such a scale allows one to record objects with DNA content as little as 1% or even 0.1% of that of G1 cells, which certainly cannot be individual apoptotic cells. Activation of TGase 2 Although apoptotic cells with activated TGase 2 can be detected using either Basic Protocol 8 or Alternate Protocol 2, the differences between these assays should be underscored. The most distinct is the difference in the length of the “time window” that may be measured by the respective assay. Namely, the detergent-based assay (see Basic Protocol 8) detects cumulative protein crosslinking, reflecting the integrated crosslinking, from its onset to the time of cell harvesting. In contrast, the assay based on incorporation of F-CDV (see Alternate Protocol 2) detects crosslinking that occurs only during the time interval when this reagent is present in the culture. Thus, if F-CDV is included at time zero, i.e., when the inducer of apoptosis is added, its incorporation is a reflection of the cumulative protein crosslinking and thus is comparable with the detergentbased assay. However, if it is added during the final hour or two, it will reflect the crosslinking that took place only during this 1- or 2-hr time window. This difference between the assays should be kept in mind when comparing fre-
Nucleic Acid Analysis
7.19.23 Current Protocols in Cytometry
Supplement 27
quency of TGase 2–positive cells, which may vary between the assays, depending on the length of the respective time window. Situations (e.g., following treatment with particular drugs) may occur in which cell proteins may become less soluble and more detergent resistant, not necessarily because of TGase 2 activation, but because of alteration by the drug. Some treatments unrelated to TGase 2 activity may also result in attachment of F-CDV to cellular proteins. Some cell types may be more resistant to detergents. In all these cases, the assay may detect the “false-positive” TGase 2–positive cells. As with other markers of apoptosis, one has to be careful and additionally identify these cells by microscopy based on the characteristic changes in their morphology. Apoptosis can be induced and may progress in some cells with no apparent TGase 2 activation (Grabarek et al., 2002). In general, the activation of TGase 2 is seen to occur in cells that show a high degree of chromatin and cytoplasm condensation leading to pronounced nuclear and cell shrinkage, and that either lack or have very limited nuclear fragmentation. In contrast, the apoptotic cells that appear larger and whose nuclei are excessively fragmented do not show activation of TGase 2. Apoptosis without activation of TGase 2 appears to occur more frequently when induced with higher drug doses, i.e., when cells enter the apoptotic process more rapidly following treatment.
Thus, a note of caution should be added, that since TGase 2 activation may not be detected in some instances of apoptosis, the absence of its activation should not be considered a marker of nonapoptotic cells.
Anticipated Results Light scatter. A decrease in forward light scatter characterizes early apoptotic cells (Fig. 7.19.1, cluster B). Late apoptotic cells and perhaps also larger apoptotic bodies show marked decrease in both forward and side light scatter (cluster C). Mitochondrial potential. A combination of PI and R123 identifies nonapoptotic cells that stain only green, early apoptotic cells whose green fluorescence is diminished, and late apoptotic or necrotic cells that stain with PI and have red fluorescence only (Fig. 7.19.2). Likewise, a combination of DiOC6(3) and PI labels live nonapoptotic cells green, early apoptotic cells dim-green, and late apoptotic and necrotic cells red (not shown). The change in binding of JC-1 is manifested by a loss of the orange fluorescence that represents the aggregate binding of this dye and that characterizes charged mitochondria (Fig. 7.19.3). JC-1 green fluorescence is expected to increase as a result of disaggregation of the complexed JC-1. However, either no change or a decrease in green fluorescence may be seen
control
CPT 103
Orange fluorescence
103
A
B
103
103
Green fluorescence
Flow Cytometry of Apoptosis
Figure 7.19.3 Detection of the collapse of mitochondrial electrochemical potential using the aggregate dye JC-1. HL-60 cells were untreated (control, left panel) or treated 3 hr with camptothecin (CPT, right panel) to induce apoptosis. Cells were then stained with JC-1 and their orange and green fluorescence was measured by cytometry, as described in Basic Protocol 2. Decreased intensity of orange fluorescence (subpopulation B) characterizes the cells with collapsed potential.
7.19.24 Supplement 27
Current Protocols in Cytometry
104
104
control
TPT
5
40
4 30
103
103
2
20
10
1 0 400
0
600
200
400
0
200
400
600
120
101
120 100
S
G2/M
G1
80
200
80 60
40
40 20
20 00
0
100
60
400
200
600
400
600
800
100
100
0
102
200
102
0
101
Activated caspase-3
3
0
0
200
400
600
600
800
PI fluorescence
104
104
Figure 7.19.4 Immunocytochemical detection of caspase-3 activation using antibody reactive with the activated (cleaved) caspase-3. Apoptosis of HL-60 cells was induced by topotecan (TPT), an analog of CPT. Zenon technology (Haugland, 2002) was used to detect caspase-3 as described in Basic Protocol 2. Top and bottom insets in each panel show cellular DNA content frequency histograms of cells with activated and nonactivated caspase-3, respectively. Note that S-phase cells preferentially contain activated caspase-3 after induction of apoptosis by TPT.
control
TPT 103
C
102
B
102
A
101
101
FAM-VAD-FMK
103
Ap
100
101
102
103
N 104
100
100
D 100
101
102
103
104
PI fluorescence
Figure 7.19.5 Binding of fluorochrome-labeled inhibitor of caspases (FLICA; FAM-VAD-FMK) and PI during apoptosis. Apoptosis of HL-60 cells was induced by TPT. The cells were stained according to Basic Protocol 3. Green (FAM-VAD-FMK) and red (PI) cellular fluorescence was measured by flow cytometry. Four cell subpopulations (A to D) can be identified, differing in their capability to bind FAM-VAD-FMK and PI. They represent sequential stages of apoptosis, starting with binding of FAM-VAD-FMK (B), loss of plasma membrane integrity to exclude PI (C), and loss of reactivity with FAM-VAD-FMK (D).
Nucleic Acid Analysis
7.19.25 Current Protocols in Cytometry
Supplement 25
if JC-1 concentration within the cell is too high, which causes quenching of its fluorescence. Caspases. Caspase-3 activation during apoptosis induced by topotecan (TPT), a camptothecin (CPT) analog, is reflected by the cells’ ability to bind antibody that is reactive with the activated (cleaved) form of this enzyme (Fig. 7.19.4). Concurrent staining of cellular DNA with PI makes it possible to correlate caspase-3 activation with the cell cycle position. Note that activation of caspase-3 occurs preferentially in S-phase cells. The bivariate distributions (scatterplots) of green and red fluorescence intensity representing cells supravitally stained with FAMVAD-FMK (FLICA) and PI reveal the presence of four distinct subpopulations (Fig. 7.19.5). Nonapoptotic cells show neither FLICA nor PI fluorescence (FLICA–/PI–; subpopulation A). Early apoptotic cells bind FAM-VAD-FMK and still exclude PI (FLICA+/PI–; subpopulation B). More advanced in apoptosis are the cells that bind FAM-VAD-FMK but lose the ability to exclude PI (FLICA+/PI+; subpopulation C). The cells most advanced in apoptosis are FAM-VAD-FMK negative and are stained with PI (FLICA–/PI+; subpopulation D). Because the late phase of apoptosis during which the plasma membrane becomes permeable to cationic dyes such as PI or trypan blue has been defined as the “necrotic stage” of apoptosis (Majno and Joris, 1995; Darzynkiewicz et al., 1997), the FLICA+/PI+ and FLICA–/PI+ cells
thus represent two consecutive phases of the “necrotic stage.” It should be noted that genuine necrotic cells, i.e., cells that die by the mode of necrosis (“accidental” cell death), not having activated caspases and unable to exclude PI (Darzynkiewicz et al., 1997), have the same properties (FLICA–/PI+) as very late apoptotic cells. As mentioned earlier, because of lack of specificity, the labeling of apoptotic cells with FAM-VAD-FMK, while likely the marker of caspase activation, is not in and of itself evidence of its binding to the active enzymatic center of caspases (Pozarowski et al., in press). PARP cleavage. Differences in intensity of PARP p85 immunofluorescence versus PI fluorescence (cellular DNA content) allow one to identify apoptotic cells and reveal the cell cycle distribution of both apoptotic (PARP p85–positive) and nonapoptotic (PARP p85–negative) cells (Fig. 7.19.6). It is quite evident that predominantly S-phase cells were undergoing apoptosis upon CPT treatment (Li and Darzynkiewicz, 2000). Annexin V. Live nonapoptotic cells stained according to Basic Protocol 5 have minimal green (annexin V-FITC) fluorescence and also minimal red (PI) fluorescence (Fig. 7.19.7; subpopulation A). At early stages of apoptosis, cells stain green but still exclude PI and therefore continue to have no significant red fluorescence (subpopulation B). At late stages of apoptosis, cells show intense green and red fluores-
103
PARP p85
0 min
120 min
150 min
102
101
100 0
Flow Cytometry of Apoptosis
200
400
600
0
200 400 600 DNA content
0
200
400
600
Figure 7.19.6 Identification of apoptotic cells by flow cytometry based on the immunocytochemical detection of the 85-kDa PARP cleavage fragment. To induce apoptosis, HL-60 cells were treated 60 min with TNF-α in the presence of CHX (Li and Darzynkiewicz, 2000). PARPp85 was detected immunocytochemically and DNA was counterstained with PI, as described in Basic Protocol 3.
7.19.26 Supplement 25
Current Protocols in Cytometry
104
104 TNF
control
Annexin V
C
D
B A 104
104
PI fluorescence
Figure 7.19.7 Detection of early and late apoptotic cells after staining with annexin V–FITC and PI. To induce apoptosis, HL-60 cells were treated 2 hr with TNF-α and CHX. Untreated (control; left panel) and TNF-α-treated (right panel) cells were then stained with annexin V-FITC and PI.
A
B
C
Number of cells
G1
G 2M
Ap
S
0
1
2
0
1
2
DNA index (DI)
Figure 7.19.8 Detection of apoptotic cells by flow cytometry based on cellular DNA content analysis. (A) Normal cell plot. (B) To induce apoptosis, HL-60 cells were treated with the DNA topoisomerase II inhibitor fostriecin (Hotz et al., 1994). Cells were fixed in 70% ethanol, suspended in high-molarity phosphate buffer to extract fragmented DNA, and then stained with PI. A subpopulation of apoptotic cells (Ap) with fractional (sub-diploid) DNA content, i.e., with DNA index (DI) <1.0 (sub-G1 cells), is apparent. Note also the increase in the proportion of S-phase cells in the nonapoptotic population. (C) The fragmented DNA extracted from the apoptotic cells by the buffer was subjected to gel elecrophoresis (Gong et al., 1994). Note “laddering" that reflects preferential DNA cleavage at internucleosomal sections, the characteristic feature of apoptosis (Arends et al., 1990). Nucleic Acid Analysis
7.19.27 Current Protocols in Cytometry
Supplement 25
0 min
120 min
150 min
DNA strand breaks
103
102 G2M G1 S
101
100
0
200
400
600
0
200
400
600
0
200
400
600
DNA content
Figure 7.19.9 Detection of apoptotic cells by flow cytometry based on the presence of DNA strand breaks. To induce apoptosis, HL-60 cells were treated 120 or 150 min with CPT (Li and Darzynkiewicz, 2000). DNA strand breaks were labeled with BrdUTP using exogenous terminal deoxynucleotidyl transferase. The cell cycle distribution of both apoptotic and nonapoptotic cell subpopulations can be estimated based on the DNA content of individual cells. Note that in analogy to PARP cleavage (Fig. 7.19.6), preferentially S-phase cells undergo apoptosis following CPT treatment.
A
B
C
100
100
0
S
32% 100
DNA content
G2 M
G1
13% 100
DNA content
Red fluorescence
68%
Red fluorescence 97% 0
Flow Cytometry of Apoptosis
100 87%
Red fluorescence
3%
0
100 DNA content
Figure 7.19.10 Detection of tissue transglutaminase (TGase 2) activation during apoptosis by the acquired resistance of the cytoplasmic proteins to detergent. Bivariate distributions illustrating red fluorescence of sulforhodamine 101 (protein content) versus blue fluorescence of DAPI (DNA content) of HL-60 cells, untreated (A) or exposed to hyperthermia (72 hr at 41.5°C) in the absence (B) and presence (C) of the cytotoxic RNase onconase (1.67 µM; Grabarek et al., 2002). Following cell lysis by Triton X-100 and staining with DAPI and sulforhodamine 101, the isolated nuclei of nonapoptotic cells from control culture (A) show low and uniform intensity of red fluorescence, reflecting low protein content. Subpopulations of apoptotic cells in B and C have their cytoplasmic protein crosslinked and therefore are resistant to detergent. They stain intensely with sulforhodamine 101. Note differences in DNA content (cell cycle) distribution of the cells with activated (top insets; cells gated above the dashed line) versus nonactivated TGase 2 (bottom insets; cells gated below the dashed line) in B and C. Percentage of cells with activated and non-activated TGase 2 in the respective cultures is indicated in each panel.
7.19.28 Supplement 25
Current Protocols in Cytometry
0
FITC-cadaverine
B 100
FITC-cadaverine
A 100
6% 94% G1 S G2M DNA content
100
59% 41%
0 DNA content
100
Figure 7.19.11 Detection of TGase 2 activity in HL-60 cells using FITC-conjugated cadaverine (F-CDV) as the enzyme substrate. Cultures of untreated (A) and hyperthermia (5 hr at 41.5°C)treated (B) HL-60 cells were incubated 5 hr with 100 µM F-CDV. Cells were then fixed and their DNA was counterstained with PI in the presence of RNase. Note the presence in the untreated culture (A) of few cells that incorporated F-CDV (spontaneous apoptosis), and large numbers of F-CDV-labeled cells in the treated culture (B). Note also that some apoptotic cells with fractional DNA content (“sub-G1” subpopulation) in the treated culture do not show incorporation of F-CDV (arrow). Percentage of cells with activated and nonactivated TGase 2 in the respective cultures is indicated. The inset shows the cellular DNA content distribution histogram of all cells (Grabarek et al., 2002).
cence (subpopulation C). It should be noted that isolated nuclei, cells with severely damaged membranes, and very late apoptotic cells stain rapidly and strongly with PI and may not bind annexin V (subpopulation D). DNA fragmentation. Apoptotic cells have decreased PI (or DAPI) fluorescence and diminished forward light scatter relative to cells in the main peak (G1; Fig. 7.19.8). Optimally, the sub-G1 peak representing apoptotic cells should be separated from the G1 peak of the nonapoptotic cell population with little or no overlap between these two. TUNEL. DNA strand breaks in apoptotic cells are strongly labeled with fluoresceinated anti-BrdU Ab that distinguishes them from the nonapoptotic cells (Fig. 7.19.9). Because of the high intensity of their green fluorescence, an exponential scale (logarithmic PMTs) often must be used for data acquisition and display. Simultaneous measurement of DNA content makes it possible to identify the cell cycle position of cells in apoptotic and nonapoptotic populations. It should be noted, however, that late apoptotic cells may have diminished DNA content because of prior shedding of apoptotic bodies (which may contain nuclear fragments), or due to such massive DNA fragmentation that small DNA fragments cannot be retained in the
cell even after fixation with formaldehyde. Such late apoptotic cells may have sub-G1 DNA content as shown in Figure 7.19.8. Tissue transglutaminase (TGase 2). Detergent (Triton X-100) treatment of nonapoptotic cells as well as apoptotic cells without activated TGase 2 results in cell lysis and release of isolated nuclei or nuclear fragments that have minimal protein content. Such isolated nuclei or nuclear fragments have very low red fluorescence after staining with sulforhodamine 101 (Fig. 7.19.10, panel A). Apoptotic cells with activated TGase 2, on the other hand, have cross-linked proteins and resist lysis under these conditions. Intensity of their red fluorescence is several times higher than that of the isolated nuclei (subpopulations represented by the scatter plots above the dashed lines in panels B and C). Note the high heterogeneity among individual TGase 2–positive cells in intensity of their red fluorescence. Because cellular DNA content (PI fluorescence) is measured concurrently with protein content, induction of protein crosslinking can be correlated with the cell cycle position. It is evident that the effects are cell cycle phase specific. Both hyperthermia and onconase lead to preferential protein crosslinking in G2/Mphase cells.
Nucleic Acid Analysis
7.19.29 Current Protocols in Cytometry
Supplement 25
Activation of TGase 2 in HL-60 cells, detected by cell labeling with F-CDV combined with cellular DNA content analysis, is shown in Fig. 7.19.11. As in the case of protein content (Fig. 7.19.10), populations of cells with activated TGase 2 either undergoing spontaneous apoptosis in control culture (Fig. 7.19.11A) or subjected to hyperthermia (Fig. 7.19.11B) are heterogeneous in terms of their TGase 2–related fluorescence. In the hyperthermia-treated culture, a large number of cells have fractional DNA content, forming a characteristic “subG1” population typical of apoptotic cells. This population is heterogeneous in terms of intensity of F-CDV fluorescence, with many FCDV-negative “sub-G1” cells (arrow). Thus, the degree of activation of TGase 2 is uneven, and many cells with apoptotic features (sub-G1 DNA content) have undetectable level of TGase activity.
Time Considerations
Flow Cytometry of Apoptosis
Basic Protocol 1 takes ∼15 min to prepare cells for incubation with mitochondrial probes followed by an additional 15- to 30-min incubation. Basic Protocol 2 (activated caspase detection by Zenon techology) requires ∼2 hr to complete (cell fixation not included). In Basic Protocol 3, FLICA (FAM-VADFMK) is added directly to cultures at least 30 min before cell centrifugation. Optimal labeling is achieved after a 1- to 2-hr incubation with FLICA. Cell rinses and staining with PI require an additional ∼15 min before cells are measured by flow cytometry. Basic Protocol 4 requires ∼4 hr to process cells from fixation through primary and secondary Ab incubations followed by staining with PI before they are analyzed by flow cytometry. Basic Protocol 5 is a rapid procedure that can be completed in 15 min. Basic Protocol 6 requires ∼40 min to carry out cell rinsing and staining with PI following removal from fixative. In Alternate Protocol 1, cell rinsing and staining can be completed in 25 to 30 min. Basic Protocol 7 takes ∼3 hr to carry out all steps of cell rinsing and incubations with the respective reagents after removal of cells from fixative. Basic Protocol 8 is a rapid procedure. Cells can be analyzed by flow cytometry ∼5 min after their removal from cultures. In Alternate Protocol 2, the reagent F-CDV is added directly to cultures at different time intervals. It takes ∼40 min following cell fixa-
tion to carry out cell rinsing and labeling with PI.
Literature Cited Alnemri, E.S., Livingston, D.I., Nicholson, D.W., Salvesen, G., Thornberry, N.A., Wong, W.W., and Yuan, J. 1996. Human ICE/CED-4 protease nomenclature. Cell 87:171-173. Arends, M.J., Morris, R.G., and Wyllie, A.H. 1990. Apoptosis: The role of endonuclease. Am. J. Pathol. 136:593-608. Bedner, E., Burfeind, P., Gorczyca, W., Melamed, M.R., and Darzynkiewicz, Z. 1997. Laser scanning cytometry distinguishes lymphocytes, monocytes and granulocytes by differences in their chromatin structure. Cytometry 29:191196. Bedner, E., Li, X., Gorczyca, W., Melamed, M.R., and Darzynkiewicz, Z. 1999. Analysis of apoptosis by laser scanning cytometry. Cytometry 35:181-195. Blagosklonny, M.V. 2000. Cell death beyond apoptosis. Leukemia 14:1502-1508. Budihardjo, I., Oliver, H., Lutter, M., and Luo, X. 1999. Biochemical pathways of caspase activation during apoptosis. Annu. Rev. Cell Dev. Biol. 15:269-290. Catchpoole, D.R. and Stewart, B.W. 1993. Etoposide-induced cytotoxicity in two human T-cell leukemic lines. Delayed loss of membrane permeability rather than DNA fragmentation as an indicator of programmed cell death. Cancer Res. 53:4287-4296. Collins, R.J., Harmon, B.V., Gobe, G.C., and Kerr, J.F.R. 1992. Internucleosomal DNA cleavage should not be the sole criterion for identifying apoptosis. Int. J. Radiat. Biol. 61:451-453. Cossarizza, A. and Salvioli, S. 2001. Analysis of mitochondria during cell death. Meth. Cell Biol. 63:467-486. Cossarizza, A., Kalashnikova, G., Grassilli, E., Chiappelli, F., Salvioli, S., Capri, M., Barbieri, D., Troiano, L., Monti, D., and Franceschi, C. 1994. Mitochondrial modifications during rat thymocyte apoptosis: A study at a single cell level. Exp. Cell Res. 214:323-330. Darzynkiewicz, Z., Staiano-Coico, L., and Melamed, M.R. 1981. Increased mitochondrial uptake of rhodamine 123 during lymphocyte stimulation. Proc. Natl. Acad. Sci. U.S.A. 78:2383-2387. Darzynkiewicz, Z., Traganos, F., Staiano-Coico, L., Kapuscinski, J., and Melamed, M.R. 1982. Interactions of rhodamine 123 with living cells studied by flow cytometry. Cancer Res. 42:799-806. Darzynkiewicz, Z., Traganos, F., and Kimmel, M. 1987. Assay of cell cycle kinetics by multivariate flow cytometry using the principle of stathmokinesis. In Techniques in Cell Cycle Analysis (J.W. Gray and Z. Darzynkiewicz, eds.) pp. 291336. Humana Press, Totowa, N.J.
7.19.30 Supplement 25
Current Protocols in Cytometry
Darzynkiewicz, Z., Bruno, S., Del Bino, G., Gorczyca, W., Hotz, M.A., Lassota, P., and Traganos, F. 1992. Features of apoptotic cells measured by flow cytometry. Cytometry 13:795-808. Darzynkiewicz, Z., Li, X., and Gong, J. 1994. Assays of cell viability. Discrimination of cells dying by apoptosis. Meth. Cell Biol. 41:16-39. Darzynkiewicz, Z., Juan, G., Li, X., Murakami, T., and Traganos, F. 1997. Cytometry in cell necrobiology: Analysis of apoptosis and accidental cell death (necrosis). Cytometry 27:1-20. Darzynkiewicz, Z., Bedner, E., Li, X., Gorczyca, W., and Melamed, M.R. 1999. Laser scanning cytometry. A new instrumentation with many applications. Exp. Cell Res. 249:1-12. Deng, Y., Lin, Y., and Wu, X. 2002. TRAIL-induced apoptosis requires Bax-dependent mitochondrial release of Smac/DIABLO. Genes Dev. 16:33-45. de Murcia, G. and Menissier-de Murcia, J.M. 1994. Poly(ADP-ribose) polymerase: A molecular nick sensor. Trends Biochem. Sci. 19:172-176. Earnshaw, W.C., Martins, L.M., and Kaufmann, S.H. 1999. Mammalian caspases: Structure, activation, substrates, and functions during apoptosis. Annu. Rev. Biochem. 68:383-424. Fadok, V.A., Voelker, D.R., Campbell, P.A., Cohen, J.J., Bratton, D.L., and Henson, P.M. 1992. Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J. Immunol. 148:22-29. Ferlini, C., Scambia, G., and Fattorossi, A. 1998. Is chloromethyl-X-rosamine useful in measuring mitochondrial transmembrane potential? Cytometry 31:74-79. Fesus, L., Thomazy, V., and Falus, A. 1987. Induction and activation of tissue transglutaminase during programmed cell death. FEBS Lett. 224:104-108. Finucane, D.M., Waterhouse, N.J., Amaranto-Mendes, G.P., Cotter, T.G., and Green, D.R. 1999. Collapse of the inner mitochondrial transmembrane potential is not required for apoptosis of HL-60 cells. Exp. Cell Res. 251:166-174. Gilmore, K. and Wilson, M. 1999. The use of chloromethyl-X-rosamine (Mitotracker Red) to measure loss of mitochondrial membrane potential in apoptotic cells is incompatible with cell fixation. Cytometry 36:355-358. Gong, J., Traganos, F., and Darzynkiewicz, Z. 1994. A selective procedure for DNA extraction from apoptotic cells applicable for gel electrophoresis and flow cytometry. Anal. Biochem.218:314319. Gorczyca, W., Bruno, S., Darzynkiewicz, R., Gong, J., and Darzynkiewicz, Z. 1992. DNA strand breaks occurring during apoptosis: Their early in situ detection by the terminal deoxynucleotidyl transferase and nick translation assays and prevention by serine protease inhibitors. Int. J. Oncol. 1:639-648.
Gorczyca, W., Bigman, K., Mittelman, A., Ahmed, T., Gong, J., Melamed, M.R., and Darzynkiewicz, Z. 1993. Induction of DNA strand breaks associated with apoptosis during treatment of leukemias. Leukemia 7:659-670. Gorman, A.M., Hirt, U.A., Zhivotovsky, B., Orrenius, S., and Ceccatelli, S. 1999. Application of a fluorometric assay to detect caspase activity in thymus tissue undergoing apoptosis in vivo. J. Immunol. Methods 226:43-48. Grabarek, J., Ardelt, B., Kunicki, J., and Darzynkiewicz, Z. 2002. Detection of in situ activation of transglutaminase during apoptosis: Correlation with the cell cycle phase by multiparameter flow and laser scanning cytometry. Cytometry 49:83-89. Haugland, R.P. 2002. Handbook of Fluorescent Probes and Research Chemicals. Ninth Edition. Molecular Probes, Eugene, OR. Hotz, M.A., Traganos, F., and Darzynkiewicz, Z. 1992. Changes in nuclear chromatin related to apoptosis or necrosis induced by the DNA topoisomerase II inhibitor fostriecin in MOLT-4 and HL-60 cells are revealed by altered DNA sensitivity to denaturation. Exp. Cell Res. 201:184191. Hotz, M.A., Gong, J., Traganos, F., and Darzynkiewicz, Z. 1994. Flow cytometric detection of apoptosis: Comparison of the assays of in situ DNA degradation and chromatin changes. Cytometry 15:237-244. Hug, H., Los, M., Hirt, W., and Debatin, K.M. 1999. Rhodamine 110-linked amino acids and peptides as substrates to measure caspase activity upon apoptosis induction in intact cells. Biochemistry 38:13906-13911. Johnson, L.V., Walsh, M.L., and Chen, L.B. 1980. Localization of mitochondria in living cells with rhodamine 123. Proc. Natl. Acad. Sci. U.S.A. 77:990-994. Kamentsky, L.A. 2001. Laser scanning cytometer. Meth. Cell Biol. 63:51-87. Kaufmann, S.H., Desnoyers, S., Ottaviano, Y., Davidson, N.E., and Poirier, G.G. 1993. Specific proteolytic cleavage of poly(ADP-ribose) polymerase: An early marker of chemotherapy-induced apoptosis. Cancer Res. 53:3976-3985. Keiji, J.F., Bell-Prince, C., and Steinkamp, J.A. 2000. Staining of mitochondrial membranes with 10-nonyl acridine orange, MitoFluor Green, and MitoTracker Green is affected by mitochondrial membrane potential altering drugs. Cytometry 39:203-210. Kerr, J.F.R., Wyllie, A.H., and Curie, A.R. 1972. Apoptosis: A basic biological phenomenon with wide-ranging implications in tissue kinetics. Br. J. Cancer 26:239-257. Knapp, P.E., Bartlett, W.P., Williams, L.A., Yamada, M., Ikenaka, K., and Skoff, R.P. 1999. Programmed cell death without DNA fragmentation in the jimpy mouse: Secreted factors can enhance survival. Cell Death Differ. 6:136-145.
Nucleic Acid Analysis
7.19.31 Current Protocols in Cytometry
Supplement 25
Kockx, M.M., De Meyer, G.R., Muhring, J., Jacob, W., Bult, H., and Herman, A.G. 1998. Apoptosis and related proteins in different stages of human atherosclerotic plaques. Circulation 97:23072315.
Liu, J., Bhalgat, M., Zhang, C., Diwu, Z., Hoyland, B., and Klaubert, D.H. 1999. Fluorescent molecular probes V: A sensitive caspase-3 substrate for fluorometric assays. Bioorg. Med. Chem. Lett. 9:3231-3236.
Komoriya, A., Packard, B.Z., Brown, M.J., Wu, M.L., and Henkart, P.A. 2000. Assessment of caspase activities in intact apoptotic thymocytes using cell-permeable caspase substrates. J.Exp.Med. 191:1819-1828.
Majno, G. and Joris, I. 1995. Apoptosis, oncosis, and necrosis. An overview of cell death. Am. J. Pathol. 146:3-16.
Koopman, G., Reutelingsperger, C.P.M., Kuijten, G.A.M., Keehnen, R.M.J., Pals, S.T., and van Oers, M.H.J. 1994. Annexin V for flow cytometric detection of phosphatidylserine expression of B cells undergoing apoptosis. Blood 84:14151420. Kroemer, G. 1998. The mitochondrion as an integrator/coordinator of cell death pathways. Cell Death Differ. 5:547-548. Lajemi, M., Demignot, S., and Adolphe, M. 1998. Detection and characterization, using fluoresceincadaverine, of amine acceptor substrates accessible to active transglutaminase expressed by rabbit articular chondrocytes. Histochem. J. 30:499-508. Lazebnik, Y.A., Kaufmann, S.H., Desnoyers, S., Poirier, G.G., and Earnshaw, W.C. 1994. Cleavage of poly(ADP-ribose) polymerase by proteinase with properties like ICE. Nature 371:346347. Li, X., and Darzynkiewicz, Z. 1995. Labelling DNA strand breaks with BdrUTP. Detection of apoptosis and cell proliferation. Cell Prolif. 28:571579. Li, X., and Darzynkiewicz, Z. 2000. Cleavage of poly(ADP-ribose) polymerase measured in situ in individual cells: Relationship to DNA fragmentation and cell cycle position during apoptosis. Exp. Cell Res. 255:125-132. Li, X., Traganos, F., Melamed, M.R., and Darzynkiewicz, Z. 1995. Single-step procedure for labeling DNA strand breaks with fluorescein- or BODIPY-conjugated deoxynucleotides: Detection of apoptosis and bromodeoxyuridine incorporation. Cytometry 20:172-180. Li, X., Melamed, M.R., and Darzynkiewicz, Z. 1996. Detection of apoptosis and DNA replication by differential labeling of DNA strand breaks with fluorochromes of different color. Exp. Cell Res. 222:28-37. Li, X., Du, L., and Darzynkiewicz, Z. 2000. During apoptosis of HL-60 and U-937 cells caspases are activated independently of dissipation of mitochondrial electrochemical potential. Exp. Cell Res. 257:290-297. Liu, X., Kim, C.N., Yang, J., Jemmerson, R., and Wang, X. 1996. Induction of apoptotic program in cell-free extracts: Requirements for dATP and cytochrome c. Cell 86:147-157.
Marguet, D., Luciani, M.-F., Moynault, A., Williamson, P., and Chimini, G. 1999. Engulfment of apoptotic cells involves the redistribution of membrane phosphatidylserine on phagocyte and prey. Nature Cell Biol. 1:454-456. Melino, E. and Piacentini, M. 1998. “Tissue” transglutaminase in cell death: A downstream or multifunctional upstream effector? FEBS Lett. 430:59-63. Nagata, S. 2000. Apoptotic DNA fragmentation. Exp. Cell Res. 256:12-18. Nicoletti, I., Migliorati, G., Pagliacci, M.C., Grignani, F., and Riccardi, C. 1991. A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry. J. Immunol. Methods 139:271-280. Oberhammer, F., Wilson, J.M., Dive, C., Morris, I.D., Hickman, J.A., Wakeling, A.E., Walker, P.R., and Sikorska, M. 1993. Apoptotic death in epithelial cells: Cleavage of DNA to 300 and/or 50 kb fragments prior to or in the absence of internucleosomal fragmentation. EMBO J. 12:3679-3684. Ormerod, M.G. 1998. The study of apoptotic cells by flow cytometry. Leukemia 12:1013-1025. Ormerod, M.G., O’Neill, C.F., Robertson, D., and Harrap, K.R. 1994. Cisplatin induced apoptosis in a human ovarian carcinoma cell line without a concomitant internucleosomal degradation of DNA. Exp. Cell Res. 211:231-237. Ormerod, M.G., Cheetham, F.P.M., and Sun, X.-M. 1995. Discrimination of apoptotic thymocytes by forward light scatter. Cytometry 21:300-304. Petit, J.M., Ratinaud, M.H., Cordelli, E., Spano, M., and Julien, R. 1995. Mouse testis cell sorting according to DNA and mitochondrial changes during spermatogenesis. Cytometry 19:304-312. Poot, M., Gibson, L.L., and Singer, V.L. 1997. Detection of apoptosis in live cells by MitoTracker Red CMXRos and SYTO dye flow cytometry. Cytometry 27:358-364. Pozarowski, P., Halicka, D.H., and Darzynkiewicz, Z. 2003. The NK-κB inhibitor sesquiterpene parthenolide induces concurrently atypical apoptosis and cell necrosis: Difficulties in identification of dead cells in such cultures. Cytometry. In press. Ratinaud, M.H., Leprat, P., and Julien, R. 1988. In situ cytometric analysis of nonyl acridine orange-stained mitochondria from splenocytes. Cytometry 9:206-212.
Flow Cytometry of Apoptosis
7.19.32 Supplement 25
Current Protocols in Cytometry
Sallman, F.R., Bourassa, S., Saint-Cyr, J., and Poirier, G.G. 1997. Characterization of antibodies specific for the caspase cleavage site on poly(ADP-ribose) polymerase: Specific detection of apoptotic fragments and mapping of the necrotic fragments of poly(ADP-ribose) polymerase. Biochem. Cell Biol. 75:451-458. Scorrano, L., Petronilli, V., Colonna, R., Di Lisa, F., and Bernard, P. 1999. Chloromethyltetramethylrosamine (Mitotracker Orange) induces the mitochondrial permeability transition and inhibits respiratory complex I. Implications for the mechanism of cytochrome c release. J. Biol. Chem. 274:24657-24663. Shi, Y. 2002. Mechanisms of caspase activation and inhibition during apoptosis. Molec. Cell 9:459470. Smolewski, P., Bedner, E., Du, L., Hsieh, T.-C., Wu, J.M., Phelps, D.J., and Darzynkiewicz, Z. 2001. Detection of caspases activation by fluorochrome-labeled inhibitors: Multiparameter analysis by laser scanning cytometry. Cytometry 44:73-82. Susin, S.A., Zamzani, N., Larochette, N., Dallaporta, B., Marzo, I., Brenner, C., Hirsch, T., Petit, P.X., Geuskens, M., and Kroemer, G. 1997. A cytofluorometric assay of nuclear apoptosis induced in a cell-free system: Application to ceramide-induced apoptosis. Exp. Cell Res. 236:397-403.
Umansky, S.R., Korol, B.A., and Nelipovich, P.A. 1981. In vivo DNA degradation in the thymocytes of gamma-irradiated or hydrocortisonetreated rats. Biochim. Biophys. Acta 655:9-17. van Engeland, M., Nieland, L.J.W., Ramaekers, F.C.S., Schutte, B., and Reutelingsperger, P.M. 1998. Annexin V–affinity assay: A review on an apoptosis detection system based on phosphatidylserine exposure. Cytometry 31:1-9. Vermes, I., Haanen, C., and Reutelingsperger, C. 2000. Flow cytometry of apoptotic cell death. J. Immunol. Methods 243:167-190. Waggoner, A.S. 1979. Dye indicators of membrane potential. Annu. Rev. Biophys. Bioeng. 8:47-68. Yang, J., Liu, X., Bhalla, K., Ibrado, A.M., Cai, J., Peng, T.I., Jones, D.P., and Wang, X. 1997. Prevention of apoptosis by Bcl-2: Release of cytochrome c from mitochondria blocked. Science 275:1129-1132. Zamzani, N., Brenner, C., Marzo, I., Susin, S.A., and Kroemer, G. 1998. Subcellular and submitochondrial mode of action of Bcl-2-like oncoproteins. Oncogene 16:2265-2282.
Contributed by Piotr Pozarowski School of Medicine Lublin, Poland
Swat, W., Ignatowicz, L., and Kisielow, P. 1981. Detection of apoptosis of immature CD4+8+ thymocytes by flow cytometry. J. Immunol. Methods 137:79-87.
Jerzy Grabarek Pomeranian School of Medicine Szczecin, Poland
Telford, W.G., Komoriya, A., and Packard, B.Z. 2002. Detection of localized caspase activity in early apoptotic cells by laser scanning cytometry. Cytometry 47:81-88.
Zbigniew Darzynkiewicz New York Medical College Valhalla, New York
Nucleic Acid Analysis
7.19.33 Current Protocols in Cytometry
Supplement 25
Analysis of Fine-Needle Aspirate Biopsies from Solid Tumors by Laser Scanning Cytometry (LSC)
UNIT 7.20
In studying the pathology of solid tumors, morphologic documentation of the analyzed events is advantageous for any cytometric analysis. This is very important if decisions on the course of further therapy depend on the results of such analysis. Additionally, in this situation, a nondestructive and nonconsumptive method of analysis is required that allows storage of the analyzed sample for later re-evaluation. Since a growing number of diagnostic and therapeutic steps are performed on an outpatient basis, the ability to obtain samples by minimally invasive approaches is also important. In surgical pathology of solid tumors, ploidy is recognized as a prognostic marker in a growing number of tumor entities. This unit describes the determination of ploidy in carcinomas by analyzing minimal sample volumes such as fine-needle aspirate biopsies (FNABs) using the slide-based technology of LSC (see Basic Protocol). These samples can be obtained on an outpatient basis, and analysis by LSC is nondestructive and nonconsumptive since the cells are immobilized on a slide and can be stored as normal cytological samples. An Alternate Protocol is also presented to describe variations from the Basic Protocol in cases where FNABs contain a high number of eosinophilic granulocytes. DETERMINATION OF PLOIDY IN CARCINOMAS BY LSC OF FINE-NEEDLE ASPIRATE BIOPSIES
BASIC PROTOCOL
FNABs are taken using a 20-ml syringe connected to a 27-G (or smaller-diameter; i.e., ≤ 0.4-mm) needle. The material is resuspended in phosphate-buffered saline (PBS) and erythrocytes are lysed. The cells are then mounted on conventional glass microscope slides, air dried, fixed, and stored in ethanol. For analysis, cells are stained for cytokeratin by immunofluorescence and for DNA by incubation with propidium iodide (PI). Analysis in the LSC is then triggered on nuclear fluorescence, and a variety of parameters are determined for the fluorescence of cytokeratin and DNA and for the forward light scatter (FS). For the interpretation of the data, the slide is counterstained with hematoxylin/eosin (H&E) and the DNA content of the cytokeratin-negative leukocytes is set to 1.0. The DNA content of the cytokeratin-positive tumor cells is then determined and the DNA index (DI) of the tumor is calculated as the ratio of tumor DNA content to leukocyte DNA content; in order to obtain more commonly used ploidy values, the DI is multiplied by the factor 2 (i.e., DI 1.00 = 2c). The morphology of the different cell types (leukocytes, tumor cells) is verified and documented with the built-in CCD camera, or any other camera, by taking micrographs after relocalization of single cells in the LSC. Materials Subject with solid tumor of interest PBS with 1% and 0.5% BSA (see recipe) Erythrocyte lysing solution: ammonium chloride lysing solution (APPENDIX 2A) or commercial lysing product (e.g., FACS Lysing Solution from Becton-Dickinson) 70% ethanol Phosphate-buffered saline (PBS; APPENDIX 2A) FITC-conjugated anti-cytokeratin antibody and negative-control isotype (both mouse lgGs; Dako) Anti-FITC antibody conjugated to Alexa Green (Molecular Probes) PBS with 0.5% BSA, 50 µg/ml PI, and 100 µg/ml RNase (see recipe) Glycerol/PBS/PI (see recipe) Contributed by Andreas O.H. Gerstner and Attila Tárnok Current Protocols in Cytometry (2002) 7.20.1-7.20.10 Copyright © 2002 by John Wiley & Sons, Inc.
Nucleic Acid Analysis
7.20.1 Supplement 22
Rubber cement (optional) Hematoxylin and eosin: purchase commercial H&E staining kit or see Zeller and Rogers (1993) Ethanol dehydration series: 70%, 80%, 90%, and 100% ethanol Xylene Permanent medium for cytological slides (e.g., Eukitt from Kindler GmbH) 20-ml syringe with 27-G (or narrower) needle Glass microscope slides, cover glasses, and grease pencil for use on glass slides Humidified chamber Laser scanning cytometer (CompuCyte) equipped with argon and helium-neon lasers and standard filter settings Additional reagents and equipment for hematoxylin/eosin staining (Zeller and Rogers, 1993) Obtain and prepare sample 1. Take FNABs by repeated multidirectional puncture of the tumor using a 20-ml syringe attached to a 27-G or narrower needle. In order to yield high cell counts, produce a vacuum within the syringe by pulling the piston after the first puncture and maintaining the vacuum all the way through; release the vacuum just at the very end by disconnecting the syringe from the needle, but reconnect it prior to retracting the needle from the tumor. Use a 27-G or smaller needle in order to avoid tumor-cell contamination into overlying tissue. Publications report skin metastases that seemingly arose from tumor puncture using needles with larger diameters (Mighell and High, 1998; Smith, 1991).
2. Resuspend tissue material in a 1.5-ml microcentrifuge tube prefilled with 100 µl PBS with 1% BSA, pH 7.40, and store at 4°C for up to 12 hr until further preparation. Resuspension in a tube rather than direct placement onto a glass slide offers the opportunity for further modification of the cell preparation (see below) as well as control over the cell density and homogeneity of the preparation on the slide. Alternatively, one may take swabs with a cotton tip from any tumor localized at the mucous membranes (during endoscopies, by direct or indirect visualization, e.g., from the larynx); to recover cells dip the cotton tip repeatedly into a 1.5-ml microcentrifuge tube prefilled with 200 ìl PBS with 1% BSA (Gerstner et al., 2002a).
3. Add 1 ml erythrocyte lysing solution to each tube and incubate 15 min at room temperature. Microcentrifuge 5 min at 250 × g, room temperature, and decant the supernatant. Resuspend the pellet in 50 µl PBS with 0.5% BSA. 4. Transfer the cell suspension onto glass slides. To facilitate and standardize this step, make use of a template drawing placed underneath the glass slide to show the areas onto which cells are regularly placed (“cell spots,” e.g., two separate squares each measuring 1.2 × 1.2 cm). Prepare at least one slide with two cell spots to allow positive staining and negative control on the same slide. Air dry slides 1 hr and store in 70% ethanol at room temperature.
Analysis of Fine-Needle Aspirate Biopsies by LSC
Only very rarely is enough material obtained to perform cell counting; therefore in only a very few cases can this step be performed in strictly quantitative fashion. As a rule of thumb, adjust the volume placed onto the slide or dilute the cell suspension in such a way that cells lie separated from each other by ∼1 cell diameter—check this by direct visualization with a standard microscope. If cells are too close, too many events will be triggered as doublets; if cells are too scattered, analysis will take longer (see Critical Parameters). For swabs, in general, no lysing step is required and specimens can be put directly onto slides as described above.
7.20.2 Supplement 22
Current Protocols in Cytometry
Stain cells on the slide 5. Remove slides from ethanol and wash twice, 5 min each time, by placing in vessels prefilled with PBS. Circle cell spots on the slide with a grease pencil, then place slides horizontally with the cell side up and pipet 200 µl PBS with 1% BSA onto each cell spot. Incubate 15 min at room temperature, then decant solution from the slide by inversion. 6. Dilute 5 µl FITC-conjugated anti-cytokeratin antibody in 95 µl PBS/0.5% BSA and add this 100 µl solution to one spot for specific staining. Dilute 5 µl FITC-conjugated negative-control isotype antibody in 95 µl PBS/0.5% BSA and add this to the other spot as a control. Incubate 30 min in a humidified chamber. Wash by adding 0.5 ml PBS to each cell spot and decanting by inversion. A humid chamber can be easily made using a 1-liter beaker. Place paper towels soaked in distilled water on the bottom and cover with aluminum foil. From this step onwards protect the slide from light.
7. For each spot on the slide, dilute 5 µl anti-FITC antibody conjugated to Alexa Green (Molecular Probes) in 95 µl PBS containing 0.5% BSA, 50 µg/ml PI, and 100 µg RNase. Add 100 µl of this solution to each spot and incubate 30 min in a humidified chamber. Wash by adding 0.5 ml PBS to each spot and decanting by inversion. 8. Cover each spot with 40 µl glycerol/PBS/PI and apply cover glasses as described below. Store slide at 4°C until analysis; if the period until data acquisition (see step 9) is >1 hr, seal the cover glasses with rubber cement in order to prevent evaporation. At this step, it is important to avoid air-bubble inclusions since they will interfere with triggering. In order to place the cover glasses in a bubble-free manner, pipet the 40 ìl glycerol/PBS/PI onto the cover glass (instead of onto the slide), turn the cover glass gently, and place it, slightly tilted, onto the slide. Do not press the cover glass on the slide using any kind of tool; this might damage the cells, leading to gross artifacts.
Acquire data 9. Within 24 hr after staining, place the slide in the motorized stage of the LSC in a standardized and reproducible manner; this is crucial for exact relocalization of single cells of interest (see Critical Parameters). Use the 20× objective. Trigger the analysis on the PI signal using the FL3-Ar (625/DF28 band-pass filter) channel with the following settings: minimal area = 20 µm2; and threshold level = 1,200 (these settings depend on the PMT settings and must be checked initially before starting the data acquisition). Acquire as many cells per spot as possible (∼3000 as a minimum). For more detailed description of the LSC, refer to review literature (Kamentsky and Kamentsky, 1991; Kamentsky et al., 1997; Tárnok and Gerstner, 2002). In general, check the PMT voltage first in order to avoid saturated pixels and adjust PMT settings as appropriate (e.g., voltage 30%, offset 2075, gain 255); however, only very rarely should changes be necessary from analysis to analysis. Next, adapt the trigger settings so that cells are neither fused to doublets nor left out from analysis; this step depends on a uniform deposition of the cells onto the slide (see step 4). Activate the appropriate channels for analysis; for FITC staining, these are FL1-Ar (530/DF30 band-pass filter; FL3-Ar [625/DF28 band-pass filter), and FS-Ar. Since the LSC is a system with high focal depth rather than a confocal system, the focal plane of the microscope does not need to be set very exactly; this also makes the system insensitive to uneven positioning of the slide on the stage.
Nucleic Acid Analysis
7.20.3 Current Protocols in Cytometry
Supplement 22
Interpret data 10. Place the slide upright for 15 min in a container prefilled with PBS and let the coverglass slip off the slide. Perform H&E staining (Zeller and Rogers, 1993, or see manufacturer’s instructions for kit). If the slide has been sealed with rubber cement (see step 8), gently pull off the rubber cement prior to this step.
11. Dehydrate the cells by dipping the slide successively in 70%, 80%, 90%, and 100% ethanol, then incubate 3 min in xylene. Remove excess xylene from the slides by decanting, but do not allow the slides to dry. Cover the slide with permanent medium. Store slides as normal cytological samples.
Cytokeratin-FITC (integral/area) 100 100000
DNA-PI (integral)
105 84 63 42 Count
negative control
100
IgG-FITC (integral/area) 100000
A
21 0
5.4
DNA-PI (integral)
5.4
DNA-PI (integral)
5.4
B
Leukocytes
Normal epithelial cells
Reference cells
Analysis of Fine-Needle Aspirate Biopsies by LSC
Tumor cells
Figure 7.20.1 Typical example of LSC analysis of an FNAB from a cervical mass (surgical level III in the right neck). One spot is stained for DNA (with PI) and for cytokeratin-FITC; the other spot is stained with PI plus a control IgG-FITC. Immunostaining is amplified by anti-FITC-Alexa for both. Cells are analyzed by the LSC. (A) Display of the LSC data: Negative control (left dot plot); specific staining (center dot plot and right histogram). Dot plots display PI-integral fluorescence versus FITC-integral fluorescence per area. The histogram shows an overlay of PI histograms from the cytokeratin-positive cells (black area) and from the cytokeratin-negative cells (black contour). Cytokeratin-positive versus negative cells are gated by a 5% cut-off set in the negative control. Numbers within the central dot plot and the histogram correspond to the cells relocalized in panel (B). (B) The slide is restained by H & E and single cells from the gates 1 to 5 are relocalized: 1, cytokeratin-negative leukocytes; 2, cytokeratin-positive normal epithelial cells; 3 to 5, hypo- and hyperploid tumor cells. This cervical mass turned out to be a metasis from a hypopharyngeal carcinoma classified as pT2 pN1 cMO.
7.20.4 Supplement 22
Current Protocols in Cytometry
12. Analyze the data to determine the cytokeratin expression and to calculate the ploidy of the tumor (see Figure 7.20.1). Make use of the relocalization feature of the LSC. Create a display showing the PI integral versus the cytokeratin integral per area. Set a 5% cut-off for IgG/FITC-integral per area with the control sample. For the specific staining, create a DNA histogram overlay for the cytokeratin-negative and the cytokeratin-positive fractions. Identify peripheral blood leukocytes within the cytokeratin-negative cells and set the DNA index of these cells to 1.0. Analyze and document the morphology of single cells representative of different subsets. Since the listmode files are in FCS 3.0 standard, data are most easily analyzed with the proprietary WinCyte software. Alternatively they could be exported and downsized to FCS 2.0 standard and then be analyzed with other convenient software. In WinCyte, set the DI in the DNA histogram by gating the peak corresponding to the leukocytes and activate the context menu; choose “DNA Index,” then “Set DI,” where one must specify the cell type as “human.” Following that step, one can display the DI of any peak in the DNA histogram as well as in dot plots showing the DNA integral on one axis. For this option, choose “Tag DI” from the sub-menu “DNA Index”; move the tag to any dot or peak as appropriate.
DETERMINATION OF PLOIDY IN CARCINOMAS BY LSC OF EOSINOPHIL-RICH SAMPLES In some patients fine-needle aspirate biopsies contain a high number of eosinophilic granulocytes. The problem with eosinophilic granulocytes is that they absorb unbound FITC, which is found in any vial of FITC-conjugated antibodies. Therefore, by detecting cytokeratin via FITC, one would also stain eosinophilic granulocytes within the sample; although they exhibit distinct morphologic features (high FS MaxPixel values due to the granules) they then could be taken as a diploid cytokeratin-positive subpopulation (Bedner et al., 1999). Of course, this misinterpretation could be corrected by means of the relocalization feature of the LSC, but this phenomenon would still interfere with the calculation of cell percentages (such as the percentage of cells with DI > 2.5, i.e., 5c-exceeding rate). Nevertheless, in order to reliably circumvent this pitfall, especially in samples with high cell counts of eosinophilic granulocytes, the authors recommend choosing allophycocyanin (APC) as the fluorochrome for detection of cytokeratin expression (Gerstner et al., 2002b).
ALTERNATE PROTOCOL
Additional Materials (also see Basic Protocol) Anti-cytokeratin antibody and negative-control isotype, both pure unconjugated IgG1 from mouse (Dako) Anti-mouse IgG antibody conjugated to biotin (Caltag) Streptavidin conjugated to allophycocyanin (APC) (Caltag) Replace steps 6, 7, and 9 of the Basic Protocol with steps 6a, 7a, and 9a below. 6a. Dilute 5 µl unconjugated anti-cytokeratin-antibody in 95 µl PBS containing 0.5% BSA and add this 100 µl solution to one spot. Dilute 5 µl of the corresponding unconjugated negative-control isotype antibody in µl PBS containing 0.5% BSA and add this to the other spot as a control. Incubate 30 min in a humidified chamber. Wash by adding 0.5 ml PBS to each cell spot and and decanting by inversion. Antibodies should be murine; otherwise the reagent used in step 7a must be changed appropriately in order to detect the first antibody.
7a. For each spot on the slide, dilute 5 µl biotinylated anti-mouse antibody in 95 µl PBS containing 0.5% BSA. Add 100 µl of this solution to each spot and incubate for 30 min in a humidified chamber. Wash by adding 0.5 ml PBS to each spot and decanting by inversion. Again for each spot on the slide, dilute 5 µl streptavidin conjugated to APC in 95 µl PBS containing 0.5% BSA, 50 µg/ml PI, and 100 µg/ml RNase. Add
Nucleic Acid Analysis
7.20.5 Current Protocols in Cytometry
Supplement 22
100 µl of this solution to each spot and incubate 30 min in a humidified chamber. Wash by adding 0.5 ml PBS to each spot and decanting by inversion. 9a. Acquire data as in step 9 of the Basic Protocol but change the display appropriately to show APC fluorescence. APC requires both lasers of the LSC; activate the channels FL3-Ar (see Basic Protocol 1 for definition), FS-Ar, and FL4-HeNe (far red, APC; 650 EFLP long-pass filter). This modification doubles the time needed for data acquisition as compared to FITC staining but prevents misinterpretation of FITC-positive diploid cells, which could be either cytokeratin-positive epithelia or eosinophils. In order to make full use of the APC modification and to discriminate the eosinophils, one can also activate channel FL1-Ar (see Basic Protocol for definition) to record the green autofluorescence of eosinophils. The green fluorescence of eosinophils can be further enhanced by adding pure FITC at one of the staining steps (Bedner et al., 1999).
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Glycerol/PBS/PI Pipet 250 µl PBS (APPENDIX 2A) into 750 µl glycerol and mix gently by repeated resuspension until the interface between the two reagents has completely disappeared. Remove 50 µl solution and replace with 50 µl of 500 µg/ml propidium iodide (PI) stock solution to obtain the final PI concentration of 25 µg/ml. Store up to 3 months at 4°C. PBS with 1% or 0.5% BSA Stock solution: Prepare 10% (w/v) stock solution of bovine serum albumin (BSA) in phosphate-buffered saline (PBS; APPENDIX 2A). Adjust pH to 7.40, filter sterilize, and store ≤6 months at −20°C in 1-ml aliquots. Working solution (1% or 0.5% BSA): In order to prepare solutions for incubation and staining, add 100 µl of 10% BSA stock solution to 900 µl PBS, pH 7.40 (for 1% BSA) or 50 µl of 10% BSA stock solution to 950 µl PBS, pH 7.40 (for 0.5% BSA) and store at 4°C; this solution should be used within 3 days. PBS with 0.5 % BSA, 50 ìg/ml PI, and 100 ìg/ml RNase Stock solutions: 10% (w/v) BSA stock (see recipe for PBS with 1% or 0.5% BSA, above) 1 mg/ml RNase A (store ≤6 months in 100-µl aliquots at −20°C; store thawed aliquots ≤3 days at 4°C) 500 µg/ml propidium iodide (PI; store up to 6 months at 4°C) Working solution: Combine 5 µl of 10% BSA stock (0.5% w/v final), 10 µl of 1 mg/ml RNase stock (100 µg/ml final), and 10 µl of 500 µg/ml PI stock (50 µg/ml final) with PBS (APPENDIX 2A), for a total volume of 100 µl. COMMENTARY Background Information
Analysis of Fine-Needle Aspirate Biopsies by LSC
The term ploidy relates to the amount of DNA present in a cell. Aneuploid cells have one or more chromosomes or segments of chromosomes absent from or in addition to a normal euploid complement. Aneuploidy is nowadays known to be a major characteristic of malig-
nancy and to play an initial role in cancer formation (Li et al., 2000; Sen, 2000), and is found mostly in transformed cells. Aneuploidy has been shown to be a prognostic marker for several tumor entities (Franzén et al., 1991; Hemmer et al., 1997; Welkoborsky et al., 1999; Millot and Dufer, 2000). Analysis of ploidy can
7.20.6 Supplement 22
Current Protocols in Cytometry
also be used in follow-up of cancer patients after treatment of a known aneuploid tumor in order to diagnose recurrence or minimal residual disease. In order to optimize the therapeutic regimen for an individual patient with a solid tumor, it would be desirable to determine the ploidy at an early stage of disease and diagnosis. Common methods for the analysis of ploidy are flow cytometry and image cytometry (e.g., after Feulgen staining). Both methods are well described and yield reliable and reproducible data. Nevertheless, ploidy rarely is a critical parameter for the decision making in the diagnostic or therapeutic regimen of carcinomas in a routine clinical setting. This might be a consequence of the drawbacks of these methods. For flow cytometry rather large samples are needed and there is no possibility of documenting cell morphology; for image cytometry, only low cell counts (<300) are feasible for routine application and there is substantial observer bias. Analysis by LSC allows a combination of the advantages of both methods but avoids their disadvantages—although only very small samples are needed, high cell counts are still achieved (≥10,000 cells). Since cells are immobilized on a slide, they are preserved for later reanalysis as well as for restaining and documentation. Due to the extremely small cell requirements, specimens can be obtained by minimally invasive procedures (FNABs and swabs). Therefore, tumors from anatomical sites which cannot be biopsied routinely (e.g., from the paratid gland, larynx, or paranasal sinuses) could be analyzed sufficiently even on an outpatient basis. For these reasons, analysis by LSC has potential to become a powerful tool in cancer diagnosis. A disadvantage of ploidy analysis by LSC is the speed of analysis; depending on the slide preparation, 10,000 cells could be analyzed in 2 to 10 min by LSC with one laser, whereas the same cell number would be analyzed within 4 to 20 sec by flow cytometry with up to three lasers. Additionally, once the data from one scanning step are calculated, all other information included in the primary pixel map is lost; therefore a complete rescanning must be performed in case trigger settings turn out to be inappropriate (see Critical Parameters). Another disadvantage is the relatively broad coefficient of variation (CV) of DNA peaks. DAPI and Hoechst dyes yield the best CVs of 1% to 2% for DNA peaks (Hemmer et al., 1999). In new LSC instruments, a violet diode laser may
be implemented that allows application of these dyes. In general, many assays originally developed for flow cytometric analysis can be adapted for analysis by LSC. The main difference is that most incubations have to be performed on the slide. Nevertheless, staining characteristics and reaction parameters of nucleic acid dyes, antibodies, and other probes (FISH, mitochondrial potential) are identical. The authors have recently reviewed clinical applications of LSC (Tárnok and Gerstner, 2002).
Critical Parameters The most prominent feature of the LSC is the ability to observe cells directly during analysis and to document their morphology; the tools are the “Scan Data” key in the WinCyte display and the relocalization process. “Scan Data” (available only during data acquisition) shows the pixel map of the actual scanning step in process at that moment; it optionally shows the user-defined contours as they are applied for analysis (see below). This is helpful in adjusting the triggering and in visualizing stainings that are too weak to be judged by eye. To make full use of the relocalization, two points are critical. First, it is crucial to place the slide on the microscope stage in a standardized and reproducible way before data acquisition. Changes of the slide position within the stage during a running data acquisition will make it impossible to relocalize a cell. Best results concerning the preservation of the x-y coordinates will be obtained by relocalizing directly after scanning. At this point, only the fluorescent staining is present. Therefore, slides will be restained with conventional cytopathological methods. The second critical factor is that restaining must be performed without shifts of the cells on the slide, or excessive cell loss. Therefore, do not drag the cover glass from the slide but let it slip off by gravity in a container filled with PBS. Nevertheless, cell loss will always be observed to some extent on relocalization, but this should neither exceed 5% nor show preference for one cell type. Like relocalization, most other critical parameters arise from the fact that analysis is slide based. As for any other cytometric assay, the triggering is of fundamental importance. In LSC, the triggering is similar to that in image analysis; first select a triggering parameter, which in this protocol is DNA fluorescence. Remember that events on the slide will be represented on a gray-scale pixel map where
Nucleic Acid Analysis
7.20.7 Current Protocols in Cytometry
Supplement 22
Analysis of Fine-Needle Aspirate Biopsies by LSC
the brightness of a pixel represents the fluorescence intensity detected at this spot on the slide. Therefore, a cell would be a cluster of pixels. Next, set the criteria that must be fulfilled in order to include a cluster of pixels in the analysis—one must define a minimal area covered continuously by pixels with a minimal brightness. This will generate a triggering contour around a cluster of pixels which in this protocol will include the nucleus. In order to include the cytoplasmic rim, a second contour will be drawn at a user-defined distance from the triggering contour (e.g., 10 pixels); this is the actual analysis contour. In order to subtract background staining per cell, an additional two contours for each individual cell are drawn in the periphery at user-defined distances. All these settings should be such that (a) the entire nucleus is included in the analysis, (b) only cells but neither larger nor smaller particles (e.g., dust or bacteria) are triggered, and (c) cells are triggered as single cells but not fused to doublets (for further discussion refer to Tárnok and Gerstner, 2002.) Obviously, these settings depend on the “density” of the cells on the slide, which presents a kind of Scylla and Charybdis quandary—if cells are too close to each other it can be almost impossible to set the triggering properly, whereas if cells are too far away from each other the data acquisition will take a very long time because a large area has to be scanned. Therefore, the initial steps of the protocol (obtaining an adequate FNAB or swab and preparing the cells on the slide) are very important. During staining it is important not to contaminate the solutions for the specific staining and for the control with one other. To this aim, the respective cell spots on the slide must be tightly encircled by the grease pencil. Additionally, slides must be prevented from drying out and be protected from light from that step onwards. Also take care that the entire spot is covered by the staining solution without bubbles. This is of highest importance for PI staining; cells that are not properly covered are incubated with reduced or even zero concentration of PI and will therefore not be properly triggered and could yield erroneously false aneuploid populations. Changes in PI concentration through dissociation of the dye from DNA could even take place after completion of staining; in order to preset an equilibrium without uncontrolled shifting of PI from the nucleus to the extracellular space (which could be a problem at the rim of the spot), add extra PI to the embedding medium as described.
Although not a primarily important parameter in this protocol, forward scatter could help in discriminating cells. In order to yield optimal forward scatter results one must adjust the optical alignment of the condensing lens, its shutter, and the photodiode underneath the condenser that detects the scattered light. In the authors’ experience, adjusting the lens almost directly underneath the microscope stage and opening the shutter completely has turned out to be optimal. The light passing through the lens should hit the obscuration bar (which is fixed just above the photodiode) in such a way that approximately 95% is blocked but the remaining 5% passes the bar and hits the photodiode. These adjustments can be performed offline using the screws of the condenser holding the lens. When analyzing a sample, check for proper setting at the beginning of the analysis by opening “Scan Data” and choosing FS; cells should resemble phase-contrast microscopy images on a dark gray background.
Troubleshooting First, there could be too few cells triggered. Check the slide by direct visualization in the fluorescence microscope and look for red nuclei (do not forget under any circumstances to realign the optical pathway for later analysis). If nuclei are not observed, but FNABs (or swabs) were properly prepared, cells might have been lost during fixation; either cell spots were not airdried sufficiently or the glass of the microscope slide was inadequate for the cell type. Try slides for frozen sections instead. However, sometimes the FNAB does not yield enough cells at the very beginning; this occurs with specific solid tumors (e.g., Schwannomas, neurinomas, recurrent local disease in previously irradiated carcinoma, or necrosis) and actually should be obvious at preparation. If cells on the slide can be seen by bright-field but not by fluorescence microscopy, the incubation with PI might have been faulty. However, if there are enough cells on the slide which are not displayed in the “Scan Data” window, check the analysis protocol and make sure that the PI channel FL3-Ar (see Basic Protocol and Alternate Protocol for the specific meaning of these channels) is activated, that the PMT is set correctly with pixels neither too bright nor too weak, and that the trigger settings are adequate for this specimen. Another problem might be that cells are triggered properly but there is no specific cytokeratin signal detectable. Besides pipetting errors and degraded antibodies, the cause could
7.20.8 Supplement 22
Current Protocols in Cytometry
be the use of denatured ethanol for fixation, which could also affect the H&E staining. Unfortunately, this effect is irreversible. As a result of some kind of preparation artifact there might exist a cytokeratin-positive as well as a cytokeratin-negative population with exactly the same DI; the negative events usually are bare nuclei stripped of their cytoplasm during preparation. Other reasons for negative cytokeratin staining are the expression of cytokeratin with altered epitopes by the tumor or a nonepithelial origin of the tumor. However, even in these situations the relocalization feature of the LSC still provides a way to determine the ploidy of the tumor cells. In these cases, there is either only one diploid (cytokeratin-negative) population in a diploid tumor; or there are two (or more) populations. One of these populations should be identifiable as the leukocytes by relocalization, whereas the other(s) would be expected to consist of aneuploid/polyploid tumor cells. Sometimes very broad DNA peaks or aneuploid sub-populations with very low cell numbers are observed. One should always exclude a staining artifact. This could be confined to a small area within the cell spot that probably was not incubated properly (e.g., air bubble) and could cause a false aneuploid sub-population. Alternatively, there could be a loss of PI intensity towards the rim of the spot. This is a problem of dye concentration which can be circumvented by confining the analysis area to the inner portion of the cell spot or by using a shaking device for slow rotation of the humidified chamber during incubation. However, if rotation is too fast, a depression within the incubation solution could result, leading to a decreased PI concentration in the center of the cell spot. In order to detect all these staining artifacts, create dot plots showing x versus y position, x position versus PI intensity, and y position versus PI intensity. Gate and colorize the suspicious subpopulations or slopes of the DNA peak and analyze their location on the slide. One should also relocalize them directly after scanning to analyze their morphology. Another possibility is contamination of the slide by bacteria, yielding irregular changes in PI intensity. In some cases, cell loss from analysis to relocalization after H&E staining might be unacceptably high. In general, this is a mechanical problem—either cells were stripped off the slide by harsh removal of the cover slip or they slipped off during the H&E staining procedures. In the latter case let slides air dry suffi-
ciently (see Basic Protocol, step 4) or try other cytological staining procedures. However, instead of real cell loss there could also be a rather virtual “relocalization loss”; this means that cells actually are still in place but cannot be found. The cause might be incorrect insertion of the slide in the microscope stage either for data acquisition or for relocalization. Reinsert the slide for re-localization first. If cells are still undetectable but one wishes to obtain micrographs from relevant cells, one should try to relocalize cells of a small subpopulation that show specific morphologic characteristics and can easily be identified in the microscope by manual searching. Such cells are eosinophils or doublets. Once the proper location of these cells is known, set a temporary offset for the x and y coordinate (choose “Visualize” from the menu, then “Temporary Offset...”) so that cells are then properly relocalized. One might also try to reposition the slide manually as it probably had been placed for analysis, but these attempts are rarely successful.
Anticipated Results Bivariate analysis of FNABs (or swabs) from solid tumors for DNA content and cytokeratin expression by this protocol should determine the DI of the tumor cells as compared to normal peripheral blood leukocytes. Setting a 5% cutoff in the negative control sample should identify a cytokeratin-positive and a cytokeratin-negative fraction in the specifically stained sample. Within the cytokeratin-negative fraction, identify the peripheral blood leukocytes and set the DI of these reference cells to 1.0. The DI of the tumor cells will then be calculated by the software and can easily be displayed. The CV of the DNA peaks should be in the range of 4% to 9%. Tumors showing a DI of 1.0 are likely to be benign, although the possibility a sampling error or a malignant diploid tumor is not excluded. Aneuploid tumors would be considered to be transformed or highly suspect as malignant.
Time Considerations Obtaining the FNAB and preparing the slides requires 45 min. Air drying takes 60 min but can be shortened by drying the slides under a fume hood. The staining procedure takes 1.5 hr for detection of cytokeratin by FITC and 2 hr for detection by APC. The data acquisition by LSC takes about 30 min for scanning the positive and the control spot; this step depends on the cell density on the slide (see Critical Parameters). H&E counterstaining of slides
Nucleic Acid Analysis
7.20.9 Current Protocols in Cytometry
Supplement 22
takes 60 min, and analysis of data with documentation of single cells takes 30 min.
Kamentsky, L.A., Burger, D.E., Gershman, R.J., Kamentsky, L.D., and Luther, E. 1997. Slidebased laser scanning cytometry. Acta Cytol. 41:123-143.
Literature Cited
Li, R., Sonik, A., Stindl, R., Rasnick, D., and Duesberg, P. 2000. Aneuploidy vs. gene mutation hypothesis of cancer: Recent study claims mutation but is found to support aneuploidy. Proc. Natl. Acad. Sci. U.S.A. 97:3236-3241.
Bedner, E., Halicka, H.D., Cheng, W., Salomon, T., Deptala, A., Gorczyca, W., Melamed, M.R., and Darzynkiewicz, Z. 1999. High affinity binding of fluorescein isothiocyanate to eosinophils detected by laser scanning cytometry: A potential source of error in analysis of blood samples utilizing fluorescein-conjugated reagents in flow cytometry. Cytometry 36:77-82. Franzén, G., Klausen, O.G., Grenko, R.T., Carstensen, J., and Nordenskjöld, B. 1991. Adenoid cystic carcinoma: DNA as an prognostic indicator. Laryngoscope 101:669-673. Gerstner, A.O.H., Machlitt, J., Laffers, W., Tárnok, A., and Bootz, F. 2002a. Analysis of minimal sample volumes from head and neck cancer by laser scanning cytometry. Onkologie 25:40-46. Gerstner, A.O.H., Machlitt, J., Welkoborsky, H.-J., Bootz, F., and Tárnok, A. 2002b. Analysis of ploidy in hypopharyngeal cancer by laser scanning cytometry on fine needle aspirate biopsies. Anal. Cell Pathol. In press. Hemmer, J., Thein, T., and van Heerden, W.F.P. 1997. The value of DNA flow cytometry in predicting development of lymph node metastasis and survival in patients with locally recurrent oral squamous cell carcinoma. Cancer 79:23092313. Hemmer, J., Nagel, E., and Kraft, K. 1999. DNA aneuploidy by flow cytometry is an independent prognostic factor in squamous cell carcinoma of the oral cavity. Anticancer Res. 19:1419-1422. Kamentsky, L.A. and Kamentsky, L.D. 1991. Microscope-based multiparameter laser scanning cytometer yielding data comparable to flow cytometry data. Cytometry 12:381-387.
Mighell, A.J. and High, A.S. 1998. Histological identification of carcinoma in 21 gauge needle tracks after fine needle aspiration biopsy of head and neck carcinoma. J. Clin. Pathol. 51:241-243. Millot, C. and Dufer, J. 2000. Clinical applications of image cytometry to human tumour analysis. Histol. Histopathol. 15:1185-1200. Sen, S. 2000. Aneuploidy and cancer. Curr. Opin. Oncol. 12:82-88. Smith, E.H. 1991. Complications of percutaneous abdominal fine-needle biopsy. Radiology 178:253-258. Tárnok, A. and Gerstner, A.O.H. 2002. Clinical applications of laser scanning cytometry (LSC). Cytometry 50:133-143. Welkoborsky, H.J., Gluckmann, J.L., Jacob, R., Bernauer, H., and Mann, W. 1999. Tumor biologic prognostic parameters in T1N0M0 squamous cell carcinoma of the oral cavity. Laryngorhinootologie 78:131-138. Zeller, R. and Rogers, M. 1993. Counterstaining and mounting of autoradiographed in situ hybridization slides. In Current Protocols in Molecular Biology (F.A., Ausubel, R., Brent, R.E., Kingston, D.D. Moore, J.G., Seidman, J.A., Smith, and K. Struhl, eds.) pp.14.5.12D14.5.5. John Wiley & Sons, New York.
Contributed by Andreas O.H. Gerstner and Attila Tárnok University of Leipzig Leipzig, Germany
The authors thank A. Tannapfel, Department of Pathology, University of Leipzig, for her helpful comments on this protocol and J. Machlitt, Department of Ear, Nose, and Throat, University of Leipzig, for her committed support in the preparation of specimens.
Analysis of Fine-Needle Aspirate Biopsies by LSC
7.20.10 Supplement 22
Current Protocols in Cytometry
Measurement of Cytogenetic Damage in Rodent Blood with a Single-Laser Flow Cytometer Double-strand DNA breaks or dysfunction of the mitotic spindle apparatus can lead to the formation of micronuclei (MN) in dividing cells. Enumeration of MN therefore serves as an index of numerical and structural chromosome damage. Erythrocytes are the cell population of choice for studying micronuclei in vivo. Erythrocyte precursors are continuously dividing and therefore at risk of forming MN. Furthermore, since the erythrocyte precursor extrudes its nucleus over the course of differentiation, MN are particularly evident in these otherwise DNA-deficient cells. This unit describes a procedure for fixing and staining rodent peripheral blood cells for flow cytometric enumeration of micronucleus-containing erythrocytes.
UNIT 7.21
BASIC PROTOCOL
In brief, cells are fixed with ultracold methanol, which also permeabilizes the membrane, making nucleic acids accessible to RNase and dye. Cells are then centrifuged to remove the fixative, and resuspended in a small volume with RNase and a FITC-labeled antibody against the CD71-defined antigen. After RNA degradation is complete, cells are resuspended with a solution containing propidium iodide (PI). This combination of reagents provides for the differential labeling and enumeration of four erythrocyte subpopulations: mature erythrocytes (normochromatic erythrocytes; NCEs), MN-containing mature erythrocytes (MN-NCEs), young erythrocytes (reticulocytes; RETs), and MN-containing young erythrocytes (MN-RETs). The protocol also describes the use of malaria-infected rodent erythrocytes, which closely mimic micronucleus-containing erythrocytes and which serve as biological standards that facilitate rational and consistent instrument calibration. Materials Mouse or Rat MicroFlowPLUS kit (Litron Laboratories) containing: Solution A (fixative; i.e., methanol) Solution B (anticoagulant) Solution C (washing/diluting buffer) Malaria Biostandard sample Experimental rodent(s) (i.e., mouse or rat) RNase/anti-CD71-FITC solution (see recipe) PI solution (see recipe) 15-ml screw-cap polypropylene tubes −70° to −85°C (preferred) chest freezer 2.5-ml microcentrifuge tubes Centrifuge (refrigerated preferred) with swinging-bucket rotor Polypropylene or polystyrene flow cytometry tubes Foil Flow cytometer: Excitation: 488 nm Filters: 530 ± 30-nm band-pass (green) and 650-nm long-pass (red) NOTE: All protocols using live animals must first be reviewed and approved by an Institutional Animal Care and Use Committee (IACUC) and must follow officially approved procedures for the care and use of laboratory animals. Nucleic Acid Analysis Contributed by Stephen Dertinger, Dorothea Torous, Nikki Hall, and Carol Tometsko Current Protocols in Cytometry (2003) 7.21.1-7.21.9 Copyright © 2003 by John Wiley & Sons, Inc.
7.21.1 Supplement 23
1. At least 1 day prior to collecting and fixing blood samples, label 15-ml screw-cap polypropylene or polystyrene tubes with rodent identifiers. Aliquot 2 ml Solution A (fixative) into each tube and store in a –70° to –85°C chest freezer. Collect blood samples 2. Label clean 2.5-ml microcentrifuge tubes with rodent identifiers. Aliquot 350 µl Solution B (anticoagulant) into each tube and store at 4°C until use. 3. Approximately 1 hr before collecting blood, transfer these tubes to room temperature, at which they will be maintained throughout the blood-collection procedure. 4. Collect ∼60 to 120 µl rodent (i.e., mouse or rat) peripheral blood into the appropriate tube containing Solution B using an IACUC-approved method (e.g., tail, vein, orbital sinus; Donavan and Brown, 1995). Cap the tube and invert several times to mix. Repeat for each animal in the study. The blood–Solution B mixture can be stored at room temperature for up to 4 hr before fixing (see below).
Fix blood samples 5. Using a micropipettor, retrieve 180 µl anticoagulated blood sample and remove the corresponding tube containing Solution A from the freezer (step 1). 6. Hold the tube of Solution A upright and position the pipet tip ∼1 cm above the surface of the fixative. Forcefully dispense the blood sample into Solution A, making sure that the pipet tip does not touch the side of the tube or the surface of the solution. 7. Recap the tube. Holding the top of the tube in one hand, strike the bottom of the tube sharply with the other several times. Alternatively, vortex briefly. 8. Immediately transfer the tube back to the freezer for storage. 9. Repeat steps 5 to 8 for the remaining samples. 10. Store samples in a –70° to –85°C freezer ≥24 hr. Blood cells fixed in Solution A must be maintained at –70° to –85°C for at least 24 hr before the fixative is washed out. Samples are stable in Solution A at least 1 year so long as this temperature has been maintained.
Wash cells 11. Incubate Solution C (washing/diluting buffer) on ice at least 45 min to stabilize the temperature. It is important that Solution C be ice cold when added to samples.
12. Remove up to three tubes of fixed blood cells from the ultracold freezer (step 10). Quickly place the capped tubes in a centrifuge rack and close the freezer. Tap each tube sharply three or four times to resuspend the cells, which settle over time in fixative at ultracold temperatures. Loosen the cap on each tube In addition to experimental samples, it is advisable to wash and stain one kit-supplied Malaria Biostandard sample each day of analysis. Note that this one tube will provide sufficient cells for instrument calibration procedures (steps 23 to 28).
13. Immediately add 12 ml ice-cold Solution C to each tube. Tighten caps and invert to mix. Immediately place the tubes on ice. Measurement of Cytogenetic Damage in Rodent Blood
7.21.2 Supplement 23
Current Protocols in Cytometry
14. Repeat steps 12 and 13 until all tubes containing experimental blood samples are prepared. Centrifuge 5 min at ∼600 × g, 4°C using a swinging-bucket rotor. When centrifugation is complete, quickly remove all tubes and immediately place on ice. A refrigerated centrifuge is preferred, but not required (i.e., the centrifugation can be performed at room temperature).
15. One tube at a time, decant the supernatant into a waste container, leaving ∼100 to 150 µl supernatant for resuspension. Recap the tube and immediately return it to ice. Continue until all supernatants have been decanted. 16. Remove one tube of washed cells from the ice. Quickly resuspend cells in the remaining supernatant by forcefully tapping the bottom of the tube. Put the tube back on ice and continue with the next tube, repeating until the series is complete. 17. After all pellets are resuspended, store tubes at 4°C or on ice until all samples are washed and ready for flow cytometric analysis. The washed samples are stable 3 days at 4°C, but it is best if they are analyzed the day they are prepared. Stained samples (see below) should be used the day they are prepared.
Label blood samples for flow cytometric analysis 18. Add 80 µl RNase/anti-CD71-FITC solution to labeled polypropylene or polystyrene flow cytometry tubes, and store covered with foil (to protect from light) at 4°C until addition of washed blood samples. 19. If tubes of washed cells are refrigerated, transfer them to ice. Tap the tubes to resuspend the cell pellets. 20. Add 20 µl washed blood sample directly into the solution in the corresponding flow cytometry tube. It is important that all of the blood come into contact with RNase/anti-CD71-FITC solution.
21. Transfer treated samples to a rack and incubate 30 min at 4°C (e.g., refrigerator) covered with foil. Return unused washed blood samples to 4°C for storage. 22. After incubation at 4°C, incubate tubes 30 min at room temperature (∼24°C) to ensure complete degradation of cellular RNA. Return all samples to 4°C until analysis. A considerable window of time (up to 6 hr) is available before the samples need to be analyzed. Stagger the addition of washed blood cells to RNase/anti-CD71-FITC solution if samples cannot be analyzed within this time.
Calibrate cytometer and collect data 23. Once the RNase/anti-CD71-FITC incubation is complete, gently tap the bottom of one Malaria Biostandard sample to loosen settled cells. Add 1 ml PI solution and immediately place sample on the flow cytometer. 24. Analyze the Malaria Biostandard to guide instrument settings, including PMT voltages and compensation values. Approximate PMT voltages by maximally resolving RETs from NCEs, as well as erythrocytes housing a single parasite from those containing no parasites (Fig. 7.21.1). This is the first step of the calibration sequence. Nucleated cells should be in the fourth decade for red fluorescence. They do not appear in Figure 7.21.1 because they have been gated out based on their high (2n) propidium iodide signal. Nucleic Acid Analysis
7.21.3 Current Protocols in Cytometry
Supplement 23
104
Anti-CD71-FITC
103
102
101
100 100
101
102
103
104
Propidium iodide Figure 7.21.1 A kit-supplied malaria biostandard sample is used to maximize the fluorescent resolution of RETs from NCEs, as well as erythrocytes with and without a micronucleus-like DNA content. Quadrant contents: lower left, NCEs; lower right, malaria-infected NCEs; upper left, RETs; upper right, malaria-infected RETs.
104
Anti-CD71-FITC
103
102
101
100 100
101
102
103
104
Propidium iodide Measurement of Cytogenetic Damage in Rodent Blood
Figure 7.21.2 A kit-supplied malaria biostandard sample is used to set appropriate compensation, which is evident when parasitized cells exhibit a vertical profile. Quadrant contents: lower left, NCEs; lower right, malaria-infected NCEs; upper left, RETs; upper right, malaria-infected RETs.
7.21.4 Supplement 23
Current Protocols in Cytometry
25. Adjust compensation to eliminate the FITC signal spilling over into the red fluorescence (UNIT 1.14). For this application, appropriate compensation is evident when cells containing one parasite exhibit consistent PI fluorescence (Fig. 7.21.2).
26. Fine-tune PMT voltage corresponding to PI fluorescence to place erythrocytes containing a single parasite into a consistent red-fluorescence channel. Adjusting parasitized cells to a consistent fluorescence channel enables investigators to use the same analysis regions for experimental blood samples, even when analysis is performed on different days. In addition to the malaria peak, it is also useful to consider the red-fluorescence peak channel for nucleated cells, which represent an internal standard and which can be monitored over the course of each analysis day to guard against instrument drift (Fig. 7.21.3).
27. After the PMT voltage is fine-tuned, re-evaluate whether the compensation is still appropriately set (step 26). 28. As demonstrated in Figure 7.21.2, use malaria-infected erythrocytes to guide the position and dimensions of quadrants or regions that are used to enumerate MN-NCE and MN-RET frequencies. If a quadrant is used to enumerate NCE, MN-NCE, RET, and MN-RET frequencies (as opposed to four regions), a gate must be used to eliminate nucleated cells based on their high (2n) propidium iodide signal. A typical stop-mode setting is the acquisition of 20,000 total RETs per sample (see Fig. 7.21.4).
50
40
Counts
parasitized cells 30
20 nucleated cells 10
0 100
101
102
103
104
Red fluorescence (PI) Figure 7.21.3 Malaria biostandard samples have a high prevalence of erythrocytes that contain micronucleus-like DNA content. PMT voltages can be carefully calibrated on each day of analysis by bringing erythrocytes which house a single parasite to a consistent propidium iodide–associated fluorescence channel.
Nucleic Acid Analysis
7.21.5 Current Protocols in Cytometry
Supplement 23
104
anti-CD71-FITC
103
102
101
100 100
101
102
103
104
Propidium iodide Figure 7.21.4 Representative blood sample from a genotoxicant-treated mouse. Quadrant contents: lower left, NCEs; lower right, micronucleated NCEs; upper left, RETs; upper right, micronucleated RETs.
29. Without changing PMT voltage or compensation settings determined with the kitsupplied Malaria Biostandard sample, add 1 ml PI solution to one experimental rodent sample (see step 24) and analyze on the flow cytometer. 30. Record the percentages, numbers, or abundances of NCEs, micronucleated NCEs, RETs, and micronucleated RETs. 31. Repeat steps 29-30 for each remaining sample. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
PI solution Combine 250 µl Solution F per 10 ml Solution C as provided in the MicroFlowPLUS kit (Litron Laboratories). Prepare sufficient solution for the number of experimental blood samples plus enough for one or two kit-supplied Malaria Biostandard blood samples (used for instrument setup), assuming 1 ml per sample. Store in the dark at 4°C until needed. Prepare fresh daily. CAUTION: Solution F (propidium iodide) is a DNA dye and should be handled using protective clothing and gloves.
RNase/anti-CD71-FITC solution Combine 10 µl each Solution D (RNase) and E (labeled antibody) per 1 ml Solution C as provided in the MicroFlowPLUS kit (Litron Laboratories). Prepare sufficient solution for the number of experimental blood samples and one or two kit-supplied Malaria Biostandard blood samples (used for instrument setup), assuming 80 µl per sample. Store in the dark at 4°C until needed. Prepare fresh daily. Measurement of Cytogenetic Damage in Rodent Blood
7.21.6 Supplement 23
Current Protocols in Cytometry
COMMENTARY Background Information The in vivo rodent micronucleus test is widely utilized to screen chemicals for genotoxic activity (Hayashi et al., 2000). Micronuclei (MN) arise when chromosome fragments or lagging whole chromosomes fail to be incorporated into daughter nuclei as cells divide. The endpoint is therefore sensitive to clastogens (e.g., cyclophosphamide), which cause doublestrand breaks in DNA, as well as to aneugens (e.g., vinblastine), which target the mitotic spindle apparatus. Traditionally, the test has been performed by staining peripheral blood or bone marrow smears and microscopically determining the frequency of MN in erythrocyte populations. Erythrocytes have become the most widely studied target population, since precursor cells are rapidly dividing (i.e., are at risk of forming MN), and MN that remain after extrusion of nuclei are readily observed with appropriate staining. Even so, it is well appreciated that the process of microscopically determining the frequency of micronucleated erythrocytes is tedious and time consuming, primarily because these events are rare. For example, the frequency of MN-RETs in most commonly studied mouse and rat strains is on the order of 0.1% to 0.3%. Clearly the highthroughput capabilities of flow cytometry provide potential for more objective and rapid analyses of these rare events. One challenge to successfully accomplishing these measurements by flow cytometry is the need to determine MN incidence separately for mature and young erythrocyte populations. The use of an immunological reagent (i.e., anti-CD71-FITC), as described in this unit, has helped make this possible (Dertinger et al., 1996). The CD71-defined antigen, also known as the transferrin receptor, is a membranebound glycoprotein that is expressed on most dividing cells and on the brain endothelium. The physiological function of CD71 is to bind Fe(apo)-transferrin, and through endocytosis, transfer iron into the cell. Since CD71 is lost from the cell surface of RETs as they mature into NCEs, fluorescent antibodies against this receptor are useful for differentially staining erythrocyte subtypes (Seligman et al., 1983; Serke and Huhn, 1992). This is critical for studies which involve rodents that are exposed to test articles in an acute or subacute treatment regimen, as exposures of short duration will not affect the frequency of MN in mature (i.e., pre-existing) erythrocytes. In these types of
studies, only the newly formed erythrocytes are at risk of forming MN in response to genotoxicant exposure. Therefore, only when mice are subchronically or chronically treated should the frequency of MN be considered in the mature (or total) erythrocyte population. Furthermore, the analysis of MN in young red blood cells is essential for all studies involving rats. Whereas genotoxicant-induced MNerythrocytes persist in the circulation of mice, they are efficiently removed by the spleen of most other mammalian species (including rat). Therefore, unless rats have been splenectomized, the frequency of MN-NCEs in this species should not be considered a sensitive indicator of genotoxic activity. Rather, analyses must be restricted to RETs (Wakata et al., 1998). In addition, there are some data to suggest that the youngest fraction of rat peripheralblood RETs may provide greater sensitivity to genotoxicants compared to total RETs (Hayashi et al., 1992; Abramsson-Zetterberg et al., 1999). A second challenge to successfully performing MN-erythrocyte measurements relates to the rarity of these events. For instance, the reticulocyte frequency is typically in the range of 1% to 5%, and MN are present in these cells at a frequency of ∼0.1% to 0.3%. These MNRETs, the double-positive population in bivariate graphs of green versus red fluorescence, are therefore extremely rare, and thus consistent optimization of PMT voltage and compensation settings is difficult. The use of malaria-infected erythrocytes has helped overcome this obstacle (Tometsko et al., 1993; Dertinger et al., 2000; Torous et al., 2001). Malaria endows the target cells of interest with a DNA content which is in the same range as that of MN; however, the incidence of parasitized cells can be very high. Together, the high frequency and uniform DNA content of infected erythrocytes enable investigators to carefully and appropriately configure instrumentation settings. These biological standards are included in commercially available kits (e.g., Mouse or Rat MicroFlowPLUS; Litron Laboratories).
Critical Parameters and Troubleshooting Having abundant experience with this protocol and having effectively taught the method to numerous other laboratories, the authors confidently characterize the fixing procedure described herein as both simple and reliable;
Nucleic Acid Analysis
7.21.7 Current Protocols in Cytometry
Supplement 23
however, it is important to emphasize that it is also exacting. Modifications to the ultracold methanol procedure are generally not tolerated well. The temperature of the fixative, which must be between –70° and about –85°C, is critical. These temperatures must be maintained as long as the samples are in fixative. Since maintenance of temperature is so important, chest freezers, which generally hold temperature better than upright freezers, are preferred. Some groups prefer to initially fix samples on dry ice rather than work out of freezers. This should be done with extreme caution. As with unacceptably high temperatures, exposure of fixative or fixed blood suspensions to CO2 vapors from subliming dry ice leads to poor sample quality. In either case, the most obvious feature of these preparations is unacceptably high amounts of cellular aggregation (evident as high FS signals). Therefore, blood samples should be added to tubes of fixative that are on dry ice only if a styrofoam shielding system is in place. An effective shield allows tubes to fit snugly through the styrofoam into dry ice flakes or pellets. The container holding the dry ice should be vented near the bottom to allow dry ice vapors to escape and fall away from the tubes of fixative. Like the fixing protocol, the labeling/staining procedure is relatively simple. RNase activity and antibody labeling occur simultaneously and in a small volume. This allows for a “washless” labeling procedure. That is, the propidium iodide solution is simply added to antibody-labeled cells and the suspensions are analyzed. This is an effective system which minimizes cell handling. Even with effective staining, it is challenging to set optimal PMT voltages and compensation and to determine appropriate dimensions/locations of regions or quadrants to enumerate MN erythrocytes. MN are rare and heterogeneous in DNA content. Staining of fixed malaria-infected erythrocytes in parallel with experimental rodent samples is therefore considered a critical component of this protocol. The PI signal of parasitized cells provides homogeneous and prevalent MN-like erythrocytes that are extremely useful for careful control of instrument settings. This reagent therefore greatly aids careful and consistent configuration of instrumentation parameters between experiments.
infected blood sample. The double-positive population represents malaria parasites in a CD71-expressing fraction of erythrocytes (that is, malaria-containing reticulocytes). The resolution of RETs and malaria-infected RETs is maximized when green-from-red compensation is appropriately set (UNIT 1.14). Appropriate compensation is evident when RETs exhibit red fluorescence similar to NCEs which are at the origin, and when malaria-infected erythrocytes exhibit a vertical profile (see Fig. 7.21.2). Overcompensation is apparent when the doublepositive cells arch back towards the y axis. Once PMT voltages, compensation, and region or quadrant locations have been determined with the malaria biostandard, experimental samples are analyzed without further adjustment. It is best to have negative and positive control samples to analyze in conjunction with experimental samples. While samples derived from untreated or vehicle-control animals should provide MN-RET measurements typically in the range of ∼0.1% to 0.3%, positive controls can run in excess of 3%. Genotoxicants that often serve as positive controls are methyl methanesulfonate and cyclophosphamide. Figure 7.21.4 illustrates typical distribution of cells from genotoxicant-treated mice. Note that when rat samples are analyzed, very low values for MN-NCEs will be observed. This is due to splenic removal and highlights the fact that enumeration of MN erythrocytes in the peripheral blood compartment of rats must focus on RETs or even the youngest fraction of RETs in order to provide a sensitive indication of genotoxicity (Hayashi et al., 1992; Abramsson-Zetterberg et al., 1999). Exposure of rodents to test articles, with or without genotoxic activity, can result in reduced erythropoiesis function. This form of bone marrow toxicity is evident as a reduction in the frequency of CD71-positive erythrocytes (i.e., %RET). Some guidelines (e.g., OECD, 1997) suggest that the top concentration for in vivo micronucleus assays should not cause greater than 80% reduction in RET frequency. Therefore, %RET values should be measured and can be useful for determining appropriate concentration ranges in preliminary studies and for monitoring bone marrow toxicity in definitive assays.
Time Considerations Measurement of Cytogenetic Damage in Rodent Blood
Anticipated Results Figure 7.21.2 shows the typical distribution of four erythrocyte populations from a malaria-
Fixation of as many as 50 blood samples can be accomplished in 2 hr. It is important to monitor freezer temperature during this process
7.21.8 Supplement 23
Current Protocols in Cytometry
and to give it time to return to baseline if necessary. Cells should remain at least 24 hr in –70° to –85°C fixative. Preparation for labeling/staining requires ∼1 hr, and staining itself requires ∼1 hr. Blood samples are typically analyzed in ∼3 min (given a stop mode of 20,000 RETs per sample). This translates to about 50 blood samples in 5 hr. Note that treatment conditions that cause appreciable bone-marrow toxicity (reduced RET frequency) can significantly affect this figure.
Literature Cited Abramsson-Zetterberg, L., Grawe, J., and Zetterberg, G. 1999. The micronucleus test in rat erythrocytes from bone marrow, spleen and peripheral blood: The response to low doses of ionizing radiation, cyclophosphamide and vincristine determined by flow cytometry. Mutat. Res. 423:113-124. Dertinger, S.D., Torous, D.K., and Tometsko, K.R. 1996. Simple and reliable enumeration of micronucleated reticulocytes with a single-laser flow cytometer. Mutat. Res. 371:283-292. Dertinger, S.D., Torous, D.K., Hall, N., Tometsko, C.R., and Gasiewicz, T.A. 2000. Malaria-infected erythrocytes serve as biological standards to ensure reliable and consistent scoring of micronucleated erythrocytes by flow cytometry. Mutat. Res. 464:195-200. Donovan, J. and Brown, P. 1995. Blood collection. In Current Protocols in Immunology (J.E. Coligan, A.M. Kruisbeek, D.H. Margulies, and W. Strober, eds.) pp. 1.7.1-1.7.8. John Wiley & Sons, New York. Hayashi, M., Kodama, Y., Awogi, T., Suzuki, T., Asita, A.O., and Sofuni, T. 1992. The micronucleus assay using peripheral blood reticulocytes from mitomycin C- and cyclophosphamidetreated rats. Mutat. Res. 278:209-213. Hayashi, M., MacGregor, J.T., Gatehouse, D.G., Adler, I.-D., Blakey, D.H., Dertinger, S.D., Krishna, G., Morita, T., Russo, A., and Sutou, S. 2000. In vivo rodent erythrocyte micronucleus assay: Aspects of protocol design including repeated treatments, integration with toxicity testing, and automated scoring. A report from the International Workshop on Genotoxicity Test Procedures (IWGTP). Environ. Molec. Mutagen. 35:234-252.
OECD, 1997. Mammalian erythrocyte micronucleus test. In OECD Guidelines for the Testing of Chemicals, Section 4, Guideline 474. Organisation for Economic Cooperation and Development (OEDC), Paris. Seligman, P., Allen, R., Kirchanski, S., and Natale, P. 1983. Automated analysis of reticulocytes using fluorescent staining with both acridine orange and an immunofluorescence technique. Am. J. Hematol. 14:57-66. Serke, S. and Huh, D. 1992. Identification of CD71 (transferrin receptor) expressing erythrocytes by mutiparameter-flow-cytometry (MP-FCM): Correlation to the quantitation of reticulocytes as determined by conventional microscopy and by MP-FCM using an RNA staining dye. Br. J. Haematol. 81:432-439. Tometsko, A.M., Torous, D.K., and Dertinger, S.D. 1993. Analysis of micronucleated cells by flow cytometry. 1. Achieving high resolution with a malaria model. Mutat. Res. 292:129-135. Torous, D.K., Hall, N.E., Dertinger, S.D., Diehl, M.S., Illi-Love, A.H., Cederbrant, K., Sandelin, K., Bolcsfoldi, G., Ferguson, L.R., Pearson, A., Majeska, J.B., Tarca, J.P., Hewish, D.R., Doughty, L., Fenech, M., Weaver, J.L., Broud, D .D., G ate h ou se, D.G., Hy nes, G.M., Kwanyuen, P., McLean, J., McNamee, J.P., Parenteau, M., Van Hoof, V., Vanparys, P., Lenarczyk, M., Siennicka, J., Litwinska, B., Slowikowska, M.G., Harbach, P.R., Johnson, C.W., Zhao, S., Aaron, C.S., Lynch, A.M., Marshall, I.C., Rodgers, B., and Tometsko, C.R. 2001. Flow cytometric enumeration of micronucleated reticulocytes: High transferability among 14 laboratories. Environ. Molec. Mutagen. 38:5968. Wakata, A., Miyamae, Y., Sato, S., Suzuki, T., Morita, T., Asano, N., Awogi, T., Kondo, K., and Hayashi, M. 1998. Evaluation of the rat micronucleus test with bone marrow and peripheral blood: Summary of the 9th collaborative study by CSGMT/JEMS-MMS. Environ. Molec. Mutagen. 32:84-100.
Contributed by Stephen Dertinger, Dorothea Torous, Nikki Hall, and Carol Tometsko Litron Laboratories Rochester, New York
Nucleic Acid Analysis
7.21.9 Current Protocols in Cytometry
Supplement 23
Analysis of Tissue Imprints by Scanning Laser Cytometry
UNIT 7.22
The diagnosis of solid tumors is based primarily on the histological analysis of formaldehyde-fixed and paraffin-embedded material. The importance of determining the biological aspects of tumors is, however, growing for the clinical practice. This also necessitates quantitative measurements obtained by well-established cytometric methodology. A major advantage of slide-based cytometry (e.g., scanning laser cytometry; UNIT 2.10) in contrast to flow cytometry is the opportunity to morphologically evaluate cell populations according to immunophenotypic characteristics following microscopic repositioning. The use of tissue imprints for this purpose significantly simplifies measurements in solid-tumor cells. The immunolabeling strategies described here are intended for the fast analysis of total nuclear DNA content and other functional aspects of tumor cells in cytological specimens using a laser scanning cytometer (LSC). The Basic Protocol describes the rapid assessment of tumor cell ploidy in any type of solid tumor using CD45-positive (CD45+) tissue leukocytes as internal reference cells with normal (2N) DNA content. An Alternate Protocol is provided for the quantitative measurement of nuclear antigen expression, such as estrogen and progesterone receptors, as well as the proliferation-related MIB1 antigen. The methods are simple and rapid and are designed to perform with success despite the variation in quality of routine surgical material. Therefore, two-color analyses of DNA content (propidium iodide; PI) and an immunological marker (fluorescein isothiocyanate; FITC) are preferred. To keep the number of slides as low as possible, multiple areas within the same slide can be designated with a grease pencil and processed with different primary antibodies. Neighboring areas on one slide can then be measured separately in the LSC. The DNA content distribution, which is the same in all the measurements, serves to identify the respective cell populations in tumors with aneuploidy. The methods rely on the analysis of tissue imprints (Support Protocols 1 and 2), but the labeling can be carried out on other freshly prepared cytological material, including smears obtained after scratching the tissue surface or following cell isolation procedures (UNIT 5.2). In this case, isolated cells from solid tumors should be deposited on slides by cytocentrifugation (2000 rpm, 7 min). To obtain large areas (≤240 mm2), a Hettich 1640 cytocentrifuge (Andreas Hettich) or similar instrument should be used. DIRECT IMMUNOFLUORESCENCE LABELING AND DNA STAINING OF TISSUE IMPRINTS FOR ANALYSIS BY SCANNING LASER CYTOMETRY
BASIC PROTOCOL
Direct immunofluorescence can conveniently be used for the analysis of highly expressed surface antigen epitopes. Although this protocol describes the use of a CD45 antibody, a similar antibody (and the appropriate isotype control) can be used. Materials Microscope slide containing tumor tissue imprint (see Support Protocols 1 and 2), smear, or cytospin preparation (UNIT 5.2) Acetone, ice cold (−20°C) 4% (w/v) BSA in PBS (APPENDIX 2A) or commercial blocking solution (e.g., Protein Block; BioGenex Laboratories) FITC-conjugated mouse CD45 antibody (Dako), diluted 1:10 with PBS FITC-conjugated mouse IgG1 isotype antibody (Dako), diluted 1:10 with PBS Contributed by Reinhard Bollmann and Gábor Méhes Current Protocols in Cytometry (2003) 7.22.1-7.22.6 Copyright © 2003 by John Wiley & Sons, Inc.
Nucleic Acid Analysis
7.22.1 Supplement 26
PI/RNase solution (see recipe) Glycerol Diamond pencil Humidified chamber: ∼15 × 30–cm stainless steel or plastic container with lid (for up to ten slides), containing one or two paper towels moistened with distilled water 37°C incubator Coverslips (10 × 10 mm or cut to fit area containing cells) Laser scanning cytometer (LSC; e.g., CompuCyte) equipped with argon (Ar) and helium/neon (HeNe) lasers; standard filter settings for the measurement of FITC and PI fluorescence Prepare specimen slide 1. With a diamond pencil mark areas for immunofluorescence and isotype control on the back of a microscope slide containing a tumor tissue imprint, smear, or cytospin preparation (e.g., capital letters or numbers). Separate areas will be needed for CD45 determination and an isotype control.
2. Fix slide 2 min in ice-cold acetone. Remove slide from acetone and air dry. 3. Apply 4% BSA or commercial blocking solution to cover the slide and incubate 15 min at 37°C in a humidified chamber. The blocking solution prevents nonspecific antibody binding. Contact between the moist paper towels of the humidified chamber and the fluid on the slide surface should be avoided.
Immunolabel 4. Remove blocking solution and add 50 µl diluted FITC-conjugated mouse CD45 antibody to the marked area of the slide. Add 50 µl FITC-conjugated mouse IgG1 isotype antibody to the appropriate area on the same slide. Incubate 30 min at 37°C in the humidified chamber. This volume adequately covers a 1.5 × 1.5–cm area. The FITC-conjugated mouse IgG1 isotype antibody serves as an isotype control antibody. After immunofluorescence staining, slides should be protected from light.
Stain for DNA 5. Remove antibody solutions and immediately add 500 µl PI/RNase solution to cover the slide. Incubate 30 min at 37°C in the dark. 6. Remove staining solution and cover both areas with glycerol and a coverslip for immediate measurement in an LSC. Slides can also be stored up to 24 hr at 4°C in the dark until they are analyzed.
Analyze fluorescence 7. Put the slide on the stage of the LSC. While viewing the area of the isotype control, focus on the PI-stained cells in the microscope. The nuclei of PI-stained cells should appear red.
8. Open the LSC files display setting. 9. Scan and set sensors to bring the FITC-related fluorescence into the negative gate as well as PI-related fluorescence intensity into a reasonable range. Stop scanning. Analysis of Tissue Imprints by Scanning Laser Cytometry
10. Go to the area stained with FITC-labeled anti-CD45. Scan at least 2000 events.
7.22.2 Supplement 26
Current Protocols in Cytometry
INDIRECT IMMUNOFLUORESCENCE LABELING AND DNA STAINING OF TISSUE IMPRINTS FOR ANALYSIS BY SCANNING LASER CYTOMETRY Indirect immunofluorescence labeling is somewhat more complex than the direct detection method. However, in instances of weakly expressed surface epitopes, as well as for intracellular (i.e., cytoplasmic or nuclear) antigens, indirect immunofluorescence is often preferred.
ALTERNATE PROTOCOL
Additional Materials (also see Basic Protocol) Unconjugated mouse antibody, such as: NCL-ER6F11 (Novocastra Labs) for estrogen receptor, diluted 1:10 in PBS (APPENDIX 2A) NCL-PGR (Novocastra Labs) for progesterone receptor, diluted 1:10 in PBS anti-Ki-67 (Linaris) or NCL-Ki67-MM1 (Novocastra Labs) for proliferating cells, diluted 1:20 in PBS Unconjugated isotype control IgG (e.g., Dako), diluted as for antibody of interest PBS (APPENDIX 2A) Biotinylated anti-mouse antibody (e.g., BioGenex Laboratories), diluted in PBS according to manufacturer’s instructions FITC-conjugated streptavidin (e.g., Dako), diluted in PBS according to manufacturer’s instructions 1. Prepare specimen slide as described (see Basic Protocol, steps 1 to 3). 2. Add 50 µl diluted unconjugated mouse antibody to the appropriate slide area. Add 50 µl diluted unconjugated isotype control IgG to the appropriate area. Incubate 1 hr at 37°C in a humidified chamber. 3. Wash two times in PBS, 5 min each. 4. Add 50 µl diluted biotinylated anti-mouse antibody. Incubate 30 min at 37°C. The authors have found that a 1:40 dilution of the BioGenex antibody works well. An optimal dilution for reagents from different suppliers may need to be determined empirically.
5. Wash two times in PBS, 5 min each. 6. Add 50 µl diluted FITC-conjugated streptavidin. Incubate 30 min at 37°C. 7. Stain for DNA and analyze fluorescence as described (see Basic Protocol, steps 5 to 10). PREPARATION OF A TISSUE IMPRINT In contrast to hematological samples, the most critical complication in the analysis of solid-tumor cells occurs because cells tightly adhere to each other and to the extracellular matrix. Therefore, analysis of whole single cells requires an isolation technique in that the cells of interest must be taken from the tissue and presented one by one for measurement. This can be done for unfixed tumor tissues by direct tissue imprint. Materials Freshly prepared solid tumor sample (prevent tissue from drying out) Positively charged microscope slides Lint-free tissues Scalpel, sharp razor blade, previously used microtome blade, or other dissection instrument Filter paper Grease pencil (e.g., ImmEdge; Vector Laboratories)
SUPPORT PROTOCOL 1
Nucleic Acid Analysis
7.22.3 Current Protocols in Cytometry
Supplement 26
1. Clean three to five positively charged microscope slides with a lint-free tissue to remove dust particles. Alternatively, ready-to-use positively charged slides coated with poly-L-lysine can be purchased (e.g., Polysciences).
2. Using a scalpel or other dissection instrument, remove any nontumor components (e.g., fat tissue, vessels) from a solid tumor sample. 3. Cut a piece of tumor with a fresh surface of no more than 8 to 10 mm in diameter. 4. Blot excess tissue fluid from the margin of the cut surface using a filter paper. 5. Touch the cut surface to the slide. Touch three to four times at regular distances to obtain imprint fields for multiple reactions within the same slide (see Basic Protocol, step 1). 6. Dry the preparation at room temperature in the dark and use a grease pencil to delineate individual fields. For optimal results, slides should be stained within 2 hr of their preparation. Slides can be stored several days in plastic containers at room temperature or up to several months at −20°C. SUPPORT PROTOCOL 2
PREPARATION OF A TISSUE IMPRINT USING A CYTOLOGICAL DEVICE The CerviSoft cell preparation technique has a substantial advantage over touch preparation in that cell density is better controlled. The formation of large clusters, which are not recognized by a laser scanning cytometer (LSC), is significantly reduced (Bollmann et al., 2002). This protocol requires a CerviSoft Foam device (Puritan Medical Products); for other needed materials, see Support Protocol 1. 1. Prepare microscope slides and a solid tumor sample as described (see Support Protocol 1, steps 1 and 2). 2. Cut a fresh tumor surface as large as possible (10 to 20 mm in diameter). 3. Roll the sponge part of a CerviSoft device over the freshly cut surface in one direction. For details on using and maintaining the CerviSoft device, see manufacturer’s instructions and Bollmann et al. (2002).
4. Transfer the attached cells to slides by rolling the device in the opposite direction. This results in an ∼15 × 30–mm field that is evenly covered with cells. This area is large enough for at least two measurements: one for immunofluorescence analysis and one for the isotype control. With one pass of the device over the tumor sample, two or three slides can be prepared before the device is discarded.
5. Dry slides and use a grease pencil to delineate individual fields. For optimal results, slides should be stained within 2 hr of their preparation. Slides can be stored several days in plastic containers at room temperature or up to several months at −20°C.
Analysis of Tissue Imprints by Scanning Laser Cytometry
7.22.4 Supplement 26
Current Protocols in Cytometry
B 75
1.16
Count
60 45 1.00 30 15 0
2.32
Green fluorescence integral
A
C 108
104
4.0 Orange fluorescence integral
4.0 Orange fluorescence integral
Figure 7.22.1 Determination of DNA ploidy and cell proliferation in tumor imprints using a laser scanning cytometer (LSC). (A) DNA histogram of a breast carcinoma indicating DNA aneuploidy. The DNA index (DI) is provided for each histogram peak. (B) Precise determination of tumor cell DNA content in any type of tumor can be performed by the demonstration of CD45+ tissue leukocytes as a reference normal cell population (upper-left; DI = 1.00). Tumor cells with slightly elevated DNA content dominate the CD45− fraction (lower-right oval gate; DI = 1.16). (C) Repositioning of gated cells confirms the LSC finding morphologically. Cell nuclei of lymphocytes (upper panel) and cancer cells (lower panel), as visualized by propidium iodide fluorescence staining, are shown. When viewed under the microscope, these nuclei are red.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
PI (propidium iodide)/RNase solution Dissolve 1 mg propidium iodide (Sigma) and 4 mg RNase A (Sigma) in 20 ml PBS, pH 7.0 (APPENDIX 2A). Store up to several weeks in a dark container at 4°C. COMMENTARY Background Information Scanning laser cytometry of tissue imprints enables the study of functional aspects of solid tumors at the protein level (Kamentsky and Kamentsky, 1991; Gorczyca et al., 1997). Because such information is becoming increasingly important in the clinical management of breast cancer, the methods presented here focus on this tumor type. DNA aneuploidy and high tumor growth fractions are independent, unfavorable prognostic markers in breast cancer (Bergers et al., 1997; Fitzgibbons et al., 2000). The demonstration of nuclear steroid receptor expression is indicative of an adjuvant anti-estrogen therapy (Nicholson et al., 1995). Using these protocols, both steroid receptor positivity and the proliferation fraction can be quantitatively measured from the same tissue piece that is selected for histological analysis following formaldehyde fixation and paraffin embedding. By the time the histological analysis is carried
out, the functional parameters as measured by the LSC are already known.
Critical Parameters and Troubleshooting The number of cells on the slide depends on the cellularity and composition of the surgical sample. In the case of excessively high cell density or in the presence of large cell groups, however, the measurement will ignore the cell aggregates because the LSC identifies single cells by excluding objects with a defined, large contour area. The high number of incubation and detection steps may negatively influence the quality of the DNA histogram. Therefore, the analysis can be better interpreted following a DNA content measurement using CD45+ cells as a reference population. Nucleic Acid Analysis
7.22.5 Current Protocols in Cytometry
Supplement 26
Anticipated Results All tumors analyzed to date contained a CD45+ fraction, which helps to define the presence of an aneuploid cell population. The practical use of this strategy is presented in Figure 7.22.1.
Time Considerations As all solutions can be stored for long periods of time, the procedures can be started immediately following tissue preparation. Information on ploidy and tumor cell content can be obtained within 90 min. Indirect immunofluorescence results are provided in 3 hr. Preparing the tissue imprints takes ∼5 min for either method.
Literature Cited Bergers, E., Baak, J.P.A., van Diest, P.J., Willig, A.J.P., Los, J., Peterse, J.L., Ruitenberg, H.M., Schapers, R.F.M., Somsen, J.G., van Beek, M.W.P.M., Bellot, S.M., Fijnheer, J., and van Gorp, L.H.M. 1997. Prognostic value of DNA ploidy using flow cytometry in 1301 breast cancer patients: Result of the prospective multicenter morphometric mammary carcinoma project. Mod. Pathol. 10:762-768. Bollmann, R., Torka, R., Schmitz, J., Bollmann, M., and Méhes, G. 2002. Determination of ploidy and steroid receptor status in breast cancer by laser scanning cytometry. Cytometry 50:201215.
Fitzgibbons, P.L., Page, D.L., Weaver, D., Thor, A.D., Allred, C.D., Clark, G.M., Ruby, S.G., O’Malley, F., Simpson, J.F., Connolly, J.L., Hayes, D.F., Edge, S.B., Lichter, A., and Schnitt, S.J. 2000. Prognostic factors in breast cancer. College of American Pathologists Consensus Statement. Arch. Pathol. Lab. Med. 124:966978. Gorczyca, W., Darzynkiewicz, Z., and Melamed, M.R. 1997. Laser scanning cytometry in the pathology of solid tumors. Acta Cytol. 41:98108. Kamentsky, L.A. and Kamentsky, L.D. 1991. Microscope based multiparameter laser scanning cytometer yielding data comparable to flow cytometry data. Cytometry 12:381-387. Nicholson, R.I., McClelland, R.A., and Gee, J.M. 1995. Steroid hormone receptors and their clinical significance in cancer. Am. J. Clin. Pathol. 48:890-895.
Contributed by Reinhard Bollmann Institute of Pathology Bonn-Duisdorf Bonn, Germany Gábor Méhes University of Pécs Pécs, Hungary
Analysis of Tissue Imprints by Scanning Laser Cytometry
7.22.6 Supplement 26
Current Protocols in Cytometry
Cell Cycle Analysis of Budding Yeast Using SYTOX Green Flow cytometric analysis of DNA content is an invaluable tool for the study of cell cycle regulation in a number of organisms. Cells can be assigned to specific cell cycle phases based on DNA content, allowing researchers to generate a profile of the cell cycle distribution within a population. These profiles can then be used to monitor alterations in cell cycle distributions or cell cycle kinetics in response to a number of experimental conditions.
UNIT 7.23
BASIC PROTOCOL
The standard flow cytometric approach is to fix cells with ethanol, remove RNA with RNase, and then stain DNA with propidium iodide (PI; UNIT 7.5). The method is simple, rapid, and reproducible, and the stability of ethanol-fixed cells allows researchers the flexibility to fix, stain, and analyze cells on different days. This method works well for mammalian cells and has been adapted for use with budding yeast (UNIT 11.10; Hutter and Eipel, 1979; Haase and Lew, 1997). Relatively large coefficients of variation (CVs) are observed in PI-stained yeast cells, however, hindering the ability to accurately assess cell cycle position. The author has also found that carbon source–shift protocols routinely used in yeast cell cycle experiments can significantly alter the fluorescence of yeast cells stained with PI (Haase and Reed, 2002). Additionally, fluorescence of PI-stained cells is extremely sensitive to variances in either stain or cell concentration, which can contribute to variability between samples (Haase and Reed, 2002). An existing protocol has been modified by the substitution of the DNA-binding dye SYTOX Green (Molecular Probes) for PI. The use of SYTOX Green instead of PI provides a significant improvement in accuracy and reproducibility of results under a number of experimental conditions. Relative to PI, SYTOX Green exhibits a substantially greater induction of fluorescence when bound to DNA and a higher quantum yield. The fluorescence of SYTOX-stained cells also appears to be far less sensitive to dye dilution at working concentrations. Thus the specific fluorescence and chemical properties of SYTOX Green probably account for the significant improvement in the accurate determination of DNA content in yeast. This protocol assumes that the researcher possesses a working knowledge of yeast cell culture (Guthrie and Fink, 1991) and flow cytometry (UNIT 11.10 and see Chapter 1). Materials ∼1 × 107 yeast cells grown in suspension 95% (v/v) ethanol RNase solution (see recipe) Protease solution (see recipe) 50 mM Tris⋅Cl, pH 8.0 (APPENDIX 2A) 1 µM SYTOX Green staining solution (see recipe) 15-ml screw-cap centrifuge tubes 1.5-ml microcentrifuge tubes 37°C incubator or water bath 12 × 75–mm tubes suitable for flow cytometer Sonicator Flow cytometer with 488-nm excitation and a band-pass filter appropriate for collecting fluorescence emission at 523 nm Nucleic Acid Analysis Contributed by Steven B. Haase Current Protocols in Cytometry (2003) 7.23.1-7.23.4 Copyright © 2003 by John Wiley & Sons, Inc.
7.23.1 Supplement 26
Fix cells with ethanol 1. Harvest ∼1 × 107 yeast cells by centrifuging 5 min at 900 × g, room temperature, in a 15-ml screw-cap centrifuge tube. The actual starting cell number is not critical; 1 × 106 cells is adequate when sample size is limited. However, cell dilutions into the SYTOX staining buffer should be adjusted accordingly at the end of the protocol (step 10) to achieve satisfactory flow rates. All subsequent centrifugations are also carried out at room temperature.
2. Resuspend cell pellet in 1.5 ml water and slowly add 3.5 ml of 95% ethanol. Mix by gently inverting tube several times. Incubate ≥1 hr at room temperature or several hours at 4°C. Cells that require additional osmotic support are often grown in medium containing 1 M sorbitol. In this case, cells should be resuspended in 1 M sorbitol instead of in water. The author routinely fixes cells overnight at 4°C. Cells can be stored in the ethanol fixation solution for weeks at 4°C without ill effects.
Wash cells and remove RNA 3. Collect fixed cells by centrifuging 15 sec at 10,000 × g, pour off ethanol fixative, and then resuspend cell pellet in 1 ml water. 4. Transfer cell suspension to a 1.5-ml microcentrifuge tube and collect cells by brief centrifugation at 10,000 × g. Pour off supernatant. 5. Resuspend cell pellet in 0.5 ml RNase solution and incubate ≥2 hr at 37°C. The RNase step is critical; fresh RNase solution must be made each day. Cells can be left in RNase for several hours, even overnight, with no ill effects.
Treat cells with protease 6. Collect cells from the RNase solution by centrifuging 15 sec at 10,000 × g. Pour off supernatant. 7. Resuspend cell pellet in 0.2 ml protease solution and incubate 20 to 45 min at 37°C. Unlike the RNase step, results are severely compromised if cells are left in protease solution too long. Incubation times in protease solution should not exceed 60 min.
Stain and analyze cells 8. Collect cells from the protease solution by brief centrifugation at 10,000 × g. Pour off supernatant. 9. Resuspend cell pellet in 0.5 ml of 50 mM Tris⋅Cl, pH 8.0. Cells can be stored up to several days in Tris buffer at 4°C.
10. Transfer 50 µl cell suspension to a 12 × 75–mm tube containing 1 ml of 1 µM SYTOX Green staining solution. Break up cell clumps by sonicating briefly on low power. 11. Analyze cells using a flow cytometer with 488-nm excitation. Collect green fluorescence emission at 523 nm. Set the flow rate on the lowest possible setting and gate out debris by setting trigger on green fluorescence. If cells with less than a G1 DNA content are to be analyzed, gate debris by setting the trigger on forward scatter. Cell Cycle Analysis of Budding Yeast Using SYTOX Green
7.23.2 Supplement 26
Current Protocols in Cytometry
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Protease solution, 5 mg/ml Dissolve 50 mg pepsin (Sigma) in 10 ml water. Add 45 µl concentrated HCl. Make solution fresh each day and discard unused portion. RNase solution, 2 mg/ml Dissolve 20 mg DNase-free RNase A (Sigma) in 10 ml of 50 mM Tris⋅Cl, pH 8.0 (APPENDIX 2A)/15 mM NaCl. Boil solution 15 min and allow to cool slowly to room temperature. Make solution fresh each day and discard unused portion. SYTOX Green staining solution, 1 ìM For stock solution: Dilute 5 mM SYTOX Green (Molecular Probes), supplied in dimethyl sulfoxide (DMSO), to 1 mM (1000×) in DMSO. Divide stock solution into 50- to 100-µl aliquots and store in the dark at –20°C. The stock solution is divided into aliquots to minimize freeze-thawing, and can be stored at −20°C for at least 1 year.
For working (1 ìM) solution: Dilute a 1 mM (1000×) stock solution to 1 µM in 50 mM Tris⋅Cl, pH 8.0 (APPENDIX 2A). Prepare fresh. COMMENTARY Background Information Because of the small genome size and the irregular cell shape of budding yeast cells, accurate determination of DNA content by flow cytometry has proven uniquely challenging. Yeast researchers now have at least three different methods for assaying DNA content using flow cytometry. UNIT 11.10 outlines a standard protocol employing propidium iodide (PI) as the fluorochrome. PI has been widely used, primarily for historical reasons. As mentioned earlier, the relatively large coefficients of variation (CVs) exhibited by PI-stained cells limits the accuracy of the analysis, and reproducibility under certain experimental conditions is a concern. More recently, two additional methods have been developed to overcome the problems associated with PI-stained cells. Both methods report significantly improved CVs, allowing researchers to assess subtle changes in cell cycle distributions more accurately. The method reported in this unit simply substitutes SYTOX Green for PI in a standard ethanol-fixation protocol. The author has made direct comparisons of the SYTOX and PI protocols and found that the SYTOX protocol yields better CVs, improved linearity, and less sample-tosample variability across a wide range of experimental conditions (Haase and Reed, 2002). Two other SYTOX stains, SYTOX Orange and SYTOX Blue, are commercially available
through Molecular Probes. These dyes have properties similar to those of SYTOX Green, but they emit at different wavelengths. The author has used SYTOX Orange and achieved results similar to those observed with SYTOX Green, but has not yet tested SYTOX Blue. These dyes may prove useful in the context of double-labeling experiments in order to avoid fluorescence overlap with other fluorochromes. A third method, presented in UNIT 11.13, utilizes the DNA-binding dye SYBR Green I (Molecular Probes). Like SYTOX Green, SYBR Green I exhibits a large induction of fluorescence upon binding to DNA and a high quantum yield. It is likely that the superior fluorescence properties of SYBR Green I account for the remarkable improvement of CVs observed in SYBR-stained cells as compared with PIstained cells. The SYBR Green method is significantly more laborious than either the PI or the SYTOX protocol, and great care must be taken to maintain a strict relationship between stain concentration and cell number to achieve good results. The sensitivity of cells stained with SYBR Green to carbon-shift protocols has not been reported. This method may have the greatest utility when very accurate measurements are required and increased labor is not prohibitive. Nucleic Acid Analysis
7.23.3 Current Protocols in Cytometry
Supplement 26
Critical Parameters and Troubleshooting The major obstacle to the accurate cytometric determination of DNA content in budding yeast is large CVs. Large CVs severely limit the ability to resolve cells into distinct cell cycle phases and are not well tolerated by some popular cell cycle analysis programs. Several problems can contribute to large CVs in the SYTOX Green method. Unlike the method utilizing SYBR Green I (UNIT 11.13), treatment with RNase is essential for optimal results when staining with SYTOX Green. Although RNase A is normally considered a very robust enzyme, the author has found that freshly prepared RNase solution is critical. The RNase step is hard to overdo, and good results can be obtained with cells that have been left in RNase solution overnight. Unlike the RNase step, care must be taken to prevent prolonged incubations in the protease solution. Cells incubated for too long in protease exhibit poor CVs. Observations suggest that the protease step facilitates binding of SYTOX Green to DNA, perhaps by partially digesting chromatin complexes and partially denaturing DNA. The author has attempted to eliminate the protease step from the protocol, but DNA staining appears to be less efficient, and the reduction in fluorescence is likely to have a negative impact on the CVs.
Data Acquisition SYTOX-stained yeast cells, like PI-stained yeast cells, are especially sensitive to fluctuations in sample pressure during data acquisition. Fluctuations in sample pressure during analysis can lead to the shifting of fluorescence peaks and may contribute to large CVs. Variances in sample pressure can be identified by monitoring the cell flow rates using a cell counter. Although flow rates normally vary somewhat during collection, large variations in flow rate usually indicate a problem. If flow rates vary, inspect the sample tube for cracks and make sure that the tube is seated properly
on the gasket at the intake nozzle. Improperly sealed sheath-fluid reservoirs or other defects in cytometer fluidics can also contribute to variations in sample pressure. Also, after changing samples wait several seconds to allow flow rates to stabilize before collecting data.
Anticipated Results The SYTOX protocol should reproducibly yield fluorescence histograms with half peak CVs of approximately 6. The ratio of mean fluorescence values for the G2 peak as compared with the G1 peak should be ≥1.9. Variations from these norms may reflect problems with the procedure (see Critical Parameters and Troubleshooting). However, the author has observed that particular mutant strains may also exhibit reproducible variances in CVs or fluorescence linearity that appear to be associated with the inherent properties of the mutant cells.
Time Considerations The Basic Protocol (ten samples) requires ∼4.5 hr for the staining procedure (assuming a minimal 1-hr ethanol fixation and a 2-hr RNase step), with an additional 1 hr required for flow cytometry and analysis.
Literature Cited Guthrie, C. and Fink, G.R. (eds.) 1991. Guide to yeast genetics and molecular biology. Methods Enzymol. Vol. 169. Haase, S.B. and Lew, D.J. 1997. Flow cytometric analysis of DNA content in budding yeast. Methods Enzymol. 283:322-332. Haase, S.B. and Reed, S.I. 2002. Improved flow cytometric analysis of the budding yeast cell cycle. Cell Cycle 1:132-136. Hutter, K.J. and Eipel, H.E. 1979. Microbial DNA determinations by flow cytometry. J. Gen. Microbiol. 113:369-375.
Contributed by Steven B. Haase Duke University Durham, North Carolina
Cell Cycle Analysis of Budding Yeast Using SYTOX Green
7.23.4 Supplement 26
Current Protocols in Cytometry
Detection of Mitotic Cells
UNIT 7.24
In preparation for cell division, nuclear chromatin undergoes a vital rearrangement required for the organization of chromosomes and their allocation to the daughter cells. These chromatin changes are initiated during the G2 phase of the cell cycle, and their most remarkable morphological manifestation is chromatin condensation, which becomes evident during prophase and is maximal during the ensuing stages of mitosis. The critical event occurring during the G2 to M transition is phosphorylation of histone H3 at Ser 10, which appears to be essential for chromatin condensation. Since the time interval during which histone H3 remains phosphorylated is restricted to mitosis, this event constitutes a specific marker discriminating mitotic cells. The use of antibody that specifically recognizes only H3 phosphorylated at Ser 10 (anti-H3-P MAb) provides the means for immunocytochemical detection of H3 phosphorylation, and thereby for detection of mitotic cells. This unit describes procedures for identification of mitotic cells with this antibody, either by flow cytometry (see Basic Protocol) or laser scanning cytometry (see Alternate Protocol). There are a number of potential applications for this methodology. It can be used to identify mitotic cells for the estimation of mitotic index (MI) in populations of cells growing exponentially or treated with drugs. It also can be used in stathmokinetic experiments in which the cells are arrested in mitosis and the rate of cell entrance to M (“cell birth rate”), emptying of the G1 compartment, and many other kinetic parameters can be estimated. Furthermore, the immunocytochemical detection of H3-P provides new opportunities for studying the role of H3 phosphorylation in chromatin condensation during mitosis. In particular, it may be helpful in identification of the kinase(s) and/or protein phosphatase(s) involved in H3 phosphorylation and dephosphorylation. Since these enzymes are a potential target for development of new antitumor drugs designed to target the G2 to M transition, the possibility of immunocytochemical detection of the activity of one or more of them may be of great value in drug screening. In addition, being a marker of mitotic cells, anti-H3-P antibody can be applied in multiparametric cytometric analysis to study H3 phosphorylation in relation to expression of other proteins critical for cell cycle progression or chromatin condensation during mitosis. FLOW CYTOMETRIC ANALYSIS OF HISTONE H3 USING anti-H3-P ANTIBODIES
BASIC PROTOCOL
This protocol describes identification and quantification of mitotic cells based on immunocytochemical detection of histone H3 phosphorylated on Ser 10 (H3-P) concurrently with differential staining of cellular DNA. The cells are briefly fixed in 1% formaldehyde, permeabilized with 70% ethanol, and immunostained with monoclonal H3-P Ab, and their DNA is counterstained with propidium iodide (PI). Bivariate analysis of cellular green fluorescence intensity (H3-P Ab; FITC-tagged) versus red (PI) fluorescence intensity allows one to distinguish G0/G1 versus S versus G2 versus M cell subpopulations. Materials Cells to be analyzed: growing in culture (APPENDIX 3B) or dissociated from solid tumor (UNIT 5.2; but do not fix dissociated cells) Phosphate buffered saline (PBS; APPENDIX 2A) 1% (v/v) formaldehyde in PBS (see APPENDIX 2A for PBS) 80% ethanol, –20°C 0.25% (v/v) Triton X-100 in PBS, pH 7.4 (store at 4°C) Nucleic Acid Analysis Contributed by Gloria Juan and Zbigniew Darzynkiewicz Current Protocols in Cytometry (2004) 7.24.1-7.24.7 Copyright ©2004 by John Wiley & Sons, Inc.
7.24.1 Supplement 28
Rinsing buffer: 1% (w/v) bovine serum albumin (BSA) in PBS, pH 7.4 (store at 4°C) Primary antibody to phosphorylated histone H3 (anti-H3-P MAb; Sigma) Mouse IgG2a (isotypic control) Secondary antibody: FITC-conjugated goat anti-mouse IgG PI staining buffer: 5 µg/ml propidium iodide (PI) and 200 µg/ml DNase-free RNase A in PBS, pH 7.4, freshly prepared 15-ml conical tubes, silanized or polypropylene Flow cytometer equipped with 488-nm argon laser and filters for collection of green (530 ± 20-nm) and red (>620-nm) fluorescence Additional reagents and equipment for trypsinizing cells (APPENDIX 3B) Prepare cell suspension for fixation 1a. For cells growing in suspension or hematologic samples: Rinse cells by centrifuging 5 min at 300 × g, room temperature. Remove supernatant, then resuspend cells in 5 ml PBS. Repeat centrifugation, remove supernatant, and resuspend at ∼1 × 106 cells/ml with PBS. 1b. For cells growing attached to tissue culture dishes: Collect cells from flasks or petri dishes by trypsinization (APPENDIX 3B) and pool the trypsinized cells with the cells floating in the medium (the latter consist of detached mitotic, apoptotic, and dead cells). Centrifuge 5 min at 300 × g, room temperature, and remove supernatant. Resuspend cells in 5 ml of serum-containing medium, repeat centrifugation, remove supernatant, and resuspend at ∼1 × 106 cells/ml with PBS. Serum is present to inactivate trypsin; other means of trypsin inactivation such as addition of protease inhibitors may be used instead of serum.
1c. For cells isolated from solid tumors: Rinse cells as described above to free them of any enzyme used for cell dissociation and resuspend cells in PBS at ∼1 × 106 cells/ml. In the final suspension in PBS, the cells should be well dispersed (not in aggregates) and should not exceed a density of 5 × 106 cells/ml.
Fix cells 2. With a Pasteur pipet, transfer 1 ml of each cell suspension to be tested into a separate 15-ml tube containing 10 ml of 1% formaldehyde in PBS at pH 7.4. Keep on ice 15 min. Include at least one tube for the negative control. 3. Centrifuge 5 min at 300 × g, room temperature, remove supernatant, then resuspend in 1 ml PBS. Post-fix by adding 10 ml of ice-cold 80% ethanol. At this stage, the sample may be stored at 4°C from 4 hr to several months.
Label cells with anti-H3P antibody To minimize cell loss, perform all of the following steps in the same tube. 4. Centrifuge 5 min at 300 × g, room temperature, remove ethanol, suspend cells in 5 ml PBS at room temperature, and centrifuge again. 5. Resuspend cell pellet (106 cells) in 1 ml of 0.25% Triton X-100 in PBS. Keep on ice 5 min, add 5 ml PBS, and centrifuge 5 min at 300 × g, room temperature.
Detection of Mitotic Cells
7.24.2 Supplement 28
Current Protocols in Cytometry
6. Remove supernatant and suspend cell pellet in 100 µl rinsing buffer containing the primary antibody at the appropriate dilution to obtain 0.5 µg antibody per sample. Incubate 60 min at room temperature with gentle agitation or at 4°C overnight. For the control cells, instead of using the anti-H3-P antibody, incubate the cells with the isotypic antibody. Label cells with FITC-conjugated goat anti-mouse IgG 7. Add 5 ml rinsing buffer and centrifuge 5 min at 300 × g, room temperature. Remove supernatant. 8. Suspend the cell pellet in 100 µl of rinsing buffer containing FITC-conjugated goat anti–mouse IgG antibody (secondary antibody) diluted 1:30. Incubate 30 min in the dark at room temperature with gentle agitation. Stain with PI 9. Add 5 ml rinsing buffer and centrifuge 5 min at 300 × g, room temperature. Remove supernatant. 10. Suspend the cell pellet in PI staining buffer. Incubate 20 min at room temperature in the dark before measurement. Perform cytometry 11. Run samples on flow cytometer using 488-nm excitation. 12. Collect green FITC fluorescence at 530 ± 20 mm (phosporylated histone H3) and red PI fluorescence above 620 nm (DNA). LASER-SCANNING CYTOMETRIC ANALYSIS OF HISTONE H3 USING ANTI-H3-P ANTIBODIES
ALTERNATE PROTOCOL
The method described above (see Basic Protocol) can be adapted to stain cells mounted on microscope slides for analysis by multiparameter laser scanning cytometry (LSC). The cells are initially attached to the slides by cytospinning, then fixed, rinsed, and stained. Additional Materials (also see Basic Protocol) Tissue culture medium with serum 70% ethanol Paraffin or gelatin-based sealer for coverslips Shandon Cytospin cytocentrifuge and Cytospin chambers Coplin jars Humid chamber (vessel with lid, containing moistened paper towels) Slides and coverslips Laser scanning cytometer (LSC) equipped with 488-nm argon laser and filters for collection of green (530±20-nm) and red (>620-nm) fluorescence 1. Place 300 µl tissue culture medium (with serum) containing ∼20,000 cells in a Cytospin chamber (assembled with microscope slide). Centrifuge 6 min in cytocentrifuge at 1000 rpm at room temperature. 2. Without allowing the Cytospin preparations on the slides to dry completely, immerse them in Coplin jars containing fixative (1% formaldehyde in PBS) for 15 min on ice, then transfer the slides to Coplin jars containing 70% ethanol. Store at 0° to 4°C for 2 hr to several weeks. Nucleic Acid Analysis
7.24.3 Current Protocols in Cytometry
Supplement 28
3. Remove the slides, rinse them in PBS, and place them horizontally. Label and stain the cells using the same reagents and incubations as for the flow procedure (see Basic Protocol, steps 5 to 10), but employ the following modifications: a. On the Cytospin area of the slides, apply small volumes (∼300 µl) of the respective buffers, rinses, or staining solutions. b. At appropriate times, remove these solutions with a Pasteur pipet (or vacuum suction pipet). c. Carry out incubations in a humid chamber. 4. At the final step of the staining procedure, mount the cells in a drop of PI staining buffer under the coverslip. Seal coverslips with warm paraffin or gelatin-based sealer. 5. Measure the cell fluorescence by LSC, using the same fluorescence excitation wavelength and emission filters as for flow cytometry (see Basic Protocol, steps 11 and 12). COMMENTARY Background Information
Detection of Mitotic Cells
A number of characteristic features of chromatin can be used to differentiate mitotic from interphase cells by flow or laser scanning cytometry. Thus, morphological features of the condensed chromatin of mitotic cells are sufficiently distinct from those of interphase chromatin to modify the laser light-scatter signal (Nusse et al., 1990) or produce different fluorescence intensity pulse-width or fluorescence intensity peak signals (Gorczyca et al., 1996; Luther and Kamentsky, 1996). These features, though, are not unique to mitosis and do not always allow one to differentiate mitotic cells from apoptotic cells, as the latter also have highly condensed chromatin. Similarly, altered DNA stainability with different fluorochromes varies depending on cell fixation and other treatments, and is similar in cells undergoing apoptosis (Darzynkiewicz et al., 1996). Cell lysis, which is needed to discriminate the clusters of mitotic chromosomes from the interphase nucleus based on the decreased protein content of the former (Roti Roti et al., 1982), makes this approach useless in experiments aimed at analysis of chromatin components other than DNA. The procedure for detection of mitotic cells based on in situ DNA denaturation is rapid and inexpensive (Darzynkiewicz et al., 1980; Juan et al., 1996), but the harsh conditions (acid or heat) required to denature DNA destroy the native structure of many cellular components, making the procedure incompatible with immunocytochemical detection of other cellular constituents. Furthermore, the DNA in apoptotic cells is also sensitive to denaturation (Darzynkiewicz et al., 1997). The method based on sensitivity of DNA in situ to
single strand–specific S1 or mung bean nucleases (Juan et al., 1996) is quite specific for identifying mitotic cells, but is rather complex and requires expensive reagents. The methods for identification of mitotic cells based on the increased reactivity with Ki-67 antibody combined with low expression of PCNA (Landberg et al., 1990) or absence of cyclin A (Gong et al., 1995) require analysis of cell fluorescence at multiple wavelengths. This is a limitation when one wishes to analyze additional features of mitotic versus interphase cells using a singlelaser instrument. Two immunocytochemical methods, one based on the increased accessibility of the chromatin constituent reacting with AF-2 antibody (Di Vinci et al., 1993) and the other on the presence of the p105 protein reacting with the antibody raised against whole nuclei of the mitogen-activated lymphocytes (Clevenger et al., 1987), do not have most of the limitations discussed above. In both cases, however, the nature of the antigen is poorly understood and the AF-2 antibody also reacts with post-mitotic cells (Di Vinci et al., 1993). Phosphorylation of histone H3 (H3) is highly correlated with the G2 to M transition (reviews, van Holde, 1989; Wolffe, 1992). In mitotic cells, H3 is specifically phosphorylated at Ser 10, near its N-terminus (Paulson and Taylor, 1982; Ajiro and Nishimoto, 1985). H3 dephosphorylation occurs quite rapidly after mitosis and Ser 10 remains unphosphorylated throughout the remainder of interphase (Balhorn et al., 1975; Gurley et al., 1975, 1978). Anti-H3-P MAb is a marker of mitotic cells which offers certain advantages over the other markers described above. It can be used in flow or laser scanning cytometry to discern these
7.24.4 Supplement 28
Current Protocols in Cytometry
1000
A
B
1000 100
100 H3-P
C M
M monocytes
10
10
lymphocytes
S
1
1
G0/G1
G0 100
G2
400
800
100
S
G2
G1 400
800
100
400
800
DNA content
Figure 7.24.1 Binding of H3-P MAb by lymphocytes, monocytes, and MOLT-4 cells. Mononuclear cells obtained after density gradient separation either (A) immediately or (B) after culturing for 48 hr in the presence of the mitogen phytohemagglutinin were fixed and stained with H3-P MAb according to the Basic Protocol. Also shown (C) are MOLT-4 cells incubated with vinblastine (as in Fig. 7.24.2), for comparison. The bivariate distributions (scatter plots) of H3-P immunofluorescence (logarithmic scale) versus cellular DNA content demonstrate that monocytes, with a G0/1 DNA content, have higher intensity of H3-P fluorescence, compared to lymphocytes. Mitotic cells, particularly arrested in mitosis by vinblastine, have the highest intensity of H3-P immunofluorescence, and can be distinguished from monocytes by G2/M DNA content.
cells, e.g., for sorting, multiparametric analysis, or rapid scoring of mitotic index, or in stathmokinetic experiments (Darzynkiewicz et al., 1986). Moreover, apoptotic cells are not reactive with anti-H3-P MAb. It should be noted that polyclonal anti-H3-P Ab is also commercially available (from Upstate Biotechnology), and it also has been successfully used by the authors and others for immunocytochemical identification of mitotic cells. The H3-P antibody, being a marker of mitotic cells, can be applied in multiparametric cytometric analysis to study H3 phosphorylation in relation to expression of other proteins critical for cell cycle progression or chromatin condensation during mitosis. Thus, the status of H3 phosphorylation can be monitored in individual cells and can be correlated with both the cell position in the cycle and the expression of other proteins essential for cells to traverse through G2 and M, such as cyclins A or B1 (Juan et al., 1998). The possibility of detecting H3-P in individual cells offers a new probe for rapid screening of potential drugs, or other agents, for their activity in live cells as inhibitors of kinases or phosphatases involved in H3 phosphorylation. This provides an alternative to in vitro analyses using lysed cells or cell fractions. It should be noted that histone H3 is also found in the phosphorylated state in monocytes; its phosphorylation in these cells appears
to be associated with chromatin changes that are associated with cell differentiation along the monocytic pathway (Juan et al., 1999). Thus, monocytes can be immunocytochemically distinguished from other white blood cells based on binding of histone H3-P Ab. Having a G1 DNA content, however, monocytes, in turn, can be distinguished from mitotic cells, the latter having a G2/M DNA content (Juan et al., 1999; Fig. 7.24.1).
Critical Parameters and Troubleshooting It is important to collect the cells floating in the medium before trypsinization in order to recover detached mitotic, apoptotic, and dead cells. Cell fixation and permeabilization are critical steps for immunocytochemical detection of intracellular proteins, and often must be customized for particular antigens. The fixative is expected to stabilize the antigen in situ and preserve its epitope in a state where it continues to remain reactive with the available antibody. The cell must be permeable, to allow access of the antibody to the epitope (Clevenger et al., 1987). A brief (15-min) treatment with 1% formaldehyde followed by 80% cold ethanol works well to avoid the loss of unbound or loosely bound proteins during the permeabilization process, especially for multiparameter
Nucleic Acid Analysis
7.24.5 Current Protocols in Cytometry
Supplement 28
A
B G1
60
60 G1
50
H3-P
G2M
50 40
40 S
S
G2M M
30
30 control
vinblastine
20
20 M
10 0
0
G1
S
10
20
30
G1
10
G2 40
50
60
0
S G2
0
10
20
30
40
50
60
DNA content
Figure 7.24.2 Bivariate distributions of DNA content versus H3-P immunofluorescence (linear scale) of MOLT-4 cells growing (A) exponentially (control cells) or (B) following exposure to vinblastine for 6 hr. Relatively few cells in mitosis (H3-P positive) are present in asynchronously growing cultures. Treatment with vinblastine increases the percentage of cells in mitosis; those cells manifest increased stainability with the anti-H3-P MAb. Insets show DNA content frequency histograms (PI integrated fluorescence) in the respective cultures. The cells were stained according to the Basic Protocol.
Detection of Mitotic Cells
analysis. It should be noted that extensive DNA-DNA or DNA-protein cross-linking occurs as the result of higher concentrations or longer fixation times in the presence of formaldehyde, which, in turn, impairs DNA stainability with intercalating dyes such as PI. Therefore, if accurate DNA distributions are required, mild formaldehyde fixation, as described above, is necessary. As with other immunocytochemical markers, each antibody used to detect histone H3-P in a particular cell type/line must first be titrated to find the optimal concentration. It is advisable to cover the 0.1- to 3.2-µg/ml range of concentration (e.g., 0.1, 0.2, 0.4, 0.8, 1.6, and 3.2 µg/ml), plot the mean immunofluorescence intensity of M cells as a function of antibody concentration, and then use the lowest concentration of this antibody before the plateau of immunofluorescence intensity is reached on the plot. When staining is done on slides, small (1 × 1–cm) pieces of thin polyethylene foil may be layered over the solutions used for cell incubations to prevent drying. Preincubation of permeabilized cells with alkaline phosphatase can be used as a control to abolish the reactivity with anti-H3-P, dem-
onstrating that H3-P does not detect the unphosphorylated epitope.
Anticipated Results Immunocytochemical staining with antiH3-P antibody is basically limited to mitotic cells, from prophase to late telophase. Prophase cells show a subtle, rather diffuse, and in some sites “thread-like” pattern of labeling. Intense staining of chromosomes appears during later stages of mitosis, with chromosomes present in metaphase and anaphase cells being the most strongly stained. Figure 7.24.2 illustrates a flow cytometric bivariate distribution of DNA content versus H3-P in untreated MOLT-4 cells (panel A) and MOLT-4 cells exposed to 50 ng/ml vinblastine for 6 hr (panel B). In untreated U-937, MOLT-4, and HL-60 cell cultures, typically 1% to 2% of cells have high H3-P immunofluorescence, which corresponds to the percentage of mitotic cells in these cultures determined by microscopy. Cell incubation in the presence of the microtubule poison vinblastine, which arrests cells in metaphase, increases the proportion of cells in mitosis. As is manifest in panel B, a cell subpopulation with a G2/M DNA content stains more intensely with anti-H3-P compared to
7.24.6 Supplement 28
Current Protocols in Cytometry
other (e.g., S-phase) cells. Preincubation of permeabilized cells with alkaline phosphatase abolishes their reactivity with anti-H3-P (not shown).
Time Considerations
The full protocol can be completed in ∼3 to 4 hr. The fixation is a good stopping point (fixed cells may be stored at 4°C from 4 hr up to several months). The cell labeling with the first antibody can be performed either at room temperature for 1 hr or at 4°C overnight.
Literature Cited Ajiro, K. and Nishimoto, T. 1985. Specific site of histone H3 phosphorylation related to the maintenance of premature chromosome condensation: Evidence for catalytically induced interchange of the subunits. J. Biol. Chem. 260:15379-15381. Balhorn, R., Jackson, V., Granner, D., and Chalkley, R. 1975. Phosphorylation of the lysine-rich histones throughout the cell cycle. Biochemistry 14:2504-2511. Clevenger, C.V., Epstein, A.L., and Bauer, K.D. 1987. Quantitative analysis of a nuclear antigen in interphase and mitotic cells. Cytometry 8:280286. Darzynkiewicz, Z., Traganos, F. and Melamed, M.R. 1980. New cell compartments identified by multiparameter flow cytometry. Cytometry 1:98108. Darzynkiewicz, Z., Gong, J. Juan, G., Ardelt, B., and Traganos, F. 1996. Cytometry of cyclin proteins. Cytometry 23:1-13. Darzynkiewicz, Z., Juan, G., Li, X., Gorczyca, W., Murakami, T., and Traganos, F. 1997. Cytometry in cell necrobiology: Analysis of apoptosis and accidental cell death. Cytometry 27:1-20. Di Vinci, A., Geido, E., Pfeffer, U., Vidali, G., and Giaretti, W. 1993. Quantitative analysis of mitotic and early-G1 cells using monoclonal antibodies against AF-2 protein. Cytometry 14:421427. Gong, J., Traganos, F., and Darzynkiewicz, Z. 1995. Discrimination of G2 and mitotic cells by flow cytometry based on different expression of cyclins A and B1. Exp. Cell Res. 220: 226-231. Gorczyca, W., Melamed, M.R., and Darzynkiewicz, Z. 1996. Laser scanning cytometer (LSC) analysis of fraction of labeled mitoses. Cell Prolif. 29:539-547. Gurley, L.R., Walters, R.A., and Tobey, R.A. 1975. Sequential phosphorylation of histone subfractions in the Chinese hamster cell cycle. J. Biol Chem. 250:3936-3944.
Gurley, L.R., Walters, R.A., Barham, S.S., and Daeven, L.L. 1978. Heterochromatin and histone phosphorylation. Exp. Cell Res. 11:373383. Juan, G., Pan, W., and Darzynkiewicz, Z. 1996. DNA segments sensitive to single strand specific nucleases are present in chromatin of mitotic cells. Exp. Cell Res. 227:197-202. Juan, G., Li, X., and Darzynkiewicz, Z. 1997. Correlation between DNA replication and expression of cyclins A and B1 in individual MOLT-4 cells. Cancer Res. 57:803-807. Juan, G., Traganos, F., James, W.M., Ray, J.M., Roberge, M., Sauve, D.M., Anderson, H., and Darzynkiewicz, Z. 1998. Histone H3 phosphorylation and expression of cyclins A and B1 measured in individual cells during their progression through G2 and mitosis. Cytometry 32:1-8. Juan, G., Traganos, F., and Darzynkiewicz, Z. 1999. Histone H3 phosphorylation in human monocytes and during HL-60 cell differentiation. Exp. Cell Res. 246:212-220. Landberg, G., Tan, E.M., and Ross, G. 1990. Flow cytometric multiparameter analysis of proliferating cell nuclear antigen/cyclin and Ki-67 antigen: A new view of the cell cycle. Exp. Cell Res. 187:111-118. Luther, E. and Kamentsky, L.A. 1996. Resolution of mitotic cells using laser scanning cytometry. Cytometry 23:272-278. Nusse, M., Beisker, W., Hoffman, C., and Tarnok, A. 1990. Flow cytometric analysis of G1- and G2/M-phase subpopulations in mammalian cell nuclei using side scatter and DNA content measurements. Cytometry 11:813-821. Paulson, J.R. and Taylor, S.S. 1982. Phosphorylation of histone 1 and 3 and nonhistone high mobility group 14 by an endogenous kinase in HeLa metaphase chromosomes. J. Biol. Chem. 257:6064-6072. Roti Roti, J.L., Higashikubo, R., Blair, O.C., and Uygur, N. 1982. Cell-cycle position and nuclear protein content. Cytometry 3:91-96. van Holde, K.E. 1989. Chromatin. Springer-Verlag, New York. Wolffe, A. 1992. Chromatin: Structure and Function. Academic Press, San Diego.
Contributed by Gloria Juan Memorial Sloan-Kettering Cancer Center New York, New York Zbigniew Darzynkiewicz Brander Cancer Research Institute New York Medical College Valhalla, New York
Nucleic Acid Analysis
7.24.7 Current Protocols in Cytometry
Supplement 28
DRAQ5 Labeling of Nuclear DNA in Live and Fixed Cells
UNIT 7.25
This unit describes the use of a novel DNA-detecting far-red-fluorescing dye, DRAQ5, which has a unique combination of properties exploitable within live- and fixed-cell cytometry. Excitation at 647 nm, close to the wavelength of maximum fluorescence excitation (Exλmax) for DRAQ5, produces a fluorescence emission spectrum extending from 665 nm out to beyond 780 nm. Thus the emission spectrum is beyond that of fluorescein, phycoerythrin, Texas Red, Cy3, and (perhaps most importantly) enhanced green fluorescent protein (EGFP). The excitation characteristics also separate the agent from propidium iodide. This live- or fixed-cell DNA probe can be deployed within a range of cell-based assays that demand the quantification, discrimination, or location of DNA using a range of analysis platforms. Here the focus is on flow cytometric assays, but typical imaging applications are also indicated because the staining protocols share the same essential features. DRAQ5 combines a high capacity to permeate the cell membrane and a high DNA binding affinity and selectivity with an excitation and fluorescence emission profile that is distinct from that of UV and visible-range fluorochromes. The high DNA binding affinity and selectivity relate to the molecular structure of DRAQ5, which, by molecular modeling, favors intercalation at AT sequences. The use of cytometers with multilaser specifications has enhanced the sophistication of subset analyses owing in part to the increased availability of fluorescent reporters with different Exλmax. The protocols described here permit use of DRAQ5 to enable discrimination of DNA for such analyses through its far-red fluorescence signature. The staining procedure described in the Basic Protocol is simple. Importantly, there have been advances in live-cell assays in which nuclear staining by agents such as DRAQ5 is used to distinguish nuclear versus cytoplasmic fluorescence signals or intracellular translocation in a range of cell analysis formats, including multiwell plates, optical biochips, and slide-based cytology. Accordingly, this unit describes the relevant characteristics of the DNA probe in conjunction with a single basic staining protocol so that the user can employ the agent in a range of fluorescence-based assays where nuclear discrimination or DNA content analysis is a parameter of value. An Alternate Protocol describes DRAQ5 staining of fixed cells. As with all new agents, the full performance envelope has yet to be explored, and the Commentary touches upon these aspects to place DRAQ5 in perspective with other options available for cell-based DNA analyses. DRAQ5 is weakly fluorescent, in keeping with the anthraquinone chromophore (Bell, 1988; Fox and Smith, 1995). Modified anthraquinones show differences in intracellular distribution and toxicity (Smith et al., 1997a,b), whereas DRAQ5 shows a unique combination of elevated DNA affinity and intracellular specificity for DNA while retaining cell-permeant and fluorescence properties (Smith et al., 2000). Thus high levels of DRAQ5 molecules bind to cellular DNA, resulting in a distinct nuclear signal without the need to remove the dye from the extracellular medium. The rapid staining potential of the dye also allows it to be used at the final stage in any multilabeling protocol. CAUTION: DRAQ5 is cytotoxic and may be carcinogenic. Use appropriate safety precautions when handling.
Nucleic Acid Analysis Contributed by Paul J. Smith, Marie Wiltshire, and Rachel J. Errington Current Protocols in Cytometry (2004) 7.25.1-7.25.11 Copyright ©2004 by John Wiley & Sons, Inc.
7.25.1 Supplement 28
BASIC PROTOCOL
PREPARATION AND DRAQ5 STAINING OF LIVE CELLS FOR ANALYSIS USING FLOW CYTOMETRY OR IMAGING Nuclear staining with DRAQ5 is straightforward, and most cultured mammalian cells label well in full culture medium (containing 10% fetal bovine serum; FBS), although for convenience pH can be maintained more easily during manipulation by including 10 mM HEPES. Staining is very rapid, reaching equilibrium within 15 min for concentrations in the 1 to 20 µM range (Smith et al., 1999, 2000). Staining in phosphate buffers is less efficient. At high cell density (i.e., high DNA concentration), the dye availability can be a limiting factor. The staining rate is enhanced at 37°C, but adequate staining can be achieved at room temperature and below. Suspension cells require no preparation and can be stained by direct addition of DRAQ5. The SU-DHL-4 cell line (Smith et al., 1999) has a high ratio of nuclear to cytoplasmic volume and can be maintained in suspension culture with a low burden of debris and apoptotic cells. Attached cells can be prepared for staining and flow cytometry by standard cell detachment methods. There is no evidence that the method of detachment affects staining potential. Attached live-cell studies using microscopy-based systems are best conducted with cells seeded onto #1.5 coverglass chambered wells (Lab-Tek Products) appropriate for high-resolution imaging. Materials Mammalian cells of interest to be grown attached or in suspension Complete medium appropriate for cells of interest, with optional 10 mM HEPES (e.g., Sigma), pH 7.2 5 mM DRAQ5 acidified stock (Biostatus); store at room temperature or 4°C Flow cytometer (e.g., FACS Vantage cell sorter; Becton Dickinson) with two lasers emitting at 488 nm and 633 nm, respectively, and appropriate software (e.g., CellQuest; Becton Dickinson) or laser scanning microscope, such as 1024MP scanning unit with LaserSharp software (Bio-Rad Laboratories) attached to a Zeiss Axiovert 135 (Carl Zeiss) with 63× 1.4–numerical aperture (NA) or 40× 1.3-NA oil-immersion lens, operating in confocal laser scanning microscope (CLSM) mode using 488-, 568-, or 647-nm lines of krypton-argon laser Additional reagents and equipment for cell culture and detachment of adherent cells 1. Seed mammalian cells of interest into a culture dish or chambered well or onto coverslips (∼1 × 105 cells/25 cm2) or initiate suspension cultures (∼1 × 105 cells/ml) prior to incubation under standard conditions. Typically, cultures are incubated for up to 48 hr. The human follicular B-lymphoma suspension cell line SUD4 (kindly supplied by Dr. F. Cotter, LRF Centre for Childhood Leukaemia at the Institute of Child Health, London, United Kingdom), which exemplifies DRAQ5 staining in this unit, was routinely cultured in RPMI medium (Gibco Life Technologies) supplemented with 10% FBS, 2 mM glutamine, 100 IU/ml penicillin, and 100 ìg/ml streptomycin and was maintained at 37°C in a humidified 5% CO2 environment. The staining procedure is a generic one appropriate for any mammalian cell type. Each cell type will have specific culture conditions; the important issue to ensure successful nuclear labeling is to use the recommended cell density and dye concentration given in steps 2 and 3.
DRAQ5 Labeling of Nuclear DNA in Live and Fixed Cells
7.25.2 Supplement 28
Current Protocols in Cytometry
2. Detach monolayer cultures (e.g., using trypsin/EDTA) and resuspend in complete medium to a concentration of ∼2-4 × 105 cells/ml. Dilute suspension cultures in complete medium to the same concentration. This cell density assumes a human diploid DNA content and an intended staining reaction with 20 ìM DRAQ5. Cells with higher ploidy levels may require higher dye concentrations or lower staining densities. Attached cell cultures (e.g., coverslip cultures or chambered wells) can be stained in situ assuming a 1- to 2-ml staining volume overlayering a 4-cm2 surface area of semi-confluent human quasi-diploid tumor cells.
3. Add 4 µl of 5 mM DRAQ5 acidified stock per ml culture medium (20 µM final). Incubate 5 to 15 min at 37°C. DRAQ5 is a dark-blue crystalline solid of molecular weight 412.54. The dye is supplied ready for use as an acidified aqueous stock solution of 5 mM (Biostatus) and can be diluted in aqueous buffers or added directly to full culture medium. Final DRAQ5 concentration will depend upon the application, but nuclear discrimination is achievable at 2.5 to 5 ìM, and it is unlikely that concentrations >30 ìM would be required. Although nuclear staining will be detectable immediately, a 5- to 15-min incubation at 37°C will yield optimal staining results. It should be noted that overstaining cannot occur.
4. Optional: Wash cells in fresh culture medium or aqueous buffer for 2 to 3 min at 37°C to remove unbound dye. Centrifuge cells 3 to 5 min at 800 × g, 37°C. Discard supernatant and resuspend in complete medium with 10 mM HEPES at 4 × 105 cells/ml. Alternatively, an appropriate aqueous buffer can be used to wash the cells. The wash step removes excess dye from the medium and allows the user to tightly define the labeling conditions if this is required. Also, if a second vital dye is to be used, the wash step avoids multiple dye complexes in the medium. Cell suspensions and attached cultures can also be examined directly in the presence of the dye.
5a. For flow cytometry: Load cells into a flow cytometer and use forward and side (90°) scatter to identify debris. Use conventional pulse analysis for doublet discrimination. Analyze parameters using appropriate software. The authors use an Enterprise II laser (Coherent Laser) emitting at 488 nm and a 127-35 helium-neon laser (maximum 35-mW output; Spectra Physics) emitting at 633 nm with a temporal separation of about 25 ìsec from that of the primary 488-nm beam. Light-scatter signals are collected as standard. The analysis optics comprise a 715-nm long-pass filter for analysis of signals originating from the primary beam, after reflection at an SP610 dichroic, and a 695-nm long-pass filter for analysis of the delayed-beam signals (DRAQ5 emission).
5b. For laser scanning microscopy: Collect DRAQ5 fluorescence images using a 695-nm long-pass filter. DRAQ5 fluorescence is routinely collected at >680 nm (peak emission extending to ∼700 nm) and is detectable within seconds of dye addition to cell suspensions. It is important to appreciate that although excitation is optimal at 633 and 647 nm, a fluorescence signal of equivalent quality may be generated by lines down to 488 nm. Accordingly, DRAQ5 can be analyzed using a range of instrumentation provided with appropriate light collection optics. For a laser scanning microscope, each optical slice typically consists of 512 × 512 pixels (x,y resolution is 0.35 ìm, z resolution is ∼1 ìm). Nucleic Acid Analysis
7.25.3 Current Protocols in Cytometry
Supplement 28
ALTERNATE PROTOCOL
DRAQ5 STAINING OF FIXED CELLS Most commonly used fixation protocols, such as 70% (v/v) ethanol or paraformaldehydebased methods, appear to be compatible with postfixation staining using DRAQ5. When using fixed samples, consider when to introduce the DRAQ5 staining component. Nuclear-located DRAQ5 is persistent and can be fixed (stabilized) in live cells and later detected after processing for immunofluorescence analysis, providing a link between the nuclear structures identified in live cells and the location of an immunofluorescence signal. In this case, a standard fixation using 4% (w/v) paraformaldehyde in PBS (APPENDIX 2A) for 30 min with resuspension in an aqueous buffer (e.g., PBS) is recommended. Alternatively, cells fixed in paraformaldehyde or 70% (v/v) ethanol and then rehydrated in PBS can readily be stained by simple addition of the dye. When staining fixed cells, similar concentrations of dye (e.g., 20 µM final concentration) and similar incubation conditions can be used as for live cells, producing similar results. Recent applications in high-throughput screening have suggested that DRAQ5 can be incorporated into the fixing medium as part of a fix-and-stain protocol (see Internet Resources). COMMENTARY Background Information
DRAQ5 Labeling of Nuclear DNA in Live and Fixed Cells
A range of DNA dyes are available for cytometric applications (reviewed by Darzynkiewicz and Kapuscinski, 1990; Latt and Langlois, 1990; Waggoner, 1990), although only a restricted number label DNA and are suitable for use in live-cell systems. Hoechst 33258, Hoechst 33342, and DAPI are the most frequently used UV-excitable fluorochromes with convenient maximum excitation and emission wavelengths (Exλmax and Emλmax). Clearly these agents demand UV excitation (sub-optimal low-violet excitation), and their emission range reaches across the visible spectrum, resulting in spectral contamination when combined with visible-range fluors. Reducing emission detection of the Hoechst dyes to narrow spectral windows in the violet or red regions of the spectrum can generate anomalous DNA reporting signals resulting from the red shift in emission maximum that occurs as the ligand titrates its binding sites in the minor groove (Smith et al., 1991). Although cell permeant to some extent, acridine orange displays metachromatic staining of nucleic acids that precludes its convenient use as a DNA-discriminating probe. The cell-permeant nucleic acid stain LDS-751 (Terstappen et al., 1988) in its DNA-bound form has an Exλmax of ∼543 nm and exhibits a far-red emission maximum (Terstappen et al., 1989; Frey, 1995). A recent study has shown that LDS-751 binds almost exclusively to mitochondria when incubated with viable, nucleated fibroblasts and monocytes, suggesting that it cannot be regarded as a nucleic acid–discriminating stain (Snyder and Small, 2001).
The introduction of the cell-permeant cyanine SYTO nucleic–acid stains (e.g., SYTO 17; Frey, 1995; Frey et al., 1995) has provided reagents with convenient red-shifted spectral characteristics. These agents passively diffuse through the membrane of most cells and can be excited by UV or visible light, but stain RNA and DNA in both live and dead eukaryotic cells. It has been stressed that the highly versatile SYTO dyes do not act exclusively as nuclear stains in live cells but may have properties that commend their use in live versus dead cell discrimination (Van Zandvoort et al., 2002). The anthraquinones are a group of synthetic DNA intercalating agents (Lown, 1985) that are only weakly fluorescent (Bell, 1988). DRAQ5 is a modified anthraquinone with enhanced DNA affinity and intracellular selectivity for nuclear DNA. In immunophenotyping laser scanning cytometry applications, a protocol has been described (see Internet Resources) in which antibody (CD4-FITC, CD8a-PE) is diluted as for flow analysis in 1:1000 diluted DRAQ5 staining solution. Cells are prepared in 30 µl of solution containing both the antibody and DRAQ5 and stained 30 min on ice. Cells are then washed once with PBS and prepared by dropping onto slides. DRAQ5 has recently been used to study histone H3 phosphorylation during Xenopus oocyte maturation and its relationship with DNA condensation (Schmitt et al., 2002). The immunofluorescence confocal microscopy of oocytes was undertaken using a Zeiss LSM510 confocal microscope (488- and 633-nm laser lines) to track co-localization of phospho-histone H3 antibody staining and nuclear structures.
7.25.4 Supplement 28
Current Protocols in Cytometry
DRAQ5 has also been used for DNA-dye fluorescence lifetime imaging (Van Zandvoort et al., 2002); DNA content analysis and multiparameter flow cytometry (Smith et al., 2000; Plander et al., 2003); blood sample analysis (Smith et al., 1999); nuclear DNA discrimination in leukemia and lymphoma cells (Allman et al., 2003; Weng et al., 2003); and mammalian, human, and amphibian cell nuclear staining (Mitta et al., 2002; Schmitt et al., 2002; Smith et al., 2002; Fitzgerald et al., 2003).
Critical Parameters The key to success in using DRAQ5 is to appreciate that although the dye is not bright its unique spectral and physical properties provide a highly advantageous signal-to-noise profile in single and multifluor combinations. Thus care should be taken to ensure that the excitation and emission optics are optimal. Lampbased excitation appears to be less efficient than laser-based excitation, although it should be noted that the deep-red fluorescence of DRAQ5 is difficult to detect by the human eye using direct observation and epifluorescence microscopy. Because DRAQ5-DNA binding is essen-
A frequency
tially stoichiometric, care should also be exercised in experiment planning to adjust the cell density and total cell number in the final staining reaction so that dye concentration is not limiting. Critical physical properties DRAQ5 is a pure synthetic compound with high affinity for DNA; it is stable at room temperature and under normal lighting conditions and is soluble in water at a biologically compatible pH. Under suboptimal solubilization conditions, DRAQ5 can appear as a fine suspension, hence its provision as a ready-foruse titrated stock solution. The solution is stable at room temperature but can be stored routinely at 4°C, although freezing should be avoided. It is recommended to use the agent directly from stock solutions rather than prepare diluted stocks for long-term storage. Although there is no need to protect it from light, dark storage is prudent. The dye can be used as a membranepermeant fluorescent dye for the rapid and convenient staining of nuclear DNA of organisms, including live or fixed mammalian cells, with minimal RNA-associated fluorescence.
B frequency
580–620 460–500 340–380 220–260 time interval 100–140 (sec) DRAQ5 fluorescence intensity
Figure 7.25.1 Flow cytometric analysis of the rapid cellular uptake of DRAQ5, demonstrating the ability to discriminate DNA content using either (A) 488-nm excitation (emission wavelength [Emλ] >715 nm) or (B) 633-nm excitation (Emλ >695 nm). DRAQ5-DNA fluorescence intensity distributions (for each channel number) of live human B cell SU-DHL-4 lymphoma cells were monitored during 40-sec acquisition periods (∼8 × 103 events) at each of the indicated periods of exposure to the dye. Cell suspensions (4 × 105 cells/ml) were prepared in complete medium supplemented with 10 mM HEPES at 37°C. The dye was added immediately prior to analysis at a concentration of 20 µM DRAQ5. Data show the rapid establishment of distributions for content analysis and the similarity of the profiles for blue- or red-line excitation.
Nucleic Acid Analysis
7.25.5 Current Protocols in Cytometry
Supplement 28
140 120
Counts
100 80 60 40 20 0
0
200 400 600 800 DNA content (channel number)
1000
Figure 7.25.2 Comparison of the DNA content distributions derived from live (solid line) and ethanol-fixed (dashed line) human B cell SU-DHL-4 lymphoma cells (4 × 105 cells/ml). Live cells were stained as described for Figure 7.25.1 and data were acquired using 488-nm excitation (>715-nm emission) after a 560-sec exposure to 20 µM DRAQ5. Fixed cells were prepared by resuspending washed live-cell pellets in 70% (v/v) ethanol in PBS and holding 30 min on ice prior to centrifugation, washing, and resuspension in PBS supplemented with 20 µM DRAQ5. Data were acquired as for live cells. The coefficient of variation (CV) for the G1 peak is <5%.
Thus an RNAse treatment of fixed or permeable cells does not need to be incorporated into any analysis and any customized protocols should be adjusted accordingly.
DRAQ5 Labeling of Nuclear DNA in Live and Fixed Cells
Critical spectral properties Before starting, ensure that appropriate analysis optics are available. It is critical to select an appropriate excitation wavelength and laser line. A wide range of excitation wavelengths is possible (e.g., 488, 514, 568, 633, or 647 nm), with optimal excitation at 647 nm. Figure 7.25.1 demonstrates the similarity in fluorescence signal obtained for 488-nm versus 633-nm excitation. The emission spectrum beyond 670 nm provides minimal overlap with emissions from visible-range dyes, including GFP. DRAQ5 shows no apparent fluorescence enhancement upon binding, although in free solution there is a small (∼10-nm) red shift in peak emission wavelength in the presence of DNA. Combining DRAQ5 with FITC does not require compensation, unlike applications using propidium iodide. The probe does not photobleach, an important consideration for gener-
ating multiple scans of cell nuclei during threedimensional image generation. There is no need to wash samples prior to analysis, an advantage for the retention of delicate cells in samples. The rapid uptake and binding of dye can be conveniently tracked over time by flow cytometry (Fig. 7.25.1). UV-excitable DNA dyes (Hoechst 33342, Hoechst 33258, and DAPI) are limited when two-photon excitation of GFP-based fluors is undertaken owing to coexcitation and overlap of the fluorescence spectra. This specific problem can be solved by using a red-excitable DNA dye such as DRAQ5, which is two-photon in the excitation range used for visible-range fluors (Albota et al., 1998; Smith et al., 2000; Bestvater et al., 2002). Vital staining properties Consider the use of DRAQ5 for systems where there may be extensive cell death. Unlike cationic DNA dyes such as propidium iodide, which show exclusion from live cells but stain dead or dying cells with disrupted membranes, DRAQ5 stains all cells. Comparisons of fixed versus live cells stained with DRAQ5 indicate
7.25.6 Supplement 28
Current Protocols in Cytometry
that they attain similar fluorescence intensities (Fig. 7.25.2). Thus DRAQ5 can be used to assess DNA degradation in apoptotic cells and the retention of normal DNA content of live cells in the same in situ assay without the need for centrifugation and fixation—procedures that can selectively remove cells or indeed generate debris during washing procedures. Nuclei can be rapidly discriminated in monolayer systems, permitting accurate counts of cell numbers, the locating of nuclei and cytoplasm in translocation assays, or the recognition of changes in cell cycle status (Fig. 7.25.3). Another related anthraquinone, DRAQ5NO (i.e., Apoptrak; Biostatus), has been described for the positive discrimination of intact cells in apoptotic cell populations (Wiltshire et al., 2000). Whole-blood preparations can be stained directly using DRAQ5, retaining light-scatter characteristics of cells and permitting live-cell immunofluorescence assays (Smith et al., 1999). The ability of DRAQ5 to stain nucleated cells in whole blood may also provide the ability to detect bloodborne parasites. Figure 7.25.4 shows the detection of intra–red blood cell (RBC) chromatin bodies of the malaria parasite Plasmodium falciparum in air-dried blood films mounted in DRAQ5. There is evidence that DRAQ5 can also penetrate into live fungal hyphae (Figure 7.25.5) for the purpose of detection of nuclear location.
A
B
Pharmacological effects Live-cell labeling is subject to micropharmacokinetic effects; these can vary with cell type and in some cases can severely restrict the intracellular staining potential of a permeant dye. The status of the cells under study should be considered when selecting staining protocols. Unlike the Hoechst DNA dyes (Morgan et al., 1989; Smith et al., 1991), DRAQ5 appears to be less subject to classical drug-resistance efflux (mdr-1), but at suboptimal concentrations nuclear accessibility may restrict binding potential. The anthraquinones such as DRAQ5 can damage DNA through the trapping of DNA topoisomerase II and may also cause DNA-protein cross-linking damage (Smith et al., 1997a). The persistence of DRAQ5 on DNA causes cytotoxicity, making it unsuitable for long-term (>12 hr) tracking experiments or viable cell sorting. The initial stress responses to dye binding may be occult or have delayed expression, eventually resulting in cell death or cell cycle arrest. In the authors’ experience, DRAQ5-stained cells can be used in a variety of functional assays, including p53-driven promoter activity reporting, cell cycle–linked GFP expression, induced calcium transients, esterase-based cytoenzymology, and functional studies on mitochondria. Dye-dye interactions The authors have observed that the AT basepair preference of DRAQ5 results in its ability to quench the fluorescence of minor groove–
C
Figure 7.25.3 Confocal microscopy of DRAQ5-stained live cells. (A) Single-plane confocal fluorescence image and (B) corresponding transmission image of live MCF-7 breast tumor cells in a microcolony (Exλ 647 nm; Emλ 680/32 nm) showing DRAQ5-stained nuclei (10 µM) and revealing nuclear architecture. Image captured 180 sec after dye addition to the culture medium containing 10 mM HEPES, pH 7.2. (C) Live human osteosarcoma cells (U2-OS) similarly stained with 20 µM DRAQ5 (600 sec), revealing nuclear features and an anaphase cell (arrow).
Nucleic Acid Analysis
7.25.7 Current Protocols in Cytometry
Supplement 28
binding agents such as Hoechst 33342, and this effect is probably associated with ligand displacement. The high affinity of DRAQ5 for DNA is such that it will also displace ethidium bromide from DNA. A recent report has suggested that DRAQ5 may act to decrease the cellular uptake of boron dipyromethane ( BODI PY)- lab eled compounds (e.g., LysoTracker Green DND 26; Molecular Probes), and this has been attributed to the dyes complexing in solution (Snyder and Garon, 2003). Health and safety Information on DRAQ5 is available from the suppliers (see Internet Resources). Because persistent intercalating agents such as DRAQ5 have the potential to damage DNA, they should be considered both cytotoxic and potentially mutagenic. Although DNA-interacting dyes often have no defined carcinogenic potential, they should always be handled with caution.
Troubleshooting The intense blue color of DRAQ5 in solution means that any attachment of the dye to a surface is easily detected. Adsorption to surfaces is a common problem with many dyes but is often not apparent to the naked eye. Sample tubes on flow cytometer systems operating with DRAQ5 may accumulate a blue coloration, although there is no evidence that the agent leaches into sample streams, and cross-con-
A
tamination has not been noticed. Some possible sources of problems for troubleshooting are: 1. The dye has been used in combination with an agent that may complex with DRAQ5. Check the order of addition of agents in the protocol and the appearance of the DRAQ5 solution to ensure that precipitation has not occurred. Check that the stock or any dilutions have not been frozen. It helps to have a biological control (e.g., a conveniently cultured human lymphoma suspension cell line) that can be used to check cell staining and analysis routines. 2. The sample has not been stained correctly. Check dilution calculations. At 20 µM, DRAQ5 can clearly be seen as a dark-blue solution and often imparts a blue coloration to the sample. The lipophilic properties of DRAQ5 may result in its partitioning with a reduction in the active concentration available for cell staining. For example, it is foreseeable that this may occur in samples carrying high loads of debris or vesicular matter or where cell staining has taken place in the presence of a large sink for the dye (e.g., a tissue mass, spheroids, or large clumps of cells). 3. The live sample has been held too long prior to analysis or held in a destaining buffer. Try to stain samples immediately prior to analysis, run in equilibrium with the dye, and hold cell suspensions on ice if storage for a few hours is required. Samples are usually stable ≥1 hr at
B
mf
tr
DRAQ5 Labeling of Nuclear DNA in Live and Fixed Cells
Figure 7.25.4 Detection of intracellular malarial parasite forms in artificially infected blood cultures using DRAQ5. (A) Fluorescence and (B) transmission images of an air-dried unfixed blood film mounted directly in PBS supplemented with 20 µM DRAQ5. The fluorescence image shows bright regions in red blood cells, representing developmental forms of Plasmodium falciparum [NF54]. Arrows indicate an early marginal form (mf) and a later trophozoite (tr) form. Images were obtained using a Bio-Rad 1024MP confocal imaging system (647-nm excitation). Blood films were kindly supplied by Dr. Laurent Rénia (Cochin, Gustave Roussy, Paris).
7.25.8 Supplement 28
Current Protocols in Cytometry
A
B
Figure 7.25.5 Live Metarhizium anisopliae mycelium mounted 30 min in 20 µM DRAQ5 in PBS. (A) Fluorescence and (B) reflectance images were obtained using a Bio-Rad 1024MP confocal imaging system (combined 647- and 568-nm laser excitation). Fungal nuclei (arrows) are shown. Cultures were kindly supplied by Dr. Tariq M. Butt (School of Biological Sciences, University of Wales, Swansea, United Kingdom).
room temperature. Track light-scatter changes to evaluate any sample degradation. 4. The staining is not matched with the purpose of the staining procedure and/or understaining is not tolerated. Relatively low concentrations (1 to 5 µM) are required to trace the presence of cellular DNA, whereas higher concentrations (20 µM) are required for DNA content reporting. 5. There are fluorescence excitation problems. Use 647-nm or 633-nm excitation if available, although in the authors’ experience wavelengths down to 488 nm will often provide even better coefficient of variation (CV) values owing to higher energy and better beam alignment. 6. There are fluorescence emission problems. Use long-pass rather than band-pass filters to optimize light collection from the wide DRAQ5 emission spectrum. 7. There are analysis problems. Ensure that cell doublets are not confusing the analysis. In some cases aneuploid and/or polyploid subpopulations may be present in cultures and revealed for the first time by direct live-cell staining methods. Changes in nuclear structure or chromatin compaction may act to prevent full reporting of cellular DNA content, and parallel staining of fixed or permeabilized samples with DRAQ5 can reveal such effects. In dual-beam systems, it is possible to acquire emissions following sequential 488-nm and 633-nm excitation and co-plot the data during acquisition. This dot plot should generate a linear relationship if there is a range of DNA
contents within the sample population. The plot will reveal whether a primary or secondary beam is out of alignment and whether anomalous populations or the effects of competing fluor signals are present in the DRAQ5 channels.
Anticipated Results Although results will depend on the particular application, Figures 7.25.1-7.25.5 provide some examples of DRAQ5 staining. Dualbeam analysis of live-cell binding of DRAQ5 versus time is shown in Figure 7.25.1. Flow cytometry demonstrates the ability of DRAQ5 to discriminate DNA content using either 488nm or 633-nm excitation. Data show the rapid establishment of DNA content distributions using DRAQ5, irrespective of excitation wavelength. Figure 7.25.2 is a comparison of DNA content distributions (488-nm excitation) derived from live and ethanol-fixed human B cell SU-DHL-4 lymphoma cells, showing the similarity in absolute fluorescence intensity and relative content distribution. Confocal laser scanning microscopy results are demonstrated in Figures 7.25.3-7.25.5.
Time Considerations The rapid staining kinetics of DRAQ5 permit direct introduction of the dye to samples at the flow cytometer and to samples already mounted for imaging. Thus protocols should be adapted to cope with the impact of time on other more critical aspects.
Nucleic Acid Analysis
7.25.9 Current Protocols in Cytometry
Supplement 28
Literature Cited Albota, M., Xu, C., and Webb, W. 1998. Two-photon fluorescence excitation cross sections of biomolecular probes from 690 to 960 nm. Appl. Opt. 37:7352-7356. Allman, R., Errington, R.J., and Smith, P.J. 2003. Delayed expression of apoptosis in human lymphoma cells undergoing low-dose taxol-induced mitotic stress. J. Cancer 88:1649-1658. Bell, D.H. 1988. Characterization of the fluorescence of the antitumor agent, mitoxantrone. Biochim. Biophys. Acta 949:132-137. Bestvater, F., Spiess, E., Strobrawa, G., Hacker, M., Fuerer, T., Porwol, T., Berchner-Pfannschmidt, U., Wotzlaw, C., and Acker, H. 2002. Two-photon fluorescence absorption and emission spectra of dyes relevant for cell imaging. J. Microsc. 208:108-115. Darzynkiewicz, Z. and Kapuscinski, J. 1990. Acridine orange: A versatile probe of nucleic acids and other cellular constituents. In Flow Cytometry and Sorting. 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 291314. Wiley-Liss, New York. Fitzgerald, K.A., Rowe, D.C., Barnes, B.J., Caffrey, D.R., Visintin, A., Latz, E., Monks, B., Pitha, P.M., and Golenbock, D.T. 2003. LPS-TLR4 signaling to IRF-3/7 and NF-κB involves the toll adapters TRAM and TRIF. J. Exp. Med. 198:1043-1055. Fox, M.E. and Smith, P.J. 1995. Subcellular localisation of the antitumour drug mitoxantrone and the induction of DNA damage in resistant and sensitive human colon carcinoma cells. Cancer Chemother. Pharmacol. 35:403-410. Frey, T. 1995. Nucleic acid dyes for detection of apoptosis in live cells. Cytometry 21:265-274. Frey, T., Yue, S., and Haugland, R.P. 1995. Dyes providing increased sensitivity in flow-cytometric dye-efflux assays for multidrug resistance. Cytometry 20:218-227. Latt, S.A. and Langlois, R.G. 1990. Fluorescent probes of DNA microstructure and DNA synthesis. In Flow Cytometry and Sorting. 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 249-290. Wiley-Liss, New York. Lown, J.W., Morgan, A.R., Yen, S.-F., Wang, Y.H., and Wilson, W.D. 1985. Characteristics of the binding of the anticancer agents mitoxantrone and ametantrone and related structures to deoxyribonucleic acids. Biochemistry 24:4028-4035. Mitta, B., Rimann, M., Ehrengruber, M.U., Ehrbar, M., Djonov, V., Kelm, J., and Fussenegger, M. 2002. Advanced modular self-inactivating lentiviral expression vectors for multigene interventions in mammalian cells and in vivo transduction. Nucleic Acids Res. 30:e113.
DRAQ5 Labeling of Nuclear DNA in Live and Fixed Cells
Plander, M., Brockhoff, G., Barlage, S., Schwarz, S., Rothe, G., and Knuechel, R. 2003. Optimization of three- and four-color multiparameter DNA analysis in lymphoma specimens. Cytometry 54A:66-74.
Schmitt, A., Gutierrez, G.J., Lenart, P., Ellenberg, J., and Nebreda, A.R. 2002. Histone H3 phosphorylation during Xenopus oocyte maturation: Regulation by the MAP kinase/p90Rsk pathway and uncoupling from DNA condensation. FEBS Lett. 518:23-28. Smith, P.J., Morgan, S.A., and Watson, J.V. 1991. Detection of multidrug resistance and quantification of responses of human tumour cells to cytotoxic agents using flow cytometric spectral shift analysis of Hoechst 33,342-DNA fluorescence. Cancer Chemother. Pharmacol. 27:445450. Smith, P.J., Blunt, N.J., Desnoyers, R., Giles, Y., and Patterson, L.H. 1997a. DNA topoisomerase IIdependent cytotoxicity of alkylaminoanthraquinones and their N-oxides. Cancer Chemother. Pharmacol. 39:455-461. Smith, P.J., Desnoyers, R., Patterson, L.H., and Watson, J.V. 1997b. Flow cytometric analysis and confocal imaging of anticancer alkylaminoanthraquinones and their N-oxides in intact human cells using 647-nm krypton laser excitation. Cytometry 27:43-53. Smith, P.J., Wiltshire, M., Davies, S., Patterson, L.H., and Hoy, T. 1999. A novel cell permeant and far red-fluorescing DNA probe, DRAQ5, for blood cell discrimination by flow cytometry. J. Immunol. Methods 229:131-139. Smith, P.J., Blunt, N., Wiltshire, M., Hoy, T., Teesdale-Spittle, P., Craven, M.R., Watson, J.V., Amos, W.B., Errington, R.J., and Patterson, L.H. 2000. Characteristics of a novel deep red/infrared fluorescent cell-permeant DNA probe, DRAQ5, in intact human cells analyzed by flow cytometry, confocal and multiphoton microscopy. Cytometry 40:280-291. Smith, P.J., Wiltshire, M., Davies, S., Chin, S.-F., Campbell, A.K., and Errington, R.J. 2002. DNA damage-induced [Zn(2+)](i) transients: Correlation with cell cycle arrest and apoptosis in lymphoma cells. Am. J. Physiol. Cell Physiol. 283:C609-C622. Snyder, D.S. and Garon, C.F. 2003. Decreased uptake of bodipy-labelled compounds in the presence of the nuclear stain, DRAQ5. J. Microsc. 211:208-211. Snyder, D.S. and Small, P.L. 2001. Staining of cellular mitochondria with LDS-751. J. Immunol. Methods 257:35-40. Terstappen, L.W., Shah, V.O., Conrad, M.P., Recktenwald, D., and Loken, M.R. 1988. Discriminating between damaged and intact cells in fixed flow cytometric samples. Cytometry 9:477-484. Terstappen, L.W., Meiners, H., and Loken, M.R. 1989. A rapid sample preparation technique for flow cytometric analysis of immunofluorescence allowing absolute enumeration of cell subpopulations. J. Immunol. Methods 123:103-112.
7.25.10 Supplement 28
Current Protocols in Cytometry
Van Zandvoort, M.A., de Grauw, C.J., Gerritsen, H.C., Broers, J.L., oude Egbrink, M.G., Ramaekers, F.C., and Slaaf, D.W. 2002. Discrimination of DNA and RNA in cells by a vital fluorescent probe: Lifetime imaging of SYTO13 in healthy and apoptotic cells. Cytometry 47:226-235.
http://www.cx.unibe.ch/dkf7/applications.html
Waggoner, A.S. 1990. Fluorescent probes for cytometry. In Flow Cytometry and Sorting. 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 209-225. Wiley-Liss, New York.
http://www.norakbioscience.com/PDF/AZSBS20 03.pdf
Weng, A.P., Nam, Y., Wolfe, M.S., Pear, W.S., Griffin, J.D., Blacklow, S.C., and Aster, J.C. 2003. Growth suppression of pre-T acute lymphoblastic leukemia cells by inhibition of notch signaling. Mol. Cell. Biol. 23:655-664.
These sites provide examples of fix-and-stain protocols using DRAQ5.
Wiltshire, M., Patterson, L.H., and Smith, P.J. 2000. A novel deep red/low infrared fluorescent flow cytometric probe, DRAQ5NO, for the discrimination of intact nucleated cells in apoptotic cell populations. Cytometry 39:217-223.
Internet Resources http://www.biostatus.co.uk/draq5_msds.html
The Flow-Cytometry Laboratory of the Department of Clinical Research, University of Bern, has provided a protocol for immunophenotyping that includes DRAQ5 staining.
http://www.norakbioscience.com/PDF/SBS03Tra nsfluor_Alpha.pdf
Contributed by Paul J. Smith, Marie Wiltshire, and Rachel J. Errington University of Wales College of Medicine Heath Park, Cardiff, United Kingdom Research supported by the UK Research Councils’ Basic Technology Research Programme Grant GR/S23483 and UK Biotechnology and Biological Sciences Research Council Grant SBRI19666.
This material safety data sheet provides health and safety information for DRAQ5.
Nucleic Acid Analysis
7.25.11 Current Protocols in Cytometry
Supplement 28
Assessment of Telomere Length, Phenotype, and DNA Content
UNIT 7.26
Telomeres are highly conserved repeats of hexameric nucleotide sequences at the ends of chromosomes, with important functions for chromosomal stability and replication. During cell division, cellular DNA is duplicated by the action of DNA polymerase; however, the polymerase does not completely replicate the 3 ends of chromosomes in the so-called telomeric region. As a consequence, telomeres get shortened during each cell division. Although mechanisms exist that slow telomere shortening, eventually somatic cells will exhaust their replication potential. Thus, telomere length can serve as an indicator of the replicative history of individual cells and is associated with cellular senescence. In recent years, various fluorescence in situ hybridization (FISH) protocols using telomere-specific peptide nucleic-acid (PNA) probes for the flow cytometric estimation of telomere length in individual cells (Flow-FISH) have been developed. DNA denaturation into single strands is achieved by heating a cell suspension to at least 80◦ C in a formamide-containing reaction mix and is followed by hybridization of the PNA telomere-specific probe to complementary DNA sequences. Utilization of flow cytometry’s greatest potential lies in simultaneous measurement of telomere length and cell surface antigen expression. Development of this technique has been difficult owing to the harsh conditions required for DNA denaturation for probe hybridization. Recently, however, protocols which overcome these obstacles have been generated. Heat-stable fluorochromes with suitable fluorescence emissions separate from the telomere fluorescence channel are used for antibody labeling. Antigen-antibody complexes are cross-linked onto the cell membrane before cells undergo probe hybridization. Telomere length can then be estimated in subsets of cells as defined by their cell surface antigen expression pattern. This unit provides a primary procedure for analyzing immunophenotype, telomere length, and DNA content (see Basic Protocol), along with a supplementary method for determining DNA ploidy (see Support Protocol).
STRATEGIC PLANNING The procedure described in the primary procedure (see Basic Protocol) is complex and requires basic understanding of cellular staining principles and flow cytometry. In addition, previous experience in setting up a flow cytometer for multicolor experiments will facilitate success. In planning a combined phenotype and telomere-length assay, it is necessary for the experimenter to consider the fluorochrome choices and the capabilities of the flow cytometer to be used for data acquisition, and to optimize the cell-surface and DNA staining protocol prior to performing the assay on samples to be studied. Only heatstable fluorochromes can be utilized in telomere Flow-FISH assays; thus, phycoerythrin or allophycocyanin is unsuitable because both are protein based. Possible fluorochrome choices include fluorescein isothiocyanate (FITC), cyanine dyes (e.g., Cy3, Cy5), and Alexa Fluor reagents (e.g., Alexa Fluor 488, Alexa Fluor 546, Alexa Fluor 647). It is critical for the reliability of the telomere length measurements that there be no fluorescence overlap of other fluorochromes into the channel used for detection of telomere fluorescence. Thus, the choice of label for the telomere-specific probe should be made before selecting the other colors. Because the conditions during FISH increase cellular autofluorescence and at the same time reduce the fluorescence intensity of the antibody labels used for cell surface staining, maximization of the specific fluorescence signal is critical. As monoclonal antibody clones directed against the same antigen can differ in their stability Nucleic Acid Analysis Contributed by Ingrid Schmid and Beth D. Jamieson Current Protocols in Cytometry (2004) 7.26.1-7.26.13 C 2004 by John Wiley & Sons, Inc. Copyright
7.26.1 Supplement 29
during Flow-FISH, it may be necessary to compare reagents from various sources. In addition, variations in reagent concentration, staining time and temperature, and the number of washes need to be investigated to optimize the brightness of the specific signal above background. To achieve sufficient signal amplification for reliable discrimination of cell subpopulations in the Flow-FISH assay, it may be necessary to utilize two-layer staining protocols that use either the streptavidin-biotin system or a species-specific second-step reagent. In addition, indirect staining protocols can result in better stabilization of cell surface antigen–antibody complexes, but they increase the staining time and potential cell losses from multiple washing steps. The selection of the appropriate nucleic-acid dye for staining will depend on its compatibility with the other fluorochromes that are used for detection of telomere-specific probe and cell surface antigens. Combinations of fluorochromes that have been used successfully for simultaneous measurement of phenotype, telomere length, and DNA content include Cy5, FITC, and propidium iodide (PI) for single-color cell surface staining (Batliwalla et al., 2001; Plunkett et al., 2001), and Alexa Fluor 488, Alexa Fluor 546, Cy5, and Hoechst 33342 (HO342) for measuring two cell-surface markers (Schmid et al., 2002). Nucleic-acid dyes are used at very low concentrations to minimize their interactions with the other fluorochromes. Titrations to find the lowest possible dye concentration that still produces adequate DNA histograms are advisable. Concentrations that have been used previously are 0.06 to 0.2 µg/ml for PI (Hultdin et al., 1998; Rufer et al., 1998), 0.06 µg/ml for 7-aminoactinomycin D (7-AAD; Rufer et al., 1998), and 0.3 µg/ml for HO342 (Schmid et al., 2002). BASIC PROTOCOL
ANALYSIS OF IMMUNOPHENOTYPE, TELOMERE LENGTH, AND DNA CONTENT This protocol describes a method for flow cytometric estimation of telomere length in cellular subsets. First, cells to be analyzed are stained for cell surface immunofluorescence. Then, antigen-antibody complexes are covalently cross-linked onto the cell surface using bis(sulfosuccinimidyl)suberate (BS3 ). Because Flow-FISH is a complex process that is sensitive to minor variations in experimental conditions, cells with long telomeres are added as an internal control to each reaction tube. Cell surface–labeled cells are mixed with unstained internal control cells that differ in telomere length from the cells of interest, resuspended in the reaction mix containing telomere-specific PNA probe, and heated 10 min at 80◦ C for DNA denaturation. Probe hybridization is performed overnight, followed by two washes at 40◦ C to remove probe that has bound nonspecifically. For determination of cellular autofluorescence, mixtures of the cells of interest and internal control cells are incubated with the hybridization solution without PNA probe. Accurate assessment of telomere length requires that only singlet cells containing one genome be considered. Thus, a nucleic-acid dye with a fluorescence emission separate from the emission of the telomere-specific probe is added to the Flow-FISH samples before sample acquisition to permit exclusion of proliferating cells and cell aggregates during data analysis. Duplicate samples are analyzed on the cytometer and the resulting telomere-specific mean fluorescence of G0/1 cells of the cell subset is then compared to the mean background fluorescence of G0/1 subset cells. At the same time, telomerespecific mean fluorescence of the G0/1 internal control cells over background fluorescence is determined.
Assessment of Telomere Length, Phenotype, and DNA Content
The amount of DNA per cell influences telomere fluorescence, because it directly correlates with the number of chromosomes and telomere ends. Consequently, relative DNA contents (DNA ploidy) of the cells of interest and internal control cells need to be reliably assessed. To improve the precision of DNA ploidy measurements, separate sample aliquots are processed in parallel to Flow-FISH using an optimized nucleic-acid staining and analysis protocol as outlined below (see Support Protocol).
7.26.2 Supplement 29
Current Protocols in Cytometry
As the final step, telomere fluorescence values obtained from Flow-FISH are then normalized for DNA content, and telomere length of the cell subset in the sample is expressed as a fraction of the telomere length of the internal control cells.
Materials PBS-washed cells Staining solution (see recipe) PBS (APPENDIX 2A), room temperature and 40◦ C Cross-linking solution (see recipe) 1 M Tris·Cl, pH 8.0 (APPENDIX 2A) Internal control cells, cryopreserved (see recipe) Hybridization buffer (see recipe) with and without fluorescently labeled, telomere-specific PNA probe, (C3 TA2 )3 (PerSeptive Biosystems or equivalent) Nucleic-acid stain compatible with other fluorochromes used (e.g., PI, 7-AAD, Hoechst 33342), in PBS 12 × 75–mm polystyrene tubes 1.5-ml polypropylene tubes 80◦ C shaking water bath 40◦ C water bath or heating block Flow cytometer with 488-nm blue excitation and 633-nm red excitation, and optional UV excitation depending on fluorochrome choices made during strategic planning of the experiments Additional reagents and equipment for staining cells (UNIT 6.2), counting cells (APPENDIX 3A), and standardizing flow cytometers (UNIT 1.3) Stain cells for cell surface immunofluorescence 1. Place ∼2.5–3 × 106 PBS-washed cells into a 12 × 75–mm polystyrene tube and resuspend in 250 µl staining solution. Perform cell-surface staining procedure (either direct or indirect) appropriate for the antigen(s) to be studied (see Support Protocol and UNIT 6.2). If cells are peripheral blood mononuclear cells (PBMCs) separated by Ficoll-Hypaque density gradient, make sure that there is no visible red-cell contamination, as it has been reported that erythrocytes interfere with FITC telomere fluorescence (see Critical Parameters and Troubleshooting). Residual red cells can be lysed by an ∼2 min treatment with ammonium chloride lysing solution (APPENDIX 2A) at 20◦ to 25◦ C. Starting cell numbers can be lower than the numbers indicated above, as cell recovery will vary depending whether a direct or an indirect staining method is used. However, it is necessary to start with sufficient cells, taking into account that 0.25 × 106 stained cells per reaction are required for Flow-FISH. Initially, follow manufacturer’s recommendation for the appropriate antibody concentration but optimize staining conditions to obtain the maximal signal separation between background and specific staining for Flow-FISH (see UNIT 4.1 for performing antibody titrations). Preparation of a sample stained with the appropriate isotype-matched control(s) is essential for determination of background staining.
2. After the last washing step, resuspend the cell pellet in 200 µl PBS. Slowly add 200 µl cross-linking solution dropwise. Incubate 30 min at 2◦ to 8◦ C protected from light, then add 8 µl of 1 M Tris·Cl, pH 8.0. Mix well and incubate 15 min at 20◦ to 25◦ C. BS3 is a highly reactive cross-linker that easily hydrolizes; therefore, it is critical that the BS3 solution be prepared immediately before use (see Critical Parameters and Troubleshooting). The reaction buffers must not contain any primary amines because they
Nucleic Acid Analysis
7.26.3 Current Protocols in Cytometry
Supplement 29
compete with the desired cross-linking action on cell surface antigen–antibody complexes. Tris buffer added at the end of the first incubation period provides an excess of primary amines to stop cross-linking.
3. Add 1 ml PBS, centrifuge cells 5 min at 300 × g, 2◦ to 8◦ C, and remove supernatant. Add 1 ml PBS. Count cells (APPENDIX 3A). Cells have to be counted carefully before fluorescence in situ hybridization (FISH) to account for cell losses during staining and cross-linking, because accurate cell numbers are critical for reliable hybridization reactions.
Denature DNA and perform FISH 4. Thaw an aliquot of cryopreserved cells from the batch to be used as an internal standard and perform a cell count (APPENDIX 3A). 5. Mix 1 × 106 cell surface–stained cells with 1 × 106 thawed internal control cells. Place 0.5 × 106 cells of mixture into a 1.5-ml polypropylene tube. Add PBS to 1 ml, mix well, and centrifuge cells 5 min at 300 × g, 2◦ to 8◦ C. Completely remove supernatant. Internal control cells are added to the reaction tubes in equal numbers compared to the cells to be studied to compensate for variations in the numerous factors that influence FISH of the telomere probe to DNA (see Critical Parameters and Troubleshooting). Samples are prepared in duplicate to reduce assay variability. Two tubes are needed for hybridization without probe, and two for hybridization with telomere-specific probe. Note that tubes have to be made of polypropylene to withstand the high temperature required for DNA denaturation. Tubes larger than 1.5 ml can be used, but as reaction conditions (i.e., evaporation of reaction mix) are influenced by tube airspace, scatter parameters will be altered.
6. Add 300 µl hybridization buffer with or without fluorescently labeled, telomerespecific probe. Mix well and incubate 10 min at 80◦ C in a shaking water bath protected from light. CAUTION: Note that formamide is a toxic substance. Observe proper precautions and perform reactions in a fume hood. Tubes containing hybridization mix without probe are needed for determination of background fluorescence. The appropriate telomere probe concentration has to be determined by titration. Concentrations of 4, 10, and 55 nM have been reported as optimal depending on the type of Flow-FISH procedure used (see Critical Parameters and Troubleshooting). Temperature needs to be maintained with great precision, because 80◦ C is considered the minimum temperature for proper DNA denaturation. Alternatively, tubes can be heated in a heating block; however, temperature differences between individual tubes will be more pronounced than in a shaking water bath (also see Critical Parameters and Troubleshooting).
7. Take the tubes out of the water bath and incubate overnight at 20◦ to 25◦ C, protected from light. 8. The next day, add 1 ml PBS, 40◦ C, to the tubes which were hybridized overnight. Mix well and incubate 10 min at 40◦ C. Wash by centrifugation 5 min at 500 × g, 20◦ to 25◦ C. Repeat once. Assessment of Telomere Length, Phenotype, and DNA Content
Stain for DNA content 9. Add 0.5 ml appropriate nucleic acid stain compatible with other fluorochromes used in PBS to all tubes. Mix well and incubate samples at least 2 to 3 hr at 20◦ to 25◦ C protected from light.
7.26.4 Supplement 29
Current Protocols in Cytometry
Which dye will be most suitable for any given experiment will depend on the other fluorochromes used for phenotyping and for detection of telomere fluorescence (see Strategic Planning, Critical Parameters and Troubleshooting, and Anticipated Results).
Analyze samples on the flow cytometer 10. Run samples on a flow cytometer with light sources and filters appropriate for the fluorochromes used. Standardize the flow cytometer for each experiment using standard particles and procedures as described in UNIT 1.3. 11. Obtain data using the following parameters: a. Collect cell-surface fluorescence and telomere fluorescence with log amplification, and DNA fluorescence with linear amplification using area and width signals for doublet discrimination. Log amplification of the telomere-specific signal is needed for simultaneous measurement of sample and internal standard.
b. Use a low sample differential during sample acquisition to reduce the coefficients of variation of DNA fluorescence measurements. 12. Analyze samples using plots and gating strategies as described (see Anticipated Results).
DNA PLOIDY DETERMINATION The protocol describes a method for assessment of relative DNA contents of cells of interest and internal control cells using a nucleic-acid staining procedure that has been optimized for high-resolution DNA analysis.
SUPPORT PROTOCOL
Materials Unstained cells Thawed internal control cells PBS (APPENDIX 2A) PI staining solution Prepare and analyze samples 1. On the day of the original experiment, place 0.5 × 106 unstained cells to be studied and 0.5 × 106 freshly thawed internal control cells into two separate tubes. 2. Wash once with 1 ml PBS by centrifugation 5 min at 300 × g, 2◦ to 8◦ C. Add 0.5 ml propidium iodide staining buffer to the cell pellets. Mix and incubate 15 min at 2◦ to 8◦ C protected from light. Alternately, perform a staining protocol for flow cytometric DNA analysis as described in Basic Protocol 1 of UNIT 7.5. Whenever analyzing malignant cell samples that may be aneuploid (e.g., samples from leukemic patients), it is advisable to include an internal DNA ploidy standard consisting of a mixture of chicken and trout erythrocytes (see Basic Protocol 2 in UNIT 7.5).
3. Acquire samples for DNA ploidy on a flow cytometer using the following parameters:
Blue excitation Detection of PI emission at linear amplification at orange-red wavelength Constant detector voltage and gain between sample runs PI area and width signal processing for doublet discrimination Low sample differential setting. The last parameter is used to obtain low coefficients of variation on DNA fluorescence measurements.
Nucleic Acid Analysis
7.26.5 Current Protocols in Cytometry
Supplement 29
4. Analyze the resulting DNA histograms with appropriate DNA histogram deconvolution software for an accurate determination of the relative positions of the G0/1 peaks.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Cross-linking solution Store bis(sulfosuccinimidyl)suberate (BS3 ; Pierce) desiccated at −20◦ C in powder aliquots suitable for the experiment sizes planned. Prepare BS3 solution immediately before use by taking out one vial of BS3 powder. Wait until the vial has warmed to ambient room temperature before opening. Add 5 mM sodium citrate buffer, pH 5, to a final concentration of 4 mM (w/v) BS3 . Hybridization buffer 70% (v/v) deionized formamide (APPENDIX 2A) 1% (w/v) BSA 10 mM Tris·Cl, pH 7.2 (APPENDIX 2A) Store 6 months in a tightly sealed container protected from light at 2◦ to 8◦ C Internal control cells Use cells with long telomeres—e.g., 1301 cells (European Collection of Cell Cultures; http://www.ecacc.org.uk) or mouse spleen cells. Generate a large batch of cells and cryopreserve by liquid-nitrogen storage of aliquots suitable for the typical experiment size planned (APPENDIX 3B). PI staining solution 0.1 g sodium citrate 0.3 ml Triton X-100 10 mg propidium iodide (PI) 2 mg DNase-free ribonuclease A (APPENDIX 2A) Adjust volume to 100 ml with H2 O Store 6 months in a tightly sealed container protected from light at 2◦ to 8◦ C Staining solution PBS without Ca2+ and Mg2+ (APPENDIX 2A) 0.1% (w/v) sodium azide 2% (w/v) newborn calf serum (e.g., Omega Scientific) Adjust pH to 7.2 with 0.1 M NaOH or 0.1 M HCl Store up to 6 months at 20◦ to 25◦ C COMMENTARY Background Information Assessment of Telomere Length, Phenotype, and DNA Content
Telomeres represent important regulatory elements for controlling how often somatic cells can divide. Although telomere shortening is counteracted by the activity of the enzyme telomerase and an alternate lengthening of telomeres pathway (Cech, 2003), a slow
but inexorable decrease in a cell’s replicative potential accompanies cell division. Thus, the average telomere length of a given cell population relates to its replication potential and is linked to cellular senescence (Effros et al., 1996, 2003). Owing to the pivotal role of telomere dynamics in tumor formation, aging,
7.26.6 Supplement 29
Current Protocols in Cytometry
and immune exhaustion pertinent to infectious disease control, interest in telomere biology has increased in recent years and has spurred the development of novel technologies for assessing telomere length. The traditional method of measuring telomere length is the determination of the mean terminal restriction fragment (TRF) size of genomic DNA by Southern blot analysis. This technique is robust but labor intensive and time consuming. In addition, the restriction enzymes used for TRF analysis cut the DNA in subtelomeric regions; as a consequence, nontelomeric sequences are included in the DNA fragments and make the measurement less accurate. Fluorescently labeled peptide nucleic-acid (PNA) probes hybridize with complementary oligonucleotide sequences at the low ionic strength unfavorable for re-annealing of the target strands and form very stable duplexes with DNA. Availability of these PNA probes facilitated the development of novel fluorescence in situ hybridization (FISH) protocols for quantitation of telomeric repeat sequences (Lansdorp et al., 1996). FISH measurements of telomere length correlate well with TRF analysis; however, a linear regression line comparing the two methods is not expected to pass through the origin, because subtelomeric sequences do not interfere with telomere-specific FISH as hexameric T2 AG3 repeats are rare except at the ends of chromosomes. Rather, the calculated line should cross the Southern blot axis at a point corresponding to the approximate mean of the subtelomeric DNA length of the samples tested (Hultdin et al., 1998; Law and Lau, 2001b; Roos and Hultdin, 2001; Schmid et al., 2002). The FISH method for estimation of telomere length was first performed by digital imaging microscopy on metaphase spreads of chromosomes and was termed QFISH, but sample preparation is complicated, lengthy, and requires a high degree of technical expertise. It is the method of choice, however, to analyze telomeric sequences on specific chromosomes. In contrast, subsequent Flow-FISH protocols (Hultdin et al., 1998; Rufer et al., 1998) that utilize rapid detection of fluorescence signals of individual cells in suspension, although complex, are less technically demanding, use standard analytic flow cytometers, are able to provide results within 24 hr, and can be combined with phenotypic analysis of cells. Nevertheless, because the conditions required for DNA denaturation during FISH are detrimental to standard staining of cell surface antigens,
until recently, analysis of telomere length in cell subpopulations had to be performed on cells that were separated prior to Flow-FISH either by cell sorting (Batliwalla et al., 2000; Rufer et al., 1999) or by magnetic bead separations (Rufer et al., 1998; Son et al., 2000). Isolating cell subpopulations by these means is time consuming and can be costly and difficult, particularly if the frequency of the cells of interest is low and when large numbers of samples have to be separated. Lately, novel methods that are able to stabilize antigen-antibody complexes on the cell surface either by covalent cross-linking (Batliwalla et al., 2001; Schmid et al., 2002) or by fixation (Plunkett et al., 2001) permit simultaneous measurement of telomere length and cell surface markers.
Critical Parameters and Troubleshooting Sample preparation The Flow-FISH method for measuring telomere length is a complex protocol with various aspects needing careful consideration for a successful experiment (Lauzon et al., 2000). The primary method (see Basic Protocol) has been successfully used on PBMC isolated by Ficoll-Hypaque density centrifugation. Whole-blood samples cannot be readily used as it has been shown that the hemoglobin contained in erythrocytes can interfere with measuring telomere length when using a FITClabeled probe, possibly by quenching probe fluorescence (Baerlocher et al., 2002); therefore, complete removal of red cells is important, and sample hematocrit should not exceed 2%. Immunophenotyping The harsh experimental conditions required for proper DNA denaturation for binding of the telomere-specific probe provided a major obstacle for the generation of reliable protocols for concomitant cell surface staining. Commonly used bright fluorochromes such as phycoerythrin (PE), allophycocyanin (APC), or tandem fluorochromes containing PE or APC are protein based and therefore destroyed by heat. In contrast, FITC and cyanine dyes are heat stable. Batliwalla et al. (2001) have used Cy5 as a fluorochrome for detection of cell surface antigens and developed a novel strategy for covalent cross-linking antigen-antibody complexes onto the cell surface membrane with bis(sulfosuccinimidyl)suberate (BS3 ). This became necessary because standard fixation with formaldehyde or commercial reagent
Nucleic Acid Analysis
7.26.7 Current Protocols in Cytometry
Supplement 29
systems such as Fix & Perm (Caltag) or FACSLysing solution (BD Biosciences) are insufficient to maintain surface fluorescence. BS3 , the water-soluble, noncleavable crosslinker utilized in the Basic Protocol, is a homobifunctional N-hydroxysuccinimide ester which predominantly reacts with the ε-amino group of lysine. BS3 is more stable in solutions with low pH, as hydrolysis increases under alkaline conditions. Nevertheless, because of its inherent instability, the generation of BS3 stock solutions is not recommended; rather, aliquots of the powder need to be stored at −20◦ C in a desiccator for retention of crosslinking activity. BS3 is excluded from intact cells and most importantly, does not hinder probe access to the DNA. Availability of heat-stable, bright Alexa Fluor–type fluorochromes has opened up the possibility to combine two cell surface markers with telomere Flow-FISH (Schmid et al., 2002). Fluorochrome brightness is important because after Flow-FISH, cell-surface fluorescence intensity still decreases markedly despite BS3 cross-linking, while cellular autofluorescence increases. Furthermore, maintenance of cell-surface fluorescence varies between different fluorochromes and antigenantibody complexes. Thus, application of the assay described in the Basic Protocol to measure telomere length in a specific cellular subset identified by dual-color staining is rather complex. In the published protocol (Schmid et al., 2002), which provides the basis for the Basic Protocol described in this unit, the authors employed an indirect staining method that, although time-consuming, permitted the detection of positive cells above background with frequency similar to that for samples that were not subjected to Flow-FISH. Nevertheless, it is possible that for staining of other antigens, directly-conjugated monoclonal antibodies are suitable. For dual-color cell-surface staining the emission spectra of both fluorochromes have to be separate, but can overlap as long as fluorescence compensation can be applied; however, neither can overlap into the channel used for detection of telomere fluorescence, as compensation of the overlap would alter the position of the telomere-specific fluorescent peak.
Assessment of Telomere Length, Phenotype, and DNA Content
Flow-FISH While Hultdin et al. (1998) exposed cells to a fixation and permeabilization step before Flow-FISH, the protocol by Rufer et al. (1998) showed that sufficient access of the telomere-specific probe can be achieved with-
out a permeabilization step. The small size of the probe and the reaction conditions used for Flow-FISH (i.e., formamide and heat) appear to cause adequate cell membrane permeability for probe access to DNA. One central experimental condition for Flow-FISH is the complete denaturation of double-stranded DNA into single strands by heat and deionized formamide for proper telomere-specific PNA probe hybridization. Early protocols exposed cell preparations to a temperature of 80◦ C and a concentration of formamide of 70% for 10 min (Hultdin et al., 1998; Rufer et al., 1998). These conditions have subsequently been used successfully by others (Son et al., 2000; Law and Lau, 2001b; Plunkett et al., 2001; Schmid et al., 2002). Recently, however, it has been indicated that temperatures of 85◦ to 87◦ C may provide further optimization of the DNA denaturation step (Baerlocher et al., 2002). Nevertheless, as higher temperatures create extremely unfavorable conditions for cells, these considerations have to be balanced against maintenance of cellular integrity and preservation of cell surface antigens in cases where subset analysis of cell populations is desired. Furthermore, tubes placed into a shaking water bath will reach the desired temperature faster and with less time and temperature difference between individual tubes than tubes placed into a heating block and therefore will require less overall heat exposure. The proper concentration of the PNA telomere-specific probe and hybridization conditions are important experimental considerations. Methods that were published by the two independent laboratories which have pioneered the Flow-FISH technique use markedly divergent probe concentrations and hybridization times for their individual protocols. Rufer et al. (1998) used 55 nM telomere-specific probe and 2 hr of hybridization, while Hultdin et al. (1998) found that overnight hybridization with a probe concentration of 4 nM was optimal for their assay. Some of these differences may be due to the cell fixationpermeabilization step and the addition of a cell line with long telomeres as an internal standard to each reaction tube by Hultdin et al. (1998). In their assay, and in the procedure described in the Basic Protocol, telomere probe is also taken up by the standard. Thus, conditions that provide optimal separation of telomere fluorescence between sample and standard may differ from those most suited for reactions without an internal standard. In a published protocol for dual-color cell surface staining and telomere length measurements (Schmid et al., 2002)
7.26.8 Supplement 29
Current Protocols in Cytometry
that also utilizes an internal standard, the authors selected a concentration of 10 nM for a Cy5-labeled telomere-specific probe as optimal for Flow-FISH. For a given probe concentration, cell numbers have to be kept extremely constant between sample tubes, because specific binding is highly dependent on cell concentration. Recent data from Baerlocher et al. (2002) show that the hybridization reaction can reach a plateau after ∼2 hr, but Hultdin et al. (1998) and Schmid et al. (2002) have successfully used overnight hybridization. Hybridization must be followed by sufficient washing steps (at least two) to remove unbound and nonspecifically bound PNA probe. While most published procedures include formamide in their washing solutions, a commercially available kit for Flow-FISH (DakoCytomation) uses formamide-free washing steps at 40◦ C, and the authors have adopted this in the published procedure (Schmid et al., 2002; also see Basic Protocol). The type of fluorochrome used for telomerespecific probe labeling can influence the ability to detect specific probe fluorescence above the high green autofluorescence generated by the Flow-FISH procedure (Mosiman et al., 1997). Using Cy5, which emits red fluorescence away from the parts of the spectrum with the highest contributions to autofluorescence, improves measurement of telomere fluorescence (Schmid et al., 2002). DNA staining Cells tend to form aggregates during the Flow-FISH procedure. Clumped cells interfere with accurate determination of telomere fluorescence and need to be gated out on plots of DNA fluorescence width versus area during data analysis. Furthermore, only cells in G0/1 which have not started to increase their DNA content can be used for an accurate determination of telomere length, as Hultdin et al. (1998) have shown that replication of telomere sequences can already begin in the early S-phase of the cell cycle. The addition of a nucleic-acid dye before sample acquisition on the cytometer can address both these problems. Dyes are added at much lower concentrations than the ones commonly used for DNA staining to minimize their interference with telomere fluorescence due to quenching or energy transfer (Baerlocher et al., 2002). Longer than usual incubation times are necessary to reach adequate staining; nevertheless, coefficients of variation of DNA distributions are not as low as with standard
staining methods and dye concentrations. Dyes are selected depending on the fluorochrome used for the detection of telomere fluorescence and cannot have notable fluorescence emissions overlapping into the telomere fluorescence channel to avoid interference with the correct determination of the mean telomere fluorescence. Assay and instrument standardization Flow-FISH of telomere length depends on the reliability of the generation and measurement of a fluorescent signal for which minor alterations or erratic or systematic errors in the procedure can lead to relatively large changes in the readout. Thus, for improvement of the accuracy of the assay, addition of an internal standard that controls for the variations between individual Flow-FISH reaction tubes is important. Hultdin et al. (1998) have introduced the use of the 1301 cell line, a subline of CCRF-CEM, as an internal standard. 1301 cells have extremely long telomeres (>25 kb), are near tetraploid, and therefore can be easily distinguished from any human sample on the flow cytometer. In principle, any cells that exhibit these features are suitable. For instance, mouse spleen cells that are known to have long telomeres could also be a useful standard. Hultdin et al. (1998) expressed the telomere length of the samples of interest as a fraction of the telomere length of the internal standard to allow for a direct comparison between samples. To compensate for differing DNA content between the samples and the 1301 cells (i.e., diploid versus tetraploid), they measured their DNA indices and normalized the telomere values for their DNA content. This is critical, because DNA content directly correlates to the number of chromosomes and telomere ends (Hultdin et al., 1998; Roos and Hultdin, 2001; Law and Lau, 2001a). DNA histograms obtained with the nucleic-acid counterstain in Flow-FISH samples are suboptimal owing to the low concentration of the stain; therefore, it is advisable to perform the determination of DNA indices using a method optimized for the determination of DNA content (Hultdin et al., 1998; Schmid et al., 2002) such as the one published by Fried et al. (1980; see Support Protocol and also protocols described in UNIT 7.5). The reliability of the fluorescent signal from Flow-FISH is also highly dependent on flow cytometer quality control and standardization (UNIT 1.3). When the telomere signal is processed linearly as described by Rufer et al. (1998) and for DNA ploidy
Nucleic Acid Analysis
7.26.9 Current Protocols in Cytometry
Supplement 29
determinations, it is advisable to check the linearity of the fluorescent scale with appropriate beads, such as DNA check beads available from BD Biosciences. When the telomere fluorescence is processed logarithmically (Hultdin et al., 1998), the correct behavior of the log amplifier must be verified by using commercially available beads and methods (UNIT 1.3), or published methods (Schmid et al., 1988). If the instrument shows deviations from expected results, adjustments have to be made by service personnel before samples are run on the cytometer. In addition, daily calibration using standard particles has to be performed to correct for day-to-day instrument variations. Data reporting and assay validation Rufer et al. (1998) reported their data in terms of arbitrary fluorescence units or molecule equivalents of soluble fluorochrome (MESF) units (Rufer et al., 1999), while Hultdin et al. (1998) used the ratio of the fluorescent values of the experimental sample and the internal standard. For a ready comparison of data obtained in different laboratories, however, it is necessary to generate a linear regression line between the Flow-FISH technique and the TRF fragment size in kilobases as measured by Southern blotting (Hultdin et al., 1998; Law and Lau, 2001b; Schmid et al., 2002). This can be achieved by parallel analysis of samples of various telomere length—e.g., human samples from individuals that differ in age or cell lines that vary in telomeres. After the correlation between TRF values and Flow-FISH data has been established, the correlation equation can then be applied to subsequent samples that are processed with the same Flow-FISH method and analyzed on the same flow cytometer.
Anticipated Results
Assessment of Telomere Length, Phenotype, and DNA Content
Figures 7.26.1 and 7.26.2 show the application of the procedure described in the Basic Protocol to the determination of telomere length in CD28 subsets of CD8+ cells in human PBMCs. CD28 is a molecule expressed on the cell surface of most T cells. It is known to be transiently downregulated after activation (Azuma et al., 1993), and an irreversible loss of CD28 on CD8+ cells has been associated with aging (Valenzuela and Effros, 2000). CD8+ CD28− cells have a diminished capacity for in vitro proliferation, raising the possibility that some of these cells may have reached a state of replicative senescence (Effros et al., 1996; Monteiro et al., 1996). Shortening of
telomeres in the CD8+ CD28− subset has been previously reported using various methods (Monteiro et al., 1996; Effros et al., 1996; Batliwalla et al., 2000). Thus, application of the method described here (see Basic Protocol) is expected to demonstrate differences in telomere length between CD8+ CD28+ and CD8+ CD28− human lymphocytes. In the example shown in Figure 7.26.1 an indirect staining strategy was used for dual-color cell-surface labeling utilizing purified CD8 antibody followed by goat anti-mouse Fab 2 Alexa Fluor 546 and biotinylated CD28 followed by streptavidin Alexa Fluor 488. Cy5labeled telomere-specific probe for FISH requires red excitation but emits around 660 nm, separate from both cell-surface fluorochromes, and was used at a concentration of 10 nM. Ultraviolet-excitable HO342 was selected for counterstaining DNA, because its blue emission does not interfere with Cy5 emission or with the emissions from the Alexa Fluor fluorochromes. The 1301 cell line was used as internal control. 1301 cells have more forward and side scatter; thus, they appear in the upper right corner of the scatter plot (Fig. 7.26.1A, gate R2), and are gated separately from lymphocytes in the lower left corner (Fig. 7.26.1A, gate R1). Next, singlet resting lymphocytes are discriminated from doublets and proliferating cells on a plot of HO342 fluorescence width versus area (Fig. 7.26.1B, gate R3) and are displayed on a plot of Alexa Fluor 488 versus Alexa Fluor 546 fluorescence for identification of CD8bright+ CD28+ and CD8bright+ CD28− subsets (Fig. 7.26.1C). Cy5 telomere fluorescence above background for each CD8+ subset is shown in plots D and E, respectively. Telomere fluorescence of singlet 1301 control cells within gates R2 and R4 is displayed in plot F. The geometric mean of the telomere fluorescence is lower in CD8bright+ CD28− cells than in CD8bright+ CD28+ cells, indicating that in the sample analyzed the average telomere length in CD28− cells is indeed shorter. DNA histograms to be used for DNA ploidy analysis are shown in Figure 7.26.2 and were generated following the procedure outlined above (see Support Protocol). The DNA index of the 1301 line is calculated by dividing the mean of the G0/1 peak of the cell line by the mean of the sample peak. The value obtained is used as the correction factor for the differences in DNA ploidy of the control cell line 1301 compared to the sample (see Critical Parameters and Troubleshooting). Calculation of sample telomere length as a percentage of the telomere length of 1301 control cells (%TL) then
7.26.10 Supplement 29
Current Protocols in Cytometry
Figure 7.26.1 Fresh human PBMCs were processed as described (see Basic Protocol) and detailed (see Anticipated Results). (A) Forward- versus side-scatter dot plot, ungated; (B) HO342 width versus area dot plot, gated on R1 or R2; (C) CD28 Alexa Fluor 488 versus Alexa Fluor 546 dot plot, gated on R1 and R3; (D) histogram overlay of Cy5 telomere-specific fluorescence of cells falling within gate R1, gate R3, and gate R5 over background fluorescence; (E) histogram overlay of Cy5 telomere-specific fluorescence of cells falling within gate R1, gate R3, and gate R6 over background fluorescence; (F) histogram overlay of 1301 cells falling within gate R2 and gate R4 over background fluorescence. RFI: relative mean fluorescence intensity.
follows the equation:
Equation 7.26.1
where tel. fluor. is the geometric mean of the telomere fluorescence of the sample, bkg is the geometric mean of the sample background, DI1301 is the DNA index of 1301 cells, tel.
fluor.1301 is the geometric mean of the telomere fluorescence of 1301 cells, bkg1301 is the geometric mean of the 1301 background, and DI is the DNA index of the sample. When this equation is applied to the current example, CD8bright+ CD28− cells have a mean telomere length of 7.7% compared to 10.1% for the CD8bright+ CD28+ cells. Conversion of these results into TRF values requires that each laboratory determine the linear regression line by parallel measurement of samples by
Nucleic Acid Analysis
7.26.11 Current Protocols in Cytometry
Supplement 29
Figure 7.26.2 Data shown are derived from parallel processing of the same human PBMCs and 1301 control cells that were used for the determination of telomere length for determination of the DNA indices as described in the Support Protocol. Relative positions of the singlet G0/1 peaks indicated are derived from histogram deconvolution using Modfit LT software, which utilizes a scale with 256-channel resolution.
Flow-FISH and TRF analysis as discussed under Critical Parameters and Troubleshooting. In the authors’ laboratory the correlation equation is determined as follows (Schmid et al., 2002): TRF size (kb) = % TL × 0.77 + 2.02. Application of this equation to the results from data shown in Figure 7.26.1 indicates that on average, telomeres in CD8bright+ CD28− cells are 2 kb shorter than those of CD8bright+ CD28+ cells.
Time Considerations Ficoll-Hypaque separation of PBMCs will take ∼2 hr, thawing of frozen samples ∼1 hr. The time required for cell surface staining will depend on the number of antigens to be labeled, whether a direct or an indirect staining protocol is used, and how many samples are processed. Thus, expect staining to take from <1 hr up to 3 hr, and cross-linking ∼1 hr. Preparing and counting the samples and the internal control cells for Flow-FISH, including DNA denaturation, takes ∼45 min. Determination of the DNA indices requires ∼30 min. The washing steps performed the next day take ∼45 min and DNA staining will be completed in 2 to 3 hr. Instrument standardization and setup will vary according to the protocols and flow cytometer used. At a minimum, expect sample acquisition to take at least 5 min per sample, as a low sample differential has to be used to optimize DNA content measurements. Assessment of Telomere Length, Phenotype, and DNA Content
Acknowledgement This work was performed in the UCLA Jonsson Comprehensive Cancer Center and Center for AIDS Research Flow Cytometry Core Facility and was supported by National
Institutes of Health awards CA-16042 and AI28697.
Literature Cited Azuma, M., Phillips, J.H., and Lanier, L.L. 1993. CD28− T lymphocytes: Antigenic and functional properties. J.Immunol. 150:11471159. Baerlocher, G.M., Mak, J., Tien, T., and Lansdorp, P.M. 2002. Telomere length measurement by fluorescence in situ hybridization and flow cytometry: Tips and pitfalls. Cytometry 47:8999. Batliwalla, F.M., Rufer, N., Lansdorp, P.M., and Gregersen, P.K. 2000. Oligoclonal expansions in the CD8+ CD28− T cells largely explain the shorter telomeres detected in this subset: Analysis by flow FISH. Hum. Immunol. 61:951958. Batliwalla, F.M., Damle, R.N., Metz, C., Chiorazzi, N., and Gregersen, P.K. 2001. Simultaneous flow cytometric analysis of cell surface markers and telomere length: Analyis of human tonsilar B cells. J. Immunol. Methods 247:103109. Cech, T.R. 2003. Beginning to understand the end of the chromosome. Cell 116:273-279. Effros, R.B., Allsopp, R., Chiu, C.-P., Hausner, M.A., Hirji, K., Wang, L., Harley, C.B., Villeponteau, B., West, M.D., and Giorgi, J.V. 1996. Shortened telomeres in the expanded CD28− CD8+ cell subset in HIV disease implicate replicative senescence in HIV pathogenesis. AIDS 10:F17-F22. Effros, R.B., Dagarag, M., and Valenzuela, H.F. 2003. In vitro senescence of immune cells. Exp. Gerontol. 38:1243-1249. Fried, J., Perez, A.G., and Clarkson, B. 1980. Quantitative analysis of cell cycle progression of synchronous cells by flow cytometry. Exp. Cell Res. 126:63-74.
7.26.12 Supplement 29
Current Protocols in Cytometry
Hultdin, M., Gr¨onlund, E., Norrback, K.-F., Eriksson-Lindstr¨om, E., Just, T., and Roos, G. 1998. Telomere analysis by fluorescence in situ hybridization and flow cytometry. Nucl. Acids Res. 26:3651-3656. Lansdorp, P.M., Verwoerd, N.P., van de Rijke, F.M., Dragowska, V., Little, M.T., Dirks, R.W., Raap, A.K., and Tanke, H.J. 1996. Heterogeneity in telomere length of human chromosomes. Hum. Mol. Genet. 5:685-691. Lauzon, W., Dardon, J.S., Cameron, D.W., and Badley, A.D. 2000. Flow cytometric measurement of telomere length. Commun. Clin. Cytom. 42:159-164.
Rufer, N., Br¨ummendorf, T.H., Kolvraa, S., Bischoff, C., Christensen, K., Wadsworth, L., Schulzer, M., and Lansdorp, P.M. 1999. Telomere fluorescence measurements in granulocytes and T lymphocyte subsets point to a high turnover of hematopoietic stem cells and memory T cells in early childhood. J. Exp. Med. 190:157-167. Schmid, I., Schmid, P., and Giorgi, J.V. 1988. Conversion of logarithmic channel numbers into relative linear fluorescence intensity. Cytometry 9:533-538.
Law, H.K.W. and Lau, Y.L. 2001a. DNA index and Q flow FISH measurement of telomere length. Cytometry 45:80.
Schmid, I., Dagarag, M.D., Hausner, M.A., Matud, J.L., Just, T., Effros, R.B., and Jamieson, B.D. 2002. Simultaneous flow cytometric analysis of two cell surface markers, telomere length, and DNA content. Cytometry 49:96-105.
Law, H.K.W. and Lau, Y.L. 2001b. Validation and development of quantitative flow cytometrybased fluorescence in situ hybridization for intercenter comparison of telomere length measurement. Cytometry 43:150-153.
Son, N.H., Murray, S., Yanovski, J., Hodes, R.J., and Weng, N.P. 2000. Lineage-specific telomere shortening and unaltered capacity for telomerase expression in human T and B lymphocytes with age. J. Immunol. 165:1191-1196.
Monteiro, J., Batliwalla, F.M., Ostrer, H., and Gregersen, P.K. 1996. Shortened telomeres in clonally expanded CD28− CD8+ T cells imply a replicative history that is distinct from their CD28+ CD8+ counterparts. J. Immunol. 156:3587-3590.
Valenzuela, H. and Effros, R.B. 2000. Telomere measurement and replicative senescence. In Aging: Methods and Protocols (Y.A. Barnett and C.R. Barnett, eds.) pp. 23-52. Humana Press, Totowa, NJ.
Mosiman, V.L., Patterson, B.K., Canterero, L., and Goolsby, C.L. 1997. Reducing cellular autofluorescence in flow cytometry: An in situ method. Cytometry 30:151-156.
Schmid et al., 2002. See above.
Plunkett, F.J., Soares, M.V.D., Annels, N., Hislop, A., Ivory, K., Lowdell, M., Salmon, M., Rickinson, A., and Akbar, A.N. 2001. The flow cytometric analysis of telomere length in antigenspecific CD8+ T cells during acute Epstein-Barr virus infection. Blood 97:700-707. Roos, G. and Hultdin, M. 2001. Flow cytometric determination of telomere length. Cytometry 45:79.
Key Reference Describes the application of the procedure presented in the Basic Protocol to the simultaneous measurement of two cell surface markers and telomere length.
Contributed by Ingrid Schmid and Beth D. Jamieson David Geffen School of Medicine University of California Los Angeles Los Angeles, California
Rufer, N., Dragowska, W., Thornbury, G., Roosnek, E., and Lansdorp, P.M. 1998. Telomere length dynamics in human lymphocyte subpopulations measured by flow cytometry. Nat. Biotech. 16:743-747.
Nucleic Acid Analysis
7.26.13 Current Protocols in Cytometry
Supplement 29
Detection of Histone H2AX Phosphorylation on Ser-139 as an Indicator of DNA Damage (DNA Double-Strand Breaks)
UNIT 7.27
This unit describes a method based on immunocytochemical detection of phosphorylated histone H2AX for revealing the presence of DNA double-strand breaks (DSBs). Concurrent measurement of DNA content and multivariate analysis of such data make it possible to correlate the presence of DSBs in individual cells with the cell cycle phase. The presence of DSBs reveals DNA damage induced by ionizing radiation (Sedelnikova et al., 2002; MacPhail et al., 2003a,b; Yoshida et al., 2003) or by treatment with antitumor drugs, such as DNA topoisomerase inhibitors (Huang et al., 2003, 2004). It should be noted, however, that DSBs can also be intrinsic (programmed), occurring in healthy nontreated cells, e.g., in the course of V(D)J and class-switch recombination during immune system development (Downs et al., 2000; Jackson, 2001; Sedelnikova et al., 2003) or in S-phase cells in association with DNA replication (McPhail et al., 2003a). In addition, DSBs are generated in the course of DNA fragmentation in apoptotic cells (Huang et al., 2003, 2004). This unit presents strategies to distinguish radiation- or drug-induced DSBs from intrinsic formation of DSBs in untreated cells or from apoptosis-associated DSBs. Histone H2AX is one of several variants of the nucleosome core histone H2A family (West and Bonner, 1980). Induction of DSBs in live cells triggers its phosphorylation (Rogakou et al., 1998). The phosphorylation is mediated by Ataxia teleangiectasia–mutated (ATM; Burma et al., 2001), ATM-Rad3-related (ATR; Furuta et al., 2003), and/or DNAdependent protein kinase (DNA-PK; Park et al., 2003), affects H2AX molecules flanking the DSBs in chromatin, and occurs on Ser-139 at the C terminus (Rogakou et al., 1998). The phosphorylated form of H2AX has been defined as γH2AX (Rogakou et al., 1999). Monoclonal or polyclonal antibodies that detect γH2AX are now commercially available (e.g., Upstate Biotechnology, Trevigen). The appearance of γH2AX in chromatin can be detected immunocytochemically, shortly after induction of DSBs, in the form of discrete nuclear foci, each presumed to represent a single DSB (Sedelnikova et al., 2002). Checkpoint and DNA repair proteins such as Rad50, Rad51, and Brca1 colocalize with γH2AX (Paull et al., 2000). In addition, the translocation of the p53 binding protein 1 (53BP1) to irradiation-induced foci is mediated by H2AX (Anderson et al., 2001). Because the loss of H2AX in mice leads to genomic instability, this histone is considered to be one of the guardians of genome integrity (Celeste et al., 2003). The intensity of γH2AX immunofluorescence from individual cells exposed to ionizing radiation was reported to correlate with the dose of irradiation used to induce DSBs (MacPhail et al., 2003b). In fact, the γH2AX immunofluorescence (γH2AX IF) measured by cytometry offers a sensitive and convenient means to detect the presence and measure the frequency of DSBs, and has been proposed as a surrogate for cell killing in viability tests for drugs that generate DSBs (Banath and Olive, 2003). Because, as mentioned, DSBs are also formed during DNA repair that involves nucleotide excision repair (NER) and nonhomologous end joining (NHEJ), analysis of γH2AX IF may report the extent of the repair process even when the primary drug-induced lesions are not DSBs, but the latter are formed during DNA repair (Huang et al., 2004).
Nucleic Acid Analysis Contributed by Xuan Huang, H. Dorota Halicka, and Zbigniew Darzynkiewicz Current Protocols in Cytometry (2004) 7.27.1-7.27.7 C 2004 by John Wiley & Sons, Inc. Copyright
7.27.1 Supplement 30
BASIC PROTOCOL
DETECTION OF H2AX PHOSPHORYLATED ON SER-139 IN RELATION TO THE CELL CYCLE PHASE This protocol describes the immunocytochemical detection of histone H2AX phosphorylated on Ser-139 (γH2AX), combined with measurement of DNA content, to identify the cells that have DNA double-strand breaks (DSBs) and concurrently assess their cell cycle phase. The detection of γH2AX is based on indirect immunofluorescence using a secondary antibody tagged with green-fluorescing FITC, while DNA is counterstained with red-fluorescing propidium iodide (PI). Cellular RNA, which otherwise may also be stained by PI, is removed using RNase A, included in the PI solution. Untreated control cells and cells treated with a drug, a carcinogen, or radiation to induce DSBs are briefly fixed in methanol-free formaldehyde and then transferred to 70% ethanol, in which they can be stored at −20◦ C for up to two weeks. While the ethanol post-fixation makes the plasma membrane permeable to the γH2AX antibody, additional permeabilization is achieved by inclusion of the detergent Triton X-100 when the cells are incubated with the antibody. After incubation with the primary γH2AX antibody followed by the FITC-labeled secondary antibody, cellular DNA is counterstained with PI in the presence of RNase A, and the cellular green and red fluorescence is measured by flow cytometry. The mean values of green fluorescence intensity for subpopulations of G1 , S, and G2 M cells (selected by differences in intensity of their red PI fluorescence) of the control (untreated) cultures, which represent the intrinsic presence of DSBs, are then subtracted from the respective means of the cells from drug-, carcinogen-, or radiation-treated cultures. These differential values of the mean green fluorescence intensity correspond to the drug-, carcinogen-, or radiation-induced H2AX phosphorylation of the G1 , S, or G2 /M cells, respectively.
Materials Cells Medium Agents expected to induce DSBs Phosphate-buffered saline (PBS; APPENDIX 2A) 1% methanol-free formaldehyde (Polysciences) in PBS, 4◦ C (store ≤2 weeks at 4◦ C) 70% ethanol, 0◦ to 4◦ C 1% (w/v) BSA and 0.2% (v/v) Triton X-100 in PBS (BSA-T-PBS; store ≤2 weeks at 4◦ C) Unconjugated primary γH2AX antibody: murine monoclonal anti–histone γH2AX antibody (Upstate Biotechnology) or rabbit polyclonal anti–histone γH2AX antibody (Trevigen) FITC-conjugated secondary antibody, appropriately titered: e.g., polyclonal goat anti-mouse or anti-rabbit-F(ab )2 , depending on the source of the primary antibody Propidium iodide (PI) staining solution: 5 µl/ml PI (Molecular Probes) and 100 µl/ml DNase-free RNase A (Sigma) in PBS (store ≤2 weeks protected from light at 4◦ C) 12 × 75–mm polypropylene tubes Flow cytometer with blue-light (e.g., 488-nm argon-ion laser or BG-12 excitation filter) excitation source and filters for collection of green (530 ± 20 nm) and red (>600 nm) fluorescence Histone H2AX Phosphorylation on Ser-139 as an Indicator of DNA Damage
Fix cells 1. In 12 × 75–mm polypropylene tubes prepare 1-ml suspensions of 1–5 × 106 cells in medium, either untreated (control) or treated with agent(s) expected to induce DSBs.
7.27.2 Supplement 30
Current Protocols in Cytometry
2. Centrifuge suspensions 4 min at 300 × g, room temperature. Decant supernatant and resuspend pellet in 0.5 ml PBS. 3. Using a Pasteur pipet, transfer the cell suspension into a tube containing 4.5 ml icecold 1% methanol-free formaldehyde in PBS. Keep cells in this fixative 15 min on ice. 4. Centrifuge cells 4 min at 300 × g, room temperature. Decant supernatant and resuspend pellet in 5 ml of 70% ethanol, 0◦ to 4◦ C. Incubate at least 2 hr (up to 2 weeks) at −20◦ C (e.g., in a freezer). 5. Centrifuge cells 4 min at 200 × g, room temperature. Decant supernatant and resuspend cell pellet in 2 ml BSA-T-PBS. 6. Centrifuge cells 4 min at 300 × g, room temperature. Decant supernatant and resuspend cell pellet in 2 ml BSA-T-PBS. Incubate 5 min at room temperature.
Stain cells with γH2AX antibody 7. Centrifuge cells 4 min at 300 × g, room temperature. Decant supernatant and resuspend cell pellet in 100 µl BSA-T-PBS containing 1 µg unconjugated primary γH2AX antibody. See Critical Parameters regarding antibody titration.
8. Cap the tubes to prevent drying and keep them overnight at 4◦ C. 9. Add 2 ml BSA-T-PBS and centrifuge 4 min at 300 × g, room temperature. 10. Decant supernatant and resuspend cell pellet in 2 ml BSA-T-PBS. 11. Centrifuge cells 4 min at 300 × g, room temperature. Decant supernatant and resuspend cell pellet in 100 µl BSA-T-PBS containing the appropriate FITC-conjugated secondary antibody. The secondary antibody will be anti-mouse or -rabbit, depending on the source of the primary antibody. See Critical Parameters regarding antibody titration.
12. Incubate 1 hr at room temperature with occasional gentle shaking. Add 5 ml BSA-TPBS and after 2 min, centrifuge 4 min at 300 × g, room temperature.
Counterstain DNA 13. Decant supernatant and resuspend cell pellet in 1 ml PI staining solution. Incubate 30 min in the dark at room temperature. 14. Set up and adjust the flow cytometer for excitation with light at blue wavelength (488 nm) and collection of green (530 ± 20 nm) and red (>600 nm) fluorescence. 15. Measure intensity of green and red fluorescence of the cells by flow cytometry. Record the data.
COMMENTARY Background Information Cytometry of γH2AX immunofluoresence As mentioned, DNA damage that leads to induction of DSBs in live cells triggers histone H2AX phosphorylation. Immunocytochemical detection of γH2AX is thus a marker of the presence of DSBs, while intensity of γH2AX immunofluorescence (γH2AX IF) is
considered to represent frequency of DSBs in the nucleus (Sedelnikova et al., 2002). This approach is much more sensitive and rapid, and much less cumbersome, compared to the alternative and most commonly used method to detect and measure DNA damage, namely the comet assay (Speit and Hartman, 1999). Flow or laser scanning cytometry offers
Nucleic Acid Analysis
7.27.3 Current Protocols in Cytometry
Supplement 30
Figure 7.27.1 Bivariate distributions of γH2AX expression in relation to cellular DNA content (cell cycle position) of untreated HL-60 cells (Ctrl) and cells cultured with 0.15 µM topotecan (TPT), a DNA topoisomerase I inhibitor, in the absence and presence of 10 nM calyculin A (Cal) for 1 or 3 hr (see Huang et al., 2004, for details). Notice increased expression of γH2AX after 1 hr treatment with TPT compared to control but no apparent effect of Cal. After 3 hr of culture, however, the expression of γH2AX (apparent in all phases of the cell cycle) is much higher in the cells treated with calyculin A. The dashed horizontal line represents the γH2AX IF level below which >95% of cells from untreated cultures (Ctrl) express γH2AX.
the possibility to rapidly quantify γH2AX IF in large cell populations, while multiparameter analysis of the data makes it possible to correlate DNA damage with other attributes of the cell. The present protocol describes the concurrent measurement of γH2AX IF and DNA content, which allows one to correlate DNA damage with cell position in the cell cycle. Specifically, subpopulations of cells in G1 versus S versus G2 /M can be selected by the differences in DNA content. By gating analysis, the mean γH2AX IF for each of these subpopulations can then be estimated (Fig. 7.27.1). It should be noted that the protocol presented in this unit can be modified to measure other cell attributes, such as surface immunophenotype or expression of a particular intracellular protein, in order to correlate them with DNA damage. Examples of other attributes that were measured concurrently with γH2AX IF are presented elsewhere (Huang et al., 2004).
Histone H2AX Phosphorylation on Ser-139 as an Indicator of DNA Damage
Changes in histone content during cell cycle Histone content doubles during the cell cycle, along with the doubling of DNA content. Unlike other proteins, however, whose content may vary in individual cells with respect to DNA content, histone synthesis is coupled with DNA synthesis and therefore the ratio of histone to DNA content remains invariable throughout the cell cycle for all cells (Marzluff and Duronio, 2002). As a consequence of their
higher histone content with the same degree of H2AX phosphorylation (i.e., the same percentage of phosphorylated H2AX molecules within the total number of H2AX molecules), the cells in S and G2 /M have 1.5× and 2.0× higher γH2AX IF, respectively, compared to G1 cells. To assess the degree of H2AX phosphorylation, and thus to make γH2AX IF independent of histone doubling during the cycle, the data may be normalized by presenting per unit of DNA (histone). This is accomplished by multiplying the mean S-phase and G2 /M-phase γH2AX IF by 0.75 and 0.5, respectively. After such normalization, one can assess whether the treatment induced a proportionally greater number of DSBs per unit of DNA at a particular phase of the cell cycle. In other words, the compensation for the change in total H2AX content during the cell cycle reveals the degree of H2AX phosphorylation, i.e., the relative ratio of phosphorylated H2AX per total number of H2AX molecules within each cell. Intrinsic versus induced H2AX phosphorylation A low level of γH2AX IF is observed in cells that are not treated with any inducers of DSBs. This represents an intrinsic (scheduled, programmed) histone H2AX phosphorylation, which appears to be primarily associated with DNA replication and thus is expressed most in S-phase cells (MacPhail
7.27.4 Supplement 30
Current Protocols in Cytometry
et al., 2003b). The level of this intrinsic γH2AX IF varies among different cell lines. For quantitative analysis of the γH2AX IF induced by external factors that damage DNA, this intrinsic γH2AX IF has to be subtracted. Towards this end, the mean values of γH2AX IF for G1 , S, and G2 /M subpopulations of untreated cells are subtracted from the respective means of the G1 , S, and G2 /M subpopulations of carcinogen-, drug-, or radiationtreated cells, respectively (Huang et al., 2004). After such subtraction, the increase in γH2AX IF (γH2AX IF) represents the treatmentinduced phosphorylation of H2AX. There is no need, therefore, to have the isotype control to estimate nonspecific antibody binding, because most likely such binding is the same for the untreated and treated cells, and is subtracted while calculating γH2AX IF.
H2AX phosphorylation versus dephosphorylation The degree of H2AX phosphorylation measured at a given time point after induction of DNA damage represents a balance between the rate of phosphorylation of histone H2AX initially induced by the damage and the dephosphorylation that occurs when DNA repair progresses. Dephosphorylation is mediated by a serine/threonine phosphatase which can be inhibited, e.g., by calyculin A, a phosphatase inhibitor (Nazarov et al., 2003; see also Fig. 7.27.1). The data in Figure 7.27.1 show that upon induction of DNA damage by topotecan (TPT), a DNA topoisomerase l inhibitor, γH2AX IF of cells incubated 1 hr in the presence or absence of calyculin A was similar, whereas after 3 hr the intensity of γH2AX IF was significantly lower in its absence. These data thus indicate that while H2AX dephosphorylation was undetectable during the first hour of the treatment, a significant degree of dephosphorylation, prevented by calyculin A, must occur between 1 and 3 hr. Therefore, to obtain a measure of the cumulative H2AX phosphorylation (as a yardstick for the total number of DSBs) in response to progressive DNA damage (e.g., as occurs during continuous cell treatment with a DNA-damaging drug), one has to incubate the cells in the presence of the protein phosphatase inhibitor to prevent γH2AX dephosphorylation. It should be noted, however, that calyculin A is cytotoxic and prolonged (>3 hr) cell incubation with this inhibitor leads to extensive chromatin condensation followed by apoptosis (Huang et al., 2004).
Distinction between drug- (or radiation-)induced and apoptosis-associated H2AX phosphorylation DNA undergoes extensive fragmentation during apoptosis as a result of activation of endonuclease(s) by caspase(s) (Gorczyca et al., 1992; Earnshaw et al., 1999). It is often desirable to distinguish between drug(or radiation-)induced primary DSBs and DSBs that are generated during the process of apoptosis. The following attributes of γH2AF IF allow one to distinguish the cells with drug(or radiation-)induced H2AX phosphorylation from cells that have additional phosphorylation of this histone triggered by DNA fragmentation during apoptosis (discussed further in Huang et al., 2004): 1. The drug- (or radiation-)induced γH2AX IF is seen very early (during the initial two hours) in the course of treatment, i.e., well prior to caspase-3 activation, the prerequisite factor for apoptotic endonuclease activation and DNA fragmentation (Earnshaw et al., 2003). 2. The intensity of drug-induced γH2AX IF is initially several-fold lower than the intensity of apoptosis-associated γH2AX IF (Huang et al., 2004). It should be noted, however, that because the intensity of the apoptosis-associated γH2AX IF decreases at late stages of apoptosis, this attribute may then fail to discriminate between drug- (or radiation-)induced and apoptosis-associated γH2AX IF. 3. The apoptosis-associated DNA fragmentation that triggers phosphorylation of H2AX may be prevented by cell treatment with the caspase inhibitor z-VAD-FMK. In its presence, thus, only the drug- (or radiation-)induced γH2AX is being detected (Huang et al., 2004). 4. The apoptosis-associated H2AX phosphorylation occurs in parallel with the concurrent activation of caspase-3 in the same cells. Multiparameter analysis of activated caspase3 versus γH2AX IF, thus, is the most direct approach to distinguish cells in which DSBs were induced by the studied agents (caspase-3 activation is undetectable) from cells that have H2AX phosphorylation additionally triggered in response to apoptotic DNA fragmentation (activated caspase-3 is present). Induction of DSBs during DNA repair Phosphorylation of H2AX may be observed following cell treatment with agents, such as cisplatin (cp) or UV light, which do not induce DSBs as the primary lesions, but rather cause DNA cross-linking or thymidine dimers (Huang et al., 2004). The repair process of
Nucleic Acid Analysis
7.27.5 Current Protocols in Cytometry
Supplement 30
these lesions, however, involves the nucleotide excision repair (NER) mechanism, known to generate single-strand (ss) DNA breaks. A fraction of ssDNA breaks, in turn, is known to be converted in the cell to DSBs (Vilenchik and Knodson, 2003). Furthermore, DNA repair may also involve nonhomologous end joining (NHEJ; Crul et al., 2003). This mechanism may additionally contribute to formation of DSBs and to the observed H2AX phosphorylation. Thus, when the primary drug-induced lesions do not involve DSBs, but the breaks are formed during DNA repair, as in the case of cp, analysis of H2AX phosphorylation may report extent of the repair process.
Critical Parameters and Troubleshooting Immunocytochemical detection of phosphorylated H2AX requires proper cell fixation and permeabilization to have the epitope of γH2AX preserved and accessible to the antibody. Prolonged cell storage at room temperature prior to fixation should be avoided. Quality of the primary and secondary antibody is of particular importance. Their ability to detect γH2AX is often lost, owing to improper transport or storage conditions. It has been the authors’ experience that few batches of the antibodies delivered by the vendor were defective. With the first use of every new batch of primary and secondary antibody, it is recommended to perform a serial dilution (e.g., over the range between 0.2 and 2.0 µg/100 µl) to find the optimal titer for detection of γH2AX. The titer recommended by the vendor is not always the optimal one. As mentioned, because γH2AX dephosphorylation may occur when repair of DNA damage progresses, the intensity of γH2AX IF may not always be a reflection of cumulative formation of DSBs induced by the drug, radiation, or mutagen. Parallel cultures in which cells were incubated in the absence and presence of a phosphatase inhibitor (e.g., calyculin A) allow one to estimate the extent of γH2AX dephosphorylation.
Anticipated Results
Histone H2AX Phosphorylation on Ser-139 as an Indicator of DNA Damage
Figure 7.27.1 illustrates the typical bivariate distributions of γH2AX IF in relation to cellular DNA content (cell cycle position) of untreated cells, as well the cells treated with a DNA-damaging drug (the DNA topoisomerase I inhibitor topotecan, TPT) in the absence and presence of calyculin A. By gating analysis, the mean values of γH2AX IF for cell populations in G1 , S, and G2 /M may be estimated in control,
TPT− , and TPT+ calyculin A–treated cultures. While no significant effect of calyculin A was seen after 1 hr, after 3 hr, the mean γH2AX IF of cells exposed to TPT and calyculin A was several-fold higher compared to cells treated with TPT alone. This indicates that during the time interval between 1 and 3 hr of incubation with TPT, the rate of H2AX dephosphorylation exceeded that of phosphorylation.
Time Considerations The minimal time of fixation is 2 hr. The overall time required for cell handling, incubations, and centrifugations after removal of the cells from fixative is ∼2 hr, excluding overnight incubation with the primary antibody. To accelerate the procedure, instead of overnight incubation at 4◦ C, the cells may be incubated 1 hr with the primary antibody at room temperature; it has been noticed, however, that somewhat better results are obtained with overnight incubation at 4◦ C. Cytometric analysis of a single sample takes ∼1 to 2 min and the data analysis may take up to 5 min per sample.
Literature Cited Anderson, L., Henderson, C., and Adachi, Y. 2001. Phosphorylation and rapid relocalization of 53BP1 to nuclear foci upon DNA damage. Mol. Cell Biol. 21:1719-29. Banath, J.P. and Olive, P.L. 2003. Expression of phosphorylated of histone H2AX as a surrogate of cell killing by drugs that create DNA doublestrand breaks. Cancer Res. 63:4347-50. Burma, S., Chen, B.P., Murphy, M., Kurimasa, A., and Chen, D.J. 2001. ATM phosphorylates histone H2AX in response to DNA double-strand breaks. J. Biol. Chem. 276:42462-67. Celeste, A., Difilippantonio, S., FernandezCapetillo, O., Pilch, D.R., Sedelnikova, O., Eckhaus, M., Ried, T., Bonner, W.M., and Nussenzweig, A. 2003. H2AX haploinsufficiency modifies genomic stabilty and tumor susceptibility. Cell 114:371-83. Crul, M., van Waardenburg, R.C.A.M., Bocxe, S., van Eijndhoven, M.A.J., Pluim, D., Beijnen, J.H., and Schellens, J.H.M. 2003. DNA repair mechanisms involved in gemcitabine cytotoxicity and in the interaction between gemcitabine and cisplatin. Biochem. Pharmacol. 65:275-282. Earnshaw, W.C., Martins, L.M., and Kaufmann, S.H. 1999. Mammalian caspases: Structure, activation, substrates and functions during apoptosis. Annu. Rev. Biochem. 68:383-424. Furuta, T., Takemura, H., Liao, Z.-Y., Aune, G.J., Redon, C., Sedelnikova, O.A., Pilch, D.R., Rogakou, E.P., Celeste, A., Chen, H.T., Nussenzweig, A., Aladjem, M.I., Bonner, W.M., and Pommier, Y. 2003. Phosphorylation of histone H2AX and activation of Mre11, Rad50, and
7.27.6 Supplement 30
Current Protocols in Cytometry
Nbs1 in response to replication-dependent DNA double-strand breaks induces by mammalian topoisomerase I cleavage complexes. J. Biol. Chem. 278:20303-20312. Gorczyca, W., Bruno, S., Darzynkiewicz, R.J., Gong, J., and Darzynkiewicz, Z. 1992. DNA strand breaks occurring during apoptosis: Their early in situ detection by the terminaldeoxynucleotidyl transferase and nick translation and prevention by serine protease inhibitors. Int. J. Oncol. 1:639-648. Huang, X., Okafuji, M., Traganos, F., Luther, E., Holden, E., and Darzynkiewicz, Z. 2004. Assesment of histone H2AX phosphorylation induced by DNA topoisomerase I and II inhibitors topotecan and mitoxantrone and by DNA cross-linking agent cisplatin. Cytometry 58A:99-110. Huang, X., Traganos, F., and Darzynkiewicz, Z. 2003. DNA damage induced by DNA topoisomerase I- or topoisomerase II- inhibitors detected by histone H2AX phosphorylation in relation to the cell cycle phase and apoptosis. Cell Cycle 2:614-619. Jackson, S.P. 2001. DNA damage signaling and apoptosis. Biochem. Soc. Transactions 29:65561. MacPhail, S.H., Banath, J.P., Yu, Y., Chu, E., and Olive, P.L. 2003a. Cell cycle-dependent expression of phosphorylated histone H2AX: Reduced expression in unirradiated but not X-irradiated G1-phase cells. Radiat. Res. 159:759-67. MacPhail, S.H., Banath, J.P., Yu, T.Y., Chu, E.H., Lambur, H., and Olive, P.L. 2003b. Expression of phosphorylated histone H2AX in cultured cell lines following exposure to X-rays. Int. J. Radiat. Biol. 79:351-358. Marzluff, W.F. and Duronio, R.J. 2002. Histone mRNA expression: Multiple levels of cell cycle regulation and important developmental consequences. Curr. Opin. Cell Biol. 14:692-699. Nazarov, I.B., Smirnova, A.N., Krutilina, R.I., Svetlova, M.P., Solovjeva, L.V., Nikiforov, A.A., Oei, S.-L., Zalenskaya, I.A., You, P.M., Bradbury, E.M., and Tomilin, N.V. 2003. Dephosphorylation of histone γ-H2AX during repair of DNA double-strand breaks in mammalian cells and its inhibition by calyculin A. Radiat. Res. 160:309317.
Paull, T.T., Rogakou, E.P., Yamazaki, V., Kirchgesser, C.U., Gellert, M., and Bonner, W.M. 2000. A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr. Biol. 10:886-985. Rogakou, E.P., Boon, C., Redon, C., and Bonner, W.M. 1999. Megabase chromatin domains involved in DNA double-strand breaks in vivo. J. Cell Biol. 146:905-916. Rogakou, E.P., Pilch, D.R., Orr, A.H., Ivanova, V.S., and Bonner, W.M. 1998. DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J. Biol. Chem. 273:5858-5868. Sedelnikova, O.A., Pilch, D.R., Redon, C., and Bonner, W.M. 2003. Histone H2AX in DNA damage and repair. Canc. Biol. Ther. 2:233-235. Sedelnikova, O.A., Rogakou, E.P., Panuytin, I.G., and Bonner W. 2002. Quantitive detection of 125IUdr-induced DNA double-strand breaks with γ-H2AX antibody. Radiat. Res. 158:486492. Speit, G. and Hartmann A. 1999. The comet assay (single cell gel test). Meth. Mol. Biol. 113:203212. Vilenchik, M.M. and Knudson, A.G. 2003. Endogenous DNA double-strand breaks: Production, fidelity of repair, and induction of cancer. Proc. Natl. Acad. Sci. U.S.A. 100:12871-12876 West, M.H. and Bonner, W.M. 1980. Histone 2A, a heteromorphous family of eight protein species. Biochemistry 19:3238-3245. Yoshida, K., Yoshida, S.H., Shimoda, C., and Morita, T. 2003. Expression and radiationinduced phosphorylation of H2AX in mammalian cells. J. Radiat. Res. (Tokyo), 44:47-51.
Contributed by Xuan Huang, H. Dorota Halicka, and Zbigniew Darzynkiewicz Brander Cancer Research Institute New York Medical College Valhalla, New York
Park, E.J., Chan, D.W., Park, J.H., Oettinger, M.A., and Kwon, J. 2003. DNA-PK is activated by nucleosomes and phosphorylated H2AX within the nucleosomes in an acetylation-dependent manner. Nucleic Acids Res. 31:6819-6827.
Nucleic Acid Analysis
7.27.7 Current Protocols in Cytometry
Supplement 30
CHAPTER 8 Molecular Cytogenetics INTRODUCTION
S
ince the development of methods for preparing and staining chromosomes, cytogenetic approaches have made significant contributions to the understanding of the chromosomal basis of constitutional diseases (e.g., Down’s syndrome) and acquired diseases (e.g., cancer). Besides clinical applications aimed at diagnosis, prevention, and therapy of disease, cytogenetic techniques have been used to monitor the biological effect of mutagenic and carcinogenic agents. The classical cytogenetics methodology of identifying quinacrine- or Giemsa-banded metaphase chromosomes and their abnormalities using a standard microscope, and creating a karyogram using photomicrography and “scissors and paste,” has been influenced by two developments. First, automated microscopy, in combination with digital image processing and pattern recognition as applied in commercial systems, allows geneticists to automatically find metaphases and identify human chromosomes with an accuracy of ~95%. In addition, image manipulation techniques and high-quality printers have greatly simplified the presentation of the resulting karyograms. Second, almost synchronously with the achievements in automation, techniques were developed to specifically stain defined nucleic acid sequences. Although these so-called hybridization methods had already been widely applied in molecular biology using radioactive reporter molecules, the introduction of fluorescence in situ hybridization (FISH) represented a breakthrough for the study of intact chromosomes (and also interphase cells). The improved resolution and flexibility of FISH combined with its still-improving sensitivity enabled this technique to replace autoradiography in many areas of research and clinical diagnosis. The Human Genome Project has been an important stimulant for the field. Techniques such as the polymerase chain reaction (PCR), chromosome sorting, microdissection, DNA sequencing, and synthesis methods (some of which may reasonably be considered spinoff technologies of the genome project) have greatly facilitated the production of nucleic acid probes specific for chromosomes or genes. It is only a matter of time before thousands of DNA probes carrying known (gene) sequences, either as recombinant clones or as chemically synthesized oligonucleotides, become widely available. Use of such reagents in multicolor FISH strategies, combined with appropriate (user-friendly) microscope imaging and analysis, will further stimulate important contributions to basic and clinically applied genetics and cell biology. Thus conventional cytogenetics based on the morphological analysis of chromosomes with a resolution determined by the size of one band (approximately a few megabases) exists side by side with so-called molecular cytogenetics, at present a complementary technique but in some cases already competing. A key element in molecular cytogenetics is in situ hybridization, which bridges the gap between molecular and classical cytogenetics. As has recently been shown (Schröck et al., 1996), by using 24 probes with different specificity, each labeled with spectrally identifiable fluorophore combinations, molecular cytogenetics is able to reduce the complex recognition and classification of the 24 human chromosomes to “simple” color identification. In the near future, conventional banding, which is based on differences in chromatin compactness or base composition, may be replaced by an artificially created color bar code by using region-specific probes created Contributed by Hans J. Tanke Current Protocols in Cytometry (2003) 8.0.1-8.0.3 Copyright © 2003 by John Wiley & Sons, Inc.
Molecular Cytogenetics
8.0.1 Supplement 26
synthetically or by microdissection. In addition to analysis of metaphase chromosomes, as in conventional cytogenetics, FISH can also be applied to the analysis of interphase cells, and as such has already given rise to the new discipline of interphase cytogenetics. Inevitably, current FISH techniques will reach their methodological endpoints and new hybridization and imaging principles will have to emerge. Examples are microarrays of individual or appropriately pooled DNA clones that replace human chromosomes in comparative genomic hybridization (CGH), thereby providing better detection sensitivity than now achievable in the analysis of compact metaphase chromosomes by light microscopy. Ultimately, microarrays containing vast numbers of oligonucleotides of defined sequences may be routinely used to sequence genes of interest. Moreover, there is reason enough to believe that new developments in molecular cytogenetics will continue to arise in the future. This chapter aims at defining the current possibilities of molecular cytogenetics with respect to its methodological aspects. An overview of the FISH methodology is provided by UNIT 8.1; this is followed by detailed discussion of basic preparation techniques (for chromosomes and interphase cells) in UNIT 8.2 and probe labeling for the various FISH methods in UNIT 8.3. UNIT 8.4 presents the two main directions for visualizing hapten-labeled probes used in molecular cytogenetics: immunoenzymatic staining for bright-field microscopy and direct/indirect immunofluorescent staining for fluorescence microscopy. This unit also introduces the recently developed peroxidase-driven tyramide signal amplification (TSA) system. One of the first and still important applications of FISH relates to interphase cytogenetics. UNIT 8.5 describes methods of preparation and staining, as well as strategies for studying aneuploidy in solid tumors on the basis of interphase cell analysis. For those needing to generate probes for use in special applications, UNIT 8.6 presents general background and methods for generating FISH probes using chromosome sorting and microdissection. To meet the increasing interest in simultaneous detection of genes and their transcripts—i.e., mRNA and cognate protein—UNIT 8.7 gives strategies and methods for combined genotyping and phenotyping. UNIT 8.13 also describes combined FISH and immunofluorescence analysis. It includes, however, an adaptation of the automated fluorescence image analysis that provides automatic scanning, detection, quantification, cell selection, storage, and relocation options. Recently, new hybridization strategies have been developed to increase the specificity and sensitivity of hybridization. UNIT 8.8, the first in a series of upcoming contributions on this topic, describes the use of padlock probes for single-nucleotide sequence discrimination, and discusses the potential for applying rolling-circle principles for signal amplification. UNIT 8.9 is the first in a series of contributions that will deal with methods for amplification
of FISH signals. This unit describes how tyramide signal amplification (TSA) systems, originally introduced in immunohistochemistry, can be successfully used for enhancement of in situ hybridization signals in cytogenetics. To improve the spatial resolution of FISH analysis, DNA molecules can be unraveled to the Watson and Crick helix, and stretched on glass slides. The technique is called “Fiber-FISH.” UNIT 8.10 describes an advance of this technology, called “molecular combing,” which provides controlled and uniform stretching of DNA molecules and facilitates quantitative genome analysis.
Introduction
8.0.2 Supplement 26
Current Protocols in Cytometry
The continuing need for improved detection sensitivity has led to the development of various labeling and signal amplification assays for molecular cytogenetic analysis. A versatile method is the PRINS reaction (PRimed IN Situ labeling), a technique that traditionally has been used for the detection of tandemly repeated target sequences by oligonucleotide probes. UNIT 8.11 describes protocols for basic PRINS methods and for combining PRINS and FISH. Originally introduced by Kallioniemi, Pinkel, and colleagues in 1992, comparative genomic hybridization (CGH) has become a powerful research tool to detect gains and losses in gene copy number such as occur in most tumors. The original method has been significantly refined, leading to better sensitivity and reproducibility. A conceptually new approach has been the introduction of array CGH, in which the target chromosomes are replaced with locus-specific probes spaced every 1-2 Mbp covering the genome. Both variants of CGH are carefully described in UNIT 8.12. Future updates to this chapter will cover further special applications such as PRINS of single-copy genes. As outlined above, new methodology and applications are expected in the coming years, thereby guaranteeing continuous updating of this chapter with useful protocols. LITERATURE CITED Schröck, E., du Manoir, S., Veldman, T., Schoell, B., Wienberg, J., Ferguson-Smith, M.A., Ning, Y., Ledbetter, D.H., Bar-Am, I., Soenksen, D., Garini, Y., and Ried, T. 1996. Multicolor spectral karyotyping of human chromosomes. Science 273:494-497.
Hans J. Tanke
Molecular Cytogenetics
8.0.3 Current Protocols in Cytometry
Supplement 26
Overview of Fluorescence In Situ Hybridization Techniques for Molecular Cytogenetics HISTORY In situ hybridization is a powerful technology for visualizing the location of specific nucleic acid sequences on chromosomes, single cells, or tissue sections through the use of a nucleic acid probe that is complementary to those sequences and has been labeled in some fashion that renders it detectable. Until the early 1980s, radioisotopes were the only labels available for nucleic acid probes and microautoradiography was the only means to detect in situ–hybridized sequences. Radioactive probes provide limited spatial resolution for in situ hybridization because the decaying particles leave tracks, not discrete spots, in the photographic emulsion. It is further limited by the size of the silver halode crystals in the emulsion. Moreover, many practical inconveniences are imposed by the use of radioactivity, such as the need to observe relatively complicated safety measures, the limited shelf life of radioisotopes, and the long exposure periods required by autoradiography. Finally, with radioactive detection it is not possible to distinguish multiple targets in one multiprobe in situ hybridization experiment. The development in the 1980s of stable nucleic acid labels that allowed nonradioactive detection through fluorescence or enzyme reactions has demolished these practical and fundamental obstacles. In situ hybridization can now be performed rapidly with multiple differently colored nucleic acid probes at maximum optical resolution, and this has permitted widespread application of this methodology in clinical and basic research. Because of the substantial advantages offered by nonradioactive detection, the presentation of in situ hybridization techniques in this chapter is limited to those methods.
DIRECT VERSUS INDIRECT METHODS In direct in situ hybridization, the fluorescent reporter molecule is bound to the nucleic acid probe so that hybrids that have formed can be visualized microscopically immediately after in situ hybridization (Wiegant et al., 1991). In indirect procedures, the probe contains an element that renders it detectable by additional labeling steps (e.g., biotin-streptavidin binding Contributed by A.K. Raap Current Protocols in Cytometry (1997) 8.1.1-8.1.6 Copyright © 1997 by John Wiley & Sons, Inc.
UNIT 8.1
or immunocytochemistry): hence the term “indirect.” A number of such hapten modifications have been described. Direct methods are also amenable to immunocytochemical amplification if antibodies against the reporter molecules are available (Raap et al., 1990; Wiegant et al., 1991). Haptens currently in use include biotin, digoxigenin, and fluorescein. Fluorescein tetramethyl rhodamine, aminomethyl coumarin acetic acid (AMCA), and a series of cyanin dyes are in widespread use as fluorochromes, providing good spectral coverage across visual and infrared wavelengths. Although chemical methods for DNA labeling exist, generally haptens and fluorochromes are incorporated enzymatically into newly synthesized DNA using haptenor fluorochrome-modified dUTP. The allylamine derivative of dUTP can be fluorochromized or haptenized, for example using N-hydroxysuccinimide esters of haptens and fluorochromes. Its use for enzymatic synthesis of nonradioactive probes (Langer et al., 1981) was a major achievement because it fit closely with existing molecular biology formats for the radioactive labeling of nucleic acids employing DNA polymerases (e.g., by nick translation or random-primed labeling). This achievement led to the widespread application of nonradioactive probes in in situ hybridization. Fluorescence in situ hybridization (FISH) methods have achieved high standards of sensitivity, resolution, and multiplicity. In the following sections, these parameters and the conditions under which they are obtained are presented, and molecular cytogenetic applications of FISH are briefly discussed.
SENSITIVITY The sensitivity of a FISH procedure is defined as the smallest nucleic acid sequence target detectable. In discussing FISH sensitivity it is useful to consider a simplified view of the human genome as consisting of repeat and unique sequences. Repeat sequences may be clustered or dispersed. Relevant examples of clustered repeats are alphoid DNAs, which are located at chromosome centromeres, and simple satellite repeats, which occur at heterochromatic regions. Their clustered nature gives them a high degree of chromosome specificity,
Molecular Cytogenetics
8.1.1
Overview of FISH for Molecular Cytogenetics
and their high copy number makes them readily detectable by FISH. Alu and Kpn repeats are examples of dispersely occurring repeats. Because of their dispersed nature, these are not useful as markers and will increase background when present in probes. They are almost invariably present in large-insert clones; therefore, an important technical issue when performing FISH with large genomic probes is the need to eliminate such dispersely occurring repeat sequences from participation in the in situ hybridization reaction. This is done by preannealing the labeled DNA with unlabeled DNA enriched for repeats—i.e., C0t1 DNA—in a process known as suppression in situ hybridization (Landegent et al., 1987; Lichter et al., 1988; Pinkel et al., 1988). In metaphase chromosomes, unique targets >30 to 40 kb (cosmid-insert size) are readily detectable by microscopy after indirect FISH. Generally, >90% of cells will show the expected 2 × 2 copy number. Occurrence of paired spots on sister chromatids is a strong sign of specificity (Landegent et al., 1985; Lichter et al., 1990), and some nonspecific background spots may be tolerated. As probe size decreases, detection efficiency drops to the point that at 1 to 2 kb of unique target, FISH detection efficiency is such that some statistical analysis is necessary to assign the probe to a chromosomal band. Successful sub-kb chromosomal FISH is rare. In interphase cells, 30 to 40 kb of target is also readily detectable by microscopy, but background should be reduced to a minimum because—unlike in chromosomal FISH—no indicator of specificity is available. With DNA Fiber-FISH (Wiegant et al., 1992), which uses naked DNA immobilized on glass object slides, sensitivity is much better, most probably as a consequence of the high accessibility of naked DNA to probes and immunological detection reagents. Genomic plasmids 1 to 2 kb in size are easily visualized (Florijn et al., 1995), although sensitivity is better when hybridizing to larger genomic clones (e.g., cosmid). Recently, unique targets 200 bp in size have been detected, providing the means for (large) exon mapping (Florijn et al., 1996). Digital imaging is recommended in such cases. Indirect FISH methods are more sensitive than direct ones (Wiegant et al., 1991). Direct methods are generally recommended for repeat targets such as satellite DNAs. Chromosomal and interphase FISH using directly labeled cosmids and YACs as probes is, however, also
feasible (Wiegant et al., 1996). Improvements in FISH sensitivity have recently been achieved through the use of hapten- or fluorochrome-labeled tyramides and horseradish peroxidase (Kerstens et al., 1995; Raap et al., 1995; van Gijlswijk et al., 1996). This approach actually combines the sensitivity potentials of fluorescence and enzyme-based detection schemes. It is of particular value in situations where the signal-to-background ratio is suboptimal.
MULTIPLICITY The multiplicity of a FISH procedure is defined as the number of DNA targets that can be distinguished on the basis of optical properties, usually fluorescence color. In the simplest application of multiplicity, different fluorochromes that are spectrally well separated are attached to separate probes either directly or indirectly. For the visible part of the electromagnetic spectrum, blue, green, and red fluorescent dyes are available, permitting a multiplicity of 3 for visual observation of FISH results (Nederlof et al., 1989). When imaging devices sensitive to infrared light rays are used, for example a CCD camera with light integration capabilities, multiplicity can be increased to 4 or 5. When the targets are spatially separated, as in well-spread metaphase chromosomes, multiplicity can be increased by combinatorial labeling of the targets (Nederlof et al., 1990; Ried et al., 1992a; Wiegant et al., 1993; UNIT 8.3). Here multiplicity is 2n − 1, where n is the number of spectrally resolvable fluorochromes; this implies that with 3 and 4 fluorochromes, multiplicities of 7 and 15, respectively, can be achieved. For such combinatorial labeling of multiple FISH targets the ratios of the fluorescence intensities of the probes are in principle not relevant, but because the fluorescence intensity ratios of differentially labeled probes recognizing the same target turn out to be fairly constant after FISH (Nederlof et al., 1992), multiplicity can easily be increased to 12 (Dauwerse et al., 1992). Recently, successful FISH imaging of all the human chromosomes in 24 colors, using combinatorial FISH with five fluorochromes and either small band excitation/emission filter sets or special spectral imaging devices, has been reported (Schröck et al., 1996; Speicher et al., 1996). Detection efficiency is an important issue in multicolor FISH, and it is generally advisable to apply high-multiplicity FISH only to large targets. For example, with FISH detection efficiencies are 0.9 and an attempted multiplicity
8.1.2 Current Protocols in Cytometry
of 6, only in 0.96 × 100, or 53%, of the cells will all six targets be visible simultaneously.
RESOLUTION FISH resolution is defined as the smallest genomic distance that must exist between two DNA sequences in order for them to be resolved microscopically. Resolution is inversely proportional to chromatin compaction and is limited by the spatial resolution of the microscope. Apart from the resolution limit, the range over which that resolution is available is a useful specification for probe ordering and mapping by FISH. For metaphase chromosomes, with their highly condensed chromatin, maximum FISH resolution has been determined to be 1 to 3 Mb, ranging up to the full genome (Lawrence et al., 1990; Lichter et al., 1990). FISH performed on more relaxed chromatin, such as that of interphase nuclei, provides a resolution of 100 kb up to 1 to 2 Mb (Lawrence et al., 1990; Trask et al., 1991; van den Engh et al., 1992). FISH to naked DNA fibers provides the highest resolution: 1 kb ranging up to 400 to 1000 kb (Wiegant et al., 1992; Parra and Windle, 1993; Haaf and Ward, 1994; Heiskanen et al., 1994; Senger et al., 1994; Florijn et al., 1995). With DNA Fiber-FISH, resolution is limited only by the spatial resolution of the light microscope; 1 kb of double-stranded DNA has a length of 0.34 µm, which corresponds to the practical resolution limit of light microscopy (Florijn et al., 1995).
APPLICATIONS With the specifications discussed above, FISH has great potential for DNA mapping as well for analysis of the molecular genetic makeup of individual cells in a morphological context (van Ommen et al., 1995). As a consequence, the applications of FISH are manyfold, and the technique has had a strong impact in clinical and research fields with a strong morphological component, such as cytogenetics, pathology, hematology, and cell biology, as well as in nonmorphological disciplines such as molecular genetics. FISH can be regarded as an elegant merger of molecular biology and cytology. In general, the ability of FISH to visualize chromosomes and chromosome segments as discrete domains is of prime importance, because it permits rapid assessment of the location, copy number, and rearrangements of genes and chromosomes in individual cells, irrespective of cell cycle stage (Cremer et al., 1986; Emanuel, 1993).
Using total DNA from human (radiation) hybrid cell lines as probe, the chromosomal origin of the human component can be rapidly mapped cytogenetically by FISH to normal metaphase chromosomes (Kievits et al., 1990). Similarly, microdissected or flow-sorted abnormal chromosomes may be characterized by “reverse painting” following universal amplification strategies (Carter et al., 1992; Melzer et al., 1993). Chromosomal imbalances in a test DNA sample can be assessed and cytogenetically mapped using comparative genomic hybridization (CGH). CGH is based on the simultaneous in situ hybridization of differentially labeled test DNA (e.g., tumor DNA; in red) and reference DNA (e.g., normal DNA; in green) to normal metaphase chromosome preparations. If over- or underrepresentation of DNA sequences occurs in the test DNA, then an imbalance in the red-over-green fluorescence intensity ratio results at the cytogenetic location where the over- or underrepresented DNA maps (Kallioniemi et al., 1992a; du Manoir et al., 1993). Probes for whole chromosomes or parts thereof can be used to help characterize complex karyotypes; particularly when performed in high multiplicity, this considerably alleviates the problems associated with conventional karyotyping (Smit et al., 1991; Lengauer et al., 1993; Schröck et al., 1996; Speicher et al., 1996). Multicolor FISH using probes for chromosome-specific repeat targets is of great value in the detection of aneusomy (Hopman et al., 1988; Arnoldus et al., 1991; Ried et al., 1992b), whereas cosmid, YAC, and especially BAC and P1 clones provide excellent FISH probes for targeted detection of regional deletions, amplifications, and structural rearrangements (Arnoldus et al., 1990; Tkachuk et al., 1990; Kallioniemi et al., 1992b; Ried et al., 1992c; Stilgenbauer et al., 1993; Bentz et al., 1994). FISH also plays an essential role in cytogenetic and physical DNA mapping, because it provides a wide resolution range (see discussion of Resolution for literature references). Apart from these molecular (cyto)genetic applications, FISH has a role to play in molecular cell biology because it permits detection of gene expression at the RNA level (Lawrence et al., 1989; Dirks et al., 1993), investigation of the timing of replication (Kitsberg et al., 1993; Selig et al., 1993), and study of functional organization (Cremer et al., 1993; Zachar et al., 1993; Zirbel et al., 1993).
Molecular Cytogenetics
8.1.3 Current Protocols in Cytometry
LITERATURE CITED Arnoldus, E.P.J., Wiegant, J., Noordermeer, I.A., Wessels, J.W., Beverstock, G.C., Grosveld, G.C., van der Ploeg, M., and Raap, A.K. 1990. Detection of the Philadelphia chromosome in interphase nuclei. Cytogenet. Cell Genet. 54:108111. Arnoldus, E.P.J., Noordermeer, I.A., Peters, A.B.C., Voormolen, J.H.C., Bots, G.T.A.M., Raap, A.K., and van der Ploeg, M. 1991. Interphase cytogenetics of brain tumors. Genes Chrom. Cancer 3:101-107. Bentz, M., Cabot, G., Moos, M., Speicher, M., Ganser, M., Lichter, P., and Dohner, H. 1994. Detection of chimeric BCR-ABL genes on bone marrow samples and blood smears in chronic myeloid and acute lymphoblastic leukemia by in situ hybridization. Blood 83:1922-1928. Carter, N.P., Ferguson-Smith, M.A., Perryman, M.T., Telenius, H., Pelmear, A.H., Leversha, M.A., Glancy, M.T., Wood, S.L., Cook, K., Dyson, H.M., Ferguson-Smith, M.E., and Willat, L.R. 1992. Reverse chromosome painting: A method for the rapid analysis of aberrant chromosomes in clinical cytogenetics. J. Med. Genet. 29:299-307. Cremer, T., Landegent, J.E., Brueckner, A., Scholl, H.P., Schardin, M., Hager, H.D., Devilee, P., Pearson, P., and van der Ploeg, M. 1986. Detection of chromosome aberrations in the human interphase nucleus by visualization of specific target DNAs with radioactive and non-radioactive in situ hybridization techniques: Diagnosis of trisomy 18 with probe L1.84. Hum. Genet. 74:346-352. Cremer, T., Kurz, A., Zirbel, S., Dietzel, S., Rinke, E., Schröck, E., Speicher, M., Mathieu, R.U., Jauch, A., Emmerlich, P., Scherthan, H., Ried, T., Cremer, P., and Lichter, P. 1993. The role of chromosome territories in the functional compartmentalization of the cell nucleus. Cold Spring Harbor Symp. Quant. Biol. 58:777-792. Dauwerse, J.G., Wiegant, J., Raap, A.K., Breuning, M.H., and van Ommen, G.J.B. 1992. Multiple colors by fluorescence in situ hybridization using ratio-labelled DNA probes create a molecular karyotype. Hum. Mol. Genet. 1:593-598.
Florijn, R.J., van de Rijke, F.M., Vrolijk, H., Blonden, L.A.J., Hofker, M.H., den Dunnen, J.T., Tanke, H.J., van Ommen, G.J.B., and Raap, A.K. 1996. Exon mapping by Fiber-FISH or LR-PCR. Genomics. In press. Haaf, T. and Ward, D.C. 1994. High resolution ordering of YAC contigs using extended chromatin and chromosomes. Hum. Mol. Genet. 3:629-633. Heiskanen, M., Karhu, R., Hellsten, E., Peltonen, L., Kallioniemi, O.P., and Palotie, A. 1994. High resolution mapping by FISH to extended DNA fibers prepared from agarose embedded cells. Biotechniques. 17:928-933. Hopman, A.H.N., Ramaekers, F.C.S., Raap, A.K., Beck, J.L.M., Devilee, P., van der Ploeg, M., and Vooijs, G.P. 1988. In situ hybridization as a tool to study numerical chromosome aberrations in solid bladder tumors. Histochemistry 89:307316. Kallioniemi, A., Kallioniemi, O.P., Sudar, D., Rutovitz, D., Gray, J.W., Waldman, F.M., and Pinkel, D. 1992a. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 258:818-821. Kallioniemi, O.P., Kallioniemi, A., Kurisu, W., Thor, A., Chen, L.C., Smith, H.S., Waldman, F.M., Pinkel, D., and Gray, J.W. 1992b. ERBB2 amplification in breast cancer analyzed by fluorescence in situ hybridization. Proc. Natl. Acad. Sci. U.S.A. 89:5321-5325. Kerstens, H.M.J., Poddighe, P.J., and Hanselaar, A.G.J.M. 1995. CARD-ISH signal amplification. J. Histochem. Cytochem. 43:347-350. Kievits, T., Devilee, P., Wiegant, J., Wapenaar, M.C., Cornelisse, C.J., van Ommen, G.J.B., and Pearson, P.L. 1990. Direct non-radioactive in situ hybridization of somatic cell hybrid DNA to human lymphocyte chromosomes. Cytometry 1:105-109. Kitsberg, D., Selig, S., Brandeis, M., Simon, I., Keshet, I., Driscoll, D.J., Nicholls, R.D., and Cedar, H. 1993. Allele-specific replication timing of imprinted gene regions. Nature 364:459-463.
Dirks, R.W., van de Rijke, F.M., Fujishita, S., van der Ploeg, M., and Raap, A.K. 1993. Methodologies for specific intron and exon localization in cultured cells by haptenized and fluorochromized probes. J. Cell Sci. 104:1187-1197.
Landegent, J.E., Jansen in de Wal, N., van Ommen, G.J.B., Baas, F., de Vijlder, G., van Duijn, P., and van der Ploeg, M. 1985. Chromosomal localization of a unique gene by non-autoradiographic in situ hybridization. Nature 317:175-177.
du Manoir, S., Speicher, M.R., Joos, S., Schröck, E., Popp, S., Doehner, H., Kovacs, G., Robert-Nicoud, M., Lichter, P., and Cremer, T. 1993. Detection of complete and partial chromosome gains and losses by comparative genomic in situ hybridization. Hum. Genet. 90:590-610.
Landegent, J.E., Jansen in de Wal, N., Dirks, R.W., Baas, F., and van der Ploeg, M. 1987. Use of whole cosmid cloned genomic sequences for chromosomal localization by non-radioactive in situ hybridization. Hum. Genet. 77:366-370.
Emanuel, B.S. 1993. The use of fluorescence in situ hybridization to identify human chromosomal anomalies. Growth Genet. Hormones 9:6-12. Overview of FISH for Molecular Cytogenetics
1995. High-resolution DNA Fiber-FISH for genomic DNA mapping and colour-bar coding large genes. Hum. Mol. Genet. 4:831-836.
Florijn, R.J., Blonden, L.A.J., Vrolijk, H., Wiegant, J., Vaandrager, J.W., Baas, F., den Dunnen, J.T., Tanke, H.J., van Ommen, G.J.B., and Raap, A.K.
Langer, P.R., Waldrop, A.A., and Ward, D.C. 1981. Enzymatic synthesis of biotin-labeled polynucleotides: Novel nucleic acid affinity probes. Proc. Natl. Acad. Sci. U.S.A. 78:6633-6637. Lawrence, J.B., Singer, R.H., and Marselle, L.M. 1989. Highly localized tracks of specific tran-
8.1.4 Current Protocols in Cytometry
scripts within interphase nuclei visualized by in situ hybridization. Cell 57:493-502.
fluorochrome-tyramides. Hum. Mol. Genet. 4:529-534.
Lawrence, J.B., Singer, R.H., and McNeil, J.A. 1990. Interphase and metaphase resolution of different distances within the human dystrophin gene. Science 249:928-932.
Ried, T., Baldini, A., Rand, T.C., and Ward, D.C. 1992a. Simultaneous visualization of seven different DNA probes by in situ hybridization using combinatorial fluorescence and digital imaging microscopy. Proc. Natl. Acad. Sci. U.S.A. 89:1388-1392.
Lengauer, C., Speicher, M.R., Popp, S., Jauch, A., Taniwaki, M., Nagaraja, R., Riethman, H.C., Donis-Keller, H., D’Urso, M., Schlessinger, D., and Cremer, T. 1993. Chromosomal bar codes produced by multicolor fluorescence in situ hybridization with multiple YAC clones and whole chromosome painting probes. Hum. Mol. Genet. 2:505-512. Lichter, P., Cremer, T., Borden, J., Manuelidis, L., and Ward, D.C. 1988. Delineation of individual human chromosomes in metaphase and interphase cells by in situ hybridization using recombinant DNA libraries. Hum. Genet. 80:224-234. Lichter, P., Tang, C.C., Call, K., Hermanson, G., Evans, G., Housman, D., and Ward, D.C. 1990. High resolution mapping of human chromosome 11 by in situ hybridization with cosmid probes. Science 247:64-69. Melzer, P.S., Guan, X.U., Burgers, A., and Trent, J.M. 1993. Rapid generation of region specific probes by chromosome microdissection and their application. Nature Genet. 1:24-28. Nederlof, P.M., Robinson, D., Abuknesha, R., Wiegant, J., Hopman, A.H.N., Tanke, H.J., and Raap, A.K. 1989. Three-color fluorescence in situ hybridization for the simultaneous detection of multiple nucleic acid sequences. Cytometry 10:20-27. Nederlof, P.M., Flier, S., Wiegant, J., Raap, A.K., Tanke, H.J., Ploem, J.S., and van der Ploeg, M. 1990. Multiple fluorescence in situ hybridization. Cytometry 11:126-131. Nederlof, P.M., Flier, S., Vrolijk, J., Tanke, H.J., and Raap, A.K. 1992. Fluorescence ratio measurements of double-labeled probes for multiple in situ hybridization by digital imaging microscopy. Cytometry 13:839-845. Parra, I. and Windle, B. 1993. High resolution visual mapping of stretched DNA by fluorescent hybridization. Nature Genet. 5:17-21. Pinkel, D., Landegent, J.E., Collins, C., Fuscoe, J., Segraves, R., Lucas, J., and Gray, J. 1988. Fluorescence in situ hybridization with human chromosome-specific libraries: Detection of trisomy 21 and translocations of chromosome 4. Proc. Natl. Acad. Sci. U.S.A.85:9138-9142. Raap, A.K., Nederlof, P.M., Dirks, R.W., Wiegant, J.C.A.G., and van der Ploeg, M. 1990. Use of haptenised nucleic acid probes in fluorescent in situ hybridisation. In In Situ Hybridization: Application to Developmental Biology and Medicine (N. Harris and D.G. Williams, eds.) pp. 33-41. Cambridge University Press, Cambridge. Raap, A.K., van de Corput, M.P.C., Vervenne, R.A.W., van Gijlswijk, R.P.M., Tanke, H.J., and Wiegant, J. 1995. Ultrasensitive FISH using peroxidase-mediated deposition of biotin- or
Ried, T., Landes, G., Dackowski, W., Klinger, K., and Ward, D.C. 1992b. Multicolor fluorescence in situ hybridization for the simultaneous detection of probe sets for chromosomes 13, 18, 21, X and Y in uncultured amniotic fluid cells. Hum. Mol. Genet. 1:307-313. Ried, T., Lengauer, C., Cremer, T., Wiegant, J., Raap, A.K., van der Ploeg, M., Groitl, P., and Lipp, M. 1992c. Specific metaphase and interphase detection of the breakpoint region in 8q24 of Burkitt lymphoma cells by triple colour fluorescence in situ hybridization. Genes Chrom. Cancer 4:69-74. Schröck, E., du Manoir, S., Veldman, T., Schoell, B., Wienberg, J.Y., Ferguson-Smith, M.A., Ning, Y., Ledbetter, D.H., Bar-Am, I., Soenksen, D., Garini, Y., and Ried, T. 1996. Multicolor spectral karyotyping of human chromosomes. Science 273:494-497. Selig, S., Okumura, K., Ward, D.C., and Cedar, H. 1993. Delineation of DNA replication time zones by fluorescence in situ hybridization. EMBO J. 11:1217-1225. Senger, G., Jones, T.A., Fidlerova, H., Sanseau, P., Trowsdale, J., Duff, M., and Sheer, D. 1994. Released chromatin: Linearized DNA for high resolution fluorescence in situ hybridization. Hum. Mol. Genet. 3:1275-1280. Smit, V.T., Wessels, J.W., Mollevanger, P., Dauwerse, J.G., van Vliet, M., Beverstock, G.C., Breuning, M.H., Devilee, P., Raap, A.K., and Cornelisse, C.J. 1991. Improved interpretation of complex chromosomal rearragements by combined GTG banding and in situ suppression hybridization using chromosome-specific libraries and cosmid probes. Genes Chrom. Cancer 3:239-248. Speicher, M.R., Gwyn Ballard, S., and Ward, D.C. 1996. Karyotyping human chromosomes by combinatorial multi-fluor FISH. Nature Genet. 12:368-375. Stilgenbauer, S., Dohner, H., Bulgay-Morschel, M. Weitz, S., Bentz, M., and Lichter, P. 1993. High frequency of monoallelic retinoblastoma gene deletion in B-cell chronic lymphocytic leukemia shown by interphase cytogenetics. Blood 81:2118-2222. Tkachuk, D.C., Westbrook, C.A., Andreeff, M., Donlon, T.A., Cleary, M.L., Suryanarayan, K., Homge, M., Redner, A., Gray, J., and Binkel, D. 1990. Detection of bcr-abl fusion in chronic myelogeneous leukemia by in situ hybridization. Science 250:559-561. Trask, B.J., Massa, H., Kenwrick, S., and Gitschier, J. 1991. Mapping of human chromosome Xq28 by two-color fluorescence in situ hybridization
Molecular Cytogenetics
8.1.5 Current Protocols in Cytometry
of DNA sequences in interphase cell nuclei. Am. J. Hum. Genet. 48:1-15. van den Engh, G., Sachs, R., and Trask, B.J. 1992. Estimating genomic distance from DNA sequence location in cell nuclei by a random walk model. Science 257:1410-1412. van Gijlswijk, R.P.M., Wiegant, J., Raap, A.K., and Tanke, H.J. 1996. Improved localization of fluorescent tyramides for fluorescence in situ hybridization using dextran sulfate and polyvinyl alcohol. J. Histochem. Cytochem. 44:389-392. van Ommen, G.J.B., Breuning, M.H., and Raap, A.K. 1995. FISH in genome research and molecular diagnostics. Curr. Opin. Genet. Devel. 5:304-308.
Rudkin, G.T. and Stollar, B.D. 1977. High resolution detection of DNA⋅RNA hybrids in situ by indirect immunofluorescence. Nature 265:472473. First description of indirect FISH using antiDNA⋅RNA antibodies. Bauman, J.G.J., Wiegant, J., Borst, P., and Van Duijn, P. 1980. A new method for fluorescence microscopical localization of specific DNA sequences by in situ hybridization of fluorochrome labeled RNA. Exp. Cell Res. 138:485-490. First report of direct FISH using fluorochrome-labeled RNA. Langer et al., 1981. See above
Wiegant, J., Kalle, W., Mullenders, L., Brookes, S., Hoovers, J.M.N., Dauwerse, J.G., van Ommen, G.J.B., and Raap, A.K. 1992. High-resolution in situ hybridization using DNA halo preparations. Hum. Mol. Genet. 1:587-591.
First description of the synthesis of allylamine dUTP, its conjugation to biotin, and its use in DNA polymerase reactions.
Wiegant, J., Ried, T., Nederlof, P.M., van der Ploeg, M., Tanke, H.J., and Raap, A.K. 1991. In situ hybridization with fluoresceinated DNA. Nucl. Acids Res. 19:3237-3241.
First report of single-copy gene detection by nonradioactive in situ hybridization using reflection microscopy.
Wiegant, J., Wiesmeijer, C.C., Hoovers, J.M.N., Schuuring, E., d’Azzo, A., Vrolijk, J., Tanke, H.J., and Raap, A.K. 1993. Multiple and sensitive fluorescence in situ hybridization with rhodamine-, fluorescein-, and coumarin-labeled DNAs. Cytogenet. Cell Genet. 63:73-76. Wiegant, J., Verwoerd, N.P., Mascheretti, S., Bolk, M., Tanke, H.J., and Raap, A.K. 1996. An evaluation of a new series of fluorescent dUTPs for fluorescence in situ hybridization. J. Histochem. Cytochem. 44:525-529. Zachar, Z., Kramer, J., Mims, I.P., and Bingham, P.M. 1993. Evidence for channeled diffusion of pre-mRNA during nuclear RNA transport in metazoans. J. Cell Biol. 12:729-742. Zirbel, R.M., Mathieu, U.R., Kurz, A., Cremer, T., and Lichter, P. 1993. Evidence for a nuclear compartment of transcription and splicing located at chromosome boundaries. Chromosome Res. 1:93-106.
KEY REFERENCES Pardue, M.L., and Gall, J.G. 1969. Molecular hybridization of radioactive DNA to the DNA of cytological preparations. Proc. Natl. Acad. Sci. U.S.A. 64:600-604.
Landegent et al., 1985. See above.
Cremer et al., 1986. See above. Introduction of “interphase cytogenetics.” Landegent et al., 1987; Lichter et al., 1988; and Pinkel et al., 1988. See above. First reports of the repeat suppression hybridization principle, also called chromosomal in situ suppression hybridization. Wiegant et al., 1992; Parra and Windle, 1993. See above. First Fiber-FISH reports. Hopman et al., 1986. See above. First bicolor FISH paper. Kallioniemi et al., 1992a. See above. First CGH report.
Contributed by A.K. Raap Leiden University Leiden, The Netherlands
Very first report of in situ hybridization using autoradiography for detection.
Overview of FISH for Molecular Cytogenetics
8.1.6 Current Protocols in Cytometry
Basic Preparative Techniques for Fluorescence In Situ Hybridization
UNIT 8.2
This unit presents protocols for preparing human metaphase chromosome slides from peripheral blood lymphocytes (see Basic Protocol 1), isolating interphase nuclei from lymphocytes and paraffin-embedded tissues (see Basic Protocol 2), and preparing DNA fibers (see Basic Protocols 3, 4, 5, 6, and 7 and Alternate Protocols 1 and 2). The protocols are designed so that the resulting preparations are amenable to fluorescence in situ hybridization (FISH; UNIT 8.3). The methods correspond to a selection of the specimens that can be analyzed with FISH techniques, and the choice of sample preparation technique is highly dependent on the molecular cytogenetic question being addressed. Normal blood lymphocyte metaphase preparations are used as target specimens for the following procedures: cytogenetic mapping of cloned DNA; chromosomal characterization of the human component of hybrid-cell DNA; chromosomal characterization of PCR-amplified, microdissected, or flow-sorted chromosomes (reverse painting); and comparative genomic hybridization (CGH). Basic Protocol 1 is well suited for these applications. However, the condensation state of the chromatin and the resolving power of the light microscope limit the mapping resolution of metaphase chromosomes to a few Mbp. Generally, interphase nuclei are used in combination with targeted, disease-specific FISH probes to detect (cyto)genetically aberrant cells in a cell population—making interphase cytogenetics (see Basic Protocol 2) diagnostically and prognostically very powerful. Indeed the strong point of interphase genetics is the ability to rapidly assess copy number and detect rearrangement of genes and chromosomes in a large number of cells, irrespective of their cell-cycle stage, by analyzing number, color, and (relative) position of FISH signals. Hence, in principle this technique allows detection and quantification of mosaicisms. Detection of low-frequency aberrant cells, however, requires tedious visual analysis of many nuclei; therefore image cytometric instrumentation that assists in the process of interphase cytogenetic analysis by automation becomes a necessity. Basic Protocol 2 is designed for both manual and automated analysis of interphase cell samples. Interphase nuclei are also used for DNA mapping (interphase mapping). They provide a mapping resolution of 50 to 100 kbp, ranging up to 1 to 2 Mbp. Fiber-FISH has a significant role to play in physical DNA mapping because it provides a 1-kb resolution, ranging up to a few hundred kbp, and in contrast to restriction mapping permits gap-sizing in that range. Furthermore, fiber-FISH permits visualization of large genomic disease regions as “color bar codes.” Hence, gene rearrangements such as translocations, deletions, and duplications can readily be studied in patient DNAs by fiber-FISH. In the few years that this method has existed, many protocols have been described for preparing DNA fibers amenable to FISH. The ones described in Basic Protocols 3 to 7 have proven to give similar results.
Molecular Cytogenetics Contributed by J. Wiegant and A.K. Raap Current Protocols in Cytometry (1997) 8.2.1-8.2.23 Copyright © 1997 by John Wiley & Sons, Inc.
8.2.1
BASIC PROTOCOL 1
PREPARATION OF METAPHASE CHROMOSOME SLIDES FROM PERIPHERAL BLOOD LYMPHOCYTES Human metaphase chromosomes are prepared for FISH applications essentially according to routine cytogenetic procedures. The protocol described here consists of culturing whole blood, metaphase arrest with Colcemid (demecolcine), hypotonic treatment, fixation (Bosman et al., 1975), and slide preparation. To increase accessibility of the target sequences for reaction with probe DNA and immunocytochemical reagents, chromosome preparations are pretreated prior to FISH with RNase A to remove any endogenous RNA that may be present. This is followed by protease treatment (e.g., with pepsin) to remove proteins, and finally a post-fixation treatment with formaldehyde to ensure preservation of chromosomal morphology (Wiegant et al., 1991). Materials Peripheral blood, collected in 10-ml heparin-coated Vacutainer (Becton Dickinson) Chromosome medium with phytohemagglutinin (PAA Laboratories) 0.0025% Colcemid working solution (see recipe) 3 mg/ml thymidine (Sigma) in RPMI 1640 medium, sterile (store ≤3 years at –20°C) Hypotonic buffer (see recipe) Methanol/acetic acid fixative: 3:1 (v/v) absolute methanol/glacial acetic acid (prepare fresh) 100%, 90%, 80%, and 70% acetic acid (optional; if needed) RNase A solution (see recipe) 2× SSC (APPENDIX 2A), room temperature and 37°C 0.01 M HCl 10% pepsin: dissolve 1 g pepsin (Boehringer Mannheim) in 10 ml H2O at 37°C; store in aliquots ≤1 year at –20°C 1× PBS (see recipe) Formaldehyde fixative (see recipe) 70%, 90%, and 100% ethanol 250-ml tissue culture flasks 15-ml polypropylene centrifuge tubes Tabletop centrifuge (e.g., IEC Clinical or HN-SII) Microscope slides, cleaned by dipping in 1:1 (v/v) ethanol/diethyl ether and wiping with lint-free Kimwipes (e.g., Fisher) Phase-contrast microscope 24 × 50–mm coverslips Rectangular staining dishes (e.g., Fisher) Moist chamber: 1-liter beaker containing paper towels moistened with 2× SSC (APPENDIX 2A), covered with aluminum foil Culture cells 1. Add 10 ml of peripheral blood (collected in heparinized Vacutainer) to 100 ml of chromosome medium with phytohemagglutinin, then divide suspension among four 250-ml culture flasks. 2. Incubate 72 hr at 37°C with the caps closed. A CO2 incubator is not necessary.
Basic Preparative Techniques for FISH
8.2.2 Current Protocols in Cytometry
Induce metaphase arrest 3a. To obtain metaphase chromosomes: After 72 hr of culture, add 80 µl of freshly prepared 0.0025% Colcemid working solution to each culture flask and incubate at 37°C for an additional 2 hr. 3b. To obtain longer (almost prometaphase) chromosomes: After 72 hr of culture, add 250 µl of 3 mg/ml thymidine in RPMI-1640 medium to each culture flask, continue incubating at 37°C for 17 hr, then add 80 µl of 0.0025% Colcemid to each flask and incubate at 37°C an additional 2 hr. Swell cells with hypotonic buffer 4. Pool the contents of the four culture flasks and mix gently. 5. Divide the cell suspension into equal aliquots among eight 15-ml polypropylene centrifuge tubes (∼14 ml per tube). Centrifuge 8 min at 90 × g, room temperature, in a tabletop centrifuge. 6. Aspirate the supernatant until 1 ml is left. Resuspend the pellet in the residual solution, add 14 ml hypotonic buffer while carefully mixing, then transfer the tubes to a 37°C water bath and incubate 20 min. Fix cells three times 7. Centrifuge for 8 min at 90 × g, room temperature. Aspirate the supernatant until 0.5 ml is left, then resuspend the pellet in the residual solution. 8. Prepare a Pasteur pipet fully filled with methanol/acetic acid fixative (∼2 ml). Add a small amount of fixative to the cell suspension in one of the tubes and immediately aspirate it back into the Pasteur pipet. Add a slightly larger amount of fixative to the cell suspension and aspirate it back into the Pasteur pipet, then repeat this process several times until the full contents of the Pasteur pipet has been added to the cell suspension. By doing this the cells are fixed and constantly kept in movement at the same time.
9. Repeat step 8 for the first tube (so that ∼4 ml of fixative has been added), then repeat the entire process with the other tubes until ∼4 ml of fixative has been added to each tube. After the first Pasteur pipet of fixative has been added to the cell suspension the actions can be carried out more rapidly. The first addition of fixative to the cell suspension is the most critical.
10. Fill all tubes up to 10 ml with fixative and resuspend with Pasteur pipet. Let tubes stand 10 min at room temperature. 11. Centrifuge 7 min at 90 × g, room temperature. Aspirate the supernatant until 0.5 ml is left, then resuspend the pellet in the residual solution. 12. Repeat steps 8 and 9 for all tubes, then pool the contents of two tubes into one tube (so that there will be four tubes, each containing ∼8 ml of fixative). Fill all tubes up to 10 ml with fixative, then let the tubes stand on crushed ice for 60 min after this second fixation. The addition of fixative in this step is done as in step 8, but the actions are carried out much more quickly.
13. Centrifuge 7 min at 90 × g, room temperature. Aspirate the supernatant until 0.5 ml is left, then resuspend the pellet in the residual solution.
Molecular Cytogenetics
8.2.3 Current Protocols in Cytometry
14. Repeat the fixation as in step 12 with another 10 ml of fixative per tube. Let the tubes stand on crushed ice for at least 30 min after this third fixation before preparing slides. During the three successive fixations the remnants of the erythrocytes dissolve. Fixed cells may be stored for ≤3 months at −20°C. This is best done with the tubes completely filled with fixative. Each time part of the cell suspension is used to prepare slides, the tube should be filled up again with fixative before storage at −20°C.
Prepare slides 15. Breathe on a cleaned microscope slide to deposit water vapor. This allows for better spreading of the chromosomes.
16. Drop 3 drops of the cold (4°C) metaphase suspension (from step 14) onto three separate spots on the slide from ∼30 cm above. Let dry at room temperature. Three drops on separate spots will cover almost the whole preparation with cells. This makes it possible to perform two FISH experiments on one preparation (on the left and right sides of the slide).
17. Inspect the metaphase slides with a phase-contrast microscope at 25× magnification for the presence of remaining cytoplasm. In optimal preparations, the chromosomes should appear grayish and have minimal overlap and no traces of residual cytoplasm (see Critical Parameters).
18a. If slides are satisfactory: Allow to air dry ≥24 hr. Slides that have simply been air dried for ≥24 hr are referred to as “fresh.” Fresh slides may be stored at room temperature for ≤2 weeks. For longer periods of time (≤3 months) storage in 70% ethanol at 4°C is recommended. Slides stored in 70% ethanol first must be dehydrated in 90% and 100% ethanol (3 min each), then air dried before continuing with preteatment.
18b. If cytoplasm is still visible or if the chromosomes are not spread sufficiently: Perform the following procedure to optimize slide preparation: a. Prepare staining dishes containing 100%, 90%, 80%, and 70% acetic acid. The optimal concentration is determined experimentally; 70% acetic acid removes more cytoplasm and gives a better spreading of the chromosomes than 100% acetic acid.
b. Prepare 10 to 15 chromosome slides as in steps 15 and 16. Put them in a horizontal position on the edge of a fume hood. c. When the last slide has been prepared, dip the first slide that was prepared (which should now be dry) into the 100% acetic acid solution, then place it in a vertical position to drain the acetic acid. Proceed with the remaining slides in the same way. d. Allow slides to air dry, then examine with phase-contrast microscope for correct chromosome morphology (see step 17 and Critical Parameters). e. Repeat if necessary with the acetic acid solutions of successively lower concentrations. f. When optimal slides have been obtained, allow to air dry ≥24 hours. Proceed to pretreatment (steps 19 to 27) or store as described above. See Critical Parameters for additional guidelines on acetic acid treatment.
Basic Preparative Techniques for FISH
Pretreat slides in preparation for FISH 19. Place 100 µl RNase A solution over chromosome preparation on a slide and cover with a 24 × 50–mm coverslip.
8.2.4 Current Protocols in Cytometry
20. Place the slide upside down in a horizontal position in a staining dish and put the jar in a moist chamber. Incubate 30 min at 37°C. 21. Add room temperature 2× SSC to the staining dish and gently agitate the slide to release the coverslip. 22. Using forceps, transfer the slide without coverslip to another staining dish containing room temperature 2× SSC. Immerse 5 min, then replace the SSC with fresh room temperature 2× SSC and immerse an additional 5 min. Finally, replace the SSC with fresh 2× SSC prewarmed to 37°C and immerse 5 min, keeping the temperature at 37°C in a water bath. 23. While the above washes are taking place, prewarm 100 ml of 0.01 M HCl to 37°C, then add 50 µl of 10% pepsin immediately before use. 24. Pour off the 2× SSC and add the pepsin solution to the slide in the staining dish. Incubate 10 min at 37°C in a water bath. 25. Wash slide by immersing 5 min in a staining dish containing 1× PBS, then transfer to another dish containing fresh 1× PBS and immerse an additional 5 min. 26. Fix by immersing 10 min in formaldehyde fixative at room temperature, then wash twice with 1× PBS as in step 25. 27. Dehydrate by immersing successively for 3 min each in 70%, 90%, and 100% ethanol. Air dry the slide. The slide is now ready for hybridization (UNIT 8.3). The RNase and pepsin pretreatments should immediately be followed by the FISH procedure; storage of any kind is not recommended (see Critical Parameters).
PREPARATION OF SLIDES FOR INTERPHASE CYTOGENETICS Here, alternative sets of steps are described for isolation of lymphocytes from fresh blood and for isolation of nuclei from paraffin-embedded tissues. These are then centrifuged onto microscope slides and subjected to proteolytic pretreatment for FISH. The resulting preparations are well suited for both manual and automated analysis of interphase cytogenetic results. Materials Peripheral blood, collected in 10-ml EDTA Vacutainer (Becton Dickinson) or paraffin-embedded tissue specimen 1× PBS (see recipe) 0.1 M KCl, 37°C (for blood specimens) 100% methanol (for blood specimens; Baker) Histo-Clear (for tissue specimens; National Diagnostics) 50%, 70%, 90%, and 100% ethanol 0.1% protease (bacterial type XXIV; Sigma; for tissue specimens) in protease buffer (see recipe) 0.3% (v/v) Nonidet P-40 (NP-40) in TE buffer, pH 7.8 (APPENDIX 2A); use for tissue specimens Methanol/acetic acid fixative: 3:1 (v/v) absolute methanol/glacial acetic acid (prepare fresh); use for tissue specimens (optional) 100% methanol (for tissue specimens) 1:1 (v/v) ethanol/diethyl ether 50% acetic acid
BASIC PROTOCOL 2
Molecular Cytogenetics
8.2.5 Current Protocols in Cytometry
0.1 M Na2B4O7⋅10H2O, unbuffered (pH 9.3); use for nuclei isolated from tissue specimens RNase A solution (see recipe) 4% formaldehyde (Merck) in 1× PBS (see recipe) 0.01 M HCl 0.1% pepsin (Sigma) in 0.01 M HCl (pH 2.0), 37°C (prepare just before use; for lymphocytes isolated from blood specimens) 1% pepsin (Sigma) in 0.01 M HCl (pH 2.0), 37°C (prepare just before use; for nuclei isolated from tissue specimens) 10-ml LeucoPrep tubes (for blood specimens; Becton Dickinson) Tabletop centrifuge 15-ml polypropylene centrifuge tubes 21-G needle and syringe Microtome with knives (for tissue specimens) 10-ml glass test tubes (for tissue specimens) 70-µm nylon-mesh cell strainer (Falcon) Microscope slides Lint-free Kimwipes (e.g., Fisher) Hettich-Universal tabletop cytocentrifuge with #1323 rotor and compatible buckets (#1266, #1271, or #1276) for depositing cells on slides (Hettich-Zentrifugen) Rectangular staining dishes (e.g., Fisher) 24 × 60–mm coverslips Moist chamber: 1-liter beaker containing paper towels moistened with 2× SSC (APPENDIX 2A), covered with aluminum foil Isolate and fix cells For isolation and fixation of lymphocytes from peripheral blood 1a. Transfer peripheral blood (collected in EDTA Vacutainer) to 10-ml LeucoPrep tubes. Centrifuge 30 min at 1400 × g, room temperature. Also see manufacturer’s instructions for the LeucoPrep tubes.
2a. Transfer the isolated lymphocytes to a 15-ml polypropylene centrifuge tube, then wash three times, each time by centrifuging 10 min at 150 × g, room temperature, removing the supernatant, adding 10 ml 1× PBS, then centrifuging again and removing the supernatant. 3a. Resuspend pellet in 5 ml of 37°C 0.1 M KCl then incubate 15 min at 37°C. 4a. Add 5 ml of 100% methanol while gently vortexing, centrifuge 10 min at 150 × g, room temperature, remove supernatant, and again add 5 ml methanol while gently vortexing. Repeat three times, but use 1 ml methanol for the final suspension. For isolation and fixation of nuclei from paraffin-embedded tissue 1b. Using a microtome, cut 50-µm sections. Place two or more of these sections in a 10-ml glass tube and add 3 ml Histo-Clear. Incubate 30 min with occasional shaking, then centrifuge 10 min at 150 × g, room temperature. Remove supernatant, add 3 ml Histo-Clear to pellet, resuspend, incubate 1 to 2 min, then centrifuge again at 150 × g. Remove supernatant, add 3 ml of 100% ethanol, resuspend, and finally centrifuge 10 min at 150 × g, room temperature. Repeat this wash procedure with 90%, 70%, and 50% ethanol, and finally twice with water. Basic Preparative Techniques for FISH
The washings with Histo-Clear and ethanol will remove the paraffin.
8.2.6 Current Protocols in Cytometry
2b. Remove supernatant, add 3 ml of 0.1% protease solution and incubate 60 min at 37°C with occasional vortexing. Centrifuge 10 min at 150 × g, room temperature, and remove the supernatant. Add 3 ml 0.3% NP-40 in TE buffer to the pellet, incubate 10 min at room temperature, then filter through a 70-µm nylon cell strainer. This treatment will release the nuclei.
3b. Centrifuge filtrate 10 min at 150 × g, and remove supernatant until ∼0.5 ml is left. Slowly add 3 ml methanol/acetic acid fixative while gently vortexing. Vortex, centrifuge 10 min at 150 × g, then remove supernatant until ∼0.5 ml is left. This methanol/acetic acid fixation is optional. Include if debris is present in the preparation; otherwise simply centrifuge the filtrate, remove the supernatant, and proceed to step 4b.
4b. Add 3 ml of 100% methanol while gently vortexing. Centrifuge 10 min at 150 × g, room temperature, and remove supernatant. Repeat methanol treatment and centrifugation three times, but use 1 ml methanol for the final suspension. 5. Draw suspension into a syringe and pass through a 21-G needle to minimize clumping, then store at 4°C until ready to proceed with slide preparation. Prepare slides using bucket centrifugation 6. Soak lint-free Kimwipes in 1:1 ethanol/diethyl ether and clean microscope slides by wiping. Allow to dry and place in centrifugation buckets for depositing cells on slides. 7. Resuspend fixed cells or nuclei in methanol by vortexing and/or syringing. Let stand for 2 min to sediment debris and remaining clumps. Resuspend a 100-µl aliquot in 8 ml of 50% acetic acid. Adjust the volume of the sample if resulting cell density is suboptimal. The 8-ml volume of 50% acetic acid is for four buckets; adjust volumes as necessary.
8a. For lymphocytes:Add 1 ml resuspended cells to the bucket holes of a Hettich-Zentrifugen cytocentrifuge bucket and centrifuge the bucket 5 min at 1400 × g in a Hettich-Universal cytocentrifuge. Pour off the 50% acetic acid solution, dismantle the bucket, remove the slide, and air dry 5 min. Immerse the slide for at least 5 min in 0.1 M Na2B407.10H2O in a staining jar, rinse with water, and air dry overnight. This treatment leads to swelling of nuclei. See Critical Parameters.
8b. For nuclei from paraffin-embedded tissue: Add 1 ml resuspended nuclei to the bucket holes of a Hettich-Zentrifugen cytocentrifuge bucket and centrifuge the bucket 5 min at 1400 × g in a Hettich-Universal cytocentrifuge. Pour off the 50% acetic acid solution, dismantle the bucket, remove the slide, and air dry overnight. Inadequate air drying leads to detachment of the cells during further processing; it is therefore preferable to air dry overnight.
Pretreat slides in preparation for FISH 9. Immerse slide 15 min in a staining dish containing 1× PBS. 10. Drain off excess fluid. Apply 100 µl RNase A solution to lymphocytes/nuclei on slide and cover with a 24 × 60–mm coverslip. Incubate 1 hr at 37°C in a moist chamber. 11. Transfer slide to a staining dish containing 1× PBS, gently shake to detach coverslip and transfer the using forceps to a second dish containing fresh 1× PBS. Molecular Cytogenetics
8.2.7 Current Protocols in Cytometry
For lymphocytes 12a. Pour off the PBS from the staining jar and replace with 4% formaldehyde in 1× PBS. Immerse slide 2 min, then pour off formaldehyde solution and immerse slide successively in three staining dishes containing fresh 1× PBS. Finally, immerse 3 min in 0.01 M HCl at room temperature. 13a. Incubate 10 min in 0.1% pepsin/0.01 M HCl at 37°C, then wash in a staining dish containing 1× PBS. For nuclei from paraffin-embedded tissue 12b. Immerse slide 3 min in 0.01 M HCl. 13b. Incubate 30 min in 1% pepsin/0.01% HCl at 37°C, then wash in a staining dish containing 1× PBS. Post-fix and dehydrate slides 14. Post-fix the preparation by immersing 2 min in 4% formaldehyde/PBS, then immerse successively for 3 min each in three staining dishes containing fresh 1× PBS. 15. Dehydrate by immersing successively for 3 min each in 70%, 90%, and 100% ethanol. Air dry the slide. The slide is now ready for hybridization (UNIT 8.3). BASIC PROTOCOL 3
PREPARATION OF SLIDES FOR DIRVISH This protocol to prepare slides for direct visual hybridization (DIRVISH) as well as Basic Protocols 4 to 7 and Alternate Protocols 1 and 2 have all been designed to achieve linearization of DNA fibers on microscope slides for in preparation for FISH (“FiberFISH”). In this protocol, unfixed cells are lysed on a horizontally positioned microscope slide. The slide is then tilted and the released DNA is allowed to run down and attach to the slide (Parra and Windle, 1993). Materials Cells for analysis 1× PBS (see recipe) Lysis buffer (see recipe) Methanol/acetic acid fixative: 3:1 (v/v) absolute methanol/glacial acetic acid (prepare fresh) 70%, 90%, and 100% ethanol RNase A solution (see recipe) Microscope slides: AES-coated (see Support Protocol 1) or poly-L-lysine–coated (see Support Protocol 3) Rectangular staining dishes (e.g., Fisher) Slide box with dessicant 1. Dilute cells to a concentration of 50 to 2500 cells/µl in 1× PBS. Place 100 to 5000 individual cells in 2 µl 1× PBS at one end of a microscope slide and air dry. 2. Place 5 µl lysis buffer on the air-dried cells and let stand for 5 min. 3. Tilt the slide with the DNA at the upper end to allow the DNA of lysed cells to run down.
Basic Preparative Techniques for FISH
4. Air dry, then fix with by immersing for 5 min in methanol/acetic acid fixative.
8.2.8 Current Protocols in Cytometry
5. Dehydrate slide by immersing successively for 3 min each in 70%, 90%, and 100% ethanol, then air dry. 6. Store DIRVISH slide at −20°C in a closed slide box containing dessicant until ready for hybridization (up to 3 months). 7. Prior to in situ hybridization pretreat the slide with RNase A and wash with 1× SSC (see Basic Protocol 1, steps 19 to 22), except carry out the last SSC wash at room temperature instead of 37°C. 8. Dehydrate slide by immersing successively for 3 min each in 70%, 90%, and 100% ethanol, air dry, and proceed with the in situ hybridization procedure (UNIT 8.3). PREPARATION OF SLIDES FOR DIRVISH (ALTERNATIVE PROCEDURE) In Basic Protocol 3 (the DIRVISH protocol of Parra and Windle, 1993) cells are air dried while suspended in PBS, which leaves salt crystals on the slide. Furthermore, the DNA fibers are obtained from one drop of cells placed upon a restricted area, which may lead to large bundles of DNA threads instead of individual fibers. To circumvent these problems, this protocol provides for an alternative in which the cells are resuspended in water and spread over the entire surface of the microscope slide. This procedure was developed by J.G. Dauwerse, Department of Human Genetics, Leiden University, the Netherlands (J.G. Dauwerse, pers. comm.).
ALTERNATE PROTOCOL 1
Additional Materials (also see Basic Protocol 3) Hair dryer Plant sprayer producing fine mist 1. Resuspend cells in water to a concentration of 1–5 × 105 cells/ml. 2. Pipet 100 µl of cell suspension onto an AES-coated slide and spread suspension out over the entire surface of the slide with a pipet tip. 3. Dry the slides completely using a warm hair dryer held at a distance of 30 to 50 cm. 4. Cover with a thin layer of lysis buffer using a plant sprayer. 5. Dry completely with hair dryer as in step 3. 6. Fix by immersing 5 min in methanol/acetic acid fixative. 7. Dehydrate through an ethanol series (see Basic Protocol 3, steps 5 and 8). This step is optional.
8. Prior to in situ hybridization pretreat the slide with RNase A (see Basic Protocol 1, steps 19 to 22). 9. Dehydrate slide by immersing successively for 3 min each in 70%, 90%, and 100% ethanol, air dry, and proceed with the in situ hybridization procedure (UNIT 8.3). Store slide at −20°C in a closed slide box containing dessicant until ready for hybridization.
Molecular Cytogenetics
8.2.9 Current Protocols in Cytometry
BASIC PROTOCOL 4
PREPARATION OF DNA FIBERS FROM FIXED CELLS In this protocol, the DNA of cells prepared according to conventional cytogenetic preparation techniques is released from the cells by the action of smearing either an alkaline solution or a formamide-containing solution (Fidlerová et al., 1994). The alkaline (sodium hydroxide) treatment results in complete disruption of nuclei, so that the slides are covered with an irregular network of chromatin fibers. With formamide, the borders of most of the disrupted nuclei can still be defined, allowing hybridization signals from the same nucleus to be identified. Materials 70% (v/v) formamide/2× SSC, pH 7 (see APPENDIX 2 for SSC recipe) 70%, 95%, and 100% ethanol 1× PBS (see recipe) NaOH/ethanol: mix 5 vol 0.07 M NaOH with 2 vol 100% ethanol (prepare fresh) 100% methanol Microscope slides: optionally AES-coated (see Support Protocol 1) or poly-L-lysine–coated (see Support Protocol 3) Rectangular staining dishes (e.g., Fisher) Long coverslips Additional reagents and equipment for harvesting and fixing peripheral blood lymphocytes (as for metaphase chromosome slide preparation; see Basic Protocol 1) 1. Harvest and fix peripheral blood lymphocytes (see Basic Protocol 1, steps 1 to 13). 2. Breathe onto a clean coated microscope slide to deposit water vapor, then place two drops of cell suspension on each half of the slide. Formamide method 3a. Wait a few seconds for cells to attach to the slide, then carefully place the slide into a staining dish filled with 2× SSC and immerse for for 2 to 3 sec. 4a. Remove slide from the SSC and allow to drain. Place one drop (~100 µl) of 70% formamide/2× SSC pH 7.0 on one end of a long coverslip. Invert the slide over the coverslip so that one end of the slide is touching the formamide/SSC, keeping an angle of ~30° between slide and coverslip. Slowly drag the slide over the formamide/SSC so that the entire slide becomes covered with formamide/SSC. Let stand ~30 sec. 5a. Drain off excess fluid, then rinse carefully with 70% ethanol by pipetting at the edges of the slide and gradually toward the middle so that the DNA does not run off the slide. 6a. Air dry, then inspect slide using phase-contrast microscope to ensure that the DNA is still on the slide and that it has been released from the nuclei. The DNA fibers can be seen by phase-contrast microscopy as long, stretched, grayish threads. This type of microscopy allows the investigator to check the number and configuration of the prepared fibers. In case the extension of the fibers is not suitable, the immersion of the slide in formamide should be prolonged.
7a. Dehydrate suitable slides by immersing successively for 3 min each in 95% and 100% ethanol. Air dry. Basic Preparative Techniques for FISH
8.2.10 Current Protocols in Cytometry
Alkaline method 3b. Wait a few seconds for cells to attach to the slide, then carefully place the slide into a staining dish filled with 1× PBS and immerse for 2 to 3 sec. Drain excess fluid. 4b. Place one drop (about 100 µl) of NaOH/ethanol solution on one end of a long coverslip. Invert the slide over the coverslip so that one end of the slide is touching the NaOH/ethanol, keeping an angle of ~30° between slide and coverslip. Slowly drag the slide over the NaOH/ethanol so that the entire slide becomes covered with NaOH/ethanol. Let stand ~30 sec. 5b. Carefully rinse slide with methanol by pipetting at the edges of the slide and gradually toward the middle so that the DNA does not run off the slide. 6b. Air dry, then inspect slide using phase-contrast microscope to ensure that the DNA is still on the slide and that it has been released from the nuclei. 7b. Dehydrate suitable slides by immersing successively for 3 min each in 70%, 95%, and 100% ethanol. Air dry. 8. Store slides dessicated up to 3 months at −20°C. Fresh slides from either procedure can be used for FISH (UNIT 8.3) after baking for at least 20 min at 65°C. In case the number of DNA fibers is too low (because of incomplete attachment of cells or DNA fibers to the glass slide), the use of coated slides is recommended.
PREPARATION OF DNA FIBERS FROM AGAROSE BLOCKS In this protocol, DNA is prepared in a manner similar to that used to prepare DNA for pulsed-field gel electrophoresis (PFGE). Cells in agarose blocks are prepared and digested with protease as for PFGE. The released DNA can either be separated by PFGE or used to make extended DNA preparations as described here. The blocks are melted and fibers of DNA are extended along a microscope slide by dragging with another slide (Heiskanen et al., 1994).
BASIC PROTOCOL 5
Materials Cells of interest (e.g., lymphocytes separated by density-gradient centrifugation or cells from cultured cell lines) 1× PBS (see recipe) 1.9% low gelling/melting temperature agarose (FMC Bioproducts) prepared in 1× PBS (see recipe) Proteinase digestion solution (see recipe) TE buffer (APPENDIX 2A) Tabletop centrifuge Plexiglas block mold (e.g., Bio-Rad 10-well sample-plug mold) 50°C water bath Poly-L-lysine coated microscope slides (see Support Protocol 3) 1. Add 20 ml of room temperature 1× PBS to the cell pellet obtained by gradient centrifugation, then centrifuge 10 min at 200 × g, room temperature. Remove the supernatant, count cells, then repeat this washing procedure. Resuspend cells in 1× PBS at a final concentration of 2 × 106 cells/µl. 2. Add an equal volume of 1.9% low gelling/melting temperature agarose in 1× PBS to cells, dispense into block mold, and allow to set. Molecular Cytogenetics
8.2.11 Current Protocols in Cytometry
Supplement 16
3. Remove blocks from mold, then incubate 48 hr in proteinase digestion solution at 50°C. 4. Place each block in a tube and wash five times with TE buffer each time by covering the block with buffer, then decanting. 5. Place a small piece of block (1⁄8 of a 10-µl block) on a coated microscope slide. 6. Prewarm the rotating glass plate in a microwave oven by running the oven empty for 4 min at ∼600 W. Add ∼20 µl water to the agarose block specimen on the slide and melt for ∼30 sec at ∼600 W. A microwave oven with a rotating plate is desirable for even melting. The melting time and temperature (i.e., power setting) and the amount of water added should be optimized for different purposes. If the melting is not efficient, unmelted agarose left on the slide will cause background problems.
7. Immediately after the block is melted, extend the DNA using the edge of another microscope slide, held at an angle of ∼30°, in a manner much like that used for making blood smears. Proceed immediately to denaturation and ISH. Do not store slides. BASIC PROTOCOL 6
PREPARATION OF DNA FIBERS FROM FIBROBLASTS BY THE HALO TECHNIQUE One of the oldest techniques used for DNA fiber-FISH mapping is called the halo preparation technique (Wiegant et al., 1992), a modified version of the technique described by Vogelstein et al. (1980). Cells growing on a glass slide are lysed in a buffer containing a detergent. This is then followed by high-salt washes to remove histones from the chromatin. The DNA in such nuclei consists of loops anchored to the nuclear matrix in a form having a high degree of negative supercoiling. Next, the nuclei are incubated in a high-salt buffer containing a DNA-intercalating dye to unwind the DNA loops by introducing positive supercoiling. This is then followed by UV irradiation to make nicks, which leads to relaxation of the DNA that has been looped out. Although in principle a true halo of DNA loops of 14 to 16 µm can be obtained, for mapping purposes it is advantageous to produce much longer loops. Materials Fibroblasts for analysis Dulbecco’s modified Eagle medium without phenol red and biotin (Life Technologies) 1× PBS (see recipe), 4°C Solutions A, B, C, D, and E for halo technique (see recipes), 4°C RNase A solution (see recipe) 70%, 90%, and 100% ethanol Microscope slides, cleaned by dipping in 1:1 (v/v) diethyl ether and wiping with lint-free Kimwipes, then sterilized by heating ≥2 hr at 160°C 50-ml centrifuge tubes or 100-ml beakers Glass plate prechilled to 0°C Slide box with dessiccant 1. Seed fibroblasts on cleaned, sterile microscope slides at a density of ∼1000 cells/mm2. Incubate overnight at 37°C in Dulbecco’s medium without phenol red and biotin in a 5% CO2 incubator.
Basic Preparative Techniques for FISH
Phenol red and biotin are omitted from the medium used to reduce autofluorescence and nonspecific avidin and streptavidin binding.
8.2.12 Supplement 16
Current Protocols in Cytometry
2. Dip slides successively for 45 sec each in two 50-ml centrifuge tubes or 100-ml beakers (kept on crushed ice) containing 4°C 1× PBS. 3. Dip slides successively for 45 sec each in 50-ml centrifuge tubes or 100-ml beakers (kept on crushed ice) containing solutions A, B, and C at 4°C. At this stage a slide can be inspected by fluoresence microscopy (with blue excitation) for the formation of DNA halos around the nuclei.
4. Place the slide on a cold (0°C) glass plate and irradiate 7 min (or optimal length of time) with UV light (254 nm, 7000 µW/cm2) at a distance of ∼5 mm (or optimal distance). The UV lamp used by the authors is a universal UV-lamp (Camag; Muttenz, Switzerland) originally designed to visualize thin-layer chromatography plates. The optimal 7-min UV irradiation has been experimentally determined using this lamp. For other light sources it is strongly advised to optimize the time of irradiation and the distance between the lamp and the microscope slide.
5. Dip slides successively for 45 sec each in 50-ml centrifuge tubes or 100-ml beakers (kept on crushed ice) containing solutions D and E at 4°C. Finally, dip successively for 45 sec each in two tubes or beakers containing water at 4°C. 6. Air dry the slides at room temperature in a vertical position. The air drying of the slides after the final wash in water in a vertical position causes the DNA to stretch down along the slide. This presumably results in the linearization of the DNA.
7. Store slides at −20°C in a slide box containing dessicant until ready for hybridization. See UNIT 8.3 for in situ hybridization techniques.
8. Prior to in situ hybridization pretreat the slide with RNase A and wash with 1× PBS (see Basic Protocol 1, steps 19 to 22). 9. Dehydrate slide by immersing successively for 3 min each in 70%, 90%, and 100% ethanol, air dry, and proceed with the in situ hybridization procedure (UNIT 8.3). PREPARATION OF DNA FIBERS FROM TUMOR CELLS BY HALO TECHNIQUE
ALTERNATE PROTOCOL 2
This protocol makes use of lymphoma cells released from cryostat sections. The cells are attached to chemically modified micrscope slides and the DNA is subsequently stretched using the halo technique (Vaandrager et al., 1996; also see Basic Protocol 6). Lymphoma cells are easily released from the tissue by incubation in PBS; tumor cells from other sources might be more difficult to suspend, and isolation procedures based on protease treatment or repeated freeze-thawing may have to be conducted for cell isolation. Additional Materials (also see Basic Protocol 6) Tumor tissue 1× PBS (see recipe), 0°C Cryostat 1.5-ml microcentrifuge tubes AES-coated microscope slides (see Support Protocol 1) 1. Cut 40- to 50-µm sections of tumor tissue using a cryostat. 2. Place section in a 1.5-ml microcentrifuge tube containing 1× PBS at 0°C. Pipet solution up and down a few times to create a single-cell suspension.
Molecular Cytogenetics
8.2.13 Current Protocols in Cytometry
Supplement 16
3. Spread cells on an AES-coated microscope slide and let them attach for 2 min. 4. Create DNA fibers (see Basic Protocol 6, steps 2 to 9). BASIC PROTOCOL 7
PREPARATION OF DNA FIBERS BY MOLECULAR COMBING Cloned DNA is allowed to bind to a microscope slide that has been silanized according to an especially rigorous procedure (see Support Protocol 2), after which fiber-FISH can be performed (Bensimon et al., 1994; Weier et al., 1995a). This method is referred to as “molecular combing” (UNIT 8.10). Materials Cosmid, P1, BAC, or YAC DNA prepared for molecular combing (see recipe) 1 × 10–6 M YOYO (Molecular Probes) 22 × 22–mm glass coverslips AES-coated microscope slides for molecular combing (see Support Protocol 2) 1. Dilute DNA with water to ∼10 pg/µl (equivalent to several thousand molecules per µl). 2. Stain the DNA with an equal volume of 10−6 M YOYO. 3. Place 2 µl of this DNA on a 22 × 22–mm coverslip. The coverslip should not be AES-treated.
4. Place an AES-coated microscope slide on top of the coverslip. The combing of the DNA and the concentration of the combed DNA can now be evaluated using a standard fluorescence microscope equipped with a filter set for FITC and a standard HBO 100 light source. The YOYO-stained DNA molecules are visible as green fibers.
5. Allow to dry overnight at 4° or 20°C. Drying at 4°C is much slower than at 20°C, but yields a better quality of combing.
6. Proceed to the in situ hybridization procedure (UNIT 8.3). If the DNA concentration is high enough, dilute 2 ìl of YOYO-stained DNA with 8 ìl AF solution (see recipe). This minimizes breakage of the DNA before further dilution with water. SUPPORT PROTOCOL 1
PREPARATION OF AES-COATED SLIDES (GENERAL PROCEDURE) This method is used for all protocols specifying aminopropyltriethoxysilane (AES)– coated slides, except that for molecular combing (see Basic Protocol 7), in which slides prepared by the more rigorous procedure (see Support Protocol 2) should be used. Materials 5% (v/v) aminopropyltriethoxysilane (AES; Sigma) in acetone Microscope slides, cleaned by dipping in 1:1 (v/v) ethanol/diethyl ether and wiping with lint-free Kimwipes (e.g., Fisher) 1. Incubate ethanol/ether–cleaned microscope slides 30 min in 5% AES. 2. Wash with water and air dry.
Basic Preparative Techniques for FISH
Preferably, fresh slides should be used.
8.2.14 Supplement 16
Current Protocols in Cytometry
PREPARATION OF AES-COATED SLIDES FOR MOLECULAR COMBING This more rigorous method for preparing aminopropyltriethoxysilane (AES)–coated slides is used only for molecular combing (see Basic Protocol 7 and UNIT 8.10).
SUPPORT PROTOCOL 2
Materials 18 M sulfuric acid 0.1% (v/v) aminopropyltriethoxysilane (AES; Sigma) in ethanol Absolute ethanol Nitrogen source Microscope slides Metal slide holders for dipping slides Rectangular staining dishes 65°C oven Slide box with dessiccant Sealable plastic bags for storing slides under nitrogen atmosphere 1. Place slides in a metal slide holder that permits dipping. Boil 10 min in distilled water, then air dry. 2. Immerse 30 to 40 min in 18 M sulfuric acid. CAUTION: Concentrated sulfuric acid is hazardous. Work in a fume hood.
3. Boil again for 10 min in distilled water and air dry. 4. Immerse 10 min in 0.1% AES/ethanol. 5. Rinse extensively in distilled water. 6. Dehydrate by immersing 3 min in absolute ethanol. 7. Dry slides in a 65°C oven in an upright position. 8. Store slides at 4°C in a sealed box inside a nitrogen-filled plastic bag. Age 2 to 6 weeks before use. The chemical modification of glass supports with silane is one of the most critical steps in the molecular combing procedure. Glass slides from different manufacturers and even from the same brand may yield very different qualities of fibers. Consequently, it is advised to prepare small batches of ∼50 slides, and test the DNA-binding and fiber-FISH performance on 2 to 3 slides using DNA molecules of known integrity. A good batch of slides binds most of the DNA molecules on one or both ends within a few minutes after the AES-treated slides are placed on top of the coverslip carrying the DNA.
PREPARATION OF POLY-L-LYSINE-COATED SLIDES Slides are cleaned with dilute HCl, water, and acetone, then coated with gelatin/sodium azide and finally with poly-L-lysine.
SUPPORT PROTOCOL 3
Materials 0.2 M HCl Acetone 0.15% (w/v) gelatin/0.3% (w/v) sodium azide (store up to 3 months at 4°C) 0.2% poly-L-lysine (Sigma; mol. wt. ∼500,000) in water Microscope slides Rectangular staining dishes
Molecular Cytogenetics
8.2.15 Current Protocols in Cytometry
Supplement 16
1. Immerse slides successively for 30 sec each in staining dishes containing 0.2 M HCl, distilled water, and acetone, then air dry at room temperature. 2. Immerse 5 min in 0.15% gelatin/0.03% sodium azide. Air dry overnight at room temperature. 3. Immerse 5 min in 0.2% poly-L-lysine. 4. Rinse by immersing 30 sec in distilled water, then air dry 1 hr at room temperature. 5. Repeat steps 3 and 4, then air dry overnight at room temperature. 6. Store slides up to 1 month at 4°C. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
AF (antifade) solution 1% (w/v) p-phenylenediamine 15 mM NaCl 1 mM NaH2PO4, pH 8.0 90% glycerol Store in glass tubes or bottles up to 1 month at 4°C Colcemid solution Stock solution (0.25%): Dissolve 25 mg Colcemid (demecolcine; Sigma) in 10 ml 1× PBS (see recipe). Filter sterilize, divide into aliquots, and store up to 5 years at −20°C. Working solution (0.0025%): Immediately before use, thaw an aliquot of stock solution and dilute 1:100 in 1× PBS (see recipe). DNA for molecular combing Isolate plasmid, cosmid, P1, BAC, or YAC DNA using alkaline lysis (Birnboim and Doly, 1979; Weier et al., 1995b). Linearize the DNA by restriction enzyme digestion, and purify by PFGE (Weier et al., 1995a,b). Establish the integrity of the prepared DNA as follows (also see Weier et al., 1995a,b). Stain small aliquots of DNA (typically 1 µl) with 0.5 µM YOYO (Molecular Probes). Resuspend DNA in AF solution (see recipe) or double-distilled water. Place the resuspended DNA on a microscope slide and cover with a coverslip. Examine on a standard fluorescence microscope equipped with a filter set for FITC and a standard HBO 100 light source. Formaldehyde fixative 1× PBS (see recipe) containing: 1% (v/v) formaldehyde (from 37% stock) 50 mM MgCl2 (from 1 M stock) Prepare fresh Hypotonic buffer Prepare (fresh for each series of slides to be prepared) a solution containing 50 mM KCl, 10 mM MgSO4, and 5 mM HEPES. Adjust the pH to 8.0 with a few drops of 1 M NaOH. Immediately before use, warm solution up to 37°C and add dithiothreitol (DTT) to a final concentration of 3 mM. Basic Preparative Techniques for FISH
8.2.16 Supplement 16
Current Protocols in Cytometry
Lysis buffer 0.5% (w/v) SDS 50 mM EDTA 200 mM Tris⋅Cl, pH 7.4 (APPENDIX 2A) Store up to 6 months at room temperature Phosphate-buffered saline (PBS), 10× 80 g NaCl 2 g KCl 15 g Na2HPO4⋅2H2O 1.2 g KH2PO4 H2O to 1 liter Adjust pH to 7.4 with 1 M HCl if necessary Store up to 1 month at room temperature Protease buffer, 10× 1 M Tris⋅Cl, pH 7.2 0.7 M NaCl Store up to 1 month at room temperature Proteinase digestion solution 2 mg/ml proteinase K 50 mM disodium EDTA 1% (w/v) N-laurylsarcosine Prepare fresh RNase A solution Dissolve 50 mg RNase A (DNase-free; Boehringer Mannheim) in 5 ml 2× SSC (APPENDIX 2A). Divide into 500 µl aliquots in microcentrifuge tubes, then put the tubes in a boiling water bath for 5 min (to remove any DNase activity). Cool to room temperature and store ≤1 year at −20°C. Solutions for halo technique Solution A: 0.5% (v/v) Nonidet P-40 (NP-40) 10 mM MgCl2 0.5 mM CaCl2 25 mM Tris⋅Cl, pH 8.0 (APPENDIX 2A) 1 mM phenylmethylsulfonyl fluoride (PMSF) Prepare fresh Solution B: 2 M NaCl 0.2 mM MgCl2 25 mM Tris⋅Cl, pH 8.0 (APPENDIX 2A) Prepare fresh
Solution C: Solution B containing 50 µg/ml ethidium bromide or 40 µg/ml propidium iodide Prepare fresh Propidium iodide causes less nonspecific background fluorescence than ethidium bromide. Solution D: 0.2 M NaCl 0.2 mM MgCl2 25 mM Tris⋅Cl, pH 8.0 (APPENDIX 2A) Prepare fresh Solution E: 0.2 mM MgCl2 25 mM Tris⋅Cl, pH 8.0 (APPENDIX 2A) Prepare fresh Molecular Cytogenetics
8.2.17 Current Protocols in Cytometry
Supplement 16
COMMENTARY Background Information Metaphase chromosomes The number of protocols for the preparation of blood metaphase chromosome slides for FISH almost equals the number of investigators who apply them. All such protocols have the following in common: stimulation of cell division of white blood cells by phytohemagglutinin, metaphase arrest by Colcemid, treatment of cells with a hypotonic solution, and fixation of the cells in a mixture of methanol and glacial acetic acid. There has been great variation during past decades regarding use of isolated lymphocytes versus whole blood for culture purposes, the composition of the chromosome culture medium, the composition of the hypotonic solution, and even the ratio of methanol and acetic acid in the fixative. Interphase cytogenetics With the advent of fluorescence in situ hybridization (FISH) techniques for the analysis of chromosome copy number or structure in interphase cells (i.e, interphase cytogenetics), the diagnostic and prognostic potential of cytogenetics has been augmented considerably (Cremer et al., 1986). The relevant molecular cytogenetic information extractable from cells after in situ hybridization is contained in the number of in situ hybridization domains (spots), their color, their spatial positions, and their intensities (Emanuel, 1993). Deviations from the normal patterns of marked regions permit the identification of cells that are aberrant in chromosomal composition. For example, in a diploid cell the deletion of a specific chromosome (or part thereof) leads to a reduction of the number of hybridization spots from 2 to 1 (monosomy), whereas a duplication leads to an increase from 2 to 3 (trisomy). Structural chromosome aberrations—e.g., translocations and inversions— can also be studied in the interphase nucleus by analyzing in situ hybridization patterns of signals (number, color, and position) from probes flanking or spanning the breakpoint (Arnoldus et al., 1990; Tkachuk et al., 1990; Bentz et al., 1994). Thus, interphase cytogenetics contributes to establishing clinico-cytogenetic relationships (see, for example, Takahashi et al., 1994).
Basic Preparative Techniques for FISH
Classification problems in interphase cytogenetics Although strategies for detection of cytogenetically aberrant cells by FISH are in theory
simple and straightforward, in practice they are fallible because false classification of hybridization spot number or patterns occurs. Causes of false cell classifications are either technical or biological in nature. Technical causes may be insufficient hybridization and detection efficiency, as well as colocalization of hybridization domains in the two-dimensional microscopic projection of the three-dimensional objects. The latter phenomenon can be enhanced by biological considerations such as specific chromatin organization of the chromosomal region under investigation. Somatic pairing of centromere regions represents an extreme example of this situation (Lewis et al., 1993). When a decision has to be made on molecular-cytogenetic normality or abnormality of a cell sample, the problem of false classification becomes particularly prominent if the fraction of aberrant cells is relatively small. Kibbelaar et al. (1993) and Carothers (1994) illustrated this point in terms of the number of cells that must be analyzed to reach a decision on the presence or absence of truly aberrant cells. For example, at a false-classification rate of 5%, ∼300 cells have to be evaluated to reach a decision on absence or presence of 5% aberrant cells at the 95% to 100% confidence interval. At a false classification rate of 1%, ∼1500 cells have to be evaluated to decide upon absence or presence of 1% aberrant cells at the same confidence interval. The manual analysis of FISH spots and/or patterns in so many cells in many patient samples is tedious. Assistance in the evaluation process by an automatic “spot counting” device is desirable. Such cytometric systems are currently under development (Vrolijk et al., 1996). Both manual and automatic screening of interphase cytogenetic results are considerably facilitated by sample preparation protocols that lead to a high density of cells. DNA for fiber-FISH Extended DNA fibers can be prepared from unfixed cells, melted pulsed-field gel electrophoresis blocks containing cells or DNA clones, cultured cells adhering to microscope slides, and pure clones of DNA. The protocols in this unit are designed to achieve (partial) linearization of DNA from fixed and unfixed cells—either mechanically or biochemically— as well as from purified cloned DNA. Although part of the DNA still remains less linearized or even unstretched, the slides prepared according to the described protocols are suited for high-
8.2.18 Supplement 16
Current Protocols in Cytometry
resolution FISH (“fiber-FISH”). In fact, it is necessary to experimentally optimize each protocol for specific cell types to determine the degree to which decondensation of the chromatin will take place. In each protocol the times for the different steps have been taken as guidelines, and therefore are not a guarantee for successful fiber-FISH. Some trial-and-error experimentation is required, for instance, to prevent loss of DNA during the entire series of procedures, including FISH itself. The use of coated microscope slides is almost a prerequisite, except in the DNA halo preparation technique, where coating of the slide is caused by proteins in the culture medium. Time, laboratory infrastructure such as culture facilities, and cell type determine which protocol is to be preferred. Obviously Basic Protocols 1, 2, and 5 seem to be the most simple, whereas Basic Protocols 3 and 4 are rather laborious and require specialized equipment. Moreover, the choice of protocol mainly depends on the question to be answered by fiber-FISH.
Critical Parameters and Troubleshooting Metaphase chromosomes Either ready-to-use culture media or media prepared in the laboratory can be used for growing the cells. For special applications, agents like thymidine, bromodeoxyuridine (Takahashi et al., 1990), and ethidium bromide may be added after culturing to get longer (pro)metaphase chromosomes. Fixation can be done at ambient temperature or 4°C. The way the fixative is added to the cell suspension varies from person to person as well. Heat lamps have been used to get a better release of chromosomes by quickly evaporating the fixative after cells have been dropped onto microscope slides. Some investigators vary the height from which the cells are dropped, to obtain the best spreading of the chromosomes. Donor-to-donor variation, the physical condition of the donor at the time the blood is taken, “ambient” temperature, and humidity also play a role in preparation of optimal chromosome slides for FISH. Obviously, a standard protocol does not exist. With blood from the same donor, Basic Protocol 1 provides more or less well defined standard chromosome preparations suitable for FISH. In Basic Protocol 1, the “standard” hypotonic buffer (75 mM KCl) has been replaced by a buffer containing KCl, MgSO4, HEPES, and dithiothreitol (DTT), which has been
adapted from chromosome-isolation protocols used for flow cytometric purposes (Trask et al., 1985). In a direct comparison, it was found that this buffer yielded many more metaphases than the “standard” 75 mM KCl. The way the fixative is added to the cell suspension, which in the authors’ view is one of the most critical steps, is designed in such a way that the cells are constantly kept in movement while gradually being fixed. This ensures a well-fixed single-cell suspension without too many cell clumps. Preparations are made by placing a few drops of cell suspension on a wet microscope slide (i.e., one that has been moistened by breathing on it or by dipping it in ice-cold water) from a distance of ∼30 cm. Variation in the distance may vary the spreading of the chromosomes; a suitable way to judge the preparation is by phase-contrast microscopy using a 25× objective lens. This provides insight into the morphological appearance of the chromosomes—i.e., shape, length and spreading of the metaphases. Under phase-contrast microscopy, residual cytoplasm is visible as greyish background speckles around and between the chromosomes. Remnants of cytoplasm negatively influence the quality of FISH. Acetic acid treatment—which is described in Basic Protocol 1 as a means of improving metaphase spreads that do not show the characteristic features of “good” spreads—is, in the experience of the authors, the most simple, reliable and flexible means of optimizing the spreads. However, the time between the drying of the preparations and the dipping of the slides into the acetic acid is critical. The step in Basic Protocol 1 for acetic acid treatment is very well suited for releasing chromosomes from old samples. It should be noted, however, that the acetic acid treatment has a negative effect on the morphology of the chromosomes. The banding of the chromosomes may deteriorate, which is of particular importance when FISH is used for mapping purposes. Once optimal conditions have been established a series of one hundred or more preparations may be made, air dried for at least 24 hr, and stored in 70% ethanol at 4°C for at least several months. For comparative genomic hybridization (CGH) it is recommended to use freshly prepared chromosome slides. The pretreatments of chromosome preparations before actual in situ hybridization, described in Basic Protocol 1, are designed to increase the accessibility of the probe fragments and immunocytochemical detection re-
Molecular Cytogenetics
8.2.19 Current Protocols in Cytometry
agents used in FISH. RNase A treatment removes endogenous RNA, and is particularly recommended when high-complexity probes are used. Cosmids, YACs, and chromosome libraries are often contaminated with vector DNA, which cross-hybridizes with endogenous RNA. Pepsin treatment is included to optimize FISH results. The optimal conditions for pretreatment should be determined experimentally (with respect to time and concentration). Batches of reagents from different manufacturers may give rise to different results, and even batch-to-batch variation from the same company has been reported. Post-fixation with formaldehyde is meant to restore chromosomal morphology that has deteriorated during the pepsin treatment. The RNase and pepsin pretreatments should immediately be followed by the FISH procedure; storage of any kind is not recommended.
Basic Preparative Techniques for FISH
Interphase cytogenetics Basic Protocol 2 integrates preparation techniques from cytogenetics and automated analytical cytology for use in interphase cytogenetic analysis; it was originally designed for automated screening. Manual screening, however, also profits considerably from its specific features. In automated analytical cytology, it is well established that cell density and overlap are decisive factors in gaining speed of analysis (Van Driel-Kulker et al., 1980). Preparation techniques derived from this field include bucket centrifugation and unit-gravity sedimentation as well as filter collection and imprint methods (see Oud et al., 1986 for literature entries in this field). The classical way of preparing cytogenetic preparations is by dropping cells that have been fixed in 3:1 methanol/acetic acid on microscope slides. This is easy to do, but it is difficult to obtain high cell density and to minimize touching or overlapping cells using this method. The bucket approach described in Basic Protocol 2 provides high cell densities with minimal overlap of cells. It can be implemented easily and requires little investment; cell loss is minimal. An additional advantage is that cells are deposited on fixed locations on the slide. For manual screening this is practical, but for automated screening it is essential because no instrument time will be wasted in trying to find objects of interest on empty parts of the slide. In situ hybridization is in principle compatible with a large variety of fixatives and storage media. The bucket centrifugation protocol
emerged as optimal for automated screening purposes following detailed analysis of a large number of preparative conditions for blood lymphocytes (Van de Rijke et al., 1996). It deviates from standard cytogenetic fixation practice in that only methanol is used to fix and store cells. Cells fixed and stored in methanol fully retain their FISH performance after >3 years of storage, whereas, in the authors’ experience, methanol/acetic acid–fixed samples have needed individual optimization (e.g., enhanced pepsin treatment). So that fresh blood cell samples may be analyzed by classical cytogenetic means as well as by interphase cytogenetics, it is recommended that the sample be split into two parts and that one be processed according to routine cytogenetic practice (i.e., mitogenic induction, culturing, metaphase arrest, and methanol/acetic acid fixation) and the other according to Basic Protocol 2. If for logistic reasons this is not feasible, methanol/acetic acid–fixed cells should preferably be resuspended in methanol for long-term storage while awaiting (automated) interphase cytogenetic analysis. Samples stored in 100% methanol are centrifuged in 50% acetic acid to prepare slides by bucket centrifugation. This measure is designed to remove cytoplasm from the cells, which in turn should improve image contrasts. The acetic acid treatment is based on the observation that conventionally prepared metaphase chromosome preparations that still have surrounding cytoplasm can be cleansed of it by rinsing with acetic acid. The 50% acetic acid centrifugation medium was found to be optimal on the basis of criteria such as cell loss and clumping, cell size, and quality of FISH signals (Van de Rijke et al., 1996). Cells fixed with 70% ethanol—as in the Vindelov preparation technique for flow cytometric DNA analysis (Vindelov and Christensen, 1990)—can be centrifuged on slides identically with similar FISH results. To minimize bias in nucleus selection by an automated system, the cells should preferably be uniform in shape and size. The optional borate swelling procedure is designed for this purpose. The effect of this on blood lymphocyte preparations is that the smaller nuclei swell, but the larger ones shrink. Cells from archival, clinical blood culture samples that have been fixed with methanol/acetic acid and centrifuged on slides in 50% actetic acid show a similar behavior in response to borate swelling. However, probably as a result of the mitogenic induction, the resulting sizes of the nuclei are
8.2.20 Current Protocols in Cytometry
much more heterogeneous. Consequently, the size criterion subsequently used in the imageanalysis program has to be adjusted to a wider range for such specimens, leading to a higher frequency of selection of doublets and artifacts. As stated earlier, the bucket centrifugation protocol was developed originally for blood lymphocytes, but works equally well for nuclei isolated from formalin-fixed, paraffin-embedded tissue stored in methanol. The borate treatment, however, has no effect on such nuclei, as a consequence of the protein cross-linking during fixation. The pepsin treatments are designed to maximize DNA target accessibility with respect to reagents such as labeled DNA probes and antibody conjugates. The concentrations and incubation times indicated in the protocol are good hints, but different concentrations should be tried if results are suboptimal. For direct fluorochrome-labeled probes, the pepsin treatment has proved to be generally not necessary. However, it reduces the autofluorescence level, which may help in improving contrasts and thus ease of analysis. Cell density is critical for automated screening. Cell or nuclei concentrations should be adjusted as necessary in the storage or bucket centrifugation media. Loss of cells or nuclei may occur during processing after bucket centrifugation when air drying is carried out for too short a period of time. Preferably air drying should be carried out overnight.
Anticipated Results In the authors’ laboratory, the metaphase chromosome procedure (see Basic Protocol 1) provides preparations suited for most FISH applications requiring metaphase chromosomes from blood cells. The interphase chromatin procedure (see Basic Protocol 2) leads to microscopic preparations that are amenable to rapid manual as well as automated screening of FISH results. Typically, densities range from 400 to 800 cells/mm. Given the spatial resolution of fluorescence microscopy (in practice equal to ∼0.3 µm), genomic resolution of FISH is determined by the degree of condensation of the target DNA. Metaphase chromosomes provide a resolution of 1 to 3 Mb (Lichter et al., 1990). For less condensed chromatin, as in interphase nuclei, resolution has been determined to be 50 kb, ranging up to 1 to 2 Mb (Trask et al., 1989). For chromatin fibers, where the DNA is (fully) decondensed, the limit of resolution has been proven to be in the order of 1 kb, ranging up to
∼ 0.5 to 1 Mb (Haaf and Ward, 1994; Florijn et al., 1995).
Time Considerations From culturing cells to beginning the actual FISH metaphase chromosome preparation (see Basic Protocol 1) takes ∼5 days, according to the following time schedule: 89 hr of culture (including 17 hr of thymidine treatment); 2 hr Colcemid treatment; 20 min incubation in the hypotonic buffer; 5 centrifugation steps at 10 min each; three 10-min fixations with a 1-hr interval between the second and the third fixation; 10 to 60 min of preparing chromosome slides (depending on whether or not acetic acid treatment is necessary); and 24 hr air drying of preparations. The pretreatments take ∼2.5 hr. Among the methods for preparing interphase cells for interphase cytogenetics (see Basic Protocol 2), preparation of methanol suspensions of blood lymphocytes and nuclei from paraffin-embedded tissues takes 3 and 4 hr, respectively. Such supensions can be stored for years without loss of FISH performance. Bucket-centrifugation takes ∼15 min, but the subsequent air drying needed for minimizing cell loss during further processing is overnight. Pretreatments of microscopic preparations for FISH take 2 to 2.5 hr. Multiple samples can be processed in parallel, the number being dependent on available equipment. In summary, hands-on time is 4 to 7.5 hr; the total time period needed is 20 to 24 hr. The different protocols to make preparations suited for fiber-FISH using cell suspensions take variable times in the following order: Basic Protocol 5, 4 hr; Basic Protocol 6, 1 hr; Basic Protocols 3 and 4, 30 min each. If the time of drying is taken into account (overnight) Protocol 7 takes >17 hr. But generally, all protocols take about the same time, as overnight drying of extended DNA preparations is recommended anyway.
Literature Cited Arnoldus, E.P.J., Wiegant, J., Noordermeer, I.A., Wessels, J.W., Beverstock, G.C., Grosveld, G.C., Van der Ploeg, M., and Raap, A.K. 1990. Detection of the Philadelphia chromosome in interphase nuclei. Cytogenet. Cell Genet. 54:108-111. Bensimon, A., Simon, A., Chiffaudel, A., Croquette, V., Heslot, F., and Bensimon, D. 1994. Alignment and sensitive detection of DNA by a moving interface. Science 265:2096-2098. Bentz, M., Cabot, G., Moos, M., Speicher, M.R., Ganser, A., Lichter, P., and Dohner, H. 1994. Detection of chimeric BCR-ABL genes on bone marrow samples and blood smears in chronic
Molecular Cytogenetics
8.2.21 Current Protocols in Cytometry
myeloid and acute lymphoblastic leukemia by in situ hybridization. Blood 83:1922-1928. Birnboim, H.C. and Doly, J. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucl. Acids Res. 7:1513-1523. Bosman, F.T., Van der Ploeg, M., Van Duijn, P., and Schaberg, A. 1975. Chromosome preparations of human blood lymphocytes; evaluation of techniques. Genetica 45:425-433. Carothers, A.D. 1994. Counting, measuring and mapping in FISH-labelled cells: Sample size considerations and implications for automation. Cytometry 16:298-304. Cremer, T., Landegent, J.E., Brueckner, A., Scholl, H.P., Schardin, M., Hager, H.D., Devilee, P., Pearson, P., and Van der Ploeg, M. 1986. Detection of chromosome aberrations in the human interphase nucleus by visualization of specific target DNAs with radioactive and non-radioactive in situ hybridization techniques: Diagnosis of trisomy 18 with probe L1.84. Hum. Genet. 74:346-352. Emanuel, B.S. 1993. The use of fluorescence in situ hybridization to identify human chromosomal anomalies. Growth Genet. Hormones 9:6-12. Fidlerová, H., Senger, G., Kost, M., Sanseau, P., and Sheer, D. 1994. Two simple procedures for releasing chromatin from routinely fixed cells for fluorescence in situ hybridization. Cytogenet. Cell Genet. 65:203-205. Florijn, R.J., Blonden, L.A.J., Vrolijk, H., Wiegant, J., Vaandrager, J-W., Baas, F., den Dunnen, J.T., van Ommen, G-J.B., and Raap, A.K. 1995. Highresolution DNA fiber-FISH for genomic DNA mapping and colour bar-coding of large genes. Hum. Molec. Genet. 4:831-836. Haaf, T. and Ward, D.C. 1994. High resolution ordering of YAC contigs using extended chromatin and chromosomes. Hum. Molec. Genet. 3:629-633. Heiskanen, M., Karhu, R., Hellsten, E., Peltonen, L., Kallioniemi, O.P., and Palotie, A. 1994. High resolution mapping using fluorescence in situ hybridization to extended DNA fibers prepared from agarose-embedding cells. Biotechniques 17:928-933. Kibbelaar, R.E., Kok, F., Dreef, E.J., Kleiverda, J.K., Cornelisse, C.J., Raap, A.K., and Kluin, P.M. 1993. Statistical methods in interphase cytogenetics: An experimental approach. Cytometry 14:716-724. Lewis, J.P., Tanke, H.J., Raap, A.K., Beverstock, G.C., and Kluin-Nelemans, H.C. 1993. Somatic pairing of centromeres and short arms of chromsome 15 in the hematopoietic and lymphoid system. Hum. Genet. 92:577-582. Lichter, P., Tang, C.C., Call, K., Hermanson, G., Evans, G.A., Housman, D., and Ward, D.C. 1990. High resolution mapping of human chromosome 11 by in situ hybridization with cosmid probes. Science 247:64-69.
Oud, P.S., Haag, D.J., Zahniser, D.J., Ramaekers, F.C.S., Huysman, A.C.L.M., Veldhuizen, J.A.M., Verheyen, R.H.M., Verrijp, K., Broers, J.L.V., Herman, C.J., and Vooijs, G.P. 1986. Cytopress: Automated slide preparation of cytologic material from suspensions. Cytometry 7:8-17. Parra, I. and Windle, B. 1993. High resolution visual mapping of stretched DNA by fluorescent hybridization. Nature Genet. 5:17-21. Takahashi, E., Hori, T., O’Connell P., Leppert, M., and White, R. 1990. R-banding and nonisotopic in situ hybridization: precise localization of the human type II collagen gene (COL2A1). Hum. Genet. 86:14-16. Takahashi, S., Qian, J., Brown, J.A., Alcaraz, A., Bostwick, D.G., Lieber, M.M., and Jenkins, R.B. 1994 Potential markers of prostate cancer aggressiveness detected by fluorescence in situ hybridization in needle biopsies. Cancer Res. 54:3574-3579. Tkachuk, D.C., Westbrook, C.A., Andreeff, M., Donlon, T.A., Cleary, M.L., Suryanarayan, K., Homge, M., Redner, A., Gray, J.W., and Pinkel, D. 1990. Detection of bcr-abl fusion in chronic myelogeneous leukemia by in situ hybridization. Science 250:559-562. Trask, B., Van den Engh, G., Landegent, J., Jansen in de Wal, N., and Van der Ploeg, M. 1985. Detection of DNA sequences in nuclei in suspension by in situ hybridization and dual beam flow cytometry. Science 230:1401-1403. Trask, B., Pinkel, D., and Van Den Engh, G. 1989. The proximity of DNA sequences in interphase nuclei is correlated to genomic distance and permits ordering of cosmids spanning 250 kilobase pairs. Genomics 5:710-717. Vaandrager, J-W., Schuuring, E., Zwikstra, E., de Boer, C.J., Kleiverda, K.K., van Krieken, J.H.J.M., Kluin-Nelemans, H.C., van Ommen, G-J.B., Raap, A.K., and Kluin, P.M. 1996. Direct visualization of dispersed 11q13 chromosomal translocations in mantle cell lymphoma by multicolor DNA fiber fluorescence in situ hybridization. Blood 88:1177-1182. Van de Rijke, F.M., Vrolijk, H., Sloos, W.C.R., Tanke, H.J., and Raap, A.K. 1996. Sample preparation and in situ hybridization for automated molecular cytogenetic analysis of white blood cells. Cytometry 24:151-157. Van Driel-Kulker, A.M.J., Ploem-Zaaijer. J.J., Van der Zwan-Van der Zwan, M., and Tanke, H.J. 1980. A preparation technique for exfoliated and aspirated cells allowing different staining procedures. Anal. Quant. Cytol. 2:243-246. Vindelov, L. and Christensen, I.J. 1990. An integrated set of methods for routine flow cytometric DNA analysis. Methods Cell Biol. 33:127-37. Vogelstein, B., Pardoll, D.M., and Coffey, D.S. 1980. Supercoiled loops and eucaryotic DNA replication. Cell 22:79-85.
Basic Preparative Techniques for FISH
8.2.22 Current Protocols in Cytometry
Vrolijk, H., Sloos, W.C.R., Van de Rijke, F.M., Mesker, W.E., Netten, H., Young, I.T., Raap, A.K., and Tanke, H.J. 1996. Automation of spot counting in interphase cytogenetics using brightfield microscopy. Cytometry 24:158-166. Weier, H-U.G., Wang, M., Mullikin, J.C., Zhu, Y., Cheng, J-F., Greulich, K.M., Bensimon, A., and Gray, J.W. 1995a. Quantitative DNA fiber mapping. Hum. Molec. Genet. 4:1903-1910. Weier, H-U.G., Rhein, A.P., Shadravan, F., Collins, C., and Polikoff, D. 1995b. Rapid physical mapping of the human trk proto-oncogene (NTRK1) gene to human chromosome 1q21-22 by P1 clone selection, fluorescence in situ hybridization (FISH) and computer-assisted microscopy. Genomics 26:390-393. Wiegant, J., Galjart, N.J., Raap, A.K., and d’Azzo, A. 1991. The gene encoding human protective protein (PPGB) is on chromosome 20. Genomics 10:345-349. Wiegant, J., Kalle, W., Mullenders, L., Brookes, S., Dauwerse, J.M.N., Van Ommen, G-J.B., and Raap A.K. 1992. High-resolution in situ hybridization using DNA halo preparations. Hum. Molec. Genet. 1:587-592.
Key References Arnoldus et al., 1990. See above.
Emanuel, 1993. See above. Elegant overview of FISH. Parra and Windle, 1993. See above. One of the first reports on fiber-FISH. Rooney, D.E. and Czepulkowski, B.H. (eds.) 1992. Human Cytogenetics. A Practical Approach. Oxford University Press, New York. A fully comprehensive manual of established cytogenetic protocols. Tkatchuk et al., 1990. See above. Along with Arnoldus et al., 1990, one of the first two reports on translocation detection in interphase cells. Van de Rijke et al., 1996. See above. Describes sample preparation techniques. Van Driel-Kulker et al., 1980. See above. Describes bucket centrifugation. Verma, R.S. and Babu, A. (eds.) 1995. Human Chromosomes. Principles and Techniques, 2nd ed. McGraw-Hill, New York. Covers the whole field of human (molecular) cytogenetics.
Along with Tkatchuk et al., 1990, one of the first two reports on translocation detection in interphase cells.
Vrolijk et al., 1996. See above.
Carothers, 1994 and Kibbelaar et al., 1993. See above.
Wiegant et al., 1992. See above.
Describes statistical approaches for interphase cytogenetics. Cremer et al., 1986. See above. Classical article on interphase cytogenetics.
Describes automated “spot” counter.
One of the first reports on fiber-FISH.
Contributed by J. Wiegant and A.K. Raap Leiden University Leiden, The Netherlands
Molecular Cytogenetics
8.2.23 Current Protocols in Cytometry
Probe Labeling and Fluorescence In Situ Hybridization
UNIT 8.3
Fluorescence in situ hybridization (FISH) is a multistep procedure that involves the following general sequence of events: 1. 2. 3. 4. 5. 6. 7.
Fixation of biological samples and preparation of microscopic specimens. Pretreatments of microscopic preparations. Probe labeling. Denaturation of in situ target DNA. In situ hybridization and post-hybridization washing. Immunocytochemistry. Microscopy.
In this unit, basic protocols for probe labeling, denaturation of in situ target DNA, in situ hybridization, and post-hybridization washing are described in detail. Support protocols for probe labeling also cover probe purification and quality controls. For further information on the development and use of FISH, see Chapter 8 introduction and UNIT 8.1. The quality of FISH results is as good as the weakest link in the chain of procedural steps. For some of the steps (e.g., probe labeling, microscopy), independent quality control is possible. For others (e.g., fixation, pretreatment, denaturation), independent quality control is not possible and, when needed, the actual FISH results are used to guide optimization. The steps not amenable to independent quality control are often interdependent and may require optimization relative to each other. One should realize that a generalized protocol for all cell types and probes does not exist. For many FISH applications, however, the multitude of experimental variables have been documented fairly accurately, so that with limited optimization studies, an operational in situ hybridization protocol can be designed rapidly. In this respect, an optimal FISH protocol is difficult to define, because a suboptimal protocol can still provide valuable FISH results. STRATEGIC PLANNING Probe Complexity and Repeat Suppression DNA with a wide range of sequence complexities can be used in FISH: oligonucleotides, small polymerase chain reaction (PCR) products, plasmids, phages, large-insert vectors (YACs and Alu-PCR products thereof, BACs, P1s), chromosome libraries, and whole genomes as well as PCR products thereof. The higher the sequence complexity—in other words, the number of base pairs (bp) contained in the probe—the higher the chance of nonspecific staining, because of the increased chance that dispersely occurring repeat sequences, such as Alu and Kpn repeats, may be present in the labeled probe DNA. However, by employing a preannealing step in which labeled repeat sequences are allowed to anneal to unlabeled DNA enriched for repeat sequences (e.g., human C0t1 DNA), these sequences can effectively be eliminated from participation in the ISH reaction. Thus without subcloning efforts, probes recognizing large target areas can be specifically detected in chromosomes and interphase nuclei with repeat suppression ISH procedures (Landegent et al., 1987; Pinkel et al., 1988; Lichter et al., 1990). Although probe complexity does not affect the condition of the probe labeling reaction, it does affect probe purification and target denaturation, and must be taken into account Contributed by J. Wiegant and A.K. Raap Current Protocols in Cytometry (1997) 8.3.1-8.3.21 Copyright © 1997 by John Wiley & Sons, Inc.
Molecular Cytogenetics
8.3.1
in deciding the composition of the hybridization solution and the in situ hybridization conditions. Therefore, a distinction is made here between high- and low-complexity probes or, in more practical terms, probes with and without dispersely occurring repeats (see Basic Protocols 5 and 6, respectively). High-complexity probes include genomic DNA clones in phage (15 kbp), cosmid (40 kbp), or YAC (up to 1 to 2 megabase pair [Mbp]) vectors, chromosome-specific DNA libraries (plasmid libraries or DOP-PCR products of flow-sorted or microdissected chromosomes), and total genomic DNA as used in comparative genomic hybridization (CGH; Kallioniemi et al., 1992; du Manoir et al., 1993). Such probes need repeat suppression during FISH. Plasmid probes with inserts ranging from a few hundred base pairs (bp) to a few kilobase pairs (kbp), and PCR products generated from such probes, are referred to as low-complexity probes. Such small probes may recognize either repetitive targets, such as chromosome-specific satellite DNAs, or unique sequences, as do cDNAs. In either case, they do not need suppression of repeats during FISH. Probe Labeling Methods Labeling of DNA probes for FISH is generally achieved by enzymatic incorporation of hapten- or fluorochrome-labeled deoxyribonucleoside triphosphates into DNA probes using E. coli DNA polymerase I and its Klenow fragment (in nick translation and random primed procedures, respectively) or Taq DNA polymerase (in PCR procedures). Mostly, dUTP labeled at the 5′ position of the uracil moiety is used as precursor. Several enzymatic labeling reactions important for FISH are presented here: nick translation (see Basic Protocol 1), random-primed labeling (see Basic Protocol 2), and labeling by the polymerase chain reaction (PCR; see Basic Protocol 3 and Alternate Protocol 1). In addition, strategies for probe labeling for multicolor FISH are presented (see Basic Protocol 4 and Alternate Protocol 2). The support protocols include probe purification by ethanol precipitation and/or gel filtration (see Support Protocol 1), preparation of probe stock solutions (see Support Protocol 1), and quality controls for size, labeling efficiency, and estimation by labeled probe concentration (see Support Protocol 2 and Support Protocol 3, respectively). The composition of the stock solutions to be prepared from the purified probe is dependent on probe complexity but not on labeling method. Therefore, Support Protocols 1 and 2 are useful for nick-translated, random-primed, and PCR-amplified probes. Dot blot assays (see Support Protocol 3) are generally useful only for nick-translated probes because the actual probe concentration is not known for random-primed or PCR probes without determination by spectrophotometric measurement or estimation by gel electrophoresis (see Support Protocol 2). The outcome of the FISH itself (especially when done with several dilutions of labeled probe) roughly indicates the amount of probe generated and its degree of labeling. The protocols for random-primed and PCR labeling are based on ones developed by Boehringer Mannheim; some information here is adapted from their technical material (Grünewald-Janho et al., 1996) by permission of Boehringer Mannheim.
Probe Labeling and Fluorescence In Situ Hybridization
8.3.2 Current Protocols in Cytometry
PROBE LABELING BY NICK TRANSLATION In nick translation, double-stranded DNA is treated with DNase I in the presence of magnesium ions. The resulting nicks provide 3′ hydroxyl groups that serve as primers for DNA synthesis catalyzed by the 5′ → 3′ polymerase activity of the Klenow fragment of DNA polymerase I. During synthesis, dNTP precursors, one of which is replaced by a labeled nucleotide, are incorporated in the growing chain of DNA while the nick is translated along the DNA by virtue of the 5′ → 3′ exonuclease activity carried by the enzyme (Rigby et al., 1977). If nonradioactive precursors are supplied to the enzyme, the resulting product is suitable as a FISH probe.
BASIC PROTOCOL 1
Nick translation kits specifically adapted for FISH are commercially available (Boehringer Mannheim and Enzo). They contain fixed amounts of the ingredients listed. These give satisfactory FISH results, but composition of the nick translation mixture by the user provides more flexibility, in particular regarding the DNase I concentration and hence the fragment size of the labeled probe. Materials 10× nick translation buffer (see recipe) 0.1 M DTT (APPENDIX 2A) Nucleotide mixture (see recipe) 1 mM labeled dUTP: biotin-16-dUTP or digoxigenin-11-dUTP (Boehringer Mannheim) or fluorochrome-labeled dUTP (Boehringer Mannheim, NEN Life Science Products, or Amersham) 1 µg/µl probe template DNA 10 U/µl DNA polymerase I (Promega) 1 µg/ml DNase I, freshly diluted (see recipe) 15°C water bath Additional reagents and equipment for preparation of labeled probe stock (see Support Protocol 1), determination of probe fragment size (see Support Protocol 2), and checking hapten incorporation by dot blot (see Support Protocol 3) 1. Prepare the following labeling mixture in a microcentrifuge tube kept on ice (50 µl total): 26 µl filtered distilled H2O 5 µl 10× nick translation buffer 5 µl 0.1 M DTT 4 µl nucleotide mixture 2 µl 1 mM hapten- or fluorochrome-labeled dUTP 1 µl 1 µg/µl probe template DNA 2 µl 10 U/µl DNA polymerase I 5 µl 1 µl/ml DNase I. Mix well and incubate 2 hr in a 15°C water bath. Adjust volumes of probe DNA and water if probe concentrations deviate from 1 ìg/ìl. It is important to accurately measure the DNA concentration. Read OD260 using a spectrophotometer that can accommodate small volumes (2 to 5 ìl). In setting up FISH with a given probe type, it is useful to analyze the effect of probe fragment size on FISH results by varying DNase concentrations (see Critical Parameters and Troubleshooting). Final concentrations of labeled dUTP and unlabeled dTTP are 40 and 8 ìM, respectively; these concentrations appear to be optimal (see Wiegant et al., 1996).
Molecular Cytogenetics
8.3.3 Current Protocols in Cytometry
2. Set aside a small aliquot to analyze the fragment size of the labeled probe (see Support Protocol 2) and to check hapten incorporation by a dot blot assay (see Support Protocol 3). 3. Purify the labeled probe (see Support Protocol 1). BASIC PROTOCOL 2
PROBE LABELING BY RANDOM PRIMING In a random-primed labeling reaction template DNA is denatured and annealed to short primers of random sequence. Starting from the 3′-OH end of the annealed primer, Klenow fragment synthesizes new DNA along the single-stranded substrate, incorporating the added hapten- or fluorochrome-labeled nucleotide (Feinberg and Vogelstein, 1983). This produces labeled probe suitable for FISH. NOTE: This protocol can be modified to accommodate equivalent random priming kits from other manufacturers. Materials Probe template DNA Digoxigenin–, biotin–, or fluorescein–high prime kit (Boehringer Mannheim) Boiling water bath Additional reagents and equipment for preparation of labeled probe stock (see Support Protocol 1), determination of probe fragment size (see Support Protocol 2), and checking hapten incorporation by dot blot (see Support Protocol 3) 1. In a 1.5-ml microcentrifuge tube, mix 1 µg probe template DNA and sterile, redistilled water to a final volume of 16 µl. For this standard protocol the amount of DNA can vary from 10 ng to 3 ìg. To label >3 ìg, scale up the volume of the sample and the amounts and volumes of all reaction components.
2. Denature the DNA by heating 5 min in a boiling water bath and chill 2 min on ice. 3. Add 4 µl of either digoxigenin–high prime, biotin–high prime, or fluorescein–high prime solution. Mix and microcentrifuge briefly. The high-prime mixture consists of 5× concentrated reaction buffer, 50% glycerol, 1 U/ìl Klenow, 1 mM each dATP, dCTP, and dGTP, 0.65 mM dTTP, and 0.35 mM X-dUTP (X = digoxigenin, biotin, or fluorescein).
4. Incubate 1 to 20 hr at 37°C. Longer incubation times (up to 20 hr) increase the yield of labeled DNA.
5. Set aside a small aliquot to analyze the labeled probe’s fragment size and to estimate probe concentration (see Support Protocol 2). 6. Purify the labeled probe (see Support Protocol 1).
Probe Labeling and Fluorescence In Situ Hybridization
8.3.4 Current Protocols in Cytometry
PROBE LABELING BY POLYMERASE CHAIN REACTION Optimal polymerase chain reaction (PCR) conditions are dependent on the sequence of the template DNA and the primers. Optimal concentrations of template, primers, Mg2+ ions, and polymerase, as well as the optimal incubation times and temperatures, should be determined experimentally for each new primer/template combination (Innis et al., 1990). The procedures given below are necessarily generalized and are for labeling with digoxigenin, biotin, or fluorescein.
BASIC PROTOCOL 3
NOTE: PCR samples are readily contaminated. Consult Innis et al. (1990) concerning techniques for avoiding PCR contamination. Materials 10× PCR buffer, pH 8.3 (see recipe) 25 mM MgCl2 10× PCR-dig, PCR-bio, or PCR-fluorescein labeling mix (see recipes) PCR primers 1 and 2 Template DNA 1 U/µl Taq DNA polymerase Mineral oil PCR thermal cycler Additional reagents and equipment for preparation of a labeled probe stock (see Support Protocol 1), determination of probe fragment size (see Support Protocol 2), and determination of labeling efficiency by dot blot (see Support Protocol 3) 1a. To label with digoxigenin or biotin: Set up the following reaction in a microcentrifuge tube placed on ice: 5 µl 10× PCR buffer, pH 8.3 2 to 10 µl 25 mM MgCl2 5 µl 10× PCR-dig or PCR-bio labeling mix 0.1 to 1 µM PCR primer 1 0.1 to 1 µM PCR primer 2 Template DNA: 1 to 100 ng human genomic or 10 to 100 pg plasmid 0.5 to 2.5 µl 1 U/µl Taq DNA polymerase H2O to 50 µl. Concentrations of MgCl2, primers, and template DNA must be determined empirically. For initial experiments, final concentrations of 1.5 mM MgCl2, 0.3 ìM of each primer, and 50 ng human genomic or 50 pg plasmid DNA are recommended.
1b. To label with fluorescein: Set up the following reaction in a microcentrifuge tube on ice: 10 µl 10× PCR buffer, pH 8.3 12 to 20 µl 25 mM MgCl2 10 µl 10× PCR-fluorescein labeling mix 0.1 to 1 µM PCR primer 1 0.1 to 1 µM PCR primer 2 Template DNA: 1 to 100 ng human genomic or 10 to 100 pg plasmid 1 to 5 µl 1 U/µl Taq DNA polymerase H2O to 100 µl. Concentrations of MgCl2, primers, and template DNA must be determined empirically, as described in step 1a.
Molecular Cytogenetics
8.3.5 Current Protocols in Cytometry
2. Mix gently and microcentrifuge briefly to collect the mixture at the bottom of the tube. Overlay with 100 µl mineral oil. 3. Carry out PCR using the following thermal cycles: Initial step: 30 cycles:
Final step:
7 min 45 sec 1 min 2 min 7 min
95°C 95°C 60°C 72°C 72°C
(denaturation) (denaturation) (annealing) (extension) (extension).
4. Stop the reaction by chilling on ice. 5. Carefully transfer the aqueous layer to a clean tube. Set aside a small aliquot to analyze the fragment size of the labeled probe and to estimate the probe concentration (see Support Protocol 2). 6. Purify the labeled probe (see Support Protocol 1). As an alternative to ethanol precipitation after the PCR reaction, the labeled probe DNA may be separated from primers and unincorporated nucleotides by gel electrophoresis on a 1% agarose gel. The band of the PCR product is then excised and separated from the agarose (Moore et al., 1994). This procedure provides not only an estimate of probe concentration but also a good impression of probe fragment size. ALTERNATE PROTOCOL 1
OPTIMIZED PCR LABELING OF PLASMID DNA FOR FISH This alternate protocol has been optimized for labeling of probes by PCR making use of plasmid template DNA, which generally gives good FISH results. Follow steps for Basic Protocol 3, substituting steps 1 and 3 as indicated below. Additional Materials (also see Basic Protocol 3) 10× PCR buffer: see recipe but adjust pH to 9 and add 1% (v/v) Triton X-100 Alternate nucleotide mixture: 2.5 mM dATP, dCTP, and dGTP and 1.5 mM dTTP 0.25 mM biotin-16-dUTP (Boehringer Mannheim) 0.25 mM digoxigenin-11-dUTP (Boehringer Mannheim) 0.25 mM fluorescein-dUTP (Boehringer Mannheim, NEN Life Science Products, or Amersham) 1 ng/µl plasmid template DNA 50 pM PCR primers 1 and 2 5 U/µl Taq DNA polymerase (e.g., Promega) 1. To a microcentrifuge tube on ice add the following: 10 µl 10× PCR buffer, pH 9, with 1% Triton X-100 10 µl alternate nucleotide mixture 5 µl 0.25 mM dig-, bio- or fluorescein-dUTP 1 µl 1 ng/µl plasmid template DNA 1 µl 50 pM primer 1 1 µl 50 pM primer 2 0.5 µl 5 U/µl Taq DNA polymerase 71.5 µl distilled H2O.
Probe Labeling and Fluorescence In Situ Hybridization
8.3.6 Current Protocols in Cytometry
2. Carry out PCR using the following thermal cycles: Initial step: 30 cycles:
Final steps:
5 min 95°C (denaturation) 1 min 55°C (annealing) 72°C (extension) 95°C (denaturation) 1 min 55°C (annealing) 5 min 72°C (extension).
PREPARATION OF PURIFIED PROBE STOCK SOLUTIONS There are two options for purification of labeled probe DNA: (1) gel filtration followed by ethanol precipitation and (2) ethanol precipitation alone. Purification by gel filtration generally leads to lower background, which may be of importance in the case of small unique FISH targets. Generally, high-complexity probes and probes for repeat targets can be purified by ethanol precipitation alone. Omitting gel filtration not only saves time, but also reduces loss of probe DNA.
SUPPORT PROTOCOL 1
The purified probe is dissolved in one of two hybridization solutions to make the probe stock solution. The solution used and the final probe concentration depend upon the probe complexity (see Reagents and Solutions, Table 8.3.1, and Commentary). Materials Probe labeling reaction mix (see Basic Protocol 1, 2, or 3 or Alternate Protocol 1) 0.5 M EDTA, pH 8 (APPENDIX 2A) MicroSpin columns (Pharmacia Biotech) 10 mg/ml sonicated and denatured fish sperm DNA (see recipe) 3 M sodium acetate, pH 5.6 (see recipe) 70% and 100% ethanol, −20°C 1 µg/µl C0t1 DNA (Life Technologies or Boehringer Mannheim) Hybridization solution I or II (see recipes and Table 8.3.1) Aspirator Perform gel filtration (optional) 1. Stop the labeling reaction by adding 5 µl of 0.5 M EDTA, pH 8. 2. Pass the labeling mixture over a MicroSpin column according to the manufacturer’s instructions. Proceed with steps 3 to 6.
Table 8.3.1
Conditions for Preparing Probe Stock Solutionsa
Probe type or target Chromosome-specific repeat Small unique Cosmid YAC (total yeast DNA) Chromosome library
% Formamide
Dextran sulfate
Excess C0t1 DNA
volume (µl)
Probe stock concentration (ng/µl)
60b
Nob
None
100
10
50c 50c 50c 50c
Yesc Yesc Yesc Yesc
None 50× 50× 3-10×
100 50 25 50
10 20 40 20
aFor PCR probes, add 2 to 10 mM EDTA (final). bAs in hybridization solution II (see complete recipe). cAs in hybridazation solution I (see complete recipe).
Molecular Cytogenetics
8.3.7 Current Protocols in Cytometry
Supplement 15
Perform ethanol precipitation 3a. Low-complexity probes: Precipitate the labeled probe by adding 50× excess sonicated fish sperm DNA, 1⁄10 vol of 3 M sodium acetate, pH 5.6, and 2.5 vol of 100% ethanol (−20°C). Mix well. 3b. High-complexity probes: Precipitate the labeled probe by adding 50× excess sonicated fish sperm DNA, 3× to 50× excess C0t1 DNA, 1⁄10 vol of 3 M sodium acetate, pH 5.6, and 2.5 vol of 100% ethanol (−20°C). Mix well. As a guideline, use a 3× to 10× excess of C0t1 DNA for chromosome libraries and 50× excess C0t1 DNA for cosmids and YACs. For other high-complexity probes, the optimal amount of C0t1 DNA needed to suppress repeats should be determined experimentally (see also Table 8.3.1).
4. Place the tube on ice for 30 min. 5. Microcentrifuge the DNA 30 min at 4°C. 6. Remove supernatant and wash the DNA pellet with a few hundred microliters 70% ethanol. Microcentrifuge an additional 5 min at 4°C. 7. Remove all traces of supernatant by suction, using a small pipet tip connected to an aspirator. Dry pellet 5 to 10 min at 37°C. When the DNA pellet is dried too long it may be difficult to dissolve the probe completely. However, small traces of ethanol may cause serious background problems when dextran sulfate is used in the hybridization solution.
8. Add the appropriate volume of hybridization solution I or II to obtain the recommended probe stock concentrations (see Table 8.3.1). Note that C0t1 DNA ethanol precipitates with the probe and is not added again at this stage. Mg2+ ions in PCR-labeled probes can lead to high background in FISH. To avoid this, add 0.5 M EDTA to a final concentration of 2 to 10 mM, depending on the amount of Mg2+ in the labeling reaction. Fairly high concentrations are used for YACs, because it is not mandatory to isolate the YAC DNA by pulsed-field gel electrophoresis. For other probe types, determine the optimal (stock) concentration experimentally.
9. Allow to dissolve 5 to 10 min at 37°C with occasional mixing and store the stocks up to several years at 4°C or −20°C. SUPPORT PROTOCOL 2
DETERMINATION OF PROBE FRAGMENT SIZE AND ESTIMATION OF PROBE CONCENTRATION In this protocol a sample of the labeled probe is run on a 1% agarose gel together with a DNA size marker and a serial dilution of a DNA standard of known concentration. This will give an impression of both the size of the fragments that have been generated and the concentration of the labeled probe. Optimally, probe fragments will be 200 to 500 bp in length.
Probe Labeling and Fluorescence In Situ Hybridization
Materials Agarose (e.g., Promega) 50× TAE buffer (APPENDIX 2A) 10 mg/ml ethidium bromide Probe labeling reaction mix (see Basic Protocol 1, 2, or 3 or Alternate Protocol 1) 6× gel loading buffer (APPENDIX 2A) DNA molecular size markers, (e.g., 1 kb DNA ladder, Life Technologies)
8.3.8 Supplement 15
Current Protocols in Cytometry
Biotin- digoxigenin-, or fluorescein-labeled DNA standard (e.g., Boehringer Mannheim) CAUTION: Ethidium bromide is a mutagen and must be handled carefully. 1. Dissolve 1 g agarose in 100 ml of 1× TAE buffer by carefully boiling. Cool to 55°C, add 3 µl of 10 mg/ml ethidium bromide, and mix. Pour gel and allow to solidify. 2. Mix 10 µl probe and 5 µl loading buffer and run sample on a 1% agarose gel, along with a DNA molecular size marker and a DNA standard, at 1 to 5 V/cm until the bromphenol blue and xylene cyanol have migrated the appropriate distance. 3. Examine and photograph the gel with UV light. For most accurate determination of fragment length, gel electrophoresis should in fact be performed under denaturing conditions, but this advantage will be gained at the cost of sensitivity of ethidium bromide fluorescence intensity. This protocol suffices for FISH purposes. See Sambrook et al. (1989) or Voytas (1988) for extensive description of gel electrophoresis.
DETERMINATION OF LABELING EFFICIENCY BY DOT BLOT ASSAY In this assay a serial dilution of the denatured hapten-labeled probe DNA is spotted on a membrane alongside a serial dilution of a standard sample, which is used as a means both of quantitatively gauging probe labeling and of controlling for success of staining in the dot blot assay itself. After immobilization of the DNA to the membrane, probe labeling is checked using a chromogenic immunocytochemical reaction. To this end, the membrane is incubated with an alkaline phosphatase (AP)-conjugated anti-hapten antibody or AP-streptavidin, after which the AP is detected with a substrate solution that produces a permanent blue precipitate. If 1 to 5 pg can be detected, the labeling is considered to be adequate. Fluorochrome incorporation can be checked conveniently after ethanol precipitation (see Support Protocol 1) by observing the fluorescence of the DNA pellet under UV illumination. Once labeling conditions have been settled, it may be useful to perform a quality control check on the newly labeled probes by FISH.
SUPPORT PROTOCOL 3
Materials Probe labeling reaction mix (see Basic Protocol 1) TE buffer (APPENDIX 2A) Hybond N+ membrane (Amersham) 2× SSC (APPENDIX 2A) 1 ng/µl biotin-, digoxigenin-, or fluorescein-labeled standard DNA (in TE buffer; Life Technologies or Boehringer Mannheim) TNB buffer (see recipe) AP-conjugated streptavidin, sheep anti-digoxigenin, or sheep anti-fluorescein (e.g., Boehringer Mannheim) TNT buffer (see recipe) NBT/BCIP substrate solution (see recipe) 80°C oven 1. After nick translation, dilute 1 µl probe-labeling reaction mix (20 ng) in 19 µl TE buffer (final 1 ng/µl). 2. Cut a 1 × 4–cm piece of Hybond N+ filter. 3. Wet the filter in 2× SSC and let it air dry. Molecular Cytogenetics
8.3.9 Current Protocols in Cytometry
4. Make serial dilutions (typically 1:2) of both the DNA standard and the probe sample in TE buffer to a final concentration of 1.4 pg/µl. 5. Denature the DNA 5 min at 100°C, then place on ice for 2 min. Microcentrifuge for a few seconds and mix. 6. Apply 1 µl of each dilution to the Hybond N+ filter. Mark orientation of spots on filter with a black ball-point pen. 7. Bake the filter 2 hr at 80°C. 8. Add 15 ml TNB buffer to a staining jar and incubate the filter 20 min at room temperature while shaking. 9. In 15 ml TNB buffer, mix the appropriate amount of AP-conjugated streptavidin (0.1 U/ml final), sheep anti-digoxigenin (0.15 U/ml final), or sheep anti-fluorescein (0.15 U/ml final). 10. Place the filter in the antibody solution in a clean jar and incubate 45 min at room temperature while shaking. 11. Wash the filter three times for 5 min each time in TNT buffer. 12. Place filter in freshly prepared NBT/BCIP substrate solution and incubate in the dark 2 to 16 hr at room temperature. To store membrane for records, wash in water, air dry, and store sealed or wrapped in plastic. BASIC PROTOCOL 4
COMBINATORIAL LABELING FOR MULTICOLOR FISH In combinatorial labeling, unique colors are created using multiple labels for a single probe. There are two strategies for combinatorially labeling a probe. It can be double-labeled by pooling two separate labeling reactions (resulting in a population of molecules with either one or the other label), or it can be double-labeled in a single reaction containing two distinctly labeled nucleotides (resulting in a population of molecules that are each double-labeled). A hybridization mix for multicolor FISH is made by combining single-, double-, and triple-labeled probes (see Table 8.3.3). Both strategies work (Nederlof et al., 1992), but the former is preferred from the practical point of view. In ratio-labeling, unique probes can be created using different ratios of the same two colors (e.g., 25% red/75% green versus 50%/50% and 75%/25%). As with combinatorial labeling, there are two strategies for preparing a mixture of ratio-labeled DNA probes. One can mix probes in the desired ratio (e.g., according to Table 8.3.4; also see Alternate Protocol 1) after each individual probe is separately labeled, or one can mix the probes after ratio-labeling during the enzymatic reaction by providing the desired ratio of haptenor fluorochrome-labeled nucleotides. Again, both strategies work, but the former is logistically preferable. This protocol describes strategies for seven-color combinatorial labeling; for five-color ratio-labeling, see Alternate Protocol 2. Both use the strategy of combining singly labeled probes. Variations can be made as desired.
Probe Labeling and Fluorescence In Situ Hybridization
8.3.10 Current Protocols in Cytometry
Materials Set of probe stock solutions, e.g.: probe 1: biotin labeled probe 2: fluorescein labeled probe 3: rhodamine labeled probe 4: biotin labeled probe 4: fluorescein labeled probe 5: biotin labeled probe 5: rhodamine labeled probe 6: fluorescein labeled probe 6: rhodamine labeled probe 7: biotin labeled probe 7: fluorescein labeled probe 7: rhodamine labeled. 1. Pipet an equal volume of each probe stock solution into a single tube and add hybridization solution to the desired concentration. RATIO-LABELING FOR MULTICOLOR FISH Materials Set of probe stock solutions, e.g.: probe 1: biotin labeled probe 2: biotin labeled probe 2: digoxigenin labeled probe 3: biotin labeled probe 3: digoxigenin labeled probe 4: biotin labeled probe 4: digoxigenin labeled probe 5: digoxigenin labeled.
ALTERNATE PROTOCOL 2
1. Using the schedule indicated in Table 8.3.4, pipet the required volume from each probe stock solution into a single tube and add to hybridization solution after the desired concentration. FISH USING LOW-COMPLEXITY PROBES AND NO REPEAT SUPPRESSION
BASIC PROTOCOL 5
Double-stranded target DNA has to be denatured prior to in situ hybridization. This can be achieved by treatment with extremes of pH or heat. Such treatments generally lead to loss of morphology. In practice, therefore, a compromise has to be found between intensity of hybridization signal and preservation of morphology. Originally, alkaline denaturations were used. In recent years thermal denaturations have become the method of choice, because of their experimental simplicity and excellent results. As for other steps in FISH procedures, there is no universal denaturation protocol. Variations in time and temperature should be tried in order to find the best conditions for a given application. For FISH with low-complexity probes (described in this protocol), probe and in situ target are denatured simultaneously. For this purpose the probe is applied to the slide, coverslipped, and heated to 80°C for the optimum time period (usually between 2 and 10 min, depending on the specific application), after which in situ hybridization is allowed to take place at 37° to 42°C. For FISH with high-complexity probes (see Basic Protocol 6), slides Molecular Cytogenetics
8.3.11 Current Protocols in Cytometry
and probes are denatured separately to enable annealing of repeats with unlabeled C0t1 DNA before the actual in situ hybridization. For discussion of the function of the different components of the hybridization solutions, see Critical Parameters and Troubleshooting. Materials Probe stock solution (see Support Protocol 1) Hybridization solution I or II (see recipes and Table 8.3.1) Biological sample or microscopic specimen on slide, pretreated for FISH (UNIT 8.2) Microscopic preparations (UNIT 8.2) 50% formamide/2× SSC, pH 7 60% formamide/2× SSC, pH 7 0.1× and 2× SSC (APPENDIX 2A) 70%, 90%, and 100% ethanol TNT buffer (see recipe) Counterstain Mounting medium 18 × 18–mm coverslips 80°C hot plate Schiefferdecker jars (Fisher) Shaking water baths, up to 60°C Additional reagents and equipment for immunocytochemistry (e.g., Watkins, 1989) 1. Dilute an aliquot of probe stock solution to a final concentration of 2 ng/µl in hybridization solution I (unique probes) or hybridization solution II (repeat probes). 2. Apply 5 or 10 µl of the diluted probe solution to slide preparation of biological sample or microscopic specimen pretreated for FISH. Cover with an 18 × 18–mm coverslip. The 5-ìl volume is used for probes dissolved in hybridization solutions without dextran sulfate. In the presence of dextran sulfate, larger volumes are required to homogenously spread fluid under the coverslip, due to the increased viscosity.
3. Denature in situ target and probe DNA simultaneously by placing the slide for 2 to 3 min on a metal plate heated to 80°C. Preferably, use a metal plate placed in a 80°C incubator. Denaturation is a critical step. Appropriate conditions are dependent on the specimen, fixation, and pretreatment. This denaturation protocol is optimized for metaphase chromosome preparations. If suboptimal results are obtained, different denaturation times separated by 30-sec intervals should be tried. The temperature may also be varied. For formalin-fixed, paraffin-embedded material, denaturation times may have to be increased to 10 min.
4. Place slide in Schiefferdecker jar. Hybridize overnight at 37°C in an environment humidified with 50% or 60% formamide/2× SSC, pH 7. Hybridization times can be reduced to a few hours for repeat targets. The percentage formamide in humidifying and washing solutions in this and all subsequent steps should be consistent with that used in the hybridization solution—i.e., 60% for repeat probes and 50% for unique probes.
Probe Labeling and Fluorescence In Situ Hybridization
To prevent probe solutions from evaporating, it is essential to humidify the environment. Place the slides horizontally in the jars and put jars in a beaker containing Kimwipes moistened with 60% or 50% formamide/2× SSC, pH 7, on the bottom. Seal the beaker tightly with aluminum foil and place at 37°C.
8.3.12 Current Protocols in Cytometry
5. Using forceps, transfer the preparations one by one from the hybridization jar to a jar containing formamide/2× SSC (pH 7), 37°C. Gently shake to detach the coverslips. Formamide used in (large-volume) washing solutions does not require deionization. However, check the pH of the formamide/SSC solutions. It should be 7. Adjust as needed with a few drops of 5 M HCl. CAUTION: Formamide is teratogenic. Use in a fume hood and dispose of properly.
6. Transfer the preparations to a fresh jar containing formamide/2× SSC (pH 7), 37°C, and gently shake for 5 min. In the same jar, repeat for a total of three washes. To reduce spurious background seen with cDNA probes, the washing temperature may be increased to 45°C. Additional washes with 0.1× SSC at 60°C may also be tried.
7. Wash twice, 5 min each, at room temperature in 2× SSC and once in TNT buffer. 8. If hapten-labeled probes are used, proceed with immunocytochemical detection (e.g., Watkins, 1989). For fluorochrome-labeled probes, dehydrate with 70%, 90%, and 100% ethanol (a few minutes each), air dry, and mount with appropriate counterstain in medium containing antifading reagent. The sequence homology of the different repeats on human chromosomes is such that under the hybridization conditions described above, besides the major binding sites for a given probe, minor binding may take place on other chromosomes (so-called “minor binding sites”). If minor binding sites appear after FISH, the following steps may circumvent their recurrence: shorten the time of denaturation; lower the probe concentration; increase the stringency of hybridization by either raising the formamide concentration to 65% or 70%, or raising the hybridization temperature to 42° to 45°C (also see Critical Parameters and Troubleshooting).
FISH USING HIGH-COMPLEXITY PROBES CONDITIONS AND REPEAT SUPPRESSION
BASIC PROTOCOL 6
Additional Materials (also see Basic Protocol 5) 70% deionized formamide (APPENDIX 2A)/2× SSC (APPENDIX 2A)/50 mM sodium phosphate, pH 7 (see recipe) 24 × 60–mm coverslips 1. Dilute an aliquot of the probe stock solution in hybridization solution I to a final concentration of 2 to 5 ng/µl for cosmids, 40 ng/µl for YACs, or 10 ng/µl for chromosome libraries. 2. Denature the probe solution 5 min at 75°C, quench 1 min on ice, microcentrifuge briefly, and mix. 3. Preanneal 30 min at 37°C. Optimal preannealing times vary between probes, and in setting up FISH with complex probes it is advised to test various preannealing times. For chromosome plasmid libraries, for example, the preannealing step can be omitted.
4. In the meantime, apply 120 µl of 70% deionized formamide/2× SSC/50 mM sodium phosphate to the preparation pretreated for FISH and cover with a 24 × 60–mm coverslip. 5. Denature the in situ DNA 2 to 3 min on an 80°C hot plate and remove coverslip by flicking it off. 6. Place preparation in ice-cold 2× SSC for 2 min, then wash in 70% ethanol (−20°C) for 3 min, followed by 90% and 100% ethanol at room temperature for 5 min each.
Molecular Cytogenetics
8.3.13 Current Protocols in Cytometry
7. Air dry and apply the preannealed probe mixture to the denatured preparation; use 10 µl probe solution per target area and cover with an 18 × 18–mm coverslip. 8. Hybridize overnight at 37°C in an environment humidified with 50% formamide/2× SSC, pH 7 (see Basic Protocol 5, step 4). Sixteen hours suffices for cosmids and chromosome plasmid libraries; 2 to 3 days is recommended for YACs (total yeast DNA) and comparative genomic hybridization.
9. Place the preparation in 50% formamide/2× SSC (pH 7), at 45°C and detach the coverslips by gently shaking. 10. Using forceps, transfer the preparations one by one from the hybridization jar to a jar containing 50% formamide/2× SSC (pH 7), 45°C. Drain the solution from the jar and wash three times, 5 min each, by gently shaking in fresh 50% formamide/2× SSC at 45°C. 11. Wash similarly three times, 5 min each, in 0.1× SSC at 60°C, and finally once in TNT. 12. If hapten-labeled probes are used, proceed with immunocytochemical detection (e.g., Watkins, 1989). For directly labeled probes, dehydrate with an ethanol series (70%, 90%, and 100% ethanol, a few minutes each), air dry, and mount with appropriate counterstain in medium containing antifading reagent. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Dextran sulfate, 50% Dissolve 50 g dextran sulfate (Pharmacia Biotech) in TE buffer (APPENDIX 2A) to a final volume of 100 ml by stirring and heating 3 hr at 70°C. Divide into aliquots and store ≤2 years at −20°C. DNase I, 1 ìg/ml Dissolve 1 mg DNase I in 1 ml of the buffer recommended by the manufacturer (Boehringer Mannheim). Store ≤1 year at −20°C. Prior to use, mix 1 µl with 1 ml ice-cold water and place at 0°C. Fish sperm DNA, 10 mg/ml, sonicated and denatured Dissolve 100 mg fish sperm DNA (e.g., herring sperm DNA, Boehringer Mannheim) in 100 ml of 0.3 M NaOH in TE buffer (APPENDIX 2A) by boiling 20 min. Neutralize with 5 ml of 2 M Tris⋅Cl, pH 7.5 (APPENDIX 2A), and 7.5 ml of 4 M HCl. Add 8 ml of 3 M sodium acetate (see recipe) and precipitate with 2 vol of 100% ethanol (−20°C) for 1 hr at 4°C. Centrifuge 10 min at 2600 × g, 4°C. Aspirate the supernatant and dry the DNA at 37°C. Dissolve DNA in 10 ml TE buffer (APPENDIX 2A). Determine the DNA concentration by measuring the optical density at 260 nm. Check the size of the DNA fragments on a 1% agarose gel (optimal size is 200 to 500 bp). Can be stored in 1-ml aliquots at −20°C for at least 10 years.
Probe Labeling and Fluorescence In Situ Hybridization
Hybridization solutions I and II For 100 ml mix: 10 ml 20× SSC (2× final; APPENDIX 2A) 10 ml 0.5 M sodium phosphate, pH 7 (50 mM final; see recipe) 50 (I) or 60 (II) ml deionized formamide (APPENDIX 2A; 50% [I] or 60% [II] final) 20 ml 50% dextran sulfate (I only; 10% final; see recipe) Store in 1-ml aliquots ≤1 year at 4°C
8.3.14 Current Protocols in Cytometry
NBT/BCIP substrate solution Mix 3 ml of 1 M Tris⋅Cl, pH 9.6 (APPENDIX 2A), 150 µl of 1 M MgCl2, and 12 ml filtered, distilled water. Add 96 µl NBT and 48 µl BCIP (ready-to-use solutions of Nitro blue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate, respectively; Promega). Make fresh immediately before use. Nick translation buffer, 10× 0.5 M Tris⋅Cl, pH 7.8 (APPENDIX 2A) 50 mM MgCl2 0.5 mg/ml BSA (bovine serum albumin), nuclease free Store in 100-µl aliquots ≤2 years at −20°C Prior to use, thaw an aliquot and place at 0°C Aliquots can be thawed and frozen repeatedly.
Nucleotide mixture Dilute 5 µl each of 100 mM dATP, dCTP, and dGTP and 1 µl of 100 mM dTTP (all from Boehringer Mannheim) to 1 ml with distilled water. Store in 100-µl aliquots ≤2 years at −20°C. Prior to use, thaw an aliquot and place at 0°C. Aliquots can be thawed and frozen repeatedly. Final concentrations: 0.5 mM each dATP, dCTP, dGTP, and 0.1 mM dTTP.
PCR buffer, 10× 100 mM Tris⋅Cl, pH 8.3 (APPENDIX 2A) 500 mM KCl Store 1-ml aliquots ≤2 years at −20°C PCR-bio and PCR-dig labeling mix, 10×, pH 7.0 2 mM each of dATP, dCTP, and dGTP (Boehringer Mannheim) 1.3 mM dTTP (Boehringer Mannheim) 0.7 mM biotin-16-dUTP or 0.7 mM digoxigenin-11-dUTP (Boehringer Mannheim) Store in 100-µl aliquots ≤1 year at −20°C Prior to use, thaw an aliquot and place at 0°C PCR-fluorescein labeling mix, 10×, pH 7.0 2 mM each dATP, dCTP, and dGTP (Boehringer Mannheim) 1.5 mM dTTP (Boehringer Mannheim) 0.5 mM fluorescein-dUTP (Boehringer Mannheim, NEN Life Science Products, or Amersham) Store in 100-µl aliquots ≤1 year at −20°C Prior to use, thaw an aliquot and place at 0°C Sodium acetate, 3 M, pH 5.6 Dissolve 408 g sodium acetate⋅3H2O in 1 liter water. Adjust pH to 5.6 with acetic acid. Autoclave and store ≤1 year at room temperature. Sodium phosphate, 0.5 M, pH 7 Mix 0.5 M NaH2PO4⋅H2O and 0.5 M Na2HPO4⋅2H2O to obtain pH 7. Store 1-ml aliquots ≤10 years at −20°C.
Molecular Cytogenetics
8.3.15 Current Protocols in Cytometry
TN buffer, 10× Dissolve 121.14 g Tris⋅Cl and 87.6 g NaCl in 800 ml water. Adjust pH to 7.5 with 6 M HCl and dilute to 1 liter with water. Store ≤1 month at room temperature. Final composition: 1 M Tris⋅Cl/1.5 M NaCl, pH 7.5.
TNB buffer Dissolve 0.5 g Boehringer Mannheim blocking reagent in 100 ml TN buffer (see recipe) for 1 hr at 60°C. Store 10-ml aliquots ≤2 years at −20°C. Optionally, 0.02% thimerosal (Sigma) may be added as a preservative; in this case the buffer can be stored ≤6 months at 4°C. Repeated thawing and freezing is not recommended.
TNT buffer Dissolve Tween 20 (Sigma) to a final concentration of 0.05% in TN buffer (see recipe). Store ≤1 month at room temperature. COMMENTARY Background Information Choice of hapten versus fluorochrome labeling of probes The haptens frequently used are biotin and digoxigenin (Langer et al., 1981; Kessler, 1990). Hapten-labeled probes used in indirect methods provide better FISH sensitivity than fluorochrome-labeled probes used in direct methods. There is no preference for biotin or digoxigenin because the immunocytochemical detection procedures are similar in sensitivity. However, for some special applications (e.g., FISH to tissue sections), use of biotin as a label may give rise to background problems due to the presence of endogenous biotin. Commer-
cially available fluorochrome-labeled nucleotides are shown in Table 8.3.2. In view of experimental simplicity and time considerations one should aim to use fluorochrome-labeled probes as much as possible. They are recommended for chromosome repeat detection and may also be used for YAC and cosmid detection, although hapten labeling and fluorescent immunocytochemical detection should be used if FISH results are unsatisfactory (Wiegant et al., 1991, 1993, 1996). If fluorescent detection is not feasible because of high levels of autofluorescence, then resort to hapten-labeled probes and use chromogenic immunocytochemical detection procedures in combination with bright-field microscopy.
Table 8.3.2 Spectral Properties of Fluorochromes Conjugated to Nucleotides
Fluorochrome
Probe Labeling and Fluorescence In Situ Hybridization
AMCA Coumarin Cy2 Fluorescein Cy3 Tetramethylrhodamine Lissamine Cy3.5 Texas Red Cy5 Cy5.5 Cy7
Color of fluorescence Blue Blue Green Green Red Red Red Red Red Dark red Dark red Near infrared
Excitation maximum (nm) 350 402 489 494 550 555 570 581 593 649 675 743
Emission maximum 455 443 506 517 570 580 588 596 612 670 694 767
8.3.16 Current Protocols in Cytometry
Choice of labeling reaction Nick translation is the most widely applied enzymatic labeling technique for FISH. By fine-tuning the DNase I concentration, optimal probe fragment sizes can be generated (200 to 500 bp for most FISH applications). In randomprimed labeling, experimental control of fragment length is probably more difficult to achieve, which may explain why probes labeled by random priming tend to give weaker signals and higher background in FISH. However, random priming offers the advantage that it results in net synthesis of DNA, unlike labeling by nick translation. PCR labeling is very well suited to generating large amounts of probes from minimal amounts of input template DNA. To optimize probe fragment length, it is advised to restrict PCR labeling of probes to be used in FISH to sequences <1 kb. Alternatively, using PCR one may generate a vast amount of DNA sequences >1 kb in the absence of a modified nucleotide and label the PCR product afterwards in a nick translation reaction. This approach can also be used conveniently in Alu- (Nelson et al., 1989) and DOP-PCR (Telenius et al., 1992) strategies for generating probes. Alu-PCR is a specific case of a more general method termed IRSPCR. By choosing primers specific for interspersed repetitive sequences (IRS), the DNA between the repeats can be amplified by PCR. In man, the most abundant family of repeats is the Alu family, estimated to comprise 900,000 elements in the haploid genome for an average spacing of 3 to 4 kb. Alu-PCR is species dependent and region specific, as Alu elements are preferentially found in the G-light bands of human chromosomes. To bypass these limitations, degenerate oligonucleotide-primed PCR
Table 8.3.3
(DOP-PCR) has been developed; this employs oligonucleotides with partially degenerate sequences as primers. This degeneracy, together with a PCR protocol using a low initial annealing temperature, ensures priming from multiple evenly dispersed sites within any given species. In general, nick translation is preferred over random priming, and PCR labeling is restricted to special applications. Multicolor FISH In its most simple form, multicolor FISH is performed by mixing differently labeled probes in a single hybridization solution (Hopman et al., 1986; Nederlof et al., 1989). As with single-color FISH, using fluorochromes rather than haptens whenever possible is recommended. Due to spectral overlap of the fluorochromes, the number of probes identifiable on the basis of (primary) fluorescence color is limited to three for the visible part of the spectrum and to five if far-red- and infrared-fluorescing dyes are included. In a more sophisticated approach, termed combinatorial labeling, each probe is labeled with a combination of haptens or fluorochromes, creating unique colors for FISH (Nederlof et al., 1990; Raap et al., 1992; Ried et al., 1992; Wiegant et al., 1993). Up to seven targets can be distinguished with only three fluorochromes/haptens (see Table 8.3.3). The combinatorial approach can be used qualitatively as well as quantitatively. In its qualitative mode the quantity of the different labels present on a given probe sequence is in principle not relevant. In its quantitative mode, also termed “ratio-labeling,” varying ratios of different labels are used for each probe to discriminate multiple probes on the basis of fluo-
Combinatorial Labeling Scheme for Seven-Color FISH
Probe number 1 2 3 4 5 6 7
Hapten/fluorochrome combination Biotin Fluorescein Rhodamine Biotin/fluorescein Biotin/rhodamine Fluorescein/rhodamine Biotin/fluorescein/ rhodamine
(Immuno)fluorescent color AMCAa + − − + + − +
Fluorescein Rhodamine − + − + − + +
− − + − + + +
aAMCA, amino-methyl coumarin acetic acid. AMCA-streptavidin is used to indirectly detect biotin-la-
beled probes with blue fluorescence.
Molecular Cytogenetics
8.3.17 Current Protocols in Cytometry
mapping of a YAC by FISH may result in two pairs of bright fluorescing spots on homologous chromosomes in a background of randomly positioned spots. The experiment would be judged optimal from the mapping point of view, but not from the technical perspective. Stringency of hybridization and washing to a large extent determine specificity of FISH results, and should be optimized as needed by varying temperature and the formamide and salt concentrations.
rescence color composition (Nederlof et al., 1992). For example, five different ratios of red and green fluorescence attached to five different probes will allow discrimination of five different targets with only two fluorochromes (see Table 8.3.4). Obviously one may combine the combinatorial and ratio-labeling approach to maximize FISH multiplicity. Although seven- to twelvecolor FISH with chromosome libraries can be accomplished fairly easily (Dauwerse et al., 1992), the demands imposed on the spatial and spectral resolution of the microscopic imaging device increase rapidly with multiplicity (Schröck et al., 1996; Speicher et al., 1996). Microscopic imaging of multicolor FISH often demands accurate spatial registration of the various fluorescent colors (e.g., bicolor cosmid mapping of cosmid). Use of double- or triple-bandpass filters in the fluorescence microscope setup is recommended for such applications (see UNIT 2.4 for discussion).
Hybridization and washing conditions During in situ hybridization, perfect hybrids form between probe and target, but imperfect ones may also form with less homologous target sequences, leading to nonspecific hybridization. Perfect hybrids are more stable than imperfect ones. To prevent the latter hybrids from forming, the stringency of the hybridization can be increased by lowering salt and raising formamide concentrations as well as by raising temperature. Remaining nonspecific hybrids can be removed by stringent post-hybridization washes. To guide the design of optimal hybridization and wash solutions, the effects of various components of hybridization solutions on rate of renaturation (hybridization) and stability of the resulting duplex DNA (hybrid stability) are briefly discussed here; see Wetmur (1991) and references therein for an extensive discussion on the thermodynamics and kinetics of denaturation and renaturation/hybridization processes. The various items are discussed for free DNA molecules in solution, but the features will be similar for immobilized nucleic acids targets as in FISH. Temperature. The maximum rate of renaturation of DNA is 25°C below the melting temperature of DNA (Tm), with Tm being a function
Critical Parameters and Troubleshooting A universal FISH procedure does not exist. Many of the critical parameters have been noted throughout the protocols. Some general issues are addressed here. FISH aims at localizing DNA sequences in histo- or cytologically preserved biological preparations. These aims are conflicting. As a rule of thumb, measures taken to obtain or improve FISH signals adversely affect cytological features. In situ denaturation is critical in this respect. Too long a denaturation time or too high a denaturation temperature may lead to suboptimal morphology. The extent to which optimization of denaturation should be pursued will depend on the scientific or diagnostic question being asked. For example, chromosomal
Table 8.3.4
Probe Labeling and Fluorescence In Situ Hybridization
Pipetting Schedule for Five-Color FISH by Ratio-Labeling with Two Labels
Probe number
% Biotin probe
% Digoxigenin probe
Volume biotin probe (µl)
Volume digoxigenin probe (µl)
1
100
0
10.0
0.0
2
75
25
7.5
2.5
3
50
50
5.0
5.0
4
25
75
2.5
7.5
5
0
100
0.0
10.0
8.3.18 Current Protocols in Cytometry
of the base composition, ionic composition, and formamide concentration. The bell-shaped curve relating renaturation rate and temperature is, however, broad, with a rather flat maximum from 16° to 32°C below Tm. Perfect hybrids are stable at temperatures a few degrees below Tm, whereas imperfect ones are not. Thus temperature can be used as a manipulator of stringency. Ionic composition. Monovalent cations interact electrostatically with the phosphate groups on the backbone of nucleic acids. Electrostatic repulsion between the two DNA strands of a hybrid decreases with increasing salt concentration. Thus, high concentrations of monovalent cations stabilize duplex DNA, whereas low concentrations destabilize it. Hence, lowering salt concentration increases stringency. Divalent cations very strongly stabilize duplex DNA and their (free) presence in the hybridization mixture should be prevented—e.g., by complexing them with citrate (present in SSC) or EDTA. In view of the poor buffering capacity of SSC solutions, it is recommended that 20 to 50 mM sodium phosphate, pH 6.5 to 7.5, be added to hybridization solutions as a buffer. Formamide. DNA of average GC content melts (denatures) in 0.1 to 0.2 M Na+ at 90° to 100°C. For in situ hybridization this would imply that the microscope preparations have to be hybridized at 65° to 75°C for prolonged periods of time, which may lead to deterioration of morphology. A solution to this problem has been found in the use of organic solvents that reduce the thermal stability of double-stranded DNA, so that in situ hybridization can be performed at lower temperatures. Formamide has been the organic solvent of choice for years. It reduces the Tm of DNA⋅DNA duplexes in a linear fashion by 0.72°C for each percent formamide. Thus, with 50% formamide present in the hybridization mixture, hybridization can be performed at 30° to 45°C. Denaturation temperature can be brought down from the 90° to 100°C needed in salt solutions to 70° to 80°C in the presence of 50% formamide. Formamide degrades over time when stored at room temperature. As a consequence, hybridization and washing solutions may have an elevated pH (9 to 11). For formamide to be used in hybridization solutions it is recommended that it be purified by ion-exchange chromatography and stored in aliquots at −20°C. The quantities of formamide used in post-hybridization washes are large, and deionization is not practicable. For washing purposes it may be used without deionization, but check the pH of
formamide-containing post-hybridization wash solutions and adjust as needed (see Reagents and Solutions). Dextran sulfate. In aqueous solutions dextran sulfate is strongly hydrated; in consequence, macromolecules such as labeled probe DNAs in dextran-containing hybridization solutions have no access to the hydrating water (this is known as the excluded volume effect). This leads to an apparent increase in probe concentration and consequently higher hybridization rates. For FISH, another advantageous effect of the increased hybridization rates obtained with dextran sulfate present has been suggested: a probe fragment may not cover its full target. On the protruding sequence (e.g., plasmid vector sequences), large networks of labeled probe fragments can form, leading to increased FISH signal. It should be noted that such a phenomenon will also occur in the absence of dextran sulfate, although at a lower rate. Use of dextran sulfate is therefore recommended for small unique and complex probes. Probe Purity. Purity of the DNA sample is essential for good hapten or fluorochrome labeling. DNA samples need to be properly RNase treated and phenol extracted, because contamination with proteins and/or RNA adversely influences labeling reactions. Fragment length. The rate of hybridization of DNA is proportional to the square root of the fragment length. Consequently, maximum hybridization rates are obtained with unfragmented intact probes. However, because probe molecules have to diffuse into the dense matrix of fixed cells and chromosomes during in situ hybridization, reduction of fragment length is a requisite for FISH. The fragment length also influences the thermal stability of hybridized fragments, as does the base composition of individual probe fragments, which may differ considerably from the overall base composition of the probe. The following formula, relating fragment length (n) in a hybrid to change in Tm, has been derived experimentally: ∂Tm = 500/n. This illustrates the fact that with commonly used fragment lengths (200 to 500 bp), effects on hybrid stability can be neglected. Concentration. The initial process in the annealing of two nucleic acid strands is a nucleation step in which a few base pairs are formed. The adjacent base pairs then form rapidly (in what is known as zippering) provided that they are in register. The nucleation reaction is the rate-limiting step. Consequently, the
Molecular Cytogenetics
8.3.19 Current Protocols in Cytometry
higher the probe concentration and the higher the chances of nucleation, the higher the hybridization rate. In practice, the concentration of probe is determined by the FISH signal-tobackground ratio needed for convenient analysis. Degree of labeling. Although fluorochrome incorporation can be conveniently checked after ethanol precipitation with the aid of a UV lamp, a dot blot assay is used to check hapten incorporation. A satisfactory outcome of a dot blot assay, however, does not always correlate with good probe performance in FISH. This can result from overlabeling of the probe or the presence of contaminating DNA. High degrees of labeling may strongly influence hydrogen bonding in base-pair formation, leading to low hybridization efficiencies. In such cases, lowering the stringency of hybridization and washes may help restore FISH performance. Presence of contaminating labeled DNA (e.g., genomic yeast DNA in the case of YACs) leads to lowering of the specific labeled probe concentration. In such cases, raising the concentration of the labeled DNA may help improve FISH performance.
Anticipated Results Carrying out the basic protocols should lead to efficient single or multi-color detection of repeat sequences, cosmids, YACs, and other large-insert probes as well as chromosome plasmid libraries. The protocol for small unique targets (e.g., cDNAs) may yield less efficient detection, as it resides at the limits of FISH sensitivity.
Time Considerations
Probe Labeling and Fluorescence In Situ Hybridization
Labeling by nick translation takes 4 to 5 hr from pipetting the required reagents to preparing probe stock solutions. Random-primed labeling can be either considerably shorter or longer depending on the amount of input template DNA or the amount of labeled probe to be generated, respectively. Labeling by PCR takes ∼6 hr from pipetting to dissolving the probe in the desired hybridization solution. Checking the fragment sizes by gel electrophoresis takes ∼3 hr, while a dot blot assay requires ∼5 hr. Excluding specimen preparation, immunocytochemical detection (for hapten-labeled probes only) and microscopy time, actual time needed for FISH varies from 2 hr (repeat probes) to several days (YACs, CGH).
Literature Cited Dauwerse, J.G., Wiegant, J., Raap, A.K., Breuning, M.H., and Van Ommen, G.J.B. 1992. Multiple colors by fluorescence in situ hybridization using ratio-labelled DNA probes create a molecular karyotype. Hum. Mol. Genet. 1:593-598. du Manoir, S., Speicher, M.R., Joos, S., Schroeck, E., Popp, S., Doehner, H., Kovacs, G., RobertNicoud, M., Lichter, P., and Cremer, T. 1993. Detection of complete and partial chromosome gains and losses by comparative genomic in situ hybridization. Hum. Genet. 90:590-610. Feinberg, A.P. and Vogelstein, B. 1983. A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132: 6-10. Grünewald-Janho, S., Keesey, J., Leous, M., Van Miltenburg, R., and Schroeder, C. 1996. PCR and high-prime labeling. In Nonradioactive in Situ Hybridization Application Manual, pp. 36-40. Boehringer Mannheim, Mannheim, Germany. Hopman, A.H.N., Wiegant, J., Raap, A.K., Landegent, J.E., Van der Ploeg, M., and Van Duijn, P. 1986. Bi-color detection of two target DNAs by non-radioactive in situ hybridization. Histochemistry 85:1-4. Innis, M.A., Gelfand, D.H., Sninsky, J.J., and White, T.J. (eds.) 1990. In PCR Protocols: A Guide to Methods and Applications. Academic Press, San Diego. Kallioniemi, A., Kallioniemi, O.P., Sudar, D., Rutovitz, D., Gray, J.W., Waldman, F.M., and Pinkel, D. 1992. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 258: 818-821. Kessler, C. 1990. The digoxigenin system: Principle and applications of the novel non-radioactive DNA labeling and detection system. Bio Technology Int. 183-194. Landegent, J.E., Jansen in de Wal, N., Dirks, R.W., Baas, F., and Van der Ploeg, M. 1987. Use of whole cosmid cloned genomic sequences for chromosomal localization by non-radioactive in situ hybridization. Hum. Genet. 77:366-370. Langer, P.R., Waldrop, A.A., and Ward, D.C. 1981. Enzymatic synthesis of biotin-labeled polynucleotides: Novel nucleic acid affinity probes. Proc. Natl. Acad. Sci. U.S.A. 78:6633-6637. Lichter, P., Tang, C.C., Call, K., Hermanson, G., Evans, G., Housman, D., and Ward, D.C. 1990. High resolution mapping of human chromosome 11 by in situ hybridization with cosmid probes. Science 247: 64-69. Moore, D., Chory, J., and Ribaudo, R.K. 1994. Isolation and purification of large DNA restriction fragments from agarose gels. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 2.6.12.6.12. John Wiley & Sons, New York. Nederlof, P.M., Robinson, D., Abuknesha, R., Wiegant, J., Hopman, A.H.N., Tanke, H.J., and Raap, A.K. 1989. Three-color fluorescence in
8.3.20 Current Protocols in Cytometry
situ hybridization for the simultaneous detection of multiple nucleic acid sequences. Cytometry 10:20-27. Nederlof, P.M., Van der Flier, S., Vrolijk, J., Tanke, H.J., and Raap, A.K. 1992. Fluorescence ratio measurements of double-labeled probes for multiple in situ hybridization by digital imaging microscopy. Cytometry 13:839-845. Nederlof, P.M., Van der Flier, S., Wiegant, J., Raap, A.K., Tanke, H.J., Ploem, J.S., and Van der Ploeg, M. 1990. Multiple fluorescence in situ hybridization. Cytometry 11:126-131. Nelson, D.L., Ledbetter S.A., Corbo, L., Victoria, M.F., Ramirez-Solis, R., Webster, T., Ledbetter, D.H., and Caskey, C.T. 1989. Alu polymerase chain reaction: A method for rapid isolation of human specific sequences from complex DNA sources. Proc. Natl. Acad. Sci. U.S.A. 86:66866690. Pinkel, D., Landegent, J.E., Collins, C., Fuscoe, J., Segraves, R., Lucas, J., and Gray, J. 1988. Fluorescence in situ hybridization with human chromosome-specific libraries: Detection of trisomy 21 and translocations of chromosome 4. Proc. Natl. Acad. Sci. U.S.A. 85: 9138-9142. Raap, A.K., Wiegant, J., and Lichter, P. 1992. Multiple fluorescence in situ hybridization for molecular cytogenetics. In Nonradioactive Labeling and Detection of Biomolecules (C. Kessler, ed.), Springer-Verlag, New York. Ried, T., Baldini, A., Rand, T.C., and Ward, D.C. 1992. Simultaneous visualization of seven different DNA probes by in situ hybridization using combinatorial fluorescence and digital imaging microscopy. Proc. Natl. Acad. Sci. U.S.A. 89:1388-1392.
J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 2.5.1-2.5.9. John Wiley & Sons, New York. Watkins, S. 1989. Immunohistochemistry. In Current Protocols in Molecular Biology (F.M. Ausubel, R.E. Brent, B. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 14.6.1-14.6.13. John Wiley & Sons, New York. Wetmur, J.G. 1991. DNA probes: Applications of the principle of nucleic acid hybridization. Crit. Rev. Biochem. Mol. Biol. 26:227-259. Wiegant, J., Ried, T., Nederlof, P.M., Van der Ploeg, M., Tanke, H.J., and Raap, A.K. 1991. In situ hybridization with fluoresceinated DNA. Nucl. Acids Res. 19:3237-3241. Wiegant, J., Wiesmeijer, C.C., Hoovers, J.M.N., Schuuring, E., d’Azzo, A., Vrolijk, J., Tanke, H.J., and Raap, A.K. 1993. Multiple and sensitive fluorescence in situ hybridization with rhodamine-, fluorescein-, and coumarin-labeled DNAs. Cytogenet. Cell Genet. 63:73-76. Wiegant, J., Verwoerd, N., Mascheretti, S., Bolk, M., Tanke, H.J., and Raap, A.K. 1996. An evaluation of a new series of fluorescent dUTPs for fluorescence in situ hybridization. J. Histochem. Cytochem. 44:525-529.
Key References Repeat-suppression FISH Landegent et al., 1987. See above. Lichter et al., 1990. See above. Pinkel et al., 1988. See above.
Multicolor FISH
Rigby, P.W.J., Dieckmann, M., Rhodes, C., and Berg, P. 1977. Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J. Mol. Biol. 113:237241.
Hopman et al., 1986. See above (bicolor FISH).
Sambrook, J., Fritsch, E.F., and Maniatis, T. 1989. Molecular Cloning: A Laboratory Manual, 2nd ed., pp. 6.3-6.35. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Ried et al., 1992, and Wiegant et al., 1993. See above (seven-color FISH).
Schröck, E., du Manoir, S., Veldman, T., Schoell, B., Wienberg, J., Ferguson-Smith, M. A., Ning, Y., Ledbetter, D.H., Bar-Am, I., Soenksen, D., Garini, Y., and Ried, T. 1996. Multicolor spectral karyotyping of human chromosomes. Science 273:494-497.
Speicher et al., 1996, and Schrock et al., 1996. See above (>24 color FISH).
Speicher, M.R., Gwyn Ballard, S., and Ward, D.C. 1996. Karyotyping human chromosomes by combinatorial multi-fluor FISH. Nature Genet 12:368-375. Telenius, H., Carter, N.P., Bebb, C.E., Nordenskjöld, M., Ponder, B.A.J., and Tunnacliffe, A. 1992. Degenerate oligonucleotide-primed PCR: General amplification of target DNA by a single degenerate primer. Genomics 13:718-725. Voytas, D. 1988. Agarose gel electrophoresis. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore,
Nederlof et al., 1989. See above (triple-color FISH). Nederlof et al., 1990, 1992. See above (combinatorial labeling).
Dauwerse et al., 1992. See above (twelve- color FISH).
Cytogenetic applications Emanuel, B.S. 1993. The use of fluorescence in situ hybridization to identify human chromosomal anomalies. Growth Genet. Horm. 9:6-12. Very useful review
Hybridization theory Wetmur, 1991. See above. See also classical articles cited therein.
Contributed by J. Wiegant (probe labeling) and A.K. Raap (FISH) Leiden University Leiden, The Netherlands
Molecular Cytogenetics
8.3.21 Current Protocols in Cytometry
Immunocytochemical Detection
UNIT 8.4
This unit presents detection protocols for in situ hybridization (ISH) to chromosomal preparations. Two ISH approaches can be distinguished: direct and indirect. In the direct method, a fluorochrome-labeled probe is detected by its fluorescence without immunocytochemical reaction. In the indirect method, a hapten-labeled probe is detected immunocytochemically using either fluorochrome- or enzyme-conjugated anti-hapten reagents for detection by fluorescence or bright-field microscopy, respectively. These anti-hapten reagents include antibodies and, in the case of biotin, streptavidin (see Critical Parameters and Troubleshooting). For bright-field microscopy, the hapten-labeled probe is detected by enzyme cytochemical reactions using a precipitating chromogenic substrate (see Basic Protocols 1 and 2). For fluorescence microscopy, the hapten-labeled probe is detected by immunofluorescent staining (see Basic Protocols 3, 4, and 5; see Alternate Protocols 1 and 2). In addition to conventional immunocytochemical detection systems, this unit also presents a newly developed and sensitive approach called tyramide signal amplification (TSA; see Basic Protocol 6), which combines enzyme cytochemistry with fluorescent detection (Kerstens et al., 1995; Raap et al., 1995). This unit exclusively contains detailed protocols for immunocytochemical detection in fluorescence in situ hybridization (FISH) and bright-field in situ hybridization (BfISH). The protocols provide details such as reagents used and incubation conditions. The specific procedures for washing and handling of the ISH-labeled sample are presented separately (see Support Protocol 1). For visualization and analysis of ISH results, see Chapter 2. Types of microscopy (UNIT 2.1), optical filter sets for (multicolor) fluorescence microscopy (UNIT 2.4), and digital fluorescence microscopy (UNIT 2.5) are described there. STRATEGIC PLANNING The choice between fluorescence or bright-field microscopical evaluation of in situ hybridization depends largely on the applications. When high sensitivity is required (e.g., mapping of a small single-copy sequence on metaphase chromosomes), fluorescence in situ hybridization (FISH) is superior to bright-field in situ hybridization (BfISH). BfISH can reach a reasonable level of sensitivity due to the turnover number of the enzymes, which determines the number of precipitating chromogenic substrate molecules catalyzed by the enzyme. A fluorescent enzyme reaction is the most sensitive immunocytochemical detection reaction. FISH is superior for multicolor hybridization as well (Nederlof et al., 1989; Wiegant et al., 1991). Different dyes used for BfISH can be spectrally separated only on the basis of their absorption characteristics, while in FISH different fluorophores can be spectrally separated on the basis of both excitation and emission. In practical terms, the multiplicity of BfISH (i.e., the maximum number of probes that can be visualized simultaneously) is limited to three (Speel et al., 1994). Multiplicity in FISH is considerably higher, and is dependent on the spatial separation of the targets. BfISH is recommended when biological objects with high levels of autofluorescence or fixative-induced fluorescence are being studied. Autofluorescence is encountered, for example, in plant cells with strongly fluorescent chlorophyll. Fixative-induced fluorescence is encountered in paraffin-embedded tissue, which is often fixed for prolonged periods of time with formalin. BfISH is also recommended when lenses of low to moderate magnification are used to evaluate large microscopic fields. Such lenses intrinsically have a low numerical aperture, and are therefore less suited to sensitive Molecular Cytogenetics Contributed by J. Wiegant Current Protocols in Cytometry (1998) 8.4.1-8.4.18 Copyright © 1998 by John Wiley & Sons, Inc.
8.4.1 Supplement 4
registration of fluorescence. BfISH is useful when preparations need to be archived for several years after ISH. The endproducts of the horseradish peroxidase (HRP) and alkaline phosphatase (AP) reactions can be stored for years at room temperature without loss of signal intensity. FISH slides cannot be stored for such long periods of time (but only up to 3 months). Finally, BfISH is of course necessary when a fluorescence microscope is not available. For both fluorescent and bright-field applications, procedures vary depending on the sensitivity of the probe (see Background Information). Probes for highly repetitive sequences (e.g., centromere-associated alpha-satellite probes) require low sensitivity. A single incubation with a conjugated anti-hapten reagent generally suffices. Probes for small and for unique sequences (e.g., unique plasmid probes, cosmid and YAC probes) require high sensitivity. In such cases, primary signals are amplified by subsequent incubations with secondary and tertiary immunoreagents to visualize the hybridized probe. BASIC PROTOCOL 1
ALKALINE PHOSPHATASE/NBT-BCIP REACTION FOR BfISH USING DIGOXIGENIN-, BIOTIN-, OR FITC-LABELED HYBRIDIZATION PROBES This enzyme-based detection reaction consists of incubation with an alkaline phosphatase (AP)–labeled anti-fluorescein antibody, anti-digoxigenin antibody, or streptavidin following ISH with a fluorescein (FITC)–, digoxigenin-, or biotin-labeled probe, respectively. In the detection reaction, alkaline phosphatase hydrolyzes 5-bromo-4-chloro-3-indolylphosphate (BCIP). The resulting 5-bromo-4-chloro-indole (BCI) reduces the yellow nitroblue tetrazolium (NBT) to a purple formazan, which precipitates at the hybridization site. A number of alternative antibody combinations are presented in this protocol. Selection depends on the label conjugated to the hybridization probe (fluorescein, digoxigenin, or biotin) and on the desired sensitivity. High sensitivity is afforded by incubations with multiple immunocytochemical reagents, resulting in the formation of large complexes containing multiple AP reaction sites. Antibody combinations for this approach are shown in Table 8.4.1, and their use is described in steps 2a through 6a. When lower-sensitivity detection is adequate, a single-antibody procedure (step 2b) can be substituted for these steps. Materials ISH-labeled preparation: chromosomes hybridized with fluorescein (FITC)–, digoxigenin-, or biotin-labeled probes (UNIT 8.3) First antibody for high sensitivity (see Table 8.4.1) TNT buffer (see recipe) Second antibody for high sensitivity (see Table 8.4.1) Third antibody for high sensitivity (see Table 8.4.1) Alkaline phosphatase (AP)–conjugated antibody for low sensitivity (select one): 1:1000 AP-sheep-anti-digoxigenin (Boehringer Mannheim), 1:200 AP-streptavidin (Vector Labs), or 1:500 AP-mouse-anti-FITC (Dako) Alkaline phosphatase (AP) buffer (see recipe) Alkaline phosphatase (AP) substrate solution (see recipe) Nuclear fast red (Sigma) Aquamount (BDH) or equivalent mounting medium for AP reaction 24 × 50–mm cover slips Moist chamber: 1-liter beaker containing paper towels moistened with TNT buffer (see recipe), covered with aluminum foil Schiefferdecker or Coplin jars (Fisher Scientific)
Immunocytochemical Detection
Additional reagents and equipment for handling of ISH samples and immunoreagents (see Support Protocol 1)
8.4.2 Supplement 4
Current Protocols in Cytometry
Table 8.4.1
Antibodies for High-Sensitivity Detection in Basic Protocol 1a,b
Dig-labeled I
Dig-labeled II
Biotin-labeled
FITC-labeled I
FITC-labeled II
1st antibody
Mouse anti-digoxigenin (1:200; BM)
Mouse anti-digoxinc (1:500; S)
Streptavidind (1:100; V)
Rabbit anti-FITC (1:250; D)
Mouse anti-FITC (1:250; D)
2nd antibody
AP-labeled goat anti-mouse (1:1000; S)
Dig-labeled sheep Bio-labeled goat anti-mouse (1:100; BM) anti-streptavidin (1:100; V)
AP-labeled goat anti-rabbit (1:500; S)
AP-labeled rabbit anti-mouse (1:500; S)
3rd antibody
AP-labeled rabbit antigoat (1:1000; S)
AP-labeled sheep antidig (1:100; BM)
None
None
AP-labeled streptavidin (1:200; V)
aSuppliers and appropriate dilutions are indicated in parentheses. Proper titration of each antibody is required to determine the optimal dilution (see
Critical Parameters and Troubleshooting). BM, Boehringer Mannheim; D, Dako; S, Sigma; V, Vector Labs. bBio, biotin; Dig, digoxigenin; FITC, fluorescein; AP, alkaline phosphatase. cIt has been demonstrated that an anti-digoxin antibody must be used in this alternate protocol to avoid background problems caused by the use of mouse anti-digoxigenin antibody. dBiotinylated goat anti-streptavidin does not recognize enzyme-conjugated (either AP or HRP) streptavidin. Therefore, the first incubation is done with unconjugated streptavidin.
Perform antibody labeling 1. Wash and block ISH-labeled preparation (see Support Protocol 1). For high-sensitivity protocol 2a. Incubate preparation 30 min at 37°C with the first antibody for high sensitivity (see Table 8.4.1; see Support Protocol 1). For hybridization probes labeled with digoxigenin or with FITC, the experimenter can select from two series of antibodies (see Table 8.4.1) that offer similar sensitivity.
3a. Wash 3 times for 5 min each with TNT buffer. 4a. Incubate 30 min at 37°C with the second antibody. For FITC-labeled hybridization probes, no third antibody is used (see Table 8.4.1). Proceed directly to steps 7 through 13.
5a. Wash 3 times for 5 min each with TNT buffer. 6a. Incubate 30 min at 37°C with the third antibody. Proceed with steps 7 through 13. For low-sensitivity protocol 2b. Incubate preparation 30 min at 37°C with the appropriate AP-conjugated antibody for low sensitivity (see Support Protocol 1). Proceed with steps 7 through 13. Each antibody should be properly titrated to determine the optimal dilution (see Critical Parameters and Troubleshooting).
Perform enzyme reaction 7. Wash 3 times for 5 min each with TNT buffer. 8. Wash 5 min with 100 ml freshly prepared AP buffer. 9. Place 100 µl AP substrate solution over the preparation and cover with a 24 × 50–mm coverslip. Incubate 1 to 4 hr at room temperature in a moist chamber in the dark. After 1 hr of incubation, microscopical monitoring of the NBT/BCIP reaction is recommended.
10. Remove coverslip and wash briefly with water.
Molecular Cytogenetics
8.4.3 Current Protocols in Cytometry
Supplement 4
11. Counterstain with 100 ml nuclear fast red 3 to 5 min in a Schiefferdecker or Coplin jar. The optimal staining time for nuclear fast red must be determined experimentally. Staining can be monitored periodically by rinsing with tap water and viewing with a microscope. If staining is too faint, increase time but be careful not to overstain.
12. Wash with running tap water for 2 min. 13. Drain off most of the water, put a few drops of Aquamount on the preparation, and cover with a 24 × 50–mm coverslip. BASIC PROTOCOL 2
PEROXIDASE/DAB REACTION FOR BfISH USING DIGOXIGENIN-, BIOTIN- OR FITC-LABELED HYBRIDIZATION PROBES In this enzyme-based detection reaction, in situ hybridization is followed by incubation with a horseradish peroxidase (HRP)–labeled anti-hapten antibody or streptavidin. The presence of HRP is detected by incubation with 3,3′-diaminobenzidine (DAB) in the presence of hydrogen peroxide (H2O2). HRP oxidizes DAB to a brown polymeric precipitate in the presence of H2O2. As with Basic Protocol 1, a number of alternative antibody combinations are presented. Selection depends on the label conjugated to the hybridization probe (fluorescein, digoxigenin, or biotin) and on the desired sensitivity. High sensitivity is obtained by incubations with multiple immunocytochemical reagents, resulting in the formation of large complexes containing multiple HRP reaction sites. Antibody combinations for this approach are shown in Table 8.4.2, and their use is described in steps 2a through 6a. When lower-sensitivity detection is adequate, a single-antibody procedure (step 2b) can be substituted for these steps. Materials ISH-labeled preparation: chromosomes hybridized with fluorescein (FITC)–, digoxigenin-, or biotin-labeled probes (UNIT 8.3) First antibody for high sensitivity (see Table 8.4.2) TNT buffer (see recipe) Second antibody for high sensitivity (see Table 8.4.2) Third antibody for high sensitivity (see Table 8.4.2) Horseradish peroxidase (HRP)–conjugated antibody for low sensitivity (select one): 1:100 HRP-sheep-anti-digoxigenin (Boehringer Mannheim), 1:100 HRP-streptavidin (Vector Labs), or 1:500 HRP-sheep-anti-FITC (NEN Life Sciences) DAB substrate solution (see recipe) Hematoxylin 70%, 90%, and 100% ethanol 1:1 (v/v) ethanol/xylene 100% xylene Fluoromount mounting medium (BDH) Schiefferdecker or Coplin jars (Fisher Scientific) 24 × 50–mm coverslips Additional reagents and equipment for handling of ISH samples and immunoreagents (see Support Protocol 1)
Immunocytochemical Detection
Perform antibody labeling 1. Wash and block ISH-labeled preparation (see Support Protocol 1).
8.4.4 Supplement 4
Current Protocols in Cytometry
Table 8.4.2 Antibodies for High-Sensitivity Detection in Basic Protocol 2a,b
Dig-labeled I
Dig-labeled II
Biotin-labeled
FITC-labeled I
FITC-labeled II
1st antibody
Mouse anti-digoxigenin (1:200; BM)
Mouse anti-digoxinc (1:500; S)
Streptavidind (1:100; V)
Rabbit anti-FITC (1:250; D)
Mouse anti-FITC (1:250; D)
2nd antibody
HRP-labeled rabbit anti-mouse (1:1000; D)
Dig-labeled sheep anti-mouse (1:100; BM)
Bio-labeled goat anti-streptavidin (1:100; V)
HRP-labeled goat anti-rabbit (1:500; D)
HRP-labeled rabbit anti-mouse (1:500; D)
3rd antibody
HRP-labeled goat antirabbit (1:1000; D)
HRP-labeled sheep anti-dig (1:100; BM)
HRP-labeled streptavidin (1:100; V)
None
None
aSuppliers and appropriate dilutions are indicated in parentheses. Proper titration of each antibody is required to determine the optimal dilution (see
Critical Parameters and Troubleshooting). BM, Boehringer Mannheim; D, Dako; S, Sigma; V, Vector Labs. bBio, biotin; Dig, digoxigenin; FITC, fluorescein; HRP, horseradish peroxidase. cIt has been demonstrated that an anti-digoxin antibody must be used in this alternate protocol to avoid background problems caused by the use of mouse anti-digoxigenin antibody. dBiotinylated goat anti-streptavidin does not recognize enzyme-conjugated (either AP or HRP) streptavidin. Therefore, the first incubation is done with unconjugated streptavidin.
For high-sensitivity protocol 2a. Incubate preparation 30 min at 37°C with the first antibody for high sensitivity (see Table 8.4.2; see Support Protocol 1). For hybridization probes labeled with digoxigenin or with FITC, the experimenter can select from two series of antibodies (see Table 8.4.2) that offer similar sensitivity.
3a. Wash 3 times for 5 min each with TNT buffer. 4a. Incubate 30 min at 37°C with the second antibody. For FITC-labeled hybridization probes, no third antibody is used (see Table 8.4.2). Proceed directly to steps 7 through 16.
5a. Wash 3 times for 5 min each with TNT buffer. 6a. Incubate 30 min at 37°C with the third antibody. Proceed with steps 7 through 16. For low-sensitivity protocol 2b. Incubate preparation 30 min at 37°C with the appropriate HRP-conjugated antibody for low sensitivity (see Support Protocol 1). Proceed with steps 7 through 16. Each antibody should be properly titrated to determine the optimal dilution (see Critical Parameters and Troubleshooting).
Perform DAB reaction 7. Wash 3 times for 5 min each with TNT buffer. 8. Wash briefly with water. 9. Prepare DAB substrate solution just before use. DAB is a potential carcinogen. Handle with caution and wear appropriate protective clothing.
10. Incubate the preparation in 100 ml DAB substrate solution in a Schiefferdecker or Coplin jar, for 20 min in the dark at room temperature. DAB staining can be monitored (e.g., after 10 min) by briefly rinsing the slide with tap water and covering it with a coverslip. Judge the intensity of the DAB staining with the aid of a bright-field microscope. If staining is too faint, proceed with the DAB staining for an additional 10 min.
Molecular Cytogenetics
8.4.5 Current Protocols in Cytometry
Supplement 4
11. Wash briefly with demineralized water. 12. Counterstain 20 sec with 100 ml hematoxylin in a Schiefferdecker or Coplin jar. 13. Rinse the preparation 10 min under running tap water. Hematoxylin counterstaining is critical and must be experimentally optimized. When staining is too faint, prolong the staining time. When staining intensity is too high, reduce by rinsing the preparation for a limited amount of time (e.g., sec) in 1% HCl/70% ethanol. After HCl/ethanol treatment, rinse another 10 min with running tap water.
14. Dehydrate using a series of 5-min washes in 70%, 90%, and 100% ethanol, followed by 15-min washes in 1:1 (v/v) ethanol/xylene and 100% xylene. 15. Air dry the preparation. 16. Put a few drops of Fluoromount mounting medium on the preparation and cover with a 24 × 50–mm coverslip. BASIC PROTOCOL 3
SINGLE-COLOR DETECTION FOR FISH USING BIOTIN-LABELED HYBRIDIZATION PROBES In this protocol, procedures are presented for fluorescent detection following FISH using biotinylated probes. As with enzyme-linked detection methods (see Basic Protocols 1 and 2), a discrimination is made between low-sensitivity and high-sensitivity fluorescence detection. Generally, highly repetitive sequences, if not detectable by direct FISH techniques (see UNIT 8.3), require only one layer of fluorochrome-labeled anti-hapten reagent (i.e., streptavidin). Middle- and single-copy sequences need higher-sensitivity detection (i.e., multiple cycles of antibody incubation). In this case, the signal from the first incubation with fluorochrome-labeled streptavidin is amplified by incubation with a biotinylated anti-streptavidin antibody, followed by a third incubation with the same fluorochrome-labeled streptavidin. Thus, the amplification is similar to the enzymelinked amplifications for biotinylated probes (see Tables 8.4.1 and 8.4.2), except that streptavidin is fluorochrome labeled at both steps. Materials ISH-labeled preparation: chromosomes hybridized with a biotinylated probe (UNIT 8.3) 1:100 fluorochrome-streptavidin (Vector Labs) TNT buffer (see recipe) 1:100 biotin-goat-anti-streptavidin (Vector Labs) Additional reagents and equipment for handling of ISH samples and immunoreagents (see Support Protocol 1) and for embedding the preparation for fluorescence microscopy (see Support Protocol 2) Perform single streptavidin labeling (high and low sensitivity) 1. Wash and block ISH-labeled preparation (see Support Protocol 1). 2. Incubate preparation 30 min at 37°C with 1:100 fluorochrome-streptavidin (see Support Protocol 1). 3. Wash 3 times for 5 min each with TNT buffer. If low-sensitivity detection is adequate, the preparation is now ready to be dehydrated and embedded for fluorescence microscopy (see Support Protocol 2).
Immunocytochemical Detection
Perform amplification of streptavidin labeling (high sensitivity only) 4. Incubate preparation 30 min at 37°C with 1:100 biotin-goat-anti-streptavidin.
8.4.6 Supplement 4
Current Protocols in Cytometry
5. Wash 3 times for 5 min each with TNT buffer. 6. Incubate preparation 30 min at 37°C with 1:100 fluorochrome-streptavidin. 7. Wash 3 times for 5 min each with TNT buffer. The high-sensitivity preparation is now ready to be dehydrated and embedded for fluorescence microscopy (see Support Protocol 2).
SINGLE-COLOR DETECTION FOR FISH USING DIGOXIGENIN-LABELED HYBRIDIZATION PROBES
ALTERNATE PROTOCOL 1
An alternate fluorescence protocol (see Basic Protocol 3) is presented for use with digoxigenin-labeled ISH probes. For lower sensitivity, the digoxigenin-labeled probe is detected by incubation with a fluorochrome-conjugated anti-digoxigenin antibody. As in Basic Protocols 1, 2, and 3, higher sensitivity is obtained by incubation with multiple antibodies. As with the first alternative antibody series for enzyme-based detection (see Tables 8.4.1 and 8.4.2), both secondary and tertiary antibodies are labeled to enhance the signal. Materials ISH-labeled preparation: chromosomes hybridized with a digoxigenin-labeled probe (UNIT 8.3) 1:100 fluorochrome-sheep-anti-digoxigenin (Boehringer Mannheim) TNT buffer (see recipe) 1:500 mouse anti-digoxin (Sigma) 1:200 mouse anti-digoxigenin (Boehringer Mannheim) 1:1000 fluorochrome-rabbit-anti-mouse (Sigma) 1:250 fluorochrome-goat-anti-rabbit (Vector Labs) Additional reagents and equipment for handling of ISH samples and immunoreagents (see Support Protocol 1) and for embedding the preparation for fluorescence microscopy (see Support Protocol 2) Perform single antibody labeling (low sensitivity only) 1a. Wash and block ISH-labeled preparation (see Support Protocol 1). 2a. Incubate preparation 30 min at 37°C with 1:100 fluorochrome-sheep-anti-digoxigenin (see Support Protocol 1). 3a. Wash 3 times for 5 min each with TNT buffer. The preparation is now ready to be dehydrated and embedded for fluorescence microscopy (see Support Protocol 2).
Perform multiple antibody labeling (high sensitivity only) 1b. Wash and block ISH-labeled preparation (see Support Protocol 1). 2b. Incubate preparation 30 min at 37°C with 1:500 mouse anti-digoxin or 1:200 mouse anti-digoxigenin (see Support Protocol 1). 3b. Wash 3 times for 5 min each with TNT buffer. 4b. Incubate 30 min at 37°C with 1:1000 fluorochrome-rabbit-anti-mouse. 5b. Wash 3 times for 5 min each with TNT buffer. 6b. Incubate preparation 30 min at 37°C with 1:250 fluorochrome-goat-anti-rabbit. 7b. Wash 3 times for 5 min each with TNT buffer. The preparation is now ready to be dehydrated and embedded for fluorescence microscopy (see Support Protocol 2).
Molecular Cytogenetics
8.4.7 Current Protocols in Cytometry
Supplement 4
ALTERNATE PROTOCOL 2
SINGLE-COLOR DETECTION FOR FISH USING FITC-LABELED HYBRIDIZATION PROBES In this alternate fluorescence protocol (also see Basic Protocol 3), fluorescein is used both as the hapten and the fluorochrome. This protocol is useful when the direct signal obtained by using FITC as a hapten is not sufficiently sensitive. The signal is amplified by incubation with an anti-fluorescein antibody, followed by a fluorescein-labeled secondary antibody. Materials ISH-labeled preparation: chromosomes hybridized with a FITC-labeled probe (UNIT 8.3) 1:250 rabbit anti-FITC (Dako) TNT buffer (see recipe) 1:1000 FITC-goat-anti-rabbit (Vector Labs) Additional reagents and equipment for handling of ISH samples and immunoreagents (see Support Protocol 1) and for embedding the preparation for fluorescence microscopy (see Support Protocol 2) 1. Wash and block ISH-labeled preparation (see Support Protocol 1). 2. Incubate preparation 30 min at 37°C with 1:250 rabbit anti-FITC (see Support Protocol 1). 3. Wash 3 times for 5 min each with TNT buffer. 4. Incubate preparation 30 min at 37°C with 1:1000 FITC-goat-anti-rabbit. 5. Wash 3 times for 5 min each with TNT buffer. The preparation is now ready to be dehydrated and embedded for fluorescence microscopy (see Support Protocol 2).
BASIC PROTOCOL 4
DUAL-COLOR DETECTION FOR FISH USING BIOTIN- AND DIGOXIGENIN-LABELED HYBRIDIZATION PROBES This procedure is a modification of single-color FISH (see Basic Protocol 3 introduction) that uses a mixture of two differently labeled immunoreagents to facilitate the simultaneous detection of one probe using green fluorescence and another probe using red fluorescence. Dual-labeling is done using appropriately paired antibody cocktails. Lowand high-sensitivity protocols are presented. Materials ISH-labeled preparation: chromosomes hybridized with one biotin-labeled and one digoxigenin-labeled probe (UNIT 8.3) 1:100 FITC-sheep-anti-digoxigenin/1:100 Texas red–streptavidin (Boehringer Mannheim and Vector Labs, respectively) TNT buffer (see recipe) 1:200 mouse anti-digoxigenin/1:100 Texas red–streptavidin (Boehringer Mannheim and Vector Labs, respectively) 1:1000 FITC-rabbit-anti-mouse/1:100 biotin-goat-anti-streptavidin (Sigma and Vector Labs, respectively) 1:250 FITC-goat-anti-rabbit/1:100 Texas red–streptavidin (Vector Labs)
Immunocytochemical Detection
Additional reagents and equipment for handling of ISH samples and immunoreagents (see Support Protocol 1) and for embedding the preparation for fluorescence microscopy (see Support Protocol 2)
8.4.8 Supplement 4
Current Protocols in Cytometry
Perform single-antibody labeling (low sensitivity only) 1a. Wash and block ISH-labeled preparation (see Support Protocol 1). 2a. Incubate preparation 30 min at 37°C with 1:100 FITC-sheep-anti-digoxigenin/1:100 Texas red–streptavidin (see Support Protocol 1). 3a. Wash 3 times for 5 min each with TNT buffer. The preparation is now ready to be dehydrated and embedded for fluorescence microscopy (see Support Protocol 2).
Perform multiple-antibody labeling (high sensitivity only) 1b. Wash and block ISH-labeled preparation (see Support Protocol 1). 2b. Incubate preparation 30 min at 37°C with 1:200 mouse anti-digoxigenin/1:100 Texas red–streptavidin (see Support Protocol 1). 3b. Wash 3 times for 5 min each with TNT buffer. 4b. Incubate 30 min at 37°C with 1:1000 FITC-rabbit-anti-mouse/1:100 biotin-goat-antistreptavidin. 5b. Wash 3 times for 5 min each with TNT buffer. 6b. Incubate 30 min at 37°C with 1:250 FITC-goat-anti-rabbit/1:100 Texas red–streptavidin. 7b. Wash 3 times for 5 min each with TNT buffer. The preparation is now ready to be dehydrated and embedded for fluorescence microscopy (see Support Protocol 2).
TRIPLE-COLOR DETECTION FOR FISH In this fluorescence protocol (see Basic Protocol 3 introduction), ISH experiments using three probes labeled with different haptens can be distinguished with blue, red, and green fluorochromes. The preparation is incubated with a mixture of two differently labeled immunoreagents facilitating the detection of a digoxigenin-labeled probe using blue fluorescence and a biotinylated probe using red fluorescence. A FITC-labeled probe is detected directly using green fluorescence. To avoid cross-reactivity of antibodies, the detection of the digoxigenin-labeled probe has been modified for the high-sensitivity protocol.
BASIC PROTOCOL 5
Materials ISH-labeled preparation: chromosomes hybridized with one digoxigenin-labeled probe, one biotinylated probe, and one FITC-labeled probe (UNIT 8.3) 1:100 aminomethyl coumarin acetic acid (AMCA)-sheep-anti-digoxigenin/1:100 Texas red–streptavidin (Boehringer Mannheim and Vector Labs, respectively) TNT buffer (see recipe) 1:500 mouse anti-digoxin/1:250 rabbit anti-FITC/1:100 Texas red–streptavidin (Sigma, Dako, and Vector Labs, respectively) 1:100 digoxigenin-sheep-anti-mouse/1:250 FITC-goat-anti-rabbit/1:100 biotin-goat-anti-streptavidin (Boehringer Mannheim, Vector Labs, Vector Labs, respectively) Additional reagents and equipment for handling of ISH samples and immunoreagents (see Support Protocol 1) and for embedding the preparation for fluorescence microscopy (see Support Protocol 2)
Molecular Cytogenetics
8.4.9 Current Protocols in Cytometry
Supplement 4
Perform single-antibody labeling (low sensitivity only) 1a. Wash and block ISH-labeled preparation (see Support Protocol 1). 2a. Incubate preparation 30 min at 37°C with 1:100 AMCA-sheep-anti-digoxigenin/1:100 Texas red–streptavidin (see Support Protocol 1). 3a. Wash 3 times for 5 min each with TNT buffer. The preparation is now ready to be dehydrated and embedded for fluorescence microscopy (see Support Protocol 2).
Perform multiple-antibody labeling (high sensitivity only) 1b. Wash and block ISH-labeled preparation (see Support Protocol 1). 2b. Incubate preparation 30 min at 37°C with 1:500 mouse anti-digoxin/1:250 rabbit anti-FITC/1:100 Texas red–streptavidin (see Support Protocol 1). 3b. Wash 3 times for 5 min each with TNT buffer. 4b. Incubate preparation 30 min at 37°C with 1:100 digoxigenin-sheep-anti-mouse/1:250 FITC-goat-anti-rabbit/1:100 biotin-goat-anti-streptavidin. 5b. Wash 3 times for 5 min each with TNT buffer. 6b. Incubate preparation 30 min at 37°C with 1:100 AMCA-sheep-anti-digoxigenin/1:100 Texas red–streptavidin. 7b. Wash 3 times for 5 min each with TNT buffer. The preparation is now ready to be dehydrated and embedded for fluorescence microscopy (see Support Protocol 2). BASIC PROTOCOL 6
TYRAMIDE SIGNAL AMPLIFICATION Tyramide signal amplification (TSA) combines enzyme cytochemistry with fluorescent detection. It is based on the peroxidase-mediated deposition of tyramide (composed of a fluorochrome or hapten attached to the phenolic compound tyramine) close to the hybridized probe. Tyramide is transformed into a highly reactive radical when incubated with bound horseradish peroxidase (HRP) in the presence of hydrogen peroxide (H2O2). The radicals deposit on the tyrosine moieties of proteins close to the bound HRP, or even on the HRP itself. A number of procedural alternatives exist that depend on the requirements of the experiment. The preparation is incubated with the appropriate HRP-conjugated anti-hapten reagent and washed with TNT buffer (see Basic Protocol 2). It is then incubated in a horizontal position with the tyramide solution in the presence of H2O2. In direct TSA, a fluorochrome-labeled tyramide is used, and the preparation is examined by fluorescence microscopy. In indirect TSA, biotin-labeled tyramide is used. By using the appropriately labeled streptavidin, it is then possible either to perform the enzyme reaction for signal detection by bright-field microscopy, or to repeat the tyramide reaction for further signal amplification. Each of these alternatives is presented in this protocol. The repeated TSA protocol that is presented uses a fluorochrome-labeled tyramide for fluorescent detection. These steps can be modified to accommodate hapten-labeled tyramides for enzyme-based detection.
Immunocytochemical Detection
8.4.10 Supplement 4
Current Protocols in Cytometry
Materials Tyramide solution (see recipe) HRP-labeled preparation (low-sensitivity procedure; see Basic Protocol 2, steps 1, 2b, and 7) TNT buffer (see recipe) 1:500 AP-, HRP-, or fluorochrome-streptavidin (Vector Labs) Additional reagents and equipment for handling of ISH samples and immunoreagents (see Support Protocol 1), embedding the preparation for fluorescence microscopy (see Support Protocol 2), and for alkaline phosphatase (see Basic Protocol 1) and peroxidase (see Basic Protocol 2) staining 1. Prepare 1 ml tyramide solution. The optimal tyramide concentration should be determined experimentally if it is not indicated by the manufacturer. For direct TSA, use any fluorochrome-tyramide. For indirect TSA, use biotin-tyramide. For fluorescent detection of a repeated TSA reaction, prepare both biotin- and fluorochrometyramides (for use in steps 2 and 5c, respectively).
2. Apply 1 ml tyramide solution to a horizontally positioned HRP-labeled preparation and incubate 30 min at room temperature. 3. Wash 4 times for 5 min each with TNT buffer at 37°C. For direct TSA 4a. Dehydrate and embed the preparation for fluorescence microscopy (see Support Protocol 2). For indirect, single-reaction TSA 4b. Incubate the preparation 30 min at 37°C with 1:500 labeled streptavidin. For fluorescent detection, use any fluorochrome-streptavidin. For bright-field microscopy, HRP- or AP-streptavidin is required.
5b. Wash 3 times for 5 min each with TNT buffer. 6b. Dehydrate and embed the preparation for fluorescence microscopy (see Support Protocol 2) or stain for alkaline phosphatase (see Basic Protocol 1) or peroxidase (see Basic Protocol 2) for bright-field microscopy. For indirect, repeated TSA 4c. Incubate the preparation 30 min at 37°C with 1:500 HRP-labeled streptavidin. 5c. Wash 3 times for 5 min each with TNT buffer. 6c. Apply 1 ml fluorochrome-tyramide solution to the horizontally positioned preparation and incubate 30 min at room temperature. 7c. Wash 4 times for 5 min each with TNT buffer at 37°C. The preparation is now ready to be dehydrated and embedded (see Support Protocol 2) for fluorescence microscopy. For bright-field microscopy, a biotinylated tyramide (NEN Life Sciences) can be used in step 6c. The wash in step 7c is then followed by incubation with AP- or HRP-labeled streptavidin and the appropriate enzyme detection procedure (see Basic Protocols 1 and 2). Molecular Cytogenetics
8.4.11 Current Protocols in Cytometry
Supplement 4
SUPPORT PROTOCOL 1
HANDLING OF PREPARATIONS AND REAGENTS FOR IMMUNOCYTOCHEMICAL DETECTION This protocol describes the handling of preparations during the entire immunocytochemical procedure, and applies to all of the Basic and Alternate Protocols in this unit. It consists of washing the preparations with the immunological buffer after the posthybridization washes (UNIT 8.3), incubating in a blocking buffer to prevent nonspecific binding of immunoreagents, and performing the immunocytochemical detection reaction(s). Materials ISH-labeled preparation: chromosomes hybridized by appropriate probe (UNIT 8.3) TNT buffer (see recipe) TNB buffer (see recipe) Immunoreagents (see recipe) Schiefferdecker or Coplin jars 24 × 50–mm glass coverslips Moist chamber: 1-liter beaker containing paper towels moistened with TNT buffer (see recipe), covered with aluminum foil Wash slides 1. After the posthybridization washes (UNIT 8.3), wash the ISH-labeled preparation 5 min at room temperature in a Schiefferdecker or Coplin jar with sufficient TNT buffer to fill the jar (∼100 ml). 2. Remove the preparation and drain off most of the buffer by putting the preparation in an upright position on tissue paper. Block nonspecific binding 3. Apply 100 µl TNB buffer to the preparation and cover with a 24 × 50–mm coverslip. Incubate the preparation in a horizontal position 30 min at 37°C in a moist chamber. Prepare diluted immunoreagents 4. Prepare fresh dilutions in TNB buffer of the anti-hapten reagent(s) and other immunoreagents according to the dilutions given in the Basic and Alternate Protocols. Microcentrifuge 1 min before applying to the preparations to prevent nonspecific binding of antibody aggregates. Perform immunocytochemical detection reaction(s) 5. Briefly wash the preparation in TNT buffer to loosen and remove the coverslip. Drain off most of the buffer by putting the preparation in an upright position on tissue paper. 6. Apply 100 µl diluted anti-hapten reagent to the preparation and cover with a 24 × 50–mm coverslip. 7. Incubate the preparation in a horizontal position in a moist chamber at 37°C for 30 min. 8. Wash the preparation briefly in 100 ml TNT buffer at room temperature to loosen the coverslip and transfer the preparation (without coverslip) to another jar containing 100 ml TNT buffer. Wash 3 times for 5 min each, with gentle agitation.
Immunocytochemical Detection
9. Proceed with either the next antibody incubation or with the procedures for brightfield microscopy (see Basic Protocols 1 and 2 for staining procedures) or fluorescence microscopy (see Support Protocol 2 for embedding procedure).
8.4.12 Supplement 4
Current Protocols in Cytometry
EMBEDDING OF PREPARATIONS FOR EVALUATION BY FLUORESCENCE MICROSCOPY
SUPPORT PROTOCOL 2
After completion of an immunofluorescent detection reaction, the preparation is washed with the immunological buffer, dehydrated through an ethanol series, and air dried. An antifading solution is then applied to the preparation to minimize photobleaching. The antifading reagent may contain a general DNA counterstain that contrasts with the specific FISH signals. The following general DNA counterstains are recommended: red propidium iodide (PI) when green fluorescent reporter molecules are used, blue DAPI or green YOYO-1 (Molecular Probes) when red fluorescent reporter molecules are used, and DAPI when both red and green reporter molecules are used. Counterstaining is omitted when aminomethyl coumarin acetic acid (AMCA) is being used as the reporter molecule, because of interference between the emissions of the counterstain(s) and AMCA (see Critical Parameters and Troubleshooting). Materials Fluorescently labeled ISH preparation (refer to appropriate labeling protocol), washed in TNT buffer 70%, 90%, and 100% ethanol Vectashield with or without DNA counterstain (see recipe) 24 × 50–mm glass coverslips 1. Dehydrate the preparation using 3-min washes in 70%, 90%, and 100% ethanol. Air dry. 2. Apply 15 µl Vectashield (with or without DNA counterstain) and cover with a 24 × 50–mm coverslip. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Alkaline phosphatase (AP) buffer 0.2 M Tris⋅Cl, pH 9.6 10 mM MgCl2 Prepare fresh before use Tris is added in the form of a 1 M stock prepared as follows: Dissolve 121.14 g Tris base in 800 ml water, adjust pH to 9.6 with 6 M HCl, and add water to 1 liter.
Alkaline phosphatase (AP) substrate solution 3.2 µl 5-bromo-4-chloro-3-indolylphosphate (BCIP) 6.4 µl nitroblue tetrazolium (NBT) 1 ml AP buffer (see recipe) Prepare fresh before use Ready-to-use BCIP and NBT solutions (both in dimethylformamide) may be stored up to several years at −20°C.
DAB (3,3′-diaminobenzidine⋅4HCl), 1% Prepare 100 ml of 1% (w/v) DAB in water. Filter through 0.22-µm filter. Store 1-ml aliquots up to several years at −20°C. Diaminobenzidine is a potential carcinogen. Handle with caution and wear appropriate protective clothing. Molecular Cytogenetics
8.4.13 Current Protocols in Cytometry
Supplement 4
DAB substrate solution 89 ml H2O 5 ml 1 M Tris⋅Cl, pH 7.5 (APPENDIX 2A) 5 ml 1% DAB (see recipe) 1 ml 1 M imidazole, pH 8 (see recipe) 167 µl 30% H2O2 Prepare fresh before use Use a fresh dilution of H2O2.
Dextran sulfate, 25% Dissolve 25 g dextran sulfate in 90 ml TE buffer (APPENDIX 2A) by stirring at least 3 hr at 70°C. Cool to room temperature and adjust the volume to 100 ml with TE buffer. Mix thoroughly. Store up to several years at 4°C. 4′,6-Diamidino-2-phenylindole (DAPI) Prepare 10 µg/ml DAPI in H2O. Store up to several years at −20°C. Imidazole, pH 8, 1 M Dissolve 6.808 g imidazole in 90 ml water and adjust pH to 8 with 6 M HCl. Adjust volume to 100 ml with H2O. Can be stored >1 year at room temperature. Immunoreagents Fluorochrome-labeled anti-hapten reagents can be stored in 100-µl aliquots up to several years at −20°C. Thawed aliquots can be stored up to several months at 4°C. Enzyme-labeled anti-hapten reagents can be stored up to 1 year at 4°C. Repeated freezing/thawing of fluorochrome-labeled reagents is strongly discouraged.
Propidium iodide (PI) solution Prepare 10 µg/ml PI in H2O. Store up to several years at −20°C. TNB buffer Dissolve 0.876 g NaCl and 1.211 g Tris base in 80 ml water. Adjust pH to 7.5 with 6 M HCl. Adjust volume to 100 ml with H2O. Add 0.5 g blocking reagent (Boehringer Mannheim, NEN Life Sciences) and stir 1 hr at 60°C to dissolve. Cool to room temperature. Add 100 µl 20% thimerosal as a preservative. Divide into 10-ml aliquots and store up to several years at −20°C. Thawed aliquots may be frozen again or can be stored several months at 4°C without significant loss of activity.
TNT buffer, 10× 121.14 g Tris base (1 M) 87.6 g NaCl (1.5 M) 800 ml H2O Adjust pH to 7.5 with 6 M HCl Add 50 ml 10% Tween 20 (final 0.5%) Bring to 1 liter with H2O Store up to several weeks at room temperature Dilute 1:10 before use
Immunocytochemical Detection
Tyramide solution 10 µl 1 M imidazole, pH 8 (see recipe) 500 µl 2× amplification diluent (ready-to-use solution; NEN Life Sciences) 400 µl 25% dextran sulfate (see recipe) 10-20 µl biotin- or fluorochrome-tyramide (NEN Life Sciences) Adjust volume to 1 ml with H2O Prepare fresh and mix thoroughly
8.4.14 Supplement 4
Current Protocols in Cytometry
Vectashield with or without DNA counterstain Vectashield (Vector Labs) comes as a ready-to-use solution. If desired, a general DNA counterstain can be added. Use 4 µl 10 µg/ml DAPI solution (see recipe) or 10 µl 10 µg/ml PI solution (see recipe) in 1 ml Vectashield and mix thoroughly. Store up to 1 year at 4°C. Alternatively, prepare a 0.1 µM YOYO-1 (Molecular Probes) solution in Vectashield immediately before use. COMMENTARY Background Information Fluorescence and bright-field in situ hybridization (FISH and BfISH, respectively) allow the detection of hybridization signals without the use of radiolabeled nucleotides (for a discussion of appropriate probe-labeling and hybridization procedures, see UNIT 8.3; for a historic overview see UNIT 8.1). Hybridization can be done with fluorochrome-labeled probes (direct FISH) or with hapten-labeled probes (indirect FISH). For indirect FISH there is a wide variety of specific procedures and antibody combinations that can be utilized, including both fluorescent and enzyme-based detection methods, as outlined in this unit. The choice of procedure will vary depending on a number of factors. The following discussions address these choices (also see Strategic Planning). Detection procedures for FISH The rich choice in antibodies conjugated to a great variety of fluorochromes makes it impossible to design uniformly applicable protocols. The detailed protocols in this unit provide the best results in the authors’ hands and are recommended as such. They can be summarized as follows. When centromere-associated alpha-satellite probes (or probes for other similarly repetitive sequences) are not detectable by direct FISH, only one fluorochrome-conjugated anti-hapten reagent incubation is required, no matter which fluorochrome is being used. However, aminomethyl coumarin acetic acid (AMCA) performs less well than fluorescein, rhodamine, or cyanine dyes. In addition, the frequently used mounting medium Vectashield (Vector Labs) shows autofluorescence when excited with UV light (Florijn et al., 1995), which obscures AMCA signals. Although the autofluorescence disappears after a few seconds of illumination, the combination of AMCA and Vectashield is not recommended. Other antifading reagents, such as Prolong (Molecular Probes), DABCO (Sigma), and Citifluor (Agar), are recommended when AMCA is used. For single- or
dual-color applications, it is best to use one of the green or red fluorochromes, such as fluorescein, rhodamine, Texas red, lissamine, and Cy3. In all cases anti-digoxigenin or anti-biotin reagents can be used. Probes for single-copy targets need conventional amplification strategies. This class of probes is best detected with the aforementioned red and green fluorochromes using successive incubations of staining reagents. For biotinylated probes, amplification is achieved using fluorochrome-streptavidin, followed by biotinylated anti-streptavidin, and another incubation in the same fluorochrome-streptavidin (Pinkel et al., 1986). For digoxigenin-labeled probes, mouse anti-digoxigenin is followed by fluorochrome-labeled rabbit anti-mouse, and then by an anti-rabbit antibody labeled with the same fluorochrome (Linsenmayer et al., 1988). For the detection of a small cDNA (1 to 3 kb), which is in practice one of the most difficult situations, sequential incubations of FITCstreptavidin, biotinylated goat anti-streptavidin, and FITC-streptavidin are strongly recommended. This combination of reagents provides for the best signal-to-noise ratio. In two-color FISH, the combination of FITC and Texas red is chosen because their emission spectra enable simultaneous visualization using double-bandpass fluorescence filter combinations without crosstalk. Detection procedures for BfISH The detection of probes for single-copy targets using bright-field microscopy is cumbersome. As discussed previously (see Strategic Planning), BfISH is on fundamental grounds less sensitive than FISH. In practice it has been demonstrated that the detection of a cosmid probe in paraffin-embedded tissue is difficult. The detection of a cDNA on chromosomes is practically impossible. The BfISH approach is more or less limited to repeat sequences, which need only one antibody incubation. As in FISH, no significant difference in detection sensitivity is found between biotin-labeled probes and digoxigenin-labeled probes.
Molecular Cytogenetics
8.4.15 Current Protocols in Cytometry
Supplement 4
TSA Tyromide signal amplification represents a new and sensitive immunocytochemical detection system for FISH that combines enzyme cytochemistry with fluorescent detection. A general tyramide-based FISH experiment has the following outline. 1. ISH with any hapten (biotin, digoxigenin, fluorescein, dinitrophenol)–labeled probe. 2. Incubation with an HRP-conjugated anti-hapten reagent. 3. Incubation with fluorochrome- or hapten-labeled tyramide in the presence of H2O2. 4a. Fluorescent detection (direct TSA for FISH) in the case of a fluorochrome labeled tyramide, or 4b. conventional immunocytochemical detection (indirect TSA for both FISH and BfISH) in the case of a hapten-labeled tyramide reaction, or 4c. a repeated tyramide reaction (also indirect TSA for both FISH and BfISH). A particular advantage of TSA detection strategies is that the probe concentration for ISH can be lowered by at least a factor of ten compared to conventional detection. In addition, when applied to the detection of repeat sequences, the hybridization time can be reduced to a few hours, enabling completion of a FISH experiment in one day. Diffusion of the tyramide reaction product can be a problem if no measures are taken to prevent it. When diffusion occurs, FISH signals may become much larger than the actual hybrid, resulting in loss of spatial information. The addition of polymers such as polyvinylalcohol or dextran sulfate dramatically decreases diffusion. Localization of the fluorescent product is thus increased and is only slightly worse than that achieved with conventional detection methods (Van Gijlswijk et al., 1996a). Recently, the use of synthetic oligonucleotides coupled to peroxidase has been introduced as a fast and elegant FISH technique for the detection of chromosome-specific repeat sequences (Van Gijlswijk et al., 1996b). The favorable hybridization kinetics of small oligonucleotides (20 to 40 bp) in combination with sensitive TSA detection allows for a short hybridization time (20 min) at a low probe concentration.
Critical Parameters and Troubleshooting Immunocytochemical Detection
because of improper use of the immunoreagents. The following guidelines can help insure proper immunocytochemical detection. In multiple detection protocols, care must be taken to prevent antibodies from cross-reacting. Dilution of reagents. A common mistake in this respect is the use of incorrect dilutions. In the protocols presented here, dilutions of immunocytochemical reagents are given with respect to stock solutions supplied by companies, rather than in weight per volume. This is done because some reagents are supplied as lyophilized powders (which have to be dissolved according to the manufacturers’ instructions) and others are supplied as solutions containing either mg protein/ml or units/ml. The indicated dilutions of immunocytochemical reagents must be used as guidelines only. Manufacturers’ instructions do not always provide guaranteed success. To achieve the maximal signal-tonoise ratio, the user is strongly urged to determine the optimal dilution experimentally each time a new reagent is introduced. Digoxigenin and digoxin. Antibodies against digoxigenin and digoxin are available from Boehringer Mannheim and Sigma, respectively. Antibody conjugates against digoxin recognize digoxigenin equally well, since the molecular structures of the two compounds show great similarity. However, it has been demonstrated that in certain protocols an antidigoxin antibody must be used to avoid background problems caused by the use of mouse anti-digoxigenin antibody. Avidin and streptavidin. In biotin detection for FISH or BfISH applications, streptavidin is preferred over avidin based on its low nonspecific binding to DNA, which results from its neutral isoelectric point. In contrast, avidin conjugates are positively charged at neutral pH because they have an isoelectric point of about ten. As a result, they show stronger nonspecific binding to negatively charged DNA. They are not recommended for single-copy probe detection. Storage of reagents. Fluorochrome-conjugated anti-hapten reagents at best can be stored in 100-µl aliquots at −20°C for up to 2 years. Once thawed, they should be stored at 4°C for up to 3 months. Enzyme-conjugated anti-hapten reagents do not survive at −20°C and should always be kept at 4°C (up to 1 year). For all reagents, quality should be carefully monitored, because some antibodies may gradually degrade at 4°C.
Immunocytochemical reactions Most immunocytochemical incubations fail
8.4.16 Supplement 4
Current Protocols in Cytometry
Enzyme reactions Alkaline phosphatase (AP) is detected in a reaction with 5-bromo-4-chloro-3-indolylphosphate (BCIP) and nitroblue tetrazolium (NBT; see Basic Protocol 1), and horseradish peroxidase (HRP) is detected in a reaction with 3,3′-diaminobenzidine (DAB; see Basic Protocol 2). The HRP/DAB reaction is generally preferred over the AP/NBT-BCIP reaction because it allows for better spatial resolution. The peroxidase/DAB endproduct localizes more closely to the probe and is more permanent. In contrast, the NBT-BCIP endproduct has a tendency to diffuse from its original site of precipitation upon prolonged storage. A proper DAB staining reaction is highly dependent on the concentration of H2O2. Diluted H2O2 solutions (and even 30% solutions) tend to degrade in time. Therefore it is strongly advised never to store or reuse diluted solutions. Choice of fluorophore Currently used optics and fluorescence filter technology in fluorescence microscopy allow for the utilization of a wide range of spectrally well-separable fluorochromes suitable for FISH. These include fluorochromes emitting in the blue, green, red, and infrared parts of the electromagnetic spectrum, though infraredemitters are detectable only by special camera systems. In addition, multibandpass fluorescence filters enable simultaneous visualization of two or three different fluorochromes, thereby
Table 8.4.3
expanding the possibilities for multicolor FISH. Table 8.4.3 shows fluorochromes commonly used for FISH, together with their excitation and emission characteristics. Currently, all of these fluorochromes are available conjugated to triphosphate-deoxyribonucleotides (enabling direct FISH) as well as to anti-hapten reagents (for indirect FISH). The choice of fluorochrome is determined by a number of factors. Generally, fluorophores with high quantum efficiencies and high molar absorption coefficients are chosen. In practice, these include fluorescein, rhodamine, and cyanine dyes, as well as their derivatives. Some potentially interesting fluorophores such as the phycobiliproteins are less well suited to FISH. Their relatively large molecular weight prevents them from effectively reaching nuclear target sequences. In addition to the physicochemical properties, some other aspects have to be considered. For instance, the human eye is most sensitive to green light. Additionally, the fluorochrome should contrast with autofluorescence of the biological object (if present). Counterstaining reagents In bright-field microscopy, brown HRP/DAB signals are counterstained with blue hematoxylin, and blue/purple AP/NBT-BCIP signals are counterstained with nuclear fast red. Both counterstain and specific signals can be viewed in one microscopic image.
Fluorochromes for FISHa
Excitation Emission (nm) (nm)
Fluorochrome
Color
Aminomethyl coumarin acetic acid (AMCA) 7-Hydroxy-coumarin Cy2 Fluorescein (FITC) Cy3 Rhodamine (TRITC) Lissamine Texas red Cy5 Cy 5.5 Cy 7
Blue
350
445
Blue Green Green Red Red Red Red Far red Near IR Near IR
388 489 495 550 555 570 591 649 675 743
447 506 520 570 580 590 612 670 694 767
aMost spectral data apply to unconjugated dyes in water. Absorption and emission
characteristics may change slightly upon conjugation to antibodies or to nucleic acids.
Molecular Cytogenetics
8.4.17 Current Protocols in Cytometry
Supplement 4
In FISH three prominent general DNA counterstains are recognized: orange/red-fluorescing propidium iodide (PI) in combination with green FISH signals, blue-fluorescing DAPI in combination with green and/or red FISH signals, and green fluorescing YOYO-1 in combination with red FISH signals. Since PI luminesces red at blue excitation, the conventional fluorescence filter block for FITC excitation can be used to visualize both red PI and green FITC. Simultaneous visualization of DAPI with green FISH signals and of YOYO-1 with red FISH signals requires specially developed multibandpass filters (see UNIT 2.4). Antifading reagents Because fluorescence is subject to fading, it is recommended that FISH preparations be embedded in an antifading reagent. Antifading reagents are often radical scavengers. Examples include Vectashield (Vector Labs), Prolong (Molecular Probes), and paraphenylene diamine (Sigma). Their performance depends on the type of fluorophore and on the applications, but fading is generally reduced by a factor of 5 to 10 (UNIT 2.5; Florijn et al., 1995). The general DNA counterstains mentioned above can often be dissolved in antifading reagents, combining antifading with counterstaining in a single step.
Anticipated Results Using the protocols in this unit, centromereassociated alpha-satellite probes can be visualized with almost 100% efficiency. In this context, efficiency is defined as the percent of objects (i.e., metaphase chromosomes or interphase nucleus) that have a specific signal. Cosmid, YAC, and library probes have a detection efficiency of ∼95%. cDNAs are much more difficult to detect, and success depends on their size. A 2-kb cDNA has a detection efficiency of ∼20% on metaphase chromosomes.
Time Considerations
Immunocytochemical Detection
From start to finish, conventional immunocytochemical detection in a BfISH experiment using only one anti-hapten reagent incubation and HRP/DAB staining is completed within 2.5 hr. The AP/NBT-BCIP system requires 3 to 4 hr. Each additional antibody incubation will take an extra 45 min, including a 15-min wash with TNT buffer. Conventional detection for FISH with only one antibody incubation takes ∼2 hr. As for BfISH, each additional antibody incubation requires an extra 45 min.
Direct TSA detection takes 2 hr; indirect TSA detection requires an additional 45 min. Any repeated TSA reaction takes another 50 min.
Literature Cited Florijn, R.J., Slats, J., Tanke, H.J., and Raap, A.K. 1995. Analysis of antifading reagents for fluorescence microscopy. Cytometry 19:177-182. Kerstens, H.M.J., Poddighe, P.J., and Hanselaar, A.G.J.M. 1995. A novel in situ hybridization signal amplification method based on the deposition of biotinylated tyramine. J. Histochem. Cytochem. 43:347-352. Linsenmayer, T.F., Fitch, J.M., and Schmid, T.M. 1988. Multiple-reaction cycling: A method for enhancement of the immunocytochemical signal of monoclonal antibodies. J. Histochem. Cytochem. 36:1075-1078. Nederlof, P.M., Robinson, D., Abuknesha, R., Wiegant, J., Hopman, A.H.N., Tanke, H.J., and Raap, A.K. 1989. Three-color fluorescence in situ hybridization for the simultaneous detection of multiple nucleic acid sequences. Cytometry 10:20-27. Pinkel, D., Straume, T., and Gray, J. 1986. Cytogenetic analysis using quantitative, high sensitivity fluorescence hybridization. Proc. Natl. Acad. Sci. U.S.A. 83:2934-2938. Raap, A.K., van de Corput, M.P.C., Vervenne, R.A.W., van Gijlswijk, R.P.M., Tanke, H.J., and Wiegant, J. 1995. Ultra-sensitive FISH using peroxidase-mediated deposition of biotin- or fluorochrome-tyramides. Hum. Mol. Genet. 4:529-534. Speel, E.J.M., Jansen, M.P.H.M., Ramaekers, F.C.S., and Hopman, A.H.N. 1994. A novel triple-color detection procedure for brightfield microscopy, combining in situ hybridization with immunocytochemistry. J. Histochem. Cytochem. 42:1299-1307. Van Gijlswijk, R.P.M., Wiegant, J., Raap, A.K., and Tanke, H.J. 1996a. Improved localization of fluorescent tyramides for fluorescence in situ hybridization using dextran sulphate and polyvinyl alcohol. J. Histochem. Cytochem. 44:389392. Van Gijlswijk, R.P.M., Wiegant, J., Vervenne, R., Lasan, R., Tanke, H.J., and Raap, A.K. 1996b. Horseradish peroxidase-labeled oligonucleotides and fluorescent tyramides for rapid detection of chromosome-specific repeat sequences. Cytogenet. Cell Genet. 75:258-262. Wiegant, J., Ried, T., Nederlof, P.M., van der Ploeg, M., Tanke, H.J., and Raap, A.K. 1991. In situ hybridization with fluoresceinated DNA. Nucl. Acids Res. 19:3237-3241.
Contributed by J. Wiegant Leiden University Leiden, The Netherlands
8.4.18 Supplement 4
Current Protocols in Cytometry
Processing and Staining of Cell and Tissue Material for Interphase Cytogenetics
UNIT 8.5
In situ hybridization (ISH) using probes for specific regions of the genome allows the targeted detection of numerical and structural chromosome aberrations in the interphase nucleus, and is generally referred to as interphase cytogenetics (Cremer et al., 1986). Genetic alterations can be deduced from counting of ISH signals (for the detection of chromosomal aneusomies and imbalances), determination of colocalization or color overlap (translocations), and signal intensity (amplifications; Tkachuk et al., 1991; Bentz et al., 1994a). Acquisition of these data is influenced by a number of factors, including the quality of material under study, the hybridization efficiency, properties of the cytochemical detection reagents, instrumentation features, and interobserver variations. Interphase cytogenetics is frequently applied to several types of malignancies in a variety of cell and tissue preparations (e.g., single cells isolated from tumors), collected either as paraffin-embedded material or as frozen tissue samples. In all cases, the general outline of the ISH procedure is as follows. (1) Select probes for interphase cytogenetics based on information concerning tumor-specific genetic aberrations. (2) Label probe. (3) Fix biological material, prepare slide, and pretreat tissue. (4) Hybridize modified probe with denatured target DNA, and follow with immunocytochemical detection. (5) Count and interpret ISH signals and evaluate the results. In this unit, protocols for the fixation of biological samples, for slide preparation, and for pretreatment of biological material (including fresh and formalin-fixed, paraffin-embedded tissue) are presented. Properly applied, the above general procedure, in which each individual step influences the final ISH result, leads to an operational protocol to obtain reliable ISH results within a limited number of experimental trials. STRATEGIC PLANNING Probe Selection The DNA probes to be used for an interphase cytogenetic study are selected on the basis of the chromosomal localization of their target and of the size of the target sequence. UNIT 8.3 describes different types of probes and nonradioactive labeling protocols in detail. Tandemly repeated DNA sequences are mostly present in the centromeric and telomeric regions of chromosomes (Willard and Waye, 1987). The sequences targeted by probes to these chromosomal regions are typically satellite sequences, which are usually repeated several hundred- to several thousand-fold. Repeat-sequence probes have been developed for most human chromosomes and are now commercially available. The ISH signal intensity with these probes is usually high and the hybridization signal is discretely localized within metaphase preparations and interphase nuclei. The probes can be routinely used to detect aneusomies within interphase cells (van Dekken et al., 1990; Poddighe et al., 1992). The stringency of hybridization affects probe specificity, while heteromorphic variability can hamper the evaluation (see Basic Protocol 9). Repeat-sequence probes can be visualized using fluorescence as well as bright-field ISH detection systems, both of which allow single-, double-, and triple-target ISH. Probes targeted to a specific locus are useful for the analysis of single-copy gene sequences varying in size from 2 kb to a few hundred kb. Inserts in plasmids, phages, and large-insert vectors (e.g., YACs, BACs, and P1s) are used for these targets. The large-insert probes can be used for analysis in interphase nuclei, because a sufficiently high signal Contributed by A.H.N. Hopman and F.C.S. Ramaekers Current Protocols in Cytometry (1998) 8.5.1-8.5.22 Copyright © 1998 by John Wiley & Sons, Inc.
Molecular Cytogenetics
8.5.1 Supplement 5
intensity can be obtained. However, the applicability of these probes is more complex than that of repeat-sequence probes, requiring fine tuning of the total sequence of reaction steps. Locus-specific probes are used for the detection of genomic deletions, amplifications, and translocations. In single-target ISH, both fluorescence and bright-field ISH detection systems can be applied. In multiple-target ISH, however, these locus-specific probes have so far been routinely detected only with fluorescence ISH procedures. Pretreatment Procedures Fine tuning of the pretreatment steps (i.e, those prior to denaturation, hybridization, and detection) is the key to obtaining a routinely applicable processing protocol and optimal results. A multitude of experimental variables have been recognized to influence the ISH steps, resulting in standardized operational ISH protocols that are applicable to cytology specimens, disaggregated nuclei from frozen and formalin-fixed archival material, as well as tissue sections from frozen and paraffin-embedded tumor material. Several standard fixatives (e.g., 3:1 methanol/acetic acid used for karyotyping, 70% ethanol for flow cytometrical analyses, methanol/acetone for cytological analyses, and 4% formaldehyde in PBS for histological analyses) are fully compatible with the ISH procedure. Depending on the degree of cross-linking of cytoplasmic and nuclear proteins, a proteolytic pretreatment step is needed to permeabilize the cell, so that macromolecular reagents such as DNA probes and antibodies can reach their targets. Digestion is also important because it will reduce background autofluorescence, which is sometimes seen with overfixation. Autofluorescence complicates the evaluation of fluorescence ISH signals, necessitating the use of bright-field ISH. With digestion and mild formaldehyde fixation, similar results for fluorescence and bright-field ISH can be obtained. However, as a result of this proteolysis step, isolated cells or parts of tissue sections may be lost during the subsequent denaturation and hybridization steps, especially when noncoated glass slides (see Support Protocol) are used. Removal of cytoplasmic and nuclear proteins from cell suspensions can be achieved by methanol/acetic acid fixation prior to spotting of the cells onto pretreated glass slides, and followed by a mild permeabilization step with a detergent or by enzymatic digestion. This approach is incorporated in nearly all ISH procedures utilizing samples such as cultured cells, bone marrow, and chorionic villi for molecular cytogenetic analysis. For large tissue fragments, individual cells need to be isolated. This can be done relatively easily by either mechanical or enzymatic disaggregation of the tissue material. The mechanically disaggregated cells (see Basic Protocol 1) are deposited onto the slide, and are then further pretreated to remove the cytoplasm. For enzymatic disaggregation of large tissue fragments or small cell samples (e.g., a single chorion villous, a small tissue biopsy, or a tissue section), the cytoplasm is removed during the enzymatic isolation procedure. Isolated nuclei are then deposited onto the slide (see Basic Protocol 2). For the efficient release of nuclei from formalin-fixed, paraffin-embedded material, a chemical soaking step is also introduced (see Basic Protocol 3). Specific variations in this general procedure are presented for sections of fresh and frozen tissue (see Basic Protocol 4), as well as for formalin-fixed, paraffin-embedded tissue (see Basic Protocol 5; see Alternate Protocol). In all protocols presented here, the total pretreatment generally results in preparations that contain nuclei from which the cytoplasm has been removed or strongly permeabilized.
Cell and Tissue Processing for Interphase Cytogenetics
8.5.2 Supplement 5
Current Protocols in Cytometry
PREPARATION OF SINGLE CELLS AND NUCLEI FOR ISH Isolation of Cells by Mechanical Disaggregation of Fresh or Frozen Tissue Samples In this protocol, a cell suspension is made from fresh tissue by mechanical disaggregation and is deposited on slides. Permeabilization with pepsin/HCl is performed after spotting, followed by formaldehyde fixation and dehydration. The slides are then ready for denaturation and hybridization protocols.
BASIC PROTOCOL 1
Materials 1-mm3 (or larger) fresh tissue specimen PBS (APPENDIX 2A) Culture medium (optional) 50 to 400 µg/ml pepsin in 0.01 M HCl (from porcine stomach mucosa, 2500 to 3500 U/mg protein, Sigma; store dilution up to several weeks at 4°C) 1% formaldehyde in PBS (freshly prepared from a 37% formaldehyde stock) 70%, 90%, and 100% ethanol 90-mm petri dishes 100-µm nylon filter Poly-L-lysine- or organosilane-coated slides (see Support Protocol) Shandon Cytospin cytocentrifuge (Life Sciences) or equivalent Coplin jars Additional reagents and equipment for cell counting (APPENDIX 3A) 1. Place a fresh 1-mm3 (or larger) tissue specimen in a 90-mm petri dish containing 1 ml PBS or culture medium. Scrape and cut with a razor blade for a few minutes to disaggregate the cells. Filter the suspension through a 100-µm nylon filter to remove large aggregates. The filtered cell suspension contains both intact cells and isolated nuclei from ruptured cells. At this stage, the nuclear suspension can be fixed in 70% ethanol (−20°C) or 3:1 methanol/acetic acid and stored (APPENDIX 3B) for up to several years at −20° to −30°C without reduction of ISH reactivity. When the number of nuclei is small, fixation in ethanol is preferable.
2. Count cells and nuclei in freshly prepared or ethanol-fixed cell suspension. Prepare a suspension of ∼5 × 105 cells and nuclei/ml PBS. 3. Spot 5- to 10-µl aliquots of suspension onto coated slides either by dropping several microliters directly onto the slides or by cytocentrifuging 200 µl PBS containing 5 to 10 µl suspension in a Shandon Cytospin cytocentrifuge 5 min at 75 to 100 × g, room temperature (also see UNITS 5.2 & 8.2). Concentration of cells and nuclei will vary with size of specimen, and aliquot volume will vary with concentration and with the cell density required on the slide. To obtain maximum nuclear recovery, do not cytocentrifuge small (<50 ìl) or large (>200 ìl) volumes. In general, cytocentrifugation will lead to a more equal distribution of cells.
4. Air dry 15 min at room temperature. If necessary, heat slides 30 to 60 min at 80°C to obtain adequate cell adhesion.
5. Incubate slides 20 min at 37°C in 100 ml of 50 to 400 µg/ml pepsin in a Coplin jar. Optimal conditions should be determined empirically. In general, 100 ìg/ml is a standard pepsin concentration. For mild permeabilization, slides can be immersed in 0.1 M HCl containing 0.05% Triton X-100 or Tween 20 for 15 min.
Molecular Cytogenetics
8.5.3 Current Protocols in Cytometry
Supplement 5
6. Dip slides 5 times in water and 5 times in PBS. 7. Postfix cells in 100 ml of 1% formaldehyde in a Coplin jar, 10 min at room temperature. 8. Dip slides 5 times in PBS. 9. Dehydrate by washing 1 min each in 100 ml of 70%, 90%, and 100% ethanol in a Coplin jar. Air dry. After fixation slides can also be rinsed in 2× SSC (APPENDIX 2A) prior to application of the probe under a coverslip. If necessary, heat slides for 30 to 60 min at 80°C to obtain adequate cell adhesion. Slides are now ready for denaturation (see Basic Protocol 6) and hybridization (UNIT 8.3). BASIC PROTOCOL 2
Isolation of Nuclei by Enzymatic Treatment of Cell and Tissue Samples This protocol combines enzymatic isolation and permeabilization of nuclei into one step that removes cytoplasm and digests part of the nuclear protein. This approach provides efficient isolation of nuclei from small tissue samples. It can be applied to a variety of fresh or frozen tissues. Enzymatically isolated nuclei can be directly spotted or cytocentrifuged onto slides, at which point they are dehydrated and fixed. The pH at which the nuclei are cytocentrifuged is critical because variations in pH strongly influence the compactness of the nucleus. During the pepsin digestion step (at pH 2.0), part of the nonhistone proteins are removed and the phosphate groups in the DNA are protonated, resulting in condensed nuclei. At pH 7 the nucleus is swollen to three times its normal size. For this reason, the nuclei in pepsin buffer are swollen by adding 50× TAE prior to cytocentrifugation. Following this protocol the morphology of the nuclei will resemble that of methanol/acetic acid–fixed cells with respect to size and to efficiency of DNA staining. Materials Tissue sample: fresh solid tissue block (≤1 mm3; mechanically disaggregate larger blocks, as necessary), tissue fragments (e.g., 30- to 50-µm-thick frozen sections), or cytological material (e.g., brush material, effusions, chorion villous, or amniotic fluid) 100 µg/ml pepsin in 0.01 M HCl (from porcine stomach mucosa, 2500 to 3500 U/mg protein, Sigma; store dilution up to several weeks at 4°C) 0.01 M HCl 50× TAE (APPENDIX 2A) 70%, 90%, and 100% ethanol 1% formaldehyde in PBS (freshly prepared from a 37% formaldehyde stock) 21-G needle Organosilane-coated slides (see Support Protocol) Shandon Cytospin cytocentrifuge (Life Sciences) or equivalent Coplin jars 1. Digest tissue sample in 1 ml of 100 µg/ml pepsin in 0.01 M HCl, 10 to 20 min at 37°C. To obtain single nuclei, mildly resuspend the tissue material by passing through a 21-G needle.
Cell and Tissue Processing for Interphase Cytogenetics
At this stage, the nuclear suspension can be fixed in 70% ethanol or 3:1 methanol/acetic acid and stored (APPENDIX 3B) up to several years at −20°C without reduction of ISH activity. When the number of nuclei is small, fixation in ethanol is preferable.
8.5.4 Supplement 5
Current Protocols in Cytometry
2. Make a test preparation by spotting 10 µl suspension on an organosilane-coated slide. Count nuclear density under a microscope. This step ensures that there will be sufficient nuclei on the slide while avoiding nuclear overlap. The optimal number of nuclei for cytocentrifuge preparations is 1 × 103.
3. Adjust an aliquot of the suspension containing 1 × 103 nuclei to a final volume of 100 µl with 0.01 M HCl. Add 2 µl of 50× TAE and mix. 4. Cytocentrifuge 100 µl suspension 5 min at 100 × g, room temperature, in a Shandon Cytospin cytocentrifuge (also see UNITS 5.2 & 8.2). To obtain maximum nuclear recovery, do not cytocentrifuge small (<50 ìl) or large (>200 ìl) volumes. Alternatively, cells can be spotted directly onto the slide. In general cytocentrifugation will lead to a more equal distribution of cells.
5. Air dry slides 10 min at room temperature. Dehydrate by washing 1 min each in 100 ml of 70%, 90%, and 100% ethanol in a Coplin jar. 6. Postfix preparations 5 min at room temperature in 100 ml of 1% formaldehyde. Postfixation can be omitted when slides are heated at 80°C for 30 min. Slides are now ready for denaturation (see Basic Protocol 6) and hybridization (UNIT 8.3).
Isolation of Nuclei by Enzymatic Treatment of Formalin-Fixed and Paraffin-Embedded Tissue
BASIC PROTOCOL 3
In formalin-fixed and paraffin-embedded tissue material, the DNA is trapped in a relatively strong matrix of cross-linked protein. If the tissue material is fixed at neutral pH for not longer than 24 hr, the DNA and proteins are internally cross-linked to a limited extent. This protocol describes such a cross-linking procedure. Incubation in formic acid/H2O2 removes protein prior to permeabilization with pepsin, improving nuclear isolation and permeabilization. Following permeabilization, formalin-fixed nuclei are brought to pH 9, resulting in a three-fold increase in nuclear size. Compare this to the pH-driven swelling of fresh nuclei, which uses a change from acidic to neutral pH (see Basic Protocol 2). For formalin-fixed nuclei, a higher pH is needed for relaxation as a result of cross-linking of histone proteins during formalin fixation. The morphology of the nuclei isolated by this protocol will resemble that of methanol/acetic acid–fixed cells with respect to size and efficiency of DNA staining. Materials 15- to 50-µm-thick, formalin-fixed, paraffin-embedded tissue section 100% xylene 70%, 90%, and 100% ethanol 100% methanol 85% formic acid/0.3% H2O2 (prepare fresh) 0.01 M HCl 4 mg/ml pepsin in 0.2 M HCl (store up to several weeks at 4°C) 0.2 M NaOH 21-G needle Organosilane-coated slides (see Support Protocol) Shandon Cytospin cytocentrifuge (Life Sciences) or equivalent Coplin jars Molecular Cytogenetics
8.5.5 Current Protocols in Cytometry
Supplement 5
1. Place a 15- to 50-µm-thick formalin-fixed, paraffin-embedded tissue section in a microcentrifuge tube. 2. Add 1 ml of 100% xylene and incubate 10 min at room temperature. Microcentrifuge 10 sec at maximum speed and remove xylene. Repeat for a total of three 10-min washes in xylene, two 10-min washes in 100% ethanol, and two 5-min washes in 100% methanol. Air dry the deparaffinized tissue section. 3. Add 1 ml of 85% formic acid/0.3% H2O2 and incubate 20 min at room temperature. Wash twice for 5 min each with 1 ml of 0.01 M HCl, microcentrifuging between washes. 4. Digest the section 20 min with 1 ml of 4 mg/ml pepsin in 0.2 M HCl. To obtain single nuclei, mildly resuspend the tissue several times through a 21-G needle. Optionally, the suspension can be filtered through a 100-ìm nylon filter to remove large aggregates. In general, no filtering is needed.
5. Pellet the nuclei by microcentrifuging 10 sec at maximum speed, and remove the supernatant, leaving ∼50 µl fluid on the nuclear pellet. Add 1.1 vol of 0.2 M NaOH to raise the pH of the suspension. 6. Make a test preparation by spotting 10 µl of the suspension on an organosilane-coated slide. Count nuclear density under a microscope. This step ensures that there will be sufficient nuclei on the slide while avoiding nuclear overlap. The optimal number of nuclei for cytocentrifuge preparations is 1 × 103.
7. Adjust an aliquot of the suspension containing 1 × 103 nuclei to a final volume of 100 µl with distilled water. 8. Cytocentrifuge 100 µl suspension 5 min at 100 × g, room temperature, in a Shandon Cytospin cytocentrifuge (UNITS 5.2 & 8.2). To obtain maximum nuclear recovery, do not cytospin small (<50 ìl) or large (>200 ìl) volumes. Alternatively, cells can be spotted directly onto the slides. In general cytocentrifugation will lead to a more equal distribution of cells.
9. Air dry slides 10 min at room temperature. Dehydrate by washing 1 min each in 100 ml of 70%, 90%, and 100% ethanol in a Coplin jar. Slides are now ready for denaturation (see Basic Protocol 6) and hybridization (UNIT 8.3).
PREPARATION OF TISSUE SECTIONS FOR ISH BASIC PROTOCOL 4
Cell and Tissue Processing for Interphase Cytogenetics
Frozen Tissue Sections In this protocol, frozen tissue sections are mildly fixed in formaldehyde by using a short fixation time (e.g., <15 min), as a means of reducing autofluorescence (see Strategic Planning). Proteolytic digestion is tuned so that the cells do not lose nuclear morphology through overdigestion. Compared to Basic Protocol 2, this protocol includes an extra acid dehydration step to avoid relaxation of the nuclei in the tissue section, which results in a loss of nuclear morphology. An optional fixation step prior to digestion with pepsin can also be performed to improve morphology and increase attachment of the cells to the slide.
8.5.6 Supplement 5
Current Protocols in Cytometry
Materials 4- to 6-µm-thick frozen tissue section 1% formaldehyde in PBS (freshly prepared from a 37% formaldehyde stock) 0.5% Tween 20/PBS (freshly prepared) 50 to 400 µg/ml pepsin in 0.01 M HCl (from porcine stomach mucosa, 2500 to 3500 U/mg protein, Sigma; store dilution up to several weeks at 4°C) 0.01 M HCl in 70%, 90%, and 100% ethanol (freshly prepared) PBS (APPENDIX 2A) 70%, 90%, and 100% ethanol Coplin jars Organosilane-coated slides (see Support Protocol) 1. Stretch a 4- to 6-µm-thick frozen tissue section onto an organosilane-coated glass slide. 2. Air dry overnight at room temperature. 3. Fix in 100 ml of 1% formaldehyde in a Coplin jar, 10 min at room temperature, to preserve morphology during hybridization. 4. Wash twice for 5 min each with 100 ml of 0.5 % Tween 20/PBS in a Coplin jar. 5. Incubate 10 min at 37°C in 100 ml of 50 to 400 µg/ml pepsin in 0.01 M HCl. Optimal pepsin concentration should be determined empirically. In general, 100 ìg/ml is a standard pepsin concentration. Alternatively, for a mild permeabilization, slides can be immersed in 0.1 M HCl containing 0.05% Triton X-100 or Tween 20 for 15 min.
6. Acid dehydrate by washing 3 min each with 100 ml of 0.01 M HCl in 70%, 90%, and 100% ethanol. Acid dehydration keeps the nuclei condensed. If the nuclei are dehydrated at this stage in a nonacidified ethanol dehydration series, the nuclei may relax, resulting in a loss of morphology. This swelling is characteristic of nuclei which are deproteinated in pepsin at pH 2.0.
7. Postfix cells in 100 ml of 1% formaldehyde, 10 min at room temperature. 8. Rinse slides by dipping 5 times in PBS and 5 times in H2O. 9. Dehydrate by washing 1 min each in 100 ml of 70%, 90%, and 100% ethanol. Air dry. If necessary, heat slides for 30 to 60 min at 80°C to obtain adequate cell adhesion. Slides are now ready for denaturation (see Basic Protocol 6) and hybridization (UNIT 8.3).
Formalin-Fixed and Paraffin-Embedded Tissue Sections A procedure is described here for the removal of proteins during pretreatment steps, as a means of reducing autofluorescence (see Strategic Planning). Although the reduction in autofluorescence is sufficient to enable fluorescence ISH detection, the application of bright-field ISH is still recommended. The chemical soaking step in formic acid/H2O2 (see Basic Protocol 3) is followed by incubation in sodium thiocyanate at 80°C. Pretreatment of the sections in this way strongly improves the effect of the subsequent pepsin digestion step, resulting in reproducible and efficient ISH results. The protocol has been tested on routinely processed malignancies from bladder, cervix, renal cell carcinoma, head and neck carcinoma, and lymphoma. The proteolytic digestion time must be
BASIC PROTOCOL 5
Molecular Cytogenetics
8.5.7 Current Protocols in Cytometry
Supplement 5
optimized for each tissue block, with digestion periods ranging from 5 to 20 min. Since different cell types (e.g., tumor cells, inflammatory cells, stromal cells) may require different pretreatment conditions, optimization is directed towards the cell type of interest. Materials 4- to 6-µm-thick formalin-fixed, paraffin-embedded tissue sections 100% xylene 100% methanol 85% formic acid/0.3% H2O2 (prepare fresh) 0.01 M HCl in 70%, 90%, and 100% ethanol 1 M sodium thiocyanate (NaSCN), prewarmed to 80°C (prepare fresh) 4 mg/ml pepsin in 0.02 M HCl (store up to several weeks at 4°C) 0.01 M HCl Organosilane-coated slides (see Support Protocol) Incubator at 56° and 80°C Plastic Coplin jars 1. Stretch a 4- to 6-µm-thick formalin-fixed, paraffin-embedded tissue section on distilled H2O, prewarmed to 40°C. 2. Pick up the sections on organosilane-coated slides. 3. Air dry and heat the slides overnight at 56°C. 4. Deparaffinize with three 10-min washes in 100 ml of 100% xylene in a plastic Coplin jar. 5. Wash twice for 5 min each with 100 ml of 100% methanol. 6. Immerse in 100 ml of 85% formic acid/0.3% H2O2 for 20 min at room temperature. H2O2 introduces nicks in the DNA, leading to DNA fragmentation. This can in turn cause reduced thermal stability of the formed hybrids. If this is the case, the formic acid/H2O2 incubation time should be reduced.
7. Acid dehydrate by washing 3 min each with 100 ml of 0.01 M HCl in 70%, 90%, and 100% ethanol. Allow slides to air dry. 8. Soak in 100 ml prewarmed 1 M NaSCN at 80°C in a plastic Coplin jar for 10 min. 9. Rinse 5 to 10 sec in 100 ml H2O in the Coplin jar. 10. Acid dehydrate and air dry as in step 7. Acid dehydration increases adhesion and is thus done to avoid losing a part of the tissue section during hybridization.
11. Incubate 5 to 20 min at 37°C in 100 ml of 4 mg/ml pepsin in 0.02 M HCl. Ten minutes is a standard digestion time. It is possible to evaluate the degree of digestion by phase contrast microscopy before moving on to the hybridization step. If there is a loss of cell borders, the sample is overdigested. If no individual nuclei can be seen, the sample is underdigested and the incubation can be prolonged. When the nuclear image becomes gray/opaque, the incubations should be stopped.
12. Rinse slides by dipping three times in 100 ml of 0.01 M HCl. Cell and Tissue Processing for Interphase Cytogenetics
13. Acid dehydrate and air dry as in step 7. If necessary, heat slides 30 min at 80°C to obtain adequate adhesion of the sections. Slides are now ready for denaturation (see Basic Protocol 6) and hybridization (UNIT 8.3).
8.5.8 Supplement 5
Current Protocols in Cytometry
Formalin-Fixed and Paraffin-Embedded Tumor Sections This alternate protocol has been optimized for tumor sections by making use of a pretreatment step of hot sodium bisulfite followed by proteinase K treatment.
ALTERNATE PROTOCOL
Additional Materials (also see Basic Protocol 5) Tumor sections 30% sodium bisulfite (Baker Chemical) in 2× SSC, prewarmed to 45°C (freshly prepared) 2× SSC (APPENDIX 2A), room temperature and prewarmed to 45°C 20 mg/ml proteinase K in 2× SSC (store aliquots at −20°C) 70%, 90%, and 100% ethanol Incubator at 45°C 1. Deparaffinize and pretreat sections (see Basic Protocol 5, steps 1 through 7). 2. Place up to four slides in a Coplin jar containing 100 ml prewarmed 30% sodium bisulfite and incubate 10 to 15 min at 45°C. 3. Wash the slides 5 to 10 sec in 100 ml of 2× SSC, room temperature. 4. Immediately add 500 µl of 20 mg/ml proteinase K to 40 ml prewarmed 2× SSC in a Coplin jar. Incubate slides 15 to 30 min, 45°C, depending on the efficiency of the digestion step, the tissue type, and fixation. Begin with a 20-min digestion time. It is possible to evaluate the degree of digestion before moving on to the hybridization procedure. Counterstain with 2.5 ìg/ml propidium iodide in 2× SSC, and evaluate the sections by phase contrast microscopy for overdigestion (loss of cell borders) or underdigestion (persistent autofluorescence). In the latter case, perform an additional pretreatment with bisulfite and proteinase K (i.e, repeat steps 2 to 4). The propidium iodide counterstaining, which persists after washing, does not interfere with hybridization.
5. Rinse slides 5 to 10 sec in 100 ml of room temperature 2× SSC and dehydrate by washing 1 min each in 100 µl 70%, 90%, and 100% ethanol. Slides are now ready for denaturation (see Basic Protocol 4) and hybridization (UNIT 8.3).
Preparation of Coated Slides As a result of the proteolysis step, isolated cells or parts of tissue sections may be lost during the subsequent denaturation and hybridization steps, especially when noncoated glass slides are used. This can be avoided by coating the slides using either poly-L-lysine or aminoalkyl silane (organosilane). For proper adhesion of biological material, slides are electrostatically charged to attract single cells or frozen, formalin-fixed tissue sections. Coated slides can be prepared by dipping in poly-L-lysine or by a chemical reaction resulting in an electrostatic activation of the glass surface. Superfrost plus slides (Menzel-Glaser) are precoated and ready for use. For practical reasons these organosilane slides are used in all ISH protocols. Materials 10% Extran MA01 (EM Science) 1 mg/ml poly-L-lysine hydrobromide (MW >150,000; Sigma) in Milli-Q-purified H2O 2% 3-aminopropyltriethoxysilane (Sigma) in anhydrous acetone Acetone 0.02% NaN3
SUPPORT PROTOCOL
Molecular Cytogenetics
8.5.9 Current Protocols in Cytometry
Supplement 5
Glass slides Incubator or water bath at 37°, 60°, and 80°C 1. Clean glass slides 2 hr in 10% Extran at 60°C. Rinse with 60°C tap water and cold distilled water, and dry at 80°C. 2a. For poly-L-lysine coating: Dip slides ∼20 sec in 1 mg/ml poly-L-lysine at room temperature. Air dry overnight at 37°C, and store up to several weeks at 4°C in a closed box. 2b. For organosilane-coated slides: Incubate slides overnight in 2% 3-aminopropyltriethoxysilane. Rinse once in acetone, and twice in distilled water. Store up to several weeks in 0.02% NaN3 at 4°C. Immediately before use, rinse the slides in distilled water and air dry. For a covalent coupling of cellular material to the glass slides, the slides can be chemically activated by immersing them 15 to 60 min in 2.5% glutardialdehyde solution in PBS at room temperature in a Coplin jar. Rinse the slides with PBS and distilled water and air dry. Glutaraldehyde-activated slides can be kept for a few days at 4°C.
HYBRIDIZATION AND DETECTION BASIC PROTOCOL 6
Probe and Target Denaturation and Hybridization The composition of hybridization buffers for repeat-sequence probes and locus-specific probes, the hybridization conditions, and the posthybridization washing conditions are described in detail in UNIT 8.3. Materials Hybridization solution (UNIT 8.3) DNA probe (UNIT 8.3) Cell or tissue sample (prepared by one of the methods described above; see Basic Protocols 1 to 5) 18 × 18–mm coverslip Additional reagents and equipment for hybridization and for posthybridization washes (UNIT 8.3) 1. Apply 5 to 10 µl hybridization solution containing 1 to 3 ng/liter DNA probe to the prepared cell or tissue sample and cover with an 18 × 18–mm coverslip. Optional: Seal the coverslip with rubber cement for overnight (or longer) hybridization periods to avoid evaporation of hybridization buffer under the coverslip.
2. Simultaneously denature target and probe DNA using the following conditions: For cell samples (see Basic Protocols 1 and 2) and frozen tissue sections (see Basic Protocol 4): 75°C (70° to 80°C) for ∼3 min (2 to 4 min). For fixed, embedded samples: 80°C (75° to 80°C) for 4 min (see Basic Protocol 3; see Alternate Protocol) or 8 min (see Basic Protocol 5).
Cell and Tissue Processing for Interphase Cytogenetics
Probe and target DNA can also be denatured separately. Target DNA is denatured by immersing the slides in 70% formamide/2× SSC (APPENDIX 2A) for 3 min, and then dehydrating in an ethanol series at 4°C. The DNA probe is denatured 4 min at 80°C in a waterbath and chilled on ice. The denatured probe is then applied to the sample under a coverslip.
3. Hybridize overnight at 37°C.
8.5.10 Supplement 5
Current Protocols in Cytometry
4. Remove coverslip and perform posthybridization washings (UNIT 8.3). If a formamide-containing hybridization buffer evaporates under the coverslip (i.e., there are air bubbles), the concentration of formamide should be increased. This reduces the thermal stability of the probe, resulting in a denaturation of the ISH hybrid under stringent washing conditions. Under these circumstances, the washing temperature should be reduced by 5°C. The slides are now ready for immunocytochemical detection (see Basic Protocols 7 and 8; UNIT 8.4).
Double-Target Fluorescence ISH A universal protocol for a double-target fluorescence ISH using biotin- and digoxigeninlabeled repeat-sequence probes is described. This protocol is applicable to all previously described procedures for sample preparation. The targets are detected using rhodamine (TRITC) and fluorescein (FITC).The DNA is counterstained with DAPI (blue). For a more detailed discussion of fluorescence detection methods for ISH, see UNIT 8.4.
BASIC PROTOCOL 7
Materials ISH-labeled sample with one biotinylated and one digoxigenin-labeled probe (UNIT 8.3) 4× SSC (APPENDIX 2A)/5% nonfat dry milk 1:100 FITC-conjugated avidin (Vector Labs) in 4× SSC/5% nonfat dry milk 4× SSC (APPENDIX 2A)/0.05% Tween 20 1:100 biotinylated goat anti-avidin (Vector Labs) in 4× SSC/5% nonfat dry milk PBS (APPENDIX 2A)/0.05% Tween 20 1:2000 mouse anti-digoxin (Sigma) in PBS/0.05% Tween 20/2% normal goat serum 1:100 TRITC-conjugated rabbit anti–mouse IgG (Dako) in PBS/0.05% Tween 20/2% normal goat serum PBS (APPENDIX 2A) 70%, 90%, and 100% ethanol Antifading mounting medium (see recipe) Coplin jars 24 × 50–mm coverslips NOTE: All buffers containing nonfat dry milk, normal goat serum, or antibodies should be prepared fresh. 1. Block ISH-labeled sample with 100 µl of 4× SSC/5% nonfat dry milk under a 24 × 50–mm coverslip, 10 min at 37°C. 2. Apply 100 µl of 1:100 FITC-conjugated avidin, coverslip, and incubate 20 min at 37°C. 3. Wash 5 min with 100 ml of 4× SSC/0.05% Tween 20 in a Coplin jar at room temperature with gentle agitation. 4. Apply 100 µl of 1:100 biotinylated goat anti-avidin under a 24 × 50–mm coverslip and incubate 20 min at 37°C. 5. Wash 1 min with 100 ml of 4× SSC/0.05% Tween 20 at room temperature with gentle agitation. 6. Repeat step 2. 7. Wash 1 min with 100 ml of 4× SSC/0.05% Tween 20 at room temperature with gentle agitation. Repeat with 100 ml of PBS/0.05% Tween 20.
Molecular Cytogenetics
8.5.11 Current Protocols in Cytometry
Supplement 5
8. Apply 100 µl of 1:2000 mouse anti-digoxin under a 24 × 50–mm coverslip and incubate 30 min at 37°C. 9. Wash twice for 1 min each with 100 ml PBS/0.05% Tween 20 at room temperature with gentle agitation. 10. Apply 100 µl of 1:100 TRITC-conjugated rabbit anti–mouse IgG and incubate 30 min at 37°C. 11. Wash twice for 5 min each with 100 ml PBS at room temperature with gentle agitation. 12. Dehydrate by washing 1 min each in 100 ml of 70%, 90%, and 100% ethanol at room temperature. Air dry. 13. Mount in antifading mounting medium. Store preparations up to several months at 4°C in the dark. BASIC PROTOCOL 8
Double-Target Bright-Field ISH A universal protocol for a double-target bright-field ISH using biotin- and digoxigeninlabeled repeat-sequence probes is described. This protocol is applicable to all previously described procedures for sample preparation. The targets are detected using alkaline phosphatase (red) and horseradish peroxidase (brown). The DNA is counterstained with hematoxylin (blue). For a more detailed discussion of bright-field detection methods for ISH, see UNIT 8.4. Materials ISH-labeled sample with one biotinylated and one digoxigenin-labeled probe (UNIT 8.3) PBS (APPENDIX 2A)/0.05% Tween 20/2% normal goat serum 1:2000 mouse anti-digoxin (Sigma) in PBS/0.05% Tween 20/2% normal goat serum PBS (APPENDIX 2A)/0.05% Tween 20 1:25 alkaline phosphatase (AP)–conjugated goat anti–mouse IgG (Dako) in PBS/0.05% Tween 20/2% normal goat serum PBS (APPENDIX 2A) New Fuchsin reagent (see recipe) 4× SSC (APPENDIX 2A)/0.05% Tween 20 1:50 horseradish peroxidase–conjugated avidin (HRP-avidin; Dako) in 4× SSC (APPENDIX 2A)/5% nonfat dry milk 3,3′-diaminobenzidine (DAB) solution (see recipe) 20% hematoxylin (Sigma) in H2O 50%, 70%, 96%, and 100% ethanol 100% xylene Entellan (organic-based mounting medium; EM Science) 24 × 50–mm coverslips Coplin jars NOTE: All buffers containing nonfat dry milk, normal goat serum, or antibodies should be prepared fresh.
Cell and Tissue Processing for Interphase Cytogenetics
1. Block ISH-labeled sample with 100 µl PBS/0.05% Tween 20/2% normal goat serum under a 24 × 50–mm coverslip 10 min at 37°C. 2. Apply 50 µl of 1:2000 mouse anti-digoxin under a 24 × 50–mm coverslip, and incubate 30 min at 37°C.
8.5.12 Supplement 5
Current Protocols in Cytometry
3. Wash twice in a Coplin jar for 1 min each with 100 ml PBS/0.05% Tween 20 at room temperature with gentle agitation. 4. Apply 50 µl of 1:25 AP-conjugated goat anti–mouse IgG under a 24 × 50–mm coverslip, and incubate 30 min at 37°C. 5. Repeat step 3. 6. Wash 1 min with 100 ml PBS at room temperature with gentle agitation. 7. Apply 100 µl New Fuchsin reagent to a slide under a 24 × 50–mm coverslip, and incubate 5 to 15 min at 37°C. After 5 min, evaluate the intensity of the red ISH signals by bright-field microscopy. If the intensity is too low, prolong the enzymatic reaction.
8. Wash 1 min with 100 ml of 4× SSC/0.05% Tween 20 at room temperature with gentle agitation. 9. Apply 50 µl of 1:50 HRP-avidin under a 24 × 50–mm coverslip, and incubate 20 min at 37°C. 10. Wash twice for 5 min each with 100 ml of 4× SSC/0.05% Tween 20 at room temperature with gentle agitation. 11. Wash 1 min with 100 ml PBS at room temperature with gentle agitation. 12. Apply 100 µl DAB solution to a slide, under a 24 × 50–mm coverslip, and incubate 5 to 15 min at 37°C. After 5 min, evaluate the intensity of the brown ISH signals by bright-field microscopy. If the intensity is too low, prolong the enzymatic reaction. CAUTION: DAB is toxic. Handle with caution, wear protective clothing, and clean instruments thoroughly with bleach.
13. Wash three times for 5 min each with 100 ml PBS at room temperature with gentle agitation. 14. Counterstain 10 sec with 100 ml of 20% hematoxylin in a Coplin jar. Wash 5 min in running tap water. To judge the staining intensity, apply a coverslip and evaluate by bright-field microscopy. If the intensity is too low, prolong the staining or increase the hematoxylin concentration.
15. Dehydrate by dipping slides once each in 100 ml of 50%, 70%, and 96% ethanol, twice in 100% ethanol, and once in 100% xylene. Embed in Entellan. Evaluation and Interpretation of Fluorescence and Bright-Field ISH Signals A specific genetic alteration is characterized based on the number of ISH signals detected, determination of color composition and signal intensity of the signals, or analysis of the relative positions of ISH signals within the nucleus. A number of basic rules exist in fluorescence as well as bright-field ISH procedures for the classification of ISH signals in an interphase cell. The following criteria are applied for a proper evaluation with minimal intra- and interobserver variation (Hopman et al., 1988). (1) Overlapping nuclei are not counted. Processing of cell samples or tissue sections influences the frequency of nuclear overlap (see Commentary). (2) Nuclei are incorporated in the evaluation only when the ISH signals have uniform fluorescence intensity or uniform size of enzymatically produced
BASIC PROTOCOL 9
Molecular Cytogenetics
8.5.13 Current Protocols in Cytometry
Supplement 5
Table 8.5.1
Troubleshooting for Single-Cell Sample Preparations
Observation
Consequence
Possible solution
High percentage of cells without ISH signal or high autofluorescence background covering part of nucleus
Underestimation of the chromosome/gene copy number
Increase concentration of proteolytic enzyme or increase digestion time
Loss of nuclear morphology
Under- or overestimation of the chromosome/gene copy number
Decrease concentration of proteolytic enzyme or decrease digestion time; check cellular adhesion (use poly-L-lysine-coated slides); check drying of preparations (optional heating at 80°C); improve postfixation steps
Heterogeneous intensity of ISH signals within the same cell (repeatsequence probes)
Underestimation of the chromosome/gene copy number
Hybridize normal lymphocytes to exclude polymorphism
Overlapping nuclei
Overestimation of the chromosome/gene copy number
Prepare a better single-cell suspension by passing through a 21G needle; digest cell suspension prior to spotting of cells onto slides; reduce cell density prior to cytocentrifugation
Minor hybridization signals
Overestimation of the chromosome/gene copy number
Hybridize under more stringent conditions (e.g., higher formamide concentration, higher hybridization temperature)
Paired occurrence of ISH signals (split spots)
Overestimation of the chromosome/gene copy number
Count ISH signals in paired arrangement as one chromosome complement
color deposits. Pretreatment efficiency and immunocytochemical detection influence the variability in signal intensity (see Commentary). (3) Minor hybridization signals (recognized by a lower intensity and/or smaller spot area) are not included in the evaluation, unless an indication for a partial deletion or polymorphism exists. Minor (false positive) hybridization signals are removed by stringent washing conditions (UNIT 8.3) or reduction of probe concentration. (4) Paired spots or split spots are counted as one spot. Paired spots are seen with locus-specific probes on chromosome arms, and are the result of replication of the target sequence during the cell cycle. Split spots, an as yet unexplained phenomenon, are commonly observed in nearly all cell samples when repeat-sequence probes are targeted in the (peri)centromeric area. With repeat-sequence probes, improper evaluation of aneusomy can have several causes, some of which result from problems in the pretreatment steps of the ISH procedure. For a summary of potential errors (e.g., under- or overestimation of the chromosome copy number) that can be recognized and avoided by paying attention to some typical indications, see Table 8.5.1. Several biological phenomena can also lead to misclassification. These include somatic pairing (false positivity for monosomy 1 and 17 in brain tissue), translocations of short-arm repeat sequences from chromosome 15 onto other D-group chromosomes (false positivity for trisomy 15), and chromosomal polymorphisms.
Cell and Tissue Processing for Interphase Cytogenetics
Evaluation of single-target fluorescence and bright-field ISH in single-cell samples Chromosomal gain (trisomy, tetrasomy, or polysomy) or loss (monosomy or nullosomy) can be detected using the same protocol. However, the percentage of cells that show one single ISH signal (suggesting monosomy) will vary within a broad range (5% to 20% or
8.5.14 Supplement 5
Current Protocols in Cytometry
even higher) even in a normal cell population. This may result from the colocalization of ISH signals, not easily recognized by visual inspection. As a consequence, a monosomy cannot be properly detected with a single-target ISH procedure. For single-target ISH on aneusomic samples, determine a threshold by performing ISH on control samples of normal cells from the same tissue origin. Process control samples in parallel with the test sample. In several control samples, count the ISH-signal copy number for 200 nuclei. Set the threshold at the mean number for the aneusomic population plus/minus three times the standard deviation. Count the chromosome ISH copy number in 200 nuclei of the test sample. If the frequency of cells with an aberration exceeds the threshold, the test sample is classified as genetically aberrant. Statistically this method for determining threshold has some drawbacks (Kibbelaar et al., 1993; Moore et al., 1996). However, because the frequency of a polysomy in normal tissue is very low, the diagnostic threshold is in most cases determined by the technical performance of the ISH. Note that if the threshold is higher than the real fraction of aberrant cells (e.g., with mixed normal and tumor cell populations), this approach will not allow the detection of a malignant (aneusomic) cell population. Evaluation of double-target fluorescence ISH in single-cell samples For the detection of deletions, amplifications, p/q arm imbalances, and translocations, specific-locus probes (i.e., large-insert vectors such as YACs, BACs, P1s) are used in ISH procedures. These aberrations can only be detected properly by double-target ISH. The criteria for scoring are based on the above-described guidelines for repeat-sequence probes. Chromosomal imbalances In 100 nuclei from the control sample, count the green and red ISH signals, representing the two chromosomal loci. For gene deletions, use a locus-specific probe to monitor the suspected deletion and a repeat-sequence probe for the internal reference chromosome signal. Calculate the ratio in spot number (mean number) for the locus-specific probe versus the repeat-sequence probe (the latter being the chromosome centromere copy number; Sauter et al., 1994). Set the threshold at the mean ratio plus/minus three times the standard deviation. Calculate the same ratio using 100 nuclei from the test sample. For aneuploid tumors, calculate the ratio after correction for the normal cells. Compare this ratio with the ratio determined in the control cells. An over- or underrepresentation (gain or loss) of one of the loci is detected when the ratio exceeds the threshold. The standard deviation depends on the efficiency of the total ISH procedure. When the ISH efficiency for one of the probes is low compared to the other, this approach will not be applicable as signals will be missed. Amplifications In 100 nuclei from the control sample, count the ISH copy number for the amplified sequence (detected with a locus-specific probe) and the signal number for the internal reference chromosome signal (generally detected with a repeat-sequence probe). Calculate the ratio between number of signals obtained for the locus-specific versus the repeat-sequence probe. Repeat this calculation for the test sample, and compare with that of control cells (Kallioniemi et al., 1992). As an example, in breast cancers with no c-erbB amplification, the signal number per cell varied between 1.7 and 3.1, while in cases with high amplifications the signal number exceeded 10.
Molecular Cytogenetics
8.5.15 Current Protocols in Cytometry
Supplement 5
Table 8.5.2
Troubleshooting for Frozen Tissue Sections
Observation
Consequence
Possible solution
Good nuclear morphology, many cells with no or only one ISH signal
Underestimation of the chromosome/gene copy number
Inefficient proteolytic digestion (improper pH, check pH of digestion buffer); inefficient denaturation (check temperature during denaturation); fundamental problem in hybridization mixture or probe modification (check modification and hybridization)
Loss of morphology, positive ISH signals
Over- or underestimation of the chromosome/gene copy number
Decrease proteolytic digestion time; increase postfixation time/concentration; reduce denaturation temperature
Poor nuclear morphology, negative ISH signals
Underestimation of the chromosome/gene copy number
Check different slide coatings; omit proteolytic digestion or use a mild permeabilization
Overlapping nuclei in tissue sections
Overestimation of the chromosome/gene copy number
Solution is difficult; combine an immunocytochemical lamin staining of the nuclear envelope with an ISH to mark the nuclear border
Translocations For detection of gene translocations, interphase cells are screened for the colocalization of ISH signals from different chromosomal regions. When ISH probe sets flank a break point, the presence of a colocalization of signals is determined (Bentz et al., 1994b). The Philadelphia chromosome is a typical example of a translocation that can be detected in interphase cells of chronic myelogenous leukemia (CML) patients. Evaluation of fluorescence and bright-field ISH signals in frozen and paraffin-embedded tissue sections Evaluation of ISH signals in tissue sections is strongly influenced by nuclear truncation. This will result in a higher percentage of cells with no or only one ISH signal (25% to 35%) as compared to single-cell ISH. No simple correction factor for nuclear truncation is available. The extent to which nuclear truncation influences the ISH signal numbers is directly correlated with nuclear size (Hopman et al., 1991). In 4- to 6-µm-thick sections of normal tissue, ∼50% of the nuclei will show the true chromosome copy number. The true chromosome copy number can be obtained by taking the maximum number of ISH signals per nucleus. A gain in copy number can be easily detected by this approach. For a summary of technical problems leading to improper interpretation of ISH signals in tissue sections, and possible solutions that lead to optimal ISH results, see Tables 8.5.2 and 8.5.3.
Cell and Tissue Processing for Interphase Cytogenetics
Evaluation of single-target ISH by comparison with internal control cells in tissue sections On the basis of single-target ISH and analysis of lymphocytes present in the tissue section, an estimation of chromosomal loss or gain in tumor cells can be made (Dhingra et al., 1994). Select areas for analysis by comparing the hybridized section with hematoxylineosin–stained serial sections for histopathological typing. Count the number of signals in a minimum of 100 tumor cell nuclei in the defined histological area. If the tissue area of interest is small (e.g., in breast cancer lobules or small ducts), count the entire microscopic field. Count in at least three different tissue areas. Divide the total number of ISH signals by the number of nuclei analyzed to obtain a chromosome index (CI). A CI between 1.3 and 1.6 is a typical value for efficient hybridization on normal disomic cells in 4-µm-thick
8.5.16 Supplement 5
Current Protocols in Cytometry
Table 8.5.3
Troubleshooting for Paraffin-Embedded Tissue Sections
Observation
Consequence
Possible solution
Good nuclear morphology; many cells with no or only one ISH signal
Underestimation of the chromosome/gene copy number
Increase proteolytic digestion time (up to 60 min, with optional refreshment of pepsin each 15 min); use different proteolytic enzymes (e.g., proteinase K instead of pepsin); use other protein pretreatment steps (see Alternate Protocol); increase denaturation temperature to 90°C
Loss of morphology, positive ISH signals
Over- or underestimation of the chromosome/gene copy number
Decrease proteolytic digestion time; use different slide coatings; reduce denaturation temperature
Poor nuclear morphology, negative ISH signals
Underestimation of the chromosome/gene copy number
Overlapping nuclei in tissue sections
Overestimation of the chromosome/gene copy number
Check different slide coatings; use denaturating agents (e.g., bisulfite) or acid treatment prior to proteolytic digestion to improve protein removal and shorten digestion time Solution is difficult; use thinner sectionsa; fix cells after proteolytic digestion in formaldehyde, or heat sections at 80°C; perform double-target ISH
aNote, however, that the use of thinner sections may itself cause underestimation of chromosome copy number, as a result of nuclear truncation.
tissue sections. Determine the CI in normal diploid lymphocytes present in the same section. Divide the CI of the tumor cells by that of the lymphocytes, to obtain a normalized chromosome index. This approach can be utilized to determine aneusomies and chromosomal imbalances. It requires a proper evaluation of the presence of normal cells. Note that the CI is not significantly increased when the proportion of aberrant cells in the section is small. In this case a more qualitative evaluation is suggested. As an alternative method, the individual ISH spot distributions in tumor cells and normal cells or the distributions for different probes can be evaluated by means of the Kolmogorov-Smirnov (K-S) test (van Dekken et al., 1992). This approach can also be used to determine chromosome losses and gains. Evaluation of double-target ISH on tissue sections An estimation of chromosomal loss or gain can also be made on the basis of a doubletarget fluorescence or bright-field ISH (Herbergs et al., 1996). Count the number of signals for both probes in the same 100 nuclei or within a tumor area of ∼2500 µm2. Calculate the ratio between the ISH spot numbers for the two different probes. In a balanced state the ratio is 1.0, while a monosomy in comparison with a disomy will give a ratio of 0.5, and a trisomy will give a ratio of 1.5. Calculate a similar ratio in control cells, and set the threshold at the mean ratio plus/minus three times the standard deviation as determined in the control cells. Compare the test ratio to the control ratio. An over- or underrepresentation (gain or loss) of one of the loci is detected when the ratio exceeds the threshold. An advantage of double-target ISH on tissue sections is that the ratio is independent of the tissue section thickness. Furthermore, a chromosomal imbalance can be detected in the presence of a nuclear overlap.
Molecular Cytogenetics
8.5.17 Current Protocols in Cytometry
Supplement 5
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Antifading mounting medium Mix 1 vol of 0.2 M Tris⋅Cl (APPENDIX 2A), pH 7.6, with 9 vol glycerol. Add 1,4-diazobicyclo-(2,2,2)-octane (DABCO; Sigma) to a final concentration of 2%. Add 40 µl of 10 mg/ml 4′,6-diamidino-2-phenylindole (DAPI) per milliliter medium (final 0.5 µg/ml). Store up to several months at 4°C. 3,3′-Diaminobenzidine (DAB) solution 1 ml 5 mg/ml 3,3′-diaminobenzidine tetrachloride (DAB; Sigma) in PBS (APPENDIX 2A; store up to several months at −20°C) 9 ml 0.1 M imidazole/PBS, pH 7.6 (store up to several months at 4°C) 10 µl 30% H2O2 Prepare immediately before use New Fuchsin (NF) reagent Dissolve 1 mg 3-hydroxy-2-naphthoic acid 2,4-dimethylanilidine phosphate (naphthol-ASMX-phosphate; Sigma) in 5 ml of 10 mM MgCl2 in 0.2 M Tris⋅Cl, pH 8.5 (APPENDIX 2A). In a separate tube, mix equal volumes of a 40 mg/ml NaNO2 solution in Milli-Q-purified H2O and a 40 mg/ml NF solution in 2 M HCl. Incubate 2 min at room temperature. Add 60 µl NaNO2/NF mixture to the naphthol-ASMXphosphate solution. Prepare immediately before use. COMMENTARY Background Information
Cell and Tissue Processing for Interphase Cytogenetics
Pretreatment procedures The most widely used preparation technique for (F)ISH is probably the hypotonic treatment of cultured cells or lymphocytes, followed by fixation in 3:1 methanol/acetic acid (UNIT 8.2). During this procedure part of the nuclear protein is removed, which leads to transparent, flat interphase nuclei that are only slightly influenced by changes in pH (see Critical Parameters and Troubleshooting). For single-cell suspensions, the standard pepsin pretreatment method (see Basic Protocols 1 and 2) has the following characteristics and advantages. (1) The clumping rate is low and there is complete removal of cytoplasmic constituents. (2) It provides high reproducibility and similar results across different types of tissue or cell material, without requiring modification of the protocol for different types of material. (3) It provides good yields and minimal loss of nuclei with easy handling. All pretreatment steps are performed in one tube to which all reagents are subsequently added, and no centrifugation steps are needed. (4) It provides efficient hybridization of cosmid probes on expanded nuclei, enabling evaluation at high magnification. (5) It is simple and reduces the
time required for pretreatment. Pretreatment time for single-cell analysis is ∼30 min, independent of the type of material used. There are two major advantages of ISH on tissue sections over ISH on isolated tumor cells. (1) Heterogeneity of focal tumor cell areas with chromosome aberrations can be recognized in the sections and can be correlated with histologic appearance. (2) No selection of cells occurs as a result of the isolation procedure. However, for tissue sections, enzymes must be applied to remove cytoplasmic and nuclear proteins. A variety of different soaking steps can be used prior to enzymatic digestion, depending on whether fresh or formalin-fixed material is to be labeled. Enzymes commonly applied for (F)ISH are shown in Table 8.5.4. Pepsin and proteinase K are the most frequently used, while subtilisin has been utilized to isolate nuclei from formalin-fixed, paraffin-embedded tissue. Subtilisin is the strongest proteolytic enzyme available and enables the total hydrolysis of strongly fixed proteins and peptides. For this reason, however, use of the enzyme is difficult to tune, and applicability is so far limited to nuclei isolated from paraffin-embedded tissue (Hyytinen et al., 1994). The choice between pepsin and proteinase K can be debated. It has been shown that similar results
8.5.18 Supplement 5
Current Protocols in Cytometry
Table 8.5.4
Enzymes for Removing Cytoplasmic and Nuclear Proteins for ISH
Protease (optimum pH)
Specificity
Biochemical application
ISH application
Pepsin (1.8-2.2)
Endopeptidase with a relatively broad specificity; preferentially cleaves bonds of Phe, Met, Leu, or Trp to other hydrophobic residues
Nonspecific hydrolysis of proteins and peptides in acidic media
Single cell suspensions, tissue sections, and nuclear preparations
Proteinase K (7.5-10.5)
Total degradation of Endopeptidase; cleaves peptide bonds primarily after the carboxyl group of N- proteins during the isolation of DNA and RNA substituted, hydrophobic aliphatic and aromatic amino acids; also active in the presence of SDS or urea
Subtilisin (7.0-11.0)
Nonspecific endopeptidase used for total hydrolysis at alkaline pH; most peptide bonds are cleaved; bonds adjacent to Asp, Glu, Ala, Gly, and Val are especially susceptible
are obtained with both enzymes, for both repeat-sequence and locus-specific probes. For standardization of the protocols, the use of pepsin can provide acceptable ISH results for different types of cell and tissue samples with minimal procedural modification. Fluorescence and bright-field ISH detection The use of fluorescence detection protocols provides a number of advantages for ISH, including easy and rapid detection of the fluorochrome-labeled probes, high sensitivity with low endogenous background, high resolution, possibility for multiple-target analysis with different fluorochromes, and the potential to quantitate signal intensity. Target sequences can be detected with directly labeled nucleic acid probes or by the application of immunocytochemistry to fluorescence- or hapten-labeled probes (Raap et al., 1989; UNITS 8.3 & 8.4). With conventional fluorescence microscopy, a target sensitivity on the order of 10 to 20 kb can be reached, which is sufficient to visualize deletions, amplifications, and translocations. These targets can be routinely detected in single cells isolated from fresh material and formalinfixed, paraffin-embedded material. In these cases, the cytoplasm must be completely or at least largely removed, so that autofluorescence is not problematic in the evaluation of ISH results (see Strategic Planning). The major advantages of cytochemical target/probe detection with enzymes include the stability of the precipitate (and thus the possibility for permanent storage of the cell prepa-
Nonspecific protease for the total hydrolysis of proteins and peptides
Single cell suspensions, tissue sections, and nuclear preparations
Nuclear preparations
rations; reviewed in Speel et al., 1995) and the fact that a standard bright-field microscope can be used for the analysis. The latter is a particular advantage in a setting where routine analyses must be performed. The in situ detection of target nucleic acid sequences requires the use of enzyme precipitation reactions that combine high sensitivity with precise localization properties. Moreover, rapid staining reactions resulting in stable precipitation products with contrasting colors are preferred. Horseradish peroxidase (HRP) and alkaline phosphatase (AP) can be efficiently used as detection enzymes for ISH. In the authors’ hands, the most appropriate enzyme reactions for ISH include the 3,3′-diaminobenzidine (DAB) and tetramethylbenzidine (TMB) reactions for HRP, and the naphthol-ASMX-phosphate/Fast Red (Fast Red) and naphthol-ASMX-phosphate/New Fuchsin (New Fuchsin) reactions for AP. The combination of peroxidase and alkaline phosphatase precipitation reactions is routinely applied in double-target ISH on single cells and tissue sections, both from frozen as well as paraffin-embedded tissue.
Critical Parameters and Troubleshooting Troubleshooting guidelines are presented in table format for single-cell preparations (see Table 8.5.1), frozen sections (see Table 8.5.2) and paraffin-embedded sections (see Table 8.5.3). In each case, observations of morphology and signals are correlated to potential evaluation errors and to possible solutions.
Molecular Cytogenetics
8.5.19 Current Protocols in Cytometry
Supplement 5
Pepsin pretreatment for single-cell suspensions For a routine application of ISH it is of the utmost importance that the different types of cell and tissue samples can be processed in a universal protocol, which leads to optimal results and constant quality. For this reason, pepsin is recommended for enzymatic digestion (see Background Information). To optimize the specificity and intensity of the ISH procedure for the detection of the chromosomal copy number, the pepsin concentration and digestion time need to be optimized. If pepsin concentration is too low, underdigestion can easily lead to underestimation of the chromosome copy number. If the sample is overdigested, high heterogeneity in chromosome copy numbers can be detected. A standard digestion schedule of 100 µg pepsin/ml in 0.01 N HCl for 20 min can be used for several tumor types, including bladder cancer, renal cell cancer, and leukemia, and several different cell lines. If nuclear morphology is completely lost as a result of the digestion step, proper evaluation of the signals cannot be made. An additional heating step or extra fixation steps can be incorporated in the protocol prior to digestion to improve morphology. The pepsin pretreatment protocol can also be used for cytological aspiration preparations, imprints, or cells isolated from chorionic villi for prenatal diagnosis.
Cell and Tissue Processing for Interphase Cytogenetics
Pepsin pretreatment for tissue sections For frozen tissue sections a standard protocol for pepsin digestion can be applied after a mild formaldehyde fixation. Most of the cytoplasmic proteins as well as some nuclear proteins are removed. Two important factors should be considered with tissue sections. First, after the digestion step the nuclei in the tissue section are very sensitive to pH changes as a result of the extensive removal of proteins. Retention of nuclear morphology is, however, guaranteed when the sections are acid dehydrated, keeping the nuclei condensed and preventing the relaxation of DNA which will take place at neutral pH. Second, when analyzing areas of tissue sections that have a high nuclear density, nuclear overlap will complicate the counting of the number of ISH signals per nucleus. In contrast to single-cell suspensions, a standardized protocol for formalin-fixed and paraffin-embedded tissue sections may have its limitations, as it has been shown that optimization of the proteolytic pretreatment step for
each paraffin block is essential. Fixation parameters (including but not limited to fixation time, pH of the formalin solution, size of the tissue blocks, and temperature of embedding in paraffin) influence the cross-linking of proteins (both histone and nonhistone) and nucleic acids. The success rate of the ISH protocol is mainly dictated by the removal of protein, required for accessibility of the target DNA for hybridization. Proteolytic enzymes such as pepsin and proteinase K are generally and efficiently used to permeabilize the tissue section to allow penetration of modified probes and antibodies. Sometimes additional steps are needed to enable an efficient retrieval of the target DNA. These include: (1) deparaffinization in warm xylene, (2) freezing and thawing of the sections, (3) soaking in hot SSC, (4) prolonged digestion time, (5) increased denaturation temperatures, and (6) HCl treatment prior to enzymatic digestion. Basic Protocols 3 and 5 and the Alternate Protocol also include extensive protein pretreatment steps prior to pepsin or proteinase K digestion. In the case of pepsin digestion, the formic acid/H2O2 and thiocyanate steps strongly improve the removal of proteins, while the acid dehydration steps guarantee retention of nuclear morphology. For proteinase K digestion, bisulfite is recommended as a protein pretreatment agent. Analysis in tissue sections Analysis of tissue sections is complicated by the introduction of nuclear truncation and nuclear overlap (Pahlplatz et al., 1995). Comparisons of the estimated chromosome copy number on the basis of single-cell protocols have shown that valid estimations of the real chromosome copy number can be made on the basis of extrapolations of ISH signal analyses in tissue sections. The best correlation between observations in single-cell suspensions and tissue sections of the same tumor is obtained in low-grade, diploid tumor cells, as the size of these nuclei is within the thickness of the tissue section (4 to 6 µm). When aneuploid cancers were studied by counting the number of ISH signals, a significant underestimation of the real copy number was made for individual chromosomes, because aneuploid nuclei are generally larger than diploid nuclei (Hopman et al., 1991). As a result, truncation of nuclei was a more significant problem in these tumor types. Techniques designed to solve the evaluation problem of sections are discussed (see Basic Protocol 9). In single-target tissue section hy-
8.5.20 Supplement 5
Current Protocols in Cytometry
bridizations, lymphocytes can serve as an internal control, while the signal distribution profile can be statistically evaluated. The best approach to obtaining a realistic impression of chromosomal aberrations is the calculation of a spot ratio by means of double-target ISH (Herbergs et al., 1996). The outcome of this approach is independent of (or at least less influenced by) variations in section thickness and ISH hybridization efficiency. Chromosomal imbalances such as monosomies and trisomies can be easily detected. However, when stemlines with high chromosome copy numbers and large nuclei are present, it is difficult to obtain reliable information concerning the true number of chromosome copies. When thick tissue sections (e.g., 50 µm) containing nontruncated nuclei are used, reliability is much higher, as it is mainly limited by ISH spot overlap and/or inefficient hybridization. On the other hand, ISH procedures performed on thin tissue sections are not seriously hampered by the latter two phenomena. The isolation of nuclei from thick tissue sections is a more valid approach for the detection of the true chromosome copy number.
Literature Cited Bentz, M., Döhner, H., Cabot, G., and Lichter, P. 1994a. Fluorescence in situ hybridization in leukemias: ‘The FISH are Spawning.’ Leukemia 8:1447-1452. Bentz, M., Cabot, G., Moos, M., Speicer, M.R., Ganser, A., Lichter, P., and Döhner. 1994b. Detection of chimeric BCR-ABL genes on bone marrow samples and blood smears in chronic myeloid and acute lymphoblastic leukemia by in situ hybridization. Blood 83:1922-1928. Cremer, T., Landegent, J., Bruckner, A., Scholl, H.P., Schardin, M., Hager, H.D., Devilee, P., Pearson, P., and van der Ploeg, M. 1986. Detection of chromosome aberrations in the human interphase nucleus by visualization of specific target DNAs with radioactive and nonradioactive in situ hybridization techniques diagnosis of trisomy 18 with probe L1.84. Hum. Genet. 74:346-352. Dhingra, K., Sneige, N., Pandita, T.K., Johnston, D.A., Lee, J.S., Emami, E., Hortobagyi, G.N., and Hittelman, W.N. 1994. Quantitative analysis of chromosome in situ hybridization signal in paraffin-embedded tissue sections. Cytometry 16:100-112. Herbergs, J., Speel, E.J.M., Ramaekers, F.C.S., de Bruïne, A.P., Arends, J.-W., and Hopman, A.H.N. 1996. Combination of lamin immunocytochemistry and in situ hybridization for the analysis of chromosome copy numbers in tumor cell areas with high nuclear density. Cytometry 23:1-7.
Hopman, A.H.N., Ramaekers, F.C.S., Raap, A.K., Beck, J.L.M., Devilee, P., van der Ploeg, M., and Vooijs, G.P. 1988. In situ hybridization as a tool to study numerical chromosome aberrations in solid bladder tumors. Histochemistry 89:307316. Hopman, A.H.N., van Hooren, E., van de Kaa, C.A., Vooijs, G.P., and Ramaekers, F.C.S. 1991. Detection of numerical chromosome aberrations using in situ hybridization in paraffin sections of routinely processed bladder cancers. Mod. Pathol. 4:503-513. Hyytinen, E., Visakorpi, T., Kallioniemi, A., Kallioniemi, O.-P., and Isola, J.J. 1994. Improved technique for analysis of formalin-fixed, paraffin-embedded tumors by fluorescence in situ hybridization. Cytometry 16:93-99. Kallioniemi, O.-P., Kallioniemi, A., Kurisu, W., Thor, A., Chen, L.-C., Smith, H.S., Waldman, F.M., Pinkel, P., and Gray, J.W. 1992. ERBB2 amplification in breast cancer analyzed by fluorescence in situ hybridization. Proc. Natl. Acad. Sci. U.S.A. 89:5321-5325. Kibbelaar, R.E., Kok, F., Dreef, E.J., Kleiverda, J.K., Cornelisse, C.J., Raap, A.K., and Kluin, P.M. 1993. Statistical methods in interphase cytogenetics: An experimental approach. Cytometry 14:716-724. Moore, D.H., Epstein, L., Reeder, J., Wheeless, L., Waldman, F.M., and the NCI bladder tumor marker network. 1996. Interlaboratory variability in fluorescence in situ hybridization analysis. Cytometry 25:125-132. Pahlplatz, M.M.M., De Wilde, P.C.M., Poddighe, P., van Dekken, H., Vooijs, G.P., and Hanselaar, A.G.J.M. 1995. A model for evaluation of in situ hybridization spot count distributions in tissue sections. Cytometry 20:193-202. Poddighe, P.J., Ramaekers, F.C.S., and Hopman, A.H.N. 1992. Interphase cytogenetics of tumors. J. Pathol. 166:215-224. Raap, A.K., Hopman, A.H.N., and van der Ploeg, M. 1989. Hapten labeling of nucleic acid probe for DNA in situ hybridization. In Techniques in Immunocytochemistry, Vol. 4 (G.R. Bullock and P. Petrusz, eds.) pp. 167-198. Academic Press, London. Sauter, G., Deng, G., Moch, H., Kerschman, R., Matsumura, K., De Vries, S., George, T., Fuentes, J., Carroll, P., Mihatsch, M.J., and Waldman, F.M. 1994. Physical deletion of the p53 gene in bladder cancer. Detection by fluorescence in situ hybridization. Am. J. Pathol. 144:756-766. Speel, E.J.M., Ramaekers, F.C.S., and Hopman, A.H.N. 1995. Cytochemical detection systems for in situ hybridization, and the combination with immunocytochemistry. “Who is still afraid of Red, Green and Blue?” Histochem. J. 27:833858. Tkachuk, D.C., Pinkel, D., Kuo, W.-L., Weier, H.U., and Gray, J. 1991. Clinical applications of fluorescence in situ hybridization. Genet. Anal. Tech. Appl. 8:67-74.
Molecular Cytogenetics
8.5.21 Current Protocols in Cytometry
Supplement 5
van Dekken, H., Pizzolo, J.G., Reuter, V.E., and Melamed, M.R. 1990. Cytogenetic analysis of human solid tumors by in situ hybridization with a set of 12 chromosome-specific DNA probes. Cytogenet. Cell Genet. 54:103-107.
Willard, F.W. and Waye, J.S. 1987. Hierarchical order in chromosome-specific human alpha satellite DNA. Trends Genet. 3:192-198.
van Dekken, H., Bosman, F.T., Teijgeman, R., Vissers, G.J., Tersteeg, T.A., Kerstens, H.M., Vooijs, G.R., and Verhofstad, A.A. 1992. Identification of numerical chromosome aberrations in archival tumours by in situ hybridization to routine paraffin sections: Evaluation of 23 phaeochromocytomas. J. Pathol. 171:161-171.
Contributed by A.H.N. Hopman and F.C.S. Ramaekers University of Maastricht Maastricht, The Netherlands
Cell and Tissue Processing for Interphase Cytogenetics
8.5.22 Supplement 5
Current Protocols in Cytometry
Advanced Preparative Techniques to Establish Probes for Molecular Cytogenetics
UNIT 8.6
The technique of fluorescence in situ hybridization (FISH) has become central to many genetic studies. Improvements in sensitivity, coupled with developments in computer software and microscope technology, have enhanced its usefulness. The utility of the FISH approach is critically dependent on the supply of appropriate probes. This unit describes a variety of advanced preparative techniques designed to provide chromosomal fragments for further experimentation, which may include FISH but could also involve the cloning of the prepared fragments to create DNA libraries (not covered in this unit). The methods covered include flow cytometry of metaphase chromosomes (see Basic Protocol 1), chromosome microdissection (see Basic Protocol 2), and the degenerate oligonucleotide– primed polymerase chain reaction (DOP-PCR) amplification method for reverse chromosome painting (see Basic Protocol 3). The application of flow cytometry to chromosomal analysis has had a major impact on cytogenetics and cancer genetics. In particular, the ability to flow-sort individual chromosomes has played a major role in providing material suitable for whole-chromosome painting and the creation of chromosome-specific DNA libraries. An important advance has been the use of two fluorochromes instead of one, which provides a greater resolution of chromosomes that have a similar overall DNA content. The emission signals from these fluorochromes are plotted as a two-dimensional histogram, called a bivariate flow karyotype. The AT-specific dye Hoechst 33258 and the CG-specific dye chromomycin A3 are commonly used in bivariate flow karyotyping of human chromosomes. With these dyes, the position of the chromosome peaks in the resulting flow karyotype provides quantitative information on the relative chromosomal DNA content and base composition of the DNA. In this way most of the human chromosomes can be distinguished, except for chromosomes 9, 10, 11, and 12, which appear in one peak. Flow sorting of individual peaks can provide chromosomal material in a form suitable for construction of chromosome-specific libraries and paints, for gene mapping, and for ongoing analysis of karyotypic changes in tumors. Basic Protocol 1 describes isolation, staining, and flow sorting of human chromosomes using the dyes Hoechst 33258 and chromomycin A3. For information on preparing the cell sorter for this analysis, see Support Protocol 1. Microdissection is the process by which a specific region of a chromosome can be removed from the cell using microneedles specially designed for such microsurgery. The technique is attractive, as it enables the operator to access directly any small region of a chromosome and remove it from the cell. The fragment of chromosome can then be analyzed using a variety of molecular approaches. Early experiments using microdissection and microcloning were time-consuming and cumbersome, requiring more than 100 fragments to be microdissected per investigation with relatively few microclones being isolated. Nowadays microdissection and microcloning techniques can be applied to any source of chromosome; the chromosomes can be GTG-banded for accurate identification and the polymerase chain reaction (PCR) can be incorporated to minimize the number of chromosome fragments that need to be microdissected. Using PCR, 15 to 20 fragments are adequate for a typical experiment. Basic Protocol 2 describes techniques for chromosome harvesting, banding, and microdissection; for preparation of the microneedles used in microdissection, see Support Protocol 2. Once whole chromosomes or fragments have been isolated by flow sorting or microdissection, it is often desirable to establish the purity. Because the quantity of DNA obtained Molecular Cytogenetics Contributed by Jan Stap, Jacob A. Aten, D. Lillington, A. Shelling, and B.D. Young Current Protocols in Cytometry (1998) 8.6.1-8.6.23 Copyright © 1998 by John Wiley & Sons, Inc.
8.6.1 Supplement 5
by either method is usually very small, some kind of PCR amplification is normally used. The amplified products can be used for FISH (reverse painting) as described in Support Protocol 3 or can be cloned for further analysis. This allows an assessment of the purity of the chromosome fragments and is an essential step if a library is to be constructed. The DOP-PCR method described in Basic Protocol 3 was initially devised as a rapid and efficient method of amplifying microdissected chromosomes without requiring excessive technical expertise. It has worked successfully in a number of laboratories, and has been shown to give strong signals when DNA amplified from flow sorted or microdissected chromosomes is hybridized back to normal metaphase chromosomes by fluorescence in situ hybridization (FISH). For FISH analysis of PCR products, see Support Protocol 3. BASIC PROTOCOL 1
PREPARATIVE TECHNIQUES FOR FLOW CYTOMETRIC ANALYSIS OF CHROMOSOMES This protocol is based on procedures published by Sillar and Young (1981) and Boschman et al. (1991, 1992) for isolating and staining chromosomes using the dyes Hoechst 33258 and chromomycin A3 for bivariate analysis and sorting. The procedure consists of five stages: (1) collection of mitotic cells, which differs from standard procedures used in conventional karyotyping methods as more cells are needed for flow karyotyping; (2) swelling of mitotic cells in a hypotonic solution; (3) stabilization with polyamines and subsequent release of the metaphase chromosomes from the mitotic cells; (4) staining of the chromosomes with Hoechst 33258 and chromomycin A3; and (5) sorting the desired chromosomes (after alignment of the sorter). The flow cytometer should be equipped with two lasers, one of which should be capable of UV illumination, in order to analyze the double-stained chromosomes. Most commercially available two-laser instruments with sorting capabilities are suitable (see Chapter 1 for more detailed information about flow cytometry instrumentation and UNIT 4.3 for a discussion of nuclear stains). For flow sorting of chromosomes, the quality of the chromosome suspension and the procedure used for sorting are equally important. Support Protocol 1 includes some general remarks about the alignment of the laser optics; however, the handbook for the particular cell sorter used must be consulted for details. The alignment should be done in advance, but it is important to check alignment on the day the sorting is carried out. It is assumed that the reader has a basic understanding of and experience in the use of flow sorters. When using bivariate flow karyotyping for the analysis of abnormal chromosomes, it may be useful to compare flow karyotypes from abnormal tissue with karyotypes from normal tissue of the same individual, because of observed variations among individuals. Normal human chromosomes can be obtained from cell cultures or stimulated lymphocytes. If possible, the isolation/staining and the measuring of normal and abnormal chromosomes should be carried out on the same day, to reduce day-to-day variation (see Critical Parameters and Troubleshooting for more detailed discussion of the variation problem).
Advanced Preparative Techniques for Cytogenetic Probes
Materials Human cell culture 10 µg/ml Colcemid (demecolcine; Sigma; store ≥8 weeks at 4°C) 37.5 mM KCl Polyamine solutions 1 and 2 (see recipe) 1 M MgSO4 500 µg/ml 4′,6-diamidino-2-phenylindole (DAPI) or 1 mg/ml propidium iodide (PI)
8.6.2 Supplement 5
Current Protocols in Cytometry
5 mg/ml (100×) chromomycin A3 (Sigma) in ethanol (store ≥6 weeks at –20°C) 100 µg/ml (100×) Hoechst 33258 in H2O (store ≥6 weeks at –20°C) 1 M sodium citrate Latex beads for aligning lasers (see Support Protocol 1 and manufacturer’s instructions) Collection buffer for sorted chromosomes (composition appropriate for subsequent analysis) containing at least 0.1% (w/v) bovine serum albumin (BSA, fraction V; Sigma) Glycerol or 100% ethanol Tabletop centrifuge Syringes with 0.7 × 40–mm (22-G) needles Fluorescence microscope Flow cytometer/cell sorter (e.g., Becton Dickinson FACStar Plus or FACSVantage) with two argon lasers and the following filters: For Hoechst 33258 detection (green fluorescence): 480-nm short-pass filter and 380- or 390-nm long-pass filter to block scattered UV laser light For chromomycin A3 detection (blue fluorescence): 490-nm long-pass filter Siliconized tubes (e.g., 1.7-ml Costar microcentrifuge tubes, no. 3207) or BSA-coated microcentrifuge tubes (see recipe) Cytocentrifugation system: e.g., Leif bucket (Leif, 1971) Microscope slides (precleaned with 1:1 ethanol/ether) Additional materials for cell culture (APPENDIX 3B) and for preparing and aligning flow sorter (see Support Protocol 1 and manufacturer’s instructions) CAUTION: All DNA stains—e.g., PI, DAPI, chromomycin A3, and Hoechst 33258—are suspected to be toxic and mutagenic and should be handled with care. See UNIT 3.3 regarding aerosol containment during flow sorting. NOTE: All solutions and equipment coming into contact with cells must be sterile and proper sterile technique should be used accordingly. NOTE: All culture incubations are performed in a humidified 37°C, 5% CO2 incubator unless otherwise specified. Collect mitotic cells (day 1) 1. Add Colcemid to the cell culture (as 10 µg/ml demecolcine solution) to a final concentration of 0.1 µg/ml to block cells in mitosis. Incubate for the appropriate amount of time. The duration of incubation with Colcemid depends on the cellular growth rate; fast-growing cells (cell-cycle time of ∼10 hr) need only 1 hr, whereas cells with a cell-cycle time of 30 hr may require up to 16 hr. To obtain a high-resolution flow karyotype, at least 25,000 mitotic cells are required. Either cell cultures or stimulated leukocytes can be used to obtain mitotic cells. Some cells may respond better to lower or higher concentrations of Colcemid or to a different mitotic inhibitor such as vindesine (Eli Lilly; to be used in a final concentration of 3.5 × 10−8 g/ml, as tested on several cell types).
2. After incubation with Colcemid, harvest the mitotic cells by shaking the flask and tapping the side of the flask firmly. Transfer cell suspension to a centrifuge tube. If it is not possible to remove the mitotic cells from the bottom of the culture flask by tapping firmly against the side of the flask, remove the culture medium and trypsinize cells (APPENDIX 3B). After an incubation just long enough to detach the cells without permeabilizing them, dilute the trypsin/EDTA solution with culture medium that has been supplemented with 0.1 ìg/ml Colcemid and try again to collect the mitotic cells by shaking them off.
Molecular Cytogenetics
8.6.3 Current Protocols in Cytometry
Supplement 5
Swell and stabilize mitotic cells and prepare sorter (day 1) 3. Centrifuge cells 5 min at 200 × g, room temperature, and remove supernatant. Resuspend pellet in 1 ml 37.5 mM KCl and incubate 15 min at room temperature to swell cells. Some cell types require longer swelling.
4. Add 9 ml polyamine solution 1, then centrifuge 5 min at 200 × g, room temperature. Remove the supernatant and add 1 ml polyamine solution 2 to the pellet. The digitonin in polyamine solution 2 permeabilizes the plasma membrane of the cells, allowing the polyamines to enter the cells and stabilize the chromosomes.
5. Add 10 µl 1 M MgSO4 (final concentration, 10 mM). Staining the chromosomes with chromomycin A3 requires the presence of divalent cations.
6. Carefully pass cell suspension 1 to 3 times (depending on the cell line) through a syringe with a 0.7 × 40–mm (22-G) needle to release the stabilized chromosomes. 7. Place a small droplet of cell suspension on a slide. Add DAPI (from 500 µg/ml stock) to a final concentration of ∼5 µg/ml or PI (from 1 mg/ml stock) to a final concentration of 50 µg/ml, then check the effects of syringing under a fluorescence microscope. The chromosomes should be very compact owing to the polyamines. If they appear as “strings” the syringing probably was too forceful. If the chromosomes are still aggregated, syringe again and check the effect under the microscope. If the membranes of cells of a specific cell type are disrupted too readily, the chromosomes might be damaged by the syringing. If that is the case, vortexing the cell suspension for 4 to 8 sec might be sufficient to release the chromosomes. Again, it is most important to check the quality of the chromosome suspension under the microscope (see Critical Parameters and Troubleshooting). At this point in the procedure the chromosome suspension may be stored overnight at 4°C.
8. Clean flow sorter tubing and incubate tubing overnight with 1 µg/ml Hoechst 33258 (see Support Protocol 1). Set up system using a sheath fluid consisting of sterile 0.9% NaCl. Make sure that the flow sorter is set up and aligned (see Support Protocol 1 and manufacturer’s instructions).
Stain chromosomes (day 2) 9. Add 10 µl of 5 mg/ml (100×) chromomycin A3 per ml of chromosome suspension (final concentration, 50 µg/ml) and 10 µl of 100 µg/ml (100×) Hoechst 33258 per ml of chromosome suspension (final concentration, 1 µg/ml). Incubate at least 2 hr at 4°C, to allow stains to equilibrate. 10. Add sodium citrate (from 1 M stock) to a final concentration of 10 mM, ∼15 min before analysis on the sorter, to enhance resolution of the peaks in the flow karyotype. It may be advisable to check the quality of the chromosomes using DAPI (see step 7) shortly before analysis on the sorter.
Advanced Preparative Techniques for Cytogenetic Probes
Perform flow cytometry and cell sorting 11. Check the alignment of the cell sorter with latex beads shortly before sorting (see Support Protocol 1 and manufacturer’s instructions). Pass the stained chromosomes through the sorter and collect the flow karyotype data. During passage of the chromosomes, make slight adjustments to alignment as necessary to obtain optimal results. For the quantitative analysis of flow karyotypes of abnormal chromosomes, several methods are possible (e.g., Boschman et al., 1992; van den Engh et al., 1990). The authors
8.6.4 Supplement 5
Current Protocols in Cytometry
recommend the method described by Boschman et al. (1992) to analyze the flow karyotypes, especially when normal and abnormal chromosomes cannot be measured on the same day.
12. Select the type of chromosome to be sorted, place a gate over the corresponding distribution, and start sorting the selected chromosomes into a siliconized or BSAcoated collection tube containing ∼100 µl of the appropriate collection buffer. The composition of the collection buffer should be adapted to the subsequent analysis of the sorted chromosomes. To obtain good recovery, it is important that the collection buffer contain at least 0.1% BSA. If, for some reason, it is not possible to use silanized tubes, nonsilanized tubes may be used if they are coated with BSA (see Reagents and Solutions). The number of chromosomes to sort largely depends on the intended subsequent analysis. For most PCR procedures at least 300 chromosomes are needed. If a precise number of sorted chromosomes is desired, it is advisable to check the number of chromosomes by counting a small portion of the sorted chromosome population using a hemacytometer (APPENDIX 3B).
Process sorted chromosomes 13a. If the sorted chromosomes are not be be used immediately: Add glycerol to a final concentration of 30% (v/v) to prevent damage to the chromosomes and store up to several days at −20°C. Alternatively, add 100% ethanol to a final concentration of 30% (v/v), to prevent the suspension from freezing, and store at −20°C. 13b. If it is desirable to deposit chromosomes on a slide individually and morphologically intact for further analysis (e.g., FISH): Centrifuge sorted chromosomes 10 min at 400 × g onto a precleaned slide using a Leif bucket or other cytocentrifugation system. Cytocentrifugation systems with filter cards are less suitable because of chromosome loss. By centrifuging chromosomes onto a slide, a large fraction of chromosomes may be lost. Loss of chromosomes can be minimized by silanizing the chambers of the bucket. Another alternative to prevent loss is to sort the chromosomes directly into a droplet on a slide, instead of using a cytocentrifuge. ∼3000 chromosomes are usually sufficient for further analysis. Let the suspension of sorted chromosomes air dry on a hot plate at ∼40°C and fix by immersing 30 min in 70% ethanol or 3 parts methanol/1 part glacial acetic acid, at room temperature. Overnight incubation in methanol/acetic acid at −20°C is helpful in making the DNA accessible for probe DNA in FISH procedures.
14. After finishing the sorting experiment, clean the cell sorter to prevent clogging of the nozzle. PREPARATION OF FLOW SORTER FOR CHROMOSOME ANALYSIS The following are general guidelines for optimizing sorter performance. For details, see manufacturer’s instructions for the sorter. See Chapter 1 for additional discussion of optical alignment and flow sorter fluidics. Additional Materials (also see Basic Protocol 1) 1:20 sodium hypochlorite or cleaning solution recommended by sorter manufacturer (e.g., FACSafe and FACSRinse; Becton Dickinson) 70% ethanol (∼0.5 liter needed for sterilizing tubing of cell sorter) Sterile distilled water 0.9% NaCl, sterile 1 µg/ml Hoechst 33258 in 1× phosphate-buffered saline (PBS; see recipe), sterile Latex beads for aligning the lasers: 1.0 ± 0.01 µm yellow/green Fluoresbrite microspheres for aligning 457-nm laser and 1.0 ± 0.01 µm bright blue Fluoresbrite microspheres for aligning the UV laser (Polysciences)
SUPPORT PROTOCOL 1
Molecular Cytogenetics
8.6.5 Current Protocols in Cytometry
Supplement 5
Perform on day 1 (see Basic Protocol 1, step 8) 1. Before attempting any analysis or sorting, thoroughly clean the sample tubing of the cell sorter using 1:20 sodium hypochlorite, or preferably according to the guidelines given by the instrument manufacturer. Remove all cleaning solutions by rinsing the tubing with distilled water. Becton Dickinson recommends rinsing with FACSafe and FACSRinse for their instruments.
2. Sterilize tubing by incubating ∼10 min with 70% ethanol, and remove all the ethanol by flushing first with sterile distilled water and subsequently with sterile 0.9% NaCl. The necessity of this step depends on the intended subsequent analysis of the sorted chromosomes. For example, when PCR techniques are applied, sterilization of the tubing and all the solutions used is necessary.
3. Incubate sample tubing overnight with sterile 1× PBS containing 1 µg/ml Hoechst 33258. This step is advisable for bivariate flow karyotyping, in order to obtain a more stable staining equilibrium during the sorting procedure. Shorter incubations may be sufficient, but overnight is often more practical.
Perform on day 2 4. Check alignment of cell sorter with latex beads shortly before sorting. In order to obtain a fluorescence distribution with narrow peaks, all chromosomes should be illuminated with the same light intensity. This means that the alignment is crucial; hence the laser beam and fluorescence pathway should be aligned using latex beads. If the alignment is correct, the electronic pulse shapes of the illuminated beads are approximately Gaussian, without irregularities. The coefficient of variation (CV) values measured with the beads should not exceed 2% for a successful analysis. The illumination of these beads in the sample stream should be homogeneous and the sample stream should be orthogonal to the laser beam and the collector lens.
5. During initial passage of chromosomes through the sorter (see Basic Protocol 1, step 11), align and adjust the sorter according to the following guidelines (also see manufacturer’s instructions for sorter). The stained chromosomes should pass through the UV laser beam first, and then through the visible-emission laser beam. This results in better CV values for the measured fluorescence distributions than with the lasers in the opposite order. Consequently, the sorter is triggered by the Hoechst signal.
a. Fluidics: To achieve precise focusing of the chromosomes in the sample stream, adjust the frequency, phase, and amplitude of the transducer until stable and narrow side streams are obtained (consult manufacturer’s instructions for details). The following instrument settings are recommended for a conventional jet-in-air sorter: 70-ìm nozzle, stable sample pressure of ∼10 psi, and sample rate <1500 particles per second. One might consider using a sheath tank with a gauge to maintain a constant sheath pressure, independent of the sheath level. This improves the sort stability. The optimal sample rate depends on the chromosome concentration in the sample. If the chromosome concentration is very low, the optimal rate may be as low as 100 chromosomes/sec.
Advanced Preparative Techniques for Cytogenetic Probes
b. Adjustment of sorting device: Make a bivariate histogram using chromosomes and place a sort gate over the fluorescence distribution of the largest chromosome (chromosome 1). Place a slide under the sorting device and mark the spot where the chromosomes are sorted. Calculate the drop delay and sort an exact number of chromosomes (e.g., 20) in a small droplet of ∼10 µl polyamine solution 1 on
8.6.6 Supplement 5
Current Protocols in Cytometry
the marked spot. Place a small coverslip on the suspension and count the chromosomes under the fluorescence microscope. If the number is not correct, repeat the adjusting procedure using latex beads until a good sort is obtained. CHROMOSOME HARVESTING, BANDING, AND MICRODISSECTION In this procedure, chromosomes are harvested according to a procedure that works well for the subsequent microdissection. To ensure adequate yields of metaphase chromosomes, the cell cultures should have an optimal density of 106 cells/ml. The isolated chromosomes are then banded. It is better to band chromosomes on the same day as they are required, preferably under sterile conditions, because the fresh chromosomes are softer and easier to cut. G banding can be achieved by treating the coverslips with trypsin and then staining in Leishman’s stain.
BASIC PROTOCOL 2
Following harvesting and banding, the chromosomes are microdissected by means of an inverted microscope fitted with micromanipulation equipment. This provides a convenient, reliable, and relatively straightforward method of microdissection. Chromosome regions are microdissected from coverslips by means of glass needles that have been prepared in the laboratory (see Support Protocol 2). The needle tip with the chromosome fragment attached is then broken into a microcentrifuge tube containing either PCR buffer or sterile water. When a sufficient number of chromosome fragments have been microdissected, they are amplified in vitro by PCR (see Basic Protocol 3). Materials Human cell culture 10 µg/ml Colcemid (demecolcine; Sigma; store ≥8 weeks at 4°C) 0.075 M KCl, 37°C 70% ethanol, ice cold Fixative: 3 parts methanol/1 part glacial acetic acid (ice cold) Trypsin solution: 1 ml Bacto Trypsin (Difco) reconstituted in 40 ml of 0.9% NaCl 0.9% NaCl Gurr buffer pH 6.8 (Bio/medical Specialties) 1 part Leishman’s stain (e.g., Baxter) in 4 parts Gurr buffer, pH 6.8 (prepare fresh) Taq DNA polymerase buffer Coverslips, cleaned with methanol and kept in sterile container in freezer Zeiss Axiovert inverted microscope with 16× eyepieces, 4× or 6× objective, 63× dry lens objective, and gliding and rotating stage Remote-controlled micromanipulator (Zeiss) and mounts for micromanipulator Vibration-free table Microdissection needles (see Support Protocol 2) NOTE: All solutions and equipment coming into contact with cells must be sterile and proper sterile technique should be used accordingly. NOTE: All culture incubations are performed in a humidified 37°C, 5% CO2 incubator unless otherwise specified. Harvest chromosomes 1. Add 50 µl of 10 µg/ml Colcemid per 5 ml of culture (using 0.1 ml to 10 ml of culture, total). Incubate 10 to 60 min at 37°C. IMPORTANT NOTE: Harvesting must be performed in sterile safety cabinet using aseptic techniques.
Molecular Cytogenetics
8.6.7 Current Protocols in Cytometry
Supplement 5
2. Centrifuge 10 min at 200 × g, room temperature. Remove supernatant and add an equal volume of 37°C 0.075 M KCl. Let stand 10 to 15 min. 3. Centrifuge again, 10 min at 200 × g. Remove supernatant and add 5 to 10 ml ice-cold 70% ethanol for prefixation. Incubate ≥30 min at −20°C. 4. Centrifuge again, 10 min at 200 × g. Remove all but ∼0.25 ml of supernatant and resuspend pellet in the residual fluid. 5. Depending on pellet size, add 0.5 to 2 ml of fresh ice-cold fixative (so that suspension is just cloudy in appearance) while constantly agitating tube. Immediately (i.e., within 20 sec of the time point at which the fixative is first added) drop onto clean, ice-cold coverslips and place a drop of ice-cold 70% ethanol on top of each drop of suspension. Coverslips rather than glass slides are preferred because glass slides are too thick for the objectives on the inverted microscope system.
6. Allow preparations to air dry in sterile hood, then store at −20°C in sterile container. Band chromosomes 7. Immerse coverslip 5 to 10 sec in a sterile tube containing trypsin solution. 8. Rinse coverslip first in fresh 0.9% NaCl, then in Gurr buffer pH 6.8. 9. Immerse coverslip 60 to 90 sec in a sterile tube containing diluted Leishman’s stain. 10. Rinse coverslip in Gurr buffer, pH 6.8, then rinse in sterile distilled water. Allow to air dry in sterile hood. 11. Set up microscope with microdissection equipment on a vibration-free table. Check banding and staining quality under microscope before using chromosomes for microdissection. Chromosomes should show discrete bands and not be fuzzy at the edges.
Microdissect chromosomes 12. Locate an easily accessible target chromosome and center it in the field of view. Rotate stage until chromosome is perpendicular relative to the direction of the needle. 13. Place needle holder in micromanipulator. Using a 6× objective, with the condenser diaphragm partly closed, manually move and lower the needle (i.e., without using the joystick) toward the coverslip until the tip of the needle is just above the chromosomes. 14. Change to a high-power (63×) dry lens and narrow the condenser diaphragm. Using the remote-control joystick for accurate control of the micromanipulator, lower the needle until the tip just comes into focus but not in contact with the coverslip. Closing the condenser diaphragm creates an apparent depth of field so the needle will be in focus well above the coverslip.
15. Open condenser to view chromosomes and lower the needle tip further using fine movement controls until the tip appears in the field of view. When the condenser is opened, the needle tip will not be in the field of view until it is lowered further towards the coverslip. Advanced Preparative Techniques for Cytogenetic Probes
16. Move needle to the desired position for microdissection—i.e., at the edge of the target chromosome region. 17. Lower needle further until the needle tip just touches the coverslip.
8.6.8 Supplement 5
Current Protocols in Cytometry
Any lowering beyond this point will cause the needle to move forward through the chromosome under its own weight. Alternatively, the needle can be moved manually through the chromosome using joystick controls. It is preferable to use the joystick to direct the microdissection. As the needle cuts through the chromosome, the fragment folds up onto the top side of the needle tip. If the fragment does not automatically adhere to the needle tip but is merely pushed out to the side of the chromosome, then it can usually be picked up by gentle prodding with the needle tip.
18. Lift the needle away from the chromosome and coverslip and break the needle tip directly into a microcentrifuge tube containing 5 µl Taq DNA polymerase buffer, using pressure against the side of the tube. Proceed to PCR amplification (see Basic Protocol 3). PREPARING MICRONEEDLES FOR MICRODISSECTION This protocol describes a method for producing microneedles from either glass rod or thick-walled glass tubing. A microelectrode puller with variable pulling force and heat setting (e.g., Campden Instruments) allows control of the length and shape of the needle tips. The needles are mounted into holders constructed to maintain the needle tip in the correct orientation when flattened sections of the stainless steel shaft are clamped into the grinder and subsequently the micromanipulator. The needle itself is clamped into a miniature crocodile clip that has had the flat ends bent to clamp around the tubular needle. After being ground as described below, the needle tip is heated to 300°C in a microoven to remove any possible contaminants from the needle before use. The microoven basically consists of a 5-mm glass tube surrounded by a heating element and insulating jacket. The temperature is controlled by a thermostat or electrothermal regulator.
SUPPORT PROTOCOL 2
Materials 70% ethanol 1- or 1.5-mm-diameter borosilicate glass rods or capillary tubes Microelectrode puller (e.g., Campden Instruments or David Kopf Instruments) Grinder (e.g., Narashige) with lens system Microneedle holders (custom made) Microoven 1. Select a suitable puller setting for the type of glass being used. Mount a 7.5-cm length of glass rod or capillary tube into the microelectrode puller. 2. Pull glass slowly to produce two short needles with fine points. 3. Mount the needles into the microneedle holders and clamp holder into grinder at an angle of 40°. 4. While viewing needle tip through the lens system and applying drops of water to the grinding wheel, lower the needle until the tip is just touching the wheel. If glass rod is used to prepare the needle, grind tip until a faint bright spot appears. If glass tube is used, grind tip until the water just begins to rise up the tube.
5. Wash the needle in a stream of 70% ethanol. 6. Carefully place needle in microoven set at 300°C for 30 sec. Allow needle to cool before using. Molecular Cytogenetics
8.6.9 Current Protocols in Cytometry
Supplement 5
BASIC PROTOCOL 3
DEGENERATE OLIGONUCLEOTIDE–PRIMED PCR (DOP-PCR) This method (modified from Guan et al., 1993) takes advantage of a primer that contains partially degenerate nucleotides to randomly amplify short fragments at frequently occurring priming sites within the genome. The principle of degenerate oligonucleotide– primed PCR and the DOP primer were first described by Telenius et al. (1992a). Briefly, priming occurs from the 3′ ATGTGG nucleotides during the initial low-annealing-temperature cycles of the PCR reaction. These sequences occur frequently within the genome, at a frequency similar to restriction endonuclease recognition sites. The six degenerate nucleotides help to stabilize the specified 3′ primer sequences by effectively allowing the primer to anneal as a 12-mer. The 5′ end of the primer contains the nucleotide recognition sequence for the XhoI restriction endonuclease, required for later cloning steps. These 5′ sequences also allow primers to anneal efficiently to previously amplified DNA, thus permitting a higher annealing temperature during later PCR cycles. Materials Microdissected chromosomal fragments in 5 µl Taq DNA polymerase buffer (see Basic Protocol 2) 50 mM MgCl2 5 U/ml Taq DNA polymerase 5 mM 4dNTP mix: 5 mM each of dATP, dCTP, dGTP, and dTTP 20 µM DOP-PCR primer (CCGACTCGAGNNNNNNATGTGG, where N = A, C, G, or T and the underlined bases represent an XhoI restriction site) Sterile H2O Mineral oil Automated thermal cycler 1. Place the tube containing microdissected chromosomal fragments in 5 µl 10× Taq polymerase buffer (see Basic Protocol 2) on ice. Some protocols suggest using a proteinase K and sodium dodecyl sulfate (SDS) step followed by phenol extraction. The authors and others have found that omitting this step has no apparent effect on amplification or subsequent probe quality.
2. Add the following reaction components to the tube (already containing 5 µl 10× Taq polymerase buffer): 2 µl 50 mM MgCl2 (2 mM final) 2 µl 5 mM 4dNTP mix (200 µM each dNTP final) 4 µl 20 µM DOP primer (1.6 µM final) 0.5 µl 5 U/ml Taq DNA polymerase (0.00005 U/µl final) 36.5 µl sterile H2O. Overlay with 100 µl mineral oil. For hot-start PCR, add the Taq polymerase to the reaction at the end of the initial 94°C denaturing step to minimize inactivation of the Taq polymerase enzyme and to prevent nonspecific primer extension during the precycling period.
Advanced Preparative Techniques for Cytogenetic Probes
8.6.10 Supplement 5
Current Protocols in Cytometry
3. Perform first-round PCR by amplifying DNA in an automated thermal cycler using the following program: Initial step: 8 cycles:
28 cycles:
Final step:
10 min 1 min 1 min 3 min 1 min 1 min 3 min 10 min
94°C 94°C 30°C 72°C 94°C 56°C 72°C 72°C
(denaturation) (denaturation) (annealing) (extension) (denaturation) (annealing) (extension) (extension).
4. Remove 1 µl of amplification product from first-round PCR above and place into a new reaction containing the same components as in step 2 (with 5 µl of 10× Taq DNA polymerase buffer). One round of amplification is usually insufficient to achieve detectable amounts of DNA.
5. Perform second-round PCR by repeating the last 28 cycles of the program in step 3, with a final 10-min extension period at 72°C. 6. Assess PCR products by separation on agarose gel. FISH ANALYSIS OF PCR PRODUCTS Following DOP-PCR amplification of microdissected chromosomal DNA or flow sorted chromosomes, fluorescence in situ hybridization (FISH) can be performed to assess the origin of the template DNA.
SUPPORT PROTOCOL 3
Materials DNA from DOP-PCR (see Basic Protocol 3) Cot-1 DNA (Life Technologies) 3 M sodium acetate, pH 5.6 (UNIT 8.3) 100% ethanol, ice-cold Hybridization buffer (see recipe) Cells containing chromosomes for hybridization RNase A solution (see recipe) Denaturing solution (see recipe; prepare fresh) 20× SSC, pH 5.3, and 2× SSC, pH 7 (APPENDIX 2A) Dehydration series: 70%, 95%, and 100% ethanol, ice-cold Rubber cement Formamide wash solution (see recipe) 4× SSCT (see recipe) Blocking solution (see recipe; prepare fresh) 5 µg/ml fluorescein isothiocyanate–conjugated avidin (avidin-FITC; Sigma or Vector Laboratories; store up to 6 months at −20°C) 5 µg/ml biotinylated anti–avidin D antibodies (Vector Laboratories; store up to 6 months at −20°C) Phosphate-buffered saline (PBS; see recipe) Citifluor/PI (see recipe) DNA purification system: Geneclean kit (Bio 101) or Promega Wizard PCR Preps DNA Purification Kit 42°, 75°, and 90°C water baths Microscope slides 32 × 22–mm coverslips
Molecular Cytogenetics
8.6.11 Current Protocols in Cytometry
Supplement 5
Moist chamber: 1-liter beaker containing paper towels moistened with 2× SSC (APPENDIX 2A), covered with aluminum foil Coplin jars Confocal or fluorescence microscope (see Chapter 2) Additional reagents and equipment for labeling of DNA by nick translation and purification of labeled probes for FISH (UNIT 8.3) and spreading of metaphase chromosomes on slides for FISH (UNIT 8.2) Prepare probe DNA 1. Purify an aliquot of DNA from DOP-PCR by use of a Geneclean kit (gel electrophoresis) or a Promega Wizard PCR Preps DNA Purification kit (column chromatography). 2. Label 1 µg of DNA with biotin-11-dUTP by nick translation (see UNIT 8.3, Basic Protocol 1). Purify labeled probe using a Sephadex G-50 column (see UNIT 8.3, Support Protocol 1). 3. Place 200 ng labeled probe into a 1.5-ml microcentrifuge tube on ice. Add 5 µl Cot-1 DNA, then add 1/10 vol 3 M sodium acetate, pH 5.6, and 2 vol ice-cold 100% ethanol. Place tube at −20°C overnight or at −70°C for 1 hr. 4. Microcentrifuge 20 min at maximum speed, 4°C. Pour off supernatant and invert tube on tissue until pellet is dry. 5. Add 15 µl hybridization buffer to the pellet and mix gently by pipetting. Denature probe DNA by placing tube in 90°C water bath for 5 min, then plunge tube into ice bath. 6. Microcentrifuge briefly at maximum speed to get all liquid to bottom of tube. Place tube in 37°C water bath for 2 to 3 hr and prepare slides during this incubation. Prepare slides and denature chromosomal DNA 7. Spread chromosomes from metaphase cells on slides using standard procedures (see UNIT 8.2, Basic Protocol 1, steps 1 to 18). 8. Place 100 µl RNase A solution on each slide and place a coverslip on top. Incubate slides in a moist chamber 30 to 60 min at 37°C. Meanwhile prepare denaturing solution in Coplin jar and place in 75°C water bath. 9. After incubation, remove coverslips from slides and wash twice, each time for 3 min, in 2× SSC, with agitation. 10. Successively immerse slides for 3 min each in ice-cold 70%, 95%, and 100% ethanol to dehydrate. Air dry. 11. Place slides in denaturing solution at 75°C for 3 min, then dehydrate through ice-cold ethanol series and air dry as in step 10. Hybridize probe to slides 12. Prewarm slides on a rack in 42°C water bath for 2 min.
Advanced Preparative Techniques for Cytogenetic Probes
13. Remove probe from 37°C water bath (see step 6) and apply to slide while still in 42°C water bath. Cover slide with a 32 × 22-mm coverslip and seal edges with rubber cement. 14. Incubate slides at 37°C overnight.
8.6.12 Supplement 5
Current Protocols in Cytometry
Perform post-hybridization washes 15. Prewarm six wash solutions to 42°C in Coplin jars—three consisting of 50 ml of formamide wash solution and three consisting of 50 ml of 2× SSC, pH 7.0. While solutions are warming up, prepare blocking solution. 16. Place slides in 2× SSC to loosen the rubber cement. 17. Remove the rubber cement carefully with forceps and soak slides 5 min in 2× SSC to loosen coverslips, then gently remove the coverslips. 18. Wash slides successively for 5 min each in the three 42°C formamide wash solutions, then successively for 5 min each in the three 42°C 2× SSC wash solutions. Detect biotin label 19. Wash slides 3 min at room temperature in 4× SSCT with agitation. 20. Place slides in blocking solution for 10 to 20 min. 21. Wash slides for 3 min with agitation in 4× SSCT. Wipe backs of slides with tissue but do not allow to dry out. 22. For each slide, mix 1 µl of 5 µg/ml avidin-FITC with 99 µl filtered blocking solution. Apply 100 µl of this solution (layer 1) to chromosome preparation on each slide and place coverslip on top. Incubate 30 min at 37°C. 23. Wash slides successively for 3 min each in three changes of 4× SSCT. Wipe backs of slides with tissue but do not allow to dry out. 24. For each slide, mix 1 µl of 5 µg/ml biotinylated anti–avidin D antibodies with 99 µl filtered blocking solution. Apply 100 µl of this solution (layer 2) to chromosome preparation on each slide and place a coverslip on top. Incubate 20 min at 37°C. 25. Wash slides successively for 3 min each in three changes of 4× SSCT. Wipe backs of slides with tissue but do no allow to dry out . 26. Apply avidin-FITC to slides as in step 22 (layer 3). Wash slides 3 min in 4× SSCT. 27. Wash slides twice, each time for 5 min, in PBS, then dehydrate through ethanol series and air dry as in step 10. 28. Mount slides in 40 µl Citifluor/PI and apply coverslips. 29. View slides on confocal microscope or fluorescence microscope. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Blocking solution 1.8 g bovine serum albumin (BSA) 60 ml 4× SSCT (see recipe) Filter sterilize using a 0.2-µm filter Prepare fresh BSA coating of tubes Incubate tubes overnight with PBS (see recipe) supplemented with 0.1% (w/v) BSA (fraction V; Sigma), remove PBS, and allow tubes to dry. Store coated tubes up to 2 weeks at 4°C.
Molecular Cytogenetics
8.6.13 Current Protocols in Cytometry
Supplement 5
Citifluor/PI 1 ml Citifluor mounting medium (Ted Pella) 8 µl 6.25 µg/ml propidium iodide (PI) stock solution (store stock solution up to 1 year at 4°C; 50 µg/ml final PI concentration) Store Citifluor/PI medium up to 1 week at 4°C in the dark Denaturing solution 35 ml formamide 5 ml 20× SSC (pH 5.3; APPENDIX 2A) 10 ml sterile distilled H2O Prepare fresh Formamide wash solution 25 ml formamide 5 ml 20× SSC (pH 5.3; APPENDIX 2A) 20 ml sterile distilled H2O Prepare fresh Hybridization buffer Prepare in sterile distilled H2O: 50% (v/v) formamide 10% (w/v) dextran sulfate 2× SSC, pH 7.0 (APPENDIX 2A) 1% (v/v) Triton X-100 Store up to 1 year at −20°C Phosphate-buffered saline (PBS), 10× 80 g NaCl 2 g KCl 15 g Na2HPO4⋅2H2O 1.2 g KH2PO4 H2O to 1 liter Adjust pH to 7.4 with 1 M HCl Store up to 1 month at room temperature Polyamine solutions Polyamine solution 1: 0.2 mM spermine (Sigma) 0.5 mM spermidine (Sigma) 50 mM KCl 20 mM NaCl 15 mM Tris⋅Cl, pH 7.2 (APPENDIX 2A) Stable for at least 3 weeks at −20°C Hoechst 33258 staining requires a neutral to alkaline environment.
Polyamine solution 2: Add 1 mg/ml digitonin to polyamine solution 1 and dissolve digitonin by incubating and mixing the solution ∼15 min at 37°C. This solution should be prepared on the same day that it is to be used. CAUTION: Digitonin is very hazardous (the target organ is the heart) and should be handled with great care. It is toxic by inhalation, in contact with skin, and if swallowed.
Advanced Preparative Techniques for Cytogenetic Probes
RNase A solution Dissolve 50 mg RNase A (DNase-free; Boehringer Mannheim) in 5 ml of 2× SSC (APPENDIX 2A). Divide into 500-µl aliquots in microcentrifuge tubes, then put the tubes in a boiling water bath for 5 min (to remove any DNase activity). When cool, store up to 6 months at −20°C.
8.6.14 Supplement 5
Current Protocols in Cytometry
SSCT, 4× Prepare 0.05% (v/v) Triton X-100 in 4× SCC, pH 7 (APPENDIX 2A). Store up to 1 month at room temperature. COMMENTARY Background Information Flow cytometric analysis and sorting Bivariate flow karyotyping is particularly useful when followed by flow sorting of chromosomes for further analysis. Because large quantities of chromosomes can be separated, the technique is very useful for sorting chromosomes to produce probe DNA for use in in situ hybridization and in molecular biology experiments. Moreover, bivariate flow karyotyping is a helpful tool in cytogenetic studies of congenital and other diseases. Chromosome abnormalities can be detected by comparison of the flow karyotypes of the abnormal cells with the flow karyotype of chromosomes derived from normal cells—e.g., stimulated lymphocytes— preferably from the same individual. In the peaks of a bivariate flow karyotype, the relative number of events may, under strictly controlled conditions, be used as a measure of the number of copies of the corresponding chromosome per cell. Aberrations in chromosome number may thus be detected as relative changes in peak volume. Structural chromosome abnormalities related to changes in DNA content or base composition may be detected because they cause shifts in peak positions. Chromosome abnormalities that do not result in changes in DNA content or base composition (e.g., inversions) remain undetectable. The composition of the aberrant chromosome(s) can be determined by sorting chromosomes for subsequent analysis using reverse painting and other FISH techniques. An example of an application in cytogenetics is described by Cotter et al. (1989). Suijkerbuijk et al. (1992), Telenius et al. (1992b), and Boschman et al. (1993) have used the method for identification of marker chromosomes. Initially application of sorting to cytogenetic and molecular genetic studies was limited because of the large amount of chromosomes needed. Sorting of large numbers of chromosomes was only possible using high-speed cell sorters (Gray et al., 1975). However, the introduction of PCR techniques has reduced the number of chromosomes required for DNA amplification to ∼300 and has resulted in the development of new applications in cytoge-
netic and molecular genetic studies that can be performed using commercially available flow cytometers. Examples include: “gene mapping” (the detection of chromosome-specific genes); “chromosome painting” (using suppressive in situ hybridization), which entails the construction of chromosome-specific recombinant DNA libraries; and “post-sorting” identification, performed directly on chromosomes, using FISH or banding techniques (Cremer et al., 1984). The identification of chromosomes using more conventional banding techniques often requires long chromosomes. For this purpose one might consider using the PI technique (see discussion of alternative isolation/staining procedures, below), although this means that the peaks in the flow karyotype representing different chromosomes will be less well-separated. Moreover, there are techniques for elongating chromosomes (also see UNIT 8.2). Rens et al. (1993) describe a technique to elongate chromosomes using trypsin, and Lucas et al. (1991) describe a method using 0.5 M KCl. These techniques, however, require experience to obtain good results; furthermore the elongated chromosomes are more vulnerable to breakage. Slit-scan flow karyotyping, a technique useful in selecting and/or sorting chromosomes on the basis of their morphological features—such as centromeres—also requires long chromosomes (Rens et al., 1993). Alternative chromosome isolation/staining procedures A different technique for the preparation of human chromosomes, also allowing dual staining with Hoechst 33258 and chromomycin A3, is described by van den Engh et al. (1984, 1988). The main difference between this technique and the technique described in Basic Protocol 1 is the use of MgSO4 instead of polyamines to stabilize the chromosomes in suspension. Excellent bivariate flow karyotypes are obtained with this method. However, if the isolated chromosomes are sorted and subsequently used in PCR procedures or spot blotting, the presence of divalent cations may be a disadvantage because they activate DNases. Chromosomes stabilized with Mg2+
Molecular Cytogenetics
8.6.15 Current Protocols in Cytometry
Supplement 5
instead of polyamines will also fall apart when sorted and stored in a buffer containing a chelating agent, such as Tris-EDTA. Another difference is the use of Triton X-100 instead of digitonin for breaking down the cell membrane to release the chromosomes. Digitonin is more effective in disrupting the cell wall. Chromosome isolation techniques as described by Aten et al. (1987) and Cram et al. (1983), which incorporate the use of fluorescent intercalators—e.g., propidium iodide (PI)—for stabilizing the chromosomes, do not allow bivariate analysis because the emission signal of PI interferes with the emission signals from chromomycin A3 and Hoechst 33258. Differentiation between different chromosomes is limited to monovariate flow karyotyping. However, the use of intercalating agents results in chromosomes that are much longer than those stabilized with polyamines or Mg2+ ions. Therefore, this method is used in slit-scan karyotyping (Rens et al., 1994). Sorted chromosomes processed by these methods are also long enough to allow identification using conventional banding techniques.
Advanced Preparative Techniques for Cytogenetic Probes
Microdissection and cloning A number of approaches to microdissection are in use, employing different microinstruments and other pieces of equipment. This unit describes microdissection using an inverted microscope equipped with a rotating stage and a remote-controlled micromanipulator. However, the reader should bear in mind that other methods are also being successfully practiced—such as laser microdissection (Hadano et al., 1991) and microdissection using a microscope that incorporates an oil chamber in which the entire process of microdissection, DNA extraction, digestion with restriction enzymes, and vector ligation is performed (Edstrom et al., 1987). A prerequisite for successful microdissection is a good supply of well-spread chromosomes. This will inevitably speed up the process of chromosome microdissection. A chromosome that is well spread, with “free space” on either side of the target region, allows the operator room to maneuver the microneedle and therefore reduces the risk of contamination from neighboring chromosomes. An abundance of chromosomes on the slide or coverslip saves much scanning time. Techniques used in the cytogenetic laboratory to produce chromosome preparations, although successful and adequate for routine cytogenetic analysis, do cause extensive damage
to the DNA. For purposes of microcloning and genome analysis, the fidelity of the DNA is paramount; hence steps must be taken to minimize the amount of DNA damage. It is now apparent that for generating region-specific probes by chromosome microdissection solely for micro-FISH analysis as described by Meltzer et al. (1992), conventional cytogenetic methods are suitable. Cell-synchronizing agents such as methotrexate, fluorodeoxyuridine, and actinomycin D, which are used to obtain elongated chromosomes, are all potentially damaging to the DNA. Where situations allow, the use of unsynchronized cultures is preferable and provides an adequate number of good quality metaphases. When high-resolution chromosomes are required, the use of thymidine to synchronize the cells is the recommended procedure (also see UNIT 8.2). Thymidine blocks the cells at a specific stage in the cell cycle; when the block is released, the cells resume their cycling in synchrony. These cells are then harvested in early metaphase to yield large numbers of divisions, with long (high-resolution) chromosomes. Prior to chromosome harvest, cells need to be arrested at metaphase using a mitotic-spindle inhibitor. Intercalating agents such as ethidium bromide should be avoided at this stage; although they enhance chromosome extension, they cause nicking of the DNA. Colcemid alone is the standard agent used in most laboratories. A 10-minute incubation with Colcemid prior to harvest yields chromosomes in the early stages of metaphase, which are more extended. Longer incubations with Colcemid, up to 1 hr, will result in a higher yield of metaphase spreads but the chromosomes will be shorter as they become increasingly contracted towards the end of metaphase. Acid causes depurination of the DNA. Chromosomes are routinely “fixed” in metaphase with 3:1 methanol/acetic acid to remove cytoplasmic debris and maintain chromosome morphology. This fixation step is prolonged to give clean preparations of good quality. For purposes of microdissection and microcloning, depurination of DNA is undesirable; hence adaptations to standard fixation procedures are essential to limit the extent of depurination. The chromosome harvesting method described in Basic Protocol 2 incorporates a prefixation in 70% ethanol (Kaiser et al., 1987) followed by a short (10- to 20-sec) fixation in methanol/acetic acid. This method of fixation is not suitable for whole-blood samples, as the length of fixa-
8.6.16 Supplement 5
Current Protocols in Cytometry
tion is insufficient to remove the debris from the red blood cells. In order to obtain chromosomes for microdissection, it is therefore advisable to remove red blood cells prior to culturing using a Ficoll gradient separation technique (UNIT 5.1). Essentially, microdissection can be performed on chromosomes obtained from any viable sample, including both normal and tumor tissue. Samples from which chromosomes are commonly derived include peripheral blood, bone marrow aspirates, lymph nodes, amniotic fluid, chorionic villus, and lymphoblastoid cell lines. The majority of samples can be cultured using standard cytogenetic techniques (Rooney and Czepulkowski, 1992), but as just mentioned, it is advisable to separate the white blood cells from the red blood cells in whole-blood samples before culturing. Chromosomes can also be prepared from frozen samples of patient material that have been viably cryopreserved in liquid nitrogen using DMSO (APPENDIX 3B). Many laboratories routinely store excess patient material in liquid nitrogen tanks for use in future research. An inverted microscope such as a Zeiss Axiovert, mounted firmly on a vibration-free table, is ideal for microdissection. The minimum attachments required include 16× eyepieces, 4× or 6× objectives, and a 63× dry lens (or similar high-power objective). Either bright-field or differential interference contrast (DIC) optics are most suitable (see Chapter 2). To enable correct orientation of the target chromosome region for microdissection, with the chromosome lined up vertically along the y axis and the needle cutting horizontally through the x axis, it is essential to be able to maneuver the coverslip in all directions. This is achieved with a gliding and rotating stage in the center of a fixed stage. The coverslip is placed on the gliding and rotating stage over a central opening with the objective underneath. Once the metaphase spread and target chromosome are selected, the stage can be rotated or dragged to position the chromosome suitably. The micromanipulator mounting points are attached to the fixed section of the stage. A long-workingdistance (LWD) condenser is preferable, to allow adequate room for the micromanipulation. The micromanipulator must be precisely controlled during operation; hence micromanipulators that require direct manual operation are less suitable. Remote-controlled electric or hydraulic micromanipulators are the best types of instrument available; Zeiss manufactures an
electrically controlled unit with extrasensitive movement control. Various additional pieces of equipment can also be included. A video camera, video printer, and television screen are particularly useful for training purposes or for demonstration. Since the first report of chromosome microdissection in 1981 using Drosophila polytene chromosomes (Scalenghe et al., 1981), technical advances, particularly the polymerase chain reaction (PCR) and DNA cloning methods, have established microdissection as a powerful tool in the analysis of the human genome. Microdissection and microcloning, in conjunction with PCR, enable the construction of region- or band-specific genomic libraries, which will aid in building high-resolution maps for the identification of disease-related genes within genomic regions (Zhang et al., 1993). DNA sequence markers are commonly generated from chromosome-specific genomic DNA libraries—derived from somatic cell hybrids, radiation-induced hybrids, or flow-sorted chromosomes—using a variety of techniques. Microdissection and microcloning offer a refinement to the existing methods, enabling much smaller chromosome regions to be studied. Recently the methodology has been finetuned to allow amplification from a single microdissected chromosome (Guan et al., 1993) in the technique known as degenerate oligonucleotide–primed PCR (DOP-PCR). However, one major disadvantage to DOP-PCR is the high risk of contamination (as with most PCR amplifications from small numbers of DNA fragments). Additionally, uneven hybridization to metaphase chromosomes has been reported (Carter et al., 1992; Rack et al., 1993), where the method often fails to paint repetitive sequences in acrocentric short arms, at the centromere, at the telomeres, and in some heterochromatic regions. Another universally applicable group of methods for DNA amplification makes use of restriction endonuclease digestion and DNA ligation directly on the microdissected material prior to amplification. As there is a minimal amount of DNA, microchemical techniques are performed on a nanoliter microdrop contained in an oil chamber. There are two main variations on the overall procedure. In the first, initially used by Lüdecke et al. (1989), the microdissected DNA is digested with the blunt-end restriction endonuclease RsaI. These fragments are then cloned into a SmaI-cut pUC13 vector and amplified by PCR using the plasmid vector sequencing primers. The amplified DNA in-
Molecular Cytogenetics
8.6.17 Current Protocols in Cytometry
Supplement 5
Advanced Preparative Techniques for Cytogenetic Probes
serts are cleaved with a second restriction enzyme (EcoRI) that flanks the cloning site, and are subcloned into a second plasmid vector to generate the microclone library. A second similar method known as linker adaptor PCR (LAPCR) was devised along similar lines by Saunders et al. (1990) and Johnson (1990) and employs ligation of microdissected DNA to linker adaptors rather than to a plasmid vector prior to PCR amplification. The dissected DNA is digested with a frequent-cutting restriction endonuclease (e.g., Sau3AI or MboI), ligated to a 5′ protruding MboI linker adaptor consisting of phosphorylated 24-mer and dephosphorylated 20-mer oligonucleotides, and amplified using the 20-mer DNA as a primer. The PCR products are then digested with MboI to remove the adaptor, and ligated into the BamHI site of a suitable plasmid vector. The procedure has been used on human chromosomes (Kao and Yu, 1991). Several libraries have been constructed using both of these universal DNA amplification PCR methods (Kao, 1993), and these typically contain large numbers of microclones. However, these methods are technically difficult, and involve working with small quantities of DNA in nanoliter microdrops contained in an oil drop. See Ausubel et al. (1998) for details of the general molecular biology techniques mentioned above. The microcloned DNA can be used for contig assembly and high-resolution mapping. Pooled microclones are used to probe and isolate genomic libraries containing larger DNA inserts—e.g., cosmids or YACs. In addition, each microclone can be sequenced and converted to sequence-tagged sites (STSs) to provide physical landmarks within the microdissected region. These STSs would then directly aid the assembly of YACs into an overlapping contig. Pools of microclones can also be hybridized to cDNA libraries to isolate expressed genes as potential candidate genes in disease. Region-specific polymorphic microsatellite probes can also be generated by microdissection using microsatellite primer probes to probe the microclone library. These can subsequently be converted to genetic markers for use in loss-of-heterozygosity or linkage-analysis studies. See Dracopoli et al. (1998) for details of these general techniques in human genetics. Any region of a chromosome can be microdissected—centromeres, telomeres, G-light bands, G-dark bands, and satellites—and the molecular structure and organization of these genomic landmarks can be identified by analyzing microclones from these regions. Mi-
crodissection and micro-FISH are valuable aids in cytogenetic analysis, permitting the characterization of many unresolved cytogenetic aberrations. The origin of cryptic translocations, ring chromosomes, derivative chromosomes, markers, homogeneously staining regions (HSRs), and double minutes (dmins) can be elucidated by micro-FISH. Microdissection has already been successfully employed to identify and generate probes from regions involved in chromosomal rearrangement and in deletions; this approach will assist in the identification of novel genes associated with various diseases. The generation of probes from translocation breakpoints such as the bcr-abl junction (Zhang et al., 1993) will be valuable in the analysis of malignant cells. The microdissection of breakpoint regions was first described by Cotter et al. (1991) who, using genespecific primers for PCR amplification, mapped translocation breakpoints in malignant disease relative to known genes. The isolation of genes underlying inherited and acquired genetic diseases will presumably aid in the diagnosis, prevention, and therapy of these diseases. Several disease loci have already been analyzed by microdissection and microcloning. This microtechnology will surely continue to be a powerful tool in achieving the goals of future research projects.
Critical Parameters and Troubleshooting Flow analysis and sorting Cell culture: A big advantage of flow karyotyping is the excellent definition of the karyotype that is obtained by analyzing a large number of chromosomes. On the other hand, the relatively large number of mitotic cells (∼25,000) needed to obtain a flow karyotype with a high resolution may present a problem for cultured cells other than lymphocytes. The main obstacle is the long cell-cycle period for most human cell types; furthermore, human cell cultures often have a large fraction of cells in G0. Another problem may be the low plating efficiency of some cell lines. In general, it is very hard to obtain sufficient mitotic cells from culturing tumors. For flow karyotyping in tumor genetics, tumor cell lines may provide a better source of chromosomes. An easy way to obtain more mitotic cells is to incubate cells for a long period in medium supplemented with Colcemid. Prolonged incubation of cells in medium with a mitotic blocker yields highly condensed chromosomes, which
8.6.18 Supplement 5
Current Protocols in Cytometry
poses a problem for conventional karyotyping. Fortunately, the length of the chromosomes is not important in bivariate flow karyotyping, so even overnight incubation in medium supplemented with Colcemid is feasible if the right concentration is used. The optimal concentration is one that prevent cells from slipping through mitosis without disturbing the progression of cells that are in other stages of the cell cycle. For long incubation periods, Colcemid is a better choice than alkaloids (e.g., vindesine or vinblastine) because the latter compounds give rise to more cell death. For short incubation periods (≤3 hr), vindesine may be an suitable alternative. Synchronized cell cultures can be used to obtain a large number of mitotic cells. Cell cultures can be synchronized by mitotic shakeoff, by sorting out cells of the same DNA content using a flow cytometer/cell sorter, or by treating the cell cultures with thymidine or hydroxyurea. It is difficult to give a general protocol for these procedures because concentrations and incubation times are largely cell line dependent. In general it is best to do initial cell kinetic experiments using a flow cytometer to evaluate the effect of the method to be used. Quality of the chromosome suspensions: A chromosome suspension should have as few chromosome aggregates and broken chromosomes as possible. The fluorescence distribution of chromosome aggregates overlaps that of larger chromosomes, whereas broken chromosomes mimic the fluorescence of smaller ones. In both cases, a flow karyotype with more background and poor resolution is the result. Aggregates may be the result of poor swelling of mitotic cells. A longer incubation period of the cells in hypotonic solution and/or incubation at 37°C instead of room temperature may help reduce clumping. In addition, a higher concentration of digitonin or more forceful syringing may be needed to break down the cell membrane. Note that syringing the cells is a critical step in the isolation procedure. Syringing more than three times will generally result in a large fraction of broken chromosomes. Inadequate syringing, on the other hand, reduces the amount of chromosomes available for analysis. If many of the cells remain unbroken after syringing three times, a higher concentration of digitonin may be needed. If the cell membranes in a certain cell line are disrupted very easily, it is better to break the cells open by vortexing, instead of syringing them, to prevent unnecessary damage to the chromosomes.
Reproducibility: To use the techniques described in this unit for investigating chromosome abnormalities—e.g., in tumors—one must deal with the problem of variation between flow karyotypes. Variations in Hoechst 33258 and chromomycin A3 fluorescence intensities are caused mainly by variations in the concentration of chemical agents and in chromosome concentration. Because of this, it is advisable to use the same cell densities in the samples to be compared. Furthermore, instrument variation resulting from technical factors such as amplifier offset, orientation and spacing of the laser beams, fluidics, laser power, and alignment of optical elements may also play a role. To reduce the effect of these variations, samples to be compared should be measured on the same day. However, measurement on the same day of both a sample containing aberrant chromosomes and its corresponding control may not always be feasible. For this reason, the authors recommend the method described by Boschman et al. (1992) as a tool for the quantitative comparison of such flow karyotypes. Sorting: One of the factors influencing the purity of a sorted chromosome population is the coefficient of variation (CV) of the peaks in the flow karyotype. High-resolution flow karyotyping requires chromosome suspensions of good quality and a cell sorter that is optimally aligned and that has a stable flow stream. Disappointing flow karyograms, with broad fluorescence distributions in which there are no distinct peaks, can be the result of either a chromosome suspension of bad quality or improper performance of the cell sorter, or, probably, a combination of these two factors. The main problem often is to decide which factors are responsible. Always check the quality of the chromosome suspension under the fluorescence microscope before sorting. It requires some experience to evaluate a chromosome suspension by eye. In short, debris may contribute to the background in the flow karyogram and a large amount of damaged chromosomes or “stringlike” chromosomes may give rise to broad fluorescence distributions. Aggregates of chromosomes can partly block the nozzle or the sample-injection needle, causing an unstable sample stream and therefore a poor resolution. Always check the alignment of the cell sorter using latex beads. Even if the alignment of the cell sorter is optimal and the quality of the chromosome suspension is good, the CV may still be poor. The origin of the troubles may be found in irregularities in the flow stream.
Molecular Cytogenetics
8.6.19 Current Protocols in Cytometry
Supplement 5
Avoiding contamination during DOP-PCR The process of PCR amplification allows the generation of microgram quantities of DNA from only a few microdissected chromosomes that represent femtogram quantities of DNA. This represents an amplification of ∼109-fold. Any contaminating DNA molecule, whether airborne from the laboratory (e.g., plasmid, bacteria, phage, or yeast), or from previously PCR-amplified material, will also be amplified. Therefore stringent procedures must be followed to minimize contamination during every step in the microdissection/microcloning pro-
Cleaning the sample tubing and the nozzle may be helpful. Unstable fluorescence distributions may present another problem, in particular when large amounts of chromosomes are needed. A slowly moving fluorescence peak will result in a sorted chromosome population containing unwanted chromosomes. Incubating the sample tubing with the Hoechst dye (see Support Protocol 1) may help; and usually the fluorescence intensity becomes more stable 15 min after starting the measurement. It has also been reported that keeping the sample in the dark may be important in this respect.
A
2
Hoechst 33258 fluorescence
4
1 3
5 6 7
X 8 13
Y
18 20
21
14 15 16 17
9-12
22 19
B
2
Hoechst 33258 fluorescence
4
1 3
5 6 X 8
7
13 18 Y 20 21
14 9-12 15a 15b 16 17
22 19 Chromomycin A3 fluorescence Advanced Preparative Techniques for Cytogenetic Probes
Figure 8.6.1 Average bivariate flow karyotypes of leukocytes from (A) a normal male donor and (B) human skin fibroblast cells. Contour lines connect channels with equal number of events at 10%, 20%, 40%, 80%, and 90% of the number of events in the peak channel of the chromosome 9 to 12 peak.
8.6.20 Supplement 5
Current Protocols in Cytometry
tocol. Along with the basic precautions normal for PCR (Sambrook et al., 1989), the following procedures are useful to reduce contamination (Kao, 1993): (1) use only sterile disposable pipets, flasks, tubes, and other labware; (2) wear gloves at all stages and change them frequently; (3) prepare buffers and reagents using aseptic techniques and place in sterile tubes for single use only; (4) expose micropipettors, buffers, and reagents to UV light (254 nm) to break down contaminating DNA before use; (5) autoclave all buffers and reagents; (6) use micropipettors dedicated for one operation only; (7) include several control reactions; (8) physically separate pre- and post-PCR manipulations; and (9) minimize the number of manipulations. A certain level of contamination is inevitable, but it can be reduced to manageable levels. Low-level contamination is not effectively a serious problem, as the clones containing human inserts can be selected by hybridization to a human genomic Southern blot. A study (Kao, 1993) demonstrated that almost all microclones shown to be of human origin were derived from the dissected region, presumably because of the high precision of the microdissection technique.
Anticipated Results Flow cytometric analysis and sorting An example of a bivariate flow karyotype is given in Figure 8.6.1. The final resolution will depend on a combination of factors, including the chromosome quality and the alignment of the flow sorter. Chromosome microdissection With care, microdissection will yield a good supply of DNA fragments from a defined chromosomal region. The subsequent DOP-PCR or similar method will allow the generation of fragments suitable for cloning or FISH. Ideally the experiment should be assessed by FISH to demonstrate the origin of the DNA.
erally more practical to do the experiment over 2 days. Furthermore, if larger numbers of sorted chromosomes are required, it will be necessary to spread the sorting over several days or longer. Although the method described is not extremely complicated, some time may need to be invested to gain hands-on experience. The authors recommend the use of fast-growing cells for this purpose. The preparation of chromosome suspensions, as well as the measurement of chromosomes, requires some expertise which may take several weeks to acquire. With experience, the isolation, staining, and measuring of chromosomes will usually take 2 days. In addition, a few hours must be allowed for analysis of the data. The most crucial factor in preparing the chromosomes is the culturing of the cells. If the cell line being used has a very low growth rate, this may present a considerable, time-consuming problem. It may be best, in a preliminary study, to establish a protocol for synchronizing the cells. The amount of sorted chromosomes required will also impose time constraints. It is of course easier to obtain 300 sorted chromosomes for use in a PCR procedure than 20,000 chromosomes for identification using banding techniques, all of which must have good morphology if the latter methods are used. It will be necessary to invest time in optimizing the cell sorter and to gain hands-on experience. Troubleshooting is easier if one is sure that the alignment of the cell sorter is correct and the flow stream is stable. Ideally a cell sorter should be dedicated solely to sorting chromosomes. Chromosome microdissection Microdissection requires considerable expertise both in the recognition of banded chromosomes and in microscopic technique. On the assumption that a trained cytogeneticist is performing the experiments, it may take several days to acquire sufficient fragments for further analysis.
Literature Cited Time Considerations Flow analysis If analysis and/or sorting of small numbers of chromosomes is planned, the whole experiment, from the collection of mitotic cells to the passage of the chromosomes through the flow cytometer, can be performed in 1 day. However, because the unstained chromosomes can be stored overnight at 4°C, it is possible and gen-
Aten, J.A., Buys, C.H.C.M., van der Veen, A.Y., Mesa, J.R., Yu, L.C., Gray, J.W., Osinga, J., and Stap, J. 1987. Stabilization of chromosomes by DNA intercalators for flow karyotyping and identification by banding of isolated chromosomes. Histochemistry 87:359-366. Ausubel, F.A., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., and Struhl, K. (eds.). 1998. Current Protocols in Molecular Biology. John Wiley & Sons, New York.
Molecular Cytogenetics
8.6.21 Current Protocols in Cytometry
Supplement 5
Boschman, G.A., Rens, W., van Oven, C.H., Manders, E.M.M., and Aten, J.A. 1991. Evaluation of variation in Hoechst 33258 fluorescence, chromomycin A3 fluorescence and relative chromosomal DNA content. Cytometry 12:559-569. Boschman, G.A., Manders, E.M.M., Rens, W., Slater, R.M., and Aten, J.A. 1992. Semi-automated detection of aberrant chromosomes in bivariate flow karyotypes. Cytometry 13:469-477. Boschman, G.A., Buys, C.H., van de Veen, A.Y., Rens, W., Osinga, J., Slater, R.M., and Aten, J.A. 1993. Identification of a tumor marker chromosome by flow sorting, DNA amplification in vitro and in situ hybridization of the amplified product. Genes Chromosomes Cancer 6:10-16. Carter, N.P., Ferguson-Smith, M.A., Perryman, M.T., Telenius, H., Pelmear, A.H., Leversha, M.A., Clancy, M.T., Wood, S.L., Cook, K., Dyson, H.M., Ferguson-Smith, M.E., and Willatts, L.R. 1992. Reverse chromosome painting: A method for the rapid analysis of aberrant chromosomes in clinical cytogenetics. J. Med. Genet. 29:299-307. Cotter, F., Nasipuri, S., Lam, G., and Young, B.D. 1989. Gene mapping by enzymatic amplification from flow-sorted chromosomes. Genomics 5:470-474. Cotter, F.E., Lillington, D.M., Hampton, G., Riddle, P., Nasipuri, S., Gibbons, B., and Young, B.D. 1991. Gene mapping by microdissection and enzymatic amplification: Heterogeneity in leukemia-associated breakpoints on chromosome 11. Genes Chromosomes Cancer 3:8-15. Cram, L.S., Bartholdi, M.F., Ray, F.A., Travis, G.L., and Kraemer, P.M. 1983. Spontaneous neoplastic evolution of Chinese hamster cells in culture: Multistep progression of karyotype. Cancer Res. 43:4824-4837. Cremer, C., Rappold, G., Gray, J.W., Muller, C.R., and Ropers, H.H. 1984. Preparative dual-beam sorting of the human Y chromosome and in situ hybridization of cloned DNA probes. Cytometry 5:572-579. Dracopoli, N.C., Haines, J.L., Korf, B.R., Moir, D.T., Morton, C.C., Seidman, C.E., Seidman, J.G., and Smith, D.R. (eds.). 1998. Current Protocols in Human Genetics. John Wiley & Sons, New York. Edstrom, J.-E., Kaiser, R., and Rohmes, D. 1987. Microcloning of mammalian metaphase chromosomes. Methods Enzymol. 151:503-516. Gray, J.W., Carrano, A.V., Steinmetz, M.A., Van Dilla, M.A., Moore, H.H. II, Mayall, B.H., and Mendelsohn, M.L. 1975. Chromosome measurement and sorting by flow systems. Proc. Natl. Acad. Sci. U.S.A. 72:1231-1234.
Advanced Preparative Techniques for Cytogenetic Probes
Guan, X.Y., Trent, J.M., and Meltzers, P.S. 1993. Generation of band-specific painting probes from a single microdissected chromosome. Hum. Mol. Genet. 2:1117-1121. Hadano, S., Watanabe, M., Yokoi, H., Kogi, M., Kondo, I., Tsuchiya, H., Kanazawa, I., Wakasa, K., and Ikeda, J. 1991. Laser microdissection and
single unique primer PCR allow generation of regional chromosome DNA clones from a single human chromosome. Genomics 11:364-373. Johnson, D.H. 1990. Molecular cloning of DNA from specific chromosomal regions by microdissection and sequence-independent amplification of DNA. Genomics 6:243-251. Kaiser, R., Weber, J., Grzeschik, K.-H., Edstrom, J.E., Driese, A., Zengerling, S., Buchwald, M., Tsui, L.C., and Olek, K. 1987. Microdissection and microcloning of the long arm of human chromosome 7. Mol. Biol. Rep. 12:3-6. Kao, F.T. 1993. Microdissection and microcloning of human chromosome regions in genome and genetic disease analysis. BioEssays 15:141-146. Kao, F.T., and Yu, J.W. 1991. Chromosome microdissection and cloning in human genome and genetic disease analysis. Proc. Natl. Acad. Sci. U.S.A. 88:1844-1848. Leif, R.C., Easter, H.N. Jr., Warters, R.L., Thomas, R.A., Dunlap, L.A., and Austin, M.F.A. 1971. Quantitative technique for the preparation of glutaraldehyde-fixed cells for the light and scanning electron microscope. J. Histochem. Cytochem. 19:203-215. Lucas, J.N., Mulliken, J.C., and Gray, J.W. 1991. Dicentric chromosome frequency analysis using slit scan flow cytometry. Cytometry 12:316-322. Lüdecke, H.-J., Senger, G., Claussen, U., and Horsthemke, B. 1989. Cloning defined regions of the human genome by microdissection of banded chromosomes and enzymatic amplification. Nature 338:348-350. Meltzer, P.S., Guan, X.Y., Burgess, A., and Trent, J.M. 1992. Rapid generation of region-specific probes by chromosome microdissection and their application. Nature Genet. 1:24-28. Rack, K.A., Harris, P.C., MacCarthy, A.B., Boone, R., Raynham, H., McKinley, M., Fitchett, M., Towe, C..M., Rudd, P., Armour, J.A.L,. Lindenbaum, R.H., and Buckle, V.J. 1993. Characterization of three de novo derivative chromosomes 16 by “reverse chromosome painting” and molecular analysis. Am. J. Hum. Genet. 52:987997. Rens, W., van Oven, C.H., Stap, J., and Aten, J.A. 1993. Effectiveness of pulse-shape criteria for the selection of dicentric chromosomes by slitscan flow cytometry and sorting. Anal. Cell. Pathol. 5:147-159. Rens, W., Boschman, G.A., Hoovers, J.M., Manders, E.M.M., Slater, R.M., Stap J., and Aten, J.A. 1994. Flow cytometric detection of chromosome abnormalities by measuring centromeric index, DNA content and DNA base composition. Anal. Cell. Pathol. 6:359-375. Rooney, D.E., and Czepulkowski, B.H. 1992. Human cytogenetics: A practical approach. In The Practical Approach Series, 2nd ed. (D. Rickwood and B.D. Hames, eds.) Vol. 1 and 2. Oxford University Press, Oxford.
8.6.22 Supplement 5
Current Protocols in Cytometry
Sambrook, J., Fritsch, E.F., and Maniatis, T. 1989. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Saunders, R.D.C. 1990. Short cuts for genomic walking: Chromosome microdissection and the polymerase chain reaction. BioEssays 12:245248. Scalenghe, F., Turco, E., Edstrom, J.E., Pirrotta, V., and Melli, M. 1981. Microdissection and cloning of DNA from a specific region of Drosophila melangoster polytene chromosomes. Chromosoma 82:205-216. Sillar, R., and Young B.D. 1981. A new method for the preparation of metaphase chromosomes for flow analysis. J. Histochem. Cytochem. 29:7478.
van den Engh, G., Trask, B.J., Lansdorp, P., and Gray, J.W. 1988. Improved resolution of flow cytometric measurement of Hoechst and chromomycin A3 stained human chromosomes after addition of citrate and sulfite. Cytometry 9:266270. van den Engh, G., Hanson, D., and Trask, B.J. 1990. A computer program for analysing bivariate flow karyotypes. Cytometry 11:173-183. Zhang, J., Meltzer, P., Jenkins, R., Guan, X.Y., and Trent, J. 1993. Application of chromosome microdissection probes for elucidation of BCRABL fusion and variant Philadelphia chromosome translocations in chronic myelogenous leukemia. Blood 81:3365-3371.
Key References
Suijkerbuijk, R.J., Matthopoulos, D., Kearney, L., Monard, S., Dhut, S., Cotter, F.E., Herbergs, J., Geurts van Kessel, A., and Young, B.D. 1992. Fluorescence in situ identification of human marker chromosomes using flow sorting and Alu element-mediated PCR. Genomics 13:355-362.
Sillar and Young, 1981. See above.
Telenius, H., Carter, N.P., Bebb, C.E., Nordenskjold, M., Ponder, B.A.J., and Tunnacliffe, A. 1992a. Degenerate oligonucleotide–primed PCR: General amplification of target DNA by a single degenerate primer. Genomics 13:718-725.
Lüdecke et al., 1989. See above.
Telenius, H., Pelmear, A.H., Tunnacliffe, A., Carter, N.P., Behmel, A., Ferguson-Smith, M.A., Nordenskjold, M., Pfragner, R., and Ponder, B.A.J. 1992b. Cytogenetic analysis by chromosome painting using DOP-PCR amplified flow-sorted chromosomes. Genes Chromosomes Cancer 4:257-263. van den Engh, G., Trask, B.J., Cram, S., and Bartoldi, M. 1984. Preparation of chromosome suspensions for flow cytometry. Histochemistry 84:501-508.
Preparative technique for chromosome sorting. Telenius et al., 1992a. See above. DOP-PCR method.
Microdissection and cloning.
Contributed by Jan Stap and Jacob A. Aten University of Amsterdam Amsterdam, The Netherlands D. Lillington, A. Shelling, and B.D. Young St. Bartholomew’s Hospital School of Medicine London, United Kingdom
Molecular Cytogenetics
8.6.23 Current Protocols in Cytometry
Supplement 5
Combination DNA/RNA FISH and Immunophenotyping
UNIT 8.7
Fluorescence-based detection methods allow the simultaneous detection of different DNA or RNA target sequences, together with proteins or other cellular constituents, in the same cell. Using combinations of fluorochromes having different excitation and emission spectra, these different cellular components can be identified on the basis of the different colors assigned to them. This approach is used in so-called genotype/phenotype analysis to identify chromosomal aberrations in subpopulations of cells present in a heterogeneous population. The genetic constitution of these cells is assessed by fluorescence in situ hybridization (FISH; UNIT 8.3) while their phenotypic nature is identified by an immunophenotypic marker. Such a marker can be directed against intracellular or cell-surface antigens. In the last few years, several protocols have been developed allowing simultaneous detection of DNA or RNA sequences together with proteins. Before applying one of these protocols it is important to consider that DNA and RNA molecules differ in chemical nature and in localization within a cell. The consequence of this is that DNA and RNA FISH have different requirements concerning pretreatment of cells and hybridization conditions. Hybridization to DNA sequences requires denaturation of the target DNA, whereas RNA sequences are single-stranded. However, owing to the high structural complexity of many mRNAs, a controlled denaturation of target sequences often improves RNA hybridization signals (Dirks et al., 1993). The genomic DNA is tightly packed within the cell nucleus, so detection of a specific DNA sequence present somewhere within this genome requires measures to obtain full accessibility of the nucleus and removal of cellular components that may cause background signals. Generally, this is achieved by fixation of cells in a fixative that does not cause protein cross-linking, like methanol, followed by removal of cellular RNA by RNase treatment, and removal of proteins by proteinase treatment. In contrast to genomic DNA, RNA sequences are easily lost from a cell due to the action of RNases or simply because they are washed out. Therefore, to retain RNAs, cells are fixed in a protein cross-linking fixative. Because this cross-linking impairs the accessibility of target mRNAs to probe sequences and antibodies, cells need to be pretreated with detergents or proteases. Clearly these conditions for DNA or RNA FISH are not fully compatible with the detection of cellular proteins. Fixation conditions, denaturation, and protease treatments may lead to changes in protein structure, with the consequence that the proteins are no longer detectable by antibodies. In practice, a compromise must be found between the detection sensitivity of DNA or RNA FISH and the preservation of antigenicity of the protein of interest. For the detection of DNA or RNA sequences together with proteins, two approaches exist—i.e., the detection of the protein of interest is performed either before or after the hybridization procedure. The former approach is the simpler, but many proteins are vulnerable to hybridization conditions (including denaturation, pretreatments, and the composition of the hybridization solution) and need to be detected with antibodies before the FISH procedure. Therefore, to preserve the antibodies bound to the protein, additional fixation steps are often required, further impairing the accessibility of the cell for probe sequences. Furthermore, antibody solutions contain RNases, which cause loss of cellular RNA. To deal with these problems, this unit describes protocols that allow simultaneous detection of DNA or RNA sequences. Molecular Cytogenetics Contributed by Roeland W. Dirks Current Protocols in Cytometry (1998) 8.7.1-8.7.14 Copyright © 1998 by John Wiley & Sons, Inc.
8.7.1 Supplement 6
Basic Protocol 1 describes a slide-based method for genotype/phenotype analysis. Identification of cells is generally achieved by immunocytochemical staining of a specific membrane or cytoplasmic antigen, while chromosome abnormalities are detected by DNA FISH. Because membrane markers are often vulnerable under hybridization conditions, these are detected first, before starting the FISH procedure. The protocol described here has successfully been applied to peripheral blood cells, bone marrow cells, and cultured cell types. For a description of the standard DNA FISH procedure and the procedure for labeling probes, the reader is referred to UNIT 8.3. Genotype/phenotype analysis provides information about the genetic constitution and identity of cells in a heterogeneous population. It does not, however, provide information about which genes are expressed in these cells—except those that encode the proteins used for identification of the cells—and at what abundance expression takes place. For this purpose an RNA FISH procedure has to be combined with immunophenotyping. Combined detection of RNAs and proteins is also used to study basic cell-biological processes like the interaction of different components involved in gene transcription, RNA processing, and translation (Xing et al., 1995; Huang and Spector, 1996; Dirks et al., 1997). RNA FISH is a microscopic technique used to monitor gene expression at the individual cell level. It involves hybridization of hapten-, fluorochrome-, or enzyme-labeled DNA or RNA probes to RNA target sequences present in cells or tissue sections, followed by the microscopic visualization of the hybridized probes. This technique for visualizing and localizing specific mRNAs in single cells has been made possible by the development of various nonradioactive labeling procedures for nucleic acid probes, the sensitivity of immunocytochemical detection systems, and the availability of advanced types of fluorescence microscopes (Lawrence et al., 1989; Dirks et al., 1993). Earlier techniques based on radioactive detection methods offered high detection sensitivity but did not allow high-resolution detection of gene transcripts; the limited resolution provided by autoradiography easily leads to misinterpretation of results, because in a heterogeneous cell population it is often difficult to trace back to which cell silver grains belong. Basic Protocol 2 describes the simplest approach to combining the detection of an mRNA species with detection of a protein. The in situ hybridization procedure is performed first, followed by the combined immunocytochemical detection of the labeled probe and the protein of interest. In a two-layer immunocytochemical detection system, the two primary antibodies—one for the probe and the other for the protein—are combined in the first layer, and the secondary antibody conjugates (with different fluorochromes) are mixed in the second antibody layer. It is important that the primary antibodies be raised in different animal species to avoid cross-reactivity of the secondary antibodies. Furthermore, this approach can be followed only when the antigenic activity of the protein is still intact after the in situ hybridization procedure. If this is not the case, as for most membrane markers, the protein should be detected first, followed by the mRNA (see Alternate Protocol).
Combination DNA/RNA FISH and Immunophenotyping
8.7.2 Supplement 6
Current Protocols in Cytometry
COMBINED DETECTION OF DNA AND PROTEIN Cells in suspension are first fixed in formaldehyde to preserve cellular morphology and then centrifuged onto microscope slides. The slides are incubated with specific antibodies for the immunocytochemical detection of a membrane or cytoplasmic marker. Cells are then post-fixed in formaldehyde and treated with pepsin to make them accessible for probe sequences. This protocol can be followed by a standard DNA-FISH procedure (see UNIT 8.3) to visualize the genetic constitution.
BASIC PROTOCOL 1
Materials 2% and 4% formaldehyde fixative (see recipe) Cell suspension: 6 × 106 cells/ml in 1× PBS (see UNIT 8.2 for PBS) 1× PBS (UNIT 8.2) containing 1% (w/v) BSA 1:1 (v/v) ethanol/ether Tris-buffered saline (TBS; APPENDIX 2A) TBS blocking buffer (see recipe) Primary antibodies: dilute cell type–specific antibody according to manufacturer’s instructions in TBS blocking buffer (see recipe for TBS blocking buffer) and microcentrifuge 3 min at 10,000 × g Secondary antibody: dilute hapten- (e.g., biotin)-labeled antibody in TBS blocking buffer according to manufacturer’s instructions (see recipe for TBS blocking buffer) and microcentrifuge 3 min at 10,000 × g Phosphate-buffered saline (1× PBS; UNIT 8.2) 0.1% pepsin (Sigma) in 0.01 M HCl (pH 2.0), 37°C 70%, 90%, and 100% ethanol 10-ml polypropylene centrifuge tubes Mechanical shaker Dust-free Kimwipes Cytospin 2 tabletop cytocentrifuge (Shandon/Lipshaw) Microscope slides 24 × 60–mm and 50 × 60–mm coverslips Schiefferdecker staining jars (e.g., Fisher) Moist chamber: 1-liter beaker containing paper towels moistened with water, covered with aluminum foil Additional reagents and equipment for cell counting (APPENDIX 3A), DNA FISH (UNIT 8.3), and probe visualization (UNIT 8.5) Fix cells 1. Add 500 µl of 4% formaldehyde fixative, slowly, with continuous stirring, to 500 µl of cell suspension containing 3 × 106 cells in a 10-ml polypropylene centrifuge tube. Fix cells for 15 min at room temperature while gently shaking continuously on a mechanical shaker. The choice of fixation method can be dependent on the cell type and the antigens to be preserved. Some antigens will lose their antigenicity when fixed in formaldehyde, so in that case other methods of fixation should be tested. Good alternatives might be 3:1 (v/v) methanol/acetone or methanol alone.
2. Add 1 ml 1× PBS/1% BSA, centrifuge 8 min at 350 × g, room temperature, remove the supernatant, and resuspend the cells in 1 ml 1× PBS/1% BSA, by breaking up the pellet. Repeat twice for a total of three washes. Finally, resuspend cells in 1 ml of 1× PBS/1% BSA, count the cells in a cell counter (APPENDIX 3A), and adjust the volume to achieve a final cell concentration of 1 × 106/ml. Molecular Cytogenetics
8.7.3 Current Protocols in Cytometry
Supplement 6
Make cytospin preparations 3. Clean microscope slides with 1:1 ethanol/ether and dry them with dust-free Kimwipes. 4. Set up the cytocentrifuge bucket with the microscope slide and filter paper, add 200 µl of 1× PBS/1% BSA to the bucket, and precentrifuge 5 min at 350 × g in a Cytospin 2 cytocentrifuge at room temperature. Add 200 µl 1× PBS/1% BSA and 50 µl of the 1 × 106 cell/ml suspension to the bucket, and again centrifuge 5 min at 350 × g. 5. Dismantle the cytocentrifuge bucket, remove the slide, and air dry. Judge the quality of the preparations using a phase-contrast microscope. Cells should be free from one another. Slides can be sealed in plastic zip-lock bags (one slide per bag) and stored at −20°C.
Incubate slides with cell type–specific antibody 6. If stored at −20°C, allow slide to adjust to room temperature while still sealed. 7. Immerse slide in TBS for 15 min, drain off excess fluid, then apply 120 µl TBS blocking buffer to the area of the slide containing the cells. Cover with a 24 × 60–mm coverslip. Place slide upside-down in a horizontal position in a staining jar and preincubate 30 min at room temperature in a moist chamber to reduce nonspecific antibody binding. 8. Add TBS and release coverslip by gently shaking. Drain off as much TBS as possible and apply 120 µl of the appropriate dilution of cell type–specific antibody in TBS blocking buffer (microcentrifuged 3 min at 10,000 × g prior to application) to the area of the slide containing the cells. Cover with a 50 × 60–mm coverslip and incubate 45 min at room temperature. Since each cell type has a unique set of marker antigens the antibody should be specific for the cell type to be identified. Microcentrifugation of antibody solutions prior to application helps to reduce immunocytochemical background staining.
9. Wash slide three times, each time by immersing for 10 min in TBS. 10. Apply the second antibody layer, consisting of 120 µl of the appropriate dilution of hapten- (e.g., biotin)-labeled secondary antibody in TBS blocking buffer (microcentrifuged 3 min at 10,000 × g prior to application), to the slide. Cover with a 24 × 60–mm coverslip and incubate 45 min at room temperature. 11. Wash cells three times, each time by immersing for 10 min in TBS, then once by immersing for 10 min in 1× PBS. Pretreat slides for FISH, perform FISH, and detect signals 12. Post-fix slide by immersing for 2 min in 2% formaldehyde fixative. Wash slide twice, each time by immersing for 3 min in 1× PBS. This fixation step will prevent loss of bound antibodies during pepsin treatment, denaturation, and probe hybridization.
13. Immerse slide in 0.1% pepsin solution, prewarmed to 37°C. Incubate 1 min at 37°C. Discard pepsin solution and wash briefly in 1× PBS. 14. Post-fix and wash slide again as in step 12. Combination DNA/RNA FISH and Immunophenotyping
8.7.4 Supplement 6
Current Protocols in Cytometry
15. Dehydrate by immersing slide successively for 3 min each in 70%, 90%, and 100% ethanol. Air dry and proceed with DNA FISH (UNIT 8.3). Note that when a biotin-labeled secondary antibody has been used in step 10, do not include a biotin-labeled DNA probe in the FISH procedure, to avoid cross-reaction.
16. After completing the DNA hybridization, visualize the FISH probes and the secondary antibody indicating the cell marker simultaneously using a combination of fluorescent-labeled antibodies (UNIT 8.5). COMBINED DETECTION OF mRNA AND PROTEIN In this protocol, the RNA FISH procedure is performed prior to the immunocytochemical detection of the protein. The order for RNA and protein detection is different from that of DNA and protein detection because antibody solutions contain RNases that lead to degradation of target RNA sequences, and as a consequence to loss of hybridization signal. However, this protocol can be followed only when the cellular protein of interest does not lose its antigenicity during the FISH procedure. Whether this is the case can only be assessed experimentally. Cells are grown on or centrifuged onto microscope slides, fixed, and pretreated with pepsin. This pretreatment is followed by hybridization with a labeled nucleic acid probe and stringent washes. Antibodies to detect the protein of interest and hybridized probe are mixed in the first antibody incubation buffer.
BASIC PROTOCOL 2
Materials Cells of interest Culture medium (e.g., complete DMEM/10% FBS; see APPENDIX 2A) without phenol red Phosphate-buffered saline (1× PBS; UNIT 8.2) Formaldehyde/acetic acid fixative (see recipe) 70%, 90%, and 100% ethanol 0.1% pepsin (Sigma) in 0.01 M HCl (pH 2.0), 37°C 4% formaldehyde fixative (see recipe; optional) Hybridization mix (see recipe) containing nick-translated probe and (optionally) without probe 50% formamide/2× SSC (see APPENDIX 2A for SSC), pH 7.0 (37°C) 2× SSC (APPENDIX 2A) Tris-buffered saline (TBS; APPENDIX 2A) Primary antibodies: against hapten used to label probe and against cell-specific protein of interest, obtained from different animal species TBS blocking buffer (see recipe) Secondary antibodies conjugated to fluorochrome or horseradish peroxidase Mounting medium (e.g., Vectashield from Vector Labs) Microscope slides 180°C oven Sterile 15 × 15–cm petri dishes 37°C humidified 5% CO2 incubator Hettich-Universal tabletop cytocentrifuge with no. 1323 rotor and compatible buckets (no. 1266, no. 1271, or no. 1276) for depositing cells on slides (Hettich-Zentrifugen) Staining jars 18 × 18–mm and 24 × 60–mm coverslips Moist chamber: 1-liter beaker containing paper towels moistened with 50% formamide, sealed with Parafilm and aluminum foil Metal plate heated to 80°C in oven
Molecular Cytogenetics
8.7.5 Current Protocols in Cytometry
Supplement 6
37°C shaking water bath Slide boxes Fluorescence microscope equipped with appropriate filter sets for red, green, and blue excitation, high-NA Plan Apo lenses, and HBO 100-W mercury arc lamp (UNIT 2.4) Camera system with ISO 640 color slide film (e.g., 3M) or CCD camera system and computer running image analysis system and Adobe Photoshop (UNIT 2.5) Additional reagents and equipment for immunocytochemical detection (UNIT 8.4) CAUTION: Formaldehyde and formamide are hazardous and should be handled with care (see manufacturer’s instructions). NOTE: All buffer solutions should be autoclaved. Prepare and fix cells 1a. To grow cells on slides (for adherent cells): Sterilize glass microscope slides by baking in a 180°C oven. Place sterile glass slides in sterile 15 × 15–cm petri dishes containing 20 ml culture medium and add cells. Culture cells to subconfluency at 37°C in a humidified 5% CO2 incubator. Most adherent cell types grow on glass without any coating. If required, cell adherence can be improved by using poly-L-lysine–coated microscope slides (UNIT 8.2).
1b. To centrifuge cells onto slides (nonadherent cells): Dilute cells to a concentration of 10,000 to 25,000 cells/ml in 1× PBS or culture medium. Set up the cytocentrifuge bucket with microscope slide, add 1 ml of the cell suspension, then centrifuge 5 min at 350 × g (1200 rpm in Hettich-Universal cytocentrifuge). In this way the cells are positioned at defined areas on the slides.
2. Place microscope slide with cells in a staining jar, add 100 ml of 1× PBS, and shake gently for 1 min. To minimize background staining caused by serum proteins fixed to the slide, it is necessary to remove culture medium by washing as described in this step, before fixation of the cells.
3. Pour off PBS and add formaldehyde/acetic acid fixative. Fix cells for 20 min at room temperature. This fixation time has been optimized for fibroblasts. Other cell types may require shorter or longer fixation times. The authors have observed that the addition of acetic acid to the fixative improves accessibility for probe and antibodies (Dirks et al., 1993).
4. Discard fixative and add 1× PBS. Wash for 2 min with gentle shaking. Replace PBS with fresh 1× PBS and wash another 2 min. 5. Discard PBS and add 70% ethanol. Wash for 2 min and replace with fresh 70% ethanol. At this stage cells can be stored in 70% ethanol at 4°C for up to several weeks.
Pretreat and hybridize cells 6. Rehydrate cells by immersing slide 2 min in 1× PBS. 7. Immerse slide in 0.1% pepsin in 0.01 M HCl, prewarmed to 37°C. Incubate 1 to 3 min at 37°C without shaking. Combination DNA/RNA FISH and Immunophenotyping
In general a balance must be found between optimal accessibility of target sequences for probe and antibodies and maintenance of cell morphology. Therefore, the pepsin concen-
8.7.6 Supplement 6
Current Protocols in Cytometry
tration and time of incubation have to be optimized for each cell type. In general, pepsin concentrations <0.1% give little improvement in accessibility.
8. Discard pepsin solution and add distilled water. Wash slide 1 min with gentle shaking, replace water with 1× PBS, and wash an additional 1 min. 9. Optional: Post-fix the cells by immersing slide for 10 min in 4% formaldehyde fixative at room temperature to preserve cell morphology. Wash slide three times, each time by immersing for 3 min in 1× PBS. 10. Dehydrate by immersing slide successively for 3 min each in 70%, 90%, and 100% ethanol. Air dry and use immediately for hybridization. 11. Optional: Boil hybridization mix (without probe) 3 min and chill in ice bath. Apply 120 µl to area of slide containing cells and cover with a 24 × 60–mm coverslip. Incubate 1 hr at 37°C in a moist chamber. To remove coverslip, rinse slide 5 min at room temperature in 2× SSC in a staining jar while gently shaking, then transfer the slide to a new jar with 2× SSC. Finally, dehydrate as in step 10. This optional prehybridization step may help to reduce noise signals.
12. Apply 10 µl hybridization mix containing nick-translated probe on a slide and cover with a 18 × 18–mm coverslip. Denature probe and target sequences simultaneously by placing the slide for 3 min on a metal plate heated to 80°C in an oven. Transfer slide to an empty staining jar prewarmed to 37°C (or other appropriate temperature; see below), then place the jar in a moist chamber containing paper towels moistened with 50% formamide (keep slide in horizontal orientation) and hybridize 2 hr to overnight at 37°C (or other appropriate temperature; see below). Two different hybridization mixtures (probes) can be applied to a slide, one on the left side and the other on the right side of the slide. Make sure that the two coverslips cannot touch one another. RNA probes are hybridized at 42°C and oligonucleotide probes at temperatures ranging from room temperature to 37°C depending on the length. Generally, oligonucleotides ranging in size from 18 to 25 bp are hybridized in 25% formamide/3× SSC at room temperature and those ranging in size from 25 to 50 bp in 25% to 50% formamide at 37°C. To prevent nonspecific hybridization signals, it is important to carefully optimize the stringency conditions for hybridization. Because HRP activity is destroyed at high temperature and in formamide, HRP-labeled oligonucleotides are not denatured together with target sequences and are preferably hybridized at room temperature in at most 30% formamide. It has repetitively been observed that denaturation of the sample results in improved detection sensitivity, probably by increasing the accessibility of target sequences. It should be noted that nuclear DNA sequences will also be denatured and may become visible after the hybridization. If required, simultaneous denaturation of probe and target sequences can be omitted. In this case the hybridization mix should be heated to 80°C for 5 min in a water bath, chilled on ice, and applied to slides just before hybridization.
Perform post-hybridization washes 13. Add 50% formamide/2× SSC, pH 7.0, 37°C, to the jar and allow coverslip to detach while gently shaking. 14. Transfer slide (without coverslip) to another jar filled with 50% formamide/2× SSC, pH 7.0, 37°C, and wash with gentle shaking for 10 min (see below for variations depending on probe used). Discard the solution and wash an additional 10 min with 50% formamide/2× SSC, pH 7.0, at 37°C. Molecular Cytogenetics
8.7.7 Current Protocols in Cytometry
Supplement 6
Optionally, to reduce nonspecific hybridization of RNA probes, slides are washed in 60% formamide/2× SSC at 45°C and incubated for 20 min with RNase A (UNIT 8.2). When oligonucleotide probes are used, the stringency of the washing conditions should be adjusted to the length of the probe. For example, 20-bp oligonucleotides can be washed in 2× SSC alone, at room temperature
15. Wash 3 min in 2× SSC at room temperature and finally for 3 min in TBS. 16a. For fluorochrome-labeled probes: Dehydrate and air dry slide as in step 10. 16b. For other types of probes: Omit dehydration and proceed with immunocytochemical detection. Perform combined immunocytochemical detection of probe and protein 17. Dilute the primary antibodies (against the hapten used to label the probe and against the cell-specific protein of interest) together in TBS blocking buffer, and microcentrifuge 3 min at 10,000 × g, room temperature. Apply 120 µl of the antibody mixture to the area of the slide containing the cells and cover with a 24 × 60–mm coverslip. Insert slide in a staining jar and incubate 1 hr in a moist chamber at 37°C. The choice of the primary antibodies will depend on the haptens and proteins to be detected. It is essential that the primary antibodies be raised in different animal species and and that they do not cross-react with each other. Such problems can be avoided when one probe is labeled with digoxigenin and detected with an anti-digoxigenin antibody and the other probe is labeled with biotin and detected with a streptavidin conjugate. Microcentrifugation of antibody solutions prior to application helps to reduce immunocytochemical background staining.
18. Wash slide 3 min in TBS with gentle shaking at room temperature to loosen coverslip, then transfer slide to a new staining jar filled with TBS and wash 5 min at room temperature with gently shaking. Discard washing solution, add TBS, and wash for another 5 min. 19. Dilute fluorochrome- or HRP-conjugated secondary antibodies, apply them to the slide, incubate as described in step 17, then wash the slide as described in step 18. 20. For HRP-labeled probes or secondary antibodies: Visualize HRP-labeled probes or antibodies using a tyramide-based detection system (UNIT 8.4). Mount specimen and examine microscopically 21. Apply 30 µl Vectashield or other appropriate antifade mounting medium to area of slide containing cells and cover with a 24 × 60–mm coverslip. Store in a slide box at 4°C. If required, nuclei can be counterstained with a general DNA counterstain like 4′,6-diamidino-2-phenylindole⋅2HCl (DAPI; for UV exitation). Different DNA counterstains and protocols are described in UNIT 8.4.
22. Examine slide using fluorescence microscope. Dual- and triple-bandpass filters can be used for simultaneous examination of multiple fluorochromes.
23. Make photomicrographs using 640 ISO color slide film. Alternatively, record images using a CCD camera connected to a computer and process images using an image analysis program and Adobe Photoshop. Combination DNA/RNA FISH and Immunophenotyping
8.7.8 Supplement 6
Current Protocols in Cytometry
DETECTION OF A PROTEIN FOLLOWED BY DETECTION OF mRNA SPECIES
ALTERNATE PROTOCOL
This protocol can be followed if a protein cannot be detected after the in situ hybridization procedure, owing to loss of antigenicity. Cells are grown on or centrifuged onto microscope slides, fixed, and incubated with a primary antibody directed against the protein of interest. This incubation is followed by a second one with a biotinylated secondary antibody. Cells are then post-fixed and treated with pepsin to increase accessibility of probe sequences. From this point one can proceed with a standard RNA-FISH protocol. Additional Materials (also see Basic Protocol 2) PBS/DEPC (see recipe) Blocking reagent for nucleic acid hybridization (Boehringer Mannheim) RNasin ribonuclease inhibitor (Promega) 2% formaldehyde fixative (see recipe) 1. Grow or centrifuge cells onto microscopic slide and fix in formaldehyde/acetic acid fixative 20 min at room temperature (see Basic Protocol 2, steps 1 to 3). The antigenicity of some proteins may be affected by acetic acid. In this case fix the cells in 4% formaldehyde in 1× PBS, without the acetic acid.
2. Wash slide in PBS/DEPC 3 min at room temperature with gentle shaking. 3. Dilute the primary antibody against the protein to the appropriate concentration in PBS/DEPC and add 0.5% (w/v) blocking reagent and 1 U/µl RNasin. Microcentrifuge the dilution 3 min at 10,000 × g. It is essential to add an RNase inhibitor to the antibody solution because most antibodies are not free of RNase activity. A preincubation with blocking reagent to avoid nonspecific antibody binding is is omitted in this procedure, to minimize exposure to RNase-containing substances.
4. Apply 120 µl of the antibody solution to the area of slide containing the cells. Cover with a 24 × 60–mm coverslip, insert the slide into a staining jar, and incubate 1 hr in moist chamber at 37°C. 5. Wash slide three times, each time by immersing for 5 min in PBS/DEPC. Prepare fluorochrome- or HRP-conjugated secondary antibody as described for the primary antibody in step 3 and apply and incubate as in step 4. HRP-conjugated antibodies are visualized using fluorochrome-conjugated tyramides (UNIT 8.4).
6. Wash slide three times, each time by immersing for 5 min in PBS/DEPC. Post-fix in 2% formaldehyde fixative for 5 min. 7. Wash slide twice, each time by immersing for 3 min PBS/DEPC. Incubate slide for 1 min in 0.1% pepsin solution prewarmed to 37°C. 8. Continue with step 8 of Basic Protocol 2.
Molecular Cytogenetics
8.7.9 Current Protocols in Cytometry
Supplement 6
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Formaldehyde/acetic acid fixative 75 ml H2O 10 ml 37% formaldehyde (Merck; acid-free and stabilized with 10% methanol and CaCl2) 5 ml glacial acetic acid 10 ml 10× PBS (UNIT 8.2) Prepare just before use Formaldehyde fixative, 2% and 4% 1× PBS (UNIT 8.2) containing 2% or 4% (v/v) formaldehyde (from 37% stock, acid-free and stabilized with 10% methanol and CaCl2; Merck). Prepare fresh. Hybridization mix Add the following to a microcentrifuge tube containing 1 µg labeled, ethanol-precipitated, dried double-stranded DNA probe (UNIT 8.3; results in mix containing 5 ng/µl probe): 10 µl H2O 20 µl 20× SSC (APPENDIX 2A; 2× final) 20 µl 0.5 M sodium phosphate, pH 7 (UNIT 8.3; 50 mM final) 100 µl deionized formamide (APPENDIX 2A; 50% final) 40 µl 50% dextran sulfate (UNIT 8.3; 10% final) 5 µl 10 mg/ml yeast tRNA 5 µl 10 mg/ml herring sperm DNA Store ≥1 year at 4°C The above proportions are for double-stranded DNA probe. Double-stranded DNA probe can be labeled by nick translation or random priming (see UNIT 8.3 for both procedures) using haptenized or fluorochrome-labeled nucleotides. RNA probes can be synthesized from a transcription vector using an RNA polymerase and haptenized nucleotides (Höltke and Kessler, 1990). When RNA probes are used for hybridization, raise the percentage of formamide in the hybridization mix to 60% and lower the probe concentration 10-fold. Ideally, probe-fragment size should range between 100 and 300 bp. Oligonucleotide probes are labeled enzymatically by terminal transferase using labeled nucleotides or chemically via a reactive amino group with a fluorochrome, hapten, or horseradish peroxidase (HRP) label. Depending on the melting temperature (Tm) of the oligonucleotide probe, the formamide concentration in the hybridization mix must be adjusted.
PBS/DEPC Add 2 ml diethylpyrocarbonate (DEPC) to 1 liter of 1× PBS (UNIT vigorously and autoclave.
8.2).
Shake
CAUTION: DEPC is a suspected carcinogen and should be handled with care.
TBS blocking buffer Dissolve 0.5 g blocking reagent for nucleic acids (Boehringer Mannheim) in 100 ml TBS (APPENDIX 2A) by stirring for 1 hr at 60°C. Store in 10-ml aliquots ≥1 year at −20°C.
Combination DNA/RNA FISH and Immunophenotyping
8.7.10 Supplement 6
Current Protocols in Cytometry
COMMENTARY Background Information Combined immunophenotyping and FISH Combined genotype/phenotype analysis provides information about the genetic composition and identity of cells present in a heterogeneous population. This type of information has diagnostic and prognostic value for tumor cytogeneticists and hematologists studying blood cancers. Additional information concerning the expression levels of genes of interest can be obtained by combining an RNAFISH procedure with immunophenotyping. These techniques can be applied to cell preparations, and in principle also to tissue sections (Weber-Matthiesen et al., 1993). The difficulty with tissue sections is, however, the high level of autofluorescence that is often encountered. Furthermore, it should be noted that due to the large variety in immunophenotypic markers and cell types, it is impossible to provide a general protocol applicable to all of them. Therefore, in practice, different types of fixation and possibly preparation techniques should be tested. The number of reports describing combined immunophenotyping and and FISH is increasing, however, and valuable practical information can be found there for specific applications (e.g., Speel et al., 1994; Hessel et al., 1996; Litle et al., 1996). RNA FISH The RNA-ISH technique is a multistep procedure in which the different steps are interdependent. In general these steps need to be optimized for each specific application. The protocol described here can therefore serve as a starting point but not necessarily as the most optimal protocol for each cell type or RNA species. Basic Protocol 2 has been developed for broad application in cell biology, which means that this procedure can be used for detecting RNA sequences present in the cytoplasm as well as in the nucleus of cells, and for the combined detection of RNA sequences and cellular proteins (Xing et al., 1995; Huang and Spector, 1996; Dirks et al., 1997). RNA detection can be achieved by using different types of probes and labels. Singlestranded oligonucleotide and RNA probes have the advantage that they do not renature during the hybridization reaction as do doublestranded DNA probes. RNA probes are frequently used in RNA-ISH studies because they
generally provide a high detection sensitivity. There are, however, also a number of drawbacks associated with the use of this probe type. First, probe sequences need to be subcloned in an RNA transcription vector, or alternatively PCR products need to be synthesized using primers with an RNA polymerase recognition sequence. Second, RNA probes are more prone to nonspecific hybridization, especially to rRNA sequences. Finally, RNA probes often require fragmentation to generate an optimal probe size. Oligonucleotide probes can more efficiently penetrate cells due to their short size, and have the advantage that they can be very specific for a target sequence, which is particularly useful when the target sequence has a great extent of sequence homology with other RNAs. Due to the limited number of labels that can be attached to an oligonucleotide, this probe type provides a lower detection sensitivity as compared with other probe types. Good results are generally obtained using double-stranded plasmid DNA probes labeled by nick translation. The number of labels that can be incorporated and the resulting probe size provide a high detection sensitivity. At present there are a large variety of labels that can be incorporated into probes. Directly fluorochrome-conjugated probes provide the advantage that no additional immunocytochemical detection steps are required after the hybridization procedure, minimizing both the time needed for the procedure and background staining. Hapten-labeled probes in conjunction with immunocytochemical detection steps, however, provide a greater detection sensitivity. Superior RNA detection sensitivity can be obtained by using peroxidase-labeled probes or antibodies and a tyramide-based detection method (Raap et al., 1995; UNIT 8.4). Potential disadvantages of this method are that it can not be used in colocalization studies and in studies that require ultimate spatial resolution.
Critical Parameters The most critical parameters in the combined immunotyping and FISH procedure are the cell preparation (sufficient cells should remain on the microscope slide during the procedure), choice of fixation method, extent of pepsin treatment, and the quality and properties of the antibodies used for immunophenotyping. These and other factors are discussed below, along with critical parameters for RNA-ISH.
Molecular Cytogenetics
8.7.11 Current Protocols in Cytometry
Supplement 6
Cell culture Best results are generally obtained when cells are grown to subconfluency in a single layer. Multiple layers of cells easily lead to nonspecific staining. It is important to wash away the culture medium prior to fixation to avoid nonspecific fluorescence signals. Ideally, the use of pH indicators in the culture medium should be limited. Fixation One of the most critical steps in the RNAISH procedure is the fixation of the cells, which is required to maintain cell morphology and to retain RNA target sequences. The most frequently used fixative is formaldehyde, a crosslinking fixative that can either be prepared fresh or purchased as a 37% solution. When a commercial solution is used, make sure that it is stabilized. Many alternative fixation methods have been explored, and a fixative consisting of 4% formaldehyde/5% acetic acid has proved to be beneficial in some aspects, especially when nuclear RNA targets have to be detected (Dirks et al., 1993). Most cell types are properly fixed in 15 to 20 min, although optimal fixation time may vary for different cell types. Glutaraldehyde, a strong cross-linking fixative, is generally not used in light-microscopical ISH studies, but finds application in electron-microscopical ISH studies. Owing to its cross-linking potential, glutaraldehyde preserves cell morphology best, but prevents optimal penetration of probe sequences and antibodies. Fixatives entirely based on ethanol, methanol, or acetone do not sufficiently retain RNA sequences in cells and are therefore not recommended (Singer et al., 1986; Dirks et al., 1993).
Combination DNA/RNA FISH and Immunophenotyping
Pretreatments To improve the accessibility of cells for probes and detection agents, cell membranes and proteins are extracted. Membranes are permeabilized by the various ethanol washes and proteins are digested by proteinases. The most consistent results are obtained with pepsin, although other proteinases such as proteinase K and pronase have been successfully used. The amount of pepsin and time of incubation need careful optimization for each different cell type. In addition to improving the accessibility of cells, pepsin treatment may also reduce background staining. Hybridization signals may further be improved by denaturing the target RNA, which is done by heating the cells together with probe sequences at 80°C. This step most likely results
in denaturation, unfolding of RNA molecules, and removal of proteins attached to them. Optimal time of denaturation may vary between cell types, and long denaturing times will severely affect cell morphology. Stringency conditions for hybridization and washing For optimal signal-to-noise ratios, it is important to adjust the stringency of the hybridization reaction rather than the washing steps. Generally, hybridization in 50% to 60% formamide at 37°C provides sufficient stringency for DNA plasmid or PCR probes. Higher-stringency conditions can be achieved by raising the temperature to 42°C without affecting cell morphology too much. RNA probes generally require high stringency conditions and are hybridized in up to 70% formamide at 45°C. Optimal stringency conditions for oligonucleotide probes should be established empirically and depend on the length and base composition. Detection High-sensitivity RNA detection can be achieved only with good-quality probes and immunocytochemical detection agents. Antibody conjugates should be of superior quality, with optimal fluorescence/protein ratios. Furthermore, it is of importance to optimize antibody dilutions to reduce the amount of nonspecific staining. Controls Results of RNA-ISH experiments can only be interpreted when proper controls are incorporated. An RNase-treated sample is an essential negative control, but is not sufficient, as it will reveal only whether or not the signals are due to staining of cellular components other than RNA. Therefore, it is necessary to incorporate hybridizations with various control probes, which may include sense RNA or oligonucleotide probes and probes that are known not to be expressed in the cells of interest. In addition, the probe of interest should be hybridized to related cell types that do not express the gene under study. One should be especially suspicious when nucleoli show positive staining after hybridization for an mRNA sequence, because this indicates nonspecific hybridization to rRNA. Like negative controls, positive controls are essential for interpreting hybridization results. Positive controls on the hybridization procedure may consist of hybridization to rRNA sequences, which are very abundantly present
8.7.12 Supplement 6
Current Protocols in Cytometry
in the cytoplasm as well as in the nucleoli of cells, or hybridization to the mRNAs of socalled housekeeping genes like actin or elongation factor. Furthermore, nucleolar staining with an rRNA probe tells that the cell is fully accessible for probe sequences.
Anticipated Results Combined DNA/RNA FISH and immunophenotyping should lead to efficient doublecolor detection of DNA/RNA sequences together with a cytoplasmic or membrane-bound antigen. Best results are obtained when target DNA sequences are highly repetitive or at least 30 kb in length, and when mRNA sequences are abundant. Generally, RNA sequences that are expressed at the level of housekeeping genes like human elongation factor and actin are easily detectable.
Time Considerations The combined FISH and immunocytochemical procedure can be completed within 2 days, starting with fixation of cells and labeling of probes and finishing with microscopic evaluation. For practical purposes the hybridization reaction is often allowed to proceed overnight. The total time of an oligonucleotide hybridization can be considerably reduced without visible reduction of hybridization signals when the hybridization reaction is performed in 1 or 2 hr.
Literature Cited Dirks, R.W., Van de Rijke, F.M., Fujishita, S., Van der Ploeg, M., and Raap, A.K. 1993. Methodologies for specific intron and exon RNA localization in cultured cells by haptenized and fluorochromized probes. J. Cell Sci. 104:1187-1197. Dirks, R.W., de Pauw, E.S.D., and Raap, A.K. 1997. Splicing factors associate with nuclear HCMVIE transcripts after transcriptional activation of the gene, but dissociate upon transcription inhibition: Evidence for a dynamic organization of splicing factors. J. Cell Sci. 110:515-522. Hessel, H., Mittermüller, J., Zitzelsberger, H., Weier, H.-U., and Bauchinger, M. 1996. Combined immunophenotyping and FISH with sex chromosome-specific DNA probes for the detection of chimerism in epidermal Langerhans cells after sex-mismatched bone marrow transplantation. Histochem. Cell Biol. 106:481-485. Höltke, H.J. and Kessler, C. 1990. Non-radioactive labeling of RNA transcripts in vitro with the hapten digoxigenin (Dig): Hybridization and ELISA-based detection. Nucleic Acids Res. 18:5843-5851. Huang, S. and Spector, D.L. 1996. Intron-dependent recruitment of pre-mRNA splicing factors to sites of transcription. J. Cell Biol. 133:719-732.
Lawrence, J.B., Singer, R.H., and Marselle LM. 1989. Highly localized tracks of specific transcripts within interphase nuclei visualized by in situ hybridization. Cell 57:493-502. Litle, V.R., Lockett, S.J., and Pallavicini, M.G. 1996. Genotype/phenotype analyses of low frequency tumor cells using computerized image microscopy. Cytometry 23:344-349. Raap, A.K., Van der Corput, M.P.C., Vervenne, R.A.W., Van Gijlswijk, R.P.M., and Tanke, H.J. 1995. Ultra-sensitive FISH using peroxidasemediated deposition of biotin- or fluorochrome tyramides. Hum. Mol. Genet. 4:529-534. Singer, R.H., Lawrence, J.B., and Villnave, C.A. 1986. Optimization of in situ hybridization using isotopic and non-isotopic detection methods. BioTechniques 4:230-250. Speel, E.J.M., Herbergs, J., Ramaekers, F.C.S., and Hopman, A.H.N. 1994. Combined immunocytochemistry and fluorescence in situ hybridization for simultaneous, tricolor detection of cell cycle, genomic, and phenotypic parameters of tumor cells. J. Histochem. Cytochem. 42:961-966. Weber-Matthiesen, K., Pressl, S., Schlegelberger, B., and Grote, W. 1993. Combined immunophenotyping and interphase cytogenetics on cryostat sections by the new FICTION method. Leukemia 74:646-649. Xing, Y., Johnson, C.V., Moen, P.T., McNeil, J.A., and Lawrence, J.B. 1995. Nonrandom gene organization: structural arrangements of specific pre-mRNA transcription and splicing with SC35 domains. J. Cell Biol. 131:1635-1647.
Key References Bakkus, M.H.C., Brakel van Peer, K.M.J., Adriaansen, H.J., Wierenga-Wolf, A.F., Van den akker, T.W., Dicke-Evinger, M.J., and Benner, R. 1989. Detection of oncogene expression by fluorescent in situ hybridization in combination with immunofluorescent staining of cell surface markers. Oncogene 4:1255-1262. Early report of a method that allows the study of gene expression in phenotypically defined cell populations. Dirks et al., 1993. See above. Describes an optimized protocol for RNA FISH on cultured cells. Harper, S.J., Pringle, J.H., Allen, A.C., Layward, L., Feehally, J., and Lauder, I. 1992. Simultaneous in situ hybridization of native mRNA and immunoglobulin detection by conventional immunofluorescence in paraffin wax embedding sections. J. Clin. Pathol. 45:114-119. Describes a method for simultaneous detection of a native mRNA by FISH and a cytoplasmic antigen by immunofluorescence in routine pathology specimens.
Molecular Cytogenetics
8.7.13 Current Protocols in Cytometry
Supplement 6
Raap, A.K., Van de Rijke, F.M., Dirks, R.W., Sol, C.J., Boom, R., and Van der Ploeg, M. 1991. Bicolor fluorescence in situ hybridization to intron and exon mRNA species. Exp. Cell Res. 197:319-322. Describes the combined detection of an RNA species with its cognate protein and with incorporated BrdU.
Weber-Matthiesen, K., Winkemann, M., MullerHermelink, A., Schlegelberger, B., and Grote, W. 1992. Simultaneous fluorescence immunophenotyping and interphase cytogenetics: A contribution to the characterization of tumor cells. J. Histochem. Cytochem. 40:171-175.
Singer et al., 1986. See above.
Provides a method that combines immunofluorescence immunophenotyping and FISH using centromere-specific cDNA probes on cytospin preparations.
Contains a detailed evaluation of hybridization parameters for RNA ISH.
Weber-Matthiesen et al., 1993. See above.
Strehl, S. and Ambros, P.F. 1993. Fluorescence in situ hybridization combined with immunohistochemistry for highly sensitive detection of chromosome 1 aberrations in neuroblastoma. Cytogenet. Cell Genet. 63:24-28. First description of a technique for double-target FISH and simultaneous immunological staining of a cell-surface antigen.
First description of a technique combining interphase cytogenetics and immunophenotyping on cryostat sections.
Contributed by Roeland W. Dirks Leiden University Medical Centre Leiden, The Netherlands
Van den Berg, H., Vossen, J.M., Van den Bergh, R.L., Bayer, J., and Van Tol, M.J.D. 1991. Detection of Y chromosome by in situ hybridization in combination with membrane antigens by twocolor immunofluorescence. Lab Invest. 64:623628. Describes a technique for simultaneous detection of the Y chromosome and differentiation-specific membrane antigens on peripheral-blood mononuclear cells.
Combination DNA/RNA FISH and Immunophenotyping
8.7.14 Supplement 6
Current Protocols in Cytometry
Single-Nucleotide Sequence Discrimination In Situ Using Padlock Probes
UNIT 8.8
Standard fluorescence in situ hybridization (FISH) techniques using cloned probes are limited in their ability to distinguish between closely similar DNA sequences, because long hybridization probes are not detectably destabilized by single mismatched base pairs. This limitation has been addressed by using short allele-specific oligonucleotide probes whose hybridization to target sequences is more sensitive to mismatches. Oligonucleotides have also been used in combination with DNA polymerases to prime in situ polymerization, incorporating labeled nucleotides. In this procedure, mismatches at the 3′ end of the primer inhibit DNA synthesis. DNA ligases are very sensitive to mismatches at the DNA ends to be joined through ligation. This mechanism has been utilized to distinguish DNA sequence variants in situ using so-called padlock probes. Padlock probes are linear oligonucleotides with targetcomplementary sequences at both ends, and a non-target-complementary segment in between. The end sequences are designed such that they are brought next to each other upon hybridization to the target DNA sequence. If the ends are perfectly matched to the target sequence, they can be joined by a DNA ligase. The circularized probe molecule becomes threaded on the target DNA strand, resulting in a hybridization-independent bond of the reacted probe molecules. Padlock probes detect target sequences with very high specificity, because both probe segments must hybridize to the target for circularization to occur. This unit presents a protocol for discrimination between closely similar DNA sequences in situ using padlock probes. The protocol is aimed at detecting variations in repetitive sequences. A discussion of means to greatly amplify the signal from circularized probes is also included. This approach is suitable for the analysis of tandemly repeated sequences with copy number ≥100 repeat units. Accordingly, most centromeric alpha satellite DNA repeats should be analyzable using this protocol. In situ studies of alpha satellite repeats are of value for elucidating the structure and function of centromeres, and for determining the ploidy of cells. Polymorphic alpha satellite repeats can be used to determine in situ the parental origin of individual chromosomes. STRATEGIC PLANNING Probe selection. When using Tth DNA ligase (see Basic Protocol), two probe sequences hybridizing to adjacent sequences should be selected so that both hybridizing segments are ∼20 nucleotides in length and have similar estimated melting temperatures. When using T4 DNA ligase (see Alternate Protocol 1), the target-complementary probe arms can be shorter (∼15 nucleotides), as hybridization and ligation are performed at lower temperatures. The ligation reaction is most sensitive to mismatches at the 3′ end of the probe (Luo et al., 1996). The probe should therefore be designed so that the diagnostic position in the target sequence is complementary to the 3′ ultimate end of the probe. The non-target-complementary linking segments in the authors’ probes are generally 10 nucleotides longer than the combined target complementary sequences (i.e., ∼50 nucleotides). Alternatively, if the probes are chemically synthesized the non-target-complementary segment can be made up of hexaethyleneglycol residues (Nilsson et al., 1994). Each hexaethyleneglycol is expected to correspond to 4 to 5 nucleotides in length, thus decreasing the number of units that have to be introduced during synthesis, and contributing to an increased yield and purity of the oligonucleotide probe. Contributed by Mats Nilsson, Ulf Landegren, and Dan-Oscar Antson Current Protocols in Cytometry (2001) 8.8.1-8.8.12 Copyright © 2001 by John Wiley & Sons, Inc.
Molecular Cytogenetics
8.8.1 Supplement 16
Probe synthesis. Padlock probes can be synthesized either by standard chemical oligonucleotide synthesis, or enzymatically using a PCR-based protocol. Although chemical synthesis produces a greater yield, it has some limitations. Chemical synthesis of probes >100 nt in length is increasingly difficult, as imperfect products accumulate with greater oligonucleotide length. Moreover, the separation of full-length oligonucleotides from truncated products through either HPLC or gel purification is also more difficult with increasing probe length. It is also relatively expensive to label padlock probes extensively with detectable functions such as haptens or fluorophores during or after chemical synthesis. Enzymatic synthesis of padlock probes, on the other hand, produces lower yield than standard chemical synthesis, but longer probes can be made. The probes can also be labeled densely during PCR by addition of labeled nucleotides in the reaction. The PCR-based method thus offers a flexible means for probe synthesis. By changing primers, template, or labeled nucleotides in the PCR, the target-complementary sequences, the overall size, and the labels added to the padlock probes can all be varied. CHEMICAL SYNTHESIS OF PADLOCK PROBES Probe labeling. Probes should be labeled in the non-target-complementary segment. This can be done by introducing detectable functions directly during synthesis or by incorporating primary amino groups that after synthesis are conjugated with N-hydroxysuccinimide (NHS) esters of haptens or fluorophores. All relevant labeling techniques can be found in Hermanson (1996). The authors prefer to introduce primary amino groups that are connected to modified nucleobases via a linker (there are several commercial sources, e.g., Glen Research), and to conjugate these to NHS esters of biotin or digoxigenin (Clontech and Boehringer Mannheim, respectively). These can then be viewed after labeling with fluorescently labeled avidin or anti-digoxigenin antibodies, as described in the protocols. 5′ Phosphate. A 5′ phosphate is required for the ligation reaction. Introduce a 5′ phosphate during synthesis or by using T4 polynucleotide kinase (T4 PNK) after synthesis (see Sambrook et al., 1989). The authors have introduced 5′ phosphates during synthesis using a reagent that allows trityl-on HPLC purification (Connolly, 1987; Guzaev et al., 1995), for which there are several commercial sources (e.g., Glen Research). The presence of 5′ phosphates can be monitored by labeling probes at the 3′ end, using terminal deoxynucleotidyl transferase (TdT), and analyzing the probes on a denaturing polyacrylamide gel. Phosphorylated probes have an increased mobility in gels corresponding to half a nucleotide. Probe purity. The purity of the oligonucleotide is crucial for successful reactions, because both ends of each probe molecule must be intact in order for ligation to occur. Trityl-on HPLC purification ensures that oligonucleotides have intact 5′ ends. A subsequent gel purification serves to remove any 3′ truncated sequences. Example of probe design. The following probe sequences were used to distinguish two alpha satellite repeats present on human chromosomes 13 and 21, based on a single nucleotide difference (Nilsson et al., 1997). achrA: 5′-P-AAAGGAGTTGAACATTTCTATTT19-biotin-T10-biotinT19TGATGTGTGTACCCAGCT 3′ Single-Nucleotide Sequence Discrimination In Situ Using Padlock Probes
achrB: 5′-P-AAAGGAGTTGAACATTTCTATTT19-digoxigenin-T10-digoxigeninT19TGATGTGTGTACCCAGCC 3′
8.8.2 Supplement 16
Current Protocols in Cytometry
ENZYMATIC SYNTHESIS OF PADLOCK PROBES Enzymatic synthesis of padlock probes is recommended when probes needed are longer, more densely labeled than is possible through chemical synthesis. Chemically synthesized padlock probes are typically ∼70 to 90 nucleotides (nt). The authors have found that a probe length of 140 nt produced maximal detection signals of a repeated sequence in metaphase chromosomes. Probes >200 nt gave rise to a low signal and high background due to unspecific binding (Antson, 2000). It should be noted, though, that the optimal probe length may differ between analyses of interphase cell nuclei, chromatin fibers, and metaphase chromosomes. Primer and template selection. Two primers, a 5′-biotinylated forward primer and a 5′-phosphorylated reverse primer, and a DNA template are required for enzymatic synthesis of padlock probes. The target-complementary sequences of the padlock probes are defined by the 5′ parts of the primers, while the 3′ ends of the primers are complementary to the PCR template. The 5′-phosphorylated primer becomes incorporated as the 5′ end of the padlock probe, and the 3′ end of the probe is the complement of the biotinylated primer. If the probes will be used to distinguish single nucleotide variants of a target sequence, then the variable nucleotide position should be the 5′-most position of the 5′-biotinylated primer, as ligases are most inhibited by mismatched 3′ termini at the site of ligation. The DNA template used to create the non-target-complementary linking segment should preferably not contain sequences present in the target chromosomes. A suitable template, used in the Support Protocol, is DNA from the bacteriophage λ. Labeling. The probe is labeled during PCR by replacing one third of the dTTP with an equal amount of labeled dUTP. Suitable haptens are dinitrophenyl and digoxigenin, but not biotin as it would bind to the streptavidin-coated paramagnetic particles used to render the double-stranded PCR product single-stranded. The probes can also be directly labeled with a fluorophore by incorporating, e.g., fluorescein-12-dUTP. To lower the reagent cost, the relatively low dNTP concentration of 50 µM is used. This also decreases the tendency of Taq polymerase to add an extra, nontemplated nucleotide at the 3′ ends of the amplification products, which would prevent ligation of the probe. The problem of adding an extra nucleotide can also be avoided if a DNA polymerase with 3′→5′ exonucleolytic activity is chosen, e.g., Pfu polymerase. However, polymerases with 3′ exonuclease activity can produce PCR products that lack one or several nucleotides at the 3′ end at low dNTP concentrations. This problem can be avoided by raising the dNTP concentration to 200 µM at the last extension step in the PCR. Proofreading DNA polymerases have a greater tendency to be inhibited by hapten-modified nucleotides, and for example Pfu DNA polymerase is inhibited by dinitrophenyl-11-dUTP at the concentration used in the Support Protocol. Example of probe design. The following primers and DNA template were used to synthesize two 128-mer padlock probes targeting a single-nucleotide difference in the centromeric alpha satellite repeats on human chromosomes 13 and 21. For clarity, the target-complementary sequences of the padlock probes are shown in bold and the complements to the target complementary sequences are shown in bold italic. Primers stat2: 5′ B-GGCTGGGTACACACATCAGGAATGGTGCAGAAATGT 3′ stat3: 5′ B-AGCTGGGTACACACATCAGGAATGGTGCAGAAATGT 3′ pad5′: 5′ P-AAAGGAGTTGAACATTTCTATTTGTTTTTTCGAGCGTTCA 3′ Molecular Cytogenetics
8.8.3 Current Protocols in Cytometry
Supplement 16
Template λ DNA 2064-GGAATGGTGCAGAAATGTCGATATCCCGTATCTGCTGGGATACTGGCGGGAT TGACCCGACCATTGTGTATGAACGCTCGAAAAAACA-2153 Nucleotide positions according to GenBank accession number J02459.
Padlock probe CenT: Primers stat3 and pad5′ 5′ P-AAAGGAGTTGAACATTTCTATTTGTTTTTTCGAGCGTTCATACACAATGGTCG GGTCAATCCCGCCAGTATCCCAGCAGATACGGGATATCGACATTTCTGCACCATTCCT GATGTGTGTACCCAGCT 3′ Padlock probe cenC: Primers stat2 and pad5′ 5′ P-AAAGGAGTTGAACATTTCTATTTGTTTTTTCGAGCGTTCATACACAATGGTCGG GTCAATCCCGCCAGTATCCCAGCAGATACGGGATATCGACATTTCTGCACCATTCCT GATGTGTGTACCCAGCC 3′ BASIC PROTOCOL
PADLOCK PROBE IN SITU LIGATION REACTION USING A THERMOSTABLE LIGASE In this protocol a thermostable DNA ligase is used to enable heat denaturation of the chromosomes in the presence of the ligase. In brief, the ligation reaction is performed on a temperature-controlled heating block, slides are washed to remove unreacted probes, and the biotin- and digoxigenin-labeled probes are then visualized using fluorophore-conjugated avidin or antibodies. Alternate Protocol 1 describes a procedure that does not require temperature cycling or the use of thermostable ligases. Materials RNase A–treated metaphase chromosome slides (UNIT 8.2), prepared fresh Tth ligation mix (see recipe) Stop buffer (see recipe) Deionized formamide (store protected from light for up to 6 months at 4°C) 2× SSC (APPENDIX 2A) Wash buffer (see recipe) 10% (w/v) blocking solution (Boehringer Mannheim; stable >1 year at −20°C) 1 mg/ml fluorescein isothiocyanate–conjugated avidin (FITC-avidin; Vector Laboratories) 200 µg/ml rhodamine-conjugated anti-digoxigenin Fab fragments (Boehringer Mannheim) 70%, 85%, and 100% (v/v) ethanol Vectashield mounting medium (Vector Laboratories) containing 100 ng/ml 4′,6-diamidino-2-phenylindole (DAPI) counterstain Programmable heating block (e.g., GeneE from Techne) 24 × 50–mm coverslips Coplin jars Fluorescence microscope with a charge-coupled device (CCD) camera and a digital imaging system (e.g., IP-lab, Vysis)
Single-Nucleotide Sequence Discrimination In Situ Using Padlock Probes
Perform in situ ligation reaction 1. Place fresh, RNase A–treated metaphase chromosome slides on a programmable heating block and preheat the slides at the ligation temperature (typically 55°C).
8.8.4 Supplement 16
Current Protocols in Cytometry
Protease treatment and postfixation of metaphase slides are not required. The thermocycler should be equipped with a flat heating block suitable for slides and should respond rapidly to temperature changes (preferably >1°C/sec). The machine should also be equipped with a lid under which a paper towel can be placed to maintain a humid atmosphere. The ligation temperature should not exceed the melting temperature of the probe arms, but the exact temperature is not critical for successful discrimination.
2. Add 55 µl freshly prepared Tth ligation mix to each slide and cover with a 24 × 50–mm coverslip. The ligation mix contains the two probes, one labeled with biotin and one labeled with digoxigenin.
3. Start a temperature program of 2 min at 92°C to denature the chromosomes, followed by 15 min at the ligation temperature. The optimal denaturation temperature has to be determined empirically for the machine used. Excessively high denaturation temperatures negatively affect the chromosome structure.
4. Transfer the slides to a Coplin jar containing stop buffer heated to the ligation temperature, and incubate 2 min. The coverslips should detach during this step.
5. Wash the slides 10 min in 2× SSC containing 30% (v/v) deionized formamide at 42°C. 6. Wash 5 min in 2× SSC at 55°C. These washing conditions (steps 5 and 6) will remove nonligated probes and have been optimized for probes with an optimal ligation temperature of 55°C. The washing conditions for probes of other hybridization stabilities must be adjusted accordingly.
7. Transfer the slides to wash buffer at room temperature. Fluorescently label the ligated probes 8. Drain the slides by tilting them on a paper towel. Do not allow the slides to dry here or at any point before the ethanol series. 9. Add 60 µl of 0.5% (w/v) blocking solution in wash buffer and cover with a 24 × 50–mm coverslip. 10. Incubate 10 min at 37°C in a humid chamber (e.g., a 1-liter beaker containing paper towel wetted with wash buffer and covered with aluminum foil). 11. Shake off the coverslips or remove them in wash buffer and drain as in step 8. Add 60 µl wash buffer containing 0.5% blocking solution, 2 µg/ml FITC-avidin, and 4 µg/ml rhodamine-conjugated anti-digoxigenin Fab fragments. 12. Incubate 30 min at 37°C in a humid chamber. 13. Wash three times, 5 min each, in wash buffer at room temperature. If necessary, signals may be amplified at this point by repeating steps 11 to 13. The signal from biotinylated probes can be amplified using biotinylated antibodies against avidin, followed by a second round of FITC-avidin (steps 11 to 13 a third time). The signals from digoxigenin-labeled probes can be amplified by using mouse anti-digoxigenin antibodies in the first detection layer, followed by sheep anti-mouse antibodies labeled with rhodamine. Molecular Cytogenetics
8.8.5 Current Protocols in Cytometry
Supplement 16
View samples 14. Dehydrate the slides in an ethanol series of 70%, 85%, and 100%, 2 min each, and then allow to air dry. 15. Add 20 µl Vectashield mounting medium containing 100 ng/ml DAPI counterstain. 16. Analyze the result using a fluorescence microscope equipped with a CCD camera and a digital imaging system (e.g., IP-lab). Compare the intensity of red and green fluorescence from the different hybridized probes. ALTERNATE PROTOCOL 1
PADLOCK PROBE IN SITU LIGATION REACTION USING T4 DNA LIGASE This protocol uses T4 DNA ligase and does not require a programmable heating block. Additional Materials (also see Basic Protocol) T4 ligation mix (see recipe) 1. Denature fresh, RNase A–treated metaphase slides 2 min in 2× SSC containing 70% (v/v) deionized formamide at 70°C. Protease treatment and postfixation of metaphase slides are not required.
2. Transfer the slides to ice-cold 2× SSC and incubate 2 min. 3. Dehydrate the slides by transferring them through a series of 70%, 85%, and 100% ethanol, 2 min each, and then allow to air dry. 4. Add 55 µl freshly prepared T4 ligation mix to each slide and cover with a 24 × 50–mm coverslip. The ligation mix contains the two probes, one labeled with biotin and one labeled with digoxigenin.
5. Incubate 30 min at 37°C in a humid chamber (e.g., a 1-liter beaker containing paper towel wetted with wash buffer and covered with aluminum foil). The ligation temperature should not exceed the melting temperature of the probe arms, but the exact temperature is not critical for successful discrimination.
6. Transfer the slides to a Coplin jar containing stop buffer heated to 37°C. Incubate 2 min. The coverslips should detach during this step.
7. Wash the slides to remove nonligated probes (see Basic Protocol, steps 5 and 6). Probes with target-complementary arms shorter than 20 nucleotides will require lowerstringency washes than those in the Basic Protocol.
8. Transfer the slides to wash buffer at room temperature. 9. Perform fluorescent labeling and view slides (see Basic Protocol, steps 8 to 16). ALTERNATE PROTOCOL 2
Single-Nucleotide Sequence Discrimination In Situ Using Padlock Probes
ENZYMATIC PADLOCK PROBE LIGATION REACTION AT LOW PROBE CONCENTRATION This protocol is recommended if hybridization has to be performed with a low probe concentration, which is desirable if probes are synthesized according to the Support Protocol. At low probe concentrations, background from nonspecifically bound probes is minimized. In order to minimize probe consumption, the probes are hybridized overnight
8.8.6 Supplement 16
Current Protocols in Cytometry
to denatured metaphase chromosomes, followed by ligation, and finally visualization using fluorophore-conjugated antibodies targeted against the hapten-modified probes. Additional Materials (also see Basic Protocol) Overnight hybridization mix (see recipe) Rubber cement (e.g., FASTIK, AB Thure Bünger) 37°C humid chamber (e.g., a 1-liter beaker containing 2× SSC–soaked paper towel and covered with aluminum foil) Tweezers 1. Perform RNase A treatment and denaturation of slides as described in steps 1 to 3 in Alternate Protocol 1. 2. Place fresh RNase A–treated and denatured metaphase chromosome slides on the lab bench. 3. Add 30 µl overnight hybridization mix as 4 drops to the slide. Overlay with 24 × 50–mm coverslip. Let the hybridization solution spread under the coverslip, then seal edges of coverslip with rubber cement. Incubate slides overnight in a 37°C humid chamber. 4. Remove rubber cement with tweezers and incubate slides in 2× SSC 5 min at 37°C to remove excess probes. Make sure that coverslip detaches from slide.
5. Dehydrate slides by transferring through a series of 70%, 85%, and 100% ethanol, 2 min each, and allow to air dry. 6. Proceed with the ligation as described in Basic Protocol, steps 1 to 4, when using a thermostable ligase, with the modifications that the probes should be omitted from the ligation mix, and no 92°C denaturing step should be performed. If ligation is performed with T4 DNA ligase, perform Alternate Protocol 1, steps 4 and 5, omitting probes in the T4 ligation mix.
7. Continue Basic Protocol, steps 4 to 16, to stop ligation reaction, wash slides, label probes with appropriate antibodies, and view samples. ENZYMATIC SYNTHESIS OF PADLOCK PROBES A pair of 5′-phosphorylated and 5′-biotinylated primers are used in a PCR to synthesize a padlock probe. The probes are then rendered single-stranded by binding the biotinylated PCR product to paramagnetic streptavidin-coated beads, and releasing the unbound strand under denaturing conditions. The probes are ready for use after neutralization of the solution and precipitation. The protocol results in amounts of a 128-mer probe sufficient for ≥100 experiments using Alternate Protocol 2. Materials 10× AmpliTaq buffer with 1.5 mM MgCl2 (PE Biosystems) dNTP mix (see recipe) 1 mM labeled dUTP (e.g., digoxigenin-11-dUTP, Boehringer Mannheim; or dinitrophenyl-11-dUTP, Molecular Probes) 5 µM 5′-phosphorylated forward primer 5 µM 5′-biotinylated reverse primer 30 pg bacteriophage λ DNA (e.g., Amersham Pharmacia Biotech) 5 U/µl AmpliTaq DNA polymerase (PE Biosystems)
SUPPORT PROTOCOL
Molecular Cytogenetics
8.8.7 Current Protocols in Cytometry
Supplement 16
2% (w/v) agarose gel 3 M sodium acetate, pH 4.6 70% and 100% ethanol, ice cold Streptavidin-coated paramagnetic particles (Dynabeads M-280 Streptavidin, Dynal) 5010 buffer (see recipe) 2 M NaCl 0.15 M lithium hydroxide (LiOH) 0.15 M ammonium chloride (NH4Cl) Glycogen (Boehringer Mannheim) TE buffer (APPENDIX 2A) 0.5-ml and 1.5-ml microcentrifuge tubes PCR thermal cycler Microcentrifuge (e.g., Biofuge 15, Heraeus) Magnetic particle concentrator rack (Dynal) Test-tube rotator (e.g., Labinco 528) Ultra microvolume cell (Amersham Pharmacia Biotech) Spectrophotometer (e.g., GeneQuant, Amersham Pharmacia Biotech) Set up and run PCR 1. Set up ten amplification reactions in 0.5-ml microcentrifuge tubes placed on ice: 5 µl 10× AmpliTaq buffer with 1.5 mM MgCl2 1.25 µl dNTP mix (see recipe) 0.83 µl 1 mM labeled dUTP 1 µl 5 µM phosphorylated forward primer 0.5 µl 5 µM biotinylated reverse primer 1 µl 30 pg template bacteriophage λ-DNA 0.2 µl 5 U/µl AmpliTaq DNA polymerase H2O to 50 µl As a negative control, add water in place of template DNA to one of the reactions. It is not possible to use biotin-11-dUTP as a labeling hapten since biotin-labeled strands will bind to the streptavidin-coated paramagnetic beads.
2. Perform the PCR using the following program: 35 cycles:
Final step:
1 min 1 min 1 min 7 min
94°C (denaturation) 55°C (annealing) 72°C (extension) 72°C thereafter 8°C (final extension)
3. Pool amplification reactions and examine PCR products on a 2% agarose gel. Concentrate particles 4. Precipitate the PCR products with 0.1 volume of 3 M sodium acetate, pH 4.6, and 2 volumes ice-cold 100% ethanol. Centrifuge 30 min at maximum speed in a microcentrifuge, 4°C. Remove supernatant and allow pellet to dry. Resuspend DNA in 50 µl sterile water. 5. Resuspend streptavidin-coated paramagnetic particles by shaking the vial. Single-Nucleotide Sequence Discrimination In Situ Using Padlock Probes
6. Pipet 200 µl resuspended particles into a clean 1.5-ml microcentrifuge tube. 7. Place the tube in a magnetic particle concentrator rack and allow particles to concentrate on the wall of tube. Remove storage solution.
8.8.8 Supplement 16
Current Protocols in Cytometry
8. Wash particles by adding 100 µl of 5010 buffer to tube and resuspend. Place tube back in the rack and concentrate particles. Remove the 5010 buffer. 9. Add 50 µl PCR product to the washed particles. Add 50 µl of 2 M NaCl. Resuspend and incubate 30 min at room temperature. Keep particles suspended during incubation by using a test-tube rotator or by vortexing the solution at regular intervals. 10. Remove the binding solution and wash the particles three times with 100 µl of 5010 buffer. 11. Add 50 µl of 0.15 M LiOH to the particles and incubate 5 min at room temperature. Remove the denaturing solution to a new 1.5-ml microcentrifuge tube and discard particles. Neutralize the denaturing solution with 50 µl of 0.15 M NH4Cl. Precipitate DNA 12. Precipitate the DNA with 0.1 volume of 3 M sodium acetate, pH 4.6, 2 volumes ice-cold 100% ethanol, and 20 µg glycogen as carrier. Microcentrifuge 30 min at maximum speed, 4°C. Remove supernatant and wash the pellet by adding 500 µl of 70% ethanol and microcentrifuge 10 min at maximum speed, 4°C. Remove supernatant and let pellet dry. Resuspend DNA in 25 µl TE buffer. 13. Dilute 2 µl resuspended DNA to 20 µl with sterile water. Quantitate by spectrophotometry using an ultra microvolume cell. The quality of the padlock probes may be assessed before using the probes, see Critical Parameters and Troubleshooting.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
5010 buffer 50 mM NaCl 10 mM Tris⋅Cl, pH 7.5 (APPENDIX 2A) Store up to 1 year at room temperature dNTP mix 6 µl 100 mM dATP 6 µl 100 mM dCTP 6 µl 100 mM dGTP 4 µl 100 mM dTTP H2O to 300 µl Store up to 1 year at −20°C Overnight hybridization mix 25 µl 20% (v/v) deionized formamide (APPENDIX 2A) in 2× SSC 3 µl 5 mg/ml sonicated salmon sperm DNA, boiled 5 min and chilled on ice 1 µl 30 fmol of each probe Prepare fresh on ice Stop buffer 3.3× SSC (see APPENDIX 2A for 20×) 50 mM EDTA (APPENDIX 2A) Store up to 1 year at room temperature Molecular Cytogenetics
8.8.9 Current Protocols in Cytometry
Supplement 16
T4 ligase buffer, 10× 100 mM magnesium acetate 500 mM potassium acetate 100 mM Tris⋅acetate, pH 7.5 Store at −20°C (stable for >1 year) T4 ligation mix 1× T4 ligase buffer (see recipe) 1 mM ATP 8.7% (w/v) glycerol 200 mM NaCl 0.1 µg/µl bovine serum albumin (BSA) 0.1 µg/µl denatured sonicated salmon sperm DNA 75 nM of each labeled padlock probe (see Strategic Planning) 0.1 Weiss U/µl T4 DNA ligase Prepare fresh on ice Sonicated salmon sperm DNA and BSA can be added from 1 mg/ml stock solutions, and ATP from a 10 mM stock solution. All are stable >1 year at −20°C.
Tth ligase buffer, 10× 1 M KCl 100 mM MgCl2 200 mM Tris⋅Cl, pH 7.9 (APPENDIX 2A) 10 mM EDTA (APPENDIX 2A) 10 mM dithiothreitol 1% (v/v) Triton X-100 Store for up to 1 year at −20°C Heat 5 min at 65°C before use to dissolve any precipitates Tth ligation mix 1× Tth ligase buffer (see recipe) 1 mM nicotinamide adenine dinucleotide (NAD) 8.7% (w/v) glycerol 0.1 µg/µl sonicated salmon sperm DNA 0.1 µg/µl bovine serum albumin (BSA) 75 nM of each labeled padlock probe (see Strategic Planning) 0.25 U/µl Tth DNA ligase (BDH Laboratory Supplies) Prepare fresh on ice Sonicated salmon sperm DNA and BSA can be added from 1 mg/ml stock solutions, which are stable >1 year at −20°C. A 100 mM NAD stock solution can be stored for >1 year at −20°C. Ligation mix should be prepared from a 10 mM NAD working solution, which can be frozen and thawed regularly for up to 2 months if stored at −20°C.
Wash buffer 2× SSC (APPENDIX 2A) 0.05% (v/v) Tween 20 Store up to 1 year at room temperature
Single-Nucleotide Sequence Discrimination In Situ Using Padlock Probes
8.8.10 Supplement 16
Current Protocols in Cytometry
COMMENTARY Background Information Compared to other reagents used in fluorescence in situ hybridization (FISH), padlock probes have unique advantages for the discrimination of closely similar DNA sequences (Nilsson et al., 1994). The selectivity (i.e., the ability to discriminate between any closely similar sequences such as single-nucleotide differences) is excellent under standard reaction conditions (Landegren et al., 1988; Luo et al., 1996). Allele-specific oligonucleotide hybridization has been successfully applied in situ to discriminate between two alpha satellite sequences differing in four nucleotide positions (O’Keefe et al., 1996). Primed in situ (PRINS) labeling (Koch et al., 1989) has been applied to discriminate between single nucleotide positions (Pellestor et al., 1995), exploiting the sensitivity of DNA polymerases to mismatches at the 3′ end of primers. Sequence distinction using padlock probes has the advantage over PRINS reactions that two or more probes can be added to the same reaction, thereby ensuring identical reaction conditions and allowing the reaction to be performed in a competitive manner. Dual-color PRINS reactions must be performed sequentially with a dideoxynucleotide incorporation step in between (UNIT 8.11). The authors believe that dual-color padlock probe reactions are more quantitative and more easily interpreted than dual-color PRINS reactions. The padlock probe design has other advantages compared to allele-specific oligonucleotide hybridization and PRINS reactions. The probe ligation reaction requires two oligonucleotide hybridization events, which ensures highly specific recognition of target sequences in complex samples such as total human genomic DNA. Moreover, circularized padlock probes are efficient templates for a rolling-circle replication reaction that can amplify the signal from padlock probes 1000-fold in an hour (Fire and Xu, 1995; Liu et al., 1996; Banér et al., 1998; Lizardi et al., 1998). In this manner, only circularized probes, and not nonspecifically bound probe molecules, will be amplified. This specific amplification should be useful to detect padlock probes that have identified single-copy genes in situ.
Critical Parameters and Troubleshooting It is important to establish the optimal target denaturation temperature on individual heating devices. However, conditions optimized for
FISH or PRINS are also suitable for padlock probe ligation reactions. The washing procedure performed after the ligation reaction should be adjusted to remove nonligated probes, but not circularized ones. Washes of too-high stringency will remove some of the circularized probes, probably due to nearby breaks of the DNA strands in the chromosomes and due to loss of DNA during the denaturation procedure (Raap et al., 1986; Nilsson et al., 1994). The concentration of ligase may need to be optimized. If too little enzyme is used, many hybridized probes will be removed in the washing procedure. If too much enzyme is used, sequence variants are less well discriminated. The optimal amount can be selected by using two different washing conditions to distinguish ligated probes from ones that only hybridize to the target sequence. The high-stringency washing conditions used in this protocol remove all nonligated probes, whereas low-stringency washing conditions (e.g., 2× SSC at 42°C for 10 min) allow nonligated probes to remain hybridized. The signals from ligated probes can be expected to be quite faint, but with less background compared to FISH or PRINS, because padlock probes introduce much less labeling per probe than do conventional FISH and PRINS reactions. Therefore, it might be necessary to amplify the signal by using additional layers of secondary labeling reagents. It is advisable to test the circularization efficiency of probes in an in vitro ligation reaction using oligonucleotide templates and 5′ radiolabeled probes. Circularized probes migrate slower than linear ones in a denaturing polyacrylamide gel (Nilsson et al., 1994).
Anticipated Results The anticipated result is highly specific but relatively faint fluorescent signals from reacted padlock probes, which readily discriminate the different sequence variants under study.
Time Considerations Starting from prepared metaphase slides and labeled padlock probes, the Basic Protocol requires 20 min for the ligation reaction, 20 min for washes, 1 hr for secondary labeling (including washes), and 15 min for dehydration and mounting of the slides (∼2 hr total). Alternate Protocol 1 requires an additional 20 min for
Molecular Cytogenetics
8.8.11 Current Protocols in Cytometry
Supplement 16
denaturation and 10 min for the reaction, adding up to 2.5 hr, and Alternate Protocol 2 additionally requires an overnight hybridization. Enzymatic synthesis of padlock probes (see Support Protocol) requires 2.5 hr for setting up and running the PCR. The subsequent examination of the PCR product on an agarose gel and preparation of single-stranded probe requires 3 hr.
Literature Cited Antson, D.-O., Isaksson, A., Landegren, U., and Nilsson, M. 2000. PCR-generated padlock probes detect single-nucleotide variation in genomic DNA. Nucl. Acids Res. 28:e58. Banér, J., Nilsson, M., Mendel-Hartvig, M., and Landegren, U. 1998. Signal amplification of padlock probes by rolling circle replication. Nucl. Acids Res. 22:5073-5078. Connolly, B.A. 1987. Solid phase 5′-phosphorylation of oligonucleotides. Tetrahedron Lett. 28:463-466. Fire, A. and Xu, S.-Q. 1995. Rolling replication of short DNA circles. Proc. Natl. Acad. Sci. U.S.A. 92:4641-4645. Guzaev, A., Salo, H., Azhayev, A., and Lönnberg, H. 1995. A new approach for chemical phosphorylation of oligonucleotides at the 5′-terminus. Tetrahedron 51:9375-9384. Hermanson, G.T. 1996. Bioconjugate Techniques. Academic Press, San Deigo. Koch, J., Kølvraa, S., Petersen, K., Gregersen, N., and Bolund, L. 1989. Oligonucleotide-priming methods for the chromosome-specific labelling of alpha satellite DNA in situ. Chromosoma 98:259-265. Landegren, U., Kaiser, R., Sanders, J., and Hood, L. 1988. A ligase-mediated gene detection technique. Science 241:1077-1080. Liu, D., Daubendiek, S.L., Zillman, M.A., Ryan, K., and Kool, E.T. 1996. Rolling circle DNA synthesis: Small circular oligonucleotides as efficient templates for DNA polymerases. J. Am. Chem. Soc. 118:1587-1594. Lizardi, P.M., Huang, X., Zhu, Z., Bray-Ward, P., Thomas, D.C., and Ward, D.C. 1998. Mutation detection and single-molecule counting using isothermal rolling-circle amplification. Nature Genet. 19:225-232. Luo, J., Bergstrom, D.E., and Barany, F. 1996. Improving the fidelity of Thermus thermophilus DNA ligase. Nucl. Acids Res. 24:3071-3078. Nilsson, M., Malmgren, H., Samiotaki, M., Kwiatkowski, M., Chowdhary, B.P., and Landegren, U. 1994. Padlock probes: Circularizing oligonucleotides for localized DNA detection. Science 265:2085-2088.
Nilsson, M., Krejci, K., Koch, J., Kwiatkowski, M., Gustavsson, P., and Landegren, U. 1997. Padlock probes reveal single-nucleotide differences, parent of origin and in situ distribution of centromeric sequences in human chromosomes 13 and 21. Nature Genet. 16:252-255. O’Keefe, C.L., Warburton, P.E., and Matera, A.G. 1996. Oligonucleotide probes for alpha satellite DNA variants can distinguish homologous chromosomes by FISH. Hum. Mol. Genet. 5:17931799. Pellestor, F., Girardet, A., Genevieve, L., Andreo, B., and Charlieu, J.P. 1995. Use of the primed in situ labelling (PRINS) technique for a rapid detection of chromosomes 13, 16, 18, 21, X and Y. Hum. Genet. 95:12-17. Raap, A.K., Marijnen, J.G.J., Vrolijk, J., and van der Ploeg, M. 1986. Denaturation, renaturation, and loss of DNA during in situ hybridization procedures. Cytometry 7:235-242. Sambrook, J., Fritsch, E.F., and Maniatis, T. 1989. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Key References Antson et al., 2000. See above. A PCR-based protocol for flexible enzymatic synthesis of padlock probes >100 or so nucleotides is described. The Support Protocol and Alternate Protocol 2 are taken from this paper. Nilsson et al., 1994. See above. Describes the padlock probe design, and demonstrates the utility of in situ hybridization using a padlock probe that specifically detects an alpha satellite repeat on human chromosome 12. Alternate Protocol 1 is taken from this paper, with some modifications. Nilsson et al., 1997. See above. Single-nucleotide discrimination ability in situ is demonstrated using two padlock probes that distinguish two alpha satellite repeats on the basis of a single nucleotide substitution present on human chromosomes 13 and 21. The Basic Protocol is taken from this paper.
Contributed by Mats Nilsson, Ulf Landegren, and Dan-Oscar Antson Uppsala University Uppsala, Sweden
Single-Nucleotide Sequence Discrimination In Situ Using Padlock Probes
8.8.12 Supplement 16
Current Protocols in Cytometry
Tyramide Signal Amplification (TSA) Systems for the Enhancement of ISH Signals in Cytogenetics
UNIT 8.9
Tyramide Signal Amplification (TSA) is a peroxidase-based signal amplification system that is compatible with all in situ hybridization (ISH) as well as immunocytochemical detection schemes. TSA utilizes the catalytic activity of the peroxidase enzyme to covalently bind a label on a solid phase. For the cytogenetic applications in this unit, the solid phase refers to protein components of cells or cellular structures that are immobilized on microscope slides. There are two basic strategies for facilitating amplified detection utilizing TSA, termed TSA-Direct and TSA-Indirect. TSA-Direct refers to the direct binding of a fluorescent label, and TSA-Indirect refers to the binding of a hapten such as biotin or dinitrophenyl (DNP). In the indirect format, the deposited hapten is detected by fluorescent- or enzyme-labeled streptavidin or antibody. This unit contains detailed protocols for bright-field detection methods, utilizing a precipitating chromogenic substrate, and for fluorescence-based detection methods. TSA detection formats are introduced for bright-field (see Basic Protocol 1) and fluorescence detection (see Basic Protocols 2 and 3). These detection formats serve as the basis for detection of single DNA targets, and can be used for all the target preparations described in UNITS 8.2, 8.3 & 8.4 (e.g., metaphase spreads and interphase nuclei). Protocols are also provided for bright-field detection using the TSA Plus System (DNP-based), with either horseradish peroxidase (HRP; see Alternate Protocol 1) or alkaline phosphatase (AP; see Alternate Protocol 2) as the detector enzyme. Fluorescence detection of DNA or RNA in cultured cells is also presented (see Basic Protocol 4). Protocols for multitarget detection are provided for DNA (see Alternate Protocol 3), RNA (see Alternate Protocol 4), and DNA and RNA (see Alternate Protocol 5). The focus of all the protocols in this unit begins with the peroxidase-catalyzed covalent binding of the label and ends with the generation of the detected signal. However, it should be understood that these steps are actually the least critical factor in achieving satisfactory results. Therefore, it is essential that one has experience in procedures for preparing the samples (UNITS 8.2 & 8.5), for probe labeling, and for fluorescence situ hybridization (UNIT 8.3). Of particular importance are methods for using hapten-labeled probes and immunocytochemical incorporation of peroxidase (UNIT 8.4). When one chooses to enhance ISH detection with TSA, the additional steps required are incorporated into the standard ISH method. In this respect, the TSA protocols in this unit are presented as part of the whole ISH method, referencing other units wherever possible. NOTE: Experiments involving RNA require careful precautions to prevent contamination and RNA degradation. STRATEGIC PLANNING The decision to use TSA-Direct or TSA-Indirect amplification depends on whether fluorescence or bright-field in situ hybridization detection is desired (see UNIT 8.4), on the sensitivity required for the particular target, and on the number of steps one wants to add to the protocol. TSA-Direct is used for fluorescence detection and requires fewer steps, but, in general, is not as sensitive as the TSA-Indirect method. TSA-Indirect can be used for fluorescence or bright-field detection, and allows for multiple rounds of TSA amplification. Molecular Cytogenetics Contributed by Mark N. Bobrow and Philip T. Moen Jr. Current Protocols in Cytometry (2000) 8.9.1-8.9.16 Copyright © 2000 by John Wiley & Sons, Inc.
8.9.1 Supplement 11
Multitarget detection is facilitated by the covalently bound label attribute of TSA. The first, often more labile, target (e.g., RNA) is detected, leaving a covalently bound label in the location of the target. The detected target can then be denatured or enzymatically degraded to enable detection of the second target. Accurate optimization of hybridization and detection reagents is essential for best results (UNIT 8.4). Often the optimal probe dilution for TSA-based detection is ten-fold or more greater than for standard detection (see Critical Parameters). BASIC PROTOCOL 1
CHROMOGENIC DETECTION USING TSA-INDIRECT Horseradish peroxidase (HRP) is incorporated into the in situ hybridization product by any of the previously described methods (UNIT 8.4). The preparation is then incubated in biotinyl-tyramide and hydrogen peroxide (H2O2). In the presence of HRP and H2O2, biotinyl-tyramide becomes covalently bound to proteins proximal to the enzyme site. The reaction occurs via the phenolic group on tyramine, and biotin serves as the label that is subsequently detected. Detection of the covalently bound biotin is achieved by incubation with streptavidin-HRP, and chromogenic detection of HRP is accomplished by incubation with 3,3′-diaminobenzidine (DAB). Materials HRP-labeled ISH preparation: nucleic acid target hybridized with hapten-labeled probe (UNITS 8.2 & 8.3), followed by the immunocytochemical incorporation of HRP (UNIT 8.4) TSA-Indirect (ISH) Kit (NEN Life Sciences), including: Stock streptavidin-HRP reagent Blocking reagent (used for TNB buffer) Amplification diluent Tyramide stock reagent (see recipe): biotinyl-tyramide TNT buffer (see recipe) TNB buffer (see recipe) 2.5% (w/v) 3,3′-diaminobenzidine (DAB) solution (NEN Life Sciences) 30% (v/v) H2O2 (Sigma) DAB buffer (see recipe) Additional reagents and equipment for counterstaining, drying, and mounting the slides (UNITS 8.3 & 8.4) Perform TSA reaction 1. Wash an HRP-labeled ISH preparation three times, 5 min each, in a Coplin jar filled with TNT buffer at room temperature. 2. Prepare biotinyl-tyramide working solution by diluting the biotinyl-tyramide stock reagent 1/50 in amplification diluent (e.g., for 300 µl, use 6 µl biotinyl-tyramide reagent plus 294 µl amplification diluent). 3. Apply 100 to 300 µl biotinyl-tyramide working solution to the slide, cover with parafilm, and incubate 10 min at room temperature. 4. Wash as in step 1.
TSA Systems for the Enhancement of ISH Signals in Cytogenetics
Label with detector enzyme 5. Prepare streptavidin-HRP working solution by diluting the stock streptavidin-HRP reagent 1/100 in TNB buffer (e.g., for each 100 µl needed, use 1 µl stock streptavidinHRP plus 99 µl TNB buffer).
8.9.2 Supplement 11
Current Protocols in Cytometry
6. Apply 100 to 300 µl streptavidin-HRP working solution to the slide, cover with parafilm, and incubate 30 min at room temperature. 7. Wash as in step 1. Perform DAB reaction 8. Prepare DAB working solution by mixing the following: 0.2 ml 2.5% DAB solution 3.5 µl 30% H2O2 10 ml DAB buffer DAB working solution should always be prepared immediately before use.
9. Apply 300 µl DAB working solution to the slide and incubate 5 to 10 min at room temperature. 10. Wash, counterstain, dehydrate, and mount the slide as described in UNITS 8.3 & 8.4. CHROMOGENIC DETECTION USING THE TSA PLUS SYSTEM WITH HORSERADISH PEROXIDASE
ALTERNATE PROTOCOL 1
This indirect chromogenic protocol utilizes the TSA Plus system in lieu of the standard TSA-Indirect detection method. As with the standard TSA protocol (see Basic Protocol 1), HRP is incorporated into the in situ hybridization preparation by any of the previously described methods (UNIT 8.4). The Plus system utilizes a dinitrophenyl (DNP) amplification reagent instead of the biotinyl-tyramide reagent. Detection of the covalently bound DNP is achieved by incubation with HRP-labeled anti-DNP, and chromogenic detection of peroxidase is accomplished by incubation with DAB. Additional Materials (also see Basic Protocol 1) TSA Plus (HRP) System (NEN Life Sciences), including: Dinitrophenyl (DNP) stock amplification reagent (see product manual) 1× Plus amplification diluent Blocking reagent (used for TNB buffer) Horseradish peroxidase–labeled anti-DNP antibody (anti-DNP-HRP) 1. Perform TSA Plus reaction as described for the standard TSA reaction (see Basic Protocol 1, steps 1 to 4), but use DNP amplification reagent working solution in place of biotinyl-tyramide working solution. Prepare DNP amplification reagent working solution by diluting the stock solution 1/50 in 1× Plus amplification diluent (e.g., for 300 µl, use 6 µl stock solution plus 294 µl diluent). 2. Label with detector enzyme as described (see Basic Protocol 1, steps 5 to 7), but use anti-DNP-HRP working solution in place of streptavidin-HRP working solution. Prepare anti-DNP-HRP working solution by diluting the stock reagent 1/100 in TNB buffer (e.g., for each 100 µl needed, use 1 µl stock anti-DNP-HRP plus 99 µl TNB buffer). 3. Perform DAB reaction and prepare slide for viewing as described (see Basic Protocol 1, steps 8 to 10).
Molecular Cytogenetics
8.9.3 Current Protocols in Cytometry
Supplement 11
ALTERNATE PROTOCOL 2
CHROMOGENIC DETECTION USING THE TSA PLUS SYSTEM WITH ALKALINE PHOSPHATASE This indirect chromogenic protocol is performed using the TSA Plus system as in Alternate Protocol 1, except that alkaline phosphatase (AP) is used in place of HRP as the final detector enzyme. Chromogenic detection of AP is accomplished by incubation with 5-bromo-4-chloro-3-indolylphosphate/nitroblue tetrazolium (BCIP/NBT; UNIT 8.4). Additional Materials (also see Basic Protocol 1) TSA Plus (AP) System (NEN Life Sciences), including: Dinitrophenyl (DNP) stock amplification reagent (see product manual) 1× Plus amplification diluent Blocking reagent (used for TNB buffer) Alkaline phosphatase–labeled anti-DNP antibody (anti-DNP-AP) AP substrate solution (UNIT 8.4) 1. Perform TSA Plus reaction as described for the standard TSA reaction (see Basic Protocol 1, steps 1 to 4), but use DNP amplification reagent working solution in place of biotinyl-tyramide working solution. Prepare the DNP amplification reagent working solution by diluting the stock solution 1/50 in 1× Plus amplification diluent (e.g., for 300 µl, use 6 µl stock solution plus 294 µl diluent). 2. Label with detector enzyme as described (see Basic Protocol 1, steps 5 to 7), but use anti-DNP-AP working solution in place of streptavidin-HRP working solution. Prepare the anti-DNP-AP working solution by diluting the stock reagent 1/100 in TNB buffer (e.g., for each 100 µl needed, use 1 µl stock anti-DNP-AP plus 99 µl TNB buffer). 3. Prepare AP substrate solution just before use (UNIT 8.4). 4. Apply 300 µl AP substrate solution to the slide and incubate 10 min at room temperature. The 10 min time is only a recommended starting point. Optimal incubation time should be experimentally determined.
5. Wash, counterstain, dehydrate, and mount the slide as described in UNITS 8.3 & 8.4. BASIC PROTOCOL 2
TSA Systems for the Enhancement of ISH Signals in Cytogenetics
FLUORESCENCE DETECTION USING TSA-INDIRECT In this protocol, the HRP incorporation into the in situ hybridization and biotinyl-tyramide reaction is the same as in Basic Protocol 1. Detection of the covalently bound biotin is achieved by incubation with streptavidin that is labeled with a fluorophore. Materials HRP-labeled ISH preparation: nucleic acid target hybridized with hapten-labeled probe (UNITS 8.2 & 8.3), followed by the immunocytochemical incorporation of HRP (UNIT 8.4) TSA-Indirect (ISH) Kit (NEN Life Sciences), including: Blocking reagent (used for TNB buffer) Amplification diluent Tyramide stock reagent (see recipe): biotinyl-tyramide Fluorescent streptavidin conjugate: fluorescein, Texas Red, or coumarin (NEN Life Sciences), or any fluorescent streptavidin conjugates available from Molecular Probes (e.g., Alexa fluors) and other vendors (e.g., cyanine dyes) TNT buffer (see recipe) TNB buffer (see recipe)
8.9.4 Supplement 11
Current Protocols in Cytometry
Additional reagents and equipment for counterstaining, drying, and mounting the slides (UNITS 8.3 & 8.4) Perform TSA reaction 1. Wash an HRP-labeled ISH preparation three times, 5 min each, in a Coplin jar filled with TNT buffer at room temperature. 2. Prepare the biotinyl-tyramide working solution by diluting the biotinyl-tyramide stock reagent 1/50 in amplification diluent (e.g., for 300 µl, use 6 µl biotinyl-tyramide reagent plus 294 µl amplification diluent). 3. Apply 100 to 300 µl biotinyl-tyramide working solution to the slide, cover with parafilm, and incubate 10 min at room temperature. 4. Wash as in step 1. Label with detector enzyme 5. Prepare the fluorescent streptavidin conjugate working solution. For fluorescein, Texas Red, and coumarin conjugates, dilute the stock reagent 1/500 in TNB buffer (e.g., for each 500 µl needed, use 1 µl stock fluorescent streptavidin plus 499 µl TNB buffer). For other conjugates, appropriate dilutions need to be determined experimentally.
6. Apply 100 to 300 µl fluorescent streptavidin conjugate working solution to the slide, cover with parafilm, and incubate 30 min at room temperature. 7. Wash as in step 1. 8. Dehydrate, mount, and counterstain the slide (UNITS 8.3 & 8.4). FLUORESCENCE DETECTION USING TSA-DIRECT In this protocol, the HRP incorporation into the in situ hybridization preparation is the same as in Basic Protocol 1. The preparation is then incubated with fluorochrometyramide and hydrogen peroxide (H2O2). In the presence of HRP and H2O2, the fluorochrome-tyramide becomes covalently bound to the solid phase, proximal to the enzyme site. The reaction occurs via the phenolic group on tyramine, and the fluorochrome is subsequently detected. This method is not as sensitive as the indirect method, but it is simpler to perform.
BASIC PROTOCOL 3
Materials HRP-labeled ISH preparation: nucleic acid target hybridized with hapten-labeled probe (UNITS 8.2 & 8.3), followed by the immunocytochemical incorporation of HRP (UNIT 8.4) TSA-Direct (ISH) Kit (NEN Life Sciences), including: Amplification diluent Tyramide stock reagent (see recipe): fluorochrome-tyramide, e.g., fluorescein (green), tetramethylrhodamine (red), coumarin (blue), cyanine 3 (red), or cyanine 5 (far red) TNT buffer (see recipe) Additional reagents and equipment for counterstaining, drying, and mounting the slides (UNITS 8.3 & 8.4) 1. Wash an HRP-labeled ISH preparation three times, 5 min each, in a Coplin jar filled with TNT buffer at room temperature.
Molecular Cytogenetics
8.9.5 Current Protocols in Cytometry
Supplement 11
2. Prepare the fluorochrome-tyramide working solution by diluting the fluorochrometyramide stock reagent 1/50 in amplification diluent (e.g., for 300 µl, use 6 µl fluorochrome-tyramide reagent plus 294 µl amplification diluent). 3. Apply 100 to 300 µl fluorochrome-tyramide working solution to the slide, cover with parafilm, and incubate 10 min at room temperature. 4. Wash as in step 1. 5. Dehydrate, mount, and counterstain the slide (UNITS 8.3 & 8.4). BASIC PROTOCOL 4
FLUORESCENCE DETECTION OF DNA OR RNA IN CULTURED CELLS This protocol details the procedure for using TSA for DNA or RNA detection by fluorescence in situ hybridization to mammalian cells grown in vitro. The procedures leading up to the TSA detection steps differ from those previously used (UNITS 8.3, 8.4, & 8.5), and are provided here in greater detail. Materials 10 µg/ml labeled probe preparation 1 mg/ml Cot-1 DNA (Life Technologies) 10 mg/ml sheared salmon sperm DNA (Sigma) 10 mg/ml yeast tRNA (Sigma) Formamide (Sigma) RNA hybridization buffer (see recipe) Vanadyl ribonucleoside complex (VRC; New England Biolabs) DNA hybridization buffer (see recipe) Cells prepared for hybridization (see Support Protocol 1) 4× SSC (APPENDIX 2A), filtered through a 0.2-µm filter unit 2× SSC/50% (v/v) formamide (Sigma), pH 7.5, freshly prepared and prewarmed to 37°C 4× SSC/0.1% (v/v) Triton X-100 (Sigma) Speedvac evaporator 95°C heating block or water bath Forceps Humidified chamber Coplin jar Additional reagents and equipment for incorporating HRP into ISH signals (UNITS 8.3 & 8.4), for preparing working tyramide solution (see Basic Protocol 3), and for dehydrating, mounting, and counterstaining slides (UNITS 8.3 & 8.4) Prepare hybridization probe 1. Combine the following in a microcentrifuge tube: 2.5 to 5 µl 10 µg/ml labeled probe preparation (25 to 50 ng total) 1 µl 1 mg/ml Cot-1 DNA 1 µl 10 mg/ml sheared salmon sperm DNA 1 µl 10 mg/ml yeast tRNA. The increased sensitivity provided by TSA generally requires a decrease in the amount of DNA probe typically used for hybridization. This is primarily to prevent nonspecific signal. Whereas ∼5 ng of labeled probe (based on a ∼10-kb plasmid probe) would typically be used per 1 ìl hybridization solution, the authors recommend ∼2 to 3 ng for use with TSA.
TSA Systems for the Enhancement of ISH Signals in Cytogenetics
2. Evaporate just to dryness in a Speedvac evaporator. Avoid excessive drying, as it may render the probe difficult to resuspend.
3. Thoroughly resuspend the dried probe in 10 µl formamide.
8.9.6 Supplement 11
Current Protocols in Cytometry
4.
Denature the probe solution 10 min at 95°C. Quickly chill 5 min on ice.
Perform hybridization 5a. For RNA hybridization: Add 8 µl RNA hybridization buffer and 2 µl VRC to the chilled probe solution. 5b. For DNA hybridization: Add 10 µl DNA hybridization buffer to the chilled probe solution. 6a. For cells on microscope slides: Pipet the probe solution onto prepared cells and cover with a square of Parafilm. Using forceps, flatten to distribute the probe solution evenly over the surface and to remove bubbles. If desired, the slide can be covered by another square of Parafilm, which can be sealed around the edges.
6b. For cells on coverslips: Pipet the probe solution onto a sheet of Parafilm. Place prepared cells face down onto the probe solution. Using forceps, flatten to distribute the probe solution evenly and to remove bubbles. If desired, the coverslips (face down) can be covered by another square of Parafilm, which can be sealed around the edges.
7.
Hybridize in a humidified chamber at 37°C. Do not use a CO2 incubator, which will adversely affect the pH of the probe solution. Note that hybridization time will vary depending on the probe and target. Abundant RNA and DNA targets may be detectable after a two-hour hybridization. Single-copy genes and rare RNAs typically require overnight hybridization.
Perform posthybridization washes 8. Using forceps, remove the slide or coverslip and place into a Coplin jar containing 2× SSC/50% formamide preheated to 37°C. Incubate 30 min at 37°C while shaking. 9.
Decant the solution, replace with 1× SSC, and incubate 30 min at 37°C while shaking.
10. Decant the solution, replace with 0.5× SSC, and incubate 30 min at room temperature while shaking. Perform TSA detection 11. Incorporate HRP into the ISH signal (UNITS 8.3 & 8.4). For example, if biotin-labeled probes are used, incorporate with streptavidin-HRP; if digoxigenin-labeled probes are used, incorporate with anti-digoxigenin-HRP.
12. Wash three times, 15 min each, in 4× SSC/0.1% Triton X-100 at room temperature. 13. Wash 5 min in 4× SSC at room temperature. 14. Prepare working tyramide solution (see Basic Protocol 3, step 2). 15a. For cells on microscope slides: Pipet 100 to 300 µl working tyramide solution onto the slide and cover with a square of Parafilm. Incubate 10 min at room temperature. 15b. For cells on coverslips: Pipet 100 µl working tyramide solution onto a sheet of Parafilm and place the coverslip face down onto the solution. Incubate 10 min at room temperature. 16. Wash three times, 15 min each, in 4× SSC/0.1% Triton X-100 at room temperature. 17. Dehydrate, mount, and counterstain the slide (UNITS 8.3 & 8.4).
Molecular Cytogenetics
8.9.7 Current Protocols in Cytometry
Supplement 11
ALTERNATE PROTOCOL 3
FLUORESCENCE DETECTION OF MULTIPLE DNA TARGETS IN CULTURED CELLS In the TSA reaction, the catalytic activity of HRP covalently deposits the fluorochrome at the site of probe hybridization. This makes TSA an ideal detection method for multiple probe or hapten visualization. Although current TSA reagents do not allow for simultaneous distinct detection of multiple targets, sequential TSA application does allow detection of multiple targets, and further provides all the benefits of signal amplification. In this protocol, hybridization of two differentially labeled probes (i.e., biotin and digoxigenin) to DNA targets is followed by TSA-Direct reactions. Two different fluorochrome tyramides must be chosen, one for detecting each target. In general, the cyanine tyramides are the most sensitive, followed by tetramethylrhodamine, fluorescein, and coumarin. It is suggested that the most abundant signal be detected using the least sensitive tyramide. This protocol also presents a method of quenching HRP activity between rounds of sequential TSA reactions. Additional Materials (also see Basic Protocol 4) Two differentially labeled, 10 µg/ml probes Two appropriate horseradish peroxidase (HRP) conjugates (e.g., streptavidin-HRP and anti-digoxigenin-HRP) Two appropriate tyramide solutions (e.g., cyanine 3 tyramide and fluorescein tyramide; see Basic Protocol 3) Sodium acetate/sodium azide buffer (see recipe) H2O2 (Sigma) Perform dual hybridization 1. Prepare hybridization probe as described for a single target (see Basic Protocol 4, steps 1 to 4), but use 2.5 to 5 µl of each of two differentially labeled, 10 µg/ml probes (25 to 50 ng each). 2. Perform hybridization as described (see Basic Protocol 4, steps 5b, 6a or 6b, and 7). 3. Perform posthybridization washes as described (see Basic Protocol 4, steps 8 to 10). Perform sequential TSA detection 4. Incorporate the first HRP into the ISH signal (UNITS 8.3 & 8.4). Note that although two differentially labeled probes are simultaneously hybridized, one label is used to incorporate HRP followed by a fluorochrome tyramide reaction, after which the second label is used to incorporate HRP followed by a different fluorochrome tyramide reaction. For example, using both biotin- and digoxigenin-labeled probes together, streptavidin-HRP is incorporated and a cyanine 3 tyramide reaction is performed first. This is followed by an anti-digoxigenin-HRP incorporation and a fluorescein tyramide reaction. An HRP-inactivation step (detailed in step 10) is performed between the HRP incorporations to prevent carryover tyramide binding.
5. Wash three times, 15 min each, in 4× SSC/0.1% Triton X-100 at room temperature. 6. Wash 5 min in 4× SSC at room temperature. 7. Prepare the first working tyramide solution (see Basic Protocol 3, step 2) and treat cells as described (see Basic Protocol 4, step 15a or 15b). 8. Wash three times, 15 min each, in 4× SSC/0.1% Triton X-100 at room temperature. TSA Systems for the Enhancement of ISH Signals in Cytogenetics
9. Wash 5 min in 2× SSC at room temperature.
8.9.8 Supplement 11
Current Protocols in Cytometry
10. Inactivate the first HRP by adding H2O2 to sodium acetate/sodium azide buffer at a final concentration of 0.3% (v/v), placing 100 to 300 µl on the slide or coverslip, and incubating 15 min at room temperature. Wash three times, 5 min each, in either PBS or 2× SSC. The H2O2 must be added to the sodium acetate/sodium azide buffer immediately before it is used.
11. Rinse twice, 5 min each, with 2× SSC at room temperature. 12. Repeat steps 4 through 9, using an HRP conjugate compatible with the second label (step 4) and a differently labeled tyramide (step 7). 13. Dehydrate, mount, and counterstain the slide (UNITS 8.3 & 8.4). FLUORESCENCE DETECTION OF MULTIPLE RNA TARGETS IN CULTURED CELLS
ALTERNATE PROTOCOL 4
In this protocol, hybridization of two differentially labeled probes (i.e., biotin and digoxigenin) to RNA targets is followed by TSA-Direct reactions. As for detection of multiple DNA targets, two different fluorochrome tyramides must be chosen, one for detecting each target. In general, the cyanine 3 tyramide is the most sensitive, followed by tetramethylrhodamine, fluorescein, and coumarin. It is suggested that the most abundant signal be detected using the least sensitive tyramide. Additional Materials (also see Basic Protocol 4 and Alternate Protocol 3) Two differentially labeled, 10 µg/ml hybridization probes 1. Prepare hybridization probe as described for a single target (see Basic Protocol 4, steps 1 to 4), but use 2.5 to 5 µl of each of two differentially labeled, 10 µg/ml probes (25 to 50 ng each). 2. Perform hybridization as described (see Basic Protocol 4, steps 5a, 6a or 6b, and 7). 3. Perform posthybridization washes as described (see Basic Protocol 4, steps 8 to 10). 4. Perform sequential TSA detection as described for multiple DNA targets (see Alternate Protocol 3, steps 4 to 13). FLUORESCENCE DETECTION OF DNA AND RNA TARGETS IN CULTURED CELLS
ALTERNATE PROTOCOL 5
This protocol presents a method for the sequential detection of both RNA and DNA targets. TSA-Direct is an ideal method to facilitate the detection of RNA and DNA. It is suggested that the more labile RNA be detected first. During the TSA reaction, the fluorochrome chosen to detect RNA is covalently bound in the location of the RNA target; therefore, it is still present after procedures used for DNA detection. RNA degradation followed by denaturation then serves to facilitate specific detection of the DNA target and inactivate the peroxidase from the RNA-dependent tyramide reaction, preventing crosscontaminating deposition of the second tyramide. NOTE: When detecting both RNA and DNA targets in the same cells, the cells must be prepared as for RNA hybridization (see Support Protocols 1 and 2). 1. Perform RNA hybridization and TSA detection as described (see Basic Protocol 4, steps 1 to 16). Molecular Cytogenetics
8.9.9 Current Protocols in Cytometry
Supplement 11
2. Denature for DNA detection and dehydrate (see Support Protocol 1, steps 1c to 3c or 1d to 7d). Note that RNA degradation is suggested for specific detection of DNA. Treatment using an RNase cocktail containing several different RNase preparations will be the most effective for complete removal of RNA. Elimination of the RNA target followed by sample denaturation allows for the diffusion of the RNA-specific probe initially used, and inactivation of the RNA-dependant peroxidase. Therefore, the same hapten label used for detection of RNA can be used for detection of DNA as well, without cross-reactivity of tyramide deposition.
3. Hybridize to the DNA target (see Basic Protocol 4, steps 1 to 17). SUPPORT PROTOCOL 1
PREPARATION OF CULTURED CELLS FOR HYBRIDIZATION Hybridization to nondenatured cells allows for discrete detection of RNA, whereas denaturation allows for detection of denatured DNA as well as RNA. If discrete DNA detection is required, two protocol variations, base denaturation and RNase treatment, can be used to degrade the majority of RNA. Materials Fixed, permeabilized cells (see Support Protocol 2) Absolute and 70% (v/v) ethanol 2× SSC (APPENDIX 2A), filtered through a 0.2-µm filter unit 2× SSC/70% (v/v) formamide (deionized; Sigma), freshly prepared, pH 7.2 0.07 N NaOH/70% ethanol RNase cocktail (Ambion) 50 mM Tris⋅Cl, pH 8.0 (APPENDIX 2A)/5 mM MgCl2, filtered through a 0.2-µm filter unit For RNA hybridization (nondenaturing) 1a. Dehydrate cells in 70% ethanol for 5 min at −20°C. 2a. Dehydrate in absolute ethanol 5 min at room temperature. Air dry cells. It is critical to prevent RNase contamination of glassware. Either use segregated instruments, or eliminate RNase using RNase-Zap (Ambion) or similar.
For DNA hybridization (denaturing without RNA degradation) 1b. Rinse cells twice in 2× SSC, 5 min each, at room temperature. 2b. Denature cells by placing the slides or coverslips in 2× SSC/70% formamide for 2 min at 70°C. The denaturing solution is heated rapidly using a microwave oven to minimize pH changes.
3b. Quickly remove the slides and place in a jar containing ice-cold 70% ethanol. Dehydrate 5 min at −20°C. 4b. Dehydrate in absolute ethanol 5 min at room temperature. Air dry cells. For discrete DNA hybridization (denaturing with RNA degradation) Base denaturation method 1c. For cells stored in 70% ethanol, incubate in freshly prepared 0.07 N NaOH/70% ethanol 5 min at room temperature. For freshly fixed cells, partially dehydrate in 70% ethanol 5 min at room temperature first. TSA Systems for the Enhancement of ISH Signals in Cytogenetics
2c. Dehydrate cells twice in 70% ethanol, 5 min each, at −20°C. 3c. Dehydrate in absolute ethanol 5 min at room temperature. Air dry cells.
8.9.10 Supplement 11
Current Protocols in Cytometry
RNase treatment method 1d. Rinse twice with 2× SSC, 2 min each at room temperature. 2d. Prepare a 1:1000 (v/v) dilution of RNase cocktail in 50 mM Tris⋅Cl, pH 8.0/5 mM MgCl2. Approximately 50 ìl is needed for a 22-mm2 coverslip. For cells on microscope slides, the volume should be adjusted according to the size of the cell surface on the slide.
3d. Pipet RNase solution onto Parafilm. Place the coverslip face down on the solution. For slides, pipet the RNase solution onto the slide and cover with a square of Parafilm. Incubate for 30 min at room temperature. 4d. Rinse with 2× SSC twice, 5 min each at room temperature. 5d. Denature the cells in 2× SSC/70% formamide 2 min at 70°C. The denaturing solution is heated rapidly using a microwave oven to minimize pH changes.
6d. Quickly remove the slides and dehydrate the cells in 70% ethanol 5 min at −20°C. 7d. Dehydrate in absolute ethanol 5 min at room temperature. Air dry cells. FIXATION AND PERMEABILIZATION OF CULTURED CELLS Using standard methods, cultured cells are mounted (or grown) on an appropriate slide or coverslip for analysis. Note that cells can be prepared for hybridization either by first extracting with detergent and then fixing, or by first fixing and then extracting. For detection of DNA or nuclear RNA targets in situ, permeabilization followed by fixation generally yields better results. For detection of cytoplasmic RNAs, fixation followed by permeabilization is the preferred order.
SUPPORT PROTOCOL 2
Materials Cells grown on or deposited onto a microscope slide or glass coverslip HBSS (APPENDIX 2A) Permeabilizing solution (select one): CSK/Triton X-100 solution (see recipe) CSK/Triton X-100/VRC solution (see recipe) DPBS/Triton X-100/VRC solution (see recipe) 4% (w/v) paraformaldehyde solution (see recipe) DPBS (see recipe) For detection of DNA 1a. Rinse the cells twice in HBSS, 2 min each, at room temperature. 2a. Permeabilize cells in freshly prepared CSK/Triton X-100 solution 3 min at room temperature. 3a. Fix the cells in 4% paraformaldehyde solution 10 min at room temperature. Cells can be used immediately for hybridization; see Support Protocol 1 to prepare cells for hybridization. Otherwise, fixed cells can be stored in 70% ethanol at either 4° or −20°C until needed. Single-copy genes have been successfully detected in cells stored for >3 months.
For detection of nuclear RNA 1b. Rinse the cells twice in HBSS, 2 min each, at room temperature. 2b. Permeabilize cells in freshly prepared CSK/Triton X-100/VRC solution 3 min at room temperature.
Molecular Cytogenetics
8.9.11 Current Protocols in Cytometry
Supplement 11
3b. Fix the cells in 4% paraformaldehyde solution 10 min at room temperature. Cells can be used immediately for hybridization; see Support Protocol 1 to prepare cells for hybridization. Otherwise, fixed cells can be stored in 70% ethanol at either 4° or −20°C until needed. Nuclear RNAs have been successfully detected in cells stored for >3 months. Note, however, that RNA abundance is a factor.
For detection of cytoplasmic RNA 1c. Rinse the cells twice in HBSS, 2 min each, at room temperature. 2c. Fix the cells in 4% paraformaldehyde solution 10 min at room temperature. 3c. Wash twice in DPBS, 2 min each, at room temperature. 4c. Permeabilize in DPBS/Trition X-100/VRC solution 5 min at room temperature. 5c. Wash twice in DPBS, 2 min each, at room temperature. Cells can be used immediately for hybridization; see Support Protocol 1 to prepare cells for hybridization. Otherwise, fixed cells can be stored in 70% ethanol at either 4° or −20°C until needed. Cytoplasmic RNAs have been successfully detected in cells stored for >3 months. Note, however, that RNA abundance is a factor.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
CSK buffer Prepare a stock solution of 0.1 M monosodium piperazine-N,N′-bis(2-hydroxypropanesulfonic acid) (PIPES; Sigma). Adjust pH to 6.8 with NaOH. Prepare stock solutions of 1 M NaCl and 1 M MgCl2. Dissolve 51.4 g sucrose in 300 ml H2O. Add 50 ml of 0.1 M PIPES, 50 ml of 1 M NaCl, and 1.5 ml of 1 M MgCl2. Add H2O to final volume of 500 ml. Filter through a 0.4-µm filter, divide into aliquots, and store for up to 6 months at −20°C. CSK/Triton X-100 solution Dissolve 10% Triton X-100 solution (membrane grade; Roche Molecular Systems) in CSK buffer (see recipe) to a final concentration of 0.5% (v/v). Prepare immediately before use. CSK/Triton X-100/VRC solution Dissolve 10% Triton X-100 solution (membrane grade; Roche Molecular Systems) in CSK buffer (see recipe) to a final concentration of 0.5% (v/v). Freshly thaw vanadyl ribonucleoside complex (VRC; New England Biolabs) and add to a final concentration of 5% (v/v). Prepare immediately before use. Diaminobenzidine (DAB) buffer Dissolve 6.80 g imidazole and 6.05 g Tris base in 1 liter water. Adjust the pH to 7.6 with HCl and filter through a 0.2-µm filter unit. Store at room temperature for up to one year. Diethylpyrocarbonate (DEPC)–treated water Add 0.05% DEPC (Sigma) to water, shake vigorously, and autoclave. TSA Systems for the Enhancement of ISH Signals in Cytogenetics
8.9.12 Supplement 11
Current Protocols in Cytometry
DNA hybridization buffer 2 ml 50% dextran sulfate (Amersham Pharmacia Biotech) 1 ml 20× SSC (APPENDIX 2A) 1 ml 20 mg/ml BSA, ultrapure (Roche Molecular Systems) 1 ml H2O Store for up to several months at 4°C DPBS (Dulbecco’s phosphate-buffered saline) Dilute 10× Dulbecco’s phosphate-buffered saline without calcium chloride and magnesium chloride (Life Technologies) 1:10 in DEPC-treated water (see recipe), and filter through a 0.2-µm filter unit. Store for up to several months at room temperature. DPBS/Triton X-100/VRC solution Dissolve Triton X-100 (Sigma) to a final concentration of 0.5% (v/v) in DPBS (see recipe). Freshly thaw vanadyl ribonucleoside complex (VRC; New England Biolabs) and add to a final concentration of 5% (v/v). Prepare immediately before use. Paraformaldehyde solution, 4% (w/v) Add 20 g paraformaldehyde powder (Sigma; store at 4°C) to 400 ml H2O while stirring (the solution will be cloudy). Add 2.5 ml of 10 N NaOH and continue stirring until the solution clears (several minutes). Add 50 ml of 10× DPBS (see recipe) and 2.5 ml of 1 M MgCl2 (the solution will become cloudy again). Adjust the pH to 7.5 by adding concentrated HCl dropwise (the solution will become clear). Add water to 500 ml total volume, filter through a 0.2-µm filter unit, and store for up to 2 weeks at 4°C. Do not heat solution to dissolve paraformaldehyde.
RNA hybridization buffer 2 ml 50% dextran sulfate (Amersham Pharmacia Biotech) 1 ml 20× SSC (APPENDIX 2A) 1 ml 20 mg/ml BSA, ultrapure (Roche Molecular Systems) Store for up to several months at 4°C Sodium acetate, 3 M, pH 5.2 Dissolve 408 g sodium acetate trihydrate in 1 liter water. Adjust pH to 5.2 with acetic acid. Filter through a 0.2-µm filter unit. Store for up to several months at room temperature. Sodium acetate/sodium azide buffer Dilute 3 M sodium acetate buffer, pH 5.2 (see recipe) 1/30 with water. Add NaN3 to a final concentration of 0.1% (w/v). Store for up to several months at room temperature. Tris/NaCl/blocking reagent (TNB) buffer 0.1 M Tris⋅Cl 0.15 M NaCl 0.5% (v/v) blocking reagent (supplied in TSA kit; NEN Life Sciences) Dissolve Tris base and NaCl in deionized water and adjust pH to 7.5 with HCl. Add blocking reagent and heat the mixture 1 hr at 60°C with stirring. Divide into aliquots and store for up to 1 month at −20°C. Specific recipes are given in UNITS 8.3 & 8.4. Alternatively, premade 1 M Tris⋅Cl buffer (Life Technologies) can be used. Molecular Cytogenetics
8.9.13 Current Protocols in Cytometry
Supplement 11
Tris/NaCl/Tween (TNT) buffer 0.1 M Tris⋅Cl 0.15 M NaCl 0.05% (v/v) Tween 20 Dissolve Tris base and NaCl in deionized water and adjust pH to 7.5 with HCl. Add Tween 20 to a final concentration of 0.05%. Filter through a 0.2-µm filter unit. Store for up to several months at room temperature. Alternatively, premade 1 M Tris⋅Cl buffer (Life Technologies) can be used to prepare this reagent.
Tyramide stock reagent Dissolve solid tyramide reagents (biotin and fluorochrome) in dimethyl sulfoxide (DMSO) according to the package directions. The volume will depend upon the size of the kit and the particular tyramide used. Store at 4°C according to the package directions. COMMENTARY Background Information
TSA Systems for the Enhancement of ISH Signals in Cytogenetics
Historically, signal amplification methods for in situ hybridization have relied on forming layers or complexes of immunocytochemical detection reagents. The novel concept of having the catalytic activity of the detector enzyme cause the formation of bound label on the solid phase was termed catalyzed reporter deposition (CARD; Bobrow et al., 1989, 1991, 1992; Bobrow and Litt, 1993, 1996, 1998). TSA is a version of CARD in which the enzyme is peroxidase and the label is bound to a tyramidebased substrate. In the course of the reaction, a free radical is formed on the tyramide substrate. The free radical is unstable and reacts with electron-rich moieties in close proximity—i.e., to the enzyme or proteins adjacent to the enzyme. The first reported application of TSA for FISH utilized both direct (biotin) and indirect (fluorochrome) detection (Raap et al., 1995). Since the initial report, a multitude of publications have described the use of TSA systems in a variety of detection formats. Notable applications include the use of a multitude of fluorochrome-labeled tyramides (van Gijlswijk et al., 1997), hybridization with HRP-labeled oligonucleotide probes (van Gijlswijk et al., 1996b), silver-enhanced gold detection (Zehbe et al., 1997), and electron microscopy (Mayer and Bendayan, 1997). There are a wide variety of procedural options and detector combinations to incorporate HRP (see UNITS 8.3 & 8.4 for several options and discussions). There are also options for the TSA component. In the indirect format, standard TSA utilizes biotin as the label, and a new version, TSA Plus, utilizes dinitrophenyl
(DNP) as the label (Mayer and Nayak, 1999). The protocol for TSA Plus is the same as for standard TSA, except the immunocytochemical detection is with anti-DNP in lieu of streptavidin. In the direct format, several different fluorochromes are available (see Basic Protocol 3). UNITS 8.3 & 8.4 discuss the selection and characteristics of fluorochromes and the selection of a detection format.
Critical Parameters The most important parameter is user proficiency in standard ISH procedures. Expecting the added sensitivity of TSA to compensate for suboptimal hybridization and immunocytochemical detection will yield only suboptimal results. TSA should be utilized to enhance difficult detection, to allow for the use of shorter probes or of probes directly labeled with HRP, to reduce total assay time, and to facilitate multitarget detection. It is also critical to keep nonspecific (background) signal to a minimum. The necessity to optimize individual steps for standard ISH and immunocytochemical detection is discussed in UNITS 8.3 & 8.4. The addition of TSA amplification makes the requirement even more critical. The added steps require reoptimization of several parameters, most notably the probe concentration. Typically, probe concentration should be two- to ten-fold lower than the optimal concentration for the standard FISH hybridization. It is advisable to gain experience with an already optimized standard ISH detection method. By applying TSA and then varying the individual parameters, one will gain an understanding of the interdependence of each of the parameters in the protocol. Applying TSA to an
8.9.14 Supplement 11
Current Protocols in Cytometry
optimized ISH protocol will likely result in unacceptable background with little or no signal increase (if the method is already providing a nearly maximal dark signal, then it is unlikely that the amplified signal will be darker). On experimentally determining a new set of optimal conditions (e.g., varying the probe concentration), one should maintain the dark signal with a concomitant reduction in background to an acceptable level. Once experience is gained in understanding the factors that influence the quality of the final result, the user will be better able to utilize TSA for his or her particular benefit. An exemplary optimization strategy for the application of TSA to the detection of lowabundance mRNAs has recently been described (Yang et al., 1999). The ISH protocol utilized digoxigenin-labeled probes, anti-digoxigeninHRP, TSA-Indirect (biotinyl-tyramide), streptavidin–alkaline phosphatase, and BCIP/NBT for visualization. The authors found that addition of TSA to their standard protocol produced unacceptable nonspecific staining. On reducing the probe concentration, they were able to eliminate the nonspecific signal while maintaining an intense specific signal. Other parameters that were optimized were tissue permeabilization, RNase digestion, and probe length and complexity. Another critical issue is the trade-off between improved detection limits and spatial resolution. Signal can be increased immensely with TSA, but at the cost of poorer resolution or signal that bleeds to adjacent structures. In practice, submicron resolution has been achieved with TSA, while still maintaining significant signal amplification. Optimization, again, is the key to obtaining quality results. As an aid to improving resolution, dextran sulfate and polyvinyl alcohol have been used as additives during the TSA reaction (van Gijlswijk et al., 1996a).
Troubleshooting For troubleshooting sample preparation, probe labeling, in situ hybridization, and immunocytochemical detection, see UNITS 8.2, 8.3 & 8.4. Unacceptable results fall into three basic categories: low signal, excess signal, and high background. When low signal is encountered, potential remedies are to increase the probe complexity; increase the probe concentration; increase the concentration of the HRP conjugate; increase the HRP conjugate incubation time; increase the tyramide concentration; increase the tyra-
mide incubation time; use multiple rounds of TSA amplification; and use TSA Plus. It is important to determine that all of the reagents are functional. This is best done on a known system that is detectable with standard ISH methods. When excess signal is encountered, potential remedies are to decrease the probe complexity; decrease the probe concentration; decrease the concentration of the HRP conjugate; decrease the HRP conjugate incubation time; decrease the tyramide concentration; and decrease the tyramide incubation time. When high background is encountered, potential remedies are to decrease the concentration of the HRP conjugate; decrease the probe concentration; decrease the probe complexity; use or lengthen an endogenous peroxidase quenching step; increase the number and length of washes; and switch to a nonbiotin system, if biotin is being used. With TSA, low signal and high background are caused by excess reagents. Lowering the concentration of critical reagents (probe, HRP conjugate) will often reduce background and increase signal.
Anticipated Results Tyramide signal amplification consistently results in an increase in hybridization detection sensitivity. Using the methods detailed above, single-copy genes have been detected using probe sequences <1 kb in length. Similarly, low-abundance cytoplasmic RNAs and nuclear RNA transcripts have been detected, both alone and in combination with single-copy gene sequences.
Time Considerations The entire ISH protocol can take from several hours to several days (UNITS 8.3 & 8.4), depending on the number of targets and the complexity of the detection scheme. The TSA component (starting with the HRP conjugate incubation following probe hybridization, washes, and detection) can take from ∼1 hr for direct detection, and from 1.5 to 2 hr for indirect detection.
Literature Cited Bobrow, M.N. and Litt, G.J. March 1993. Method for the detection or quantitation of an analyte using an analyte dependent enzyme activation system. U.S. patent 5,196,306. Bobrow, M.N. and Litt, G.J. December 1996. Method for the detection or quantitation of an analyte using an analyte dependent enzyme activation system. U.S. patent 5,583,001.
Molecular Cytogenetics
8.9.15 Current Protocols in Cytometry
Supplement 11
Bobrow, M.N. and Litt, G.J. March 1998. Catalyzed reporter deposition. U.S. patent 5,731,158. Bobrow, M.N., Harris, T.D., Shaughnessy, K.J., and Litt, G.J. 1989. Catalyzed reporter deposition, a novel method of signal amplification. Application to immunoassays. J. Immunol. Methods 125:279-285. Bobrow, M.N., Shaughnessy, K.J., and Litt, G.J. 1991. Catalyzed reporter deposition, a novel method of signal amplification. Application to membrane immunoassays. J. Immunol. Methods 137:103-112. Bobrow, M.N., Litt, G.J., Shaughnessy, K.J., Mayer, P.C., and Conlon, J. 1992. The use of catalyzed reporter deposition as a means of signal amplification in a variety of formats. J. Immunol. Methods 150:145-149. Mayer, G. and Bendayan, M. 1997. Biotinyl-tyramide: A novel approach for electron microscopic immunocytochemistry. J. Histochem. Cytochem. 45:1449-1454. Mayer, P. and Nayak, Y. 1999. Recent developments in signal amplification. Am. Biotechnol. Lab. 17:18-20. Raap, A.K., van de Corput, M.P.C., Vervenne, R.A.W., van Gijlswijk, R.P.M., Tanke, H.J., and Wiegant, J. 1995. Ultra-sensitive FISH using peroxidase-mediated deposition of biotin- or fluorochrome-tyramides. Hum. Mol. Genet. 4:529-534. van Gijlswijk, R.P.M., Wiegant, J., Raap, A.K., and Tanke H.J. 1996a. Improved localization of fluorescent tyramides for fluorescence in situ hybridization using dextran sulfate and polyvinyl alcohol. J. Histochem. Cytochem. 44:389-392. van Gijlswijk, R.P.M., Wiegant, J., Vervenne, R., Lasan, R., Tanke H.J., and Raap, A.K. 1996b. Horseradish peroxidase-labeled oligonucleotides and fluorescent tyramides for rapid detection of chromosome-specific repeat sequences. Cytogenet. Cell Genet. 75:258-262. van Gijlswijk, R.P.M., Zijlmans, H.J.M.A.A., Wiegant, J., Bobrow, M.N., Erickson, T.J., Adler, K.E., Tanke, H.J., and Raap, A.K. 1997. Fluorochrome-labelled tyramides: Use in immunocytochemistry and fluorescence in situ hybridization. J. Histochem. Cytochem. 45:375-382.
Yang, H., Wanner, I.B., Roper, S.D., and Chaudhari, N. 1999. An optimized method for in situ hybridization with signal amplification that allows the detection of rare mRNAs. J. Histochem. Cytochem. 47:431-445. Zehbe, I., Hacker, G.H., Su, H., Hauser-Kronberger, C., Hainfeld, J.H., and Tubbs, R. 1997. Sensitive in situ hybridization with catalyzed reporter deposition, Streptavidin-Nanogold, and silver acetate autometallography. Am. J. Path. 150: 1553-61.
Key References Mayer and Bendayan, 1997. See above. Electron microscopy. Villnave-Johnson, C., Singer, R.H., and Lawrence, J.B. 1991 Fluorescent detection of nuclear RNA and DNA: Implications for genome organization. Methods Cell Biol. 35:73-99. In situ hybridization to DNA and RNA. Xing, Y., Johnson, C.V., Moen, P.T. Jr., McNeil, J.A., and Lawrence, J.B. 1995. Nonrandom gene organization: Structural arrangements of specific pre-mRNA transcription and splicing with SC35 domains. J. Cell Biol. 131:1635-1647. Simultaneous detection of DNA and RNA in situ. Yang, et al., 1999. See above. Optimization of hybridization and detection conditions. Zehbe et al., 1997. See above. Silver-enhanced gold detection.
Internet Resources http://www.nen.com The most current information for reagents, protocols, and applications.
Contributed by Mark N. Bobrow and Philip T. Moen Jr. NEN Life Science Products Boston, Massachusetts
TSA Systems for the Enhancement of ISH Signals in Cytogenetics
8.9.16 Supplement 11
Current Protocols in Cytometry
Molecular Combing
UNIT 8.10
Direct manipulation and visualization of single DNA molecules by fluorescence microscopy have advanced the understanding of the physical and biological properties of DNA (Strick et al., 1996). Uniform stretching of a large number of molecules obtained from a single sample of genomic DNA on a solid surface has led to several useful applications, particularly in molecular genetics and, more recently, in the study of DNA replication (Florijn et al., 1995; Herrick and Bensimon, 1999; Herrick et al., 2000a). This unit describes a process called molecular combing, which is an important advance in fiber-FISH technology. Molecular combing is a natural process, rather like long hair being pulled down a swimmer’s back as she emerges from a pool, and involves two basic steps: binding single DNA molecules by one or both extremities to a surface and using a receding meniscus to extend each molecule in a uniform and parallel manner over the surface (Bensimon et al., 1994). The principal advantage of this method is that reliable and quantitative information for genome-wide studies can be obtained without using other techniques. Because DNA molecules can be stretched on a surface in a controlled and uniform way, a large number of genomes can be combed and analyzed. Thus, a large number of accurate measurements can be obtained in a single experiment. The utility of molecular combing for high-resolution genomic studies is already well established (Michalet et al., 1997). The basic procedure for molecular combing (see Basic Protocol) is followed by alternate protocols that describe high-resolution physical mapping (see Alternate Protocol 1) and gene dosage approaches (see Alternate Protocol 2). The support protocols outline the silanization procedure for glassware (see Support Protocol 1), the preparation of DNA solutions (see Support Protocol 2), and the labeling of probes (see Support Protocol 3). DYNAMIC MOLECULAR COMBING OF DNA The following protocol describes the manner in which genomic DNA can be combed on treated glass surfaces. The procedure is identical for different sources of DNA of various lengths. Some of the properties of molecular combing are important; the action of the meniscus on the molecule is localized in the immediate vicinity of the interface. This property has the consequence that the stretching of a molecule anchored at one end is uniform (Allemand et al., 1997). The extension of the molecules is proportional to their number of base pairs and is sequence independent, as expected for homogeneous stretching, which is reasonable for the large forces involved (Bensimon et al., 1995). The specific binding of DNA by its extremities on various surfaces is a physical process controlled by the pH of the solution (Allemand et al., 1997), the optimization of which is surface dependent. Thus, molecular combing is a very general phenomenon; it does not require any modification of the molecule, is gentle to the molecules, and is easy to perform on a great variety of surfaces. The critical parameters are the quality of the DNA and the pH of the solution.
BASIC PROTOCOL
A combing apparatus is recommended, but any apparatus providing an adjustable and constant speed for lifting the surfaces out of the solution will be appropriate. Materials YOYO-1-stained DNA solution, pH 5.5 (see Support Protocol 2 for DNA preparation) Superglue Molecular Cytogenetics Contributed by Chiara Conti, Sandrine Caburet, Catherine Schurra, and Aaron Bensimon Current Protocols in Cytometry (2001) 8.10.1-8.10.23 Copyright © 2001 by John Wiley & Sons, Inc.
8.10.1 Supplement 16
Combing apparatus with 1.5-ml Teflon reservoirs (Pasteur Institute) Silanized glass surfaces (see Support Protocol 1) Inverted epifluorescence microscope Optical filter suitable for green fluorescence (FITC) Video camera (e.g., SIT 68, Dage-MTI) Video recording device (Panasonic) 60°C incubator Microscope slides 1. Gently pour the YOYO-1-stained DNA solution (from Support Protocol 2) into the Teflon reservoir specifically adapted for the combing apparatus. 2. Lower a vertical silanized glass surface into the DNA solution and incubate 5 min, to allow DNA molecules to anchor to the surface. 3. Pull out the surface using the combing device at the constant speed of 300 µm/sec. Fibers are combed on both sides of the surface.
4. Check the quality of the combing process with an inverted epifluorescence microscope equipped with a FITC filter and connected to a video camera. Use a 100× objective with immersion oil, but check only one side of the coverslip to keep one side clean for further use. Molecules should appear as green fibers similar to those seen in Figure 8.10.1.
5. Record some images for each surface, for further reference. 6. Dry the surface overnight in a 60°C incubator. This time can be reduced to 1.5 hr.
7. Superglue the surface (oiled side down) onto a glass microscope slide and store indefinitely at −20°C.
Figure 8.10.1 Human genomic combed DNA. Green fibers are molecules stretched and vertically aligned. YOYO-1 is an intercalating molecule that allows visualization of deproteinized DNA as green fibers using an epifluorescence microscope equipped with an optical filter for green fluorescence and an intensifying camera. Differences in fluorescence intensity can be attributed to local bundling of a few DNA molecules. The scale bar represents 25 µm.
Molecular Combing
8.10.2 Supplement 16
Current Protocols in Cytometry
HIGH-RESOLUTION GENOMIC MAPPING BY FLUORESCENCE IN SITU HYBRIDIZATION (FISH) ON COMBED DNA
ALTERNATE PROTOCOL 1
FISH was first performed on combed YAC DNA fragments that were isolated by pulsed-field gel electrophoresis (Weier et al., 1995). In these initial experiments, hybridization and fluorescent antibody detection were performed according to standard FISH protocols. The results of this study demonstrated that clones could be precisely ordered along a YAC molecule with the ability to directly measure the physical distances between the hybridization signals. A precision of a few kilobases was obtained from these measurements. Likewise, P1 clones were combed and cosmid probes were hybridized to the combed molecules to demonstrate the utility of combing for rapidly assembling subclones into a contig. The conditions required for FISH on combed human genomic DNA proved to be somewhat different. In such cases, classical FISH protocols yielded signals that were inadequate for quantification. To optimize probe signals from hybridized genomic DNA, standard FISH protocols had to be modified. It was also found that several consecutive layers of fluorophore-conjugated antibody are necessary for an optimal signal-to-noise ratio. It is not clear why stretched DNA requires these modifications, but it was found that only under these conditions could a statistically adequate number of measurable signals be obtained. Optimal conditions for labeling probes for hybridization to genomic DNA, on the other hand, are the same as in standard protocols. Adequate signals are obtained using either nick-translated probes or probes labeled by random primer extension (UNIT 8.3). However, for more sensitive assays that require a strong signal-to-noise ratio, nick-translated probes are preferable. These probes give less background fluorescence, and the signals are more uniform and complete. The application of molecular combing to high-resolution mapping follows the Basic Protocol for combing DNA and applies a FISH protocol to the resulting DNA fibers. Signals are collected and measured using CartographiX software (see Figure 8.10.2A and B). Histograms are obtained showing the distribution of probe lengths, gaps, or overlaps (see Figure 8.10.2C). Additional Materials (also see Basic Protocol) Probe cosmids, labeled with two (biotin and digoxigenin) or three (biotin, digoxigenin, and FITC) different haptens 10 mg/ml herring sperm DNA (Table 8.10.1) 1 mg/ml human cot-1 DNA (Table 8.10.1) 3 M sodium acetate, pH 5.2 (Table 8.10.1) 70%, 90%, and 100% ethanol, cold Hybridization buffer (Table 8.10.1) Slides with combed DNA stored at −20°C (from Basic Protocol) 50% and 70% (v/v) formamide in 2× SSC Rubber cement 2× SSC (Table 8.10.1) Blocking solution (Table 8.10.1) 0.05% (v/v) Tween 20 in 4× SSC Avidin coupled to Texas Red (Av-TR; Table 8.10.1) Mouse anti-digoxigenin coupled to FITC (mαdig-FITC; Table 8.10.1) Dilution buffer (DB; Table 8.10.1) Biotinylated anti-avidin (AAB; Table 8.10.1) Donkey anti-mouse coupled to FITC (dαm-FITC; Table 8.10.1)
Molecular Cytogenetics
8.10.3 Current Protocols in Cytometry
Supplement 16
A
B
digoxigenin probe, detected with FITC
biotinylated probe, detected with TR
Fluorescence intensity 700.0 600.0 500.0 400.0 300.0 200.0 100.0 0.0 644.5
C
0
25 50 75 100 125 150 175 200 225 250 275 300 325 Size (pixels)
Number of signals 38.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0
0
5
10
15
20
25
30
35
40
45
50 Size (µm)
size of the red probe size of the green probe size of the gap between the two probes Figure 8.10.2 Gene mapping analysis. (A) Image showing two probes hybridized on combed DNA and detected as two different colors. (B) CartographiX supplies a profile of fluorescence intensity related to each probe. The profile is used to establish the length of the probe. (C) Histogram obtained after measuring several red and green signals. CartographiX supplies a curve for each probe and for the gap between them. Molecular Combing
8.10.4 Supplement 16
Current Protocols in Cytometry
Mouse anti-rabbit coupled to FITC (mαrb-FITC; Table 8.10.1) Rabbit anti-FITC (F1; Table 8.10.1) Anti-rabbit coupled to FITC (F2; Table 8.10.1) Avidin coupled to AMCA (Av-AMCA; Table 8.10.1) Mouse anti-digoxigenin coupled to Texas Red (mαdig-TR) Anti-mouse coupled to digoxigenin (αm-dig) 1× PBS (APPENDIX 2A) Vectashield (Vector Labs) Transparent nail polish Speedvac evaporator (Savant) 76°C water bath Humid chamber (any tightly closing box protecting the slide from light and desiccation) Coverslips 37°C incubator Epifluorescence microscope equipped with a CCD video camera Image acquisition software (CartographiX, see Background Information) Specific software for data analysis (e.g., CartographiX, see Background Information) Prepare probes 1. Prepare a mix with 400 ng of each labeled probe cosmid, 1 µl of 10 mg/ml herring sperm DNA, and 1 mg/ml human cot-1 DNA (final 5 times the total quantity of probes) per slide. Add 1/10 total volume 3 M sodium acetate, pH 5.2, and 3 volumes ice-cold 100% ethanol. 2. Freeze 15 min at −75°C. 3. Centrifuge 15 min at 15,500 × g, 4°C. 4. Dry the pellet 5 min in a Speedvac evaporator. 5. Resuspend the pellet in 20 µl hybridization buffer per slide. 6. Denature the probes 5 min in boiling water and place on ice 10 min. Prepare slides 7. Take the slides with combed DNA out of −20°C and let warm to room temperature. 8. Heat 70% formamide in 2× SSC to 76°C. 9. Immerse slides 2 min in the heated 70% formamide in 2× SSC at 76°C. 10. Wash 3 min each in 70%, 90%, and 100% ethanol at room temperature. Allow to air dry. 11. Put 20 µl probe mix on the slide, cover with a coverslip, and seal with rubber cement. Incubate in a moist chamber overnight at 37°C. Remove coverslip prior to washing.
Wash probe 12. Wash the slides three times, 5 min each, in 50% formamide in 2× SSC. 13. Wash three times, 3 min each, in 2× SSC at room temperature. Molecular Cytogenetics
8.10.5 Current Protocols in Cytometry
Supplement 16
14. Add 40 µl blocking solution, cover with a coverslip, and incubate in a moist chamber ≥30 min at 37°C. 15. Rinse in 0.05% Tween 20 in 4× SSC at room temperature. For double amplification 16a. Dilute Av-TR 1:50 (v/v) and mαdig-FITC 1:50 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 20 min in a moist chamber at 37°C. 17a. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 18a. Dilute AAB 1:50 (v/v) and dαm-FITC 1:50 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 20 min in a moist chamber at 37°C. 19a. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 20a. Dilute Av-TR 1:50 (v/v) and mαrb-FITC 1:50 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 20 min in a moist chamber at 37°C. 21a. Wash three times for 3 min in 0.05% Tween 20 in 4× SSC at room temperature. 22a. Dilute AAB 1:50 (v/v) and F1 1:100 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 20 min in a moist chamber at 37°C. 23a. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 24a. Dilute Av-TR 1:50 (v/v) and F2 1:100 (v/v) in DB. Add 25 µl to each slide, cover with a coverslip, and incubate 20 min in a moist chamber at 37°C. 25a. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. For triple amplification 16b. Dilute Av-AMCA 1:10 (v/v), mαdig-TR 1:50 (v/v), and F1 1:100 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 15 min in a moist chamber at 37°C. 17b. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 18b. Dilute AAB 1:50 (v/v), αm-dig 1:50 (v/v), and F2 1:100 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 15 min in a moist chamber at 37°C. 19b. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 20b. Dilute Av-AMCA 1:10 (v/v), mαdig-TR 1:50 (v/v), and F1 1:100 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 15 min in a moist chamber at 37°C. 21b. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 22b. Dilute AAB 1:50 (v/v), αm-dig 1:50 (v/v), and F2 1:100 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 15 min in a moist chamber at 37°C. 23b. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 24b. Dilute Av-AMCA 1:10 (v/v) and mαdig-TR 1:50 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 15 min in a moist chamber at 37°C. 25b. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. Molecular Combing
8.10.6 Supplement 16
Current Protocols in Cytometry
Mount slides 26. Wash 3 min in 1× PBS at room temperature. 27. Add one drop of Vectashield. 28. Cover with a coverslip, dry any excess liquid, and seal with transparent nail polish. Scan and acquire images, measure length, and analyze 29. Scan the slides using an epifluorescence microscope equipped with a CCD camera. Analyze fluorescent signals using CartographiX (see Background Information) or other program that can measure combed fluorescent signals. 30. Measure combed fluorescent signal using a home-written program. Adobe Photoshop can be used for an estimation of signal length. Data are organized in a histogram of probe length, gap, or overlap (see Fig. 8.10.2C).
GENE DOSAGE BY MOLECULAR COMBING An approach for the quantification of subtle gains and losses of genomic DNA has been developed. This approach relies on the Basic Protocol and consists of applying fluorescence hybridization to the combed DNA by using probes to identify the amplified region (Herrick et al., 2000b). The quantity of each probe is increased to saturate target sequences on the surface. Measurements are then made on the linear hybridization signals to ascertain the exact size of the region. Genomic amplification ranging from subtle duplications as small as 50 kb in size to gross amplifications involving oncogenes or whole chromosomes can be precisely mapped and quantified with this approach. The principle behind probe length measurement (PLM) is described in Background Information.
ALTERNATE PROTOCOL 2
Additional Materials (also see Alternate Protocol 1) Probes specific for a control region and the target region, labeled with two different haptens Additional reagents and equipment for preparation of molecular combing (see Basic Protocol), DNA combing solution (see Support Protocol 2), and probe labeling (see Support Protocol 3) Prepare probes 1. Prepare a mix with either 700 ng of each labeled cosmid probe or 1 µg of each labeled BAC probe (see Support Protocol 3), 1 µl herring DNA, and human cot-1 DNA (five times the total quantity of probes) per slide. Add 1/10 volume 3 M sodium acetate, pH 5.2, and 3 volumes cold glacial ethanol. 2. Freeze 15 min at −75°C. Centrifuge 15 min at 15,500 × g, 4°C. 3. Dry the pellet 5 min in a Speedvac evaporator. 4. Resuspend the pellet in 20 µl hybridization buffer. 5. Denature the probes 5 min in boiling water and place 10 min on ice. Prepare slides 6. Take the slides with combed DNA out of storage at −20°C (see Basic Protocol) and let them warm to room temperature. DNA is combed at a very high density, to obtain as many genomes as possible on one surface. In fact, for a statistical reason, it is not possible to compile data obtained from more than one slide.
Molecular Cytogenetics
8.10.7 Current Protocols in Cytometry
Supplement 16
7. Heat 70% formamide in 2× SSC at 76°C. 8. Immerse slides 2 min in the heated 70% formamide in 2× SSC at 76°C. 9. Wash slides 3 min each in 70%, 90%, and 100% ethanol. Allow to air dry. 10. Put the probes on the slide, cover with a coverslip, and seal with rubber cement. Incubate in a humid chamber overnight at 37°C. Coverslips must be removed prior to washing.
Detect and amplify 11. Wash the slides three times, 5 min each, in 50% formamide in 2× SSC. 12. Wash three times, 3 min each, in 2× SSC at room temperature. 13. Add 40 µl blocking solution and incubate ≥30 min in a humid chamber at 37°C. 14. Rinse in 0.05% Tween 20 in 4× SSC. 15. Dilute Av-TR 1:50 (v/v) and mαdig-FITC 1:50 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 20 min in a humid chamber at 37°C. 16. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 17. Dilute AAB 1:50 (v/v) and dαm-FITC 1:50 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 20 min in a humid chamber at 37°C. 18. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 19. Dilute Av-TR 1:50 (v/v) and mαr-FITC 1:50 (v/v) in DB and add 25 µl to each slide. Cover with a coverslip and incubate 20 min in a humid chamber at 37°C. 20. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 21. Dilute AAB 1:50 (v/v) and F1 1:100 (v/v) in DB. Add 25 µl to each slide, cover with a coverslip, and incubate 20 min in a humid chamber at 37°C. 22. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 23. Dilute Av-TR 1:50 (v/v) and F2 1:100 (v/v) in DB. Add 25 µl to each slide, cover with a coverslip, and incubate 20 min in a humid chamber at 37°C. 24. Wash three times, 3 min each, in 0.05% Tween 20 in 4× SSC at room temperature. 25. Wash 3 min with 1× PBS. 26. Add one drop of Vectashield, cover with a coverslip, and press slightly to remove any excess. Seal with nail polish. Scan and acquire image 27. Produce digital images using an epifluorescence microscope connected to a video camera.
Molecular Combing
Scan successive lines of fields of view (FOVs) first with the FITC filter and then with the TR filter. This order has to be respected until the whole slide has been viewed. Image acquisition is performed using CartographiX and digital images are stored until measurements. Each fluorescent signal is measured on its recorded image using CartographiX. A specific application of this software, Length Ratio Analysis, allows one to present all data plotted on a curve (Figure 8.10.3) which represents the copy number of the target region per haploid genome as a function of the number of genomes that have been scanned (see Background Information).
8.10.8 Supplement 16
Current Protocols in Cytometry
2.5
2.0
1.5
1.0
0.5
0.0 0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
Figure 8.10.3 Gene dosage by probe length measurement performed in a control genome, where the target region is supposed to be not amplified. Horizontal axis: number of genomes that have been analyzed. Vertical axis: copy number of the target region per haploid genome. The curve oscillates when a few genomes have been measured, but converges after ∼20 genomes. The envelopes define the interval in which the curve can oscillate according to the number of analyzed genomes.
CLEANING AND SILANIZATION OF GLASS SURFACES The glass surfaces used for molecular combing are normal 22 × 22–mm glass coverslips. However, their cleanliness is very important and the recommended supplier provides the cleanest coverslips that the authors have tested thus far. Still, coverslips need further cleansing to suppress remaining dust that could cause nonhomogeneous silane treatment (see Critical Parameters, Combed DNA quality). Coverslips are cleansed with acetone, thoroughly rinsed, and dried well with medical oxygen. Residual water also causes nonhomogeneous silanization. The cleansing is achieved with ozone, obtained by UV irradiation of medical oxygen. Since silane is highly sensitive to oxygen, ozone is then replaced by argon, a neutral gas, before silane is injected.
SUPPORT PROTOCOL 1
NOTE: Care must be taken to avoid any dust on the coverslips. Handle the acetone and the oxygen in a well-ventilated area, such as a fume hood. Wear powder-free gloves throughout to avoid additional dust that could cause fluorescent background. Materials Acetone Pyrolyzed water 7-Octenyltrichlorosilane (silane; Table 8.10.1) Chloroform Ethanol 22 × 22–mm coverslips (ESCO) Ceramic coverslip holders Argon and oxygen supplies Flow metering units Gas hydration device
Molecular Cytogenetics
8.10.9 Current Protocols in Cytometry
Supplement 16
UV lamp covered with quartz Connecting tubing and appropriate valves Closed treatment cell (1-liter volume) Small Erlenmeyer (silane reservoir) 100-µl glass syringe (Hamilton) Clean coverslips 1. Select 22 × 22–mm glass coverslips for cleanliness and place on ceramic racks. Ceramic racks were chosen for facility of use and cleansing.
2. Incubate the racks 30 min to 2 hr in pure acetone at room temperature. 3. Set up a treatment system consisting of a closed treatment cell supplied with adjustable flows of oxygen, argon, and humidified argon and with a quartz-covered UV lamp for irradiation (see Fig. 8.10.4 for an example). 4. After the incubation, rinse the racks rapidly in three baths of pyrolyzed water. 5. Extensively dry off the coverslips and the racks with a flow of medical oxygen. 6. Place the racks in the treatment cell, along with little silane reservoirs (small Erlenmeyer flasks) wrapped with aluminum foil. Close the cell tightly. 7. Fill the cell by applying a 15 liter/min flow of oxygen for 1 min followed by an 8 liter/min flow for 5 min. 8. Reduce the flow to 0.160 liter/min and irradiate treatment cell for 1 hr with UV light in order to produce an ozone atmosphere, needed for additional cleansing. Silanize 9. Stop the oxygen flow and flush 10 min with 2 liter/min argon to remove the ozone. 10. Apply an argon flow of 1 liter/min with 10% hydration for 20 min. Stop the flow and leave to rest 10 min.
oxygen meter oxygen
dry argon meter
outlet
argon treatment cell humidified argon meter
Molecular Combing
Figure 8.10.4 Schematic representation of the device used for the treatment of the glass surfaces in gas phase. The oxygen supply and the argon supply are regulated by flow meters, allowing for fine tuning of the gas flow. The argon supply can be diverted to pass through a humidifier to provide the 10% humidity necessary for treatment. The gas supplies are connected to a closed cell equipped with a UV lamp.
8.10.10 Supplement 16
Current Protocols in Cytometry
11. Apply an argon flow of 0.9 liter/min for 10 min. 12. With a glass syringe inject 50 µl silane into each reservoir. Rinse the syringe three times in chloroform and three times in 100% ethanol at room temperature. 13. Close the cell and let the silane vaporize overnight (up to 72 hr). Handle the silane very quickly as it is highly sensitive to oxygen.
14. After the silane has vaporized, rinse the cell 5 min with an argon flow of 8 liter/min. 15. Take out the coverslip rack and leave 1 hr in a fume hood to dry. Take out the aluminum foil reservoirs and dip them quickly in water. 16. Package the coverslips in aluminum foil and store protected from dust until use. PREPARATION OF DNA SOLUTION Molecular combing of DNA fibers is a process independent of DNA origin. Thus, it has been possible to stretch lambda phage DNA, bacterial DNA, Saccharomyces cerevisiae, Xenopus laevi, and mammalian DNA. The different cells used successfully for combing mammalian genomic DNA are as follows.
SUPPORT PROTOCOL 2
Human fibroblasts HeLa and 4XY Human lymphoblastoids Human lymphocytes CEMCl3 Human RCC tumor cells Murine embryonic fibroblasts Hamster fibroblasts 42 In addition, the various chromosomes in yeast can be sorted by pulse-field gel electrophoresis. The important point is to extract unsheared DNA from these cells, and the protocol to obtain intact DNA varies with cell type until staining of DNA with YOYO-1. This unit provides different protocols to prepare high-molecular weight DNA from human and yeast cells. These protocols are based upon the inclusion of cells in an agarose block prior to enzymatic digestion of the cellular structure instead of mechanical manipulation of DNA during extraction. NOTE: Wear non-powdered gloves throughout. Materials Desired cells 1× PBS, pH 7.4, ice cold Low melting point agarose (LMP agarose; FMC Bioproducts) EDTA/sarcosyl/proteinase K buffer (ESP solution; Table 8.10.1) TE/phenylmethylsulfonyl fluoride solution (TE/PMSF; Table 8.10.1) 0.5 M EDTA, pH 8.5 (APPENDIX 2A) Sorbitol/citrate/EDTA buffer (SCE; Table 8.10.1) Zymolyase 20T and buffer (Table 8.10.1) 14.3 M β-mercaptoethanol Dodecyl lithium sulfate buffer (DLS; Table 8.10.1) 1× TE (Table 8.10.1) 1.3% (w/v) agarose gel in 0.5× TBE 0.5× TBE (Table 8.10.1) 1 mM YOYO-1 in T40 E2 (Table 8.10.1) 10× β-agarase 1 buffer (NEB) β-agarase 1
Molecular Cytogenetics
8.10.11 Current Protocols in Cytometry
Supplement 16
0.5 M 2-[N-morpholino]ethane sulfonic acid (MES; Table 8.10.1), pH 5.5 37°C and 50°C waterbaths 100-µl pulsefield block molds (BioRad) Rotating wheel 15-ml tubes 2-ml round-bottom Eppendorf tubes Prepare mammalian genomic DNA 1a. Count cells (APPENDIX 3A) and resuspend in ice-cold 1× PBS at 2 × 106 cells/ml. Place on ice. The cell concentration has to be determined for each application. A low concentration is used when individual fibers must be dissociated; high concentrations are used when a high number of signals per slides is needed (gene dosage). See Commentary for explanations about the applications.
2a. Incubate cell suspension 5 min in 1× PBS at 37°C. 3a. Add an equal volume of melted LMP agarose and resuspend cells quickly. 4a. Pipet cell suspension into 100-µl block formers (106 cells/block). 5a. Chill molds 45 min at 4°C to set. 6a. Knock out each set block into at least 250 µl predigested ESP solution (preheated 2 hr at 50°C). Incubate 24 hr at 50°C. 7a. On the next day, change ESP solution and incubate another 24 hr with freshly predigested ESP solution. 8a. Rinse blocks three times with 0.5 M EDTA, pH 8.5. 9a. Wash three times, 30 min each, in 50 ml TE/PMSF solution at 50°C. 10a. Discard the last TE/PMSF wash and replace with 0.5 M EDTA, pH 8.5. Blocks can be stored indefinitely at 4°C in EDTA 0.5 M, pH 8.5.
Prepare total yeast DNA 1b. Count cells, centrifuge 2 min at 2000 rpm, and resuspend in SCE at 8 × 109 cells/ml on ice. 2b. Warm yeast cells 3 min at 38°C. Prepare molten 1% LMP agarose and keep at 45°C. 3b. Add an equal volume of agarose to the cell solution and pipet up and down to mix. Add 10 U/ml zymolyase 20T. 4b. Make 100-µl blocks (to obtain ∼4 × 107 cells/ml). Chill blocks 45 min at 4°C to set. 5b. Incubate in SCE buffer with 0.126% of 14.3 M β-mercaptoethanol and 10 U/ml zymolyase 20T (100 µl stock solution). 6b. Incubate 24 hr at 37°C. Discard the solution and repeat steps 5 and 6. 7b. Rinse blocks in 20 ml of 0.5 M EDTA, pH 8.5. 8b. Discard, add 20 ml DLS, and incubate 24 hr at 50°C. Repeat. 9b. Replace the solution with 45 ml of 1× TE. Wash 2 hr on a rotating wheel, room temperature. Repeat. Molecular Combing
10b. Discard TE and store in 0.5 M EDTA, pH 8.5, 4°C. Color of blocks should be transparent.
8.10.12 Supplement 16
Current Protocols in Cytometry
Prepare single yeast chromosome 1c. Prepare a pulsed-field gel electrophoresis at 1.3% agarose to run yeast genomic DNA in 0.5× TBE. 2c. Run gel 16 hr at 190 V, 231 mA (pulse of 60 min). 3c. Cut the band containing the chromosome of interest from the agarose gel and store in 0.5 M EDTA, pH 8.5. 4c. Proceed to step 11. The agarose gel fragment is further referred to as “block,” by analogy with the two preceding protocols.
Stain DNA with YOYO-1 11. Wash one block in 1× TE overnight (or ≥2 hr) in a 15-ml tube on a rotating wheel. 12. Discard the supernatant and place the block in a 2-ml round-bottom Eppendorf tube. 13. Add 10 µl of a 1:10 dilution of YOYO-1 in T40E2 and 160 µl T40E2 in order to cover the block. 14. Incubate 45 min at room temperature. Digest agarose 15. Discard supernatant, wash with 1× TE, and add 150 µl of 1× TE. 16. Melt the agarose 20 to 30 min at 68°C. Melting of agarose is performed at 65°C for DNA labeled with thymidine analogs to limit damaging impact on labels. Low molecular weight DNA, such as cosmids or λ DNA, is already in solution and is directly stained with YOYO-1.
17. Add 25 µl (1/10 total volume) prewarmed 10× β-agarase 1 buffer. Do not mix. In this and the following steps, avoid any kind of mechanical mixing (inverting the tube, stirring, vortexing), since it increases the shearing of the DNA.
18. Incubate 15 min at 40°C to equilibrate the temperature of the solution. 19. Add 2 µl β-agarase 1 and incubate 4 hr or overnight at 40°C. Do not mix. 20. Add 0.5 M MES, pH 5.5, up to 1.5 ml. Do not mix. The final volume of MES to add depends on the specific application, according to the needed concentration of combed molecules.
21. Heat the DNA solution 30 min at 65°C. 22. Allow the DNA solution to cool to room temperature. The DNA solutions can be stored up to 10 to 12 months at 4°C for later use.
Molecular Cytogenetics
8.10.13 Current Protocols in Cytometry
Supplement 16
SUPPORT PROTOCOL 3
PROBE LABELING The probes used for FISH on combed DNA can be labeled with various haptens, mainly biotin-16-dUTP and digoxigenin-11-dUTP, using random priming or PCR. The resulting labeled fragments are purified and quantified by gel electrophoresis. The quality of the labeling is assessed with an alkaline phosphatase assay by comparison with control labeled DNA. Materials Bioprime DNA labeling system (Life Technologies), including: 2.5× random primers Klenow fragment 10× dNTP mix (with biotin-16-dUTP included) Extracted DNA (cosmids, BAC, YAC) 0.5 M EDTA, pH 8.0 (APPENDIX 2A) 3 M sodium acetate, pH 5.2 (Table 8.10.1) 100% ethanol 1× TE (Table 8.10.1) Digoxigenin-11-dUTP 4.4 M lithium chloride (LiCl; Table 8.10.1) Fluorescein-11-dUTP (FluoroGreen, Amersham) QIAquick PCR purification kit (QIAGEN) 100 mM dATP Li-salt, 100 mM dCTP Li-salt, 100 mM dGTP Li-salt, 100 mM dTTP Li-salt in distilled water UltraPure agarose (Life Technologies; also see Table 8.10.1) 1× TBE buffer (Table 8.10.1) OrangeG (Sigma; Table 8.10.1) 0.25 µg/µl λ DNA 10 mg/ml ethidium bromide (Table 8.10.1) 5 ng/µl dig-labeled control DNA BLUGENE nonradioactive nucleic acid detection system (Life Technologies) Hybond membranes (Amersham Pharmacia Biotech) 6% BSA (Table 8.10.1) 0.4 N NaOH 5× SSC (Table 8.10.1) 1× PBS Anti-digoxigenin-AP Fab fragments (Table 8.10.1) AP-DB buffer (Table 8.10.1) Bromochloroindolyl phosphate (BCIP; Table 8.10.1) Nitroblue tetrazolinium (NBT; Table 8.10.1) Speedvac evaporator Electrophoresis apparatus UV transilluminator Label probe with biotin-16-dUTP 1a. Prepare a mixture with 20 µl 2.5× random primers from the Life Technologies kit, 500 ng DNA to label, and distilled water up to 44 µl. 2a. Denature the solution 8 min at 100°C. 3a. Keep 10 min on ice.
Molecular Combing
8.10.14 Supplement 16
Current Protocols in Cytometry
4a. Add 1 µl Klenow fragment and 5 µl 10× dNTP mix with biotin-16-dUTP from the Life Technologies kit. 5a. Incubate 4 hr (or overnight) at 37°C. Add 2 µl of 0.5 M EDTA to stop the reaction. 6a. Precipitate the probes by adding 1/10 volume 3 M sodium acetate, pH 5.2, and three volumes cold 100% ethanol. 7a. Keep 15 min at −75°C. Centrifuge 15 min at 15,500 × g. 8a. Dry the pellet 5 min in a Speedvac evaporator and resuspend in 50 µl 1× TE. Probe labeling with Alexa nucleotides consists of using the Bioprime labeling kit but substituting the Alexa dNTP for the biotin-16-dUTP.
Label probe with digoxigenin-11-dUTP 1b. Prepare a mixture with 20 µl of 2.5× random primers from the Life Technologies kit, 500 ng DNA to label, and distilled water up to 42 µl. 2b. Denature the solution 5 min at 100°C. Keep 10 min on ice. 3b. Add 1 µl Klenow fragment from the Life Technologies kit, 2 µl of 1 mM digoxigenin11-dUTP, and 5 µl 10× dNTPs mix (1 µl of 100 mM dATP, 1 µl of 100 mM dCTP, 1 µl of 100 mM dGTP, 6.5 µl of 10 mM dTTP, 90.5 µl distilled water). 4b. Incubate 4 hr (or overnight) at 37°C. Add 2 µl of 0.5 M EDTA to stop the reaction. 5b. Precipitate the probes by adding 1/10 volume 4.0 M LiCl and 3 volumes cold 100% ethanol. 6b. Keep 15 min at −75°C. 7b. Centrifuge 15 min at 13,000 × g. 8b. Dry the pellet 5 min in a Speedvac evaporator and resuspend in 50 µl of 1× TE. Label probe by fluorescein-11-dUTP Probes are labeled by random priming (Bioprime DNA labeling system, Life Technologies) with fluorescein-11-dUTP (FluoroGreen, Amersham Pharmacia Biotech) and purified with the QIAquick PCR purification kit (Qiagen). Each probe of 900 ng is ethanol precipitated with 12 times excess 1 mg/ml human cot-1 DNA (Life Technologies) to be hybridized (see Alternate Protocol 1). Label by PCR Prepare the mix with 140 µM dCTP, 60 µM dCTP-14-biotin (or 130 µM dTTP), and 70 µM dUTP-dig. Follow manufacturer’s instructions. Labeled nucleotides will replace the unlabeled ones in the dNTP mix used for random priming.
Quantify labeled probe 9. Prepare a 0.6% gel with UltraPure agarose in 1× TBE and keep 0.5 hr at 4°C to solidify. 10. Prepare a mix with 4 µl labeled DNA and 2 µl OrangeG for the 0.25 µg/µl λ control DNA and each probe. 11. Charge the gel and run 1 hr at 50 V. Molecular Cytogenetics
8.10.15 Current Protocols in Cytometry
Supplement 16
12. Stain 3 min with ethidium bromide (65 µl stock solution in 300 ml distilled water) and wash 1 min in distilled water. Run 0.5 hr more. 13. Verify the gel under a UV transilluminator and measure ethidium bromide fluorescence to quantify DNA. Verify label with alkaline phosphatase assay For digoxigenin-labeled probe 14a. Prepare 4 successive 1:3-dilutions of the 5-ng/µl dig-labeled control DNA with 1 µl sample and 2 µl water. Prepare a dilution of the labeled probe to get a concentration of 5 ng/µl and 4 successive 1:3-dilutions with 1 µl sample and 2 µl water. For a biotinylated probe 14b. Prepare 4 successive 1:2-dilutions of the 0.2-ng/µl labeled control (DNA BLUGENE nonradioactive nucleic acid detection system) with 1 µl sample and 1 µl water. Prepare a dilution of the labeled probe to get a concentration of 0.2 ng/µl and 4 successive 1:2-dilutions with 1 µl sample and 1 µl water. 15. Charge the Hybond membrane with 1 µl of each dilution of either the biotinylated or digoxigenin probe and its own control DNA. 16. Incubate the membrane 1 hr at 60°C to fix the DNA. 17. Wash the membrane 20 min in 10 ml of 0.4 N NaOH and several times in 5× SSC, to get a pH of 7. 18. Add 10 ml of 6% BSA and incubate 1 hr (or overnight) at 60°C. 19. Add 10 µl SA-AP conjugate (DNA BLUGENE nonradioactive nucleic acid detection system) to biot-probes and 10 µl anti-digoxigenin-AP Fab fragments to dig-probes. 20. Incubate 30 min at 37°C. 21. Wash the membrane five times, 10 min each, with 1× PBS. 22. Prepare a mixture with 7 ml AP-DB, 25 µl BCIP, and 33 µl NBT. 23. Add this mix to the membrane and incubate ∼8 min for the dig-probes and 15 min for the biot-probe at 37°C. 24. Wash with distilled water and add 2 ml of 0.5 M EDTA, pH 8. Dry. Stop the incubation with the enzyme when the color of spots corresponding to the control DNA decreases. The probes are correctly labeled if the intensity of the spots is equal or superior to that of the spots for control labeled DNA.
Molecular Combing
8.10.16 Supplement 16
Current Protocols in Cytometry
8.10.17
Current Protocols in Cytometry
Supplement 16
1.3% 1:50 (v/v) 75 U/10 ml 1:10 (v/v)
RT −20°C +4°C −20°C
+4°C
1000 U/ml 10× Powder 1.6 mg/ml 150 U/200 µl
β-agarase 1
β-agarase 1 buffer
Agarose
Anti-digoxigenin made in mouse, coupled to FITC (mαdig-FITC)
Anti-digoxigenin-AP Fab fragments
Avidin coupled to 5 mg/ml 7-amino-4-methylcoumarin-3-acetic acid (Av-AMCA)
Anti-FITC made in rabbit (F1)
-
AP-DB buffer
Avidin coupled to Texas Red (Av-TR)
2 mg/ml
1:50 (v/v)
−20°C
1.5 mg/ml
Anti-rabbit made in mouse, coupled to FITC (mαrb-FITC)
Argon
1:50
−20°C
200 µg/ml
Anti-mouse made in sheep, coupled to digoxigenin (αm-dig)
−20°C
RT 1/50 v/v
-
1:50 (v/v)
−20°C
1.4 mg/ml
Anti-mouse made in donkey, coupled to FITC (dαm-FITC)
RT
1:100 (v/v)
+4°C
1:100 (v/v) or 1:400 (v/v)
1×
8 U/ml
Anti-rabbit made in goat, coupled to FITC (F2)
0.4 mg/ml
-20°C
−20°C
4°C
Powder
ABS
6%
1:50 (v/v)
−20°C
Stock solution
Anti-avidin made in goat, coupled 0.5 mg to biotin (AAB)
Name
Concentration in reaction
Reagents and Solutions
Storage temperaturea
Table 8.10.1
DB (see recipe)
-
DB (see recipe)
DB (see recipe)
DB (see recipe)
DB (see recipe)
DB (see recipe)
DB (see recipe)
6% BSA
DB (see recipe)
1× TBE
TE
10× NE buffer
1× PBS
DB (see recipe)
Reaction buffer
37°C
RT
-
37°C
37°C
37°C
37°C
37°C
37°C
37°C
37°C
RT
40°C
40°C
RT
37°C
continued
Made fresh
Temperaturea Pre-treatment
8.10.18
Supplement 16
Current Protocols in Cytometry
Powder
See manufacturer’s instructions 1 mM dATP, 1 mM dCTP, 1 mM dGTP, 0.65 mM dTTP 1:2 (v/v) 0.05% Tween 20 in 4× SSC/blocking solution 1 nmol/µl Freshly made 100 mM NaCl, 50 mM EDTA, 100 mM Tris, 1% dodecyl lithium sulfate, water, pH 8 4 ml proteinase K (10 mg/ml), 1% Freshly made N-lauroylsarcosine, 10 mM Tris 10 mg/ml
Blocking solution
BLUGENE nonradioactive nucleic acid detection system
dNTP
DB (dilution buffer)
Digoxigenin-11-dUTP
Dodecyl lithium sulfate (DLS)
EDTA/sarcosyl/proteinase K buffer (ESP)
Ethidium bromide
Hybridization buffer (see recipe)
10 µg per reaction
−20°C
Herring DNA
10 mg/ml
See manufacturer’s Water instructions
−20°C
Fluorescein-11-dUTP
37°C
37°C
RT
60 µl in 300 ml water 0.5%
+4°C
50°C
0.5 M EDTA, pH 8
37°C
−20°C
Water
37°C
37°C
37°C
Freshly made 2 nmol/50 µl
1:10 (v/v)
−20°C
Water
See manufacturer’s 6% BSA instructions
37°C
1.5%
RT
continued
30 min at 50°C
2 hr at 68°C
Made fresh
Temperaturea Pre-treatment
37°C 0.05% Tween 20 in 4× SSC, pH 7.2
TP detection buffer (see recipe)
Reaction buffer
See manufacturer’s instructions
Concentration in reaction
+4°C
RT in powder, −20°C when diluted
−20°C
See manufacturer’s instructions
Bioprime DNA labeling system
Storage temperaturea −20°C
Stock solution
Reagents and Solutions, continued
Bromochloroindolyl phosphate (BCIP)
Name
Table 8.10.1
8.10.19
Current Protocols in Cytometry
Supplement 16
RT
4.4 M 1% 14.3 M 0.5 M, pH 5.5 3 M, pH 5.2
Lithium chloride (LiCl)
LMP agarose
β-mercaptoethanol
2-[N-morpholino]ethane sulfonic acid (MES)
Sodium acetate
RT
100% Powder 10×, pH 7.4 20×
7-Octenyltrichlorosilane (silane)
OrangeG
PBS
SSC
RT
RT
+4°C
−20°C
Nitroblue tetrazolinium (NBT)
RT
Powder, at RT
RT
+4°C
−20°C
Labeled probes
12.5 ml deionized formamide, 2.5 −20°C ml Tween 20, 6.25 ml dextran sulfate, 2.5 ml 20× SSC, up to 25 ml with water
Hybridization buffer
Hybridization buffer (see recipe)
Reaction buffer
5×, 2×, 4× 0.05% Tween 20
1×
0.35% (w/v)
100%
1:10 (v/v)
0.126%
0.5%
1:10 (v/v)
Water
Water
30% (w/v) glycerol in water
TP detection buffer
Mix for hybridization solution
Water
SCE (see recipe)
1× TBE
Ethanol
Cosmids: 400 ng Hybridization buffer on yeast, 700 ng on (see recipe) human BAC: 1 µg
20 µl per surface
5× quantity of probes
−20°C
1 mg/ml
Human cot-1 DNA
Concentration in reaction
Storage temperaturea
Stock solution
Reagents and Solutions, continued
Name
Table 8.10.1
RT
−20°C
RT
37°C
-
+4°C
37°C
-
−70°C
37°C
37°C
37°C
continued
42°C
Precipitated in 100% ethanol and high salt concentration
Temperaturea Pre-treatment
Supplement 16
TE (see recipe)
4 ml 1 M Tris, 0.4 ml 0.5 M RT EDTA, up to 100 ml with distilled water; autoclave 40 mg/ml in isopropanol 100 mM Tris⋅Cl, 100 mM NaCl, 50 mM MgCl2 1.3% 1 mM Powder
0.126% (v/v) SCE buffer, 12.6 µl RT β-mercaptoethanol in 100 ml final 10 U/ml zymolyase 20T
T40E2
Phenylmethylsulfonyl fluoride solution (PMSF)
AP-DB buffer
UltraPure agarose
YOYO-1
Zymolyase 20T
Zymolyase buffer
aAbbreviation: RT, room temperature.
Water
10 ml 1 M Tris, 2 ml 0.5 M EDTA, up to 1 liter with distilled water; autoclave
1 µl in 150 µl
1:1000 (v/v)
+4°C in powder, 1 U/µl −20°C in solution
+4°C
RT
RT
RT
RT
0.5×, 1×
50 mM Tris⋅Cl, pH 7.4, 1 mM EDTA, pH 8, 50% glycerol, sterilize
T40E2 (see recipe)
1× TBE
Water
Water
TE
RT
Stock
20×
Freshly made
TBE
Reaction buffer
1 M sorbitol, 10 mM EDTA, 100 mM sodium citrate, up to 100 ml water, pH 8.5
Concentration in reaction
Sorbitol/citrate/EDTA buffer (SCE)
Storage temperaturea
Stock solution
Reagents and Solutions, continued
Name
Table 8.10.1
Molecular Combing
8.10.20
Current Protocols in Cytometry
37°C
RT
50°C
RT
-
RT
37°C
Temperaturea Pre-treatment
COMMENTARY Background Information The “drop method” and “dynamic molecular combing” (DMC) Molecular combing is based on the use of a receding interface to stretch single deproteinized DNA molecules. Molecules are anchored at one end to a silanized coverslip while a meniscus (the interface between air and water) stretches them by applying a hydrodynamic force. This force is greater than the entropic forces that keep the DNA in its random coil configuration. The first method used for molecular combing was the drop process. A drop of DNA solution is deposited on a silanized coverslip, and an untreated coverslip is allowed to float on top of it. As the solution evaporates, the receding meniscus extends the bound molecules (Bensimon et al., 1994). Dynamic molecular combing (DMC) is faster than the drop method. Silanized coverslips are dipped into a buffered solution containing DNA and after 5 min of incubation are pulled out at a vertical constant speed of 300 µm/sec. The meniscus exerts a constant force on the immersed part of molecules, while the emerged part comes out with the surface (Michalet et al., 1997). These processes result in irreversibly fixed, parallel DNA fibers, aligned in a single direction across the entire surface. CartographiX CartographiX is a software package specifically developed in the authors’ laboratory for image acquisition and analysis of FISH on combed molecules. An intensifying camera, whose shutter is controlled by the software, is used to detect signals. Black and white images are acquired separately for each fluorochrome, and after contrast enhancement, are colored with artificial colors. Separate images are then merged to obtain multicolored images. Signals (see Fig. 8.10.2A) are measured on the recorded image according to a profile of the fluorescence that the software is able to define (see Fig. 8.10.2B). Data are then compiled in histograms showing the probe’s length, gap, or overlap (see Fig. 8.10.2C). CartographiX also allows statistical analysis of these data. However, any home-written program that can measure combed fluorescent signals could be as useful. To the authors’ knowledge, no commercial software is available for such a
specific application. Adobe Photoshop can be used for an estimation of signal length. CITool This home-made software allows one to view molecules as colored bars and classify them according to appropriate parameters. Every software developed to measure fluorescent signals can be used to analyze combed signals. There are several available programs that can be used for this purpose. CartographiX and CITool were developed for an easier approach to the analysis of combed molecules. These two programs are currently not commercially available. Advantages and disadvantages of the combing method With respect to other methods for stretching deproteinized DNA molecules, molecular combing has three main advantages. (1) The stretching factor is constant, which means that the system is calibrated once and forever as 1 µm = 2 kb; (2) the stretching is independent of the length of the fibers; and (3) a large number of genomes can be combed per coverslip, which allows for a reliable statistical analysis. The principal disadvantage of molecular combing is the DNA shearing due to mechanical stress during manipulation. Consequently, fibers have to be manipulated very carefully. Gene dosage principle This is an approach developed for the quantification of subtle gains and losses of genomic DNA (Herrick et al., 2000b). The approach consists of applying fluorescence in situ hybridization to the combed DNA by using probes to identify the amplified region. The principle behind probe length measurement (PLM) is to determine the effective copy number of an over-represented region and to compare it with the effective copy number of a nonamplified control region. The effective copy number of the control region is calculated as the number of screened genomes on the coverslip. This is done by measuring the total length of the hybridized control segments and dividing by the known length of the probe. This value is compared with that obtained from the target region and the ratio is represented as a function of the number of scanned genomes.
Molecular Cytogenetics
8.10.21 Current Protocols in Cytometry
Supplement 16
Typically, a good estimation of this ratio is obtained when ∼20 genomes have been analyzed.
Critical Parameters Reproducibility The number of cells per block is a major parameter to get the expected density of fibers on the surface and to guarantee reproducibility. Consequently, it is important to count cells carefully, assuring the expected number of cells per block. Combed DNA quality Depending on the analysis to be performed, DNA shearing can be a limiting factor, especially in gene mapping. Consequently, it is better to get long molecules that allow for visualizing and measuring long distances. On the contrary, broken fibers do not affect the gene dosage analysis. Hence, cells are included in agarose blocks that are melted and digested with β-agarase after DNA staining, to protect the DNA from further manipulations that could induce mechanical breakage. The efficiency of combing depends also on the quality of the surfaces. They must be as clean and dry as possible before silanization. Any dust causes the silane layer to be irregular, thus preventing correct combing, and adds nonspecific background to immunofluorescent detection. Residual water also causes a nonhomogeneous silanization. Therefore, coverslips are thoroughly cleansed and dried.
Molecular Combing
pH The specific binding of DNA to specifically treated surfaces via its unmodified extremities can be achieved on a great variety of surfaces by a judicious choice of pH. DNA was specifically bound to several different surfaces: hydrophobic surfaces, glass coated with vinyl silanes, polystyrene, polydimethylsiloxane; or hydrophilic surfaces, glass coated with amino silanes, polyhistidine; or plain cleaned glass. On all these surfaces the same behavior is observed; at low pH, DNA molecules adsorb strongly and nonspecifically, whereas at high pH they adsorb very weakly or not at all. In between there exists a narrow pH range where DNA binds to the surface strongly and specifically by its extremities (Allemand et al., 1997). On hydrophobic surfaces, the best binding efficiency is reached at pH 5.5. The pH range in which strong nonspecific adsorption of DNA exists is commonly used for
chromosome spreading on various surfaces, particularly on polylysine-coated ones. Molecular combing should not be confused with these techniques. They rely on an uncontrolled flow to stretch the molecules as they adsorb on the surface and yield nonreproducible results. The adsorbed molecules are unevenly stretched and aligned, and are often broken. Molecular combing does not require a shear flow. It relies on a receding meniscus to stretch DNA spontaneously bound to the surface by its extremities. This occurs in a very narrow range of pH, typically 0.2 units on a hydrophobic surface, with the best binding pH varying slightly with surface treatment.
Troubleshooting Bundles of DNA fibers DNA molecules have a tendency to remain bundled in solution. The DNA solution can then be heated 30 min at 70°C to improve mixing. Alternatively, it can be kept several days at 4°C before combing. DNA concentration When the DNA solution is too dense, dilute it by pouring half the solution back into a 2-ml tube, and adding 0.5 M MES, pH 5.5 (see Support Protocol 2). If this dilution results in sheared DNA, prepare a new solution from half or a third of a block, keeping the final volume to 1.5 ml. When the DNA solution is too dilute, concentrate it by leaving the open reservoir in a 60°C incubator for several hours, until the desired volume is attained. Alternatively, prepare a DNA solution from two blocks, keeping the final volume to 1.5 ml. Low binding efficiency The binding of DNA to a treated surface for combing is pH dependent. Consequently, a low binding efficiency can be due to a nonappropriate pH of the solution (for more detail see Allemand et al., 1997). However, silanization treatment can be highly variable from batch to batch, or even from surface to surface in the same batch. Be careful to use very clean and dry surfaces for silanization. A typical response to bad binding is to try at least three different surfaces from various batches. Shearing of DNA This phenomenon is derived from either mechanical stress during manipulation or
8.10.22 Supplement 16
Current Protocols in Cytometry
DNase contamination during storage of DNA solution. Try to handle the DNA solution in a more gentle way.
measurements take more time, yielding thus ∼50 to 100 measurements per day.
Literature Cited Widely dispersed measurements This can be due to DNA shearing or to the “beads on a string” aspect of the fluorescent signals. For these signals, it can be difficult to locate the ends of the measured signal. A tool displaying a profile corresponding to the fluorescence intensity is thus very useful to visualize the difference between hybridization dots and background noise. High background can be avoided by using washes with shaking during immunoamplification.
Anticipated Results According to DNA origin and to the concentration of the used DNA solution, the visual results are very different. Short DNA molecules, such as lambda DNA or the shortest yeast chromosomes (up to 200 kb) are expected to be unbroken in every field of view. Longer yeast chromosomes and human genomic DNA should yield fibers ranging typically from 200 to 500 kb. Fibers should be straight and unbundled, except for gene dosage, where a very high density is needed. Signal fluorescence should be intense enough to be visible on the microscope without using the intensification of the camera. Hybridization signals are expected to be in a “beads on a string” way, where the beads are several aligned fluorescent dots along the fiber.
Time Considerations Preparation of blocks takes 2 days but requires very little hands-on time. The combing process by itself is rather short, ∼8 to 10 min per surface. Baking takes longer, and can vary from 1.5 to 12 hr. Hybridization is currently done overnight but can be reduced to 6 to 7 hr. Detection steps take 4.5 hr for FISH double and triple detection. The most time-consuming part is the scanning of the surface, which can take 1 to 2 days per slide. However, this time should be doubled for inexperienced users to become familiar with the aspect and intensity of the signals. Simple measurements (such as gene dosage or lambda molecules) are fast enough to expect ∼500 measurements per day, provided the measurement tool is user friendly. Mapping
Allemand, J.F., Bensimon, D., Jullien, L., Bensimon, A., and Croquette, V. 1997. pH-dependent specific binding and combing of DNA. Biophys. J. 73:2064-2070. Bensimon, A., Simon, A., Chiffaudel, A., Croquette, V., Heslot, F., and Bensimon, D. 1994. Alignment and sensitive detection of DNA by a moving interface. Science 265:2096-2098. Bensimon, D., Simon, A., Croquette, V., and Bensimon, A. 1995. Stretching DNA with a receding meniscus: Experiments and models. Phys. Rev. Lett. 76:4754-4795. Florijn, R.J., Bonden, L.A., Vrolijk, H., Wiegant, J., Vaandrager, J.W., Baas, F., den Dunnen, J.T., Tanke, H.J., van Ommen, G.J., and Raap, A.K. 1995. High-resolution DNA fiber-FISH for genomic DNA mapping and colour bar-coding of large genes. Hum. Mol. Genet. 4:831-836. Herrick, J. and Bensimon, A. 1999. Single molecule analysis of DNA replication. Biochimie 81:859871. Herrick, J., Stanislawski, P., Hyrien, O., and Bensimon, A. 2000a. Replication fork density increases during DNA synthesis in X. laevis egg extracts. J. Mol. Biol. 300:1133-1142. Herrick, J., Michalet, X., Conti, C., Schurra, C., and Bensimon, A. 2000b. Quantifying single gene copy number by measuring fluorescent probe lengths on combed genomic DNA [published erratum appears in Proc. Natl. Acad. Sci. U.S.A. 97:4410]. Proc. Natl. Acad. Sci. U.S.A. 97:222227. Michalet, X., Ekong, R., Fougerousse, F., Rousseaux, S., Schurra, C., Hornigold, N., van Slegtenhorst, M., Wolfe, J., Povery, S., Beckmann, J.S., and Bensimon, A. 1997. Dynamic molecular combing: Stretching the whole human genome for high-resolution studies. Science 277:1518-1523. Strick, T.R., Allemand, J.F., Bensimon, D., Bensimon, A., and Croquette, V. 1996. The elasticity of a single supercoiled DNA molecule. Science 271:1835-1837. Weier, H.U., Wang, M., Mullikin, J.C., Zhu, Y., Cheng, J.F., Greulich, K.M., Bensimon, A., and Gray, J.W. 1995. Quantitative DNA fiber mapping. Hum. Mol. Genet. 4:1903-1910.
Contributed by Chiara Conti, Sandrine Caburet, Catherine Schurra, and Aaron Bensimon Institut Pasteur Paris, France
The authors wish to thank Sophie Gad, Emmanuel Cornillot, and Arielle Maho for providing useful remarks and tips on improvements of the protocols.
Molecular Cytogenetics
8.10.23 Current Protocols in Cytometry
Supplement 16
Principles and Applications of PRINS in Cytogenetics
UNIT 8.11
PRINS (PRimed IN Situ labeling) is a technique that traditionally has been used for the detection of tandemly repeated target sequences in chromosomes and nuclei (“in situ”) by oligonucleotide probes. With a simple short procedure rendering virtually 100% of potential targets labeled, such sequences can be detected down to a target size of a few hundred bases. More recently it has been found that single-copy sequences can also be detected at high sensitivity, if signals are enhanced with the Tyramide system (Kadandale et al., 2000a,b; also see UNIT 8.4 and UNIT 8.9), and according to recent research results, doing the PRINS reaction in a rolling-circle format enables the detection of single-copy targets at single-nucleotide resolution. This unit provides protocols for the labeling reaction in the form of basic PRINS or dideoxy-PRINS (ddPRINS) as well as the novel rolling-circle PRINS (see Basic Protocol 4). Basic PRINS and ddPRINS are quantitative reactions, and in particular, ddPRINS is useful for quantification of target DNA (Q-PRINS; see Basic Protocol 3). The labeling reactions can be combined with each other, or with FISH, in a variety of ways for differential multicolor detection of multiple targets (M-PRINS; see Basic Protocol 2) as described here. PRINS may also be combined with immunostaining, by simply adding the PRINS reaction on top of the immunostaining after the latter has been “secured” at the relevant sites by fixation in paraformaldehyde. Rolling-circle PRINS (see Basic Protocol 4) is a new tool under development to provide allele-discriminating single-molecule detection in individual cells in situ. STRATEGIC PLANNING PRINS has much in common with FISH, and often either of the two techniques can be applied for the detection of a particular DNA target. However, there are also important differences between the two techniques, which are worth noting. First of all, the two techniques have different requirements in terms of specimen quality. The PRINS technique relies heavily on the intact nature of the target sequences. Not only is nicked or broken DNA a poor template for chain elongation in situ, but nicks in the chromosomal DNA may also considerably raise the otherwise low level of background staining associated with PRINS through nick translation in situ. This feature may be different with the rolling-circle PRINS, where the reaction is inverted so that the probe provides the template for the probe-dependent site-specific DNA synthesis, with the target providing the primer (Fig. 8.11.1). As accessibility of the target is generally not a problem in PRINS on standard chromosome spreads, fairly freshly prepared slides are the preferred starting point for an optimal reaction. This is unlike FISH, which works best on “aged” slides treated with proteinases to optimize probe access to target. PRINS is highly reliant on a strict temperature control during the reaction, both to provide optimal denaturation of target sequences without destroying the polymerase and to provide optimal target selection by the primer. Again, this feature may be different with the rolling-circle PRINS. Suitable computerized incubators are available from a range of companies (the author uses incubators from Hybaid and MJ Research). Fewer probes are commercially available for PRINS than for FISH. However, since a probe in PRINS is simply an oligonucleotide, PRINS probes can be custom designed and custom made on demand from basically any oligonucleotide synthesis facility, making PRINS a flexible low-cost alternative to FISH. Molecular Cytogenetics Contributed by Jørn Koch Current Protocols in Cytometry (2004) 8.11.1-8.11.13 Copyright © 2004 by John Wiley & Sons, Inc.
8.11.1 Supplement 27
B 3′ 5′ ligate I. Form padlock probe at target sequence B
B
3′
3′ 5′
5′
II. Extend free 3′-end in target stand to obtain rollingcircle PRINS reaction tagging the target site with copies of the probe backbone B.
3′ B B 5′
3′
B
III. Detect these copies, e.g., by FISH
B B
Principles and Applications of PRINS in Cytogenetics
B
Figure 8.11.1 Reaction outline of the target-primed rolling-circle PRINS (see Basic Protocol 4). Padlock probes (UNIT 8.8) are hybridized and ligated into closed circles to define the location of the target sequence(s) after the target DNA has been made single-stranded. The target sequence is broken (e.g., with a restriction enzyme), either before or after hybridization and ligation, to produce a free 3′-end next to the padlock probe. If a break already exists in that region this step is unnecessary, and in its absence the reaction would be a functional test for whether the target DNA originally was intact or broken in that region (SPRINS, Andersen et al., 2002). Hybridized circles can now be replicated in a rolling-circle PRINS reaction from the endogenous 3′-ends, to produce tandem repeat copies of the padlock probe. This DNA synthesis is so extensive that some products subsequently may be visible as coiled structures under the fluorescence microscope, and with extended incubation they may reach a size where they extend out of the chromatin of metaphase chromosomes, hanging down the side of the chromosome, looking like a yo-yo anchored through its string to the initiating locus. Finally, the rolling-circle products are visualized, either by a FISH reaction or by a second PRINS reaction. In either case, the necessary oligonucleotide probes are best directed against the backbone part of the padlock probe, since the copy of that was neither present in the specimen prior to the reaction nor added to it during the reaction, but exists only at sites where rolling-circle PRINS of the padlock probe actually occurred.
8.11.2 Supplement 27
Current Protocols in Cytometry
BASIC PRINS ON CHROMOSOME SPREADS The basic PRINS protocol is in principle applicable to the detection of all tandem repeat sequences in situ, as well as to the detection of SINE and LINE repeats. For the detection of simple repeats void of one or more of the four bases of DNA (such as telomere repeats and trinucleotide repeats) it is preferable to use the ddPRINS protocol. For ddPRINS, replace the nucleotide(s) not needed for DNA synthesis from correctly hybridized probes with the corresponding dideoxynucleotides (e.g., replace dGTP with ddGTP when synthesizing the CCCTAA element of the human telomere).
BASIC PROTOCOL 1
Materials Slides containing chromosome spreads (UNIT 8.2) Reaction mixture (see recipe) Stop buffer (see recipe) Wash buffer (see recipe) Blocking solution (see recipe) 2 ng/ml fluorochrome-labeled anti-digoxigenin (Fab fragment; Roche) or fluorochrome-labeled streptavidin (Roche, Vector Laboratories) Antifade mounting medium (Vectashield from Vector Laboratories or p-phenylenediamine from Sigma) containing either of the following: 0.4 µM DAPI (Sigma; for blue counterstaining of DNA) 0.5 µg/ml propidium iodide (Sigma; for red counterstaining of DNA) Coverslips of appropriate size (e.g., 24 × 60 mm) 94°C and 55° to 65°C heating blocks and insulating lids to cover slide and block or dedicated PRINS/in situ PCR machine (Hybaid, MJ Research, or Techne) Fluorescence microscope with standard excitation and emission filters (e.g., 81000 filter set from Chroma Technology) Perform PRINS 1. Decide how large a region of the slide is to be used, and choose a coverslip and an amount of reaction mixture that fit the area. In general, use 1 µl reaction mixture for each mm in coverslip length. To cover a standard slide completely, prepare 60 µl reaction mixture and cover with a 24 × 60–mm coverslip. To cover a smaller area prepare less reaction mixture (and use a smaller coverslip). 2. As soon as the reaction mixture has been spread with the coverslip (by placing the coverslip over the spot of solution and allowing the spot to spread), denature the slide by placing it on a 94°C heating block covered with a lid for 4 min. A heating block with a metal surface to ensure good heat transfer must be used; temperature should be scrupulously maintained, since it must not deviate more than 1° to 2°C from the desired value. To ensure this, it is important that the slide and the hot plate be covered with an insulating lid. A special PRINS/in situ PCR machine is available from Hybaid, MJ Research, or Techne. This can be preset (simulated slide function) to incubate the slide at 91° to 94°C (optimally 92.5°C) for 2 to 4 min. This is the more convenient setup and ensures absolute reproducibility of the reaction conditions. The program operates from an estimated slide temperature, and the slide is slightly cooler than the hot plate, hence the slightly lower temperature in comparison to that used in the manual procedure above. If a specialized incubator is used, the machine “handles” this and the following step automatically, so just spread the reaction mixture with the coverslip, put the slide in the machine, press the button to start the reaction, and transfer the slide to stop buffer when the program is completed. Either type of incubator should be covered with a lid (if a lid is not built in), to optimize temperature control.
Molecular Cytogenetics
8.11.3 Current Protocols in Cytometry
Supplement 27
3. Transfer the slide quickly to a 55° to 65°C block and incubate 5 to 60 min (depending on the probe) for probe annealing and chain elongation. See annotation to step 2 regarding options for incubation.
4. Place the slide in stop buffer, preheated to the temperature used in step 3, for 1 min to terminate the PRINS reaction. 5. Transfer the slide to 50 ml wash buffer and wash at least 3 min at room temperature. The reaction can be paused at this step, and the slide stored overnight in wash buffer at 4°C.
6a. If fluorochrome-labeled nucleotides have been used: Counterstain, mount, and evaluate the slide under the microscope. Omit steps 7 and 8. 6b. If digoxigenin- or biotin-labeled nucleotides have been used: Visualize with antidigoxigenin antibody or streptavidin as in steps 7 and 8. Visualize digoxigenin- or biotin-labeled PRINS product 7. Apply 50 µl of 2 ng/ml fluorochrome-conjugated anti-digoxigenin (or fluorochromeconjugated streptavidin) in blocking solution to the slide. Incubate under a coverslip 30 to 60 min in the dark. 8. Wash twice, 5 min each time, in 50 ml wash buffer. The slide is now ready for counterstaining and mounting.
Counterstain and mount slides 9a. For blue counterstaining of DNA: Mount slide in 20 µl antifade mounting medium containing 0.4 µM DAPI. 9b. For red counterstaining of DNA: Mount slide in 20 µl antifade mounting medium containing 0.5 µg/ml propidium iodide. 10. Examine slides by fluorescence microscopy using standard excitation and emission filters (e.g., 81000 filter set from Chroma Technology). BASIC PROTOCOL 2
Principles and Applications of PRINS in Cytogenetics
MULTICOLOR PRINS (M-PRINS) AND PRINS-FISH REACTIONS There are two basically different ways of doing combinatorial labeling involving PRINS: either (1) all probes are hybridized together, or (2) the probes are hybridized sequentially. Simultaneous hybridization requires, first, that the probes have the same optimal hybridization conditions. It is, therefore, generally not possible to hybridize an oligonucleotide together with a cloned probe. Second, the different probes must be distinguishable, and since the chain elongation in situ will label all probes indiscriminately, regardless of what target they detect, single-reaction M-PRINS has long been considered impossible. However, discrimination of large numbers of probes is indeed possible if the following three modifications are employed, either separately or in combination (Serakinci et al., 2002). (1) The chain elongation copies the DNA flanking the probe, and the nucleotide composition in the PRINS product therefore reflects the base composition of that DNA. Feeding the polymerase two or more labeled nucleotides (e.g., fluorescein-dCTP and rhodaminedUTP) will, thus, result in a mixed-color ratio labeling of the site with the color blend differing from site to site, depending upon the base composition at the site. (2) A 5′ label on the probe produces enough signal for the detection of high-copy-number tandem repeat targets. Targets that label similarly in the chain elongation may thus be discriminated through different 5′ labels. (3) The 5′-labeled probe may additionally be blocked at the 3′ end (e.g., with a dideoxy residue) so that it cannot initiate primer extension. Such 3′-blocked probes are readily distinguishable from standard unblocked probes with the same 5′ label. Single reaction M-PRINS is thus done according to these protocols, simply by adding 5′ labels to the primers and increasing the number of labeled nucleotides.
8.11.4 Supplement 27
Current Protocols in Cytometry
When probes cannot be mixed, the multicolor reactions must be done by applying the probes sequentially. It is crucial that the preparation be denatured only once, as the signal from the first PRINS reaction would otherwise be lost when the preparation is redenatured for the subsequent PRINS or FISH reactions. The sequential procedure is exemplified in the following protocol. In a PRINS-FISH combination it is furthermore important that PRINS be done first, as the FISH probes would otherwise function as primers in the PRINS reaction. It is generally not necessary to do any proteinase treatments prior to PRINS-FISH. In the author’s laboratory PRINS-FISH is used in PRINS-painting (Hindkjær et al., 1995) and PRINS-CGH formats to combine painting data with locus-specific data (J. Koch, unpub. observ.). Materials Labeled nucleotides with two different labels (e.g., digoxigenin-11-dUTP and fluorescein-dUTP) to be used in preparation of reaction mixture (see recipe) Ethanol series, ice cold (see recipe) Additional labeled nucleotide with rhodamine-dUTP (Roche) or biotin-dUTP (Enzo) label (optional) Additional reagents and equipment for PRINS (see Basic Protocol 1) 1. Perform Basic Protocol 1, steps 1 to 4, using, e.g., digoxigenin-11-dUTP as the labeled nucleotide. 2. After incubation in stop buffer (see Basic Protocol 1, step 4), immediately dehydrate the slide in a −20°C ethanol series (70%, 90%, and 99%, 3 min each). 3. Remove the slide from the 99% ethanol, drain, and air dry. 4. Prepare and mix a new reaction mixture as before, but using a different type of labeled dNTP (in this case, e.g., fluorescein-dUTP). 5. Again perform Basic Protocol 1, steps 1 to 4, with the following variations: a. Do not denature the slide prior to the second PRINS as in Basic Protocol 1, step 2, but preheat to 55° to 65°C for 1 min. b. Preheat the second reaction mixture (prepared in step 4 of this protocol) to 55° to 65°C for 1 min. c. Spread the preheated reaction mixture on the slide with a coverslip. d. Incubate as in Basic Protocol 1, step 3. 6. Place the slide in stop buffer, preheated to the temperature used in step 5, for 1 min to terminate the second PRINS reaction. 7. If a third PRINS reaction is desired, repeat steps 2 to 8, using rhodamine-dUTP or biotin-dUTP as the labeled nucleotide. 8. Wash the slide 3 min at room temperature in wash buffer. 9. Perform antibody staining and counterstaining/mounting and examine the slide by fluorescence microscopy (see Basic Protocol 1, steps 6 to 10).
Molecular Cytogenetics
8.11.5 Current Protocols in Cytometry
Supplement 27
BASIC PROTOCOL 3
QUANTITATIVE PRINS (Q-PRINS)
BASIC PROTOCOL 4
ROLLING-CIRCLE PRINS
Principles and Applications of PRINS in Cytogenetics
With the PRINS design, high concentrations of small probes that penetrate the specimen well can be used, making targets effectively saturated and the reaction quantitative. This is particularly true for the ddPRINS, which is used for measuring the size of telomeric repeat domains at individual chromosome ends, as well as for sizing of other simple repeat domains that vary in size, such as trinucleotide repeats. Evaluation of Q-PRINS reactions can be done in either of two ways. One way is by counting signals, and calculating the fraction of potential targets actually stained (the author typically counts 100 potential targets). The higher the staining frequency, the larger the target (Therkelsen et al., 1995). With dideoxy-PRINS, this often becomes impractical as the staining efficiency in many cases is close to 100% (e.g., 99.8% for telomeric repeats in normal blood lymphocytes). In such cases, it is preferable to measure the light intensities of the PRINS signals instead, and average those over ten metaphases. The author uses the preexisting image analysis program QUIPS from VYSIS. This program was designed for CGH and compares the intensity of a test sample to the intensity of a reference sample. For PRINS analysis, the signals are designated as “tester” and the DAPI counterstain as “reference,” and the result is thus a signal-to-counterstain ratio that is directly proportional to the amount of target sequence (Krejci and Koch, 1998). The adaptation to the program has the derived advantage that some of the factors causing artificial variation in signal intensities (e.g., uneven illumination) have a similar influence on the counterstain, leaving the ratio less affected than the absolute values. In collaboration with Peter Lansdorp, DAKO has released an analysis program tailor-made for telomere signal quantification. This program operates with absolute intensity values.
The rolling-circle PRINS protocol given here was originally optimized, in collaboration with the inventors of padlock probes (UNIT 8.8), for the detection of single-copy target sequences in mitochondrial DNA in ethanol-fixed human cells. A corresponding procedure starting at step 9, below, has produced results on standard human metaphase chromosomes. The procedure is under continuous further development and revisions of this protocol are likely to appear in the future. The hybridization conditions here have been optimized for padlock probes where each of the two hybridizing segments is ∼20 nucleotides long, the backbone ∼50 nucleotides long, and the identifier oligonucleotides ∼23 bases long. Materials Slides containing ethanol-fixed cells (see, e.g., UNIT 7.5 for fixation technique) Low-salt buffer: e.g., PBS or SSC (see APPENDIX 2A for recipes) 0.1% (w/v) pepsin in 0.1 M HCl: freshly prepared and preheated to 37°C (it may take several minutes to dissolve pepsin even at 37°C) Ethanol series (see recipe) RNase A mixture (see recipe) Restriction enzyme mixture (see recipe) containing suitable restriction enzyme (see step 7) λ-exonuclease mixture (see recipe) Padlock probe(s) (UNIT 8.8) Hybridization mixture (see recipe) DNA ligase mixture (see recipe) Padlock removal solution: 2× SSC (APPENDIX 2A) containing 30% (v/v) deionized formamide Reaction mixture for rolling-circle PRINS (see recipe) Identifier oligonucleotides (see annotation to step 17)
8.11.6 Supplement 27
Current Protocols in Cytometry
Antifade mounting medium (Vectashield from Vector Laboratories or p-phenylenediamine from Sigma) containing either of the following: 0.4 µM DAPI (Sigma; for blue counterstaining of DNA) 0.5 µg/ml propidium iodide (Sigma; for red counterstaining of DNA) Coverslips of appropriate size (e.g., 24 × 60 mm to completely cover a standard slide) 37°C humidified incubator Fluorescence microscope with standard excitation and emission filters (e.g., 81000 filter set from Chroma Technology) Perform rolling-circle PRINS 1. Decide how large a region of the slide is to be used, and choose a coverslip that fits the area. Plan out amounts of the different hybridization and enzyme reaction mixtures that fit under a coverslip of this size. In general use, 1 ìl of mixture for each mm in coverslip length. To cover a standard slide completely, one would prepare 60 ìl of each mixture and cover with a 24 × 60–mm coverslip. To cover a smaller area, prepare less mixture (and use a smaller coverslip).
2. Rinse slide in a fairly neutral, low-salt buffer without excessive buffering capacity (e.g., PBS or SSC). 3. Put slide into preheated pepsin solution and incubate 1 min at 37°C. The duration of this step is critical. Try to avoid overdigestion or underdigestion of samples.
4. Rinse the slide in the same buffer used in step 2. Dehydrate the specimen by immersing the slide successively in an ethanol series (3 min each in 70%, 90%, and 99% ethanol). Drain off the ethanol and air dry the slide. The procedure can be paused at this step, and the slide stored dry.
5. Incubate the slide with the RNase A mixture under the coverslip for 30 min at 37°C in a humidified incubator. 6. Rinse, dehydrate, and air dry the slide as in step 4 (slide can be stored if necessary). 7. Digest the target DNA on the slide with a suitable restriction enzyme so that the hybridization target will be in the vicinity of a free 3′-end by incubating with restriction enzyme mixture under the coverslip for 30 min at 37°C (or other temperature suitable for the particular enzyme). Bloch and Grossman (1995) contains a detailed tabulation of restriction enzyme recognition sequences and reaction conditions.
8. Rinse, dehydrate, and air dry the slide as in step 4 (slide can be stored if necessary). 9. Remove the target complementary DNA strand by incubating with λ-exonuclease mixture under the coverslip 30 min at 37°C. 10. Rinse, dehydrate, and air dry the slide as in step 4 (slide can be stored if necessary). 11. Hybridize the padlock probe(s) at a concentration of 10 to 200 nm in hybridization mixture under the coverslip 30 min at 37°C. 12. Rinse away excess probe by incubating 5 min each in two changes of wash buffer at 37°C, then dehydrate and air dry the slide as described in step 4. 13. Incubate the slide with DNA ligase mixture under the coverslip 30 min at 37°C. 14. Remove noncircularized padlock probes by washing 15 min each in two changes of padlock removal solution at 42°C. Rinse, dehydrate, and air dry slide as in step 4. 15. Incubate the slide with the rolling-circle PRINS reaction mixture under the coverslip for 30 min at 37°C.
Molecular Cytogenetics
8.11.7 Current Protocols in Cytometry
Supplement 27
16. Rinse, dehydrate, and air dry the slide as in step 4 (slide can be stored if necessary). 17. Incubate with identifier oligonucleotides at a concentration of 10 to 200 nm in hybridization solution 30 min at 37°C. The identifier oligonucleotides are used for the identification of the rolling circle product from a given padlock probe (which may be applied as a padlock probe within a mixture of padlock probes that are all hybridized and rolled in parallel). The criterion for the selection of these identifiers is that they should specifically recognize a given rolling circle product, and, essentially, they need not be longer than necessary for that. The 23-mers mentioned in the introduction to this protocol therefore have this size for historic reasons—anything from 16-mers and up should be able to do the job. Since it is possible to cohybridize and co-roll multiple padlock probes on the same specimen, it is preferable to generate a class of unique identifiers that can also be cohybridized for the identification of the individual rolling circle products in situ. Each identifier is synthesized with one fluorophore attached to it, and the fluorophores are chosen among the fluorophores usually used for FISH (or standard PRINS). Instead of using labeled oligonucleotides and FISH to identify the rolling circle products, the identifier oligonucleotides may also be used as PRINS primers (see Basic Protocol 2, steps 5 to 9). However, the FISH procedure is the best-tested at this point and the most multiplexable approach. For PRINS detection it may be preferable to design the backbone from only three nucleotides so it can be detected by ddPRINS.
18. Rinse away excess probe by incubating in two changes of wash buffer for 5 min each at 37°C. Counterstain and mount slides 19a. For blue counterstaining of DNA: Mount slide in 20 µl antifade mounting medium containing 0.4 µM DAPI. 19b. For red counterstaining of DNA: Mount slide in 20 µl antifade mounting medium containing 0.5 µg/ml propidium iodide. 20. Examine slides by fluorescence microscopy using standard excitation and emission filters. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Blocking solution Dissolve 5% (w/v) nonfat dry milk in wash buffer (see recipe). Centrifuge 2 min in a microcentrifuge at maximum speed, and use supernatant. This can be stored at −20°C for years. When in use it is preferable to store the blocking solution at 4°C and not use it for more than 1 week (the milk turns sour with time).
Denhardt’s solution, 4× 400 mg Ficoll 400 400 mg polyvinylpyrrolidone 400 mg BSA (fraction V) H2O to 500 ml Filter sterilize Store at –20°C in 25-ml aliquots Principles and Applications of PRINS in Cytogenetics
8.11.8 Supplement 27
Current Protocols in Cytometry
DNA ligase mixture 39 µl H2O 5 µl 1 µg/µl BSA 5 µl 10× ligase buffer (supplied with enzyme from Roche) 1 µl (5 U) T4-DNA ligase (Roche) Mix gently by tapping a finger on the microcentrifuge tube, and use immediately after addition of the ligase. The author prepares this mixture fresh every time. If the slide tends to dry out during incubation, glycerol (5% to 10%) can be added as in the other enzyme reaction mixtures.
dNTPs for basic PRINS, 10× 1.0 mM each dATP, dCTP, and dGTP (lithium salts; Roche) 0.1 mM hapten- or fluorochrome-labeled dUTP (digoxigenin-dUTP, biotindUTP, fluorescein-dUTP, rhodamine-dUTP; Roche) Mix in 50% (v/v) glycerol Store up to 6 months at −20°C The glycerol keeps the liquid from solidifying so that it will be possible to aliquot from a stock without cycles of thawing and freezing.
dNTPs for dideoxy-PRINS (with ddGTP), 10× 1.0 mM each dATP and dCTP (lithium salts; Roche) 1.0 mM ddGTP 0.1 mM hapten- or fluorochrome-labeled dUTP (digoxigenin-dUTP, biotindUTP, fluorescein-dUTP, rhodamine-dUTP; Roche) Mix in 50% (v/v) glycerol Store up to 6 months at −20°C Ethanol series for PRINS, ice-cold Prepare 70%, 90%, and 99% (v/v) ethanol in distilled water and store at −20°C in the container that is to be used for the dehydration for at least one half hour before it is used, as it is essential that the ethanol be very cold. λ-Exonuclease mixture 34 µl H2O 5 µl 10× incubation buffer (supplied with enzyme) 5 µl 1 µg/µl BSA 5 µl concentrated glycerol 1 µl (5 U) λ-exonuclease (New England Biolabs) Mix gently by tapping a finger on the microcentrifuge tube, and use immediately after addition of the nuclease.
Hybridization mixture 12 µl H2O 25 µl 4× SSC (APPENDIX 2A) 25 µl 40% (v/v) formamide 25 µl 4× Denhardt’s solution (see recipe) 12 µl 2 µg/µl salmon sperm DNA 1 µl of each relevant DNA probe (from 1 µM stock solutions) Mix gently by tapping a finger on the microcentrifuge tube, and use soon after preparation.
Molecular Cytogenetics
8.11.9 Current Protocols in Cytometry
Supplement 27
Reaction mixture for PRINS 38 µl H2O 5 µl 10× polymerase buffer (supplied with enzyme from Roche) 5 µl 10× dNTP for basic PRINS (see recipe) or for dideoxy-PRINS (see recipe) 1 µl DNA probe (from 500 ng and up) 1 µl (1 U) Tth or Taq DNA polymerase (Roche) Mix gently by tapping a finger on the Eppendorf tube, and use immediately after addition of the polymerase. Store reaction mixture without polymerase up to several months at −20°C. Reaction mixture for rolling-circle PRINS 28 µl H2O 5 µl 10× polymerase buffer 5 µl 1 µg/µl BSA 1 µl 50 mM DTT 5 µl concentrated glycerol 5 µl 10× dNTPs for basic PRINS (see recipe) 1 µl φ29 DNA polymerase (New England Biolabs) Mix gently by tapping a finger on the microcentrifuge tube, and use immediately after addition of the polymerase. The author prepares this mixture fresh every time.
Restriction enzyme mixture 39 µl H2O 5 µl 1 µg/µl BSA 5 µl 10× restriction buffer (supplied with enzyme from Roche) 1 µl (1 U) restriction enzyme (Roche) Mix gently by tapping a finger on the microcentrifuge tube, and use immediately after addition of the enzyme. The author prepares this mixture fresh every time. If the slide tends to dry out during incubation, glycerol can in principle be added as in the other enzyme mixtures, but care should be taken not to induce star activity of the enzyme. The effect varies significantly with the enzyme, but the rule of thumb is to avoid having the enzymes make up more than 10% of the reaction mixture (corresponding to a glycerol concentration of 5%). Enzymes prone to star activity often come with a warning in the product insert.
RNase A mixture 40 µl H2O 5 µl 1 µg/µl RNase A 5 µl PBS (APPENDIX 2A) Mix gently by tapping a finger on the microcentrifuge tube, and use immediately after addition of the nuclease. The author has prepared this mixture fresh every time, but it should be possible to store it at –20°C. If the slide tends to dry out during incubation, glycerol (5% to 10%) can be added as in the other enzyme reaction mixtures.
Stop buffer 50 mM NaCl 50 mM EDTA, pH 8.0 (see APPENDIX 2A for 500 mM) Store up to 1 year at room temperature Wash buffer 4× SSC, pH 7.0 (see APPENDIX 2A for 20×) 0.05% (v/v) Tween 20 Store up to 1 year at room temperature Principles and Applications of PRINS in Cytogenetics
8.11.10 Supplement 27
Current Protocols in Cytometry
COMMENTARY Background Information The PRINS concept was first introduced by Koch et al. (1989) for the study of minute sequence variations in centromere DNA in situ. The potential for discrimination among closely related DNA sequences in situ has more recently also been illustrated by Pellestor et al. (1994, 1995), and by Koch et al. (1995), both discriminating between the alpha satellite DNA of chromosome 13 and that of chromosome 21 on the basis of a single-base difference. A further advantage of the PRINS technique is that the probes used are very small, and thus penetrate easily to the target, even if this is in a very tight formation. This feature of the technique is being exploited to enable efficient staining of target sequences in sperm cells (e.g., see Pellestor et al., 1997). More recently, a new version of the PRINS technique, the “dideoxy-PRINS” (ddPRINS), has been developed. The advantage of ddPRINS derives from the basic principle that PRINS operates from unlabeled probes, which thus do not themselves label the hybridization target. Labeling happens only when the probe is turned into a primer by a DNA polymerase and elongated with the DNA flanking the hybridization target as template for the in situ synthesis of DNA. Supplying this DNA synthesis with labeled nucleotides results in site-specific labeling where the probe was elongated by the polymerase. In ddPRINS, one or more of the nucleotides fed to the polymerase are dideoxynucleotides. Consequently, only oligonucleotides hybridizing at sites providing a template for DNA synthesis that can be copied in the presence of this (these) dideoxynucleotide(s) will be elongated and induce labeling at the binding site. Thus, for example, if ddCTP is included in the reaction mixture, only probes hybridizing next to DNA lacking Gresidues can be copied. In general, this added selectivity increases the signal-to-noise ratio of the PRINS reaction by about an order of magnitude (Koch, 1999). The ddPRINS approach allows for prolonged incubations, thus ensuring that targets are saturated with primer and that the chain elongation has proceeded to an end. Consequently, what is obtained is a uniquely sensitive detection of simple repeat DNA, including telomeric repeats and trinucleotide repeats (Serakinci and Koch, 1999), with a labeling intensity reflecting the amount of target sequence at a particular site (QPRINS; Krejci and Koch, 1998). Combining
this with the high discriminatory power of the PRINS technique, it is possible to selectively detect and measure small closely related sequence elements and to study their organization (Krejci and Koch, 1999; Serakinci et al., 1999). Staining of very small target sequences has been employed for the study of variant telomeric repeats at individual chromosome ends in a variety of species. Furthermore, since one strand of the target is flanked at its 5′-end by one sequence, and the other strand by another, it is possible to also determine the orientation of the target—for example, the relative orientation of various telomeric repeat variants with respect to each other—by this method (Krejci and Koch, 1999; Serakinci and Koch, 1999; Serakinci et al., 1999). Target-primed rolling-circle PRINS essentially is a decade-old invention by this author, which now, through a series of collaborations, has been developed to sufficient efficiency and reproducibility to be published here. Its primary advantages come from the ability of padlock probes to discriminate individual alleles in situ in a multiplexable format (Nilsson et al., 1997), where they provide unbroken templates for PRINS reactions at the sites binding the probes, and from the target-primed approach, which both eliminates structural hindrances to the PRINS reaction and provides optimal anchoring of the PRINS product, which is a direct continuation of the hybridization target. The full development of this approach will provide a major new research tool capable of genotyping multiple alleles and loci in individual cells, while at the same time reporting the amount and spatial position and relations of the gene products. PRINS has much in common with FISH and often either of the two techniques can be applied for the detection of a particular DNA target. Fewer probes are commercially available for PRINS than for FISH. However, since a probe in PRINS is simply an oligonucleotide, PRINS probes can be custom designed and custom made on demand from basically any oligonucleotide synthesis facility, making PRINS a flexible low-cost alternative to FISH.
Critical Parameters and Troubleshooting With the possible exception of the rollingcircle PRINS, PRINS reactions rely heavily on the intact nature of the target sequences. Not only is nicked or broken DNA a poor template
Molecular Cytogenetics
8.11.11 Current Protocols in Cytometry
Supplement 27
for chain elongation in situ, but nicks in the chromosomal DNA may also considerably raise the otherwise low level of background staining associated with PRINS through the occurrence of nick translation in situ. As accessibility of the target is generally not a problem in PRINS on standard chromosome spreads, fairly freshly prepared slides are the preferred starting point for an optimal reaction. PRINS is also highly reliant on a strict temperature control during the reaction, both to provide optimal denaturation of target sequences without destroying the polymerase and to provide optimal target selection by the primer. Suitable computerized incubators are available from a range of companies (e.g., Hybaid, MJ Research). The most recurrent problem in PRINS is a “self-labeling” of certain regions of satellite DNA. In human chromosomes, satellite III on chromosome 9, and more rarely, satellite II on chromosomes 1 and 16, may self-label. This self-labeling is somewhat primer dependent, but may occur with any primer, though it would seem less likely with the rolling-circle PRINS. The self-labeling is seen (only) in low-quality chromosome spreads, and the solution to the problem is simply to make better spreads.
less. A single-color PRINS reaction with directly labeled nucleotides thus takes 1 hr, whereas a reaction with hapten-labeled nucleotides takes 2 hr. A sequential dual-color reaction simply places two single-color PRINS reactions head to tail, and thus takes 2 to 4 hr, depending on the labels chosen. Because of the many steps involved in the rolling-circle PRINS, this reaction takes a full working day, though none of the incubations takes longer than 30 min.
Anticipated Results
Kadandale, J.S., Wachtel, S.S., Tunca, Y., Wilroy, R.S. Jr., Martens, P.R., and Therapel, A.T. 2000a. Localization of SRY by primed in situ labeling in XX and XY sex reversal. Am. J. Med. Genet. 95:71-74.
Targets containing highly repeated DNA, such as centromeric repeats and classical satellite DNA, are labeled with virtually 100% efficiency. With the dideoxy-PRINS, the labeling efficiency at telomeric repeat domains likewise approaches 100%. In normal blood lymphocytes, for example, the main telomeric repeat is labeled at an efficiency of 99.8% (Krejci and Koch, 1998). For less-repeated targets (e.g., variant telomeric repeats and trinucleotide repeats), the labeling efficiency is <100%. On such targets the efficiency varies in parallel with the size of the target, and the frequency of labeled sites can thus be used as a measure of the relative size of individual repeat domains (Therkelsen et al., 1995). The efficiency of the rolling-circle PRINS is currently lower than the efficiency of the other PRINS reactions, but it should be recalled that it has been designed for the detection of targets that can not be detected with these techniques.
Time Considerations Principles and Applications of PRINS in Cytogenetics
Each PRINS reaction, including denaturation, chain elongation, and washing, takes 1 hr or less. If the reaction includes an indirect label, such as biotin or digoxigenin, a complete antibody staining of the label takes another hour or
Literature Cited Andersen, C.L., Wandall, A., Kjeldsen, E., Mielke, C., and Koch, J. 2002. Active, but not inactive, human centromeres display topoisomerase II activity in vivo. Chromosome Res. 10:305-312. Bloch, K.D. and Grossman, B. 1995. Digestion of DNA with restriction endonucleases. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 3.1.13.1.21. John Wiley & Sons, New York. Hindkjær, J., Brandt, C.A., Koch, J., Lund, T.B., Kølvraa, S., and Bolund, L. 1995. Simultaneous detection of centromere specific probes and chromosome painting libraries by a combination of PRimed IN Situ labeling and chromosome painting (PRINS-painting). Chromosome Res. 3:41-44.
Kadandale, J.S., Tunca, Y., and Therapel, A.T. 2000b. Chromosomal localization of single copy genes SRY and SOX3 by primed in situ labeling (PRINS). Microb. Comp. Genomics 5:71-74. Koch, J. 1999. PRINS: PRimed IN Situ labeling and hybridization in one step. In Nonradioactive Labelling and Detection of Biomolecules, 2nd ed. (C. Kessler, ed.) pp. 407-416. Springer Verlag, Heidelberg. Koch, J., Kølvraa, S., Gregersen, N., and Bolund, L. 1989. Oligonucleotide-priming methods for the chromosome-specific labelling of alpha satellite DNA in situ. Chromosoma 98:259-265. Koch, J., Hindkjær, J., Kølvraa, S., and Bolund, L. 1995. Construction of a panel of chromosome specific oligonucleotide probes (PRINS-primers) useful for the identification of individual human chromosomes in situ. Cytogenet. Cell Genet. 71:142-147. Krejci, K. and Koch, J. 1998. Improved detection and comparative sizing of human chromosomal telomeres in situ. Chromosoma 107:198-203. Krejci, K. and Koch, J. 1999. An in situ study of variant telomeric repeats. Genomics 58:202-206.
8.11.12 Supplement 27
Current Protocols in Cytometry
Nilsson, M., Krejci, K., Koch, J., Kwiatkowski, M., Gustavson, P., and Landegren, U. 1997. Padlock probes reveal single-nucleotide differences, parent of origin and in situ distribution of centromeric sequences in human chromosomes 13 and 21. Nature Genet. 16:252-255. Pellestor, F., Girardet, A., Andréo, B., and Charlieu, J.P. 1994. A polymorphic alpha satellite sequence specific for human chromosome 13 detected by oligonucleotide primed in situ labelling (PRINS). Hum. Genet. 94:346-348. Pellestor, F., Girardet, A., Lefort, G., Andréo, B., and Charlieu, J.P. 1995. Rapid in situ detection of chromosome 21 by PRINS technique. Am. J. Med. Genet. 56:1-8. Pellestor, F., Girardet, A., Andréo, B., Lefort, G., and Charlieu, J.P. 1997. Incidence of chromosome 1 disomy in human sperm estimated by the primed in situ (PRINS) labeling technique. Cytogenet. Cell Genet. 76:192-195. Serakinci, N. and Koch, J. 1999. Detection and sizing of human telomeric repeat DNA in situ. Nature Biotechnol. 17:200-201.
Serakinci, N., Krejci, K., and Koch, J. 1999. Telomeric repeat organization: A comparative in situ study between man and rodent. Cytogenet. Cell Genet. 86:204-211. Serakinci, N., Ostergaard, M., Larsen, H., Madsen, B., Pedersen, B., and Koch, J. 2002. Multiple telomeric aberrations in a telomerase-positive leukemia patient. Cancer Genet. Cytogenet. 138:11-16. Therkelsen, A.J., Nielsen, A., Koch, J., Hindkjær, J., and Kølvraa, S. 1995. Staining of human telomeres with primed in situ labeling (PRINS). Cytogenet. Cell Genet. 68:115-118.
Contributed by Jørn Koch Laboratory of Molecular Pathology, Institute of Pathology Aarhus Kommunehospital Aarhus, Denmark
Molecular Cytogenetics
8.11.13 Current Protocols in Cytometry
Supplement 27
Comparative Genomic Hybridization (CGH)—Detection of Unbalanced Genetic Aberrations Using Conventional and Micro-Array Techniques
UNIT 8.12
Comparative genomic hybridization (CGH) represents a genome-wide screening technique to mine archived tumor material for genetic aberrations (Kallioniemi et al., 1992). Pathologists have been collecting invaluable treasures for longer than a thousand and one nights. Numerous tissue samples for many different tumor types and stages, results of histopathological investigations, and information about patient clinical courses have been stored. The application of CGH is like using Ali Baba’s magic password “open sesame” to gain access to the comprehensive pathological tumor archives and to allow for a systematic analysis of tumor-specific genetic aberrations. Figure 8.12.1 introduces the principle of CGH. Tumor and control genomes are labeled with two different haptens or fluorescent dyes (green and red). Equal amounts of both DNAs are hybridized onto normal targets, either metaphase preparations (chromosomeCGH) or gene- or region-specific DNA clones, e.g., BAC clones (array-CGH). The nonspecific hybridization of repetitive elements is suppressed by adding an excess of unlabeled highly concentrated repetitive DNA sequences termed Cot-1 DNA. Tumor DNA and control DNA sequences bind to their respective normal target sequences. Copy number differences in specific regions between the two genomes are reflected by increased or decreased green fluorescent intensities compared to the red fluorescent values. Intensity ratios are calculated and interpreted as gain or loss of these same regions in the tumor genome. Over the last 8 years, CGH has been applied to investigate virtually every different kind of tumor—leukemias, lymphomas, mesenchymal tumors, and epithelial cancers. Several hundred scientific publications have identified recurrent, tumor-specific, and tumorstage-specific chromosomal gains and losses (for reviews see Forozan et al., 1997; Ried et al., 1997, 1999; Gray and Collins, 2000). This increase in information regarding the genetics of cancer has paved the way to new diagnostic applications and investigations of mechanisms involved in tumorigenesis. CGH has recently been extended to the analysis of animal models of human cancer (Donehower et al., 1995; Weaver et al., 1999). The testing of new treatment strategies and the functional analysis of genes causing cell transformation, tumor initiation, and progression are performed on mouse and rat tumors. CGH serves here as an invaluable tool for identifying the characteristic genetic rearrangements or copy number changes in these tumor models. Using conventional CGH on chromosomes, the resolution for the detection of a loss is in a range of ∼5 to 10 Mbp. Gains and amplifications are easier to identify if the copy number is higher; the resolution limit here is in a range of ∼1 to 5 Mbp. Micro-array CGH allows for a tremendous improvement towards the detection of oncogene amplifications and deletions of tumor suppressor genes on a single-gene level (Solinas-Toldo et al., 1997; Pinkel et al., 1998; Pollack et al., 1999; Albertson et al., 2000). Target chromosomes are being replaced by locus- and gene-specific DNA clones covering the genome. Using BAC clones, the gains and losses are mapped with a resolution of ∼100 kbp and may be gene specific. Since these BAC clones also carry at least one STS marker, they provide immediate links to the physical maps. Thus, array-CGH will tremendously improve the Contributed by Evelin Schröck, Zoë Weaver, and Donna Albertson Current Protocols in Cytometry (2001) 8.12.1-8.12.30 Copyright © 2001 by John Wiley & Sons, Inc.
Molecular Cytogenetics
8.12.1 Supplement 18
A −
tumor DNA
+ 1
2
3
6
7
8
13
14
15
9
10
16 control DNA
19
20
21
4
5
11
12
17
18
22 X
Y
B tumor DNA
control DNA
Figure 8.12.1 Principle of CGH. (A) Hybridization of tumor and control DNA onto normal human chromosomes. Chromosomal regions shown in cross-hatched lines are gained in the tumor genome, e.g., chromosomes 7 and 12, chromosome arms 3q and 20q, and bands 11q14-22. In contrast, losses (portrayed in a pattern of dots) are mapped to chromosomes 5 and Y, chromosome arms 3p, 8p, and 17p, and on band 13q14. (B) Hybridization of tumor and control DNA onto arrayed BAC clones demonstrating gains (cross-hatched lines) and losses (dotted pattern) with a resolution of ∼100 kb. Chromosomal regions and DNA clones that are recurrently gained or lost are targets for molecular genetic approaches towards identification and cloning of potential oncogenes and tumor suppressor genes as well as for diagnostic and prognostic purposes.
common efforts to identify genes that are responsible for tumor initiation, progression, and metastasis.
Identification of Copy Number Changes in Tumor Genomes by CGH
This unit therefore contains the basic protocols for the conventional CGH technology on chromosomes as well as specific information about the CGH analysis of mouse tumors and the development of array-CGH. Figure 8.12.2 outlines the overall procedure.
8.12.2 Supplement 18
Current Protocols in Cytometry
conventional CGH
array CGH
tumor DNA preparation (Basic Protocol 1 and Alternate Protocol 1)
preparation of control DNA (Basic Protocol 2)
preparation of target human or mouse chromosomes and chromosome identification probes (Basic Protocol 3 and Alternate Protocol 2)
preparation of arrayed BAC clones
nick-translation of tumor and control DNA (Basic Protocol 4 and Alternate Protocol 3)
fluorescence in situ hybridization (Basic Protocol 6 and Basic Protocol 7)
image acquisition (Basic Protocol 8)
data analysis and interpretation (Basic Protocol 9)
Figure 8.12.2 Flow chart outlining the course of CGH protocols.
CGH ON HUMAN CHROMOSOMES AND MOUSE CHROMOSOMES For CGH, tumor DNA may be extracted from fresh or frozen tumor material (see Basic Protocol 1), formalin-fixed tumor tissue (see Alternate Protocol 1), or even cultured tumor cells. In principle, the protein digestion and the DNA preparation follow standard procedures. Control DNA and metaphase chromosomes may be extracted from peripheral blood of healthy donors (see Basic Protocol 2). In order to obtain accurate results for DNA copy numbers of the X- and Y-chromosomes, it is recommended that the sex of all genomes involved—tumor DNA, control DNA, and target chromosomes—be matched. The basic methodology for CGH on mouse chromosomes is the same as for human chromosomes, with the exception of the procedures for preparation of target metaphase chromosomes and control DNA. It is important to note that the control DNA should be prepared from the same mouse strain as the source of the tumor DNA (e.g., a wild-type littermate), to minimize the occurrence of bright bands resulting from repetitive sequences that differ between strains. For the preparation of the target chromosomes, any mouse strain may be used; in the authors’ experience lymphocytes from spleens of C57BL/6 mice give consistently good chromosome spreads. Spleens contain an abundant supply of T and B cells that can be expanded in a short-term culture. The spleens should be sex-matched to the tumor and control material, and should be from a mouse between 3 and 6 weeks old.
Molecular Cytogenetics
8.12.3 Current Protocols in Cytometry
Supplement 18
BASIC PROTOCOL 1
TUMOR DNA PREPARATION FROM FROZEN TISSUE Tissue sample is mechanically disaggregated, washed, and digested. DNA is extracted with phenol and isolated. This protocol may be followed for mouse frozen tissue as well. Materials Frozen tissue sample(s) of interest Culture medium (e.g., RPMI 1640) Phosphate-buffered saline (PBS; APPENDIX 2A) DNA buffer I (see recipe) 20 mg/ml proteinase K 10% (w/v) sodium dodecyl sulfate (SDS) Phenol 24:1 (v/v) chloroform/isoamyl alcohol 3 M sodium acetate, pH 5.2 (see recipe in UNIT 8.3) 100% and 70% ethanol Petri dishes 15-ml centrifuge tubes, sterile and phenol-safe 55°C water bath 1.5-ml microcentrifuge tubes (e.g., Eppendorf) Speedvac evaporator (Savant) Rotating shaker Additional reagents and equipment for assessing DNA concentration (UNIT 4.5) and running a 1% agarose gel (UNIT 8.3) Disaggregate and digest tissue 1. Put 60 to 80 mg of tissue into a petri dish with culture medium and cut the tissue into small pieces using a sterile knife or scissors. 2. Transfer the suspension into 2 sterile 15-ml centrifuge tubes and centrifuge 2 min at 300 × g, 4°C. Use only phenol-safe tubes and pipets made of polypropylene.
3. Remove supernatant, add 1 ml PBS or DNA buffer I, and centrifuge as above. Repeat. 4. Remove supernatant and resuspend the pellet in 2.06 ml DNA buffer I. 5. Add 100 µl of 20 mg/ml proteinase K and 240 µl of 10% SDS. 6. Mix gently and incubate overnight in a 55°C water bath. If there are still some tissue parts left, add more proteinase K, gently mix, and incubate an additional night at 55°C.
Extract DNA with phenol 7. Add 2.4 ml of phenol. Mix well by inverting the tube for 10 min. 8. Centrifuge 5 min at 1000 × g, room temperature. Transfer the supernatant into a new 15-ml tube and repeat step 7. 9. Centrifuge 5 min at 1000 × g, room temperature. Transfer the supernatant into a new 15-ml tube. Identification of Copy Number Changes in Tumor Genomes by CGH
10. Add 2.4 ml of 24:1 chloroform/isoamyl alcohol. Mix well by inverting the tube for 10 min.
8.12.4 Supplement 18
Current Protocols in Cytometry
11. Centrifuge 5 min at 1000 × g, room temperature. Transfer the supernatant into a new 15-ml tube. Isolate DNA 12. Add 25 µl of 3 M sodium acetate, pH 5.2, and 5 ml of 100% ethanol. Gently invert the tube until the DNA precipitates. 13. Use a standard automatic pipettor with a sterile pipet tip to transfer the DNA into 1.5-ml microcentrifuge tubes. 14. Wash the DNA in 70% ethanol and microcentrifuge 5 min at maximum speed, room temperature. 15. Remove supernatant and dry pellet in Speedvac evaporator. 16. Dissolve DNA in 0.5 to 1 ml sterile water overnight at 4°C on a rotating shaker. 17. Measure DNA concentration using a photometer and sterile water as a blank (UNIT 4.5) and run ∼400 ng on a 1% agarose gel to double-check concentration and molecular weight (UNIT 8.3). TUMOR DNA PREPARATION FROM PARAFFIN-EMBEDDED TISSUE Tumor DNA can also be prepared from archival tissue. A tissue sample is first treated to remove paraffin, washed, and incubated with sodium thiocyanate. Following tissue digestion with proteinase K, the DNA is extracted with phenol and isolated. DNA from formalin-fixed tumors will not give results as good as those from DNA in fresh or frozen tissue.
ALTERNATE PROTOCOL 1
Additional Materials (also see Basic Protocol 1) Formalin-fixed, paraffin-embedded tissue sample(s) of interest Xylene 100% methanol 1 M sodium thiocyanate DNA buffer II (see recipe) 100 µg/ml RNase A in 2× SSC (optional; see recipe for 20 mg/ml RNase A stock in Reagents and Solutions; see APPENDIX 2A for 20× SSC) Isopropanol Remove paraffin from tissue 1. Prepare 50-µm slices from formalin-fixed, paraffin-embedded tumor samples. If possible, perform microdissection (UNIT 8.6) to increase the amount of tumor or aberrant cells and decrease the number of normal cells in the sample.
2. Place the 50-µm slices in 1.5-ml microcentrifuge tubes and add 1 ml xylene. Incubate 15 min in a 55°C water bath to remove the paraffin. 3. Microcentrifuge ∼10 min at maximum speed, room temperature. Remove supernatant. Repeat xylene step 2 if any paraffin is still left.
4. Wash in 100% methanol, vortex, and microcentrifuge ∼10 min at maximum speed, room temperature. 5. Remove supernatant, wash in 100% ethanol, vortex, and microcentrifuge ∼10 min at maximum speed, room temperature. Dry pellet in a Speedvac evaporator.
Molecular Cytogenetics
8.12.5 Current Protocols in Cytometry
Supplement 18
Digest tissue 6. Resuspend pellet in 1 ml of 1 M sodium thiocyanate and incubate overnight in a 37°C water bath. 7. Microcentrifuge 10 min at maximum speed, room temperature. Remove supernatant and dry pellet in a Speedvac evaporator. 8. Resuspend in 400 µl DNA buffer II. 9. Optional: Perform standard RNase treatment. Incubate 1 hr with 20 µl of 20 mg/ml RNase A at 37°C. 10. Add 50 µl of 20 mg/ml proteinase K and 50 µl of 20% SDS, vortex briefly, and incubate overnight in a 55°C water bath. Tissue should be completely dissolved. Additional proteinase K can be added during incubation as needed.
Extract DNA with phenol 11. Add 500 µl (or equal volume) phenol and mix by inverting the tube for 10 min. 12. Centrifuge 5 min at maximum speed, room temperature. Transfer the supernatant into a new tube. 13. Repeat steps 11 and 12. 14. Add 500 µl of 24:1 chloroform/isoamyl alcohol. Mix by inverting the tube for 10 min. 15. Microcentrifuge 5 min at maximum speed, room temperature. Transfer the supernatant into a new tube. Isolate DNA 16. Add 10 µl of 3 M sodium acetate, pH 5.2. 17. Estimate the total volume and add an equal volume of isopropanol. 18. Gently shake until the DNA precipitates or store 30 min at −80°C. 19. Use a standard automatic pipettor with a sterile pipet tip to transfer the precipitated DNA into a new 1.5-ml microcentrifuge tube. Alternatively, microcentrifuge the tube that was stored for 30 min at −80°C at maximum speed, 10°C. Remove supernatant.
20. Wash DNA in 70% ethanol and centrifuge 5 min at maximum speed, room temperature. 21. Remove supernatant. Dry the pellet in a Speedvac evaporator. 22. Dissolve the DNA in ∼50 to 200 µl sterile water overnight at 4°C on a rotating shaker. 23. Measure DNA concentration using a photometer and sterile water as a blank and run ∼400 ng on a 1% agarose gel (UNIT 8.3). BASIC PROTOCOL 2 Identification of Copy Number Changes in Tumor Genomes by CGH
PREPARATION OF CONTROL DNA FROM PERIPHERAL BLOOD Normal whole peripheral blood is lysed to remove erythrocytes. The remaining white cells are digested with proteinase K. The DNA is extracted with phenol and isolated.
8.12.6 Supplement 18
Current Protocols in Cytometry
Control DNA of mice may be prepared by applying the tumor DNA preparation protocol (see Basic Protocol 1 or Alternate Protocol 1) to the spleen or liver tissue of a wild-type mouse. Some investigators use DNA from tail preparations, but these may not be pure enough in all cases and can be resistant to digestion by DNase I during nick translation. CAUTION: When working with human blood, cells, or infectious agents, biosafety practices should be followed; see UNIT 5.1. Materials Normal whole blood Lysis buffer (see recipe) SE buffer (see recipe) 20 mg/ml proteinase K 20% (w/v) SDS Phenol 24:1 (v/v) chloroform/isoamyl alcohol 3 M sodium acetate, pH 5.2 (see recipe in UNIT 8.3) Isopropanol 70% ethanol 50-ml blood collection tube containing EDTA, heparin, or sodium citrate anticoagulant 37°C water bath 1.5-ml microcentrifuge tubes (e.g., Eppendorf) Speedvac evaporator (Savant) Additional materials and equipment for assessment of DNA concentration (UNIT 4.5) and 1% agarose gel electrophoresis (UNIT 8.3) Draw, lyse, and process whole blood 1. Draw 10 ml whole blood into collection tubes containing EDTA, heparin, or sodium citrate as anticoagulant, transfer into 50-ml tubes, and add 30 ml lysis buffer. Shake gently. 2. Incubate 30 min on ice. 3. Centrifuge 10 min at 300 × g, 4°C and remove supernatant (blood waste). 4. Add 10 ml lysis buffer and resuspend the pellet. 5. Centrifuge 10 min at 300 × g, 4°C and remove supernatant (blood waste). 6. Add 5 ml SE buffer and resuspend the pellet. 7. Centrifuge 10 min at 300 × g, 4°C and remove supernatant (blood waste). 8. Add 5 ml SE buffer and resuspend the pellet. Digest cell pellet 9. Add 250 µl of 20 mg/ml proteinase K and 250 µl of 20% SDS. Shake gently. 10. Incubate overnight in a 37°C water bath. Extract DNA with phenol 11. Add 4.5 ml SE buffer and 10 ml phenol. Mix by inverting the tube for 10 min. 12. Centrifuge 5 min at 1500 × g, 10°C. Transfer the supernatant into a new 50-ml centrifuge tube.
Molecular Cytogenetics
8.12.7 Current Protocols in Cytometry
Supplement 18
13. Add 10 ml phenol. Mix by inverting the tube for 10 min. 14. Centrifuge 5 min at 1500 × g, 10°C. Transfer the supernatant into a new 50-ml centrifuge tube. 15. Add 10 ml of 24:1 chloroform/isoamyl alcohol. Mix 10 min by hand. 16. Centrifuge 5 min at 1500 × g, 10°C. Transfer the supernatant into a new 50-ml centrifuge tube. Isolate DNA 17. Add 300 µl of 3 M sodium acetate, pH 5.2, and 10 ml isopropanol. Gently shake until the DNA precipitates. 18. Use a standard automatic pipettor with a sterile pipet tip to transfer the precipitated DNA into a 1.5-ml microcentrifuge tube. 19. Wash DNA in 70% ethanol and microcentrifuge 5 min at maximum speed, 10°C. 20. Remove the supernatant and dry pellet in a Speedvac evaporator. 21. Dissolve DNA in 0.5 to 1 ml sterile water overnight at 4°C on a rotating shaker. 22. Measure the DNA concentration and run ∼400 ng on a 1% agarose gel (UNIT 8.3). BASIC PROTOCOL 3
PREPARATION OF NORMAL TARGET METAPHASE CHROMOSOMES Peripheral blood lymphocytes of normal female or male donors are cultured to prepare metaphase chromosomes. Cell division is induced using phytohemagglutinin. Colcemid treatment, which inhibits microtubule formation, causes mitotic arrest. The cellular membrane becomes fragile during hypotonic swelling of the cell. The metaphase chromosomes are fixed by dehydration and proteins are removed using a mixture of methanol and glacial acetic acid. Finally, the suspension is dropped onto clean glass slides and the spreading takes place when the methanol/acetic acid evaporates. CAUTION: When working with human blood, cells, or infectious agents, biosafety practices should be followed; see UNIT 5.1. Materials RPMI 1640 medium (Life Technologies) 100× antibiotic-antimycotic: 10,000 U/ml penicillin G sodium, 10,000 µg/ml streptomycin sulfate, 25 µg/ml amphotericin B (Life Technologies) Fetal bovine serum (FBS): qualified, heat-inactivated, sterile-filtered (Life Technologies) Phytohemagglutinin (PHA; Murex Diagnostics Ltd.) Normal whole blood (heparin anticoagulated) 10 µg/ml KaryoMAX colcemid solution (Life Technologies) 0.4% (w/v) KCl, 37°C 3:1 (v/v) methanol/glacial acetic acid fixative, freshly prepared 1:1 (v/v) ethanol/ether 70%, 90%, and 100% ethanol
Identification of Copy Number Changes in Tumor Genomes by CGH
75-cm2 tissue culture flasks 50-ml and 15-ml centrifuge tubes 37°C and 60°C water bath Microscope slides
8.12.8 Supplement 18
Current Protocols in Cytometry
NOTE: All incubations are performed in a humidified 37°C, 5% CO2 incubator unless otherwise specified. Process blood and coat culture flask 1. Prepare a 75-cm2 tissue culture flask containing 40 ml RPMI 1640 medium supplemented with 0.4 ml antibiotics and 8 ml FBS. Add 400 µl PHA. 2. Collect 10 ml whole blood in a tube with heparin as anticoagulant. Centrifuge 10 min at 300 × g, room temperature, or allow to settle 2 to 3 hr at room temperature. Discard serum and save lymphocyte layer (buffy coat) just above the erythrocytes. 3. Add 2 ml lymphocyte layer to the 75-cm2 flask from step 1. Final volume in the flask is 50 ml.
Culture and process lymphocytes 4. Culture 72 hr in a 37°C, 5% CO2 incubator. Shake flasks once a day. 5. Add 500 µl of 10 µg/ml KaryoMAX colcemid solution and mix well but gently by shaking or pipetting. 6. Immediately transfer into two 50-ml centrifuge tubes and incubate 20 min at 37°C. 7. Centrifuge 12 min at 300 × g, room temperature. Remove supernatant, leaving at least 5 ml in tube, and resuspend pellet by vortexing or pipetting until it is completely dissolved. 8. Add 5 ml prewarmed 0.4% KCl to each tube and mix by shaking tube, vortexing, or pipetting up and down. 9. Add more 0.4% KCl to a total of 40 ml. Mix well. Incubate 10 min in a 37°C water bath. 10. Repeat steps 7 and 8. Fix cells 11. Add 5 ml freshly prepared 3:1 methanol/glacial acetic acid fixative per tube and mix by shaking tube, vortexing, or pipetting up and down. 12. Add more 3:1 methanol/glacial acetic acid fixative to a total of 25 ml per tube. 13. Repeat steps 7, 11, and 12 two more times. 14. Transfer cell suspension to 15-ml centrifuge tubes. 15. Repeat steps 7, 11, and 12 using 10 ml of 3:1 methanol/glacial acetic acid fixative four more times, always leaving at least 2 ml in tube. 16. Resuspend pellet in 4 to 6 ml of 3:1 methanol/glacial acetic acid fixative. Prepare slides and observe chromosomes 17. To prepare slides, dip in 1:1 ethanol/ether or 3:1 methanol/glacial acetic acid fixative and wipe with paper towels. Drop ∼100 µl cell suspension onto a clean slide. 18. Check the number of metaphases, the spreading of the chromosomes, and the amount of cytoplasm left around the metaphases. Adjust dropping conditions; change temperature, humidity, and dilution. Frequently, good results are obtained by placing the slides on a rack in a 60°C water bath, dropping ∼100 ìl per slide. Alternatively, drop suspension onto slides and cover with foil
Molecular Cytogenetics
8.12.9 Current Protocols in Cytometry
Supplement 18
to prolong drying time. The slides should be dry after 1 min and the chromosomes should appear gray using a phase-contrast microscope.
19. Wash slides in 70%, 90%, and 100% ethanol, 3 min each. 20. Store slides 3 to 4 days in the 37°C incubator. 21. Store slides for immediate hybridization in 100% ethanol at 4°C for ∼2 weeks. For longer storage, freeze slides with Drierite in sealed plastic bags at −80°C. ALTERNATE PROTOCOL 2
PREPARATION OF NORMAL TARGET CHROMOSOMES FROM MOUSE SPLEEN Mouse chromosomes are prepared in a similar way as human chromosomes. However, phytohemagglutinin is replaced by concanavalin A and lipopolysaccharide for induction of mitosis. Materials RPMI 1640 medium 100× antibiotic-antimycotic: 10,000 U/ml penicillin G sodium, 10,000 µg/ml streptomycin sulfate, 25 µg/ml amphotericin B (Life Technologies) Fetal bovine serum qualified, heat-inactivated, sterile-filtered (FBS; Life Technologies) 0.4% (v/v) KCl, 37°C Mouse spleen 1 mg/ml concanavalin A (Sigma; see recipe) 25 mg/ml lipopolysaccharide (LPS; see recipe) 0.5% 2-mercaptoethanol 10 mg/ml 5-bromo-2′-deoxyuridine (BrdU; see recipe) 0.1 mg/ml 5-fluoro-2′-deoxyuridine (FUdR; see recipe) 10 µg/ml KaryoMAX colcemid solution (Life Technologies) Fixative solution: 3:1 (v/v) methanol/glacial acetic acid, freshly prepared Mouse spleen homogenizer 15-ml conical centrifuge tube 125-cm2 tissue culture flasks 37°C, 5% CO2 incubator 37°C water bath 3-ml plastic transfer pipet Prepare spleen cells 1. Prepare medium by adding 5 ml of 100× antibiotic-antimycotic and 100 ml FBS to 500 ml RPMI 1640. 2. Crush each mouse spleen separately in a mouse spleen homogenizer in ∼5 ml medium (step 1). Let any insoluble material settle to the bottom of the glass tube and then transfer the solution to a 15-ml conical centrifuge tube. 3. Centrifuge samples 10 min at 300 × g, room temperature. 4. Remove supernatant and resuspend each pellet in 10 ml RPMI 1640 medium as prepared in step 1.
Identification of Copy Number Changes in Tumor Genomes by CGH
5. Transfer each resuspended spleen pellet to a separate 125-cm2 tissue culture flask and add medium up to a total of 100 ml. Add 0.6 ml of 1 mg/ml concanavalin A, 0.1 ml of 25 mg/ml LPS, and 120 µl of 0.5% 2-mercaptoethanol.
8.12.10 Supplement 18
Current Protocols in Cytometry
Culture cells 6. Culture cells 48 hr in a 37°C, 5% CO2 incubator. After 48 hr, the medium should contain a concentrated population of proliferating cells, which may appear to float in clumps.
7. Add 0.3 ml of 10 mg/ml BrdU and 150 µl of 0.1 mg/ml FUdR to each flask, and culture cells an additional 5 hr in a 37°C, 5% CO2 incubator. 8. Add 1 ml of 10 µg/ml KaryoMAX colcemid solution to each flask and incubate 15 min in a 37°C, 5% CO2 incubator. 9. Transfer the cells from each flask to two 50-ml centrifuge tubes and centrifuge 10 min at 300 × g, room temperature. 10. Remove all but ∼1 ml of the supernatant in each tube and gently resuspend the cell pellet. 11. Add 25 ml of 0.4% KCl to each tube. Incubate 5 min in a 37°C water bath. 12. Add 1 ml fixative solution (always prepare fresh) to each tube and centrifuge 10 min at 300 × g, room temperature. Fix cells 13. Remove supernatant leaving 5 ml in tube and resuspend the pellet by vortexing or pipetting using a 3-ml plastic transfer pipet. Transfer to 15-ml centrifuge tubes. 14. Fill tubes with fixative and centrifuge 10 min at 300 × g, room temperature. 15. Repeat steps 13 and 14 four more times to thoroughly fix the chromosome preparations. Leave 2 ml of fixative solution in tube when removing supernatant. 16. Drop slides as described in Basic Protocol 3. 17. Store slides 1 week at room temperature or in a dry 37°C incubator prior to hybridization. Continue to store slides in 100% ethanol for ∼2 weeks at 4°C or freeze slides in a sealed plastic bag containing Drierite at −80°C.
NICK TRANSLATION OF TUMOR AND CONTROL DNA FOR BOTH HUMAN AND MOUSE CHROMOSOMES
BASIC PROTOCOL 4
Tumor and control genomic DNA are labeled by nick translation with different fluorochromes or haptens. Simultaneously, single-strand nicks are introduced by DNase I, and haptens or directly labeled nucleotides are incorporated into the DNA as a function of the DNA polymerase. In addition, the high molecular-weight DNA is cut into shorter fragments. The length of these fragments is crucial for the quality of the hybridization and should always be monitored using gel electrophoresis. For optimal results, a similar length of the tumor and control DNA is desired. When using fresh or frozen material, DNA fragments should be in a range of 500 to 1000 bp after nick translation. However, when using formalin-fixed tissue, DNA will already be degraded to a certain extent before the labeling reaction. In order to allow for incorporation of a high number of modified nucleotides, shorter DNA fragments in the range of 300 to 600 bp should be generated.
Molecular Cytogenetics
8.12.11 Current Protocols in Cytometry
Supplement 18
Materials 1 mg/ml DNase I from bovine pancreas (see recipe) Genomic DNA 10× NT buffer (see recipe) 10× dNTP mix (see recipe) 0.1 M 2-mercaptoethanol (see recipe) Biotin-16-dUTP (Roche Diagnostics) Digoxigenin-11-dUTP (Roche Diagnostics) Kornberg polymerase (Roche Diagnostics) Lambda HindIII DNA marker 0.5 M EDTA 1.5-ml microcentrifuge tube (e.g., Eppendorf) Additional reagents and equipment for DNA concentration and 1% agarose gel electrophoresis (UNIT 8.3) 1. Prepare working solution of DNase I by diluting the 1 mg/ml stock solution 1:1000 in cold sterile water. Keep on ice. 2. Add to a 1.5-ml microcentrifuge tube the following: 2 µg genomic DNA 10 µl 10× NT buffer 10 µl 10× dNTP mix 10 µl 0.1 M 2-mercaptoethanol 4 µl biotin-16-dUTP for the tumor DNA or 4 µl digoxigenin-11-dUTP for the control DNA Sterile water to bring the total volume including DNase I and DNA polymerase (added in next step) to 100 µl 3. Vortex, centrifuge briefly at room temperature, and keep tubes on ice. Add 0.5 to 20 µl DNase I working solution (from step 1) and 2 µl Kornberg polymerase. 4. Vortex, centrifuge briefly at room temperature, and incubate 1.5 to 2 hr at 15°C. 5. Run ∼5 µl DNA sample and 2 µl lambda HindIII DNA marker on a 1% agarose gel electrophoresis. Ideally, the length of the DNA should be between 300 and 1000 bp after nick translation. If the DNA is too large, add more DNase I working solution and incubate 10 to 20 min at 15°C. Run gel electrophoresis again.
6. Stop the nick translation with 1 µl of 0.5 M EDTA or incubate 10 min at 65°C. 7. Store labeled DNA at −20°C or continue with the CGH hybridization procedure. ALTERNATE PROTOCOL 3
Identification of Copy Number Changes in Tumor Genomes by CGH
MOUSE CHROMOSOME IDENTIFICATION PROBES Chromosome-specific P1 and BAC clones have been identified and are publicly available as reagents for mouse chromosome identification (Jackson Laboratory, http://www.informatics.jax.org/, and Research Genetics, http://www.resgen.com). It is highly recommended that investigators cohybridize DNA probes as part of the mouse CGH procedure, as mouse chromosomes are all acrocentric and of similar size, making them especially difficult to karyotype without some experience. For probe preparation, follow the nick translation procedure (see Basic Protocol 4) but incorporate Cy5-dUTP as a direct label in place of Bio-dUTP or Dig-dUTP. As long as the probes are not made from repeat sequences, they may be precipitated (use 10 µl of the 100 µl nick translation reaction volume) together with the control and tumor DNA.
8.12.12 Supplement 18
Current Protocols in Cytometry
A
B
Figure 8.12.3 Typical metaphase spread suited for CGH experiments. (A) Chromosomes are surrounded by cytoplasm (gray layer around metaphase chromosomes) before pretreatment. (B) After pepsin treatment note the lower amount of cytoplasm.
PRETREATMENT OF TARGET CHROMOSOME SLIDES FOR BOTH HUMAN AND MOUSE CHROMOSOMES
BASIC PROTOCOL 5
The quality of metaphase chromosome preparations is crucial for the success of CGH experiments (Karhu et al., 1997). A high number of well-spread metaphases and a low amount of cytoplasm are required (Fig. 8.12.3). In order to improve access by the labeled DNA fragments, a pepsin or trypsin treatment should be performed prior to hybridization. The amount of pepsin and the incubation time need to be determined for each new batch of slides and tested for different denaturation times and temperatures. Materials Metaphase chromosome preparations (see Basic Protocol 3) 2× SSC (see APPENDIX 2A for 20× recipe) 20 mg/ml RNase A stock (see recipe) 10% pepsin (see recipe) 0.01 M HCl, pH 2.0, 37°C PBS (APPENDIX 2A) 1× PBS/MgCl2 (see recipe) Formalin (37% formaldehyde) 24 × 60–mm coverslip 37°C incubator Coplin jars 37°C water bath 1. Equilibrate metaphase chromosome slides 5 min (see Basic Protocol 3) in 2× SSC at room temperature. Dilute the 20 mg/ml RNase A stock 1:200 in 2× SSC. 2. Apply 120 µl diluted RNase A to a 24 × 60–mm coverslip. Touch slide to coverslip. 3. Incubate slide 45 min in a 37°C incubator. 4. Lift coverslip and wash slides three times, 5 min each, in 2× SSC in a Coplin jar at room temperature with shaking.
Molecular Cytogenetics
8.12.13 Current Protocols in Cytometry
Supplement 18
5. Put 5 to 50 µl of 10% pepsin into an empty beaker, add 100 ml prewarmed 0.01 M HCl, pH 2.0, mix well, transfer into a Coplin jar, and place in a 37°C water bath. For mouse, do not exceed 10 ìl/100 ml of 0.01 M HCl as the mouse chromosomes can be quite sensitive to pepsin digestion. If slides <1 week old must be used, leave out the pepsin entirely.
6. Incubate slides 4 to 10 min in the Coplin jar in a 37°C water bath. 7. Wash slides two times, 5 min each, in 1× PBS at room temperature with shaking. 8. Wash slides 5 min in 1× PBS/MgCl2 at room temperature with shaking. 9. Prepare solution of 1% formalin in 1× PBS/MgCl2 by adding 2.7 ml formalin to 97.3 ml 1× PBS/MgCl2. 10. Incubate slides 10 min in Coplin jar at room temperature. 11. Wash slides 5 min in 1× PBS at room temperature with shaking. 12. Dehydrate slides in 70%, 90%, and 100% ethanol, 3 min each. Air dry slides. BASIC PROTOCOL 6
FLUORESCENCE IN SITU HYBRIDIZATION FOR BOTH HUMAN AND MOUSE CHROMOSOMES Using nick translation, the tumor DNA was labeled with biotin-dUTP and the control DNA with digoxigenin-dUTP (see Basic Protocol 4). In this protocol, the two genomes are combined and the single-copy sequences will bind to their matching DNA sequences located on the normal metaphase chromosomes. Thus, additional or missing DNA sequences in the tumor compared to the control DNA will result in increased or decreased fluorescence intensities, respectively, at very specific chromosomal sites. Repetitive DNA sequences, however, will bind in a nonspecific manner across the genome and are, therefore, removed beforehand by adding an excess of Cot-1 DNA to the hybridization mixture. The highly repetitive sequences anneal and become double-stranded in a timedependent manner before the hybridization mixture is applied to the slides. Materials 500 ng tumor DNA labeled with biotin via nick translation (see Basic Protocol 4) 500 ng control DNA labeled with digoxigenin via nick translation (see Basic Protocol 4) 1 mg/ml human Cot-1 DNA (Life Technologies) 10 mg/ml salmon testes DNA (Sigma) 3 M sodium acetate, pH 5.2 (UNIT 8.3) 100%, 90%, and 70% ethanol, ice cold and room temperature Hybridization mixture (see recipe) Probe DNA 70% formamide/2× SSC (see recipe) Centromere enumeration probe (see Background Information) Rubber cement 1.5-ml microcentrifuge tubes (Eppendorf) Speedvac evaporator (Savant) 24 × 60–mm coverslips 18 × 18–mm coverslips
Identification of Copy Number Changes in Tumor Genomes by CGH
8.12.14 Supplement 18
Current Protocols in Cytometry
1. Put in a 1.5-ml microcentrifuge tube the following: 500 ng tumor DNA labeled with biotin via nick translation 500 ng control DNA labeled with digoxigenin via nick translation 30 µl 1 mg/ml human Cot-1 DNA 1 µl 10 mg/ml salmon testes DNA 2. Add 1/10 volume 3 M sodium acetate, pH 5.2, and 2.5 to 3.0 total volumes of 100% ethanol. 3. Vortex and store overnight at −20°C or ≥15 to 30 min at −80°C. 4. Centrifuge 30 min at 14,000 × g, 4°C. 5. Pour off supernatant, remove leftover ethanol by pipetting, and dry pellet 5 min in a Speedvac evaporator. 6. Add 10 µl hybridization mixture and incubate 30 min at 37°C, vortexing occasionally. Pellet needs to be completely dissolved.
7. Denature probe DNA 5 min at 80°C, centrifuge briefly, and preanneal 2 hr at 37°C. 8. For slide denaturation, apply 120 µl of 70% formamide/2× SSC to a 24 × 60–mm coverslip, touch slide to coverslip, and incubate slides 1.5 to 2.0 min on a hot plate at 80°C. 9. Place slides immediately in ice-cold 70% ethanol, followed by 90% ethanol and 100% ethanol, 3 min each, and air dry slides. 10. Prepare centromere enumeration probe, denature 5 min at 80°C, place on ice immediately, and add to preannealed probe DNA. 11. Apply preannealed probe DNA to denaturated slides and cover with 18 × 18–mm coverslip. Seal coverslip with rubber cement. 12. Hybridize 2 to 3 nights at 37°C. DETECTION OF HYBRIDIZED DNA SEQUENCES FOR BOTH HUMAN AND MOUSE CHROMOSOMES
BASIC PROTOCOL 7
In order to visualize the hybridization results and to allow for the measurement of fluorescence intensities along metaphase chromosomes, the haptens are coupled to fluorescent dyes. Precisely, the highly specific biotin-avidin complex formation is utilized to connect avidin-FITC (green fluorescence) to the biotin-16-dUTP that was incorporated into the tumor DNA. The application of additional layers of biotin and avidin-FITC results in an increase of fluorescence intensities by combining the sandwich effect with the capacity of avidin to bind four biotin molecules. The control DNA, labeled with digoxigenin-11-dUTP, is detected by adding two to three layers of monoclonal or polyclonal antibodies linked to red fluorescent dyes such as Texas Red. Materials Hybridized slides (see Basic Protocol 6) 50% formamide/2× SSC solution (see recipe), prewarmed 0.1× SSC (see APPENDIX 2A for 20× recipe) 4× SSC/Tween 20 (see recipe) Blocking solution (see recipe) Antibodies/fluorescent dye solution 1 (see recipe) Antibodies/fluorescent dye solution 2 (see recipe)
Molecular Cytogenetics
8.12.15 Current Protocols in Cytometry
Supplement 18
Antibodies/fluorescent dye solution 3 (see recipe) DAPI working solution (see recipe) 70%, 90%, and 100% ethanol Antifade: 1,4-phenylenediamine (see recipe) Coplin jars Hybridization chamber 37°C water bath 24 × 60–mm coverslips Prepare slides 1. Remove rubber cement and lift coverslips from hybridized slides (see Basic Protocol 6). Scratches can be prevented by dipping slides in 50% formamide/2× SSC solution.
2. Wash slides three times, 5 min each, in prewarmed 50% formamide/2× SSC in Coplin jars with shaking. 3. Wash slides three times, 5 min each, in 0.1× SSC with shaking. 4. Dip slides in 4× SSC/Tween 20. Do not let them dry. 5. Add 120 µl blocking solution to 24 × 60–mm coverslips, touch slides to coverslips, and incubate in a hybridization chamber ∼30 min in a 37°C water bath. 6. Dip slides in 4× SSC/Tween 20. Do not let them dry. Apply fluorescent dyes 7. Add 120 µl prepared antibody/fluorescent dye solution 1 per 24 × 60–mm coverslip, touch slide to coverslip, and incubate in hybridization chamber 45 to 60 min in a 37°C water bath. 8. Wash slides three times, 5 min each, in 4× SSC/Tween 20 with shaking. 9. Repeat steps 7 and 8 for each antibody/fluorescent dye solution 2 and antibody/fluorescent dye solution 3. 10. Stain with DAPI by incubating slides 10 min in DAPI working solution at room temperature in the dark. 11. Wash slides 5 min in sterile water with shaking. 12. Dehydrate slides in 70%, 90%, and 100% ethanol, 3 min each. 13. Air dry and apply 30 to 35 µl antifade (1,4-phenylenediamine), cover with 24 × 60–mm coverslips, and store in the dark up to 2 to 4 weeks at 4°C. Best results are obtained when image acquisition is performed within 1 to 2 weeks. BASIC PROTOCOL 8
Identification of Copy Number Changes in Tumor Genomes by CGH
CGH IMAGE ACQUISITION In order to generate the average ratio profile for a single tumor, 8 to 10 metaphases are acquired and analyzed. DAPI, FITC, Texas Red, and Cy5 images (optional) are captured by a black-and-white CCD camera for each metaphase. The DAPI image provides a G-like banding pattern that is intensified by a band enhancement function implemented in CGH software packages to facilitate chromosome classification. Utilization of signals from centromere enumeration probes captured in the Cy5 image improve the speed and the accuracy of chromosome karyotyping (see Fig. 8.12.4). The FITC image visualizes the DNA copy number information of the tumor, indicating gains and losses of specific
8.12.16 Supplement 18
Current Protocols in Cytometry
1
2
3
6
7
8
13
14
15
19
20
4
9
21
5
10
11
12
16
17
18
22 X
Figure 8.12.4 Karyogram of a normal metaphase used as CGH target. Note that centromere enumeration probes (labeled with Cy5-near infrared, indicated by arrows) were added for several chromosomes that are difficult to differentiate from their neighbors (e.g., nos. 4 and 5; nos. 7, 8, 9, 10, and X; nos. 13, 14, and 15; nos. 19 and 20; nos. 21 and 22). One chromosome 4 was missing in this metaphase.
chromosomal regions. In contrast, the Texas Red image shows similar intensity values for all image points within chromosome masks, because normal DNA contains two copies of each chromosome. Before and during image acquisition, the following points are of particular importance. (1) The lamp should be carefully aligned. Intensity deviations within the image frame of the CCD camera should be <5% when measuring an image area without nuclei and chromosomes using a long exposure time and at least half of the dynamic range of the camera. (2) Heat-protection filters (KG1, BG38) and the field diaphragm should be used to increase the signal-to-noise ratio. (3) The optical filters should be carefully selected for optimal excitation and emission wavelengths and also to prevent cross-talk between FITC and Texas Red that would diminish the tumor-specific gains and losses. (4) The metaphase chromosomes selected for image acquisition should be well spread with ≥70% of the chromosomes appearing as single objects. (5) The exposure times should be adjusted before acquisition, so that the complete dynamic range of the CCD camera is being used. It is best to have identical exposure times for the acquisition of all FITC or all Texas Red images, respectively, of one tumor throughout the session. (6) All images for one tumor should be acquired within the same session to keep conditions as similar as possible (e.g., age of the mercury lamp). IMAGE ANALYSIS The data are subjected to a comprehensive analysis that consists of a number of steps. For more detailed explanations of the optical, physical, and mathematical background, see du Manoir et al. (1995a,b) and Piper et al. (1995). Commercial software packages are available for the CGH analysis of human tumors and cancers of other species. First, the images need to be aligned in case there is any image shift during image acquisition. Background subtraction procedures are crucial and determine the quality of the results.
BASIC PROTOCOL 9
Molecular Cytogenetics
8.12.17 Current Protocols in Cytometry
Supplement 18
1
4
3
2 >
6
>
7
X
8
9
10
11
14
12 >
< 13
5
15
16
17
18
22
Y
< 19
20
21
Figure 8.12.5 Scheme for chromosome identification based on DAPI banding, indicating the typical landmarks.
Identification of Copy Number Changes in Tumor Genomes by CGH
An average background value is calculated and subtracted from all image points, or the image is divided into smaller image regions, and then the average background value will be determined and subtracted for each region separately. The measurement mask and the central chromosome axis have to be determined and should be based on the segmentation of the reference image (UNIT 10.9). All chromosomes will be stretched to a predefined length. The user needs to interactively classify the DAPI-banded chromosomes. A simple scheme with the basic landmarks is provided in Figure 8.12.5. After chromosome karyotyping, the FITC and Texas Red intensity profiles as well as the average ratio values are calculated. Finally, the ratio profile is plotted next to the chromosome idiograms (ISCN) and all average ratio values are collected in a histogram. The most frequent average ratio value in the histogram is defined as the ratio of 1.0 and applied for data normalization. Using known cell lines and cytogenetically characterized tumors, thresholds for gains and losses have been determined empirically as being 0.75 or 0.8 for a loss and 1.25 or 1.2 for a gain (Kallioniemi et al., 1992; du Manoir et al., 1993). For example, a threshold of 1.25 for a gain means that the average ratio profile indicates this gain, if 50% of the cells that were used for DNA extraction carry this aberration. Commercial CGH software packages include most of these features without user involvement in any of the processes. However, the user is asked to check that all chromosomes are identified correctly and to interpret the data. For that, the average ratio profiles may be used to look at every single metaphase and determine if all the single profiles for each chromosome match the average ratio profile. If a single profile appears different, the chromosome might be misclassified, show some background spots on the hybridization images, or be mechanically disturbed, or the position of the centromere and the measurement mask were not correctly established. Either the data should be corrected or the chromosome should be deleted. The decision as to which chromosomal bands show gains and losses should be made by comparing the average ratio profile with the very best single metaphase spread that is characterized by a smooth hybridization pattern and by long chromosomes that do not overlap. The resolution in a single image is higher than when looking at an average ratio profile. Finally, data for several tumors of the very same biological background and
8.12.18 Supplement 18
Current Protocols in Cytometry
a similar clinical course can be compared with different tumors, depending on the biological question of interest. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Antibodies/fluorescent dye solutions Microcentrifuge all stocks of fluorescent dyes for 3 min at maximum speed and use supernatant to prepare the following dilutions solutions: Solution 1: 1:200 avidin-FITC (Jackson Laboratories) + 1:200 mouse-anti-DIG (Sigma) Solution 2: 1:200 biotinylated anti-avidin (Vector) + 1:200 rabbit anti-mouse Texas Red (Jackson Laboratories) Solution 3: 1:200 avidin-FITC (Jackson Laboratories) + 1:200 donkey anti-rabbit Texas Red (Jackson Laboratories) Prepare fresh Antifade (1,4-phenylenediamine) Mix 10 ml of 0.5 M sodium bicarbonate, pH 8.13, and 40 ml of 0.5 M sodium carbonate, pH 11.32, to make a carbonate-bicarbonate buffer, pH 9.0. Filter sterilize. Dissolve 100 mg 1,4-phenylenediamine (Sigma) in 4 ml of 1× PBS (APPENDIX 2A). Adjust pH slowly with carbonate-bicarbonate buffer to 8.0. Add 1× PBS (APPENDIX 2A) to a final volume of 10 ml. Mix with 90 ml of 86% (v/v) glycerol. Dispense aliquots into UV-safe tube, store ≤3 months at −20°C, discard when brown. Blocking solution (3% BSA) Add 0.3 g bovine serum albumin (BSA) to 10 ml prewarmed 4× SSC/Tween 20 (see recipe). Heat 30 min at 37°C. Vortex until dissolved. Prepare fresh. BrdU (bromodeoxyuridine), 10 mg/ml Dissolve 10 mg/ml bromodeoxyuridine (Sigma) in sterile 1× PBS (APPENDIX 2A) and filter sterilize; store up to 1 year at −20°C. Concanavalin A, 1 mg/ml Resuspend in sterile 1× PBS to 1 mg/ml, then filter sterilize. Store up to 1 year at −20°C. DAPI solutions Stock solution: Prepare 2 mg DAPI/10 ml sterile water. Store up to 1 year at −20°C. Working solution:Add 20 µl DAPI stock solution to 100 ml of 2× SSC (see APPENDIX 2A for 20×). Store at 4°C; will be usable at least 2 weeks. DNA buffer I 200 µl 1 M Tris⋅Cl, pH 8.0 (APPENDIX 2A; 0.2 M final) 200 µl 0.5 M EDTA (0.1 M final) 600 µl sterile water Prepare fresh
Molecular Cytogenetics
8.12.19 Current Protocols in Cytometry
Supplement 18
DNA buffer II 1.5 ml 5 M NaCl (75 mM final) 5 ml 0.5 M EDTA (25 mM final) 0.5 ml Tween 20 (0.5% final) 93 ml sterile water Prepare fresh DNase I, 1 mg/ml 10 mg DNase I from bovine pancreas (Roche Diagnostics) 1.5 ml 1 M NaCl (0.15 M) 5 ml glycerol (50% v/v final) Add sterile water to a final volume of 10 ml Store aliquots up to 1 year at −20°C dNTP mix, 10× Combine the following (all available from Roche Diagnostics): 5 µl each of 100 mM dATP, dCTP, dGTP (0.5 mM each dNTP, final) 1 µl of 100 mM dTTP (0.1 mM final) 984 µl sterile water Prepare aliquots, store up to 1 year at −20°C Formamide, 50%/SSC, 2× 30 ml 20× SSC (APPENDIX 2A) 120 ml sterile water 150 ml formamide Adjust pH to 7.0 with 1 M HCl Heat 30 min at 45°C Prepare fresh Formamide, 70%/SSC, 2× 70 µl deionized formamide 3 µl 20× SSC (APPENDIX 2A) 27 µl sterile water Adjust pH to 7.0 with 1 M HCl Store up to 1 year at −20°C FUdR (fluorodeoxyuridine), 0.1 mg/ml Dissolve 0.1 mg/ml fluorodeoxyuridine (Sigma) in sterile 1× PBS (APPENDIX 2A) and sterile filter; store up to 1 year at −20°C. Hybridization mixture 50 ml deionized formamide 5 ml 20× SSC (APPENDIX 2A) 25 ml sterile water Adjust pH to exactly 7.0 Add 20 ml of 50% (w/v) dextran sulfate (Oncor) Stir, autoclave, dispense into aliquots, and store up to 6 months at 4°C LPS (lipopolysaccharide), 25 mg/ml Resuspend 25 mg/ml lipopolysaccharide (Sigma) in sterile 1× PBS (APPENDIX 2A); store up to 1 year at −20°C. Identification of Copy Number Changes in Tumor Genomes by CGH
8.12.20 Supplement 18
Current Protocols in Cytometry
Lysis buffer 8.29 g NH4Cl (155 mM final) 1 g KHCO3 (10 mM final) 0.034 g Na2EDTA (0.1 mM final) or 200 µl 0.5 M EDTA 1000 ml sterile water Adjust pH 7.4 with 1 M HCl or NaOH Prepare fresh 2-Mercaptoethanol, 0.1 M Pipet 34.7 µl of 99% (14.4 M) stock solution into sterile water to a total volume of 5 ml. Dispense into 1-ml aliquots, store up to 1 year at −20°C. NT buffer, 10× 500 µl 1 M Tris⋅Cl, pH 8.0 (APPENDIX 2A; 0.5 M final) 100 µl 0.5 M MgCl2 (50 mM final) 50 µl 1% (10 mg/ml) BSA (0.5 mg/ml final) 350 µl sterile water Dispense aliquots, store up to 1 year at −20°C PBS/MgCl2, 1× 50 ml 1 M MgCl2 950 ml 1× PBS Prepare fresh Pepsin, 10% Dissolve 100 mg pepsin (Sigma) in 100 ml sterile water. Keep on ice; divide into 50-µl aliquots and store up to 1 year at −20°C. RNase A stock, 20 mg/ml Dissolve 20 mg RNase A (Roche Diagnostics) per ml sterile water. Boil 15 min, cool to room temperature, divide into aliquots, and store up to 1 year at −20°C. SE buffer 4.39 g NaCl (75 mM final) 8.41 g Na2EDTA (25 mM final) or 50 ml 0.5 M EDTA 1000 ml sterile water Adjust pH 8.0 with 1 M NaOH Prepare fresh SSC/Tween 20, 4× 100 ml 20× SSC (APPENDIX 2A) 400 ml sterile water 0.5 ml Tween 20 Heat 30 min at 45°C Prepare fresh COMMENTARY Background Information The basic scheme of comparative genomic hybridization (CGH) is the comparison of DNA copy number changes between two genomes. As with loss of heterozygosity (LOH) analysis, DNA from the tumor only is used to perform
the analysis. It is not necessary to obtain control DNA from the very same patient, however; any sex-matched normal DNA can be used. The achieved results are relative copy numbers of regions in the test or tumor genome. Molecular Cytogenetics
8.12.21 Current Protocols in Cytometry
Supplement 18
CGH has truly become a valuable tool for studying the genetic makeup of thousands of archived tumors. It is the only method that reveals gains and losses of genomic regions using DNA from formalin-fixed tissue (Speicher et al., 1993). The combination of histomorphology, tissue microdissection, and screening for genetic aberrations by CGH allows for a phenotype-genotype correlation of tumors (Heselmeyer et al., 1996; Ried et al. 1997, 1999). Thus, identification of tumor-specific and stage-specific genetic aberrations has facilitated gene-mapping strategies and the application of interphase cytogenetics in clinical diagnostics. With the development of arrayCGH, a precise definition of the genomic region of interest became possible, and tumor suppressor gene and oncogene cloning efforts will be much more efficient. Furthermore, the correlation of gene-specific gains and losses with gene expression data and clinical information will result in a big step towards an individualized molecular medicine for cancer patients.
Identification of Copy Number Changes in Tumor Genomes by CGH
Chromosome identification based on DAPI banding and co-hybridization with centromere enumeration probes Correct identification of chromosomes is extremely important because it is the basis for accurate CGH results. Figure 8.12.5 provides a simple scheme showing landmarks that are useful for chromosome classification based on DAPI banding. This figure was originally a sketch for teaching noncytogeneticists how to identify normal chromosomes. Later on, the layout was brought into a more professional shape and placed onto the Internet by Dr. Iver Petersen (http://amba.charite.de/cgh/ protocol/02/class.html). In order to further facilitate accuracy and speed of chromosome identification steps during CGH image analysis, the cohybridization of centromere enumeration probes (CEP) for certain chromosomes is performed as part of the CGH hybridization procedure. DNA probes specific for human centromeres are prepared using PCR with alpha-satellite-specific primers or can be purchased from commercial suppliers (e.g., Vysis, Inc.). Mouse chromosome identification probes are described in Alternate Protocol 3. These probes should be labeled with a third color besides green and red, preferably with Cy5 (near infrared, Cy5-dUTP, Amersham Pharmacia Biotech). Figure 8.12.4 demonstrates the usefulness of such a strategy. Imageanalysis software for CGH has been and could
be adjusted to handle this additional information (Leica Microsystems; Applied Imaging Corp.). Accuracy of automated and user-interactive karyotyping based on DAPI banding will be improved from ∼60% to ∼95% using the additional information from centromeric probe signals or BAC clones, respectively. In turn, gains and losses will be mapped correctly to chromosomal regions that harbor oncogenes and tumor suppressor genes, thus facilitating their identification. Historical development CGH was developed by Anne and Olli Kallioniemi in the laboratory of Dan Pinkel of the University of California at San Francisco (Kallioniemi et al., 1992). Other groups followed and presented technical and software developments for the application of CGH (du Manoir et al., 1993, 1995a,b; Kallioniemi et al., 1994b; Lundsteen et al., 1995; Tirkkonen et al., 1996; Moore et al., 1997; Van Dekken et al., 2000). The analysis of cell lines and primary tumors by a number of research groups has resulted in a large data pool (Kallioniemi et al., 1994a; Schröck et al., 1994; Heselmeyer et al., 1996; for reviews see Forozan et al., 1997; Ried et al., 1997, 1999). Possibilities for further improvements include the removal of repetitive DNA sequences by affinity chromatography (Craig et al., 1997) and optimized DOP-PCR strategies for amplification of genomic DNA (Huang et al., 2000). Recently, the application of CGH for single-cell analysis was successfully accomplished (Klein et al., 1999). An important step towards the genetic analysis of animal models of human cancer was the development of mouse CGH (Donehower et al., 1995; Weaver et al., 1999; compare Figs. 8.12.7 and 8.12.8) and the use of other animal models (e.g., rat). Here, the comparison between tumors with a defined genetic makeup and the transient expression or suppression of tumor-specific genes allow for studies of gene function and genetic pathways. The latest breakthrough, however, has been accomplished with the development of array-CGH, resulting in a tremendous improvement of resolution (SolinasToldo et al., 1997; Pinkel et al., 1998; Albertson et al., 2000). CGH on microarrays The use of metaphase chromosomes as a representation of the genome limits the power of CGH to detect alterations involving small regions of the genome (<12 Mbp; Bentz et al.,
8.12.22 Supplement 18
Current Protocols in Cytometry
14
3
12
2.5 Normalized ratio
Normalized ratio
A 10 8 6 4 2 0
0
0.2
0.4 FLpter
0.6
2 1.5 1 0.5 0
1
0
0.2
0.4 FLpter
0.6
1
B 14
S21 B4087 B4195 B4198
B4121
12
10 8 6 4 2
14
S50
B4097
Normalized ratio
Normalized ratio
12
B4121
10 8 6 4
B4097
B4090
2 B4087
0
ZNF218 ZNF217 NABC1 CYP24 centromere Position 20q13.2
0 telomere
ZNF218 ZNF217 NABC1 CYP24 centromere Position 20q13.2
telomere
Figure 8.12.6 Array-CGH mapping of copy number alterations occurring on chromosome 20 in breast cancer. (A) Copy number profiles on chromosome 20 from two breast tumors determined using arrays of genomic clones spaced at 1 to 3 Mbp along chromosome 20. Both tumors show high-level amplification at 20q13.2 (after Pinkel et al., 1998). (B) Higher resolution analysis of copy number in the breast tumors using an array of contiguous genomic clones across a ∼2 Mbp region at 20q13.2. Each clone in the array is represented by a horizontal bar that indicates the location and length of the clone as determined by STS content mapping. The high-resolution copy number profiles show narrow peaks, suggesting selection for amplification of genes in these regions. The peak in tumor S50 maps to the proximal region of the contig and is centered on ZNF217. In tumor S21, on the other hand, there is elevated copy number at ZNF217, but the copy number maximum maps more distally and includes CYP24, a gene involved in regulation of vitamin D signaling. Note that the array used for the low-resolution analysis of S21 did not contain the clones (RMC20B421 and RMC20B4087) that mapped to the copy number maximum in S21. The names of some clones are shown with the RMC20 prefix omitted (after Albertson et al., 2000).
1998), to resolve closely spaced aberrations, and to relate copy number changes to genetic or genomic markers. To overcome these limitations, several approaches have recently been described in which arrays of clones, rather than metaphase chromosomes, are used to represent the genome (Solinas-Toldo et al., 1997; Pinkel et al., 1998; Pollack et al., 1999; Albertson et al., 2000; Heiskanen et al., 2000). Hybridization is carried out to the microarrays of the mapped clones and copy number is related to the test/reference fluorescence ratio on each clone. The resolution of CGH is then determined by the length of the clones and their spacing along the chromosome. For example, the entire genome might be scanned for copy number changes using a set of ∼3000 clones spaced at ∼1 Mbp intervals, while a greater
density of clones might be used to provide higher resolution measurements in regions of particular interest. Figure 8.12.6 shows the application of two types of arrays to measure copy number aberrations occurring on chromosome 20 in breast tumors. Both tumors showed copy number changes on 20q. When the whole chromosome was scanned, one tumor showed several regions of amplification, while in the other a single region was detected (Pinkel et al., 1998). A higher-resolution analysis was carried out with arrays of overlapping clones spanning a ∼2 Mbp region at 20q13.2 (Fig. 8.12.6B) and provided precise information on the locations of both the boundaries and the maxima of the amplicons, thereby resolving two closely spaced regions of amplification (Albertson et al., 2000).
Molecular Cytogenetics
8.12.23 Current Protocols in Cytometry
Supplement 18
Table 8.12.1
BAC Resources
Source
Website
Reference
GenMap DB Korenberg Laboratory
http://genomics.med.upenn.edu/genmapdb/ http://www.csmc.edu/genetics/korenberg/ korenberg.html#A National Cancer Institute http://www.ncbi.nlm.nih.gov/CCAP/ Cancer Chromosome Aberration Project (CCAP) National Cancer Institute http://www.nci.nih.gov/dcbBACRESOR.htm Division of Cancer Biology BAC Resource
A
B
C
D
Cheung et al. (1999) Korenberg et al. (1999) Kirsch et al. (2000)
Figure 8.12.7 Example of a CGH analysis of a mammary tumor obtained from a mouse conditionally mutated for Brca1. (A) DAPI-banding picture including chromosome identification. (B) FITC image showing the hybridization pattern of the tumor DNA. (C) TRITC image indicating the normal control-DNA hybridization. (D) FITC/TRITC overlay demonstrating, e.g., a gain on chromosome 6 and a gain close to the centromere on chromosome 9 (arrowhead). In contrast, increased intensity values are found on chromosome 1 in the FITC and the TRITC images (arrow), reflecting hybridization of repetitive DNA sequences in both images that do not result in a change of DNA copy numbers in the average ratio profile (compare to Figure 8.12.8; Xu et al., 1999). See color plate.
Identification of Copy Number Changes in Tumor Genomes by CGH
8.12.24 Supplement 18
Current Protocols in Cytometry
Arrays of clones for CGH have used either large-insert genomic clones such as BACs, P1s, and cosmids (Solinas-Toldo et al., 1997; Pinkel et al., 1998; Albertson et al., 2000) or cDNAs (Pollack et al., 1999; Heiskanen et al., 2000). In addition, an array of BAC clones for analysis of copy number at loci of known oncogenes is commercially available (Vysis). Robust hybridization signals are generally obtained using the large-insert genomic clones (BACs, etc.), so that changes in copy number, e.g., from two copies to one copy, can be measured reliably on individual clones (Pinkel et al., 1998). Several resources of mapped BAC clones suitable for CGH are listed in Table 8.12.1. However, preparing sufficient quantities of DNA from a large number of BACs that are low copy number vectors remains a challenge. Thus, one advantage of using cDNA arrays for CGH is the availability of arrays of these clones, as well as the convenience of a single type of array for measuring copy number or gene expression. Currently, the measurement precision is lower on the smaller cDNAs than on the large-insert genomic clones, due to reduced fluorescence
signal intensity of the hybridization. Reliable detection of deletions on these arrays can be achieved by averaging ratios across several clones (Pollack et al., 1999). The development and application of arrays for the analysis of DNA copy number is a new and rapidly evolving field. Therefore, it is likely that future improvements in both the production and arraying of BACs will facilitate production of these types of arrays, while methods for enhancing signals on cDNA clones will increase their ability to reliably measure small copy number changes.
Critical Parameters and Troubleshooting For tumor DNA extraction, it is important to collect samples showing a high amount of tumor cells, preferably after performing tissue microdissection (UNIT 8.6). Normal cells present in the sample will diminish the chance of detecting chromosomal aberrations. DNA copy number changes that occur in <40% of the tumor cells might not be detected at all.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
X
Figure 8.12.8 Average ratio profile of the mouse mammary tumor shown in Figure 8.12.7. Numerous gains were identified on almost all chromosomes except for chromosome 17. Specifically, a gain was detected on chromosome 11 in band E, a region that is syntenic to the long arm of chromosome 17 in humans. This region is recurrently gained in human breast cancers and harbors the oncogene Her2-neu. Interphase cytogenetics using the DNA clone for Her2-neu reveals a consistent gain and amplification of this gene in mouse mammary tumors (Weaver et al., manuscript in preparation).
Molecular Cytogenetics
8.12.25 Current Protocols in Cytometry
Supplement 18
A
1
2
3
6
7
8
13
14
15
19
20
4
9
10
16
21
5
11
12
17
18
22 X
B
1
Identification of Copy Number Changes in Tumor Genomes by CGH
2
6
7
13
14
19
20
3
4
8
9
15
10
16
21
22
5
11
17
12
18
X
Y
Figure 8.12.9 Legend on next page.
8.12.26 Supplement 18
Current Protocols in Cytometry
Figure 8.12.9 (Figure on facing page) (A) Karyogram of an overlaid FITC/Texas Red image visualizing the DNA copy number changes in the tumor (SKBR3, breast cancer cell line) compared to the control DNA (normal DNA). Gains in the tumor DNA appear in green (e.g., gene amplifications on 8q, gain of chromosome arms 5p and 7p), whereas losses appear in red (e.g., loss of chromosome arms 4p and 9p). Regions with equal copy numbers show an orange mixed color (e.g., regions on chromosome arms 2q and 12q). The second chromosome 4 was missing in this metaphase. See color plate. (B) Average ratio profile (black) indicating numerous gains and losses of specific chromosomal regions. The straight dotted dark lines to the right of the ratio value of 1.0 show the threshold for a gain (ratio value of 1.2), whereas the straight dashed lighter lines to the left visualize the border for a loss (ratio value of 0.8). Similarly, the gains and losses of chromosomal regions are also depicted by straight dotted dark and straight dashed lighter bars next to the chromosome ideogram based on the comparison of the average ratio values in B with the appearance of gains and losses in a single metaphase spread shown in A.
The hybridization quality will be better when using high-molecular-weight DNA obtained from fresh or frozen tissue. It is more difficult to obtain acceptable results from formalin-fixed tumors. The formalin used for tissue fixation should be prepared using PBS (buffered formalin), and the fixation time may be rather short. If available, 1 to 2 µg DNA might be labeled and used for hybridization. The spreading of metaphase chromosomes onto glass slides should be performed within 1 week after chromosome preparation to obtain well-spread metaphases. Lower air pressure and higher humidity as experienced on rainy days will facilitate the spreading. If available, a temperature and humidity-controlled environment may allow for optimal dropping conditions. One of the important steps during CGH experiments is the blocking of repetitive DNA sequences. The labeled tumor and control genomic DNA and the target chromosomes contain several types of repetitive DNA sequences. Without any suppression, these sequences will bind to target sequences on all chromosomes in a random manner, making it difficult to identify DNA copy number changes for single-copy sequences. With incomplete suppression, an R-banding pattern will be visible, in particular as bright fluorescent regions and bands on chromosomes 1pter, 16, 19, 22, and also 7 (see Fig. 8.12.7). These regions are visible in the green and in the red images, indicating that these reflect hybridization problems rather than tumor-specific aberrations. It is therefore of particular importance to use high-quality Cot-1 DNA (Life Technologies) in an excess amount in order to achieve a nearly complete blocking of the repetitive DNA sequences. This aspect is even more important when performing mouse CGH. Using Cot-1 DNA from one strain is not enough to completely suppress repetitive sequences of another
strain. It would therefore be best to prepare a pool of all mouse strains used in the laboratory and prepare Cot-1 DNA out of these mixed DNAs. Several commercial software packages are now available for CGH image acquisition and analysis of mouse chromosomes (e.g., Leica Microsystems and Applied Imaging). These programs apply the same quantitation methods used for human CGH but map the results to the mouse ideograms. The authors have found that even though hybridization is performed with an excess of mouse Cot-1 DNA, the suppression at the mouse centromeres is never complete. The final profiles often show a peak at the centromere that may extend slightly below it, depending on the strain used for the target chromosomes (see Fig. 8.12.8, chromosomes 4, 7, 8, 10, 14, 18, and X). It is important that the software account for this “background” by either (1) starting the measurements right below the centromere, (2) including the option to normalize the test profile against a “control” profile (wild-type DNA hybridized to wildtype DNA), or (3) simply shading the centromere region of the profile to indicate that the information from this portion of the profile is not accurate. In any case, it is wise with any program to test the thresholds and the integrity of the final profile by analyzing a cell line that has known gains or losses or by hybridizing XX versus XY DNA.
Anticipated Results Figure 8.12.9 exemplifies the CGH analysis of SKBR3, a breast cancer cell line that is used as a positive control in the authors’ laboratory when hybridizing unknown tumor samples.
Time Considerations
DNA extraction takes ∼3 days when using paraffin-embedded tissue and ∼2 days when using frozen tissue and performing an over-
Molecular Cytogenetics
8.12.27 Current Protocols in Cytometry
Supplement 18
night proteinase K digestion. For preparation of normal target chromosomes (72 hr of culture), 4 days are needed, which includes the time needed for dropping of several hundred slides. Nick translation and hybridization can be performed in 1 day, if the right size of the DNA is achieved within the first experiment (400 to 1000 bp). If the DNA is too short, the nick translation needs to be repeated; if the DNA is too long, additional DNase I should be added to cut the fragments to a shorter size. However, the optimum time frame for dUTP incorporation lies between 1.5 and 2 hr. If the cutting needs to continue, it would be better to start over from the beginning with a higher amount of DNase I. Hybridization should be allowed to take place at least 2 nights, or 48 hr, to allow most of the single-copy sequences to find their match. Fluorescence detection of indirectly labeled nucleotides requires ∼6 hr. Image acquisition of 10 metaphase spreads in all four colors will take the experienced user ∼1.5 to 2 hr. An additional 2 to 3 hr will be needed for data analysis of one tumor, depending on the user-friendliness of the software. Finally, data interpretation and comparison between different tumors might also require some time depending on the questions asked.
Literature Cited Albertson, D.G., Ylstra, B., Segraves, R., Collins, C., Dairkee, S.H., Kowbel, D., Kuo, W.-L., Gray, J.W., and Pinkel, D. 2000. Quantitative mapping of amplicon structure by array CGH identifies CYP24 as a candidate oncogene. Nature Genet. 25:144-146. Bentz, M., Plesch, A., Stilgenbauer, S., Dohner, H., and Lichter, P. 1998. Minimal sizes of deletions detected by comparative genomic hybridization. Genes Chrom. Cancer 2:172-175. Cheung, V.G., Dalrymple, L., Narasimhan, S., Watts, J., Schuler, G., Raap, A.K., Morley, M., and Bruzel, A. 1999. A resource of mapped human bacterial artificial chromosome clones. Genome Res. 9:989-993. Craig, J.M., Kraus, J., and Cremer, T. 1997. Removal of repetitive sequences from FISH probes using PCR-assisted affinity chromatography. Hum. Genet. 100:472-476. Donehower, L.A., Godley, L.A., Aldaz, C.M., Pyle, R., Shi, Y.P., Pinkel, D., Gray, J., Bradley, A., Medina, D., and Varmus, H.E. 1995. Deficiency of p53 accelerates mammary tumorigenesis in Wnt-1 transgenic mice and promotes chromosomal instability. Genes Dev. 9:882-895. Identification of Copy Number Changes in Tumor Genomes by CGH
du Manoir, S., Speicher, M.R., Joos, S., Schröck, E., Popp, S., Dohner, H., Kovacs, G., Robert-Nicoud, M., Lichter, P., and Cremer, T. 1993. Detection of complete and partial chromosome
gains and losses by comparative genomic in situ hybridization. Hum. Genet. 90:590-610. du Manoir, S., Kallioniemi, O.P., Lichter, P., Piper, J., Benedetti, P.A., Carothers, A.D., Fantes, J.A., Garcia-Sagredo, J.M., Gerdes, T., Giollant, M., et al. 1995a. Hardware and software requirements for quantitative analysis of comparative genomic hybridization. Cytometry 19:4-9. du Manoir, S., Schröck, E., Bentz, M., Speicher, M.R., Joos, S., Ried, T., Lichter, P., and Cremer, T. 1995b. Quantitative analysis of comparative genomic hybridization. Cytometry 19:27-41. Forozan, F., Karhu, R., Kononen, J., Kallioniemi, A., and Kallioniemi, O.P. 1997. Genome screening by comparative genomic hybridization. Trends Genet. 13:405-409. Gray, J.W. and Collins, C. 2000. Genome changes and gene expression in human solid tumors. Carcinogenesis 21:443-452. Heiskanen, M.A., Bittner, M.L., Chen, Y., Khan, J., Adler, K.E., Trent, J.M., and Melzer, P.S. 2000. Detection of gene amplification by genomic hybridization to cDNA microarrays. Cancer Res. 60:799-802. Heselmeyer, K., Schröck, E., du Manoir, S., Blegen, H., Shah, K., Steinbeck, R., Auer, G., and Ried, T. 1996. Gain of chromosome 3q defines the transition from severe dysplasia to invasive carcinoma of the uterine cervix. Proc. Natl. Acad. Sci. U.S.A. 93:479-484. Huang, Q., Schantz, S.P., Rao, P.H., Mo, J., McCormick, S.A., and Chaganti, R.S. 2000. Improving degenerate oligonucleotide primed PCR-comparative genomic hybridization for analysis of DNA copy number changes in tumors. Genes Chrom. Cancer 28:395-403. Kallioniemi, A., Kallioniemi, O.P., Sudar, D., Rutovitz, D., Gray, J.W., Waldman, F., and Pinkel, D. 1992. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 258:818-821. Kallioniemi, A., Kallioniemi, O.P., Piper, J., Tanner, M., Stokke, T., Chen, L., Smith, H.S., Pinkel, D., Gray, J.W., and Waldman, F.M. 1994a. Detection and mapping of amplified DNA sequences in breast cancer by comparative genomic hybridization. Proc. Natl. Acad. Sci. U.S.A. 91:21562160. Kallioniemi, O.P., Kallioniemi, A., Piper, J., Isola, J., Waldman, F.M., Gray, J.W., and Pinkel, D. 1994b. Optimizing comparative genomic hybridization for analysis of DNA sequence copy number changes in solid tumors. Genes Chrom. Cancer 10:231-243. Karhu, R., Kahkonen, M., Kuukasjarvi, T., Pennanen, S., Tirkkonen, M., and Kallioniemi, O. 1997. Quality control of CGH: Impact of metaphase chromosomes and the dynamic range of hybridization. Cytometry 28:198-205.
8.12.28 Supplement 18
Current Protocols in Cytometry
Kirsch, I.R., Green, E.D., Yonescu, R., Strausberg, R., Carter, N., Bentley, D., Leversha, M.A., Dunham, I., Braden, V.V., Hilgenfeld, E., Schuler, G., Lash, A.E., Shen, G.L., Martelli, M., Kuehl, W.M., Klausner, R.D., and Ried, T. 2000. A systematic, high-resolution linkage of the cytogenetic and physical maps of the human genome. Nature Genet. 24:339-340.
Schröck, E., Thiel, G., Lozanova, T., du Manoir, S., Meffert, M.C., Jauch, A., Speicher, M.R., Nurnberg, P., Vogel, S., Jänisch, W., Donis-Keller, H., Ried, T., Witkowski, R., and Cremer, T. 1994. Comparative genomic hybridization of human malignant gliomas reveals multiple amplification sites and nonrandom chromosomal gains and losses. Am. J. Pathol. 144:1203-1218.
Klein, C.A., Schmidt-Kittler, O., Schardt, J.A., Pantel, K., Speicher, M.R., and Riethmuller, G. 1999. Comparative genomic hybridization, loss of heterozygosity, and DNA sequence analysis of single cells. Proc. Natl. Acad. Sci. U.S.A. 96:4494-4499.
Solinas-Toldo, S., Lampel, S., Stilgenbauer, S., Nickolenko, J., Benner, A., Dohner, H., Cremer, T., and Lichter, P. 1997. Matrix-based comparative genomic hybridization: Biochips to screen for genomic imbalances. Genes Chrom. Cancer 20:399-407.
Korenberg, J.R., Chen, X.-N., Sun, Z., Shi, Z.-Y., Ma, S., Vataru, E., Yimlamai, D., Weissenbach, J.S., Shizuya, H., Simon, M.I., Gerety, S.S., Nguyen, H., Zemsteva, I.S., Hui, L., Silva, J., Wu, X., Birren, B.W., and Hudson, T.J. 1999. Human genome anatomy: BACs integrating the genetic and cytogenetic maps for bridging genome and biomedicine. Genome Res. 9:9941001.
Speicher, M.R., du Manoir, S., Schröck, E., Holtgreve-Grez, H., Schoell, B., Lengauer, C., Cremer, T., and Ried, T. 1993. Molecular cytogenetic analysis of formalin-fixed, paraffin-embedded solid tumors by comparative genomic hybridization after universal DNA-amplification. Hum. Mol. Genet. 2:1907-1914.
Lundsteen, C., Maahr, J., Christensen, B., Bryndorf, T., Bentz, M., Lichter, P., and Gerdes, T. 1995. Image analysis in comparative genomic hybridization. Cytometry 19:42-50. Moore, D.H. II, Pallavicini, M., Cher, M.L., and Gray, J.W. 1997. A t-statistic for objective interpretation of comparative genomic hybridization (CGH) profiles. Cytometry 28:183-190. Pinkel, D., Segraves, R., Sudar, D., Clark, S., Poole, I., Kowbel, D., Collins, C., Kuo, W.-L., Chen, C., Zhai, Y., Dairkee, S.H., Ljung, B.-M., Gray, J.W., and Albertson, D.G. 1998. Quantitative high resolution analysis of DNA copy number variation in breast cancer using comparative genomic hybridization to DNA microarrays. Nature Genet. 20:207-211. Piper, J., Rutovitz, D., Sudar, D., Kallioniemi, A., Kallioniemi, O.P., Waldman, F.M., Gray, J.W., and Pinkel, D. 1995. Computer image analysis of comparative genomic hybridization. Cytometry 19:10-26.
Tirkkonen, M., Karhu, R., Kallioniemi, O., and Isola, J. 1996. Evaluation of camera requirements for comparative genomic hybridization. Cytometry 25:394-398. Van Dekken, H., Krijtenburg, P.J., and Alers, J.C. 2000. DNA in situ hybridization (interphase cytogenetics) versus comparative genomic hybridization (CGH) in human cancer: Detection of numerical and structural chromosome aberrations. Acta Histochem. 102:85-94. Weaver, Z.A., McCormack, S.J., Liyanage, M., du Manoir, S., Coleman, A., Schröck, E., Dickson, R.B., and Ried, T. 1999. A recurring pattern of chromosomal aberrations in mammary gland tumors of MMTV-cmyc transgenic mice. Genes Chrom. Cancer 25:251-260. Xu, X., Wagner, K.U., Larson, D., Weaver, Z., Li, C., Ried, T., Hennighausen, L., Wynshaw-Boris, A., and Deng, C.X. 1999. Conditional mutation of Brca1 in mammary epithelial cells results in blunted ductal morphogenesis and tumour formation. Nature Genet. 22:37-43.
Pollack, J.R., Perou, C.M., Alizadeh, A.A., Eisen, M.B., Pergamenschikov, A., Williams, C.F., Jeffrey, S.S., Botstein, D., and Brown, P.O. 1999. Genome-wide analysis of DNA copy-number changes using cDNA microarrays. Nature Genet. 23:41-46.
Key References
Ried, T., Liyanage, M., du Manoir, S., Heselmeyer, K., Auer, G., Macville, M., and Schröck, E. 1997. Tumor cytogenetics revisited: Comparative genomic hybridization and spectral karyotyping. J. Mol. Med. 75:801-814.
du Manoir et al., 1995b. See above.
Ried, T., Heselmeyer-Haddad, K., Blegen, H., Schröck, E., and Auer, G. 1999. Genomic changes defining the genesis, progression, and malignancy potential in solid human tumors: A phenotype/genotype correlation. Genes Chrom. Cancer 25:195-204.
Development of CGH du Manoir et al., 1993. See above. du Manoir et al., 1995a. See above. Kallioniemi et al., 1992. See above. Piper et al., 1995. See above.
CGH using DNA from formalin-fixed tissue Speicher et al., 1993. See above.
Genotype-phenotype correlation Heselmeyer et al., 1996. See above. Ried et al., 1997. See above. Ried et al., 1999. See above. Molecular Cytogenetics
8.12.29 Current Protocols in Cytometry
Supplement 18
Single-cell CGH analysis Klein et al., 1999. See above.
Array CGH development
Contributed by Evelin Schröck Institute for Genetic Medicine, Charité Berlin, Germany
Heiskanen et al., 2000. See above. Pinkel et al., 1998. See above. Solinas-Toldo et al., 1997. See above.
Internet Resources http://www.aicorp.com Web site for Applied Imaging.
Zoë Weaver National Cancer Institute (NCI/NIH) Bethesda, Maryland Donna Albertson University of California San Francisco, California
http://www.leica-microsystems.com Web site for Leica Microsystems Imaging Solutions Ltd., Cambridge, U.K. http://www.metasystems.de Web site for Metasystems GmbH. http://www.vysis.com Web site for Vysis, Inc.
The authors wish to thank Dr. Dan Pinkel (Cancer Center, UCSF, San Francisco, CA) and Dr. Thomas Ried (Genetics Department, NCI/NIH, Bethesda, MD) for continued support and collaboration.
Identification of Copy Number Changes in Tumor Genomes by CGH
8.12.30 Supplement 18
Current Protocols in Cytometry
Combined Immunofluorescence and FISH: New Prospects for Tumor Cell Detection/Identification
UNIT 8.13
The term cytomics has recently been introduced to define developments in the field of cell-based analyses that integrate genomics and proteomics. Improvements in microscope technology (e.g., digital image analysis systems, confocal laser microscopy, and automation) facilitate in-depth studies at the cellular level. The authors of this unit have paid special attention to detection/quantification and molecular characterization using fluorescence microscopy–based technology. This includes the detection of rare tumor cells in the blood, bone marrow, or apheresis products of cancer or leukemia patients, as well as of circulating fetal cells from maternal blood for noninvasive prenatal testing. Furthermore, certain biological features, i.e., proliferation activity or apoptotic rate, can be tested and correlated with clinical features. These studies are facilitated by a microscopy device for which the authors propose the term FLAME (fluorescence-based automatic microscope). This unit describes methods that have been found to be particularly useful for the demonstration of various phenotypic and genotypic features of a specific target cell. Through the use of several immunological and molecular biological protocols, the complex biological nature of single cells can be elucidated. The authors give an overview of basic biotechnological hardware and software requirements that make simultaneous and/or sequential immunological/molecular cytogenetic analyses possible. Simultaneous and/or sequential demonstration of protein expression and rearrangements of the genome can be achieved using multicolor fluorescence detection systems. The phenotype and the functional status (e.g., proliferation capacity, apoptosis, resistance genes, and overexpression of certain genes) can thus be correlated with the genetic aberrations of these cells (e.g., gained, lost, or amplified genes/chromosomal regions or even specific translocations) and with the cytomorphology. Moreover, cytogenetic aberrations can be studied in immunologically defined cell populations with complex immunophenotype (e.g., disseminated tumor cells, rare cells, mixed/biphasic hematological disorder, and heterogeneous tumors). A method is first described (see Basic Protocol 1) for immunofluorescence labeling of cells. Immunofluorescence labeling can be followed by fluorescence in situ hybridization (FISH) to enable targeted evaluation of FISH signals in a preselected population of cells (see Basic Protocol 2). FISH may also be performed sequentially with multiple fluorochrome-labeled probes (see Alternate Protocols 1 and 2). IMMUNOFLUORESCENCE DETECTION AND CHARACTERIZATION OF RARE CELLS This protocol presents a basic method for immunofluorescence labeling of cells to be evaluated by visual inspection. Captured images and recorded cell coordinates allow exact correlation of cell shape and staining properties with the subsequent information provided by further immunological characterization or FISH analysis.
BASIC PROTOCOL 1
The extent and intensity of particular immunofluorescence reactions can be sufficient such that a simultaneous FISH study can be performed without substantial loss of immunofluorescence labeling. This, however, should be tested for all individual labeling experiments. For simultaneous presentation of immunophenotype and molecular cytogeMolecular Cytogenetics Contributed by Peter F. Ambros and Gábor Méhes Current Protocols in Cytometry (2003) 8.13.1-8.13.11 Copyright © 2003 by John Wiley & Sons, Inc.
8.13.1 Supplement 26
netic aberrations, see Knuutila (1993), Strehl and Ambros (1993), Weber-Matthiesen et al. (1993), Nylund et al. (1994), and Speel et al. (1994). Materials Cytological specimen: cytospin preparation (see, e.g., Support Protocol 1), smear, or touch slide 3.7% (v/v) formaldehyde in PBS (see APPENDIX 2A for PBS) Phosphate-buffered saline (PBS), pH 7.0 (APPENDIX 2A) 2% and 4% (w/v) bovine serum albumin (BSA; Sigma) in PBS (see APPENDIX 2A); filter sterilize before use Primary antibody: e.g., anti-cytokeratin (MNF 116, available from Dako) or anti-GD2 (available from Dr. R. Reisfeld, Scripps Clinic, La Jolla, Calif.) Fluorescently labeled secondary antibody directed against species in which primary antibody was raised: e.g., FITC-conjugated anti-mouse 0.2 µg/ml DAPI (Sigma) in PBS (see APPENDIX 2A) Moist chamber: e.g., rectangular plastic box with lid, containing paper towels moistened with deionized H2O, accommodating up to 10 slides Glass coverslips Antifade mounting medium containing DAPI (e.g., Vectashield, Vector Laboratories) Fluorescence microscope (e.g., Axioplan, Zeiss) Digital image-analysis workstation (optional; see Support Protocol 2) Fix cells 1. Fix the cells by overlaying them with 150 µl ice-cold 3.7% formaldehyde for at least 10 min. For some antigens (e.g., GD2), a fixation overnight at 4°C is recommended. The fixation of several slides in a single Coplin jar may lead to a transposition of positive cells from one slide to another. Therefore individual fixation of slides is recommended.
2. Wash slides twice in PBS, 5 min each time. 3. Block nonspecific labeling by flooding the cell-containing area of the slide with 4% BSA in PBS and incubate 15 min at 37°C in a moist chamber. Stain with antibodies 4. Carefully shake off the BSA solution and pipet the primary antibody (or antibodies), appropriately diluted in 2% BSA/PBS, onto the slide. Cover with a glass coverslip. Incubate at least 30 min at 37°C in a moist chamber. 5. Wash slides twice in PBS, 5 min each time. 6. Pipet secondary antibody, appropriately diluted in 2% BSA/PBS, onto the slide and incubate at least 30 min at 37°C. 7. Wash slides twice in PBS, 5 min each time. 8. Fix again in formaldehyde as described in step 1, for at least 5 min but ideally up to 30 min. 9. Wash slides twice in PBS, 5 min each time, then drain off excess PBS but do not let slides dry out at any point. Immunofluorescence and FISH: For Tumor Cell Detection/ Identification
8.13.2 Supplement 26
Current Protocols in Cytometry
Counterstain with DAPI 10. Stain at least 10 min with 0.2 µg/ml DAPI in PBS. 11. Incubate slides 5 min in 1× PBS. 12. Cover with mounting medium containing DAPI (e.g., Vectashield). Add coverslip, but do not seal. 13. Evaluate fluorescence staining by visual inspection using a fluorescence microscope. Images can also be stored after an automated procedure (see Support Protocol 2 and Background Information) has been performed. The coordinates of the cells of interest are recorded (manually or automatically) and an image is captured to allow an exact correlation of the cell shape and the staining properties with the subsequent information gained after further immunological characterization or FISH analysis.
FISH AFTER IMMUNOFLUORESCENCE STAINING This protocol describes FISH following immunofluorescence labeling to enable targeted evaluation of FISH signals in a preselected population of cells.
BASIC PROTOCOL 2
Materials Immunofluorescently labeled slides (see Basic Protocol 1) Phosphate-buffered saline (PBS), pH 7.0 (APPENDIX 2A) 0.05% pepsin solution (see recipe) 70%, 96%, and 100% ethanol Hybridization mix (see recipe) containing biotin- or digoxigenin-labeled DNA probes (ideally BAC, PAC, YAC, or cosmid clones or repetitive sequences) Rubber cement 2× and 4× SSC (APPENDIX 2A) Post-hybridization wash solution: e.g., 50% (v/v) formamide in 2× SSC, prewarmed to 42°C 2% and 4% (w/v) bovine serum albumin (BSA; Sigma) in PBS (see APPENDIX 2A); filter sterilize before use Primary antibodies: e.g., mouse anti-biotin (Dako), FITC-conjugated sheep anti-digoxigenin (e.g., Boehringer Mannheim) Secondary antibodies: fluorochrome-conjugated antibodies directed against species from which primary antibody was derived (e.g., TRITC-conjugated rabbit anti-mouse, FITC-conjugated rabbit anti-sheep (e.g., Dako) 0.1% (v/v) Tween 20 in 4× SSC (see APPENDIX 2A for SSC) 3.7% (v/v) formaldehyde in PBS (see APPENDIX 2A for PBS) Antifade mounting medium containing DAPI (e.g., Vectashield, Vector Laboratories) Coplin jars Moist chamber: e.g., rectangular plastic box with lid, containing paper towels moistened with deionized H2O, accommodating up to 10 slides Coverslips 1. Put the slides vertically in a Coplin jar containing PBS and let coverslips detach from the slide. Wash once, 5 min in PBS. 2. Put slides into prewarmed (37°C) pepsin working solution. Incubate 30 sec to 1 min if cells were fixed only briefly or 5 to 10 min if cells were fixed overnight. 3. Wash slides twice in PBS, 5 min each time at room temperature. Molecular Cytogenetics
8.13.3 Current Protocols in Cytometry
Supplement 26
4. Dehydrate slides through an ascending ethanol series (70%, 96%, and 100% ethanol), immersing the slides 2 min in each ethanol dilution. Air dry. 5. Pipet 3 to 5 µl hybridization mix onto the slide. 6. Cover with coverslip of appropriate size (not too small, not too large) to allow the hybridization mix to spread over the whole area covered by cells. Avoid air bubbles. 7. Seal the coverslip with rubber cement and allow to dry. 8. For denaturation, put slides on a heating plate prewarmed to 78°C, for 8 min. Denaturation temperature and time may be modified. When only those cells located at the periphery display strong hybridization signals, while cells located in the center of the slide or those arranged in clusters show only weak signals, this is an indication that the denaturation temperature and/or time or the time of the pepsin pretreatment step was not sufficient. One may always increase one parameter per experiment to learn the optimal time or temperature. Accordingly, when cells in the center or in aggregates show strong signals and those at the periphery weak signals, overtreatment is very likely, the cause.
9. Transfer slides quickly to a moist chamber prewarmed to 37°C and hybridize overnight at 37°C. 10. Carefully remove rubber cement and put slides vertically in a Coplin jar filled with 2× SSC. Let coverslips detach from the slides. Detachment of coverslips can take 15 min or longer.
11. Transfer slides to a Coplin jar containing prewarmed post-hybridization wash solution (e.g., 50% formamide in 2× SSC; see UNIT 8.3), and incubate 15 min at 42°C without agitation. 12. Transfer slides to a Coplin jar containing pre-warmed 2× SSC and wash by incubating 7 min without agitation at 42°C. Repeat this washing step. 13. Take slides out of the jar, drain off as much liquid as possible, and transfer them to a moist chamber. Handle slides individually to avoid drying out. 14. To block unspecific binding of the antibody, flood slide with 100 µl of 4% BSA/PBS and incubate 15 min at 37°C. Remove the BSA/PBS solution. 15. Pipet 100 µl of the primary antibodies at the appropriate dilution in 2% BSA/PBS onto the slide. Cover with a coverslip and incubate 30 min at 37°C in a moist chamber. For example, mouse anti-biotin would be diluted 1:20 and FITC-conjugated sheep antidigoxigenin would be diluted 1:100 (also see UNIT 8.3 and manufacturer’s instructions). A FITC-conjugated primary antibody is used in order to obtain a stronger FITC signal.
16. Remove coverslip and wash slides twice in 0.1% Tween 20 in 4× SSC. 17. Drain slides and pipet 100 µl of the secondary (detection) antibodies (e.g., TRITCconjugated rabbit anti-mouse, FITC-conjugated rabbit anti-sheep) diluted in 2% BSA on the slides. Cover with coverslip and incubate 30 min at 37°C. 18. Remove coverslip and wash slides twice in 0.1% Tween 20 in 4× SSC. 19. Fix the cells by incubating the slides 5 min in a Coplin jar containing ice-cold 3.7% formaldehyde. 20. Wash slides in PBS, then dehydrate in ascending ethanol series (70%, 96%, and 100% ethanol) and air dry. Immunofluorescence and FISH: For Tumor Cell Detection/ Identification
21. Cover with antifade medium containing DAPI (e.g., Vectashield) and a coverslip. Seal with rubber cement. 22. Evaluate FISH signals following repositioning of the selected cells (see Support Protocol 2 and Background Information).
8.13.4 Supplement 26
Current Protocols in Cytometry
SEQUENTIAL FISH USING DIRECTLY LABELED DNA PROBES FOLLOWING IMMUNOFLUORESCENT ANALYSIS Home-made or commercially available fluorochrome-labeled probes can be used with this protocol, greatly simplifying the demonstration of specific chromosomal regions.
ALTERNATE PROTOCOL 1
Additional Materials (also see Basic Protocol 2) DNA probes with direct fluorochrome labeling (purchase from Qbiogene, use LSI probes from Vysis, or prepare home-made probes as described in UNIT 8.3) ; store DNA probes at –20°C; store ready-to-use probe mixtures at 4°C 1. Hybridize and wash slides (see Basic Protocol 2, steps 1 to 12) using DNA probes with direct fluorochrome labeling instead of biotin- or digoxigenin-labeled probes. 2. After the last wash, treat slides according to the manufacturer’s recommendations for the labeled probes. Alternatively, rinse slides briefly with deionized water and dry with a stream of air. 3. Mount slides and evaluate FISH signals (see Basic Protocol 2, steps 21 and 22). MULTIPLE SEQUENTIAL FISH FOLLOWING IMMUNOLABELING The demonstration of multiple cytogenetic aberrations within the same cell using only two fluorochromes can be performed by sequential hybridization of different DNA probes. For reliable hybridization results, the pre-existing hybridization signals have to be erased. The same in situ hybridization protocol described in Basic Protocol 2 can be applied with different double-target DNA probes after a formamide washing step.
ALTERNATE PROTOCOL 2
Additional Materials (also see Basic Protocol 2) 65% (v/v) formamide in 2× SSC (see APPENDIX 2A for SSC) 1. Put the slides vertically in a Coplin jar and let the coverslip sink to the bottom of the jar. 2. Transfer the slides to a prewarmed (60°C) Coplin jar containing 65% formamide/2× SSC and wash 10 min. 3. Wash twice in 2× SSC, 5 min each time at room temperature, dehydrate in an ascending ethanol series (70%, 96%, and 100% ethanol), and air dry. 4. Pipet 3 to 5 µl of the next hybridization mixture onto the slide and cover with a coverslip of appropriate size (not too small, not too large) to allow the hybridization mix to spread over the whole area covered by cells. Avoid air bubbles. Seal the coverslip with rubber cement and allow to dry. 5. Proceed with FISH (see Basic Protocol 2, steps 8 to 22). MONONUCLEAR CELL PREPARATION FOR SEQUENTIAL IMMUNOLABELING
SUPPORT PROTOCOL 1
Materials Lymphoprep (Nycomed) Phosphate-buffered saline (PBS), pH 7.0 (APPENDIX 2A) Erythrolysis buffer (see recipe) RPMI 1640 medium containing 10% FBS (APPENDIX 3B) 240-mm2 Hettich Cytospin cytocentrifuge 15-ml conical polypropylene tubes Benchtop centrifuge
Molecular Cytogenetics
8.13.5 Current Protocols in Cytometry
Supplement 26
Glass slides (e.g., HistoBond from Paul Marienfeld, Heidelberg, Germany) Vacuum aspirator Additional reagents and equipment for counting cells using a Coulter counter (APPENDIX 3A) 1. Place 4 ml Lymphoprep in a 15-ml conical polypropylene tube and carefully layer 3 to 5 ml bone marrow or peripheral blood on top of it. Centrifuge 25 min at 1619 × g, room temperature. 2. Collect the interphase layer containing the mononuclear cells, place it in a centrifuge tube, and mix gently with 1× PBS (fill up the tube). Centrifuge 8 min at 259 × g, room temperature. Discard supernatant, fill tube with PBS, and repeat centrifugation. Discard supernatant. 3a. If the cell suspension contains erythrocytes (red color): Gently overlay the cell pellet with 10 ml erythrolysis buffer and resuspend the cells. Wait 10 min for complete erythrolysis, then spin as in step 4 and wash once more with PBS (wash twice). Remove supernatant. 3b. If cell suspension contains no erythrocytes (no red color): Proceed directly to step 4 without erythrolysis. 4. Add 2 ml RPMI containing 10% FBS to the cell pellet, mix, and count the cells in a Coulter counter (APPENDIX 3A). Keep cells on ice to avoid aggregates. 5. For 240-mm2 Hettich Cytospin centrifuge, adjust cell concentration to 1 × 106 cells per ∼300 µl RPMI/10% FBS for each preparation and put 300 µl in each cytospin chamber. The volume should be chosen to give an adequate cell number per slide, i.e., 0.5 to 1 × 106 mononuclear cells.
6. Spin cells onto the slides by centrifuging 6 min at 2500 to 2700 rpm. 7. Using a vacuum aspirator, carefully remove medium from the cytospin chamber. Do not touch the cells on the slide. 8. Remove the cytospin chamber, centrifuge again for 2 min at 3300 rpm to remove remaining liquid, and allow the cells to dry. 9. Store at least several hours at room temperature, then proceed to fixation and immunofluorescence staining (see Basic Protocol 1). SUPPORT PROTOCOL 2
Immunofluorescence and FISH: For Tumor Cell Detection/ Identification
AUTOMATED FLUORESCENCE IMAGE ANALYSIS A schematic drawing of the different steps of FLAME (fluorescence-based automatic microscope) is given in Figure 8.13.1. The different steps include an automatic scanning/detection/quantification, cell selection/storage, and relocation option. After feeding the slides into the stage holder and setting the parameters, the focus plane is determined by the nuclear stain (e.g., DAPI). The cell counting is done in the same fluorescence channel. Cells with appropriate immunofluorescence (e.g., FITC) are selected by the automated image analysis system (e.g., Metafer4/RCDetect, MetaSystems, Altlussheim, Germany). The cells are selected according to predefined criteria (e.g., fluorescence pattern and fluorescence intensity) and parameters from a second (e.g., DAPI) or optionally third (e.g., CY3, TRITC) channel are also taken into account. Cells that fulfill the search criteria (e.g., FITC and DAPI positivity, size and contour of the FITC-fluorescent object, and ratio of FITC and DAPI fluorescence area) are automatically captured by the system and the slide positions are recorded. In case the cell does not fulfill the search criteria, the next field is measured. The parameters for the search criteria have to be set carefully and verified in a number of control experiments to avoid false-positive and
8.13.6 Supplement 26
Current Protocols in Cytometry
focus plane determination, cell counting; DAPI channel image acquisition for FITC fluorescence next field
FITC positive?
cell detection
no
image acquisition for DAPI fluorescence
DAPI positive?
cell selection
no
storage of position and image of the FITC/DAPI positive cell display: total cell count number of FITC/DAPI positive cells gallery of all FITC/DAPI positive cells further analysis: other immunological stains, FISH, TUNEL
relocation of the FITC positive cells visual inspection
Figure 8.13.1 Schematic presentation of the work flow of the automatic microscope. 21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Figure 8.13.2 Gallery of automatically selected breast carcinoma cells showing double-positive immunofluorescence staining with cytokeratin (FITC) and mucin (TRITC), and DAPI-positive nuclei. This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www.interscience.wiley.com/c_p/colorfigures.htm
Molecular Cytogenetics
8.13.7 Current Protocols in Cytometry
Supplement 26
A
B
Figure 8.13.3 (A) Breast carcinoma cell surrounded by hematopoietic cells visualized by cytokeratin-positive staining (FITC). (B) The tumor origin was verified by sequential FISH using a chromosome 17 (TRITC) and an X-specific (FITC) DNA probe. This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www. interscience.wiley.com/c_p/colorfigures.htm
A
Immunofluorescence and FISH: For Tumor Cell Detection/ Identification
B
C
Figure 8.13.4 (A) Leukemia cell: positive staining with CD10 (TRITC) and proliferation-associated marker Ki-67 (FITC). (B)The leukemia specific genetic aberrations: trisomy of chromosome 16 (TRITC) and monosomy X (FITC). (C) TEL/AML fusion (arrow). Aberrations and fusion were demonstrated sequentially on the same cell after automatic relocation of the immunologically positive cell, thus providing the ultimate proof of the leukemic nature of the target cell. This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www.interscience.wiley.com/c_p/colorfigures.htm
8.13.8 Supplement 26
Current Protocols in Cytometry
especially false-negative results. These search criteria are usually specific for a certain tumor type but have to be newly defined for most tumor entities. The fluorochromes can be chosen according to the laboratory preferences, and three-color analyses (e.g., FITC+/TRITC+/DAPI+ fluorescence) are also possible. After these fully automatic steps, the observer can have a quick lookup of the cells on the screen (gallery images of FITC+/TRITC+/DAPI+ -positive cells are presented in Fig. 8.13.2). A close lookup of every positive cell in the microscope is enabled by the precise relocation of the automatic microscope. However, this microscopical analysis following the repositioning of isolated cells frequently does not support tumor-typical cell morphology. Features such as irregular broad cytoplasms, kidney-shaped small nuclei, condensed chromatin, or no prominent nucleoli could refer to tumor cells damaged due, e.g., to the cytotoxic treatment, but also to falsely positive macrophages or other hematopoietic cells. Therefore, such cells are defined as “ambiguous.” To achieve a high degree of reliability, one can either use another antigen or subject the cells to sequential FISH analysis directed toward tumor-typical cytogenetic aberrations found in the primary tumor. To enable the sequential genetic testing by FISH, slides are taken out of the stage and reinserted after successful FISH. The automatic relocation function allows the repositioning of all previously immunologically positive cells. In this way every questionable cell can be re-evaluated and the genetic and biological makeup can be studied in detail (Fig. 8.13.3B, 8.13.4B). REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Erythrolysis buffer 4.15 g NH4Cl 0.5 g KHCO3 100 µl 0.5 M EDTA 500 ml distilled H2O Store up to 1 year at 4°C Hybridization mix For BAC/PAC or cosmid clones: Prepare master mix: 500 µl formamide 100 µl 20× SSC (APPENDIX 2A) 200 µl dextran sulfate 200 µl distilled H2O Mix 1 µg of biotin- or digoxigenin-labeled probe (including Cot-1 DNA) in 20 µl of master mix. For repetitive sequences or commercial probes: Dilute 1 µg of biotin- or digoxigenin-labeled probe in 20 µl Hybrisol VI or VII (Oncor). Store DNA probes at –20°C; store ready-to-use probe mixtures at 4°C.
Pepsin solution, 0.05% Combine the following: 25 µl of 100 mg/ml pepsin stock solution (Sigma; store up to 6 months at −20°C) 49.5 ml H2O 0.5 ml 1 N HCl Prepare fresh Molecular Cytogenetics
8.13.9 Current Protocols in Cytometry
Supplement 26
COMMENTARY Background Information
Immunofluorescence and FISH: For Tumor Cell Detection/ Identification
In situ hybridization (ISH; see other units in Chapter 8), including fluorescence in situ hybridization (FISH), is an extremely powerful technique for the detection of genomic changes in interphase cells (Hopman, 1994). Unfortunately, the demonstration of smaller cell populations in heterogenous samples is limited, as genetic changes cannot be directly related to the cell phenotype unless other methods are used in combination with FISH. Some published techniques use the advantages of both bright-field and fluorescence microscopy for the combined analysis (Knuutila, 1993; Strehl and Ambros, 1993; Weber-Matthiesen et al., 1993; Nylund et al., 1994; Speel et al., 1994). Immunohistochemical detection of antigens has the advantage of permanence (no bleaching), in contrast to fluorescence techniques. However, substrate precipitates from immunoenzymatic procedures may significantly interfere with probe penetration. The advantages of fluorescence-based methods are, on the other hand, manifold. The number of fluorescencelabeled antibodies and FISH probes is constantly growing, and the number of fluorescence colors is constantly increasing as well, allowing the unambiguous and simultaneous detection of a great number of fluorochromes. Already, the sequential use of only two standard fluorochrome-labeled (e.g., FITC-, TRITC-, or Cy3-labeled) antibodies, plus DAPI as a counterstain, makes possible a combined analysis of as many as eight relevant genetic parameters per single cell after dual-color immunophenotyping. This is achieved by the use of six different DNA probes in three independent hybridization experiments (Méhes et al., 2000). Multiple immunofluorescence with FISH analysis is technically simple and widely applicable, provided that the target cells can be repositioned exactly. Repositioning can be done by recording cell coordinates on the slide. For this purpose, in addition to the traditional but laborious-to-use stage vernier scale or England Finder (from Graticules, Ltd., a division of SPI Supplies), computer-based systems are in current use. The requirements for scanning, capturing, and repositioning cells in microscopic preparations are fulfilled by automated image-analysis systems (Méhes et al., 2000, 2001; Ambros et al., 2001, 2003). As few as one single tumor cell in the background of one million normal blood cells can be detected and analyzed for chromosomal aberrations by FISH
using the automated procedure. The selection of cells can be done by automated image analysis according to cellular fluorescence parameters, such as area, size, and intensity of fluorescence, background fluorescence, nuclear size, overlap of nuclear fluorescence with immunological stain, and other parameters.
Critical Parameters and Troubleshooting Peripheral blood or bone marrow smears Smears made from peripheral blood or bone marrow can be used for simultaneous and sequential immunofluorescence/FISH, but frequently the morphological quality is not adequate. Fixation may improve the quality of the slides, as unwanted erythrocytes may dissolve with adequate fixatives. Smears should be prepared on clean and dry glass slides. After drying, the slides should be processed as soon as possible or kept at −20° or −70°C, depending on the antigen. The stability of the antigen must be evaluated individually. Smears heavily contaminated by erythrocytes can be cleared by fixation in 3.7% formaldehyde/PBS. However, after 5 to 10 days without fixation, erythrocytes no longer dissolve in formaldehyde and may produce intense greenish-red autofluorescence. Cytospins and cell adherence One possible problem with this procedure is loss of individual cells during the sequential steps of slide processing. The following precautions may prevent this from happening. (1) Cells should adhere tightly to the glass surface. This adherence can be improved by the use of different slide pretreatments (e.g., silane, poly-L-lysine; see UNIT 8.2). (2) The speed of cytocentrifugation is also critical. In general, cell adherence is improved by the elevation of speed (rpm). However, some cell surface antigens cannot sustain high g forces and will no longer be detectable in preparations spun at a speed higher than 2500 rpm. Highly expressed, robust antigens frequently tolerate 2500 rpm or even more. Cells with less intense immunological staining should be spun gently (2000 rpm or less) in the cytocentrifuge. (3) Longer drying and prolonged fixation time may also result in better adherence, but the exact procedures should be tested for each particular antigen.
8.13.10 Supplement 26
Current Protocols in Cytometry
(4) The sequential staining procedures include several steps that require the changing of coverslips. The removal of coverslips by using shearing forces should be avoided, as this could lead to damage of the cell layer and detachment of cells. This will also make it impossible to reposition individual cells.
Ambros, P.F., Méhes, G., Ambros, I.M., and Ladenstein, R. 2003. Disseminated tumor cells in the bone marrow: Chances and consequences of microscopical detection methods. Cancer Lett. 197:29-34.
Anticipated Results
Knuutila, S. 1993. Simultaneous detection of immunophenotype and genome by the MAC method. J. Histochem. Cytochem. 41:1715-1716.
Fluorochrome-labeled immunopositive cells that are intended for FISH analysis should be clearly distinguishable from other cells. The exact position, and preferably the image, of these cells should be stored for later comparison. Automatic repositioning will enable the simple evaluation of the FISH results cell by cell. The FISH picture can be stored together with preexisting data. Finally, the exact number of analyzed cells and immunopositive cells, and the proportion of cells presenting with specific genetic aberrations should be given.
Time Considerations Immunofluorescence staining can be performed in 2 hr. If the material contains no immunopositive cells, no further analysis is necessary. In case of doubtful immunoreactions, the genetic makeup of the target cell needs to be studied by FISH. Further processing of slides with immunopositive cells, however, requires another 2.5 hr, plus overnight hybridization and the time required for detection and analysis of the FISH signals. Immunostained slides should be evaluated within 1 to 2 days as the immunofluorescence may fade. FISH preparations kept in the dark at 4°C, on the other hand, can be analyzed within the next few weeks or in some cases months.
Hopman, A.H., Voorter, C.E., and Ramaekers, F.C. 1994. Detection of genomic changes in cancer by in situ hybridization. Mol. Biol. Rep. 19:3144.
Méhes, G., Lörch, T., and Ambros, P.F. 2000. Quantitative analysis of disseminated tumor cells in the bone marrow by automated fluorescence image analysis. Cytometry 42:357-362. Méhes, G., Luegmayr, A., Hattinger, C.M., Lörch, T., Ambros, I.M., Gadner, H., and Ambros, P.F. 2001. Combined automatic immunological and molecular cytogenetic analysis allows exact identification and quantification of tumor cells in the bone marrow. Clin. Cancer Res. 7:19691975. Nylund S.J., Wessman, M., and Larramendy, M.L. 1994. Analysis of genotype and phenotype on the same interphase or mitotic cell: A manual of MAC (morphology antibody chromosomes) methodology. Cancer Genet. Cytogenet. 72:115. Speel, E.J.M., Herbergs, J., Ramaekers, F.C.S., and Hopman, A.H.N. 1994. Combined immunocytochemistry and fluorescence in situ hybridization for simultaneous tricolor detection of cell cycle, genomic and phenotypic parameters of tumor cells. J. Histochem. Cytochem. 42:961-966. Strehl, S. and Ambros, P.F. 1993. Fluorescence in situ hybridization combined with immunohistochemistry for highly sensitive detection of chromosome 1 aberrations in neuroblastoma. Cytogenet. Cell Genet. 63:24-28.
Literature Cited
Weber-Matthiesen, K., Deerberg, J., Müller-Hermelink, A., Schlegelberger, B., and Grote, W. 1993. Rapid immunophenotypic characterization of chromosomally aberrant cells by the new FICTION method. Cytogenet. Cell Genet. 63:123125
Ambros, P.F., Méhes, G., Hattinger, C.M., Ambros, I.M., Luegmayr, A., Ladenstein, R., and Gadner, H. 2001. Unequivocal identification of tumor cells in the bone marrow by combining immunological and genetic approaches: Functional and prognostic information. Leukemia 15:275277.
Contributed by Peter F. Ambros Children’s Cancer Research Institute (CCRI) St. Anna Children’s Hospital Vienna, Austria Gábor Méhes University of Pécs Pécs, Hungary
The help of Andrea Luegmayr, Elisabeth Vitasek, and Rita Narath is greatly acknowledged. This work was supported by the Children’s Cancer Research Institute (CCRI).
Molecular Cytogenetics
8.13.11 Current Protocols in Cytometry
Supplement 26
CHAPTER 9 Studies of Cell Function INTRODUCTION
T
his chapter deals with measurements of cellular function; these are generally studies of living cells, in which a dye probe is used as an indicator of cell physiology or behavior (see overview in UNIT 9.1). Studying such characteristics provides a spectrum of information that is of increasing interest to cell biologists. This information is both different from and complementary to that obtainable from analysis of cell surface markers or fixed cells. Prior to the advent of cytometric analyses, functional cellular measurements were performed in bulk assays, yielding a mean value for a mixed population of cells. The cytometer has the unique capacity to measure the physiologic parameters of large numbers of individual living cells (although, when needed, small cell numbers can suffice). The physiologic functions measured can be as basic as cell viability (UNIT 9.2) or as detailed as the function of specific intracellular organelles, including mitochondria (UNITS 9.14 & 9.15). Often, the same methods may be applied equally well to the study of normal and malignant cells, an example being an analysis of phagocytosis that may be adapted to examine ability of cancer cells to degrade and internalize fluorescently labeled extracellular matrix (UNIT 9.13). These methods are uniquely suited to detect heterogeneity within a cell population; combined with analysis of cell surface markers, this provides great power in the elucidation of functional subsets of cells. An exciting example, and one that is of great topical interest, is the identification and sorting of stem cell populations, based on their Hoechst staining characteristics (UNIT 9.18). Other examples of the elucidation of functional heterogeneity are the analysis of apoptosis (UNITS 9.14 & 9.15), of cell activation—as detected by alterations in pH (UNIT 9.3), intracellular calcium (UNIT 9.8), or cytokine production (UNITS 9.9 & 9.21)—and of cell replication by dye dilution (UNIT 9.11). The field of immunology finds many uses for cytometry (see Chapter 6), and it is thus no surprise that there are important applications for the functional assessment of subsets of immune cells. These include the identification and function of dendritic cells (UNIT 9.17), and the study of natural killer (NK) cell target-effector interactions (UNITS 9.10 & 9.16). Likewise, flow cytometry is the method of choice in quantitating bacterial or zymosan particle phagocytosis in immunophenotypically defined cell subsets (UNIT 9.19). Detection of cellular heterogeneity is also a principal feature of the analysis of gene expression using either enzymatic reporters (UNIT 9.5) or green fluorescent protein (UNIT 9.12). Another theme emphasized by many units in this chapter is that cell function need not be a static process; indeed, many assays are performed by perturbing a population and examining the response over time. The perturbation is often a physiologic stimulus—for example, an activating signal—in which case the cytometer can be used both to determine the magnitude of response and to probe the pathways that mediate the response: e.g., pH (UNIT 9.3), membrane potential (UNIT 9.6), oxidative burst (UNIT 9.7), intracellular calcium (UNIT 9.8), or calcium mobilization in platelets (UNIT 9.20). The molecular interactions and communications involved in cell activation have been an exciting area of growth in knowledge (UNIT 9.16), and flow cytometry can be an important tool in elucidating these mechanisms. In a similar fashion, flow cytometry has become increasingly valuable as a method to probe the sequence of physiological events (e.g., changes in mitochondrial function) taking place during apoptosis (UNITS 9.4, 9.14, & 9.15). Sequential observations of cells by either imaging or flow cytometry can be facilitated by hardware and/or software, Contributed by Peter S. Rabinovitch Current Protocols in Cytometry (2003) 9.0.1-9.0.2 Copyright © 2003 by John Wiley & Sons, Inc.
Studies of Cell Function
9.0.1 Supplement 25
so that true kinetic analyses can be performed. When the multiparameter nature of cytometry is harnessed effectively, this can become the only experimental modality that is capable of producing the desired information. The growth of functional cell measurement by cytometry has been greatly facilitated by advances in the chemistry of dye probes. The introduction of a new generation of probes has been responsible in some cases for the development of whole new arenas of cytometric study. Continuing probe development has yielded dyes that are more sensitive, have alternative spectral properties, or both. New excitation or emission characteristics often permit the simultaneous use of pairs of functional probes, or functional probes paired with cell markers. The combination of sensitivity, ability to discriminate cell heterogeneity, and simultaneous acquisition of multiple parameters has given cytometry great impetus; in fields in which analysis of cell function is important, cytometry has emerged as a powerful tool for cell biologists. The editors of this manual believe that this will be evident from the units contained within this chapter, and expect that it will become increasingly apparent as more units are added in successive updates. Peter S. Rabinovitch
Introduction
9.0.2 Supplement 25
Current Protocols in Cytometry
Overview of Functional Cell Assays Flow cytometers can be used to measure a variety of functional parameters that are of increasing interest to cell biologists. Some of these parameters are also of interest in the clinical laboratory, and more will become clinically relevant in the future. The distinguishing feature of these assays is that the functional behavior of the cell itself, rather than a phenotypic marker present on the cell, provides the information of interest. These analyses are thus of great interest to cell physiologists for elucidating mechanisms of cellular response to different signals or environments; they can also be of great utility in establishing variation or differences in cell populations for which there may not be a static marker. The ability to use cytometry to look at functional cellular responses also implies that many of the experimental protocols described in these units are implemented very differently from other cytometric measurements. Especially in the case of measurements of pH, membrane potential, oxidative burst, and intracellular ionized calcium, cells are first examined in an unperturbed state and subsequently followed over a period of time after the introduction of an agonist. Before and during functional assays, therefore, cells must be maintained in a normal physiologic state so as to preserve their native baseline function and response to stimuli. In recent years, the increasing availability of new fluorescent probes has permitted more sensitive and accurate measurements of cellular and subcellular function, usually with some choice of excitation or emission wavelengths. This has also enabled a larger spectrum of multiparameter measurements: correlating two or more functional measurements or correlating functional measurements with immunophenotype or cell cycle. Prior to the advent of flow cytometric analyses, the techniques that were available to measure functional cellular parameters utilized large numbers of cells and resulted in a mean value for a mixed population of cells. The flow cytometer has the unique capacity to measure the physiologic parameters of large numbers of individual living cells; it is therefore especially suited to reveal heterogeneity within the cell population. Combined with identification of immunophenotypic subsets, this provides great power in the elucidation of functional variants. In addition, because the cytometer has the capacity to sort popula-
tions of cells, subpopulations can be used in other subsequent assays, allowing cytometrically derived results to be correlated with many other types of analysis. The reader is also referred to a number of prior reviews on this subject (Rabinovitch et al., 1992; Maftah et al., 1994; Petit et al., 1993).
CELL VIABILITY The discrimination of live from dead cells is perhaps the most widely used physiologic measurement in cytometry and can be performed in more ways than any other functional assay. The nucleic acid stain propidium iodide (PI) is commonly coupled with analysis of immunophenotype; cells stained red by PI do not have an intact plasma membrane, and gating on this indicator of viability loss allows separate immunophenotypic analysis of both live and dead cells. In UNIT 9.2, the basic protocol describes the use of PI, and alternate protocols make use of dye exclusion in fresh or fixed cells or describe identification of viable cells by their functional activity. See Table 9.1.1 for a list of cell viability markers and their excitation/emission wavelengths.
OXIDATIVE METABOLISM All cells possess antioxidant mechanisms designed to protect them from the damaging action of oxygen radicals and other excited species of oxygen. Some cells, however, are capable of producing very large quantities of these dangerous molecules as part of their normal function. Polymorphonuclear leukocytes and macrophages are probably the most active in this regard, because their roles are closely related to protection of the body from foreign microbial attack; the oxygen species are produced to destroy invading organisms. Measurement of the production of these molecules can provide valuable information as to the ability of phagocytic cells to operate. Several methods for measuring superoxide (O2−) and hydrogen peroxide (H2O2) will be described in a future supplement to this chapter. The probes discussed (see Table 9.1.2) work well with phagocytic cells, but their usefulness is not limited to white blood cell suspensions; they can also be used with cell cultures, such as HL60 cells or endothelial cells (Carter and Robinson, 1994). Flow cytometry, fluorescence image analysis, and confocal microscopy are all potential applications of these probes.
Contributed by Peter S. Rabinovitch and J. Paul Robinson Current Protocols in Cytometry (1997) 9.1.1-9.1.6 Copyright © 1997 by John Wiley & Sons, Inc.
UNIT 9.1
Studies of Cell Function
9.1.1
Table 9.1.1
Probes for Cell Viability
Probe Dye-exclusion probes Propidium iodide 7-amino actimomycin D Functional probes Fluorescein diacetate Rhodamine 123 Probes usable with fixed cells Ethidium monoazide LDS-751
Table 9.1.2
Emission wavelength (nm)
488 488
>590 >650
488 488
520-540 520-540
488 488
>630 >650
Probes for Oxidative Metabolism
Probe Dyes for H2O2 Dichlorofluorescin diacetate Dihydrorhodamine 123 Dyes for O2− Hydroethidine
MEMBRANE POTENTIAL
Overview of Functional Cell Assays
Excitation wavelength (nm)
Resting cells maintain large gradients between intracellular and extracellular concentrations of a variety of ions. Potassium ions, for example, are concentrated within cells by action of the Na-K ATPase; the leakage of K+ ions establishes an electron countergradient and the cytoplasm becomes electron negative with respect to the external medium. This K+ electrochemical gradient provides the most significant contribution to the negative membrane potential of most mammalian cells. Maintenance of a large negative transmembrane potential has been postulated to be a control mechanism to arrest cells in an inactive stage, and changes in cell membrane potential that occur rapidly in various cell types after binding of ligands to transmembrane receptors have been suggested to be mediators of subsequent physiologic cellular responses. Detailed investigation of membrane potential in small cells has been made feasible by the development of membrane potential–sensitive indicator probes. These probes are charged lipophilic molecules that partition between the cell and surrounding medium according to the Nernst equation, Cc/Co = e−nεF/RT (where Cc and Co are the cytosolic and extracellular indicator concentrations, n is the charge of the indicator, ε is the membrane
Excitation wavelength (nm)
Emission wavelength (nm)
488 488
525 525
488
590
potential, and F, R, and T are the Faraday constant, gas constant, and temperature, respectively). For a cationic indicator, such as the cyanine dyes (Table 9.1.3), the cellular concentration falls as the membrane potential declines towards zero, and rises if the cell hyperpolarizes (i.e., the cytosol becomes more electronegative with respect to the medium). For the negatively charged oxonol dyes (Table 9.1.3), responses are in the opposite direction. Flow cytometry was first demonstrated to be applicable to analysis of membrane potential by Shapiro et al. (1979), and the techniques used subsequently, which will be included in a forthcoming supplement, remain fundamentally the same. Today, an increasing fraction of membrane potential studies are performed by flow cytometry, taking advantage of the sensitivity of this methodology, the ability to recognize heterogeneity in cellular responses, and opportunities for multiparameter analysis (Shapiro, 1994).
INTRACELLULAR IONIZED CALCIUM Ionized calcium has an important role as a mediator of transmembrane signal transduction, and elevations in intracellular ionized calcium concentration ([Ca2+]i) regulate diverse
9.1.2 Current Protocols in Cytometry
Table 9.1.3
Membrane Potential Indicator Dyes
Probe
Emission Excitation wavelength (nm)a wavelength (nm)a
Cyanine DiOC5(3) DiOC6(3) DiIC1(3) DiIC1(5) DiIC5(3) DiSC3(5)
488 488 488 633-647b 488, 514 568, 633
520-530 520-530 575-585 660-680 540-580 >590, >680
Oxonol DiBAC4(3) DiBAC4(5) DiSBAC2(3) DiTBAC4(3) DiTBAC4(5)
488 568-595 568 488 633-647
520-530 610-640 590-630 575-585 670-680
aCommonly used laser lines and emission detection ranges. bDepending on specific laser source.
cellular processes. Measurement of [Ca2+]i in living cells is thus of considerable interest to a broad range of investigators. Until the development of the first practical indicator, quin2, by Tsien et al. (1982), microelectrodes were required to measure [Ca2+]i and measurements in small intact cells were impossible. Tsien also introduced the strategy of loading small intact cells with the acetoxymethyl (AM) ester of the dye (Tsien, 1981), an approach now widely used for many dyes. The uncharged AM form of the indicator diffuses freely into the cytoplasm, where it is hydrolyzed by esterases to yield a hydrophilic form which is trapped inside the cell. However, quin2 has several disadvantages that limit its application in flow cytometry. In 1985, Grynkiewicz et al. described a new family of highly fluorescent calcium chelators, inaugurating the present popularity of cytometric measurement of [Ca2+]i . One of these dyes, indo-1 ([1-[2-amino-5-[6-carboxylindol-2-yl]-phenoxy]-2-[2′-amino-5′-methyl phenoxy]ethane N,N,N′,N′-tetraacetic acid]), exhibits large changes in fluorescent emission wavelength upon calcium binding; analysis of the ratio of fluorescence intensities at two wavelengths (∼400 and 500 nm) allows calculation of [Ca2+]i independent of variability in cellular size or intracellular dye concentration. The ratiometric approach thus provides this analysis with considerable accuracy. The only significant drawback to the use of indo-1 is the requirement for ultraviolet (UV) excitation. The introduction of a fluorescein-based cal-
cium-sensitive probe, fluo-3, provided a visibly excited alternative to indo-1 (Minta et al., 1989). This dye exhibits an increase in fluorescence intensity with increasing [Ca2+]i but does not have the advantages of permitting ratiometric determinations; the fluorescence intensity distributions are wider, and calibration is more complicated because the signal is proportional to cell size and dye concentration as well as to [Ca2+]i. Strategies to minimize this problem have been developed, based on the simultaneous use of a second dye to serve as an indicator of the magnitude of dye loading in an individual cell (Rijkers et al, 1990), or the simultaneous use of Fura Red, a dye that exhibits a decrease in fluorescence intensity with increasing [Ca2+]i. The latter combination results in a close simulation of a ratiometric analysis (Novak and Rabinovitch, 1994). Calcium-sensitive dyes useful for flow cytometry are summarized in Table 9.1.4. Protocols for use of indo-1, fluo-3, and fluo-3 combined with Fura Red will be included in a forthcoming supplement. Measurements of [Ca2+]i have been widely performed as multiparameter analyses, with additional fluorochromes used for the determination of cellular immunophenotype. This allows alterations in [Ca2+]i to be examined in specific immunophenotypic subsets (Rabinovitch et al., 1986). Fluorescein isothiocyanate (FITC) and phycoerythrin (PE)-conjugated antibodies can be used with indo-1, and PE can be used with fluo-3 and the fluo-3/Fura Red combination. Numerous examples of the
Studies of Cell Function
9.1.3 Current Protocols in Cytometry
Supplement 7
Table 9.1.4
Calcium Indicator Dyes Useful for Flow Cytometric Applications
Indicator
Indo-1 Fluo-3 Calcium Green-1 Calcium Orange Calcium Crimson Fura Red Fluo-3/Fura Red
Emission response to elevated calcium
Excitation wavelength (nm)
Emission wavelength (nm)
Ratio Increase Increase Increase Increase Decrease Ratio
325-360 488 488 550 590 488 488
390/520 530 530 575 610 660 530/660
Calcium affinity, Kd (nM)a 22°C
37°C
NA
∼250 ∼860
∼400 ∼250 ∼330 ∼200 ∼400 ∼400
NA NA NA NA NA
aNA, not available.
analysis of [Ca2+]i in immunophenotypically defined subsets have been described (Rabinovitch and June, 1990). This approach has been used extensively in the demonstration of differences between [Ca2+]i activation requirements of different cell subsets, and subset specificities of activation pathways. The combination of sensitivity, reliability, and ability to analyze large numbers of cells within cell subsets has made flow cytometric assays of [Ca2+]i the preferred technique for a broad spectrum of research applications and has spurred interest in developing clinical applications of [Ca2+]i analysis.
INTRACELLULAR pH
Overview of Functional Cell Assays
The intracellular pH (pHi) of mammalian cells is closely regulated to ∼7.2 by at least three mechanisms: the Na+/H+ antiporter and sodium-dependent Cl−o/HCO3−i exchanger accomplish acid extrusion, while the HCO3−o/Cl−i exchanger has the major role for base extrusion. All of these mechanisms appear to be stimulated by a variety of growth factors and by phorbol esters, presumably through the activation of protein kinase C. The first commonly used pHi probes were modifications of fluorescein. Fluorescein diacetate was the first generation pH probe, followed by carboxyfluorescein diacetate (CFDA) and 2′,7′-bis-carboxyethyl-5(6)-carboxyfluorescein (BCECF). There is a pH-dependent shift in the excitation wavelength of BCECF. However, for excitation near 500 nm, emission is also pH dependent, allowing ratiometric fluorescence emission analysis (Table 9.1.5). In both cases, though, the magnitude of pH-dependent ratio shifts is relatively modest, and these dyes have therefore largely been supplanted by newer probes. The most useful
of several UV-excited pH probes that have been developed is 1,4-diacetoxy-2,3-dicyanobenzene (ADB). The cell-permeant ADB is hydrolyzed and trapped intracellularly to yield 2,3dicyanohydroquinone (DCH; Valet et al., 1981). Ratiometric determination of pH using a single excitation source can be achieved by measuring fluorescence emission at blue and green wavelengths (Table 9.1.5). As with measurement of [Ca2+]i, many laboratories find the requirement for UV excitation to be the primary limitation. A later development was SNARF-1 (SemiNaphthoRhodaFluor), the most widely used probe for pHi measurement, which was first introduced by Whitaker et al. (1988). SNARF-1 has convenient excitation spectra (488 or 514 nm) and exhibits large changes in pH-dependent fluorescence. The ratio of orange and red emissions is used as the pHi indicator, with the advantages of ratiometric determination, as described previously. The use of SNARF-1 and BCECF is detailed in UNIT 9.3. Because the fluorescence emission properties of SNARF-1 allow simultaneous excitation of FITC probes with the same 488-nm laser, analysis of pHi in immunophenotypically defined cell subsets using FITC-labeled monoclonal antibodies is very straightforward. Because of the large shifts in pHi that take place upon activation of granulocytes, studies of intracellular pH in analysis of granulocyte function are probably among the most interesting applications of this methodology (Rothe et al., 1990; Robinson et al., 1994; Carulli et al., 1995).
INTRACELLULAR ORGANELLES Evaluating cellular function not only involves studies of cell products, responses, or functional mechanisms, but can also include
9.1.4 Supplement 7
Current Protocols in Cytometry
Table 9.1.5
Fluorochromes for Ratiometric Determination of pH by Flow Cytometry
Probe
pKa
Fluorescence (nm) Excitation
Emission 525
Fluorescein diacetate (FDA)
6.3
Ratio 436/495
Carboxyfluorescein diacetate (CFDA) Bis-carboxyethylcarboxyfluorescein acetoxymethyl ester (BCECF AM)
6.4 6.98
Ratio 441/488 535 Ratio 439/490 535 488 Ratio 520/620
Hydroxycoumarin (4-methylumbelliferone) (4-MU)
7.8
∼350 ∼350
Ratio 430/470 Ratio 450/560
Diacetoxydicyanobenzene (ADB) (yields dicyanhydroquinone [DCH] after de-esterification)
8.0
∼350
Ratio 425/540
Carboxy SNARF-1 acetoxymethyl ester
7.50
514 or 530
Ratio 575/670
identification of specific organelles. Depending upon the organelle of interest, various techniques are available to study function. In UNIT 9.4, a number of methods for both flow and image evaluation of organelle function are provided. Techniques for evaluating mitochondria rely on fluorescent dyes such as rhodamine 123, dihydrorhodamine 123, MitoTracker Green FM, MitoFluor, nonyl acridine orange, and MitoTracker Red. Mitochondria maintain an inside negative membrane potential, allowing lipophilic fluorescent cations to accumulate inside the mitochondria. Thus, dyes such as rhodamine 123 provide an excellent assessment of the physiological state of the mitochondrion because labeling decreases with loss of membrane potential. In addition to demonstrating detailed methods for evaluating mitochondrial function using flow cytometry, this unit shows how to use microscopy on cells attached to slides or culture dishes. Studies of lysosomes are made possible by probes such as fluorescein di-β-D-galactopyranoside (FDG), which becomes fluorescent upon enzymatic cleavage of the sugars. Alternatively, lysosomes can be labeled with LysoTracker dyes. These are plasma membrane– permeant bases that accumulate in lysosomes because of the acidic lysosomal pH, which is required for function of lysosomal enzymes. Although assays of lysosomal activity provide rapid data, they are more qualitative than quantitative because of the membrane barriers that the substrates must penetrate to become active. Two other organelles discussed in UNIT 9.4 are the Golgi apparatus and the endoplasmic reticulum. The Golgi apparatus consists of complex stacks of membranes and associated vesicles that comprise the factory for production of
complex carbohydrates, modification of glycoproteins, and transport of glycoproteins to the cell surface. The lipid composing the outer Golgi membrane is targeted by the fluorescent Golgi probes. Typical probes are NBD C6 ceramide (NBD) or the BODIPY ceramides. One unique feature of the latter is that their fluorescence emission changes from green to red after high-level dye accumulation within the Golgi, because of excimer formation of dye molecules stacked onto one another. The endoplasmic reticulum (ER), which acts as a scaffold for protein and lipid synthesis, must be stained in order to be viewed by light microscopy. The large amount of lipid present prevents many dyes from accumulating within the ER. Carbocyanine probes such as DiOC6(3) are not ideal for ER staining because such dyes also accumulate in other membraneous organelles such as mitochondria. However, these probes are suitable for most aqueous cell culture systems and thus can provide meaningful information; this is particularly true when using image analysis as opposed to flow cytometry, because visual discrimination of structure can be made by fluorescent or confocal microscopy.
GENE REPORTER ASSAYS Whereas intrinsic cellular characteristics are described in most of the functional assays in this chapter, in UNIT 9.5, assays are presented that may be used to detect cells with new functions introduced by gene transfer. Flow cytometry has unique advantages for the latter application: (1) cytometry can be used to quantitate the proportion of cells that express a gene reporter, the level of expression of the reporter, and any heterogeneity in level of expression; (2) cytometry can viably sort and select cells on the
Studies of Cell Function
9.1.5 Current Protocols in Cytometry
Supplement 21
basis of levels of gene reporter expression; (3) the multiparameter nature of cytometry allows gene expression to be correlated with other phenotypic or functional cell assays. This combination of features often makes cytometric assay of gene reporters the method of choice for maximizing information return while minimizing the length of time required to assay the result of gene transfer. There are three general types of gene reporters and associated strategies for their cytometric detection: a reporter enzyme that converts a substrate to a fluorescent product, an intrinsically fluorescent gene reporter, and a reporter that codes for a tag, usually a novel protein expressed on the cell surface. Protocols for analysis of gene reporter surface tags are the same as those for analysis of other cell surface markers using immunofluorescent antibodies (see UNIT 6.2). Two enzyme-based reporters are presented in UNIT 9.5: fluorescein di-β-D-galactopyranoside (FDG) to detect genetically introduced β-galactosidase enzymatic activity and fluorescein di-β-D-glucuronide (FDGlcU) to detect genetically introduced β-glucuronidase. While the former “FACS-Gal” assay has been more widely used, FDGlcU can be applied to cell types for which FACS-Gal has not been particularly successful. Both enzymatic systems offer the advantage of highly sensitive detection, because only a few molecules of enzyme are necessary to convert substrate to detectable amounts of fluorescent product.
LITERATURE CITED
Overview of Functional Cell Assays
Carter, W.O. and Robinson, J.P. 1994. Intracellular hydrogen peroxide and superoxide anion detection in endothelial cells. J. Leukocyte Biol. 55:253-258. Carulli, G., Minnucci, S., Vanacore, R., and Ambrogi, F. 1995. The role of flow cytometry in the study of physiopathology of neutrophilic granulocytes. Recent Prog. Med. 86:208-216. Grynkiewicz, G., Poenie, M., and Tsien, R.Y. 1985. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J. Biol. Chem. 260:3440-3450. Maftah, A., Huet, O., Gallet, P.F., and Ratinaud, M.H. 1993. Flow cytometry’s contribution to the measurement of cell functions. Biol. Cell 78:8593. Minta, A., Kao, J.P.Y., and Tsien, R.Y. 1989. Fluorescent indicators for cytosolic calcium based on rhodamine and fluorescein chromophores. J. Biol. Chem. 264:8171-8178. Novak, E.J. and Rabinovitch, P.S. 1994. Improved sensitivity in intracellular ionized calcium measurement using fluo-3/Fura Red fluorescence ratios. Cytometry 17:135-141.
Petit, J.M., Denis-Gay, M., and Ratinaud, M.H. 1993. Assessment of fluorochromes for cellular structure and function studies by flow cytometry. Biol. Cell 78:1-13. Rabinovitch, P.S. and June, C.H. 1990. Measurement of intracellular free calcium and membrane potential. In Flow Cytometry and Cell Sorting, 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 651-668. Wiley-Liss, New York. Rabinovitch, P.S., June, C.H., Grossmann, A., and Ledbetter, J.A. 1986. Heterogeneity among T cells in intracellular free calcium responses after mitogen stimulation with PHA or anti-CD3. Simultaneous use of indo-1 and immunofluorescence with flow cytometry. J. Immunol. 137:952-961. Rabinovitch, P.S., June, C.H., and Kavanagh, T.J. 1992. Measurements of cell physiology: Ionized calcium, pH and glutathione. In Clinical Flow Cytometry: Principles and Applications (K.D. Bauer, R.E. Duque, and T.V. Shankey, eds.) pp. 505-534. Williams & Wilkins, Baltimore. Rijkers, G.T., Justement, L.B., Griffioen, A.W., and Cambier, J.C. 1990. Improved method for measuring intracellular Ca++ with fluo-3. Cytometry 11:923-927. Robinson, J.P., Carter, W.O., and Narayanan, P.K. 1994. Oxidative product formation analysis by flow cytometry. Methods Cell Biol. 41:437-447. Rothe, G., Kellermann, W., and Valet, G. 1990. Flow cytometric parameters of neutrophil function as early indicators of sepsis- or trauma-related pulmonary or cardiovascular organ failure. J. Lab. Clin. Med. 115:52-61. Shapiro, H.M. 1994. Cell membrane potential analysis. Methods Cell Biol. 41:121-133. Shapiro, H.M., Natale, P.J., and Kamentsky, L.A. 1979. Estimation of membrane potentials of individual lymphocytes by flow cytometry. Proc. Natl. Acad. Sci. U.S A. 76:5728-5730. Tsien, R.Y. 1981. A non-disruptive technique for loading calcium buffers and indicators into cells. Nature 290:527-528. Tsien, R.Y., Pozzan, T., and Rink, T.J. 1982. Calcium homeostasis in intact lymphocytes: Cytoplasmic free calcium monitored with a new, intracellularly trapped fluorescent indicator. J. Cell Biol. 94:325-334. Valet, G., Raffael, A., Moroder, L., Wünsch, E., and Ruhenstroth-Bauer, G. 1981. Fast intracellular pH determination in single cells by flow-cytometry. Naturwissenschaften 68:265-266. Whitaker, J.E., Haugland, R.P., and Prendergast, F.G. 1988. Seminaphtho-fluoresceins and-rhodafluors: Dual fluorescence pH indicators. Biophys. J. 53:197a.
Contributed by Peter S. Rabinovitch University of Washington Seattle, Washington J. Paul Robinson Purdue University West Lafayette, Indiana
9.1.6 Supplement 21
Current Protocols in Cytometry
Assessment of Cell Viability
UNIT 9.2
The method used to determine cell viability (and to a degree, the definition of viability) is often related to the phenomenon studied. Frequently, cell viability is thought of in somewhat negative terms. That is, one needs to exclude dead cells because they generate artifacts as a result of nonspecific binding and/or uptake of fluorescent probes. In addition to simple enumeration of live or dead cells present, there is a broad range of biologically relevant cytometric procedures that are related to the physiological state of the cells measured. For example, one may wish to measure cell morbidity in apoptosis, cell survival as a result of cytotoxicity, the potential of bacteria or microalgae to survive environmental insult and grow once normal conditions have been restored, or the potential of intracellular protozoa to undergo division cycles as part of the infective process. Cell viability may be judged by morphological changes or by changes in membrane permeability and/or physiological state inferred from the exclusion of certain dyes or the uptake and retention of others. Here methods are presented for staining nonviable cells by dye exclusion (indicative of an intact membrane) using the fluorescent, DNA-binding probes propidium iodide (PI) (see Basic Protocol) and 7-amino actinomycin D (7-AAD; see Alternate Protocol 1). These probes may also be used in cells labeled with phycoerythrin (PE)-conjugated antibodies (see Alternate Protocol 2). The next two protocols present different aspects of physiological state that can be used to assess viability; one is based on esterase activity (see Alternate Protocol 3) and the other on mitochondrial membrane potential (see Alternate Protocol 4). For fixed cells, the state of viability prior to fixation can be determined using DNA-binding probes either before (see Alternate Protocol 5) or after (see Alternate Protocol 6) fixation. Each of the above protocols requires basic understanding in cell handling and flow cytometry for which they are designed. In contrast, the final method presented (see Alternate Protocol 7) is a dye exclusion procedure for microscopy using trypan blue and a hemacytometer (APPENDIX 3A & 3B). ASSESSMENT OF CELL VIABILITY USING PROBES FOR MEMBRANE INTEGRITY Live cells with intact membranes are distinguished by their ability to exclude dyes that easily penetrate dead or damaged cells. Staining of nonviable cells with propidium iodide (PI) has been performed on most cell types. Its broad application is most likely due to ease of use: the procedure is very simple, and the stained cells are bright red and easy to identify. Alternatively, the longer-wavelength emission of 7-amino actinomycin D (7AAD) makes it easier to use simultaneously with phycoerythrin (PE) as a surface marker, as detailed in Alternate Protocol 2, or with fluorescein isothiocyanate (FITC). Although 7-AAD is not as bright as PI, permeable cells are easily distinguishable from live cells. Propidium Iodide Staining of Nonviable Cells
BASIC PROTOCOL
Materials 2 mg/ml propidium iodide (PI) in PBS (store wrapped in foil ≤1 month at 4°C) Cell suspension PBS (APPENDIX 2A) 13 × 100–mm polystyrene culture tubes CAUTION: Propidium iodide is a suspected carcinogen and should be handled with care. In particular, be careful of particulate dust when weighing out the dye. Use gloves when handling it.
Contributed by David M. Coder Current Protocols in Cytometry (1997) 9.2.1-9.2.14 Copyright © 1997 by John Wiley & Sons, Inc.
Studies of Cell Function
9.2.1 Supplement 15
Number of cells
2000
1500 live 1000
dead
1 2
500
0 1
10
100
1000
10,000
PI fluorescence
Figure 9.2.1 Identification of nonviable cells with propidium iodide (PI). Nonviable cells are more than two decades brighter than the unstained, viable cells. Gating on a one-parameter histogram is sufficient to identify the viable population. Region 1: viable cells; region 2: nonviable cells.
1. Add 1 µl of 2 mg/ml propidium iodide (2 µg/ml final) to approximately 106 washed cells suspended in 1 ml PBS in 13 × 100–mm polystyrene culture tubes. To use this procedure with cells that are also labeled with PE-conjugated antibodies, see Alternate Protocol 2.
2. Incubate ≥5 min in the dark on ice. 3. Analyze on flow cytometer with excitation at 488 nm and emission collected at >550 nm. PI is easily excited at 488 nm. The dye has a broad fluorescence emission and can be detected with photomultiplier tubes (PMTs) normally used for phycoerythrin (~585 nm) or at longer wavelengths (≥650 nm). Amplify the PMT signal logarithmically to distinguish populations of permeable (and presumed dead) cells from viable cells. Adjust the PMT high voltage such that bright cells are well separated from dim, viable cells. It is often easy to identify nonviable cells on a bivariate plot of forward light scatter (see Fig. 9.2.1). Dead cells can be live-gated, but unless one is absolutely sure of the viable population, it is always much better to collect ungated listmode data and perform gating after the raw data files are collected. Populations that may not be obvious on the flow cytometer display will be seen during subsequent data analysis. Moreover, any gating can be done and redone without losing cells. ALTERNATE PROTOCOL 1
7-AAD Staining of Nonviable Cells Additional Materials (also see Basic Protocol) 1 mg/ml 7-amino actinomycin D (7-AAD; see recipe) 1. Add 1 µl of 1 mg/ml 7-AAD (1 µg/ml final) to approximately 106 washed cells suspended in 1 ml PBS in 13 × 100–mm polystyrene culture tubes.
Assessment of Cell Viability
To use this procedure with cells that are also labeled with PE-conjugated antibodies, see Alternate Protocol 2.
9.2.2 Supplement 15
Current Protocols in Cytometry
2. Incubate ≥30 min in the dark on ice. 3. Analyze on flow cytometer with excitation at 488 nm. Collect fluorescence emission with a 650-nm long-pass or a 670 ± 20–nm band-pass filter. 7-AAD is easily excited at 488 nm. The fluorescence emission of the dye has a peak at ∼670 nm.
A 180 160
Number of cells
140 120
live
100
1
80 60
2
40
dead
20 0 10
1
100 7-AAD fluorescence
10,000
C
B 10,000
1
2
1
live PE fluorescence
1000
2 live
1000
100
10 3
4
1 1
10
100
1000 10,000
7-AAD fluorescence
4
3 1
10
100
1000 10,000
7-AAD fluorescence
Figure 9.2.2 Effects of gating and compensation with 7-AAD. (A) Gating discriminates live cells. One-parameter histogram of logarithmically amplified 7-AAD fluorescence using a 650-nm longpass filter. Mouse spleen cells are labeled only with 7-AAD. Note the peak of dead cell population in region 2 at a relative brightness between 100 and 200 (about ten-fold dimmer than what is expected for propidium iodide). The live cell population that occupies the first decade in the histogram (region 1) is 7-AAD negative and constitutes the majority of the cells in the population. (B) Uncompensated phycoerythrin fluorescence in the presence of 7-AAD. A bivariate plot of mouse spleen cells labeled with 7-AAD and a PE-labeled antibody to a cell surface antigen. Note the two small populations of dead cells in regions 2 and 4 and the large population of live cells that occupies region 2. If gating was performed before adequate compensation was achieved, then most of the viable PE-positive cells could be lost. (C) Compensation of PE with 7-AAD. Distribution of cells from same sample as in B. Note the dead cells are in the same location as in B, but the live cells are now clearly resolved from 7-AAD positive populations. At this point, the live cell gate defined in region 2 of A is valid.
Studies of Cell Function
9.2.3 Current Protocols in Cytometry
Use logarithmic amplification to distinguish permeable and bright cells from nonpermeable cells. Adjust the PMT high voltage to resolve a population of viable cells in the first decade of the 7-AAD fluorescence histogram as shown in region 1 of Figure 9.2.2A; nonviable cells are in region 2 of Figure 9.2.2A. ALTERNATE PROTOCOL 2
Use of PI or 7-AAD for Cells Labeled with PE-Conjugated Antibodies Additional Materials (also see Basic Protocol) PE-labeled cell suspension (UNIT 6.2) 1. Stain and incubate PE-labeled cells with PI (see Basic Protocol, steps 1 and 2) or 7-AAD (see Alternate Protocol 1, steps 1 and 2). 2. Analyze on flow cytometer with excitation at 488 nm. Use a 585 ± 20–nm band-pass filter for detection of PE fluorescence and a 650-nm long-pass filter for PI or 7-AAD. PI, 7-AAD, and PE are all excited by 488-nm light. Despite the separation by two detection filters, there is substantial overlap between the PE and PI or 7-AAD, requiring compensation between the two detectors. This problem is illustrated for PE and 7-AAD in Figure 9.2.2 (parts B and C), though the same procedure is used with PI. Note the typical discrimination of viable/nonviable cells. If one were to set a gate region on the viable cells in region 1 and use that for analysis in the absence of proper compensation, most PE-positive cells would be eliminated as nonviable (see Fig. 9.2.2B). Using a bivariate histogram of log 7-AAD versus log PE fluorescence, adjust the PE-7-AAD compensation such that PE-positive, 7-AAD-negative cells are above the PE-negative, 7-AAD-negative population (see Fig. 9.2.2C). Once proper compensation of the long-wavelength component of PE is subtracted from the output of the detector for 7-AAD, the problem disappears.
ASSESSMENT OF CELL VIABILITY USING PROBES OF PHYSIOLOGICAL STATE These protocols describe the use of probes that require a specific cellular function in addition to an intact membrane. Alternate Protocol 3 describes the use of fluorescein diacetate (FDA), which requires cellular esterase activity, and Alternate Protocol 4 describes the use of rhodamine 123 as a probe for mitochondrial membrane potential. ALTERNATE PROTOCOL 3
Fluorescein Diacetate Staining of Viable Cells Cell viability can be assessed directly through the presence of cytoplasmic esterases that cleave moieties from a lipid-soluble nonfluorescent probe to yield a fluorescent product. The product is charged and thus is retained within the cell if membrane function is intact. Hence, viable cells are bright and nonviable cells are dim or nonfluorescent. Typical probes include fluorescein diacetate (FDA, described here), carboxyfluorescein, and calcein. Variations in uptake or retention of the dye among individual cells or under different conditions affect the efficacy of particular probes. Additional Materials (also see Basic Protocol) 1 mg/ml fluorescein diacetate (FDA; prepare fresh in acetone in a 13-mm glass culture tube and cover with foil) Cell suspension in culture medium appropriate for the cell type 1. Add 2 µl of 1 mg/ml FDA (2 µg/ml final) to approximately 106 cells in 1 ml medium in a 13 × 100–mm polystyrene culture tube. 2. Vortex to mix and incubate 15 min at 37°C.
Assessment of Cell Viability
3. Analyze on flow cytometer immediately with excitation at 488 nm. Collect fluorescence using a 530 ± 20–nm band-pass filter.
9.2.4 Current Protocols in Cytometry
FDA is excited by 488-nm light and fluoresces green. Filters used for measuring fluorescein (e.g., 530 ± 20–nm band-pass) are sufficient to visualize the nonfluorescent cells on the same scale. Use logarithmic amplification of the PMT output. Cells that take up and retain free fluorescein are very bright (approximately two decades brighter on a logarithmic scale) and should be easily distinguishable from nonfluorescent, nonviable cells. Unless one is absolutely sure about the location of viable cells in the histograms, it is preferable to collect listmode data files of all samples and perform gating after the raw data files are collected, to avoid the danger of inadvertent loss of viable cells.
Rhodamine 123 Staining of Viable Cells Another property of viable cells is the maintenance of electrochemical gradients across the plasma membrane. Functional subsets of this general phenomenon include the maintenance of pH and other ion gradients as well as the capacity for energy-yielding metabolism in mitochondria. These physiological processes can be exploited to distinguish viable from nonviable cells. One of the most commonly used probes for identifying viable cells is rhodamine 123, a cationic lipophilic dye that partitions into the low electrochemical potential of mitochondrial membranes. Active mitochondria in viable cells are stained bright green; loss of gradients within nonviable cells results in loss of the dye.
ALTERNATE PROTOCOL 4
Additional Materials (also see Basic Protocol) 1 mg/ml rhodamine 123 (prepare fresh in distilled water) Cell suspension in culture medium appropriate for the cell type 1. Add 5 µl of 1 mg/ml rhodamine 123 (5 µg/ml final) to approximately 106 cells in 1 ml medium in a 13 × 100–mm polystyrene culture tube. 2. Vortex to mix and incubate 5 min at 37°C; return to room temperature. 3. Analyze immediately with excitation at 488 nm. Collect fluorescence using a 530 ± 20–nm bandpass filter. Rhodamine 123 absorbs 488-nm light and fluoresces green. Collect fluorescence through a filter that transmits at ~530 nm, as for FITC. PMT output should be amplified logarithmically. Viable cells are brighter than nonviable cells, though with some cell types there may be some overlap (see Fig. 9.2.3A). A bivariate plot of forward scatter versus rhodamine 123 fluorescence can be useful to distinguish viable from nonviable cells (see Fig. 9.2.3B). Unless one is absolutely sure about the location of viable cells in the histograms, it is preferable to collect listmode data files of all samples and perform gating after the raw data are collected, to avoid the danger of inadvertent loss of viable cells.
ASSESSMENT OF CELL VIABILITY IN FIXED CELLS For reasons of safety or convenience, it is frequently necessary to fix cells prior to analysis. Data analysis is less ambiguous if nonviable or damaged cells can be eliminated, but the methods discussed above will not work, because fixation will render all cells permeable. There are, however, DNA probes that penetrate and stain dead or damaged cells and that can withstand fixation. Ethidium monoazide (EMA) is positively charged and penetrates the membranes of dead or damaged cells but not live ones. EMA can be photochemically cross-linked with short exposure to visible light; after the excess dye is washed away, the cells are fixed. Another DNA-binding fluorochrome used for staining nonviable cells is laser dye styryl-751 (LDS-751). The procedure is somewhat more simple, as staining is done after fixation and does not require cross-linking. Studies of Cell Function
9.2.5 Current Protocols in Cytometry
ALTERNATE PROTOCOL 5
Ethidium Monoazide Staining of Nonviable Cells Prior to Fixation Additional Materials (also see Basic Protocol) 50 µg/ml ethidium monoazide (EMA; see recipe) 1% (w/v) paraformaldehyde in PBS (see APPENDIX 2A for PBS recipe; store mixture ≤1 week at 4°C and discard if precipitate forms) 40-W fluorescent light 1. Add 10 µl of 50 µg/ml EMA (~5 µg/ml final) to approximately 106 washed cells suspended in 100 µl PBS in a 13 × 100–mm polystyrene culture tube. Preparations of EMA dye vary. The exact concentration needed may range from 1 to 5 ìg/ml.
A 3000
Number of cells
2500 2000 1500 ungated
live gate
1000 500 0 10
1
100
1000
10,000
Rhodamine 123 fluorescence
Rhodamine 123 fluorescence
B
10,000
1000 live 100
10 dead 1 0
50
100
150
200
250
Forward light scatter
Assessment of Cell Viability
Figure 9.2.3 Effects of gating with rhodamine 123. (A) Identification of live cells after gating. Rhodamine 123 may not always completely resolve viable from nonviable cells as indicated in the ungated histogram (dotted line). Gating on forward light scatter versus rhodamine 123 fluorescence helps separate both populations. Note the histogram of the gated population of viable cells (solid line) overlaid on the ungated population. (B) A bivariate plot of forward light scatter versus rhodamine 123 fluorescence helps to resolve live (rhodamine 123–bright) and dead (rhodamine 123–dim) populations. Debris is gated out at the same time.
9.2.6 Current Protocols in Cytometry
2. Place tubes on ice ~18 cm beneath a 40-W fluorescent light for 10 min. 3. Wash and fix cells by adding 1 ml µl of 1% paraformaldehyde to the cell pellet. Incubate 1 hr at room temperature. 4. Analyze on flow cytometer with excitation at 488 nm. Collect fluorescence emission using a ≥630-nm long-pass filter; amplify PMT output logarithmically. EMA excites with 488-nm light and fluoresces well into the red region of the spectrum. EMA does not fluoresce as brightly as propidium iodide, so discrimination of nonviable, EMA-bright cells from viable cells may be less obvious. A bivariate plot of forward light scatter versus EMA fluorescence may aid in distinguishing viable (EMA-negative) cells. Dead cells can be live-gated, but unless one is absolutely sure of the viable population, it is always better to collect ungated listmode data and perform gating after the raw data files are collected. Populations that may not be obvious on the flow cytometer display will be seen during subsequent data analysis. Moreover, any gating can be done and redone without losing cells.
LDS-751 Staining of Previously Nonviable Cells After Fixation
ALTERNATE PROTOCOL 6
Additional Materials (also see Basic Protocol) 1% (w/v) paraformaldehyde in PBS (see APPENDIX 2A for PBS recipe; store mixture ≤1 week at 4°C and discard if precipitate forms) 2 µg/ml LDS-751 (laser dye styryl-751) working solution (see recipe) 1. Wash approximately 106 cells in PBS in a 13 × 100–mm polystyrene culture tube. 2. Fix cells by adding 1 ml µl of 1% paraformaldehyde to the cell pellet. Incubate 1 hr at room temperature. 3. Add 10 µl of 2 µg/ml LDS-751 working solution to 1 ml fixed cells at a concentration of approximately 106 cells per ml. 4. Incubate overnight at room temperature in the dark. 5. Analyze on flow cytometer with excitation at 488 nm. Collect fluorescence emission using a 650-nm long-pass filter. LDS-751 is excited with 488-nm light and emits in the red portion of the spectrum; the 650-nm long-pass filter is adequate to separate red fluorescence from other fluorochromes and scatter laser light. Amplify the PMT output logarithmically to distinguish bright, nonviable cells from dim, viable cells and from nonfluorescent red cells or debris. A bivariate plot of forward scatter versus log LDS-751 fluorescence can help identify populations. Dead cells can be live-gated, but unless one is absolutely sure of the viable population, it is always much better to collect ungated listmode data and perform gating after the raw data files are collected. Populations that may not be obvious on the flow cytometer display will be seen during subsequent data analysis. Moreover, any gating can be done and redone without losing cells.
ASSESSMENT OF CELL VIABILITY BY MICROSCOPY Using Trypan Blue Staining Assessment of cell viability may be accomplished with a microscope, using dyes that mark nonviable cells by dye exclusion. The most commonly used dye is trypan blue, but others may be used as well; see Background Information for details on other useful stains. Viable cells have intact membranes and exclude the dye; nonviable cells are labeled with the dye and are visible with brightfield optics. As well as being useful as a means of assessing functional integrity, trypan blue exclusion is widely used as an objective method of determining viable cell count prior to using cells; a simple protocol for this application is presented, along with other basic cell culture techniques, in APPENDIX 3B.
ALTERNATE PROTOCOL 7
Studies of Cell Function
9.2.7 Current Protocols in Cytometry
Additional Materials (also see Basic Protocol) 0.4% (w/v) trypan blue in PBS (store up to 1 year at room temperature in the dark; filter if a precipitate forms; for PBS recipe, see APPENDIX 2A) Serum-free culture medium (APPENDIX 3B; optional) Additional materials for cell counting with a hemacytometer (APPENDIX 3A) 1. Add an equal volume of 0.4% trypan blue to a cell suspension at a concentration of approximately 106 cells per ml. Use PBS or a serum-free medium for the cell suspension. Serum proteins may stain with trypan blue, resulting in falsely depressed viable counts.
2. Incubate at room temperature for ~3 min and load into a hemacytometer. Using brightfield optics, count cells in three separate fields (see APPENDIX 3A for use of hemacytometer). Count nonviable, deep blue cells as well as viable, clear cells. 3. Calculate viability: % viable = (number viable cells/number total cells) × 100. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
7-AAD (7-amino actinomycin D), 1.0 mg/ml Dissolve 1.0 mg 7-AAD in 50 µl of dimethyl sulfoxide (DMSO), and then add 950 µl of PBS (APPENDIX 2A). Store ≤1 month in the dark at 4°C. As 7-AAD is not water soluble, an organic solvent is required. Although the mutagenicity of 7-AAD is unknown, caution should be exercised when handling the dye.
EMA (ethidium monoazide), 50 ìg/ml Dissolve ethidium monoazide at 5 mg/ml in PBS (APPENDIX 2A). Dilute to 50 µg/ml (working strength) and divide into 0.5-ml aliquots. Wrap in aluminum foil and store ≤6 months at −20°C. Thaw immediately before use. Discard unused workingstrength dye. EMA is very light sensitive; keep wrapped in foil or in the dark.
LDS-751 (laser dye styryl-751) working solution, 2 ìg/ml Stock solution: Dissolve LDS-751 (Exciton Corp.) at 0.2 mg/ml in methanol. Store up to 1 month at 4°C in the dark. Working solution: Add 0.5 ml stock solution to 50 ml PBS (APPENDIX 2A; 2 µg/ml final). Store up to 1 week in the dark at 4°C. COMMENTARY Background Information
Assessment of Cell Viability
Assessments of viability depend on one or both of two cellular properties: (1) the intactness of the cell membrane, and (2) the physiological state of the cell. Dye exclusion methods are based on the fact that only intact membranes are impermeable to large or charged molecules. Intact membranes also maintain cytoplasmic gradients with respect to the surrounding medium, thus retaining intracellular concentrations of ions and small molecules. This latter property also reflects the physiological state of the cells in that energy is
required to maintain gradients. Thus, methods that assay physiological properties of the cell also are dependent upon and indicative of an intact membrane. Probes for membrane integrity Permeability of the cytoplasmic membrane is commonly exploited to mark cells that are moribund or dead. The reagents most often used for this purpose are dyes such as trypan blue or a variety of fluorochromes that will penetrate only damaged, permeable membranes of nonviable cells. These are then easily identified
9.2.8 Current Protocols in Cytometry
visually by the presence of blue color (with trypan blue) in a simple brightfield microscope, or by bright fluorescence seen by fluorescence microscopy or flow cytometry. Fluorescent probes include a wide range of dyes that label DNA of membrane-damaged cells. Tetrabromofluorescein (eosin Y) is a fluorescein derivative that has been used to identify nonviable Candida blastospores (Costantino et al., 1995). In addition, the intracellular penetration of enzymes such as DNase or trypsin can indicate the loss of membrane integrity and thus nonviability (Frankfurt, 1990; Darzynkiewicz et al., 1994; Johnson, 1995). Along the same lines, the penetration of probes for cytoplasmic markers (actin, tubulin, or cytokeratin) has been used to identify cells with damaged plasma membranes (O’Brien and Bolton, 1995). The most widely used group of fluorescent probes are those that label nucleic acids (for further discussion of nucleic acid stains, see UNIT 4.3). The most straightforward labeling methods use propidium iodide (PI) or 7-amino actinomycin D (7-AAD) to identify dead cells, which are hundreds or thousands of times brighter than viable cells. Propidium iodide is in widespread use with many mammalian cell types (Jacobs and Pipho, 1983; Massaro et al., 1989; Coco-Martin et al., 1992; Darzynkiewicz et al., 1994; Stewart and Stewart, 1994; O’Brien and Bolton, 1995), bacteria (Vesey et al., 1994; Nebe-von Caron and Badley, 1996), and protozoa (Armstrong et al., 1991; Humphreys et al., 1994). 7-AAD is a useful alternative to PI. Like PI, 7-AAD penetrates only dead cells, but 7AAD fluorescence is both less intense and at a longer wavelength (~670 nm, versus ~610 nm for PI). These latter two properties make 7AAD preferable as a viability marker when FITC and PE are used to label surface antigens (Schmid et al., 1992). It has been reported that the dye can be used for fixed cells as well (Fetterhoff et al., 1993); see section on viability assays of fixed cells for further discussion. SYBR-14, a recently introduced DNA stain from Molecular Probes, penetrates viable cells (Garner et al., 1994). The dye can be used in conjunction with propidium iodide to unambiguously differentiate viable from dead or moribund sperm cells (Garner et al., 1994). Another dye that can differentiate apoptotic and nonapoptotic cells is SYTO 16 (Molecular Probes), of the SYTO series of dyes (Frey, 1995). The dyes YOYO-1 and TOTO-1 (Molecular Probes; both derivatives of thiazole orange) have also been used successfully in viability assays (Becker et al., 1994). As with other membrane exclu-
sion/DNA binding probes, the dyes do not penetrate viable cells, and remain nonfluorescent until they bind to DNA. Variants of these dyes, YO-PRO-1 and TO-PRO-1 (Molecular Probes), have even higher affinities for nucleic acids and have been used successfully with mammalian and bacterial cells (Vesey et al., 1994; O’Brien and Bolton, 1995; see Haugland, 1994, for details of dyes). The use of YO-PRO-1 for the identification of apoptotic cells seems to have the advantage of preserving the proliferation capacity of living cells (Idziorek et al., 1995). Ethidium bromide (EB) used with low concentrations of acridine orange (AO) identifies normal (AO at high fluorescence level, EB low), early apoptotic (AO low, EB low), and late apoptotic/necrotic (AO low, EB high) cells (Liegler, et al., 1995; Olivier, 1995). Ethidium bromide staining due to loss of membrane integrity identifies the population of necrotic cells; the mechanism of decreased AO staining is not clear but may be related to loss of DNA integrity (Liegler et al., 1995). Probes of physiological state Viable cells can be identified directly using fluorescent probes that identify properties of normal cells. Two principle properties are the integrity of the plasma membrane and the presence of metabolic processes. Cell viability can be assessed based on the presence of cytoplasmic esterases that cleave moieties from a lipidsoluble, nonfluorescent probe to yield a charged fluorescent product that is retained within the cell if membrane function is intact. Hence, viable cells are bright and nonviable cells are dim or nonfluorescent. The most common of these probes is fluorescein diacetate (FDA). It has been used with bacteria (Diaper et al., 1992; Diaper and Edwards, 1994; Vesey et al., 1994; Nebe-von Caron and Badley, 1996), protozoa (Armstrong et al., 1991; Humphreys et al., 1994), phytoplankton (Yentsch and Pomponi, 1994), plants (Galbraith, 1994; Brigham et al., 1995; Kodama and Komamine, 1995), and a variety of mammalian cells (Coco-Martin et al., 1992; Darzynkiewicz et al., 1994; Johnson, 1995). Once FDA diffuses into cells, nonspecific esterases in the cell cytoplasm generate free fluorescein. The dye works well in some instances, but the rate at which fluorescein diffuses out of cells varies greatly. To circumvent this problem, dye variants such as BCECF (Molecular Probes) and carboxyfluorescein diacetate, which require energy-dependent ef-
Studies of Cell Function
9.2.9 Current Protocols in Cytometry
Assessment of Cell Viability
flux of the fluorescent dye (Massaro et al., 1989; Breeuwer et al., 1994), have been developed. Another approach to minimize dye loss is to use the acetoxymethyl ester of calcein (calcein AM). The ester group facilitates uptake of the dye and is cleaved in the cytoplasm to give free calcein. The fluorochrome has an increased retention time (a 3-hr half-life is reported) and less sensitivity to pH (Haugland, 1994; Holló et al., 1994). Molecular Probes provides the dye as a kit in conjunction with ethidium homodimer (EukoLight). Only live cells retain calcein and are labeled green, while dead cells are labeled red because their permeable membranes allow the ethidium homodimer to penetrate and label DNA. In addition, Molecular Probes has produced a variety of kits for determining viability in bacteria (BacLight) or fungi (FungoLight, FunLight). Dihydroethidium is taken up by viable cells and cleaved by esterases to generate ethidium monomer, which binds to DNA and is retained in the nucleus (Bucana et al., 1986). Dead cells do not produce the monomer and remain nonfluorescent. Viable intracellular parasites (Babesia bovi) can be identified with flow cytometry (Wyatt et al., 1991). The dye has been used in conjunction with carboxyfluorescein diacetate to identify viable populations of sperm from frozen samples (Ericsson et al., 1989). Questions regarding the potential toxicity of the dye were raised when the dye was found to inhibit the oxygen uptake of sperm cells stained with dihydroethidium (Downing et al., 1991). Another dye that seems to have promise is Vita Blue (Becton Dickenson; Lee et al., 1989). This dye is excited by red light, thereby making possible the simultaneous use of green and orange emitting dyes that are excited by 488nm light. Viable cells maintain electrochemical gradients across the plasma membrane. Functional subsets of this general phenomenon include maintenance of pH and other ion gradients as well as the capacity for energy-yielding metabolism in mitochondria. These physiological processes can be exploited to distinguish viable from nonviable cells. Fluorochromes useful in viability assays include those used to measure membrane potential and intracellular pH. These probes are typically lipophilic, charged molecules that preferentially partition into cells having a negative potential difference with respect to the surrounding environment, so that
the dye becomes concentrated in the cytoplasm or internal organelles. One of the most commonly used probes for identifying viable cells is rhodamine 123. This is a cationic lipophilic dye that partitions into mitochondria because of their low potential. Active mitochondria in viable cells are stained bright green; loss of gradients within the cell results in loss of the dye. Rhodamine 123 can indicate viable cells among bacteria (Diaper et al., 1992; Diaper and Edwards, 1994; Vesey et al., 1994; Porter et al., 1995a,b; Nebe-von Caron and Badley, 1996) or a variety of mammalian cells (Darzynkiewicz et al., 1994; Johnson, 1995). The presence of aliphatic side chains on fluorochromes may strongly influence retention within the cells. For example, side chains of the cyanine dye DiOC6(3) cause the probe to partition into the mitochondrial membranes, resulting in increased concentration and brightness—the brightness related to better fluorescence in a lipid environment (Sims et al., 1974). A possible disadvantage is that increased affinity for membranes may also slow the loss of the dye if mitochondria lose their metabolic capacity. Any of the assays employing dye uptake to label viable cells can be coupled with assays using dyes to label nonviable cells. Thus, one may use rhodamine 123 with, for example, PI or ethidium bromide. Instead of viable cells being labeled with cationic dyes, nonviable cells can be labeled with the lipophilic anionic dye oxonol; this has been done in bacteria (Deere et al., 1995; Nebevon Caron and Badley, 1996) and protozoa (Humphreys et al., 1994). Loss of negative potential with respect to the outside causes accumulation of oxonol within dead cells. It has been reported, however, that starved bacteria can contain populations of oxonol-positive bacteria that are PI-negative (Nebe-von Caron et al., 1996). In some cells, dyes that accumulate in live cells may be pumped out by the glycoprotein pump (Holló et al., 1994; Shapiro, 1995). Thus, cells that do not stain brightly with a dye such as rhodamine 123 may still be viable. It has also been reported that some nonspecific staining of rhodamine 123 in natural particulate environments may complicate the identification of live bacteria (Porter et al., 1995b). Conversely, cells in the presence of glutathione may have hyperpolarized mitochondria and hence enhanced uptake of dye into mitochondria (Pieri et al., 1992). Valinomycin has been used to hyperpolarize bacteria to enhance their dye uptake (Por-
9.2.10 Current Protocols in Cytometry
ter et al., 1995b). There is some indication that the toxicity of rhodamine appears to be low (Downing et al., 1991). Probes for fixed cells Often cells must be fixed or permeabilized before analysis. In both cases analysis of other markers is less ambiguous if nonviable or damaged cells can be identified and rejected. Several methods have been successful in the labeling of dead or damaged cells prior to fixation. A modification of Hoechst/PI staining methods permits the use of ethanol as a fixative (Pollack and Cianco, 1990). Prior to fixation, dead cells are labeled with PI. After ethanol fixation, all cells label with Hoechst, but the PI in dead cells quenches Hoechst fluorescence; viable cells are reported to have insignificant PI fluorescence. Other DNA probes that penetrate and stain dead or damaged cells brightly are ethidium monoazide (EMA) and laser dye styryl-751 (LDS-751). Ethidium monoazide is positively charged and penetrates the membranes of dead or damaged cells but not live cells. EMA intercalates into DNA and can be photochemically crosslinked with short exposure to visible light (Riedy et al., 1991). In contrast, LDS-751 penetrates both damaged and live cells, but labels the DNA of damaged or dead cells much more brightly (Terstappen et al., 1988). A novel use of 7-AAD for labeling nonviable fixed cells involves addition of a molar excess of actinomycin D (AD) while cells are being fixed (Fetterhoff et al., 1993). It is thought that the higher concentration of AD prevents binding of 7-AAD to viable cells; for the short term, nonviable cells retain bound 7-AAD. A different approach is based on the penetration of large molecules into dead or damaged cells. Streptavidin-Tricolor (SA-TR; Caltag) irreversibly penetrates nonviable cells, staining them bright red; subsequent washing, fixation, and permeabilization do not result in significant dye loss (Levelt and Eichmann, 1994). The mechanism for this retention is not clear. This technique permits prelabeling damaged cells prior to labeling intracellular antigens. Methods for microscopy In determining the viability of cells, one should not overlook the obvious. Much information can be obtained by direct observation of the cells in a microscope. In many cases, obviously misshapen or bloated cells or cells that have lost refractility in phase contrast indicate severe problems that obviate more so-
phisticated approaches. In other cases, morphological changes are very useful in following physiological processes. For example, certain morphological changes are hallmarks of apoptosis (Darzykiewicz et al., 1994). In plant cells, changes in shape are easily detectable, and the loss of metabolically driven processes such as cytoplasmic streaming can indicate the loss of viability (Brigham et al., 1995). Assessment of cell viability under the microscope can be accomplished with stains that mark nonviable cells. Dyes include the very common trypan blue (McGahon et al., 1995), nigrosin (Johnson, 1995), and erythrosin B (Bochner et al., 1989). Viable cells have intact membranes and exclude the dyes. Nonviable cells are labeled and are visible with brightfield optics. Fluorescent probes in common use for flow cytometry can also be used in the microscope. These include FDA (Humphreys et al., 1994), YO-PRO-1 (Idziorek et al., 1995), and dihydroethidine (Bucana et al., 1986).
Critical Parameters and Troubleshooting Using PI or 7-AAD with phycoerythrin (PE) Although the emission spectrum of PE overlaps the shorter-wavelength end of PI and 7AAD emission spectra, PE can be used in conjunction with PI or 7-AAD for dead cell discrim ination . Using a gate for PI- or 7-AAD-negative cells allows for the examination of PE label on presumed viable cells. It is important that there be appropriate compensation between PE and PI or 7-AAD detectors, as both fluorochromes will be detected by both detectors. (For illustrations of compensation, see Fig. 9.2.2 and comments in Alternate Protocol 2.) To get compensation for dual-labeled cells, prepare a tube without PI or 7AAD. Adjust the compensation of PE into the PI/7-AAD channel such that PE-positive cells are in the range of PI- or 7-AAD-negative or viable cells (see Fig. 9.2.2C). If expected nonviable cells are not found, demonstrate the efficacy of the dye by heating. Take one tube of cells ready for analysis and heat 10 min in a 45°C water bath. Cool to room temperature and reanalyze. All cells are now nonviable and should be a brighter red. 7-AAD should also be usable with dyes that emit in the longer-wavelength region if they are excited by a second laser that is not colinear with the 488-nm laser. For example, a heliumneon (HeNe) laser emitting at 633 nm, a krypton (Kr) laser emitting at 647 nm, or a diode
Studies of Cell Function
9.2.11 Current Protocols in Cytometry
laser operating in the same region would not excite 7-AAD. One may also be able to use a HeNe laser that is colinear with a 488-nm beam if the filters in front of the PMT that detects 7-AAD emission exclude the 633-nm laser line. Probes of physiological state Cells must be kept under optimal conditions for assays that reflect their physiological state. That is, some cells may survive well in PBS or HBSS, but others may require serum supplementation or other factors to remain healthy. Because FDA can leak from cells, it is important to analyze immediately after the incubation period. When FDA is used together with cell surface markers labeled with PE, the FDA fluorescence overlaps the PE emission range. Thus, setting proper compensation is crucial. Compensation should be checked with samples labeled with each fluorochrome. The glycoprotein pump in some cells may pump rhodamine 123 out of live cells (Holló, 1994; Shapiro, 1995). Thus, cells that do not stain brightly may be viable. When in doubt, try an alternate reagent for nonviable cell identification, such as PI or 7-AAD. Probes for fixed cells Although EMA emission can be used in conjunction with surface-labeled antibodies, the emission of PE overlaps that of EMA; hence, appropriate compensation is required (see Alternate Protocol 2 for further discussion of compensation). LDS-751 stains all nucleated cells, so all will be positive. Dead or damaged cells are about ten-fold brighter than viable cells. If nonviable cells are present, then two populations should be present; mature red cells will be negative (see Terstappen et al., 1988, for illustrations). Artifacts in cell surface labels have been observed in formaldehyde-fixed cells labeled with LDS-751 (McCarthy et al., 1994). If alteration of the surface labeling pattern is suspected, compare the staining pattern of unfixed, labeled cells stained with LDS-751. Methods for microscopy After staining, count cells within 5 min. On sitting, some viable cells may become permeable and take up dye, appearing as false nonviables.
Anticipated Results
Assessment of Cell Viability
Probes for membrane integrity. Nonviable cells are stained bright red and viable cells are nonfluorescent. With PI, there should be about a 2-log difference in brightness between viable and nonviable cells; 7-AAD-positive cells are
dimmer. Discrimination of viable cells, nonviable cells, and debris can be done easily on a bivariate plot of forward scatter (typically linear scale) versus log PI or 7-AAD fluorescence (see Fig. 9.2.1). Probes of physiological state. FDA and rhodamine 123 are both excited by 488-nm light and fluoresce green. Filters used for measuring fluorescein (e.g., 530 ± 20–nm band-pass) are adequate to detect fluorescence of either dye. With logarithmic amplification of the fluorescence signal, viable FDA-positive cells are very bright, but viable rhodamine 123–positive cells will be dimmer. A bivariate plot of forward scatter versus rhodamine 123 fluorescence may make it simpler to distinguish rhodamine 123– positive viable cells from nonviable cells (see Fig. 9.2.3B). Probes for fixed cells. Because the fluorescence of EMA is not as bright as that of propidium iodide, discrimination of nonviable, EMA-bright cells from viable cells may be less obvious. A bivariate plot of forward light scatter versus EMA fluorescence may aid in distinguishing viable (EMA-negative) cells. When stained with LDS-751, viable cells can be identified on a bivariate plot of light scatter and red fluorescence as a population of intermediate brightness. Dead or damaged cells stain more brightly, and enucleate red blood cells are unstained. Methods for microscopy. The procedure is very straightforward, but phase contrast optics can aid in the identification of viable cells that do not stain with trypan blue. Nonviable cells are blue and phase dense; viable cells are phase bright.
Time Considerations Probes for membrane integrity. Cell staining should take <5 min with PI or 30 min with 7-AAD. Instrument setup (and compensation, if employed) should take 5 min. If data are collected as listmode files, then subsequent analysis may require another 5 to 10 min depending on the complexity of the gating and the degree of automation of the analysis software employed. Probes of physiological state. Cell staining should take 5 min with rhodamine 123 or 15 min with FDA. Instrument setup (and compensation, if employed) should take 5 min. If data are collected as listmode files, then subsequent analysis may require another 5 to 10 min depending on the complexity of the gating and the degree of automation of the analysis software employed.
9.2.12 Current Protocols in Cytometry
Probes for fixed cells. EMA labeling of cells takes ~10 min to stain and cross-link the dye, followed by ~30 mins for washing and fixing. Labeling of cells with LDS-751 takes ~20 min for staining and incubation. Instrument setup (and compensation, if employed) should take 5 min. If data are collected as listmode files, then subsequent analysis may require another 5 to 10 min depending on the complexity of the gating and the degree of automation of the analysis software employed. Methods for microscopy. Staining and counting cells can be done in 5 to 10 min.
Diaper, J.P., Tither, K., and Edwards, C. 1992. Rapid assessment of bacterial viability by flow cytometry.Appl. Microbiol. Biotechnol. 38:268-272.
Literature Cited
Frankfurt, O.S. 1990. Flow cytometric measurement of cell viability using DNase exclusion. Methods Cell Biol. 33:13-18.
Armstrong, M.Y., Koziel, H., Rose, R.M., Arena, C., and Richards, F.F. 1991. Indicators of Pneumocystis carinii viability in short-term cell culture. J. Protozool. 38: 88S-90S. Becker, B., Clapper, J., Harkins, K.R., and Olson, J.A. 1994. In situ screening assay for cell viability using a dimeric cyanine nucleic acid stain. Anal. Biochem. 221:78-84. Bochner, B.S., McKelvey, A.A., Schleimer, R.P., Hildreth, J.E.K., and MacGlashan, D.W., Jr. 1989. Flow cytometric methods for the analysis of human basophil surface antigens and viability. J. Immunol. Methods 125:265-271. Breeuwer, P., Drocourt, J.L., Rombouts, F.M., and Abbe, T. 1994. Energy-dependent, carrier-mediated extrusion of carboxyfluorescein from Saccharomyces cerevisiae allows rapid assessment of cell viability by flow cytometry. Appl. Environ. Microbiol. 60:1467-1472. Brigham, L.A., Woo, H.H., and Hawes, M.C. 1995. Root border cells as tools in plant cell studies. Methods Cell Biol. 49:377-387. Bucana, C., Saiki, I., and Nayar, R. 1986. Uptake and accumulation of the vital dye hydroethidine in neoplastic cells. J. Histochem. Cytochem. 34:1109-1115. Coco-Martin, J.M., Oberink, J.W., van der Veldende Groot, T.A., and Beuvery, E. 1992. Viability measurements of hybridoma cells in suspension cultures. Cytotechnology 8:57-64. Costantino, P.J., Budd, D.E., and Gare, N.F. 1995. Enumeration of viable Candida albicans blastospores using tetrabromofluorescein (eosin Y) and flow cytometry. Cytometry 19:370-375. Darzynkiewicz, Z., Li, X., and Gong, J. 1994. Assays of cell viability: Discrimination of cells dying by apoptosis. Methods Cell Biol. 41:15-38. Deere, D., Porter, J., Edwards, C., and Pickup, R. 1995. Evaluation of the suitability of bis-(1,3dibutylbarbituric acid) trimethine oxonol, (diBA-C4(3)-), for the flow cytometric assessment of bacterial viability.FEMS Microbiol. Lett. 130:165-169. Diaper, J.P. and Edwards, C. 1994. Survival of Staphylococcus aureus in lakewater monitored by flow cytometry. Microbiology 140: 35-42.
Downing, T.W., Garner, D.L., Ericsson, S.A., and Redelman, D. 1991. Metabolic toxicity of fluorescent stains on thawed cryopreserved bovine sperm cells. J. Histochem. Cytochem. 39:485-489. Ericsson, S.A., Garner, D.L., Redelman, D., and Ahmad, K. 1989. Assessment of the viability and fertilizing potential of cryopreserved bovine spermatozoa using dual fluorescent staining and twoflow cytometric systems. Gamete Res.22: 355-368. Fetterhoff, T.J., Holland, S.P., and Wiles, K.J. 1993. Fluorescent detection of non-viable cells in fixed cell preparations. Cytometry (Suppl. 6):204b.
Frey, T. 1995. Nucleic acid dyes for the detection of apoptosis in live cells. Cytometry 21:256-274. Galbraith, D.W. 1994. Flow cytometry and sorting of plant protoplasts and cells. Methods Cell Biol. 42:539-561. Garner, D.L., Johnson, L.A., Yue, S.T., Roth, B.L., and Haugland, R.P. 1994. Dual DNA staining assessment of bovine sperm viability using SYBR-14 and propidium iodide. J. Androl. 15:620-629. Haugland, R.P. 1994. Spectra of fluorescent dyes used in flow cytometry. Methods Cell Biol. 42:641-663. Holló, Z., Homolya, L., Davis, C.W., and Sarkadi, B. 1994. Calcein accumulation as a fluorometric functional assay of the multidrug transporter. Biochim. Biophys. Acta 1191:384-388. Humphreys, M.J., Allman, R., and Lloyd, D. 1994. Determination of the viability of Trichomonas vaginalis using flow cytometry. Cytometry 15:343-348. Idziorek, T., Estaquier, J., De Bels, F., and Ameisen, J.C. 1995. YOPRO-1 permits cytofluorometric analysis of programmed cell death (apoptosis) without interfering with cell viability. J. Immunol. Methods 185:249-258. Jacobs, D.B. and Pipho, C. 1983. Use of propidium iodide staining and flow cytometry to measure anti-mediated cytotoxicity: Resolution of complement-sensitive and resistant target cells. J. Immunol. Methods 62:101-108. Johnson, J.E. 1995. Methods for studying cell death and viability in primary neuronal cultures. Methods Cell Biol. 46:243-276. Kodama, H. and Komamine, A. 1995. Synchronization of cell cultures of higher plants. Methods Cell Biol. 49:315-329. Lee, L.G., Berry, G.M., and Chen, C.H. 1989. Vita Blue: A new 633-nm excitable fluorescent dye for cell analysis. Cytometry 10:151-164. Levelt, C.N. and Eichmann, K. 1994. StreptavidinTricolor is a reliable marker for nonviable cells subjected to permeabilization or fixation. Cytometry 15:84-86.
Studies of Cell Function
9.2.13 Current Protocols in Cytometry
Liegler, T.J., Hyun, W., Yen, T.S., and Stites, D.P. 1995. Detection and quantification of live, apoptotic, and necrotic human peripheral lymphocytes by single-laser flow cytometry. Clin. Diag. Lab. Immunol. 2:369-376.
Porter, J., Pickup, R., and Edwards, C. 1995b. Membrane hyperpolarisation by valinomycin and its limitations for bacterial viability assessment using rhodamine 123 and flow cytometry. FEMS Microbiol. Lett. 132:259-262.
Massaro, E.J., Elstein, K.H., Zucker, R.M., and Bair, K.W. 1989. Limitations of the fluorescent probe viability assay. Mol. Toxicol. 2:271-284.
Riedy, M.C., Muirhead, K.A., Jensen, C.P., and Stewart, C.C. 1991. Use of a photolabeling technique to identify nonviable cells in fixed homologous or heterologous cell populations. Cytometry 12:133-139.
McCarthy, D.A., Macey, M.G., Cahill, M.R., and Newland, A.C. 1994. Effect of fixation on quantification of the expression of lymphocyte function-associated surface antigens. Cytometry 17:39-49. McGahon, A.J., Martin, S.J, Bissonnette, R.P, Mahboubi, A., Shi, Y., Mogil, R.J., Nishioka, W.K., and Green, D.R. 1995. The end of the (cell) line: Methods for the study of apoptosis in vitro. Methods Cell Biol. 46:153-185. Nebe-von Caron, G. and Badley, R.A. 1996. Bacterial characterization by flow cytometry. In Flow Cytometry Applications in Cell Culture (M. AlRubeai and A.N. Emery, eds.) pp. 257-290. Marcel Dekker, New York. Nebe-von Caron, G., Badley, R.A., and Powell, J.R. 1996. Identification and viability assessment of bacterial species in natural samples by flow cytometry and sorting. Cytometry (Suppl. 8):115. O’Brien, M.C. and Bolton, W.E. 1995. Comparison of cell viability probes compatible with fixation and permeabilization for combined surface and intracellular staining in flow cytometry. Cytometry 19:243-255. Olivier, R. 1995. Flow cytometry technique for assessing effects of N-acetylcysteine on apoptosis and cell viability of human immunodeficiency virus–infected lymphocytes. Methods Enzymol. 251:270-278. Pieri, C., Moroni, F., and Recchioni, R. 1992. Glutathione influences the proliferation as well as the extent of mitochondrial activation in rat splenocytes. Cell. Immunol. 145:210-217. Pollack, A. and Ciancio, G. 1990. Cell cycle phase– specific analysis of cell viability using Hoechst 33342 and propidium iodide after ethanol preservation. Methods Cell Biol. 33:19-24. Porter, J., Edwards, C., and Pickup, R.W. 1995a. Rapid assessment of physiological status in Escherichia coli using fluorescent probes. J. Appl. Bacteriol. 79:399-408.
Schmid, I., Krall, W.J., Uittenbogaart, C.H., Braun, J., and Giorgi, J.V. 1992. Dead cell discrimination with 7-amino-actinomycin D in combination with dual color immunofluorescence in single laser flow cytometry. Cytometry 13:204-208. Shapiro, H.M. 1995. Practical Flow Cytometry, 3rd ed. Wiley-Liss, New York. Sims, P.J., Waggoner, A.S., Wang, C.-H., Hoffman, J.H. 1974. Studies on the mechanism by which cyanine dyes measure membrane potential in red blood cells and phosphatidylcholine vesicles. Biochemistry 13:3315-3330. Stewart, C.C. and Stewart, S.J. 1994. Cell preparation for the identification of leukocytes. Methods Cell Biol. 41:39-59. Terstappen, L.W.M.M., Shah, V.O., Conrad, M.P., Rechtenvald, D., and Loken, M. 1988. Discriminating between damaged and intact cells in fixed flow cytometric samples. Cytometry 9:477-484. Vesey, G., Narai, J., Ashbolt, N., Williams, K., and Veal, D. 1994. Detection of specific microorganisms in environmental samples using flow cytometry. Methods Cell Biol. 42:489-522. Wyatt, C.R., Goff, W., and Davis, W.C. 1991. A flow cytometric method for assessing viability of intraerythrocytic hemoparasites. J. Immunol. Methods 140:23-30. Yentsch, C.M., and Pomponi, S.A. 1994. Strategies for flow cytometric analysis of marine microalgae and sponge cells. Methods Cell Biol. 42:523538.
Contributed by David M. Coder University of Washington School of Medicine Seattle, Washington
Assessment of Cell Viability
9.2.14 Current Protocols in Cytometry
Flow Cytometric Measurement of Intracellular pH
UNIT 9.3
Cells actively maintain intracellular pH (pHi) within a narrow range of values. However, changes in pHi do occur during certain fundamentally important biological processes, such as cell division, and in response to cell signaling. Intracellular pH can be measured by flow cytometry using the probes carboxy SemiNaphthoRhodaFluor acetoxymethyl ester (carboxy SNARF-1 AM) and 2′,7′-bis(2-carboxyethyl)-5-(and -6)-carboxyfluorescein acetoxymethyl ester (BCECF AM). These pH probes are weak acids with pKa values close to 7.0. Their protonated and free base forms have different emission spectra; taking the ratio of the two resulting emissions gives a signal that is proportional to pHi and independent of cellular dye content. The acetoxymethyl (AM) ester form of the dyes enters the cells readily and is hydrolyzed by nonspecific cellular esterases to yield the free fluorescent dye. The overall approach to pHi calibration is similar to that for calcium measurements: i.e., generation of a calibration curve by imparting known changes in pHi. The pHi value of the test sample can then be interpolated from the calibration curve. The first two protocols detail pH calibration methods using SNARF-1: the traditional pH calibration method using highpotassium buffers and the proton ionophore nigericin (see Basic Protocol) and a more recently developed technique called the pseudo null method, which involves resuspension of cells in defined mixtures of weak acids and weak bases (see Alternate Protocol 1). Alternatively, BCECF can be substituted for SNARF-1 (see Alternate Protocol 2). Like SNARF-1, BCECF is useful for either calibration method. pH MEASUREMENTS WITH SNARF-1 USING NIGERICIN CALIBRATION For most purposes, SNARF-1 is the most satisfactory pHi probe currently available. It has a clear isobestic point (i.e., a fluorescence emission wavelength that is insensitive to pH change), and it is very sensitive to pHi changes within the physiological range. Calibration is achieved by resuspending dye-loaded cells in high-potassium buffers in the presence of the proton ionophore nigericin. Nigericin allows exchange of H+ for K+ across their concentration gradients, so that when external and internal potassium ion concentrations are approximately equal, there is free movement of hydrogen ions and intracellular pH is the same as extracellular pH. This feature allows the experimenter to set pHi in a series of calibration standards to a set of known values, and thus to construct a calibration curve of fluorescence versus pHi.
BASIC PROTOCOL
Materials Cell sample PBS (APPENDIX 2A) 2 mM carboxy SNARF-1 AM (see recipe) High-potassium calibration buffers (see Support Protocol 1) 1 mg/ml nigericin (see recipe) 12 × 75–mm polypropylene or polystyrene tubes (Falcon) as required by flow cytometer Flow cytometer with 488-nm argon laser and ratio parameter, if available Bandpass filters centered at or around 580 and 640 nm Load cells with intracellular pH indicators 1. Resuspend single cells at 1 × 106 cells/ml in PBS. Use half a million cells per sample. Usually 3 × 106 cells are enough for a five-point calibration curve and one test sample. Contributed by Sue Chow and David Hedley Current Protocols in Cytometry (1997) 9.3.1-9.3.10 Copyright © 1997 by John Wiley & Sons, Inc.
Studies of Cell Function
9.3.1
2. For each milliliter of cell suspension, add 2.5 µl of 2 mM carboxy SNARF-1 AM (5 µM final) and incubate 30 min at 37°C. 3. Centrifuge at room temperature, selecting time and g force as appropriate for the cell line used. Resuspend dye-loaded cells in 10 µl PBS for each milliliter of cell suspension stained (1 × 108 cells/ml final). Set up calibration and test samples 4. Pipet 0.5 ml of each high-potassium calibration buffer into a separate 12 × 75–mm polypropylene or polystyrene tube. Include one tube with 0.5 ml PBS alone for test sample. 5. Add 5 µl cell suspension (0.5 × 106 cells) and 1 µl of 1 mg/ml nigericin (2 µg/ml final) to each tube containing high-K+ calibration buffers. Nigericin is added at this point to the cell suspension rather than to the stock high-K+ calibration buffers because it sticks to plastic, resulting in an unknown available concentration. Nigericin allows exchange of H+ for K+ across their concentration gradients, so that when external and internal potassium ion concentrations are approximately the same, free hydrogen ion movement results in intracellular pH equal to the pH of the high-K+ calibration buffer.
6. Add 5 µl cell suspension (0.5 × 106 cells) to PBS tube. Do not add nigericin to this tube. 7. Allow pHi to equilibrate 20 min at room temperature before running on flow cytometer. Cytotoxicity of nigericin is enhanced if cells are incubated at 37°C.
Acquire fluorescence data on flow cytometer 8. Set up flow cytometer. Use forward light scatter and 90° side scatter to look at the size and granularity distribution of the cells. Collect emission fluorescences at 580 and 640 nm using linear amplification, and determine the 640/580 fluorescence ratio. If ratio measurements cannot be made directly on the flow cytometer, use an off-line program such as WinList (Verity Software House). This fluorescence ratio increases with increased pHi. Reversing the ratio will produce the opposite effect.
9. Optimize flow cytometer settings for ratiometric methods. To optimize sensitivity, it is advisable to make maximum use of the available channel numbers for each fluorescence signal. However, as illustrated in Figure 9.3.1, the calibration procedure imparts significant shifts in the two emissions, and it is important that these remain on scale for the full range of the calibration series. For this reason it is advisable to determine the optimal instrument settings from the extremes of the range in the calibration series before running the whole series.
10. Run the calibration series and the test sample. 11. Exclude dead cells by gating out cells that fail to retain SNARF-1 fluorescence (see Fig. 9.3.1).
Flow Cytometric Measurement of Intracellular pH
It is “cleaner” to gate on fluorescence than on the scatters. It is important to exclude both cells with too little SNARF-1 fluorescence and cells with fluorescence in the overflow (uppermost) channel to avoid erroneous ratio values.
9.3.2 Current Protocols in Cytometry
G:
G:A 29
G:A pH 7.0
52
G:A pH 7.0
169
pH 7.0
Count
488 SC
64
0
0
64
0
28
pH 8.0
33
pH 8.0
76
pH 8.0
Count
580 nm
A
0
0
0 0
64 FALS
0 0
1024 580 nm
0 0
1024 640 nm
0
1024 Ratio 640/580
Figure 9.3.1 Changes in fluorescence emissions at 580 and 640 nm in response to nigericin calibration procedure. CCRF-CEM wild-type cells were loaded with SNARF-1 AM and resuspended in buffers at pH 7.0 (top) and 8.0 (bottom). Note that at pH 8.0 there is a decrease in the 580 nm emission and an increase in the 640 nm emission. This is reflected in the increase in the 640/580 fluorescence ratio. The single parameter histograms are gated on the A region (in lower left panel) based on forward-angle light-scatter (FALS) vs. 580 nm fluorescence rather than the two scatter signals. 488 SC, 488 nm side scatter; G:, gate; G:A, gated on A.
Analyze listmode files and interpolate pHi of test sample 12. Take the mean fluorescence ratio of each standard and plot against known pHi (see Fig. 9.3.2). 13. Read pHi of test sample from calibration curve. pH MEASUREMENT WITH SNARF-1 USING PSEUDO NULL CALIBRATION
ALTERNATE PROTOCOL 1
Because of concerns regarding the nigericin calibration method (see Commentary), we developed an alternative calibration technique called the pseudo null method. This involves resuspension of cells in defined mixtures of weak acids and weak bases. These impart changes in pHi that are predictable from the Henderson-Hasselbalch equation, and are independent of intracellular potassium concentration. Additional Materials (also see Basic Protocol) 10 mM HEPES buffer (see recipe) HDFBS: 10 mM HEPES buffer supplemented with 10% dialyzed fetal bovine serum (FBS) 6× pseudo null calibration solutions (see recipe) Load cells with intracellular pH indicators 1. Resuspend single cells at 1 × 106 cells/ml in 10 mM HEPES buffer. Usually 2 × 106 cells are enough per calibration assay.
2. For each milliliter of cell suspension, add 2.5 µl of 2 mM carboxy SNARF-1 AM (5 µM final). Incubate 30 min at 37°C. 3. Centrifuge at room temperature, selecting time and g force as appropriate for the cell line used. Resuspend dye-loaded cells at 2 × 106 per milliliter HDFBS. Incubate 10 min for clinical samples or 20 min for cell lines at 37°C.
Studies of Cell Function
9.3.3 Current Protocols in Cytometry
800
BCECF
Mean fluorescence ratio
750
700 SNARF-1 650
600
550
500
450 7.0
7.2
7.4
7.6
7.8
8.0
pH
Figure 9.3.2 Representative calibration curves obtained for CCRF-CEM wild-type cells stained with SNARF-1 AM or BCECF AM using the nigericin method. Interpolating the fluorescence ratio value for the test sample gave almost the same pHi result for the two dyes, but note that SNARF-1 has a steeper calibration curve in this range, indicating greater sensitivity and accuracy in measurement. Mean fluorescence ratio is expressed in arbitrary units.
Resuspension of dye-loaded cells in HEPES buffer supplemented with 10% dialyzed fetal bovine serum (HDFBS), followed by incubation at 37°C, helps the cells to reestablish physiological conditions.
4. Pipet 100 µl pseudo null calibration solutions into separate 12 × 75–mm polypropylene or polystyrene tubes. Prepare duplicate or triplicate tubes of the highest and lowest pseudo null value. Include one tube with 100 µl HDFBS for test sample. Incubate tubes at 37°C. Keep caps on tubes to prevent evaporation. The extra tubes of solutions at the extremes of the calibration range are for setting up the flow cytometer.
Acquire fluorescence data on flow cytometer 5. Set up flow cytometer. Use forward light scatter and 90° side scatter to look at the size and granularity distribution of the cells. Collect emission fluorescences at 580 and 640 nm using linear amplification, and determine the 640/580 fluorescence ratio. If ratio measurements cannot be made directly on the flow cytometer, use an off-line program such as WinList (Verity Software House). This fluorescence ratio increases with increased pHi. Reversing the ratio will produce the opposite effect.
6. Optimize flow cytometer settings for ratiometric methods. Flow Cytometric Measurement of Intracellular pH
In order to improve sensitivity, it is advisable to make maximum use of the available channel numbers for each fluorescence signal. However, the calibration procedure imparts signifi-
9.3.4 Current Protocols in Cytometry
cant shifts in the two emissions, and it is important that these remain on scale for the full range of the calibration series. For this reason it is recommended to determine the optimal instrument settings from the extremes of the range in the calibration series before running the whole series.
7. Run calibration series and test sample. Add 100 µl stained cells (0.2 × 106 cells) to the first tube (containing prewarmed pseudo null solution or HDFBS). Immediately vortex and load onto the flow cytometer. Run 20 sec after cells are added to the solution. Repeat for remaining tubes until calibration series and test sample have all been run. For the pseudo null method, the steepness of the calibration curve is dependent on the concentration of weak acids and bases. Equal volumes of stained cells are added to 6× pseudo null solution for a final 3× concentration; this gives good resolution without overstressing the cells. The shift in pHi imparted by the pseudo null solution is transient, unlike the nigericin method, probably because cellular regulatory mechanisms readjust pHi rapidly. Therefore, consistent timing is important. Dye-loaded cells are added to prewarmed pseudo null solution 20 seconds before data acquisition on the flow cytometer, and each tube of cells can be read only once.
Analyze listmode files and interpolate pHi of test sample 8. Exclude dead cells by gating out cells that fail to retain SNARF-1 fluorescence (see Fig. 9.3.1). It is “cleaner” to gate on fluorescence than on the scatters. It is important to exclude both cells with too little SNARF-1 fluorescence and cells with fluorescence in the overflow (uppermost) channel to avoid erroneous ratio values.
9. Take the mean fluorescence ratio of each standard and plot against known pHi. Calibration curves are similar to the example shown for the nigericin method (see Fig. 9.3.2).
10. Read pHi of test sample from calibration curve. pH MEASUREMENT WITH BCECF BCECF does not have an isobestic point, but its green fluorescence is more pH-sensitive than its red fluorescence, allowing it to be used as a ratiometric dye. Because its pKa is ~6.8, BCECF is more sensitive for pHi measurements in the acidic range. It can be used in place of SNARF-1 with either nigericin or pseudo null calibration (see Basic Protocol and Alternate Protocol 1). However, the staining concentration and conditions are different, and bandpass filters centered at or around 525 and 640 nm are required. For each milliliter of cell suspension, add 2 µl of 1 mg/ml BCECF AM (2 µg/ml final; see recipe) and incubate 15 min at 37°C. Collect emission fluorescences at 525 and 640 nm using linear amplification, and determine the 525/640 fluorescence ratio. This ratio increases with increased pHi; reversing the ratio will produce the opposite effect.
ALTERNATE PROTOCOL 2
PREPARATION OF HIGH-POTASSIUM CALIBRATION BUFFER SERIES
SUPPORT PROTOCOL
This support protocol describes the solutions required for the nigericin calibration method (see Basic Protocol). 1. To 1 liter deionized, distilled water add: 10.44 g KCl (140 mM) 0.952 g MgCl2 (1 mM) 0.294 g CaCl2⋅2H2O (2 mM) 0.9 g α-D-glucose (5 mM).
Studies of Cell Function
9.3.5 Current Protocols in Cytometry
2. Divide into two 500-ml batches. To one batch, add 1.952 g 2-[N-morpholino] ethanesulfonic acid (MES; 20 mM final) to make acidifying solution. To second batch, add 1.211 g tris(hydroxymethyl)aminomethane (Tris base; 20 mM final) to make alkalinizing solution. Filter sterilize solutions with 0.22-µm filters. 3. Prepare calibration series at room temperature: Pour ∼30 ml of each solution into a 100-ml glass beaker. Measure pH with a pH electrode while stirring on a magnetic stir plate. While continuing to measure pH, slowly add an appropriate amount of the alkalinizing or acidifying solution to achieve the desired pH. Pour 50 ml into a polypropylene tube. Repeat steps to make the whole calibration series. Store up to several months at 4°C. Although microbial growth in these solutions is not a major concern and there is no need to filter sterilize each standard solution, practice sterile technique.
REAGENTS AND SOLUTIONS For aqueous solutions, use Milli-Q double-filtered or deionized, distilled water. For solvents, use anhydrous dimethyl sulfoxide (DMSO) and absolute ethanol. For sterilization, use 0.22-ìm filter.
BCECF AM (2′,7′-bis(2-carboxyethyl)-5-(and -6)-carboxyfluorescein acetoxymethyl ester), 1 mg/ml Dissolve BCECF AM (mol. wt. ∼620; Molecular Probes) in DMSO at 1 mg/ml. Store protected from light up to several weeks at −20°C. Dye can be purchased in 50-ìg aliquots to avoid aliquoting or refreezing.
n-Butyric acid, 1 M Add 4.6 ml of 10.9 M n-butyric acid (pKa = 4.82; Sigma) to 40 ml water. Titrate to pH 7.4 with NaOH. Add water to 50 ml final volume. Filter sterilize and store up to several months at 4°C. Carboxy SNARF-1 AM (SemiNaphthoRhodaFluor-1 acetoxymethyl ester), 2 mM Dissolve 50 µg SNARF-1 (mol. wt. 567; Molecular Probes) in 44 µl DMSO. Store protected from light up to several weeks at −20°C. Dye can be purchased in 50-ìg aliquots to avoid aliquoting or refreezing.
HEPES (N-[2-hydroxyethyl]piperazine-N′-[2-ethanesulfonic acid]), 10 mM 2.383 g HEPES (10 mM final) 7.802 g NaCl (133.5 mM final) 0.298 g KCl (4 mM final) 0.166 g NaH2PO4⋅H2O (1.2 mM final) 0.144 g MgSO4 (1.2 mM final) 1.981 g α-D-glucose (11 mM final) 0.294 g CaCl2⋅2H2O (2 mM final) H2O to 1 liter Titrate to pH 7.4 with NaOH Filter sterilize Store up to several months at 4°C Nigericin, 1 mg/ml Dissolve nigericin (mol. wt. 747; Sigma) in ethanol at 1 mg/ml in a glass vial and store up to several months at −20°C. Flow Cytometric Measurement of Intracellular pH
9.3.6 Current Protocols in Cytometry
Table 9.3.1
Sample
Composition of Pseudo Null Calibration Solutions
6× concentration BAa/TMAb (mM)
0.5 log [A]/[B]c
Pseudo null pHc
BAa,d (µl)
TMAb,d (µl)
HDFBSd,e (ml)
1
6/96
−0.6
8.0
60
960
8.98
2 3
6/24 6/6
−0.3 0
7.7 7.4
60 60
240 60
9.7 9.88
4 5
24/6 96/6
0.3 0.6
7.1 6.8
240 960
60 60
9.7 8.98
aButyric acid (from 1 M stock solution; see recipe). bTrimethylamine (from 1 M stock solution; see recipe). cPseudo null pH values are calculated from Eisner’s null equation: pH = pH − 0.5log{[A]/[B]}, where pH = 7.4 (the i e e pH of HDFBS) and [A] and [B] are the concentrations of acid and base, respectively. dVolume of solution required to make 10 ml of 6× concentrated standard. eHDFBS is 10 mM HEPES buffer (see recipe) supplemented with 10% dialyzed fetal bovine serum (APPENDIX 2A).
Pseudo null calibration solutions, 6× Prepare 10 ml of each 6×-concentrated standard solution by adding the appropriate amounts of 1 M butyric acid (see recipe) and 1 M trimethylamine (see recipe) to HDFBS (10 mM HEPES buffer [see recipe] supplemented with 10% dialyzed fetal bovine serum) as shown in Table 9.3.1. This will yield a calibration series of pseudo null pH 6.8 to 8.0 (see Background Information for further discussion). Use sterile labware and appropriate sterile technique to avoid the need for further filter sterilization. Store indefinitely at 4°C. The additions of weak acid and base do not change the external pH because all three solutions were originally titrated to pH 7.4; therefore, pipet carefully.
Trimethylamine, 1 M Add 12.3 ml of 4.06 M trimethylamine (pKa = 9.8; Aldrich) to 30 ml water. Titrate to pH 7.4 with HCl. Add water to 50 ml final volume. Filter sterilize and store up to several months at 4°C. COMMENTARY Background Information General Considerations Regulation of intracellular pH (pHi) is of fundamental biological importance (Busa and Nuccitelli, 1984; Lagarde and Pouyssegur, 1986; Sutherland, 1986; Grinstein et al., 1989). Because many critical macromolecules have isoelectric points close to the physiological range, relatively minor fluctuations in hydrogen ion concentrations may have major effects on the conformations or interactions of proteins and nucleic acids. For this reason, intracellular pH is normally regulated to within a very narrow range of values, and many important biological events, such as cell signaling or the initiation of mitosis, are associated with changes in pHi (Grinstein et al., 1989).
Intracellular pH measurement by flow cytometry is a generally straightforward procedure that can be done using any instrument equipped with a 488-nm argon laser. Ratio measurements can be made directly with some flow cytometers; alternatively, they can be obtained from listmode files using programs such as WinList (Verity Software House). Choice of calibration method The most widely used calibration method is the nigericin technique. Nigericin is an ionophore that allows 1:1 exchange of potassium ions for hydrogen ions against their concentration gradients (Pressman and de Guzman, 1977; Boyer and Hedley, 1994). When cells are suspended in medium whose potassium ion concentration is equal to that inside the cells,
Studies of Cell Function
9.3.7 Current Protocols in Cytometry
Supplement 1
there is no net movement of potassium ions, and hydrogen ions can move freely down their concentration gradient. The pHi is therefore set at the pH of the buffer. A series of calibration standards is made by suspending dye-loaded cells in high-K+ calibration buffers of known pH. Fluorescence is measured and plotted against the set pHe. This method depends on knowing the intracellular potassium concentration. Although this is usually assumed to be ~140 mM, it is not generally measured and significant errors can result. Additional problems with using nigericin are that it can be toxic and that the conditions (e.g., the concentration of and the length of exposure to nigericin) needed to set intracellular pH equal to extracellular pH can vary between cell lines. For these reasons the authors have developed an alternative calibration procedure for flow cytometric pHi measurement that is termed the pseudo null method. This is based on the null point method for calibrating spectrofluorometric pHi measurements proposed by Eisner et al. (1989), which was in turn developed from earlier work by Szatkowski and Thomas (1986). The pseudo null method involves exposing dye-loaded cells to mixtures of weak acids and bases. Only the uncharged forms of the weak acids and bases can diffuse into the cells, where they dissociate in accordance with the Henderson-Hasselbalch equation: pH = pKa + log[A−]/[HA] where A− and HA are the protonated and deprotonated forms of an acid. From this, Szatkowski and Thomas derived the following relationship: pHi = pHe + 0.5{log[BT]/[AT] + log[∆pHa/∆pHb] + ∆pHa − ∆pHb}
Flow Cytometric Measurement of Intracellular pH
where pHi is dependent on the extracellular pH (pHe), the ratio of total concentration of weak base [BT] and weak acid [AT] present, and the changes in pHi caused by adding either the weak acid (∆pHa) or the weak base (∆pHb) alone. Addition of a weak base will produce intracellular alkalosis (acceptance of protons) and a weak acid will produce intracellular acidosis (donation of protons). The simplest situation is where a concentration of weak base and a concentration of weak acid produce an equal but opposite change in pHi (i.e., ∆pHa = ∆pHb); pHi can then be calculated simply from the ratio of the concentration of acid and base, because the ∆pH terms cancel out. If the ∆pH terms are associated with equal changes in fluorescence ratio—i.e., fluorescence is linearly related to
pHi—these observations allow calibration of fluorescence against pHi. Eisner et al. (1989) developed this concept further by determining the mixture of weak acid and base that produces no change in the signal obtained from cells loaded with a fluorescent pH indicator, measured using a spectrofluorometer. This is achieved by exposing the cells to prepared solutions with a range of acid/base ratios. The ratio that gives no change in pHi can be interpolated between those that lead to the smallest increase or decrease in pHi. This “null point” method is particularly useful when the indicator response is not a simple function of pH and is presumably independent of mechanisms that regulate pHi, because pHi does not change at the null point. From the equation above, the value of pHi at the null point is as follows: pHi = pHe − 0.5 log[AT]/[BT] The major disadvantage of the null point method when using flow cytometry is the requirement to observe the direction of change of pHi after adding several mixtures of acid and base in order to obtain a single data point, the true null point. However, it is not necessary to find the molar ratio that gives no change in pHi, as a calibration curve can be derived from “pseudo null” values whose pHi is defined by the previous equation (Chow et al., 1996). This process, the pseudo null calibration method, is rapid, technically simple, and reproducible. Because it is not influenced by the intracellular potassium concentration, it may give a more reliable estimate of the absolute value of pHi.
Critical Parameters There are a number of technical points that need to be remembered in performing flow cytometric measurement of pHi.. Importantly, biologically relevant changes in pHi occur over a much narrower range of values than those for [Ca2+]i, so the demands on the calibration procedure are more stringent. Handling of samples for pHi measurements This is a live cell assay, and appropriate biohazard precautions need to be taken when handling the cells. Regulation of intracellular pH is an active process; therefore, to retain metabolic health, cells should be isolated, stained, and analyzed as quickly as possible. The pHi values obtained depend on resuspension buffer (HEPES or PBS), the presence or absence of 10% dialyzed fetal bovine serum (FBS), and the temperature (room temperature
9.3.8 Supplement 1
Current Protocols in Cytometry
or 37°C; Chow et al., 1996). In general, pHi is lower at 37°C and with HEPES. Addition of FBS aids maintenance of physiological conditions; however, it should be dialyzed to remove bicarbonate, which can potentially influence pHi regulation. Furthermore, serum should be omitted during cell labeling as it can interfere with the loading of esterified probes (see Troubleshooting). Results from the authors’ laboratory suggest that the optimum conditions for measuring absolute pHi values for cells under physiological conditions comprise initial dye loading in serum-free HEPES buffer at 37°C, subsequent resuspension in HEPES buffer supplemented with 10% dialyzed FBS for 10 to 20 min at 37°C, and measurement by pseudo null calibration (see Alternate Protocol 1). Choice of pHi indicator The most widely used dyes for flow cytometric pHi measurements are SNARF-1 and BCECF. Figure 9.3.2 shows pH calibration curves of CCRF-CEM wild-type cells, a human T lymphoblastic cell line, stained with BCECF or SNARF-1 using the nigericin calibration method (see Basic Protocol). For most purposes, SNARF-1 is the indicator of choice. It has a clear isobestic point, has a large dynamic range of the ratio response to pH shifts, and is particularly sensitive to shifts in pHi that are in the physiological range (Wieder et al., 1993). SNARF-1 can be measured using filter combinations suitable for phycoerythrin and a redemitting dye such as PerCP (e.g., at 580 and 640 nm). The older dye BCECF does not have an isobestic point, and although it can be used for ratio measurements due to the greater pH sensitivity of its green emission, it is less sensitive in the physiological range (Musgrove et al., 1987). However, under acidic pHi conditions BCECF is more sensitive than SNARF-1, probably because of its relatively low pKa value. Emissions for BCECF are collected at 525 and 640 nm.
staining is cell-type- and concentration-dependent, it is important to have an accurate cell count. Clinical materials tend to be more difficult to stain. Poor coefficient of variation and resolution suggest insufficient dye uptake. Doubling the amount of stain and/or extending incubation for 10 to 15 min may help.
Anticipated Results There is no universal calibration curve for either the nigericin or the pseudo null method. As calibration is cell-type- and stain-dependent, a calibration curve should be generated for each test sample. Under the conditions given here, the pseudo null method usually produces a well-resolved calibration curve, whereas the nigericin method is less reliable. Because pHi is a influenced by pH values in intracellular compartments, no absolute measures of pHi can be made without perturbing the cell. Therefore, there will be greater confidence in measuring changes in pHi than in determining an absolute value. With these reservations, and using HDFBS at 37°C as described in Alternate Protocol 1, the authors have observed pHi values of up to 7.6 for rapidly growing cell lines and between 7.2 and 7.4 for blasts freshly obtained from leukemia patients (Chow et al., 1996). SNARF-1 is a more sensitive indicator than BCECF in the physiological range. Figure 9.3.2 shows that for the one pH unit from 7.0 to 8.0, SNARF-1 gives ~300 channels of resolution with the nigericin method, whereas BCECF gives half that number. The pseudo null method gives similar results. A pH calibration curve is usually smooth. In Figure 9.3.2, the mean fluorescence ratio of the pH 7.4 calibration point for both curves is lower than expected, indicating that this calibration solution is “off.” When this occurs with the nigericin method, the pH of the particular high-K+ buffer should be measured and readjusted; for the pseudo null method, it is advisable to make a fresh standard solution, paying close attention to the amounts of weak acid and base added.
Troubleshooting As with most esterified probes, SNARF-1 and BCECF may be difficult to load in the presence of fetal bovine serum because this often contains significant esterase activity. An additional problem in dye loading can occur using cells that express the P-glycoprotein (Pgp) efflux pump, because the hydrolyzed forms of both dyes are Pgp substrates. This can be circumvented by using a Pgp-inhibiting agent such as verapamil or cyclosporin A. As
Time Considerations Staining requires ~45 min; the calibration series should be prepared during this period. Instrument alignment and cytosetting optimization take ~15 min, each sample is acquired in 1 to 2 min, and analysis takes ~15 min. Intracellular pH of a test sample can thus be determined in ~2 hr or less after preparation of single cells. Studies of Cell Function
9.3.9 Current Protocols in Cytometry
Boyer, M.J. and Hedley, D.W. 1994. Measurement of intracellular pH. Methods Cell Biol. 41:135149.
Musgrove, E.A., Rugg, C.A., and Hedley, D.W. 1987. Flow cytometric measurement of intracellular pH: A critical evaluation of available fluorochromes. Cytometry 7:347-355.
Busa, W.B. and Nuccitelli, R. 1984. Metabolic regulation via intracellular pH. Am. J. Physiol. 246:R409-R438.
Pressman, B.C. and de Guzman, N.T. 1977. Biological applications and evolutionary origins of ionophores. Adv. Exp. Med. Biol. 84:285-300.
Chow, S., Hedley, D.W., and Tannock, I.F. 1996. Flow cytometric calibration of intracellular pH measurements in viable cells using mixtures of weak acids and bases. Cytometry 24:360-367.
Sutherland, R. 1986. Importance of critical metabolites and cellular interactions in the biology of microregions of tumors. Cancer 58:1668-1680.
Literature Cited
Eisner, D.A., Kenning, N.A., O’Neill, S.C., Pocock, G., Richards, C.D., and Valdeolmillos, M.A. 1989. A novel method for absolute calibration of intracellular pH indicators. Pflügers Archiv. Eur. J. Physiol. 413:553-558. Grinstein, S., Rotin, D., and Mason, M.J. 1989. Na+/H+ exchange and growth factor–induced cytosolic pH changes. Role in cellular proliferation. Biochim. Biophys. Acta 988:73-97. Lagarde, A.E. and Pouyssegur, J.M. 1986. Mini Review: The Na+:H+ antiport in cancer. Cancer Biochem. Biophys. 9:1-14.
Szatkowski, M.S. and Thomas, R.C. 1986. New method for calibrating pHi from accurately measured changes in pHi induced by a weak acid and base. Pflügers Archiv. Eur. J. Physiol. 407:59-63. Wieder, E.D., Hang, H., and Fox, M.H. 1993. Measurement of intracellular pH using flow cytometry with carboxy-SNARF-1. Cytometry 14:916921.
Contributed by Sue Chow and David Hedley Ontario Cancer Institute and Princess Margaret Hospital Toronto, Ontario, Canada
Flow Cytometric Measurement of Intracellular pH
9.3.10 Current Protocols in Cytometry
Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
UNIT 9.4
Functional analysis of cellular organelles can be accomplished by staining cells with suitable organelle-specific dyes and then analyzing the fluorescence of the stained cells with a flow cytometer. With this methodology it is possible to resolve suspected heterogeneity in organelle function or content within a population of cells. Flow cytometry does not provide morphological information; if that is desired, quantitative microscopy—using a video microscope with digital image analysis system, or a confocal microscope—should be employed. MITOCHONDRIA Mitochondria are the organelles that generate most of the energy needed for a cell to function. The distinguishing features of mitochondria include a high concentration of NADH, a specific membrane lipid (i.e., cardiolipin), and a negative inside membrane potential. NADH can be detected as blue autofluorescence after excitation with UV light. Cardiolipin can be specifically stained with nonyl acridine orange (NAO). Because of their negative inside membrane potential, mitochondria accumulate fluorescent cations such as rhodamine 123, JC-1, and MitoTracker CMXRos dyes. A second group of mitochondrial stains accumulate in mitochondria by a mechanism that is independent of the physiological state of the organelle. The amount of fluorescence obtained with these dyes, therefore, may reflect the amount of mitochondrial material in a cell. A third group consists of reduced dyes that become fluorescent after oxidation inside the mitochondrion. See Commentary for discussion of specific mitochondrial stains. Five protocols for mitochondrial analysis by flow cytometry are presented here for: analysis of NADH levels (see Basic Protocol 1); analysis of cardiolipin levels (see Basic Protocol 2); analysis of the normalized mitochondrial membrane potential based on MitoTracker Green FM and MitoTracker CMXRosamine staining (see Basic Protocol 3); analysis of mitochondrial oxidative turnover by H2-CMXRosamine staining (see Basic Protocol 4); and analysis of leakage of oxidant from mitochondria (see Basic Protocol 5). A sixth protocol describes analysis of mitochondrial mass or function by microscopy (see Basic Protocol 6). A procedure for mitochondrial staining followed by cell fixation is given in Alternate Protocol 1. LYSOSOMES Lysosomes are organelles devoted to degradation of cellular macromolecules and subsequent recycling of their components. The degradative enzymes of lysosomes all operate at an acidic pH. These features offer two ways of specifically staining lysosomes. First, basic fluorophores that are plasma membrane permeant will accumulate inside lysosomes. That means that the amount of fluorescence obtained with a plasma membrane–permeant base will be related to the amount of lysosomal mass in a cell. Second, the enzymatic activity of lysosomes can be monitored using nonfluorescent dyes containing a chemical group that can be removed by an acidic hydrolase to generate a fluorophore. Procedures for assessing the amount of lysosomal mass in a cell using LysoTracker dyes, which are plasma membrane–permeant bases (see Basic Protocol 7), and for staining cells with a fluorogenic lysosomal enzyme substrate (see Alternate Protocol 2) are described. As staining is dependent upon lysomal activity, fixation is not possible. Studies of Cell Function Contributed by Martin Poot Current Protocols in Cytometry (2000) 9.4.1-9.4.24 Copyright © 2000 by John Wiley & Sons, Inc.
9.4.1 Supplement 14
GOLGI APPARATUS The Golgi apparatus is involved in a highly regulated traffic of lipids such as ceramides. This characteristic offers a way to specifically label the Golgi apparatus. By N-acylation of sphingosine with a lipophilic fluorochrome, a family of fluorescent probes for the Golgi apparatus has been created. Procedures for fluorescent labeling of the Golgi apparatus in living cells (see Basic Protocol 8) and fixed cells (see Alternate Protocol 3) are presented. These protocols have been used for microscopic analysis of cells grown on coverslips, but may also be amenable to modification for analysis of cell suspensions by flow cytometry. ENDOPLASMIC RETICULUM The endoplasmic reticulum is an elaborate intracellular network that serves as a scaffold for cellular functions such as protein and lipid synthesis. In addition, the endoplasmic reticulum is a major intracellular calcium store. Because the endoplasmic reticulum consists of a network of membranes, it can be stained with any highly lipophilic dye. Methods for fluorescent labeling of the endoplasmic reticulum (see Basic Protocol 9) and for labeling of fixed cells (see Alternate Protocol 4) are described. These protocols have been used for microscopic analysis of cells grown on coverslips. NOTE: To maintain cells in a physiologically optimal state and to avoid photochemical damage to stained cells, all incubations are performed at 37°C under subdued light in fresh culture medium. Unless stated otherwise, all solutions should be prewarmed to 37°C to conserve cell morphology and function. NOTE: All protocols described have been performed on cultured animal cells; limited data exist regarding the use of these methods in plant cells and yeast. Expertise is assumed for basic techniques in flow cytometry as well as in cell culture and harvesting (of both suspension cultures and those grown adherent on coverslips; APPENDIX 3B), immunocytochemistry (Watkins, 1989), and fluorescence microscopy (UNIT 2.4). CAUTION: Dimethylsulfoxide (DMSO) and dye solutions are toxic to humans. Use (nitrile) gloves and wear eye protection at all stages of handling. Seek medical advice if dye or dye solution is ingested or inhaled. The dyes mentioned are for in vitro use only; do not administer either externally or internally. TO DISPOSE: All staining solutions should be poured through a funnel with a filter containing activated charcoal. When the solution passing through becomes fluorescent, the filter should be incinerated or disposed of according to applicable rules for environmental hygiene and a fresh filter installed. BASIC PROTOCOL 1
ANALYSIS OF NAD(P)H LEVELS IN LIVE CELL MITOCHONDRIA BY FLOW CYTOMETRY After excitation with UV light (∼360 nm), NADH and NADPH emit blue fluorescence. UV-excited blue autofluorescence as detected in individual cells by flow cytometry has been shown to reflect cellular NAD(P)H levels (Thorell, 1983). Materials Cells in suspension (see APPENDIX 3B) Cell culture medium supplemented with 10% FBS (APPENDIX 2A), 37°C
Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
15-ml screw-cap centrifuge tubes 37°C water bath with cover Flow cytometer with mercury arc lamp or argon laser (tuned to 360 nm) as excitation source
9.4.2 Supplement 14
Current Protocols in Cytometry
12 × 75–mm polypropylene test tubes suitable for the flow cytometer Computer for data collection and processing Additional reagents and equipment for cell culture and harvesting (APPENDIX 3B) 1. Harvest cultured cells by standard procedures (APPENDIX 3B) in 15-ml screw-cap centrifuge tubes and centrifuge 5 min at 200 × g, room temperature. 2. Resuspend the pellet at 0.5 to 1.0 × 106 cells/ml in prewarmed cell culture medium with 10% FBS. Place the tubes in a 37°C water bath for ≥5 min. Since the functional state of the mitochondria is to be monitored, it is advisable to keep cell suspensions at their optimal temperature (i.e., 37°C) and to allow them to recover for a brief moment after harvesting.
3. Set up and optimize the flow cytometer. Carefully resuspend the cell sample by gently pipetting up and down a few times immediately before analysis and transfer to a 12 × 75–mm polypropylene test tube. Ultraviolet (360 nm)-excited blue autofluorescence (collected with a 450-nm-centered band-pass filter) is proportional to cellular NAD(P)H content. Due to the wide variation in cellular contents of mitochondria, it is advisable to use logarithmic signal amplification for the signal channels collecting mitochondria-related fluorescence. During harvesting, cells tend to clump; to obtain meaningful data on a per-cell basis it is essential to resuspend cells immediately before analysis.
4. Run sample and collect data. ANALYSIS OF CARDIOLIPIN LEVELS IN LIVE CELLS Nonyl acridine orange (NAO) specifically stains the mitochondrial membrane lipid, cardiolipin. NAO emits a broad range of fluorescent light, but its red emission (i.e., ~640 nm) appears to reflect most accurately the cardiolipin content of mitochondrial membranes.
BASIC PROTOCOL 2
Materials Cells in suspension (see APPENDIX 3B) Cell culture medium supplemented with 10% FBS (APPENDIX 2A), 37°C 1 mM nonyl acridine orange dye stock solution (NAO; Table 9.4.1) in DMSO; store at −20°C in the dark 15-ml screw-capped centrifuge tubes 37°C water bath with cover 12 × 75–mm polypropylene test tubes suitable for the flow cytometer Flow cytometer with mercury arc lamp or argon laser (tuned to 488 nm) as excitation source Band-pass filter centered at ~530 nm 630-nm long-pass filter Computer for data processing and collection Additional reagents and equipment for cell culture and harvesting (APPENDIX 3B) 1. Harvest cultured cells by standard procedures (APPENDIX 3B) in 15-ml screw-capped centrifuge tubes and centrifuge 5 min at 200 × g, room temperature. 2. Resuspend the pellet at 0.5 to 1.0 × 106 cells/ml in prewarmed cell culture medium. Place the tubes in a 37°C water bath for ≥5 min. Since the functional state of the mitochondria is to be monitored, it is advisable to keep cell suspensions at their optimal temperature (i.e., 37°C) and to allow them to recover for a brief moment after harvesting.
Studies of Cell Function
9.4.3 Current Protocols in Cytometry
Supplement 14
3. Thaw 1 mM NAO dye solution at room temperature, keeping it protected from light (e.g., in a drawer). Dye solutions decompose rapidly if exposed to light.
4. Pipet 1-ml aliquots of pre-warmed cell suspension (from step 2) into 12 × 75–mm polypropylene tubes. Add 1 µl 1 mM NAO dye solution to each aliquot. Mix immediately by briefly vortexing at maximal speed. Incubate 15 to 30 min at 37°C in the dark or under subdued light. After staining, put tubes with cell suspensions in a melting ice bath. During the staining period cells tend to clump; to obtain meaningful data on a per-cell basis it is essential to resuspend cells by gently pipetting up and down a few times immediately before analysis.
5. Set up and optimize the flow cytometer. Excite NAO-stained samples at 488 nm and collect NAO fluorescence using a band-pass filter centered at ∼530 nm and a 630-nm long-pass filter. Due to the wide variation in cellular contents of mitochondria, it is advisable to use logarithmic signal amplification for the signal channels collecting mitochondria-related fluorescence Cells stained with NAO show a broad fluorescence emission; the ratio of red (above 630 nm) to green (∼530 nm) fluorescence appears to best reflect cardiolipin levels (see Commentary). BASIC PROTOCOL 3
ANALYSIS OF NORMALIZED MITOCHONDRIAL MEMBRANE POTENTIAL MitoTracker Red CMXRosamine fluorescence is sensitive to changes in mitochondrial membrane potential, whereas MitoTracker Green FM fluorescence stains aldehyde-fixed cells, in which the mitochondrial membrane potential has been lost. Combination of MitoTracker Green FM and CMXRosamine allows better resolution between apoptotic and “normal” cells than does staining with CMXRosamine alone. This protocol requires basic expertise in cell handling, microscopy, and flow cytometry. Materials Cells in suspension (see APPENDIX 3B) Cell culture medium supplemented with 10% FBS (APPENDIX 2A), 37°C 200 µM MitoTracker Green FM (Table 9.4.1) dye stock solution in DMSO; store at −20°C in the dark 200 µM CMXRosamine (Table 9.4.1) dye stock solution in DMSO; store at −20°C in the dark 15-ml screw-cap centrifuge tubes 37°C water bath with cover 12 × 75–mm polypropylene test tubes suitable for the flow cytometer Flow cytometer with mercury arc lamp or argon laser (tuned to 488 nm) as excitation source Band-pass filter centered at ~530 nm 630-nm long-pass filter Computer for data collection and processing Additional reagents and equipment for cell culture and harvesting (APPENDIX 3B)
Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
1. Harvest cultured cells by standard procedures (APPENDIX 3B) in 15-ml screw-capped centrifuge tubes and centrifuge 5 min at 200 × g, room temperature. 2. Resuspend the pellet at 0.5 to 1.0 × 106 cells/ml in prewarmed cell culture medium. Place the tubes in a 37°C water bath for ≥5 min.
9.4.4 Supplement 14
Current Protocols in Cytometry
Table 9.4.1
Key Features of Mitochondrial Dyesa
Excitation Emission maximum maximum (nm) (nm)
Dye
Excitation sourceb (nm)
Collection Sensitivity wavelengthc toward functional state (nm)
Rhodamine 123
506
530
488 (argon)
530
+
MitoTracker Red CMXRos
594
608
610
+
MitoTracker Green FM MitoFluor
480
516
488, 514 (argon) 543, 594 (HeNe) 488 (argon)
530
480
516
488 (argon)
497
519
488 (argon)
Nonyl acridine orange
Fixability Photostabilityd with aldehydes −
−
−
+
+
530
−
−
+
530
−
−
+
aFor further discussion, see Commentary. bOutput wavelengths of most commonly used flow cytometry lasers that are compatible with the excitation spectra of mitochondrial dyes. cFor flow cytometry, band-pass filters should be centered around the indicated wavelength. d+, photographable with ordinary skill; −, difficult to photograph.
Since the functional state of the mitochondria is to be monitored, it is advisable to keep cell suspensions at their optimal temperature (i.e., 37°C) and to allow them to recover for a brief moment after harvesting.
3. Thaw the dye solutions at room temperature, keeping them protected from light (e.g., in a drawer). Dye solutions decompose rapidly if exposed to light.
4. Pipet 1-ml aliquots of cell suspension into 12 × 75–mm polypropylene tubes and keep warm. Add 1 µl each of the 200 µM MitoTracker Green FM and 200 µM CMXRosamine dye stock solutions to the cell suspension. Mix immediately by briefly vortexing at maximal speed. Incubate 15 to 30 min at 37°C in the dark or under subdued light. After staining, put tubes with cell suspensions in a melting ice bath. During the staining period cells tend to clump; to obtain meaningful data on a per-cell basis it is essential to resuspend cell samples by gently pipetting up and down a few times immediately before analysis. Dye concentrations in the range of 100 to 200 nM are recommended, since at higher concentrations nonmitochondrial staining may occur.
5. Set up and optimize the flow cytometer. Excite MitoTracker Green FM– and CMXRosamine-stained samples at 488 nm. Collect fluorescence from MitoTracker Green FM using a band-pass filter centered at ∼530 nm; for CMXRosamine use a 630-nm long-pass filter. Due to the wide variation in cellular contents of mitochondria, it is advisable to use logarithmic signal amplification for the signal channels collecting mitochondria-related fluorescence. Cells stained with MitoTracker Green FM show maximal emission at 516 nm; cells stained with CMXRosamine show maximal absorption at 594 nm and maximal emission at 608 nm; they also exhibit significant absorption in the UV region of the spectrum and may be excitable with a mercury arc lamp.
Studies of Cell Function
9.4.5 Current Protocols in Cytometry
Supplement 14
BASIC PROTOCOL 4
ANALYSIS OF MITOCHONDRIAL OXIDATIVE TURNOVER IN LIVE CELLS This protocol uses the fact that H2-CMXRosamine becomes fluorescent in proportion to the mitochondrial oxidative turnover rate. This protocol requires basic expertise in cell handling, microscopy, and flow cytometry. Materials Cells in suspension (see APPENDIX 3B) Cell culture medium supplemented with 10% FBS (APPENDIX 2A), 37°C 200 µM H2-CMXRosamine (Table 9.4.1) dye stock solution in DMSO; store at −20°C in the dark 15-ml screw-cap centrifuge tubes 37°C water bath with cover 12 × 75–mm polypropylene test tubes suitable for the flow cytometer Flow cytometer with mercury arc lamp or argon laser (tuned to 488 nm) as excitation source 630-nm long-pass filter Computer for data collection and processing Additional reagents and equipment for cell culture and harvesting (APPENDIX 3B) 1. Harvest cultured cells by standard procedures (APPENDIX 3B) in 15-ml screw-cap centrifuge tubes and centrifuge 5 min at 200 × g, room temperature. 2. Resuspend the pellet at 0.5 to 1.0 × 106 cells/ml in prewarmed cell culture medium. Place the tubes in a 37°C water bath, for ≥5 min. Since the functional state of the mitochondria is to be monitored, it is advisable to keep cell suspensions at their optimal temperature (i.e., 37°C) and to allow them to recover for a brief moment after harvesting.
3. Thaw dye solution at room temperature, keeping it protected from light (e.g., in a drawer). Dye solutions decompose rapidly if exposed to light.
4. Pipet 1-ml aliquots of cell suspension into 12 × 75–mm polypropylene tubes and keep warm. Add 1 µl 200 µM H2-CMXRosamine stock solution to cell suspension. Mix immediately by briefly vortexing at maximal speed and incubate 15 to 30 min at 37°C in the dark or under subdued light. After staining, put tubes with cell suspensions in a melting ice bath. Dye concentrations in the range of 100 to 200 nM are recommended, since at higher concentrations nonmitochondrial staining may occur. During the staining period cells tend to clump; to obtain meaningful data on a per-cell basis it is essential to resuspend cells by pipetting up and down a few times immediately before analysis.
5. Set up and optimize the flow cytometer. Excite H2-CMXRosamine-stained samples at 488 nm and collect H2-CMXRos fluorescence with a 630-nm long-pass filter. Due to the wide variation in cellular contents of mitochondria, it is advisable to use logarithmic signal amplification for the signal channels collecting mitochondria-related fluorescence. Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
Cells stained with H2-CMXRosamine show maximal absorption at 594 nm and maximal emission at 608 nm; they also exhibit significant absorption in the UV region of the spectrum and may be excitable with a mercury arc lamp.
9.4.6 Supplement 14
Current Protocols in Cytometry
LEAKAGE OF OXIDANT FROM MITOCHONDRIA This protocol uses the fact that 5-(and-6)-carboxy-2′,7′-dichlorodihydrofluorescein diacetate is a nonfluorescent, electronically neutral molecule that is readily taken up by intact cells, after which the acetates are cleaved. Thus, a negatively charged molecule is formed that will be repelled by mitochondria, which have a negative inside membrane potential. Upon oxidation, the 5-(and-6)-carboxy-2′,7′-dichlorodihydrofluorescein becomes the fluorescent dye 5-(and-6)-carboxy-2′,7′-dichlorofluorescein. This protocol requires basic expertise in cell handling, microscopy, and flow cytometry.
BASIC PROTOCOL 5
Materials Cells in suspension (see APPENDIX 3B) Cell culture medium supplemented with 10% FBS (APPENDIX 2A), 37°C 10 mM 5-(and-6)-carboxy-2′,7′-dichlorodihydrofluorescein diacetate stock solution in PBS; store at −20°C in the dark 15-ml screw-cap centrifuge tubes 37°C water bath with cover 12 × 75–mm polypropylene test tubes suitable for the flow cytometer Flow cytometer with mercury arc lamp or argon laser (tuned to 488 nm) as excitation source Band-pass filter centered at ~530 nm Computer for data collection and processing Additional reagents and equipment for cell culture and harvesting (APPENDIX 3B) 1. Harvest cultured cells by standard procedures (APPENDIX 3B) in 15-ml screw-capped centrifuge tubes and centrifuge 5 min at 200 × g, room temperature. 2. Resuspend the cell pellet at 0.5 to 1.0 × 106 cells/ml in prewarmed cell culture medium. Place the tubes in a 37°C water bath for ≥5 min. Since the functional state of the mitochondria is to be monitored, it is advisable to keep cell suspensions at their optimal temperature (i.e., 37°C) and to allow them to recover for a brief moment after harvesting.
3. Thaw dye solution at room temperature, keeping it protected from light (e.g., in a drawer). Dye solutions decompose rapidly if exposed to light.
4. Pipet 1-ml cell suspension aliquots into 12 × 75–mm polypropylene tubes and keep warm. Add 5 µl 10 mM 5-(and-6)-carboxy-2′,7′-dichlorodihydrofluorescein diacetate to cell suspension. Mix immediately by briefly vortexing at maximal speed and incubate 15 to 30 min at 37°C in the dark or under subdued light. After staining, put tubes with cell suspensions in a melting ice bath. During the staining period cells tend to clump; to obtain meaningful data on a per-cell basis it is essential to resuspend cells by gently pipetting up and down a few times immediately before analysis. Dye concentrations in the range of 50 ìM give optimal sensitivity.
5. Set up and optimize the flow cytometer. Excite 5-(and-6)-carboxy-2′,7′-dichlorofluorescein at 488 nm and collect fluorescence with a band-pass filter centered at ∼530 nm. Due to the wide variation in cellular contents of mitochondria, it is advisable to use logarithmic signal amplification for the signal channels collecting mitochondria-related fluorescence.
Studies of Cell Function
9.4.7 Current Protocols in Cytometry
Supplement 14
Cells stained with 5-(and-6)-carboxy-2′,7′-dichlorodihydrofluorescein diacetate show maximal absorption at 488 nm and maximal emission at 530 nm. This dye is excitable either by an argon laser tuned to 488-nm excitation or by a mercury arc lamp. BASIC PROTOCOL 6
STAINING OF LIVE CELLS FOR MEASUREMENT OF MITOCHONDRIAL MASS OR FUNCTION BY MICROSCOPY This protocol presents a general method for staining cultured mammalian cells with mitochondrion-specific dyes. The cells are grown on coverslips for subsequent analysis with microscopes equipped with appropriate filters (Table 9.4.1). This protocol requires basic expertise in cell handling, microscopy, and flow cytometry. Materials Cells grown in monolayer (APPENDIX 3B) on coverslips Cell culture medium supplemented with 10% FBS (APPENDIX 2A), 37°C 9:1 (v/v) HBSS/FBS (freshly prepared; see APPENDIX 2A for HBSS recipe) Melted wax or nail polish Mitochondrial dye stock solutions (see recipe): 0.5 mM stock solutions of rhodamine 123, dihydrorhodamine 123, MitoTracker Green FM, MitoFluor, nonyl acridine orange, or MitoTracker Red (CMXRos or CMXRos-H2) 15 mM propidium iodide (aqueous; stable >1 year at 4°C) or 50 µM SYTOX Green working solution (prepare immediately before use from purchased stock solution) 18 × 18–mm coverslips sterilized by dipping into absolute ethanol and subsequent flaming 35-mm cell culture dishes Fluorescence microscope (UNIT 2.4) Computer for data collection and processing Additional reagents and equipment for cell culture and harvesting (APPENDIX 3B) CAUTION: Due to the potential mutagenicity of propidium iodide and SYTOX Green (and other nucleic acid stains), it is preferable to purchase a concentrated stock rather than to prepare one. 1. Culture adherent cells overnight on sterilized 18 × 18–mm coverslips. Alternatively, larger coverslips and culture dishes can be used; in this case the volumes of the staining and other solutions should be adjusted accordingly. This protocol assumes that 35-mm dishes are used throughout.
2. Take coverslip with cells out of the cell culture dish, rinse once with 37°C cell culture medium, and place in a 35-mm cell culture dish with 1 ml prewarmed cell culture medium. 3. Thaw the dye solutions at room temperature, keeping them protected from light (e.g., in a drawer). Dye solutions decompose rapidly if exposed to light.
4. Add 1 µl 0.5 mM mitochondrial dye stock solution and swirl immediately to distribute the dye solution evenly.
Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
5. Incubate 15 to 30 min at 37°C in the dark or in subdued light. After staining, briefly rinse the coverslip three times with prewarmed 9:1 (v/v) HBSS/FBS. At this stage, samples can be costained ≥15 min at room temperature with a DNA dye: 5 ìM propidium iodide (final) if rhodamine 123, MitoTracker Green, MitoFluor, or nonyl acridine orange is used (final), or 0.5 ìM SYTOX Green (final) if CMXRos or CMXRos-H2
9.4.8 Supplement 14
Current Protocols in Cytometry
is used, to distinguish dead cells; however, these can generally be distinguished by morphology.
6. Invert coverslip and mount onto a slide in 9:1 HBSS/FBS, leaving some clearance. Do not apply pressure. Seal the coverslip by a method regularly used in the laboratory (e.g., melted wax or nail polish). Proceed to fluorescence microscopy. To observe cells stained with rhodamine 123, MitoTracker Green FM, MitoFluor, or nonyl acridine orange by fluorescence microscopy, use fluorescein excitation and emission filters (ex. 465 to 495 nm, em. 515 to 555 nm). Cells stained with CMXRos are best observed with Texas Red excitation and emission filters (ex. 540 to 580 nm, em. 600 to 660 nm). Use a band-pass filter centered around 620 nm for propidium iodide, or around 530 nm for SYTOX Green.
MITOCHONDRIAL STAINING WITH CELL FIXATION AND SECONDARY LABELING
ALTERNATE PROTOCOL 1
This protocol describes fixation of cells that have been labeled with a fixable mitochondrial dye (see Table 9.4.1). Additionally, cells stained by this method can be secondarily labeled with an antibody of choice. These methods can be used for flow cytometry or microscopy, and require basic expertise in those areas as well as in cell handing. Additional Materials (also see Basic Protocols 1 to 6) 3.7% formaldehyde (Fluka) in PBS (prepared fresh), 37°C. PBS (APPENDIX 2A), 37°C Anhydrous acetone (Sigma), −20°C Additional reagents and equipment for immunocytochemistry (Watkins, 1989) Fixing Cells for Flow Cytometry 1a. Stain harvested cells in suspension with 1 µl 0.5 mM fixable mitochondrial dye per ml (see Table 9.4.1). Incubate 15 to 30 min at 37°C in the dark or under subdued light, then place in a melting ice bath. 2a. Centrifuge cells 5 min at 200 × g, room temperature. Discard most of the staining solution and resuspend the cell pellets in the residual solution by gently tapping. 3a. Add 5 to 10 ml of 37°C 3.7% formaldehyde/PBS dropwise while vortexing at medium speed. Incubate ≥10 min at room temperature. Formaldehyde is recommended because glutaraldehyde requires quenching of autofluorescence with NaBH4, which affects mitochondrial morphology. Dropwise addition and vortexing of fixative minimize the formation of cell clumps.
4a. Centrifuge fixed cells 5 min at 200 × g, room temperature. 5a. Discard most of the fixative and resuspend the cell pellet in a small amount of residual fixative by gently tapping. Add 1 ml PBS. After fixation, PBS may be at room temperature. At this stage, cells can be either analyzed directly by flow cytometry or counterstained.
6. To permeabilize for immunocytochemistry (optional): Add 5 to 10 ml of −20°C acetone and incubate ≥15 min at room temperature. Centrifuge 5 min at 200 × g, room temperature. Discard most of the acetone and add 5 to 10 ml PBS; repeat incubation and centrifugation. Discard most of the PBS and resuspend the cell pellet in the residual PBS by gently tapping. Studies of Cell Function
9.4.9 Current Protocols in Cytometry
Supplement 14
At this stage, cell suspensions can be treated with blocking solution and incubated with the antibody of choice. For a discussion of immunocytochemistry principles and techniques, refer to Watkins (1989). To give antibodies full access to intracellular epitopes, it is necessary to permeabilize all cell membranes. Among possible permeabilizing agents, acetone has been found to preserve cellular morphology best.
Fixing Cells for Microscopy 1b. Stain cells on coverslips with a fixable mitochondrial dye (see Basic Protocol 6, steps 1 to 5, and Table 9.4.1). 2b. Add 1 ml of 37°C 3.7% formaldehyde/PBS dropwise to the staining solution already on the coverslips. Alternatively, pour off the staining solution and replace with 37°C PBS, then replace PBS with 1 ml 37°C 3.7% formaldehyde/PBS. Formaldehyde is recommended because use of glutaraldehyde necessitates quenching of autofluorescence with NaBH4, which affects mitochondrial morphology. It is critical to add prewarmed fixative to the cells in order to fully conserve their morphology. Addition of cold or room temperature fixative may cause cells to “wrinkle.”
3b. Incubate ≥15 min at room temperature. 4b. After fixation, pour off the formaldehyde solution. View cells directly in a drop of 9:1 HBSS/FBS (see Basic Protocol 1, step 6b), or wash twice with PBS at room temperature and proceed with secondary labeling (step 5b, below). Fixed and stained slides can be stored at 4°C short term (1 to 2 days) for later analysis. Longer-term stability of the label has not been determined.
5b. To permeabilize for immunocytochemistry (optional): Replace PBS with 1 ml of −20°C acetone and incubate ≥15 min at room temperature. Wash the coverslip three times with room temperature PBS. At this stage, coverslips can be treated with blocking solution and incubated with the antibody of choice. For a discussion of immunocytochemistry principles and techniques, refer to Watkins (1989). To give antibodies full access to intracellular epitopes, it is necessary to permeabilize all cell membranes. Among possible permeabilizing agents, acetone has been found to preserve cellular morphology best. BASIC PROTOCOL 7
STAINING OF LIVE CELLS FOR MEASUREMENT OF LYSOSOMAL MASS OR FUNCTION BY FLOW CYTOMETRY OR MICROSCOPY This protocol presents two methods for staining cultured mammalian cells with fluorescent plasma membrane–permeant bases: one performed with cells in suspension, suitable for subsequent analysis by flow cytometry, and one performed with cells on coverslips, suitable for subsequent microscopic analysis. The protocols can be adapted to staining of plant and yeast cells. Samples can be analyzed with flow cytometers equipped with an excitation light source matching the dye excitation wavelength range (see Table 9.4.2) or with microscopes equipped with appropriate filters. This protocol requires basic expertise in cell handling, microscopy, and flow cytometry.
Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
Materials Cells in suspension (APPENDIX 3B; for flow cytometry procedure) or grown in monolayer (APPENDIX 3B) on coverslips (for microscopy procedure) Cell culture medium supplemented with 10% FBS (APPENDIX 2A), 37°C
9.4.10 Supplement 14
Current Protocols in Cytometry
Table 9.4.2
Spectral Properties of Lysosomal Dyes
Excitation Emission maximum (nm) maximum (nm)
Dye
Excitation sourcea (nm)
Collection wavelengthb (nm)
LysoTracker Blue
376
422
360 (argon, mercury arc lamp)
420
LysoTracker Green
501
511
488 (argon)
510
LysoTracker Yellow
528
551
488, 514 (argon), 543 (HeNe)
550
LysoTracker Red
577
592
543 (HeNe)
590
FDG
490
530
488 (argon)
530
aOutput wavelengths of most commonly used flow cytometry lasers that are compatible with the excitation spectra of
LysoTracker dyes and FDG. bFor flow cytometry, band-pass filter should be centered around the indicated wavelength.
100 × LysoTracker Blue, Green, Yellow, or Red dye working solution (see recipe) 15 mM propidium iodide (aqueous; stable >1 year at 4°C) or 50 µM SYTOX Green working solution (prepare immediately before use from purchased stock solution) 9:1 (v/v) HBSS/FBS, 37°C (freshly prepared, for microscopy; see APPENDIX 2A for HBSS recipe) Melted wax or nail polish (for microscopy) 15-ml screw-cap centrifuge tubes and 12 × 75–mm polypropylene tubes (for flow cytometry) 18 × 18–mm coverslips sterilized by dipping into absolute ethanol and subsequent flaming (for microscopy) 35-mm cell culture dishes (for microscopy) Flow cytometer with either a mercury arc lamp, an argon-ion laser, or a HeNe laser as excitation source, or fluorescence microscope (UNIT 2.4) Computer for data collection and processing Additional reagents and equipment for cell culture and harvesting (APPENDIX 3B) CAUTION: Due to the potential mutagenicity of propidium iodide and SYTOX Green (and other nucleic acid stains), it is preferable to purchase a concentrated stock rather than to prepare one. Staining for Flow Cytometry 1a. Harvest cultured cells by standard procedures into 15-ml screw-cap centrifuge tubes and centrifuge 5 min at 200 × g, room temperature. 2a. Resuspend the cell pellet at 0.5–1.0 × 106 cells/ml in 37°C cell culture medium. Leave cell suspension at 37°C for ≥5 min. Because the functional state of the lysosomes is to be monitored, it is advisable to keep cell suspensions at their optimal temperature (37°C) and allow them to recover for a brief moment after harvesting.
3a. Thaw LysoTracker dye working solution at room temperature, keeping it protected from light (e.g., in a drawer). Dye solutions decompose rapidly if exposed to light. Do not refreeze thawed solutions, because the dyes decompose during freeze-thaw cycles.
Studies of Cell Function
9.4.11 Current Protocols in Cytometry
Supplement 14
4a. Divide cell suspension into 1-ml aliquots in 12 × 75–mm polypropylene tubes, add 10 µl dye working solution, and mix immediately by briefly vortexing at maximum speed. Dye concentrations in the range of 0.05 to 0.1 ìM are recommended, because nonlysosomal staining may occur at higher concentrations.
5a. Incubate 15 to 30 min at 37°C in the dark or in subdued light. After staining, place tube in a melting ice bath. 6a. Optional: Exclude dead cells by costaining ≥15 min at room temperature with 5 µM propidium iodide (final) if LysoTracker Blue, Green, or Yellow is used or 0.5 µM SYTOX Green (final) if LysoTracker Red is used. 7a. Set up and optimize the flow cytometer (see Table 9.4.2). Because of the wide variation in cellular lysosome content, use logarithmic signal amplification for the signal channels collecting lysosome-related fluorescence. Use a band-pass filter centered around 620 nm for propidium iodide or around 530 nm for SYTOX Green.
8a. Carefully resuspend the cell sample by gently pipetting up and down a few times immediately before analysis. Cells tend to clump during staining; to obtain meaningful data on a per-cell basis, it is essential to resuspend cells immediately before analysis.
Staining for Microscopy 1b. Culture adherent cells overnight on sterilized 18 × 18–mm coverslips. Alternatively, larger coverslips and culture dishes can be used; in this case the volumes of the staining and other solutions should be adjusted accordingly. This protocol assumes that 35-mm dishes are used throughout.
2b. Take coverslip with cells out of the cell culture dish, rinse once with 37°C cell culture medium, and place in a 35-mm cell culture dish with 1 ml of 37°C cell culture medium. 3b. Thaw LysoTracker dye working solution at room temperature, keeping it protected from light (e.g., in a drawer). Dye solutions decompose rapidly if exposed to light. Do not refreeze thawed solutions, because the dyes decompose during freeze-thaw cycles.
4b. Add 10 µl dye working solution to the dish and swirl immediately to distribute the dye evenly. 5b. Incubate 15 to 30 min at 37°C in the dark or in subdued light. After staining, briefly rinse three times with 37°C 9:1 (v/v) HBSS/FBS. At this stage, cells can be labeled with propidium iodide or SYTOX Green (see Basic Protocol 2, step 6a) to distinguish dead cells; however, these generally can be distinguished by morphology.
6b. Invert coverslip and mount onto a slide in a drop of 9:1 HBSS/FBS while leaving some clearance. Do not apply pressure. Seal the coverslip by a method regularly used in the laboratory (e.g., melted wax or nail polish).
Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
7b. To observe cells stained with LysoTracker dyes by fluorescence microscopy, use excitation and emission band-pass filters covering wavelength ranges compatible with the data displayed in Table 9.4.2. Use a band-pass filter centered around 620 nm for propidium iodide or around 530 nm for SYTOX Green.
9.4.12 Supplement 14
Current Protocols in Cytometry
STAINING OF LYSOSOMAL β-GALACTOSIDASE ACTIVITY WITH FDG This protocol describes the staining of lysosomes based on function by assaying endogenous β-galactosidase activity. The nonfluorescent substrate, fluorescein di-β-D-galactopyranoside (FDG), is cleaved into active fluorescein when it enters lysosomes with functional β-galactosidase. Samples are analyzed by flow cytometry.
ALTERNATE PROTOCOL 2
Additional Materials (also see Basic Protocol 7) 5 to 10 mM fluorescein di-β-D-galactopyranoside (FDG) working solution (in culture medium; prepared fresh from purchased stock solution) 1. Harvest cultured cells by standard procedures in 15-ml screw-cap centrifuge tubes and centrifuge 5 min at 200 × g, room temperature. 2. Resuspend the cell pellet at 0.5–1.0 × 106 cells/ml in 37°C cell culture medium. Leave cell suspensions at 37°C for ≥5 min. Because the functional state of the lysosomes is to be monitored, it is advisable to keep cell suspensions at their optimal temperature (37°C) and to allow them to recover for a brief moment after harvesting.
3. Thaw FDG working solution at room temperature, keeping it protected from light (e.g., in a drawer). Dye solutions decompose rapidly if exposed to light; FDG solutions have been found to decompose during repeated freeze-thaw cycles.
4. Divide cell suspension into 1-ml aliquots in 12 × 75–mm polypropylene tubes, add 10 µl FDG working solution, and mix immediately by briefly vortexing at maximum speed. Dye concentrations in the range of 50 to 100 ìM are recommended to obtain a sufficiently strong signal. FDG working solution should be prepared immediately before use and kept on ice. It is stable for only 1 to 2 hours.
5. Incubate 15 to 30 min at 37°C in the dark or in subdued light. After staining, place tube in a melting ice bath. 6. Optional: Exclude dead cells by costaining ≥15 min at room temperature with 5 µM propidium iodide (final). 7. Set up and optimize the flow cytometer (see Table 9.4.2). Because of the wide variation in lysosomal enzyme activity, use logarithmic signal amplification. Use a band-pass filter centered around 620 nm for detection of propidium iodide.
8. Carefully resuspend the cell sample by gently pipetting up and down a few times immediately before analysis. Cells tend to clump during staining; to obtain meaningful data on a per-cell basis, it is essential to resuspend cells immediately before analysis.
STAINING OF LIVE CELLS FOR MEASUREMENT OF GOLGI APPARATUS FUNCTION BY MICROSCOPY
BASIC PROTOCOL 8
This protocol details the staining of cultured mammalian cells with fluorescent ceramides. Samples can be analyzed by an epifluorescence microscope equipped with excitation and emission filters matching the dye excitation and emission wavelength range (see Table 9.4.3). The protocol may also be suitable for analysis of cell suspensions by flow cytometry. This protocol requires basic expertise in cell handling and microscopy. See Figure 9.4.1 for an example of labeled Golgi apparatus.
Studies of Cell Function
9.4.13 Current Protocols in Cytometry
Supplement 14
Figure 9.4.1 Fluorescence photomicrograph of a human skin fibroblast treated with a BODIPY-labeled analog of ceramide, N-[5-(5,7-dimethyl BODIPY)-1 pentanoyl]-D-erythro-sphingosine (C5DMB-Cer). Monolayer cultures were incubated 30 min at 4°C with 2 µM C5-DMB-Cer, washed, and further incubated 30 min at 37°C in a balanced salt solution. The cells were then observed under the fluorescence microscope (λex = 450 to 490 nm; λem ≥ 520 nm). The bright red/orange perinuclear structure corresponds to the Golgi apparatus (see Pagano et al., 1991). Figure by permission of Dr. Richard E. Pagano, Department of Biochemistry and Molecular Biology, Mayo Clinic and Foundation, Rochester, Minn.
Materials Cells grown in monolayer (APPENDIX 3B) on coverslips Cell culture medium without pH indicator or serum, room temperature and ice-cold Ceramide-BSA complex solution (see recipe) Melted wax or nail polish 18 × 18–mm coverslips sterilized by dipping into absolute ethanol and subsequent flaming 35-mm cell culture dishes Fluorescence microscope (UNIT 2.4) Additional reagents and equipment for adherent cell culture and harvesting (APPENDIX 3B) 1. Culture adherent cells overnight on sterilized 18 × 18–mm coverslips. Alternatively, larger coverslips and culture dishes can be used; in this case the volumes of the staining and other solutions should be adjusted accordingly. This protocol assumes that 35-mm dishes are used throughout. Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
2. Take coverslip with cells out of the cell culture dish, rinse twice with room temperature serum-free cell culture medium, and place in a 35-mm cell culture dish with 1 ml ice-cold cell culture medium. Put the dishes and some cell culture medium in a melting ice bath.
9.4.14 Supplement 14
Current Protocols in Cytometry
Table 9.4.3
Spectral Properties of Fluorescent Ceramides
Excitation maximum (nm)
Emission maximuma (nm)
Excitation sourceb (nm)
NBD C6 ceramide BODIPY FL C5 ceramide
466 505
536 511
488 (argon) 488 (argon)
BODIPY TR ceramide
590
616
543, 594 (HeNe)
Dye
aAll fluorescence emission spectra have been recorded with methanol as the solvent. BODIPY FL C5
ceramide and BODIPY TR ceramide exhibit a shift in fluorescence emission color after accumulation inside the Golgi apparatus. This is best observed with a band-pass filter centered around 620 nm. bOutput wavelengths of most commonly used flow cytometry lasers that are compatible with the excitation
spectra of the ceramides.
It is critical to remove any traces of serum from the coverslips, because the serum protein will interfere with the ceramide-BSA staining.
3. Thaw the ceramide-BSA complex solution at room temperature, keeping it protected from light (e.g., in a drawer). Add 100 µl ceramide-BSA complex to a 35-mm dish containing a coverslip. Incubate 30 min on ice protected from light. The ceramide-BSA complex may decompose if exposed to light. Do not refreeze thawed aliquots of ceramide-BSA complex because it decomposes during repeated freeze-thaw cycles.
4. Remove staining solution and rinse the coverslip at least three times with ice-cold serum-free cell culture medium. Incubate for ≥30 min at 37°C protected from light. At this stage, cells can be labeled with propidium iodide (if NBD C6 or BODIPY FL C5 is used) or SYTOX Green (if BODIPY TR is used; see Basic Protocol 6) to distinguish dead cells; however, these can generally be distinguished by morphology.
5. Rinse the coverslip with room temperature cell culture medium, invert coverslip, and mount onto a slide in cell culture medium while leaving some clearance. Do not apply pressure. Seal the coverslip by a method regularly used in the laboratory (e.g., melted wax or nail polish). 6. To observe cells stained with ceramide-BSA complexes by fluorescence microscopy, use excitation and emission band-pass filters covering the wavelength ranges denoted in Table 9.4.3. Use a band-pass around 620 nm for the detection of propidium iodide or around 530 nm for SYTOX Green.
GOLGI STAINING AND SECONDARY LABELING FOLLOWING CELL FIXATION
ALTERNATE PROTOCOL 3
In this procedure, cells are fixed prior to labeling of the Golgi apparatus. After Golgi labeling, the cells can be secondarily labeled with any desired antibody. The protocol is intended for microscopic analysis but may be suitable for flow cytometry. It assumes basic expertise in immunocytochemistry. Additional Materials (also see Basic Protocol 8) 3.7% formaldehyde (Fluka) in PBS (prepared fresh) or glutaraldehyde fixative solution (see recipe) 0.5 mg/ml NaBH4 (sodium borohydride) in unsupplemented cell culture medium (prepared fresh)
Studies of Cell Function
9.4.15 Current Protocols in Cytometry
Supplement 14
Defatted BSA solution (see recipe) Additional reagents and equipment for immunocytochemistry (Watkins, 1989) Culture and fix cells 1. Culture adherent cells overnight on sterilized 18 × 18–mm coverslips. 2. Take coverslip with cells out of the cell culture dish, rinse twice with room temperature serum-free cell culture medium, and add 1 ml of 3.7% formaldehyde/PBS or glutaraldehyde fixative solution. Incubate ≥15 min at room temperature. Chill some serum-free cell culture medium in a melting ice bath. Glutaraldehyde gives a sturdier fixation than formaldehyde. However, it leads to autofluorescence that must be quenched by incubation with NaBH4 (see step 4 below). Fixatives that will dissolve or extract lipids (e.g., acetone, ethanol, and methanol) should be avoided, since they will obviate staining of the Golgi apparatus.
3. Rinse coverslips with room temperature serum-free cell culture medium. 4. Optional: If glutaraldehyde is used, incubate 15 min in a melting ice bath in 0.5 mg/ml NaBH4 to reduce autofluorescence. 5. Place dish with coverslip in a melting ice bath and rinse at least three times with ice-cold serum-free cell culture medium. Label Golgi 6. Add 100 µl ceramide-BSA complex to a dish containing 1 ml ice-cold serum-free cell culture medium and incubate 30 min on ice under subdued light. 7. Rinse several times with room temperature serum-free cell culture medium. 8. Remove medium and add 1 ml defatted BSA solution. Incubate ≥30 min at room temperature. Repeat with fresh defatted BSA for a total of four incubations. This step removes nonspecifically bound excess ceramide-BSA complex.
Use immunocytochemistry (optional) and view 9. Rinse the coverslips with serum-free cell culture medium at room temperature. At this stage cells can be incubated in blocking buffer as the first step in labeling with an antibody. For a discussion of immunocytochemistry principles and techniques, refer to Watkins (1989). In this case cells cannot be permeabilized with acetone, ethanol, or methanol before antibody labeling, because that will remove the ceramide label.
10. If cells are to be viewed immediately, invert coverslip and mount onto a slide in cell culture medium while leaving some clearance. Do not apply pressure. Seal the coverslip by a method regularly used in the laboratory (e.g., melted wax or nail polish). 11. To observe cells stained with ceramide-BSA complexes by fluorescence microscopy, use excitation and emission band-pass filters covering the wavelength ranges listed in Table 9.4.3. Fixed, stained slides can be stored at 4°C short term (1 to 2 days) for later analysis. Longer term stability of the label has not been determined.
Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
9.4.16 Supplement 14
Current Protocols in Cytometry
STAINING OF LIVE CELLS FOR ENDOPLASMIC RETICULUM FUNCTIONAL ANALYSIS BY MICROSCOPY
BASIC PROTOCOL 9
This protocol details the staining of cultured mammalian cells with a carbocyanine dye, DiOC6(3). Samples can be analyzed by an epifluorescence microscope equipped with “fluorescein” excitation and emission filters. This protocol requires basic expertise in cell handling and microscopy. Materials Cells grown in monolayer (APPENDIX 3B) on coverslips Cell culture medium with and without 10% FBS (APPENDIX 3B), 37°C 1 mM DiOC6(3) stock solution (see recipe) Melted wax or nail polish 18 × 18–mm coverslips sterilized by dipping into absolute ethanol and subsequent flaming 35-mm cell culture dishes Fluorescence microscope (UNIT 2.4) Additional reagents and equipment for cell culture and harvesting (APPENDIX 3B) 1. Culture adherent cells overnight on sterilized 18 × 18–mm coverslips. 2. Take coverslips with cells out of the cell culture dish and place in a 35-mm cell culture dish in 1 ml prewarmed serum-free cell culture medium. 3. Dilute DiOC6(3) stock solution into serum-free cell culture medium to make a 1 µM staining solution and keep this in the dark. Staining solution may decompose if exposed to light. Do not refreeze stock or staining solution, because it decomposes during repeated freeze-thaw cycles.
4. Replace culture medium with 1 ml staining solution. Incubate ~5 min in the dark at 37°C. 5. Remove the coverslip from the staining solution, mount it on a microscope slide, and view the extent of staining by microscopy. If only the mitochondria are stained, put the coverslip back into the staining solution and continue to incubate. If cells do not appear to be sufficiently stained after 10 min, increase the dye concentration in the staining medium to 5 ìM and repeat staining with a fresh coverslip. If cells are rounding up, or the mitochondria appear swollen, repeat staining with a lower dye concentration. At this stage, cells can be labeled with 5 ìM propidium iodide (final; see Basic Protocol 6) to distinguish dead cells; however, these can generally be distinguished by morphology.
6. After sufficient cell staining is obtained, take the coverslip out of the staining dish, invert coverslip, and mount onto a slide in cell staining medium while leaving some clearance. Do not apply pressure. Seal the coverslip by a method regularly used in the laboratory (e.g., melted wax or nail polish). 7. To observe cells stained with DiOC6(3) by fluorescence microscopy, use fluorescein excitation and emission band-pass filters. Use a band-pass filter centered around 620 nm for the detection of propidium iodide.
Studies of Cell Function
9.4.17 Current Protocols in Cytometry
Supplement 14
ALTERNATE PROTOCOL 4
ENDOPLASMIC RETICULUM STAINING FOLLOWING CELL FIXATION This protocol provides additional steps for fixation of cells prior to staining of the ER. Like the live cell staining procedure (see Basic Protocol 4), it is intended to be used for microscopy. Additional Materials (also see Basic Protocol 9) Glutaraldehyde fixative solution (see recipe) 0.5 mg/ml NaBH4 (sodium borohydride) in unsupplemented cell culture medium (prepared fresh) PBS (APPENDIX 2A) 1. Culture adherent cells overnight on sterilized 18 × 18–mm coverslips. 2. Take coverslip with cells out of the cell culture dish, rinse twice with room temperature serum-free cell culture medium, and add 1 ml glutaraldehyde fixative solution. Incubate ≥15 min at room temperature. Fixatives that will dissolve or extract lipids (e.g., acetone, ethanol, and methanol) should be avoided, as they will obviate staining of the endoplasmic reticulum. After fixation with formaldehyde, endoplasmic reticulum often appears swollen and vesiculated.
3. Remove fixative solution and rinse coverslips with room temperature unsupplemented cell culture medium. 4. To reduce the autofluorescence that develops during fixation with glutaraldehyde, incubate 15 min at room temperature with 0.5 mg/ml NaBH4. 5. Rinse several times, at least 10 min each time, with room temperature PBS. 6. Proceed with staining and microscopy (see Basic Protocol 9, steps 3 to 7). In following the steps of Basic Protocol 9, the use of propidium iodide should be omitted in this fixation protocol. If desired, cells can be stained with an antibody. For a discussion of immunocytochemistry principles and techniques, refer to Watkins (1989). Fixed, stained slides can be stored short term (1 to 2 days) at 4°C for later analysis. Longer-term stability of the label has not been determined.
REAGENTS AND SOLUTIONS Use deionized, doubly distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2; for suppliers, see SUPPLIERS APPENDIX.
Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
Ceramide-BSA complex solution Dissolve NBD C6 ceramide, BODIPY FL C5 ceramide, or BODIPY TR ceramide in 19:1 (v/v) chloroform/methanol to make a 1 mM solution. Pipet 50 µl into a small glass test tube, then dry by flushing with nitrogen and then in a Speedvac for ≥1 hr. Add 200 µl absolute ethanol and vortex vigorously to dissolve the ceramide completely. Dissolve 240 µg of defatted BSA (Sigma) into 10 ml cell culture medium without serum or pH indicator in a 50-ml plastic centrifuge tube. While vortexing at maximum speed, inject 200 µl ceramide solution into the BSA solution. Rinse the glass test tube with the ceramide-BSA complex. Dialyze overnight at 4°C against 500 ml serum-free cell culture medium. Dilute the dialysate into 10 ml serum-free cell culture medium, divide into 100-µl aliquots, and store up to 6 months in the dark at −20°C. Do not refreeze.
9.4.18 Supplement 14
Current Protocols in Cytometry
Defatted BSA (bovine serum albumin) solution Dissolve 1.2 mg defatted BSA (Sigma) into 50 ml cell culture medium without serum or pH indicator in a 50-ml plastic centrifuge tube. Use fresh. DiOC6(3) stock solution, 1 mM Dissolve DiOC6(3) (Molecular Probes) in absolute ethanol to make a stock solution of 1 mM. Divide into convenient aliquots (this will depend on the number of samples to be stained) and store up to 6 months at −20°C in the dark. Do not refreeze. Glutaraldehyde fixative solution Prepare a 10% (w/v) solution of sucrose in 100 mM PIPES. Adjust pH to 7.0. Dilute glutaraldehyde stock solution (microscopy grade; Sigma) into sucrose/PIPES to obtain 0.5% glutaraldehyde. Store at 4°C until a precipitate forms or as long as solution remains colorless and odorless. LysoTracker dye working solutions, 100× Dilute the LysoTracker stock solution as provided by the manufacturer (Molecular Probes) 200-fold in PBS (APPENDIX 2A) to give a 100× (5-µm) working solution. Keep working solution in the dark on ice or at 4°C and discard at the end of the day; it cannot be stored overnight or longer. Divide stock solution into aliquots upon receipt to avoid refreezing.
Mitochondrial dye stock solutions, 0.5 mM Dissolve rhodamine 123, dihydrorhodamine 123, MitoTracker Green, MitoFluor, nonyl acridine orange, chloromethyl-X-rosamine (CMXRos), or dihydrochloromethyl-X-rosamine (CMXRos-H2; all from Molecular Probes) in dry DMSO (Sigma) or absolute ethanol to make a stock solution of 0.5 mM. Store at −20°C in the dark; stable >1 year. To minimize exposure of cell samples to high concentrations of solvent, dye stock solutions should be prepared at 0.5 mM (100×). Stock solutions of reduced dyes should be flushed with inert gas (e.g., nitrogen or argon) immediately after preparation. To minimize hazards during handling, most dyes are now supplied as small weighed aliquots of dye powder or small aliquots of dye solution. Mitochondrial dye stock solutions can be refrozen. Reduced dyes should again be flushed with inert gas prior to refreezing.
COMMENTARY Background Information Mitochondria Most of the energy needed for the proper functioning of cells is generated by the mitochondria through the oxidation of reduced nucleotides (e.g., NADH2 and FADH2) and the generation of a negative inside membrane potential. Notwithstanding decades of intense biochemical and cell biological research, the mechanisms by which mitochondria function have not been fully elucidated. Interest in these organelles rose to new heights upon the discovery that some severely debilitating inherited neuromuscular diseases arise from mutations in the mitochondrial genome that result in mitochondrial dysfunction (Wallace, 1995). In
addition, the mitochondrion appears to be involved in the process of apoptosis (Green and Kroemer, 1998; Green and Reed, 1998). One approach to fluorescent labeling of mitochondria is based upon their negative inside membrane potential, which allows lipophilic fluorescent cations to accumulate inside mitochondria. Because there is now a variety of mitochondrial dyes with differing properties available, Table 9.4.1 has been included as a guide to dye selection. The first dye to be described as a specific probe for mitochondria was rhodamine 123 (Johnson et al., 1981). The amount of fluorescence obtained with this dye corresponds to the physiological state of the mitochondrion, as predicted from the assumption that the mito-
Studies of Cell Function
9.4.19 Current Protocols in Cytometry
Supplement 14
Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
chondrial membrane potential is the driving force behind dye accumulation inside the cell. During induction of apoptosis, which entails a decrease in mitochondrial membrane potential, a paradoxical increase in rhodamine 123 fluorescence has been found (Vander Heiden et al., 1997). This is due to the fact that rhodamine 123 fluorescence can undergo self-quenching when the dye accumulates to a high concentration. “Dilution” of the dye then leads to a higher fluorescence level, caused by “de-quenching.” Vander Heiden and coworkers (1997) observed swelling of the mitochondria and a concomitant increase in rhodamine 123 fluorescence during apoptosis. These observations should give rise to caution when interpreting results obtained with rhodamine 123. Almost a decade later a second dye was found to respond to the functional state of mitochondria: 5,5′,6,6′-tetrachloro-1,1′,3,3′tetraethylbenzimidazolylcarbocyanine iodide (JC-1; Reers et al., 1991). The JC-1 dye exhibits green fluorescence (excitation wavelength 490 nm/emission 527 nm) when present in low concentrations, and red fluorescence (excitation 490 nm/emission 590 nm) when accumulating at higher concentrations. The intramitochondrial concentration of the dye, and thus its fluorescence maximum, depends in turn upon the mitochondrial potential and thus mitochondrial function (Smiley et al., 1991). However, in addition to accumulating inside the mitochondrion, JC-1 and other carbocyanine dyes were found to stain the endoplasmic reticulum (Chen, 1989). This feature limits the usefulness of these dyes for studies of mitochondrial physiology by flow cytometry. Nevertheless, this dye is being used to detect changes in mitochondrial function during apoptosis (Cossarizza et al., 1994; UNIT 9.14). Although rhodamine 123 proved to be quite valuable in flow cytometric studies, its high sensitivity to illumination by a microscope lamp limited its use in applications involving microscopy. In addition, rhodamine 123 is lost from a stained specimen as soon as cell fixation is attempted. A series of dyes developed in recent years have overcome these drawbacks. The MitoTracker dyes chloromethyltetramethylrosamine (MitoTracker Orange) and CMXRos (MitoTracker Red) exhibit good photostability, are retained after cell fixation, and appear to respond to changes in mitochondrial function (Bossy-Wetzel, 1998; Macho et al., 1996; Poot et al., 1996). The MitoTracker Orange dye appears to inhibit respiratory complex I and to induce the mitochondrial permeability
transition (Scorrano et al., 1999). This obvious dye toxicity precludes the use of MitoTracker Orange as a tool to measure changes in mitochondrial membrane potential (Scorrano et al., 1999). The MitoTracker Red dye CMXRos, on the other hand, did not show obvious cytotoxicity in short-term assays (M. Poot, unpub. observ.). A number of investigators found CMXRos to be suitable for monitoring changes in mitochondrial membrane potential during apoptosis (Macho et al., 1996; Gilmore and Wilson, 1999; Poot and Pierce, 1999). Because CMXRos contains a chloromethyl moiety that can bind to reduced thiol groups of proteins, this dye will be retained after cell fixation (Macho et al., 1996; Poot et al., 1996). Therefore, amounts of dye retained after fixation are likely to be a function of both uptake (driven by the mitochondrial membrane potential) and the number of reduced protein thiol groups available. Thus, it is conceivable that the fluorescence intensity retained after cell fixation does not truly reflect the mitochondrial membrane potential before cell fixation. Gilmore and Wilson (1999) found this to be the case. The authors conclude that cell fixation may dramatically alter the amount of CMXRos dye retained and suggest that using CMXRos fluorescence after cell fixation is inappropriate and will lead to spurious results (Gilmore and Wilson, 1999). The MitoTracker Green FM and MitoFluor dyes are both well excited by the 488-nm line of argon lasers; they emit in the green region of the spectrum and show little overlap with the orange-red region (see Table 9.4.1). The fluorescence emission range of MitoTracker Green FM and MitoFluor dyes is much narrower than that of rhodamine 123 (which emits significantly at wavelengths up to 620 nm). Therefore, the MitoTracker Green FM and MitoFluor dyes can be used readily in multicolor applications. Although MitoTracker Green FM is retained after fixative treatment, MitoFluor fluorescence vanishes. The mechanism by which these dyes label mitochondria is not known; however, neither dye appears to respond to alterations in the physiological state of mitochondria (M. Poot, unpub. obser.). Therefore it appears that MitoTracker Green FM fluorescence reflects mitochondrial mass. This feature has been used to normalize CMXRos fluorescence (which is red and mitochondrial membrane potential sensitive) to the amount mitochondrial mass per cell (Poot and Pierce, 1999). Nonyl acridine orange (Maftah et al., 1989) accumulates in
9.4.20 Supplement 14
Current Protocols in Cytometry
mitochondria due its specific, high-affinity binding to cardiolipin (Petit et al., 1992). Because cardiolipin is a phospholipid situated at the inner mitochondrial membrane, the fluorescence of nonyl acridine orange depends upon the amount of cardiolipin present. In other words, the fluorescence intensity of nonyl acridine orange has been taken as a direct measure of the amount of mitochondrial lipid in a cell (Petit et al., 1992). Nonyl acridine orange (NAO) emits a broad range of fluorescent light that covers the green, yellow, and red parts of the visible spectrum. The red emission (∼640 nm) appears to reflect most accurately the cardiolipin content of membranes (Gallet et al., 1995). Recently, the green and red fluorescence of NAO was shown to be sensitive to a variety of drug treatments (Keij et al., 2000). The ratio of red to green fluorescence of NAO appeared to be quite stable among the different treatments. The possible sensitivity of NAO fluorescence to alterations in mitochondrial function clearly needs more investigation. For now, it appears prudent to take the ratio of red to green fluorescence of NAO-stained cells as a measure of mitochondrial cardiolipin content. In addition, reduced forms of some dyes are available (e.g., dihydrorhodamine 123, H2CMXRos), that yield a fluorescent response only after they are oxidized by functioning mitochondria (Whitaker et al., 1991). Fluorescence of H2-CMXRos proved to be strongly sensitive to inhibition of mitochondrial electron flux with antimycin A (Poot and Pierce, 1999). Thus, H2-CMXRos may allow functional monitoring of the cytochrome c oxidase– dependent respiration complex IV, which is affected in some muscular diseases (Morin et al., 1993; Capaldi et al., 1995). Lysosomes Since their discovery in the early 1950s (Berthet et al., 1951), lysosomes have enjoyed continuous interest from biochemists and cell biologists. Indications of the possible involvement of lysosomal dysfunction in a large number of inherited diseases of infancy, childhood, and early adulthood that have in common a progressively debilitating neurologic decline have added incentive to the research into lysosomal structure and function. Elucidation of the versatile mechanisms whereby cellular m acro mo lecules are degraded inside lysosomes has long been hampered by the paucity of biochemical methods available for assessing lysosomal enzymes. Initially, these en-
zymes could be studied only with the aid of chromogenic substrates that were not plasma membrane permeant. Thus, only bulk information derived from cell homogenates could be obtained. Fluorescent substrates, which were developed during the 1970s, also could not be used with intact cells and therefore still limited the analysis of lysosomes to studies in homogenates or permeabilized cells. With the synthesis of mono- and digalactopyranosides of fluorescein, however, the first plasma membrane–permeant substrates for lysosomal enzymes became available (Hofmann and Sernetz, 1983). Addition of a sugar moiety to a fluorescein molecule entails a significant reduction in fluorescence; bis-conjugation of fluorescein leads to an essentially nonfluorescent product, fluorescein di-β-Dgalactopyranoside (FDG; Hofmann and Sernetz, 1983). Enzymatic cleavage of the sugars leads to the full recovery of fluorescence. This process, termed fluorochromasia, allows detection of very low levels of substrate turnover. FDG was the first substrate used in a flow cytometric assay of lysosomal enzyme activity (Jongkind et al., 1986). In addition to fluorescein-based fluorogenic substrates for glycosidases, rhodamine-based substrates for exoand endoproteinases have been developed (Leytus et al., 1983) and successfully used in flow cytometry (Assfalg-Machleidt et al., 1992). Data obtained by flow cytometry of intact cells incubated with these plasma membrane– permeant fluorogenic substrates do not necessarily reflect intralysosomal enzyme activity (Jongkind et al., 1986). This discrepancy may be caused by the fact that the substrate has to pass several membrane barriers before it can be converted by its cognate lysosomal enzyme. In addition, the fluorescence from fluorescein shows a pKa value of 6.4, whereas the pH of the lysosome is below 5.0. Therefore, the fluorescence generated by enzymic action upon the substrate inside the lysosome is strongly quenched. Unless the substrate transportation problem and the pH sensitivity of the fluorophores are remedied, only semiquantitative information can be obtained (Jongkind et al., 1986). Golgi apparatus The Golgi apparatus is a complex conglomerate of stacks of membranes and associated vesicles. It has been implicated in the synthesis of complex carbohydrates, modification of glycoproteins, and routing of matured glycoproteins to lysosomes, secretory vesicles, and the cell
Studies of Cell Function
9.4.21 Current Protocols in Cytometry
Supplement 14
surface. Adding to its unique and highly fascinating features is the lipid composition of the membranes that make up the Golgi apparatus. A strategy to specifically label the Golgi apparatus has been devised based on this lipid composition. The strategy entails the modification of a sphingosine with a lipophilic fluorophore. The first probe used successfully in this way was N-[7-(4-nitrobenzo-2-oxa-1,3-diazole)]-6-aminocaproyl sphingosine (NBD C6 ceramide; Lipsky and Pagano, 1985). With NBD C6 ceramide the Golgi apparatus has been labeled specifically in viable and fixed cells (Pagano, 1989) and via an ATP-dependent uptake of probe-loaded liposomes (Kobayashi and Pagano, 1988). The NBD C6 ceramide probe, however, has several problems. First, the NBD moiety rapidly loses is fluorescence during viewing with an epifluorescence microscope. Second, the probe appears to rapidly label intracellular membraneous structures such as the endoplasmic reticulum and mitochondria. This property may result in rapid metabolic conversion of the NBD C6 ceramide (Kobayashi and Arakawa, 1 99 1) . Sy nth esis o f B ODIPY- labeled ceramides made it possible to observe the labeled Golgi apparatus for much longer time periods (Pagano et al., 1991). In addition to improved photostability, BODIPY ceramides offer another useful feature: after accumulation to a relatively high level inside the Golgi apparatus, their fluorescence emission range changes from green to red (Pagano et al., 1991). This is due to excimer formation of dye molecules, which are stacked onto each other. Thus, the Golgi apparatus exhibits red labeling while other cell labeling sites emit green fluorescence.
Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
Endoplasmic reticulum The endoplasmic reticulum is an extensive membraneous network that serves as a scaffold for protein and lipid synthesis. The endoplasmic reticulum is not visible by regular microscopy, hence the need for a staining procedure. Because the endoplasmic reticulum consists of membranes, it will accumulate any lipophilic dye. The very lipophilicity of the dyes needed makes them hard to handle in a cell culture environment, which is essentially aqueous. The carbocyanine dye DiOC6(3) fulfills the two conflicting requirements of accumulating readily in the endoplasmic reticulum and being water soluble enough to be used in a regular cell culture system (Terasaki et al., 1984). Because the mechanism of staining of the endoplasmic reticulum involves passive accumu-
lation in cellular membranes, organelles such as mitochondria and lysosomes will be stained as well. Because of this lack of selectivity for the endoplasmic reticulum, this procedure cannot be recommended for quantitation of the amount of endoplasmic reticulum in a cell by flow cytometry. Staining of cells after fixation, which abolishes the mitochondrial membrane potential, improves the selectivity of a DiOC6(3)-based procedure for staining endoplasmic reticulum. Some carbocyanine dyes have been found to exert significant mitochondrial toxicity (Terasaki et al., 1984; Anderson et al., 1993). Therefore, some caution has to be applied in performing studies with DiOC6(3).
Critical Parameters and Troubleshooting Experimental parameters As analysis of mitochondria and lysosomes by flow or image cytometry is intended to generate information on the physiological state of the cells under study, it is paramount to prepare fresh staining solutions and to prewarm these to 37°C at all times. A cold shock may cause mitochondria to “wrinkle” rapidly and can affect cell staining accordingly. Another potential problem to keep in mind is that dye solutions can be subject to photodegradation. It is therefore advisable to store all dye solutions in a freezer (not a “No-Frost” freezer) in the dark. Shortly before use, dye solutions can be thawed at room temperature while being kept in the dark (e.g., in a drawer). Solutions of reduced dyes have to be flushed with inert gas (e.g., nitrogen or argon) to prevent oxidation. Because analysis of the Golgi apparatus and the endoplasmic reticulum is based upon the specific uptake of a group of complex lipids (the ceramides), it is critical to keep the lipid composition of the cell intact when performing such studies. In addition, labeling the Golgi apparatus involves the use of highly lipophilic probes. Viable cells can be labeled only if the the probe is contained within a carrier such as a liposome or defatted BSA. The procedures outlined in this unit offer a relatively simple and straightforward way of handling such probes. Instrument function If little or no fluorescence is found in stained cells, check whether the excitation wavelength and output power of the laser or arc lamp are compatible with the dye used; check filter combinations in front of the photomultiplier tubes. If those parameters suit the dye’s features and
9.4.22 Supplement 14
Current Protocols in Cytometry
still little or no signal is obtained, try to increase the signal amplification; the sensitivity of the detection system of the flow or image cytometry system may vary. If the signal is still weak, check cells by fluorescence microscopy to see whether cell staining is dim or bright.
Anticipated Results With suspensions of cells stained with a mitochondrial dye, a single population of signals with a coefficient of variation in the range of 20% to 40% is obtained. In some cases a much broader signal distribution will be observed; sometimes two or more signal populations can even be resolved. With suspensions of cells stained with a lysosomal dye, a single population of signals with a coefficient of variation in the range of 30% to 50% is obtained. Again, in some cases a much broader signal distribution, or even two or more resolvable signal populations, will be observed. Some cell types, such as lymphocytes, contain very few lysosomes and may therefore not be amenable to the study of lysosome function. Good results are generally obtained with fibroblasts, provided that the donor of the biopsy is not deficient for the lysosomal enzyme activity under study. Figure 9.4.1 provides an example of what a labeled Golgi apparatus should look like. For an example of a DiOC6(3)-labeled endoplasmic reticulum, see Terasaki et al. (1984).
Time Considerations Mitochondria, lysosomes, and the endoplasmic reticulum For warm-up and laser alignment, 1 hr should be allowed. The time needed to harvest, stain, and analyze cells varies according to the number of samples to be analyzed. If a microscope-based system is to be used, the warm-up time may be shorter, but data acquisition time per sample may be longer. The Golgi apparatus The preparation of the labeled ceramideBSA complex will take several hours. Depending on the procedure used, cell labeling may take several hours to a full day.
Literature Cited Anderson, W.M., Delinck, D.L., Benninger, L., Wood, J.M., Smiley, S.T., and Chen, L.B. 1993. Cytotoxic effect of thiocarbocyanine dyes on human colon carcinoma cells and inhibition of bovine heart mitochondrial NADH-ubiquinone reductase activity via a rotenone-type mecha-
nism by two of the dyes. Biochem. Pharmacol. 45:691-696. Assfalg-Machleidt, I., Rothe, G., Klingel, S., Banati, R., Mangel, W.F., Valet, G., and Machleidt, W. 1992. Membrane permeable fluorogenic rhodamine substrates for selective determination of Cathep sin L. Biol. Chem. Hoppe-Seyler 373:433-440. Berthet, J., Berthet, L., Appelmans, F., and De Duve, C. 1951. Tissue fractionation studies. 2. The nature of the linkage between acid phosphatase and mitochondria in rat-liver tissue. Biochem. J. 50:182-189. Bossy-Wetzel, E., Newmeyer, D.D., and Green, D.R. 1998. Mitochondrial cytochrome c release in apoptosis occurs upstream of DEVD-specific caspase activation and independently of mitochondrial transmembrane depolarization. EMBO J. 17:37-49. Capaldi, R.A., Marusich, M.F., and Taanman J.W. 1995. Mammalian cytochrome-c oxidase: Characterization of enzyme and immunological detection of subunits in tissue extracts and whole cells. Methods Enzymol. 260:117-132. Chen, L.B. 1989. Fluorescent labeling of mitochondria. Methods Cell Biol. 29:103-123. Cossarizza, A., Kalashnikova, G., Grassilli, E., Chiappelli, F., Salvioli, S., Capri, M., Barbieri, D., Troiano, L., Monti, D., and Franceschi, C. 1994. Mitochondrial modifications during rat thymocyte apoptosis: A study at the single cell level. Exp. Cell Res. 214:323-330. Gallet, P.F., Maftah, A., Petit, J.-M., Denis-Gay, M., and Julien, R. 1995. Direct cardiolipin assay in yeast using the red fluorescence emission of 10-N-nonyl acridine orange. Biochem. J. 228:113-119. Gilmore, K. and Wilson, M. 1999. The use of chloromethyl-X-rosamine (MitoTracker Red) to measure loss of mitochondrial membrane potential in apoptotic cells is incompatible with cell fixation. Cytometry 36:355-358. Green, D. and Kroemer, G. 1998. The central executioners of apoptosis: Caspases or mitochondria? Trends Cell. Biol. 8:267-271. Green, D.R. and Reed, J.C. 1998. Mitochondria and apoptosis. Science 281:1309-1312. Hofmann, J. and Sernetz, M. 1983. A kinetic study on the enzymatic hydrolysis of fluorescein diacetate and fluorescein-di-β-D-galactopyranoside. Anal. Biochem. 131:180-186. Johnson, L.V., Walsh, M.L., Bockus, B.J., and Chen, L.B. 1981. Monitoring of relative mitochondrial membrane potential in living cells by fluorescence microscopy. J. Cell Biol. 88:526-535. Jongkind, J.F., Verkerk, A., and Sernetz, M. 1986. Detection of acid β-galactosidase activity in viable human flibroblasts by flow cytometry. Cytometry 7:463-466. Kobayashi, T. and Arakawa, Y. 1991. Transport of exogenous fluorescent phosphatidylserine analogue to the Golgi apparatus in cultured fibroblasts. J. Cell Biol. 113:235-244.
Studies of Cell Function
9.4.23 Current Protocols in Cytometry
Supplement 14
Kobayashi, T. and Pagano, R.E. 1988. ATP-dependent fusion of liposomes with the Golgi apparatus of perforated cells. Cell 55:797-805. Leytus, S.P., Melhado, L.L., and Mangel, W.F. 1983. Rhodamine-based compounds as fluorogenic substrates for serine proteinases. Biochem. J. 209:299-307. Lipsky, N.G. and Pagano, R.E. 1985. A vital stain for the Golgi apparatus. Science 228:745-747. Macho, A., Decaudin, D., Castedo, M., Hirsch, T., Susin, S.A., Zamzami, N., and Kroemer, G. 1996. Chloromethyl-X-Rosamine is an aldehyde-fixable potential-sensitive fluorochrome for the detection of early apoptosis. Cytometry 25:333-340. Maftah, A., Petit, J.-M., Ratinaud, M.-H., and Julien, R. 1989. 10 N-Nonyl-acridine orange: A fluorescent probe which stains mitochondria independently of their energetic state. Biochem. Biophys. Res. Commun. 164:185-190. Morin, C., Mitchell, G., Larochelle, J., Lambert, M., Ogier, H., Robinson, B.H., and De Braekeleer, M. 1993. Clinical, metabolic, and genetic aspects of cytochrome C oxidase deficiency in Saguenay-Lac-Saint-Jean. Am. J. Hum. Genet. 53:488-496. Pagano, R.E. 1989. A fluorescent derivative of ceramide: Physical properties and use in studying the Golgi apparatus of animal cells. Methods Cell Biol. 29:75-85. Pagano, R.E., Martin, O.C., Kang, H.C., and Haugland, R.P. 1991. A novel fluorescent ceramide analogue for studying membrane traffic in animal cells: Accumulation at the Golgi apparatus results in altered spectral properties of the sphingolipid precursor. J. Cell Biol. 113:1267-1279. Petit, J.-M., Maftah, A., Ratinaud, M.-H., and Julien, R. 1992. 10 N-Nonyl-acidine orange interacts with cardiolipin and allows the quantification of this phospholipid in isolated mitochondria. Eur. J. Biochem. 209:267-273. Poot, M. and Pierce, R.H. 1999. Detection of changes in mitochondrial function during apoptosis by simultaneous staining with multiple fluorescent dyes and correlated multiparameter flow cytometry. Cytometry 35:311-317. Poot, M., Zhang, Y.-Z., Krämer, J., Wells, K.S., Jones, L.J., Hanzel, D.K., Lugade, A.G., Singer, V.L., and Haugland, R.P. 1996. Analysis of mitochondrial morphology and function with novel
fixable fluorescent stains. J. Histochem. Cytochem. 44:1363-1372. Reers, M., Smith, T.W., and Chen, L.B. 1991. J-aggregate formation of a carbocyanine as a quantitative fluorescent indicator of membrane potential. Biochemistry 30:4480-4486. Scorrano L., Petronilli, V., Colonna, R., Di Lisa, F., and Bernardi, P. 1999. Chloromethyltetramethylrosamine (Mitotracker Orange) induces the mitochondrial permeability transition and inhibits respiratory complex I. Implications for the mechanism of cytochrome c release. J. Biol. Chem. 274:24657-24663. Smiley, S.T., Reers, M., Mottola-Hartshorn, C., Lin, M., Chen, A., Smith, T.W., Steele, G.D. Jr., and Chen, L.B. 1991. Intracellular heterogeneity in mitochondrial membrane potentials revealed by a J-aggregate-forming lipophilic cation JC-1. Proc. Natl. Acad. Sci. U.S.A. 88:3671-3675. Terasaki, M., Song, J., Wong, J.R., Weiss, M.J., and Chen, L.B. 1984. Localization of endoplasmic reticulum in living and glutaraldehyde-fixed cells with fluorescent dyes. Cell 38:101-108. Thorell, B. 1983. Flow-cytometric monitoring of intracellular flavins simultaneously with NAD(P)H levels. Cytometry 4:61-65. Vander Heiden, M.G., Chandel, N.C., Williamson, E.K., Schumacher P.T., and Thompson, C.B. 1997. Bcl-XL regulates the membrane potential and volume homeostasis of mitochondria. Cell 91:627-637. Wallace, D.C. 1995. Mitochondrial DNA variation in human evolution, degenerative disease, and aging. Am. J. Hum. Genet. 57:201-223. Watkins, S. 1989. Immunohistochemistry. In Current Protocols in Molecular Biology (F.M. Ausubel, R.E. Brent, B. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 14.6.1-14.6.13. John Wiley & Sons, New York. Whitaker, J.E., Moore, P.L., Haugland, R.L., and Haugland, R.P. 1991. Dihydrotetramethylrosamine: A long wavelength, fluorogenic peroxidase substrate evaluated in vitro and in a model phagocyte. Biochem. Biophys. Res. Commun. 175:387-393.
Contributed by Martin Poot University of Washington Seattle, Washington
Analysis of Intracellular Organelles by Flow Cytometry or Microscopy
9.4.24 Supplement 14
Current Protocols in Cytometry
Reporters of Gene Expression: Enzymatic Assays
UNIT 9.5
Reporter genes are widely used in studies of gene expression to quantitate the regulation of genetic elements. The value of a reporter gene is based on the existence of a specific property of the encoded protein that can be accurately and precisely measured: for instance, the protein may have enzymatic activity (hence, the measurement would be conversion of a substrate) or may be capable of being directly measured (e.g., be inherently fluorescent). Reporter gene systems were initially established for the purpose of quantitating promoter strength in bacteria. Recently, they have been utilized in eukaryotic cells, including mammalian cells, to reveal genetic regulation after signal transduction or during differentiation, or simply to mark cells for identification after another procedure, such as reconstitution of a host with stem cells. Measuring β-galactosidase (β-gal) activity in individual viable cells by flow cytometry allows the determination of gene expression on a cell-by-cell basis, which has several useful consequences not accomplished by simple bulk assays: (1) determination of the distribution of enzyme activity (and thus gene activity) within a population of viable cells, (2) resolution of heterogeneities in gene expression within a population, (3) concomitant measurement of other cellular parameters, such as surface immunophenotype, DNA content, or intracellular glutathione content, and (4) sorting of cells with defined levels of gene expression for subsequent studies—in essence, using the reporter gene as a selectable marker. For reporter genes to be useful in flow cytometric studies, the encoded protein product must directly or indirectly generate sufficient levels of intracellular fluorescence for quantitative analysis. Currently, two classes of reporter genes are routinely used for this purpose: hydrolases, such as β-gal, that cleave a fluorogenic substrate, and intrinsically fluorescent proteins, such as green fluorescent protein (GFP). This unit discusses the use of hydrolases (β-gal or β-glucuronidase; see Basic Protocol and see Alternate Protocol, respectively) in flow cytometric studies, the advantages and disadvantages of these reporter genes, and the current application of these systems in clinical and basic research applications. The use of a fluorescence plate reader and methylumbelliferone-based β-gal and β-glucuronidase substrates for quantitation of reporter gene expression in the context of cell lysates is also described (see Support Protocol). This assay is used to standardize hydrolase assays, allowing the measured fluorescence to be converted into number of protein molecules per cell, as discussed in the Commentary (see Anticipated Results). STRATEGIC PLANNING When embarking on the construction of novel vectors for reporter gene analysis, several considerations are pertinent to the choice of reporter gene. (1) Perhaps the most important issue is sensitivity: how strong is the expression element? The hydrolase assays are extremely sensitive, capable of detecting as few as five to ten molecules of enzyme per cell. On the other hand, it is likely that more than a thousand GFP molecules per cell is the minimum detectable; thus, with relatively weak promoters, GFP will not be detected. (2) The second issue involves vector limitations: is there a size limit to the reporter gene coding segment? The coding sizes are 3.4 kb for β-gal, 1.8 kb for β-glucuronidase, and 0.8 kb for GFP. (3) In which tissues is expression to be measured? For example, it is virtually impossible to detect β-gal expression in primary lymphocytes, whereas GFP has not shown any such problems. (4) Are noncytometric assays of activity going to be used? Contributed by Matt Lorincz and Mario Roederer Current Protocols in Cytometry (1998) 9.5.1-9.5.22 Copyright © 1998 by John Wiley & Sons, Inc.
Studies of Cell Function
9.5.1 Supplement 6
Biochemical bulk assays, such as those using 4-methylumbelliferyl β-D-galactopyranoside (MUG) or 5-bromo-4-chloro-3-indolyl β-D-galactopyranoside (Xgal) as β-gal substrates, are very useful for prescreening purposes. There is no bulk assay for GFP. On the other hand, GFP can be used in microscopic localization studies for which hydrolases are not well suited. (5) What assays need to be done concurrently with the reporter gene assay? If another assay requires the fluorescein channel on the cytometer, then the only suitable reporter genes are spectral mutants of GFP. In the future, hydrolases may be used with UV-based fluorogenic substrates, but this is still under development. Also, if postprocessing steps such as fixation or permeabilization are performed, then the hydrolase-based assays (in which the product may leak) may not be well suited. (6) All other issues being equal, the GFP reporter genes are easier to use because cells can be directly assayed on the cytometer without preparation. BASIC PROTOCOL
USE OF FLUORESCEIN DI-â-D-GALACTOPYRANOSIDE AND FLOW CYTOMETRY TO QUANTITATE â-GALACTOSIDASE ENZYMATIC ACTIVITY In this protocol, E. coli lacZ–encoded β-gal activity is measured in individual viable eukaryotic cells by flow cytometry. The assay depends on the introduction of fluorescein di-β-D-galactopyranoside (FDG), a nonfluorescent fluorogenic substrate, into the cell cytoplasm. This substrate is hydrolyzed and retained intracellularly. The presumption underlying the assay is that the rate of hydrolysis (and, therefore, the rate of accumulation of fluorescence) is proportional to the cellular concentration of the reporter enzyme, as long as the substrate has not been exhausted. In this system, β-gal serves both as a reporter gene to quantitate gene expression and as a selectable marker for the sorting of fluorescence-labeled cells based on the extent of substrate hydrolysis. To load FDG, cells in suspension at 37°C are briefly exposed to hypotonic medium containing the substrate. Following this treatment, the cell suspension is diluted at least 10-fold with ice-cold isotonic medium to restore isotonicity and chill the cells (“quenching”). These two effects serve to restrict further substrate entry and to prevent leakage of the hydrolyzed product. The protocol includes descriptions of (1) the use of phenylethyl β-D-thiogalactopyranoside (PETG), a competitive inhibitor of β-gal activity, (2) the inhibition of interfering mammalian β-galactosidases by the weak base chloroquine, and (3) the use of two-color flow cytometry measurements to reduce detection of false positives. The protocol requires expertise in flow cytometric techniques and cell handling techniques (see APPENDIX 3B), and special attention to the kinetics of substrate hydrolysis (see Critical Parameters and Troubleshooting, discussion of quantitation). Materials Single-cell suspensions of transfected and control cultures Staining medium (see recipe) 2 mM fluorescein di-β-D-galactopyranoside (FDG) working solution (see recipe) Quenching solution (see recipe) 50 mM phenylethyl β-D-thiogalactopyranoside (PETG; Sigma; prepare in H2O and store indefinitely at −20°C) 12 × 75–mm polystyrene tubes (Falcon) Flow cytometer with sample chilling capability and 488-nm excitation source Additional reagents and equipment for immunophenotyping (UNIT 6.2)
Reporters of Gene Expression: Enzymatic Assays
Load cells with FDG 1. Harvest cells by appropriate method (e.g., trypsinize; APPENDIX 3B) and centrifuge 10 min at 1000 × g, 4°C, in 12 × 75–mm polystyrene tubes.
9.5.2 Supplement 6
Current Protocols in Cytometry
An appropriate number is 106 cells per sample; there is virtually no reasonable upper limit to the number of cells that can be used per sample. Always include a cell sample not expressing β-gal to establish the background “activity” levels in the cells being analyzed. As fibroblasts typically express high levels of almost any transfected β-gal construct, β-gal-positive fibroblasts (such as transfected NIH 3T3 cells) are ideal for establishing the assay as well as being a positive control in future experiments. The contribution of endogenous activity (mammalian lysosomal β-galactosidase) should first be assessed without chloroquine. If cells have relatively high endogenous activity, then chloroquine should be added at a final concentration of 300 ìM to the staining medium and the quenching solution.
2. Resuspend the cells in 50 µl staining medium (with chloroquine, if needed) and move to a 37°C water bath. Incubate for 20 min in the presence of chloroquine, or for 5 min in its absence (to equilibrate the temperature). 3. Warm 2 mM FDG working solution to 37°C and add 50 µl to each sample. Incubate 60 sec at 37°C, then add 1 ml ice-cold quenching solution and transfer the tube to 0°C. The timing of this incubation is critical; quenching must be done as close to 1 min after addition of FDG as possible. With a large number of samples, it is most convenient to have two persons perform the assay: one adds FDG every 10 to 15 sec, the other adds quenching solution to each sample 1 min later. When working with cells with high enzymatic activity, the quenched cells can be pelleted (at 4°C) and resuspended in 1 ml ice-cold quenching solution.
4. Incubate for an additional period on ice. Stop the reaction by adding 2 µl of 50 mM PETG. The length of time at 0°C, during which FDG continues to be hydrolyzed, depends on the activity of the cells. Typical reaction times are 10 to 120 min (for cells with 100 molecules of enzyme per cell, use a 120-min reaction time; for cells with more enzyme, use a shorter reaction time). If cells have very high activity, PETG can be included in the quenching solution.
5. Optional: Stain cells with monoclonal antibodies using standard methodology (e.g., UNIT 6.2). The cells must be kept at 0° to 10°C to prevent leakage of fluorescein from the cells. Although PETG is difficult to wash out of cells, inclusion of PETG during antibody staining and subsequent washes is recommended. If long incubations are desired in step 4, antibody staining can be done during this incubation. Repeated washing of the cells has no effect on the assay and FDG will continue to be hydrolyzed (in the absence of PETG).
Set up the flow cytometer 6. Set up flow cytometer for standard fluorescein detection (488-nm excitation laser; 530 ± 15 nm bandpass filter) and cell sorting. If two-color flow cytometry is possible, use a 575 ± 13 nm filter in front of the second photomultiplier tube (see Critical Parameters and Troubleshooting, discussion of background fluorescence). Dead cells can be excluded on the basis of propidium iodide (PI) fluorescence (excitation by 488-nm laser; emission at 562 to 588 nm). Fluorescence compensation between the fluorescein and PI channels can be used to reduce the contribution of autofluorescence and increase sensitivity of the assay. If cells are to be sorted, all solutions should be sterilized with a 0.22-ìm filter prior to the assay. When using flow cytometry to quantitate absolute enzyme levels, it must be remembered that there is a nonlinear relationship between enzyme content and generated fluorescence:
Studies of Cell Function
9.5.3 Current Protocols in Cytometry
Supplement 6
i.e., fluorescence per cell is proportional to roughly the square of the number of β-gal molecules per cell (see Anticipated Results).
Analyze/sort FDG-loaded cells 7. If possible, maintain cells on ice throughout the data collection period. Note the incubation time prior to analysis. If levels of activity are insufficient for detection, place cells back on ice and incubate to allow further hydrolysis of the substrate. 8. Resuspend sorted cells directly in the culture medium of choice. Washing out intracellular fluorescein is not necessary for viability. When the FACS-Gal assay is performed on cells without lacZ, there is a background of rare cells that are higher in fluorescence than the rest of the population and appear to be weakly positive for β-gal activity. These “rare false positives” are generally found at a frequency between 0.01% and 1%. They can reduce the effectiveness of the FACS-Gal assay as a selection system because they reduce purity when selecting for infrequent positive cells. Avoiding cells with low green fluorescence and above-average yellow fluorescence is the best approach for selecting true β-gal–expressing cells when they are very infrequent in the population (see Critical Parameters and Troubleshooting, discussion of background fluorescence). ALTERNATE PROTOCOL
USE OF FLUORESCEIN DI-â-D-GLUCURONIDE AND FLOW CYTOMETRY TO QUANTITATE â-GLUCURONIDASE ENZYMATIC ACTIVITY Although lacZ has proved to be of general utility in the study of developmentally regulated gene expression, β-gal activity has not been detected in lymphocytes isolated from a large number of independently generated transgenic or chimeric mice in which the lacZ gene had been inserted into the genome, even though under the control of promoters/enhancers known to be expressed in lymphocytes. This absence of activity has been described for lacZ under the control of tissue-specific or ubiquitous regulatory elements. The clear lack of β-gal expression in developing lymphocytes in many systems is an impediment to analysis of immune function. The E. coli gus gene can be used as an alternative to lacZ. The Gus assay protocol is nearly identical to that for β-gal, with the exception of the substrate (fluorescein di-β-D-glucuronide [FDGlcU]) and inhibitor (1,4-saccharolactone [1,4-SL]) employed. Although high levels of lysosomal β-glucuronidase activity have been reported for a number of mammalian tissue types, the authors find very low Gus activity in lysates of mammalian cell lines under the assay conditions described here. In combination with flow cytometry, the gus reporter gene system can be used both in quantitative expression studies and as a selectable marker. Furthermore, β-galactosidase and β-glucuronidase enzymes will hydrolyze only their cognate substrates, allowing for independent detection of these hydrolases within the same cell. Additional Materials (also see Basic Protocol) 2 mM fluorescein di-β-D-glucuronide (FDGlcU) working solution (see recipe) 250 mM 1,4-saccharolactone (1,4-SL; see recipe) Load cells with FDGlcU 1. Prepare cells for FDGlcU loading as for FDG (see Basic Protocol, steps 1 and 2). 2. Warm 2 mM FDGlcU working solution to 37°C and add 50 µl to each sample. Incubate 60 sec at 37°C, then add 1 ml ice-cold quenching solution and transfer the tube to 0°C.
Reporters of Gene Expression: Enzymatic Assays
If the cells to be analyzed have high levels of Gus activity, 1,4-SL can be added to the cell suspension at a final concentration of 0.1 to 5 mM depending on the level of activity. The inhibitors can be added with FDGIcU or quenching solution, also depending on the level of activity. This additional step slows the hydrolysis of FDGlcU, and maintains the correlation between fluorescence levels and enzymatic activity within the cell population.
9.5.4 Supplement 6
Current Protocols in Cytometry
3. Incubate for an additional period at 10°C. To stop the reaction, transfer cells to 0°C. In contrast to β-gal, Gus is not active at 0°C. However, the enzyme shows a reduction in activity of only 2-fold at 10°C versus 37°C. Thus, if hydrolysis beyond the first minute of loading is required, incubation can be carried out at or above 10°C. The length of time at 10°C depends on the activity of the cells and must be determined empirically.
4. Optional: Stain cells with monoclonal antibodies using standard methodology (e.g., UNIT 6.2). Cells must be kept at 0° to 10°C to prevent leakage of fluorescein from the cells.
Set up the flow cytometer and analyze/sort FDGlcU-loaded cells 5. Set up as for cells loaded with FDG (see Basic Protocol, step 6). If possible, maintain cells on ice throughout the data collection period. Note the incubation time prior to analysis. If levels of activity are insufficient for detection, return cells to 10°C and incubate to allow further hydrolysis of the substrate. 6. Resuspend sorted cells directly the culture medium of choice. Washing out intracellular fluorescein is not necessary for viability.
USE OF THE MUG OR MUGlcU ASSAY FOR INDEPENDENT DETERMINATION OF â-GAL OR GUS ACTIVITY
SUPPORT PROTOCOL
These assays are performed on bulk lysates of cells and rely on the conversion of the nonfluorescent β-gal substrate 4-methylumbelliferyl β-D-galactopyranoside (MUG) or the nonfluorescent Gus substrate 4-methylumbelliferyl β-D-glucuronide (MUGlcU) to the highly fluorescent 4-methylumbelliferone. The resulting solution can be measured on any fluorometer capable of UV excitation. The assay is designed for 96-well microtiter plates with measurement by a fluorescence plate reader. The major advantage of the plate reader is that several thousand assays can be done easily in a few hours. This allows for a high degree of replication of samples. Another advantage is its versatility with respect to cell preparation. This protocol describes three alternate methods. Cells can be dispensed from cultures into wells, or they can be cultured, stimulated or treated, and lysed directly in the 96-well microtiter plates, avoiding cell transfers prior to the assay. Additionally, cells can be taken presorted directly from the cell sorter. For batch screening of clones, in which precise quantitation (e.g., activity per cell) is less important, the assay can be modified to be performed directly in culture medium (Kerr et al., 1991). Additional Materials (also see Basic Protocol) Cell cultures: β-gal- or Gus-transfected and control 0.13% Triton X-100/Z buffer prepared fresh with 1 part 1% (v/v) Triton X-100/Z buffer stock and 7 parts Z buffer 250 mM 1,4-saccharolactone (1,4-SL; see recipe) 30 mM 4-methylumbelliferyl β-D-galactopyranoside (MUG) or 30 mM 4-methylumbelliferyl β-D-glucuronide (MUGlcU; see recipes) Z buffer (see recipe) Stop buffer (see recipe) 96-well flat-bottom microtiter plate Microtiter plate centrifuge Multichannel pipettor Fluorescence plate reader capable of excitation at 355 nm and emission at 460 nm (e.g., Fluoroskan II, Flow Labs) Studies of Cell Function
9.5.5 Current Protocols in Cytometry
Supplement 6
Prepare 96-well microtiter plate with 1,000 to 50,000 cells/well To grow cells directly in plate: 1a. Grow desired number of adherent or nonadherent cells in individual wells. For nonadherent cells only, centrifuge 10 min at 1000 × g, 4°C, in a microtiter plate centrifuge. The number of cells required per well depends on the activity per cell. The Fluoroskan can easily detect 106 molecules of β-gal per well in this assay. To determine the number of enzyme molecules per cell, include standard dilutions of purified enzyme (Sigma) on the same plate (dilute the enzyme into the same volume as the cells).
2a. Aspirate medium/supernatant and add 120 µl of 0.13% Triton X-100/Z buffer. To harvest cells from other carriers: 1b. Harvest appropriate number of cells. Centrifuge 10 min at 1000 × g, 4°C. 2b. Resuspend in 0.13% Triton X-100/Z buffer at the desired concentration. Pipet 120-µl samples into wells of a 96-well microtiter plate. To use presorted cells: 1c. Sort appropriate number of cells directly from cell sorter into separate wells. Centrifuge 10 min at 1000 × g, 4°C, in a microtiter plate centrifuge. Aspirate supernatant. 2c. Bring volume in each well to 120 µl with 0.13% Triton X-100/Z buffer. As an alternative to centrifugation, determine the volume in each well and add Triton X-100 in Z buffer to the cultures for a final volume of 120 ìl with a final Triton X-100 concentration of ∼0.1%. Ideally, the cells will have been cultured in <100 ìl of medium, such that the assay can be performed as detailed below (150 ìl final).
Perform assay 3. Add 15 µl of 50 mM PETG (for the β-gal assay) or 5 µl of 250 mM 1,4-SL (for the Gus assay) to one well per plate as a blank control. More inhibitor may be necessary to stop the reaction completely; a titration should be performed to determine the appropriate amount for each experiment.
4. Prepare 5× substrate by diluting 30 mM MUG or 30 mM MUGlcU 1:10 with Z buffer. 5× MUG and MUGlcU will remain in solution for only a limited time (∼30 to 60 min at room temperature).
5. Using a multichannel pipettor, add 30 µl of 5× substrate to each well. As the increase in fluorescence is directly proportional to time, note the times of addition of the substrate and of the stop buffer carefully.
6. Optional: To assess extent of substrate hydrolysis, read the plate on a fluorescence plate reader at 450 nm during the reaction. The reaction must be stopped before the substrate has been exhausted and before measurement of fluorescence by the plate reader saturates. On the Fluoroskan, saturation is at ∼5000 fluorescence units after the reaction has been stopped. The addition of stop buffer increases the fluorescence at least 6-fold; thus, the reactions should be stopped before the wells reach 1000 fluorescence units, unstopped. The reaction proceeds linearly with time for at least 96 hr (data not shown). To increase the reaction rate ∼4-fold, incubate at 37°C.
7. Using a multichannel pipettor, add 75 µl stop buffer to all wells. Reporters of Gene Expression: Enzymatic Assays
The stop buffer completely halts further hydrolysis of substrate and increases the pH to >10, as required for deprotonation to obtain maximal 4-methylumbelliferone fluorescence. The resulting fluorescence is stable and can be read several hours later, if needed.
9.5.6 Supplement 6
Current Protocols in Cytometry
8. Read fluorescence on a fluorescence plate reader at 450 nm. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Chloroquine, 30 mM Dissolve chloroquine (Sigma) in H2O and store up to several months at −20°C. Fluorescein di-â-D-galactopyranoside (FDG) working solution, 2 mM Stock solution: Dissolve 100 mM FDG (mol. wt. 657; Molecular Probes) in anhydrous dimethyl sulfoxide (DMSO) and store up to several months at −20°C. Working solution: Dilute stock solution 1/50 in ice-cold H2O in a 6-ml polystyrene tube (2 mM final) immediately before use. Keep on ice. Stock solution can be thawed and refrozen several times. A yellow working solution is indicative of contaminating fluorescein, which can cause high background staining and decrease the sensitivity of the assay. The solution can be laserbleached to alleviate this problem. Hold the polypropylene tube containing the substrate in front of a 488-nm laser such that the laser beam intersects the meniscus of the substrate solution within the tube. Continue holding the solution in the beam for 1 to 2 min, moving the tube to distribute the light throughout. Properly bleached solution should be clear. CAUTION: Wear protective eye wear when bleaching the substrate! Use only approved laser-blocking goggles that completely block laser light.
Fluorescein di-â-D-glucuronide (FDGlcU) working solution, 2 mM Prepare stock and working solutions, store, and bleach as for FDG (see recipe). FDGlcU (mol. wt. 685) is available from Molecular Probes.
4-Methylumbelliferyl â-D-galactopyranoside (MUG), 30 mM Dissolve MUG (mol. wt. 338; Sigma) at 10 mg/ml in dimethylformamide (DMF). Divide into aliquots and store up to several months at −20°C. Use glass containers, as DMF dissolves many plastics.
4-Methylumbelliferyl â-D-glucuronide (MUGlcU), 30 mM Dissolve MUGlcU (mol. wt. 352; Molecular Probes) at 10 mg/ml in dimethylformamide (DMF). Aliquot and store up to several months at −20°C. Use glass containers, as DMF dissolves many plastics.
Quenching solution Per 100 ml freshly prepared staining medium (see recipe), add: 1 ml 100 µg/ml propidium iodide (1 µg/ml final) 1 ml 30 mM chloroquine, if needed (300 µM final; see recipe) 0.2 to 2 ml 50 mM phenylethyl β-D-thiogalactopyranoside (PETG), if needed (0.1 to 1 mM final) Prepare 1 ml fresh per sample and keep on ice Propidium iodide and PETG are prepared as aqueous stocks and can be stored indefinitely at −20°C. See Critical Parameters and Troubleshooting for discussion of when to use PETG and chloroquine.
1,4-Saccharolactone (1,4-SL), 250 mM Dissolve 1,4-SL (mol. wt. 210; Aldrich) in H2O at 250 mM and store up to several months at −20°C. Studies of Cell Function
9.5.7 Current Protocols in Cytometry
Supplement 6
Staining medium 960 µl RPMI 1640 deficient in biotin, phenol red, and riboflavin 40 µl FBS (APPENDIX 2A; 4% final) 10 µl 1 M HEPES acid, pH 7.4 (10 mM final) 10 µl 30 mM chloroquine, if needed (300 µM final; see recipe) Prepare fresh PBS or complete RPMI (APPENDIX 2A) can also be used; however, the above formulation results in a reduction of fluorescence contributed by phenol red and riboflavin. See Critical Parameters and Troubleshooting for discussion of chloroquine.
Stop buffer 4.4 g EDTA (15 mM final) 22.5 g glycine (300 mM final) H2O to 1 liter Adjust pH to 11.2 with NaOH Store indefinitely at room temperature Z buffer 16.1 g Na2HPO4⋅7H2O or 21.5 g Na2HPO4⋅12 H2O (60 mM final) 5.5 g NaH2PO4⋅H2O (40 mM final) 0.75 g KCl (10 mM final) 0.246 g MgSO4⋅7H2O (1 mM final) H2O to 1 liter Adjust pH to 7.0 with NaOH or HCl Store indefinitely at room temperature COMMENTARY Background Information
Reporters of Gene Expression: Enzymatic Assays
â-Galactosidase Reporter genes have been used to analyze the expression of genetic control elements for several decades. Historically, reporter genes were measured by biochemical means. Recently, however, methodologies have been developed that utilize flow cytometry to detect and quantitate reporter gene expression (Nolan et al., 1988; Fiering et al., 1991). Flow cytometry enhances the use of reporter genes in three significant ways. First, when a functionally heterogeneous population of cells is analyzed by flow cytometry, the distribution of expression is resolved by virtue of the fact that gene expression is determined on a cell-by-cell basis. In contrast, conventional reporter analyses yield a single (biochemical) measurement of the bulk cell activity, which is assumed to be representative of the activity of every cell in the population. Second, the sorting capabilities of the cytometer can be used to viably isolate cells based on these distinguishable expression levels. This makes reporter genes double as nontoxic selectable markers, whereby selection can easily be performed on a range of expression levels rather than on a threshold. As analysis is
conducted on intact cells, sorting of cells can be followed by biochemical assays; thus, gene expression levels can be correlated with almost any desired parameter (e.g., RNA level, presence of DNA constructs by polymerase chain reaction, or metabolic enzyme activity). Third, the multiparametric nature of fluorescencebased cell sorting allows the simultaneous measurement of cell phenotypes such as cell size, surface antigens, DNA content, and calcium flux with reporter gene activity. Thus, the gene activity within a functionally or phenotypically heterogeneous population can be assigned to specific subsets of cells. The bacterial lacZ gene, encoding β-galactosidase (β-gal), is one of the most widely utilized reporters for a number of reasons: its enzymatic activity can be easily and sensitively measured, it can be expressed and assayed in virtually any type of cell, and its activity is unaltered by making N-terminal fusion polypeptides. Fluorescence-based flow cytometric analysis of β-gal activity (FACS-Gal) is performed using fluorescein di-β-D-galactopyranoside (FDG) as a substrate. FDG is nonfluorescent, but upon crossing the membrane of a β-gal-ex-
9.5.8 Supplement 6
Current Protocols in Cytometry
pressing cell, it is cleaved to yield the fluorescent molecule fluorescein. FACS-Gal has been used in a wide variety of systems to answer a number of questions and problems in modern biology. These include (1) control of gene expression by known promoters and enhancers (Emilie et al., 1989; Bierer et al., 1990; Chun et al., 1990; Fiering et al., 1990; Ko et al., 1990; Mattila et al., 1990; Roederer et al., 1990; Staal et al., 1990; Yancopoulos et al., 1990; Milan and Nicolas, 1991; Karttunen et al., 1992); (2) detection of novel developmentally or stress-regulated genes and enhancers (Brenner et al., 1989; Kerr et al., 1989, 1991; Reddy et al., 1991); (3) expression of a reporter gene in untransformed cells in vivo (Kerr et al., 1989, 1991; Strair et al., 1990; Zenke et al., 1990; Krasnow et al., 1991; Brombacher et al., 1994; Abe et al., 1996) and in embryonic stem cells (Mansour et al., 1990), or to follow localization of tumor cells (Kruger et al., 1994); (4) isolation of antigen-responsive T cell hybridomas (Karttunen et al., 1992) and recombinase-active B cells (Yancopoulos et al., 1990); (5) development of cell lines responsive to viral particles (Rocancourt et al., 1990; Milan and Nicolas, 1991); (6) study of antiviral agents (Savatier et al., 1989; Roederer et al., 1990; Staal et al., 1990); (7) study of gene expression during embryonic development of Drosophila (Krasnow et al., 1991); and (8) quantitation of DNA transfection methods (Zenke et al., 1990; Schachtschabel et al., 1996; Floch et al., 1997, 1998), viral infection (Saalmuller and Mettenleiter, 1993; Gojo et al., 1996), and of episomal DNA replication (Heinzel et al., 1991). FACS-Gal has been successfully used in a number of cell types other than mammalian cells, including chicken (Zenke et al., 1990), Drosophila (Krasnow et al., 1991), yeast (Nir et al., 1990), and bacteria (Nir et al., 1990; Russo-Marie et al., 1993). There are two other commonly used biochemical assays for β-gal activity. Each has advantages and disadvantages. For assays on bulk lysates of cells, the 4-methylumbelliferyl β-D-galactopyranoside (MUG) assay is the most convenient and sensitive. A protocol for this assay is included in this unit (see Support Procotol). It is especially useful for screening large numbers of cell aliquots or clones growing in 96-well microtiter plates, as it is suitable for use with a fluorescence plate reader. Because it is a nonviable assay, plates must be replicated prior to assay if cell recovery is desired; for more details, see Kerr et al. (1991).
The two assays performed on intact cells are the 5-bromo-4-chloro-3-indolyl β-D-galactopyranoside (Xgal) assay and the FACS-Gal assay. Xgal is considerably less sensitive, with cells requiring ∼800 molecules of enzyme to appear blue. It is also a nonviable stain, requiring cells to be fixed. The advantage of Xgal is that subcellular localization of the β-gal can be ascertained; this is useful if localization sequences have been introduced on the lacZ coding sequence. Such localization is not possible with FDG, as fluorescein is diffusible. The FACS-Gal assay can detect lacZ-expressing cells that contain as few as five molecules of β-gal; thus, it is conceivable that the assay could detect a cell with a single lacZ mRNA molecule. The upper limit of quantitation by FACS-Gal is ∼10,000 molecules per cell (because cells will hydrolyze all FDG before even the loading step is finished). With the addition of phenylethyl β-D-thiogalactopyranoside (PETG) to slow the reaction, cells expressing in excess of 106 molecules of enzyme can be analyzed. â-Glucuronidase The E. coli gus gene, which encodes β-glucuronidase, has been used extensively as a reporter gene in plants (Jefferson et al., 1986; Jefferson, 1989), because the majority of plants tested have no endogenous cellular glucuronidase activity. Although high levels of lysosomal β-glucuronidase activity have been reported for a number of mammalian tissue types (Paigen, 1989), very low endogenous Gus activity has been found in several mammalian cell lines using the assay conditions described here (Lorincz et al., 1996). Several properties of the Gus reporter system are useful in studies of mammalian gene expression. First, the gus gene is only 1800 bp in size (versus the 3400-bp lacZ gene), which allows it to be used in DNA constructs with functional size constraints, such as retroviral vectors. Second, the Gus enzyme, like β-gal, is extremely stable under a variety of conditions—e.g., in the presence of 0.1% (v/v) Triton X-100—and over a broad pH range (Roederer et al., 1991). Third, although the Gus enzyme is tetrameric in its native state, it retains enzymatic activity when fused to the C-terminus of heterologous polypeptides (Jefferson et al., 1987), much like β-gal (Norton and Coffin, 1985), allowing for translational fusions to genes of interest in heterologous promoter expression studies.
Studies of Cell Function
9.5.9 Current Protocols in Cytometry
Supplement 6
The authors have found that the Gus enzyme has a stronger temperature dependence than β-gal, with almost no detectable activity at 0°C (see Critical Parameters and Troubleshooting, discussion of substrate loading). Furthermore, under the same conditions, the FDG substrate is loaded into cells at significantly higher concentrations than the fluorescein di-β-D-glucuronide (FDGlcU) substrate. As the KM values of β-gal and Gus versus their cognate fluorescein substrates are 17 µM and 133 µM, respectively, the sensitivity of the β-gal system is likely to be higher than that of the Gus reporter system. Thus, unless other factors preclude the use of β-gal, the FACS-Gal system should be tested first. An independent assay for Gus activity, using the substrate 4-methylumbelliferyl β-D-glucuronide (MUGlcU), is described in the Support Protocol. This assay is parallel to the MUG assay for β-gal (see discussion of β-galactosidase, above).
Reporters of Gene Expression: Enzymatic Assays
â-Lactamase Recently, Zlokarnik et al. (1998) described the use of a new reporter gene, β-lactamase, for fluorescence imaging analysis and flow cytometric analyses. These authors also describe the use of a resonance energy tandem (RET) fluorogenic substrate. A RET substrate is a covalent linking of two distinct fluorophores, in which the energy of excitation of one substrate (the donor) is transferred nonradiatively to the other (the acceptor) for fluorescent emission. The complex thus has the excitation spectrum of the donor but the emission spectrum of the acceptor. In the Zlokarnik application, a coumarin molecule was covalently linked to a fluorescein molecule through a linker that could be hydrolyzed by β-lactamase. This substrate has the UV excitation spectrum of coumarin and the visible (green) emission spectrum of fluorescein: i.e., an emission at 530 nm when excited in the UV. However, once the link is hydrolyzed, the two fluorophores are separated and the cell acquires both independent fluorescence spectra: with emission at 460 nm (from the coumarin) when excited in the UV and at 530 nm (from the fluorescein) when excited at 488 nm. Thus, the rate of hydrolysis (proportional to β-lactamase content) is reflected by the conversion of the RET fluor into the separate components, and can be most sensitively and quantitatively measured by determining the ratio of UV-excited emissions at 530 nm (unhydro-
lyzed RET substrate) to that at 460 nm (hydrolyzed free coumarin). In terms of sensitivity, the β-lactamase system approaches that for β-galactosidase, with the capability to distinguish cells expressing as few as 50 molecules of the enzyme. An advantage of the system is that hypotonic loading is not required; the substrate freely diffuses into cells. However, the accuracy of quantification was not reported; in contrast, an advantage of the hypotonic-shock loading method employed in the protocols above is that there is relatively small variation in the amount of substrate per cell, resulting in high accuracy as well as precision of the measurement. A second disadvantage of the RET fluorogenic system is the need for two colors to assay enzyme activity, although measuring only the decrease in the RET emission should provide a reasonable measure of conversion (with lower sensitivity). Finally, the potential effect of relative differences in leakage of RET and of free coumarin has not been evaluated. Use of reporter genes as selectable markers Reporter gene analyses based on sorting of fluorescence-labeled cells have two major advantages over other reporter gene methods: the ability to determine the distribution (and possible heterogeneity) of expression of the reporter gene within a population, and the use of the flow cytometer to sort (select) desired cells from within a population without killing either subpopulation, thereby making the reporter genes nontoxic selectable markers. Selectable markers are genes that allow the selection of uncommon cells expressing a particular phenotype encoded by that gene. Selection by flow cytometer has several advantages over standard methods used with drug-based selectable markers. Generally, selectable marker systems rely on a combination of toxic compounds and genes that encode resistance or sensitivity to the toxic compound. The systems work by killing or halting the growth of all cells that do (or do not) produce some threshold amount of the protein encoded by the selectable marker gene. There are multiple disadvantages of such a survival-based selection system. Because the selection system is inherently toxic, nonexpressing cells are dead and unavailable for further study, and the selection environment is often toxic even to cells expressing the resistance gene. Furthermore, within the selected population, it is difficult or impossible to select for different levels of the marker expression.
9.5.10 Supplement 6
Current Protocols in Cytometry
Finally, the cells must be able to reproduce in culture and selection generally takes weeks. On the other hand, flow cytometry allows the isolation of cells based on virtually any level of expression of the nontoxic β-gal or Gus enzymes. This provides capabilities that are not available using standard drug selection methods. For example, cells that rapidly change expression of the reporter gene in response to a stimulus are viable in both states and can be selectively sorted. Additionally, all sorted cells are available for the concomitant determination of other parameters (e.g., by biochemical assays of lysates). For instance, by sorting cells with different expression levels and then quantifying mRNA levels in the sorted cells, a comparison of message levels within defined subsets of a population can be made; for example, see Fiering et al. (1990). The flow cytometer can analyze and sort from as many as 5000 cells per second; thus, 1 cell in 10 million can be selected in an hour of sorting, making cytometry as efficient as selection using drug-based selectable markers. Sensitivity is also very good. Cells expressing very low levels of β-gal (e.g., ten molecules) that might not be sufficient to confer drug resistance can be detected and sorted by flow cytometry. Use of reporter genes in transgenic mouse tissues A basic technique of gene expression studies is isolation of regulatory elements that control expression of a given gene, and attachment of these elements to a reporter gene. Mammalian in vivo gene expression studies in the last decade have frequently utilized transgenic mice. Studies of the control of gene expression in the functioning immune system have been hampered by a lack of appropriate reporter gene systems (i.e., those that do not interfere with function and are sufficiently sensitive). Because FACS-Gal combines the measurement of gene expression with cell identification—e.g., cell surface immunophenotyping (Kerr et al., 1991)—it would seem to be ideal for studying gene expression in vivo. Although β-gal activity has been detected in the lymphocytes of several murine transgenic strains with the lacZ gene under the control of endogenous promoters with ubiquitous expression patterns (Weintraub et al., 1994), the authors and others have found that lacZ is difficult to express in the immune system of transgenic mice (Fiering, 1990; Roederer et al., 1991). This is completely unexpected, because lacZ has been expressed in a wide variety of
(nonhematopoeitic) tissues of transgenic mouse embryos from many different constructs. The lacZ gene has been expressed in some adult tissues, but often only under the control of very strong promoters. Although lacZ remains a useful reporter gene for many tissues in the mouse, its use in stable expression studies in lymphoid tissues is not advisable. As the gus gene shares only limited homology with lacZ, and the Gus enzyme has a different glycolytic substrate specificity from that of β-gal, it is likely that the Gus reporter system (see Alternate Protocol) will circumvent the limitations ascribed to lacZ in hematopoietically derived cells.
Critical Parameters and Troubleshooting Substrate loading The FACS-Gal and FACS-Gus assays depend on the introduction of nonfluorescent fluorogenic substrates (FDG and FDGlcU, respectively) into the cytoplasm of cells. Hydrolysis of either substrate generates two monosaccharides and the highly fluorescent fluorescein. The resulting fluorescent product is trapped intracellularly and can be measured in individual cells using a cell sorter. The rate of fluorescence accumulation, equivalent to the rate of hydrolysis, is presumed to be proportional to the cellular concentration of the reporter enzyme, as long as the substrate has not been exhausted. The loading procedure is designed both for optimal loading of the substrate (FDG) and for retention of the hydrolyzed fluorescein product. Fluorescein is membrane permeable at 37°C, at which temperature it is lost from cells with a half-time of ∼3 min (Nolan et al., 1988), but not at 0°C, at which temperature cells do not lose significant fluorescence even over a 17-hr period (Fiering et al., 1991). The amount of FDG loaded into cells is roughly proportional to the length of time that cells are exposed to the hypotonic medium (after a short lag phase; Fig. 9.5.1). For most cell types, there is no significant loss in viability with treatment times as long as 3 min. However, a 1-min treatment is generally adequate, because it is sufficiently long to load enough FDG or FDGlcU for the sensitivity required by most systems, and sufficiently short to prevent the product from leaking out of cells. This can be a concern in cell lines with high activity during longer loading times, as hydrolysis begins immediately upon introduction of substrate. The 1-min loading technique results in a very uni-
Studies of Cell Function
9.5.11 Current Protocols in Cytometry
Supplement 6
form loading of substrate, with <3-fold cell-tocell variation in the amount of FDG (Fiering et al., 1991). Some of this variation may be cell volume–dependent. The other parameters affecting the amount of FDG loaded into cells are the degree of hypotonicity and the concentration of the substrate (Fig. 9.5.1). The greater the hypotonicity, the greater the amount of FDG loaded. However, viability can be affected below 50% isotonicity. The amount of FDG loaded is also directly proportional to the concentration of
FDG during the hypotonic treatment. The authors have estimated the intracellular concentration of FDG to be 5 µM when the standard loading conditions (i.e., 1 min, 50% hypotonic treatment, at 1 mM FDG) are used (Fiering et al., 1991). After the cells are loaded with FDG and quenched at 0°C (to restore isotonicity), hydrolysis by β-gal proceeds linearly with time. In fact, hydrolysis can continue for at least 17 hr if the concentration of β-gal is low enough (Fiering et al., 1991). During this time, cells can be stained for other types of analyses (i.e.,
A
B
Final fluorescence
300 400 200 200
100
0
0 0 50 100 150 200 Duration of hypotonic treatment (sec)
Fluorescence per min
C
20
40 60 80 100 Percent of isotonicity
10
5
0 0
Reporters of Gene Expression: Enzymatic Assays
1
2 3 [FDG], mM
4
5
Figure 9.5.1 Biochemistry of FACS-Gal. Cells from a line uniformly expressing high amounts of β-gal were loaded with FDG under a variety of conditions. The incubation was long enough to assure complete hydrolysis of the substrate; thus, differences in the final fluorescence reflect differences in substrate loading. (A) The hypotonic loading step was stopped after different time points. After a 30-sec lag during which no substrate enters the cells, a linear increase in substrate loading is observed. The standard loading time is 60 sec. Increased sensitivity can be achieved by using longer loading times, but this may decrease cell viability. Shorter loading times are preferable for cells with high activity to prevent product release during the 37°C loading procedure. Alternatively, PETG can be used to slow the reaction. (B) The amount of substrate loaded is proportional to the degree of hypotonicity during the loading step. The standard condition is 50%. If reduction in cell viability is unacceptable, then a higher isotonicity can be used, but less substrate will enter the cells. (C) The final fluorescence is linearly dependent on the FDG concentration during loading, indicating that the concentration achieved inside the cells is well below the KM for the enzyme. Concentrations >4 mM are difficult to achieve because of the limited solubility of FDG. Figure adapted from Fiering et al. (1991).
9.5.12 Supplement 6
Current Protocols in Cytometry
surface staining with fluorescent monoclonal antibodies or staining with DNA-specific dyes). As long as the cell membrane is intact and the temperature of the cells is maintained below ∼10°C, the fluorescein will not leak from the cytoplasm. The FACS-Gus system behaves similarly, except that the Gus enzyme is not active at 0°C. However, the enzyme shows only a 2-fold reduction in activity at 10°C versus 37°C (Fig. 9.5.2). Thus, if hydrolysis beyond the first minute of loading is required, incubation can be carried out at or above 10°C. The two considerations to be addressed when optimizing a FACS-Gal or FACS-Gus experiment are the requisite sensitivity and the effect on viability. Sensitivity (i.e., the amount of fluorescence over background) can be increased by increasing the amount of substrate loaded, increasing the hypotonic treatment time, increasing the degree of hypotonicity, or increasing the substrate concentration (and thus DMSO) in the hypotonic treatment. Higher sensitivity can also be achieved by increasing the time of incubation after quenching. How-
ever, each of these steps could adversely affect viability; thus, if sensitivity is not an issue, more gentle loading conditions can be utilized. Background fluorescence There are three common sources of background that can detract from the sensitivity of cytometry-based assays. The first is cell autofluorescence. Sensitivity can be significantly improved through autofluorescence compensation (see below). The second is the presence of endogenous β-gal and Gus activities, contributed by endogenous hydrolases that localize to endosomes and lysosomes. The third background activity is the generation of “rare brights,” rare cells that falsely appear to have reasonably high reporter enzyme activities. Autofluorescence compensation An excellent method for increasing sensitivity of reporter gene assays where the measured fluorescence is only slightly above autofluorescence is autofluorescence compensation. Roederer and Murphy (1986) and Alberti et al. (1987) showed that standard compensation
MU fluorescence (relative)
2.0 70Z/3 Gus+ (r = 0.95) 70Z/3 parental (r = 0.98)
1.5
1.0
0.5
0 0
5
10
15
20
25
Temperature ( C)
Figure 9.5.2 Temperature dependence of β-glucuronidase. 70Z/3 MFG-Gus6 cells were lysed in Z buffer and titrations of cell extract were equilibrated to 0°, 10°, 14°, 18°, and 25°C. MUGlcU was added to a final concentration of 0.6 mM. At 30, 90, and 240 min, aliquots were removed from each titration point (equivalent to 2.5 × 105, 1.25 × 105, and 6.25 × 104 cell equivalents) and the reactions were stopped by the addition of ice-cold stop buffer. Each point on the graph represents the reaction rate determined by measuring 4-methylumbelliferone (MU) fluorescence at each time point on a Fluoroskan fluorometer, normalized to 2.5 × 105 cell equivalents for all samples. 70Z/3 parental (gus) cells were also measured to determine endogenous β-glucuronidase activity at 0°, 14°, and 25°C. Vertical bars denote the standard deviation of the fluorescence levels normalized to 2.5 × 105 cell equivalents. Because of the lack of activity at 0°C, incubation after the loading step must be done at a permissive temperature that does not allow the product to leak from cells—i.e., 10°C. Figure adapted from Lorincz et al. (1996).
Studies of Cell Function
9.5.13 Current Protocols in Cytometry
Supplement 6
Figure 9.5.3 (right) Increasing sensitivity for reporter gene detection by compensation for autofluorescence and cell volume. Highly autofluorescent cells can overlap with cells expressing low levels of a reporter gene. In addition, larger cells will generally load with more substrate, and therefore with more of any contaminating product present in the substrate. By estimating autofluorescence in a second channel, the contribution of autofluorescence in the measurement channel can be removed. These panels show the increased sensitivity obtained by compensation for these effects. Only a few cells actually express the reporter gene. In each case, the gate region was drawn based on a similarly analyzed negative control sample. (Top) Uncompensated signal from the reporter gene and autofluorescence (left) and side scatter (right). Note the excellent correlation between the fluorescence in the reporter gene channel (product fluorescence) and each of these parameters; this is a requirement for compensation to work optimally. (Middle) Compensation with the autofluorescence channel removes the correlation between the product fluorescence and the autofluorescence; some size dependence remains. A slight increase in the positive population is found compared to uncompensated calculations. (Bottom) Full three-way compensation for size and autofluorescence results in removing all correlation between the product fluorescence signal of the negative population and either autofluorescence or side scatter. The result is a much better discrimination of the true positive population; nearly 5% of the cell population is actually positive. Thus, 1.7% of the positive population overlapped with the negative population in the uncompensated collection (top). Note: The diagonal streaks in the bottom right cell cluster (arrow) are a consequence of digital compensation and do not affect quantification in any way.
Reporters of Gene Expression: Enzymatic Assays
techniques could be used to estimate the autofluorescence contribution in the measurement channel based on the autofluorescence emission in a different channel. Thus, autofluorescence is treated as another fluorophore. In a typical experiment using fluorescein-based substrates (or even GFP), the reporter fluorescence is collected using a filter centered at about 530 nm. The autofluorescence can then be estimated by measuring fluorescence at 580 nm. Based on the proportionate relationship between autofluorescence at 580 nm and at 530 nm, the contribution of autofluorescence at 530 nm can be removed. Alternatively, Roederer and Murphy (1986) showed that an excellent estimate of autofluorescence can be obtained from the side-scatter profile: cells with more autofluorescence tend to be larger. For substrate-based assays, it is also true that larger cells will take up more substrate than smaller cells. Since many substrates are contaminated with hydrolyzed product, the net result is that larger cells will tend to have more background fluorescence. This can be corrected for by incorporating the side-scatter measurement into the compensation. Figure 9.5.3 shows how this compensation can be achieved. For flow cytometers that do not allow direct compensation between side scatter and a fluorescence channel, the compensation must be performed using software (e.g., FlowJo from Tree Star) after data collection. The ability to deconvolute the size and autofluorescence contributions to the reporter measurement aids in the discrimination of very dull populations that overlap with nonexpress-
ing, large, and/or autofluorescent cells. Therefore, the percentage of positive events increased from 3.2% to almost 5%, reflecting better discrimination of low-intensity events. Endogenous hydrolases The lysosomes of most mammalian cells possess endogenous β-gal and Gus activities that have an acidic pH optimum (lysosomal pH being ≤5). Because they can hydrolyze the reporter substrates, these enzymes contribute to background activity (nonreporter-dependent hydrolysis of the substrates). The amount of endogenous activity varies considerably by cell type. In general, lymphocytes and lymphocytic cell lines have very low activity, whereas macrophage and adherent cell lines tend to have higher activity, and primary fibroblasts can have very high activity. Culture conditions can also affect endogenous activity. Because confluent cells accumulate lysosomes (Roederer et al., 1989), cells that have been maintained at high density or allowed to reach confluency tend to have high activity (Fig. 9.5.4). Cells that have been serum-starved or otherwise “mistreated” also tend to accumulate lysosomal activity and thereby have higher endogenous activity. Endogenous activity contributes to the background present in all cells in a population. This reduces the signal-to-background ratio, thus reducing the sensitivity to reporter enzyme activity. The background activities of lysosomal hydrolases can be reduced by taking advantage of their relatively acidic pH optimum. Treatment of cells with a lysosomotropic weak base
9.5.14 Supplement 6
Current Protocols in Cytometry
Side scatter
Autofluorescence
3.1%
3.2%
4.8%
Product fluorescence
Studies of Cell Function
9.5.15 Current Protocols in Cytometry
Supplement 6
chloroquine varies from experiment to experiment, but not within an experiment (Fiering et al., 1991).
such as chloroquine will raise the pH of acidic compartments (chloroquine freely crosses cell membranes, unless protonated; thus, it accumulates in acidic vesicles and buffers the pH toward neutrality). In the assay illustrated in Figure 9.5.4, treatment of cells with chloroquine significantly reduces the hydrolysis of FDG by the endogenous β-gal. Because chloroquine does not inhibit hydrolysis of FDG by the lacZ-encoded β-gal, or of FDGlcU by Gus, its use increases the signal-to-background activity in cell types where endogenous activity is a significant fraction of total hydrolytic activity. The inhibition of endogenous activity by
Rare brights Another type of background activity, the presence of “rare brights,” reduces the ability to detect and sort infrequent lacZ-positive cells. When the FACS-Gal assay is done on cells without lacZ, there can be a small fraction (0.01% to 1%) of rare cells that are higher in fluorescence than the rest of the population (Fig. 9.5.5). Why or how these rare brights arise is not known; however, they occur only when
Relative cell number
A
B
low density high density
100 1 10 Endogenous β- galactosidase activity (relative fluorescence)
Percentage of endogenous β- galactosidase activity
100
cell line 293
80
NIH 3T3
60 40 20 0 0
33
100
300
Chloroquine concentration (µM)
Reporters of Gene Expression: Enzymatic Assays
Figure 9.5.4 Endogenous hydrolase activity. (A) Cells that do not express the bacterial lacZ gene were assayed by FACS-Gal under standard conditions. Cells grown at high density to confluence showed considerably higher background activity from mammalian β-gal than cells grown in log phase (low density) due to an accumulation of hydrolytic compartments such as lysosomes (Roederer et al., 1989). (B) The background activity can be partially inhibited by preincubating cells with chloroquine, which neutralizes the acidic pH of lysosomal compartments. The chloroquine does not inhibit the activity of transduced β-gal, which has a neutral pH optimum for its activity. Figure adapted from Fiering et al. (1991).
9.5.16 Supplement 6
Current Protocols in Cytometry
Yellow fluorescence (562-588 nm)
cells are loaded with FDG and are therefore substrate-induced artifacts. The presence of rare brights decreases the purity with which lacZ-expressing cells can be sorted. However, successive sorts (after cells have been grown and reassayed) will enrich for the true lacZ-expressing cells because the progeny of sorted rare brights (which are viable) are not themselves enriched for rare brights: i.e., this property is not heritable (Kerr et al., 1989). FACS-Gal has been used to select for lacZ-expressing cells occurring at a frequency of <0.1%, even in the presence of rare brights (Yancopoulos et al., 1990; Kerr et al., 1991). Two fluorescence characteristics of rare brights help differentiate them from lacZ-expressing cells (Fig. 9.5.5): they are generally less fluorescent than cells with enough activity to have completely hydrolyzed the loaded FDG, and they tend be brighter than negative cells for emission in both the yellow wavelength band (562 to 588 nm) and the green (fluorescein) band (515 to 545 nm). The green
emission from lacZ-positive cells does not correlate with the yellow emission if compensation is appropriately adjusted for fluorescein. Thus, the highest ratio of lacZ-positive cells to rare brights is obtained by sorting cells with high 530-nm emission and low 575-nm emission (Fig. 9.5.5). The presence of rare brights is cell type–dependent and is weakly correlated with endogenous activity. Adherent cells, which typically have higher endogenous activity, include more rare brights. The percentage of rare brights is increased in high-density or confluent cultures. For FACS-Gal assays, the authors find that it is advisable to maintain cultures in exponential growth, because these growth conditions minimize both endogenous activity and the frequency of rare brights. Quantification At this stage, absolute quantification—the calibration of cellular fluorescence into an absolute number of protein molecules per cell—is
0.1% false positives
100
10
1
1
10 100 Green fluorescence (fluorescein) (515-545 nm)
Figure 9.5.5 Rare brights artifactually appear as hydrolase-positive cells. Under certain conditions, cells can exhibit artifactually positive activity for the FACS-Gal assay. When sorted and cultured, these cells do not subsequently have high activity upon reassay. Although these cells interfere with the ability to recover rare positive cells, they have a different spectral emission from true positives; i.e., they tend to have correlated yellow fluorescence whereas true positives have only green fluorescence. Thus, optimal purity for sorting can be achieved by using a sort gate as shown by the shaded panel. Because of the potential contamination by rare brights, isolation of low-frequency lacZ-expressing cells may require multiple rounds of enrichment (Kerr et al., 1991). Figure adapted from Fiering (1990).
Studies of Cell Function
9.5.17 Current Protocols in Cytometry
Supplement 6
Reporters of Gene Expression: Enzymatic Assays
not possible with GFP. However, it is reasonable to assume that the fluorescence measured above background autofluorescence is linearly proportional to the number of GFP molecules per cell. Thus, relative quantitation is simple. However, the sensitivity of the GFP assay has not yet been adequately characterized. It is not yet known exactly how active a promoter must be in order to provide minimum detectable level of protein, or what that detectable level is. In order for FACS-Gal or FACS-Gus to quantitatively measure β-gal or Gus activity on a per cell basis, several criteria must be met: (1) the substrate must be uniformly and rapidly loaded, (2) the product must not leak from the cells, (3) hydrolysis of the substrate must proceed at a constant rate such that the final fluorescence can be related to absolute reporter enzyme activity, (4) fluorescence must be measured at a known time after addition of the substrate, (5) fluorescence must be measured prior to the exhaustion of substrate by any cells in the population, and (6) appropriate controls for background activity must be performed. Because these are kinetic assays, precise timing of incubations is required. When dealing with a large number of samples, it is too difficult to analyze each sample at known times with respect to substrate addition. Instead, the reaction is stopped for all samples at the same time relative to the addition of substrate. Samples can be analyzed as much as several hours later. As the Gus enzyme is inactive at 0°C, the FACS-Gus assay can be stopped simply by placing the samples on ice. For β-gal, the use of a competitive inhibitor to stop the hydrolysis of FDG is an important adjunct to the protocol. Phenylethyl β-D-thiogalactopyranoside (PETG) is ideal for this purpose for several reasons (Fiering et al., 1991). It is sufficiently hydrophobic to enter cells rapidly even at 0°C, but is sufficiently hydrophilic to dissolve stably in aqueous solution. It is a competitive inhibitor of lacZ-encoded β-gal, with a low Ki (concentration at half-maximal inhibitions; 2.5 µM). It has a sulfur atom in place of an oxygen atom at the hydrolytic site, which makes it nonhydrolyzable by glycosidases. It is a reversible inhibitor and can be washed out of the cells at room temperature or above, allowing β-gal activity to proceed (for unknown reasons, it cannot be washed out of cells at 0°C, even though it readily enters cells at that temperature; Fiering et al., 1991). It is nontoxic, allowing viable cell sorting, and it is inexpensive and readily available. PETG only slightly inhibits mammalian lysosomal β-gal activity. Addition of PETG at
1 mM final concentration to cells hydrolyzing FDG completely stops further hydrolysis (Fiering et al., 1991), thereby “freezing” the reaction at that time point. Halting enzyme reactions facilitates analysis for other situations besides a large sample number. Cell sorts requiring a large amount of time normally require continual adjustment of sort gates as cell fluorescence continues to increase during the course of sorting. This can be avoided by stopping the reaction. Additionally, cells with high β-gal activity can be analyzed easily, as the reaction can be stopped immediately after loading (i.e., by including PETG in the quenching solution). Because a limited amount of substrate is loaded into cells, both assays have a limited range. Given enough time, cells will hydrolyze all available substrate into fluorescein, and will appear uniformly bright. Although useful for distinguishing reporter enzyme–positive from reporter enzyme–negative cells, saturation completely obviates quantitation of activity. The reaction must be stopped (or at least measured) before any cells reach saturation. Saturation can be determined by letting a reporter gene–expressing cell line incubate until fluorescence no longer increases. This point will vary for different cell lines, and depends on the amount of substrate loaded. For cells with very high levels of β-gal activity, accurate activity measurements are impossible, because the cells will have hydrolyzed all available FDG before the end of the 1-min loading time (i.e., before PETG can be added). This problem can be surmounted by adding concentrations of PETG to the loading mixture that slow the reaction. Because the Ki of PETG is 2.5 µM, the presence of 25 µM PETG in the loading and quenching solutions will slow the reaction ∼10-fold. PETG can be added later at a final concentration of 1 mM to essentially stop the reaction. Similar principles apply to the FACS-Gus assay, which utilizes the inhibitor 1,4-SL to maintain the Gus enzyme/fluorescein product correlation. Finally, accurate quantitation requires determination of the background fluorescence, which will be subtracted from sample fluorescences. Background fluorescence has three contributions: (1) autofluorescence, (2) free fluorescein contamination in the FDG or FDGlcU stocks, and (3) hydrolysis of these substrates by endogenous hydrolases. Each background can be determined with relative ease through the use of appropriate controls. For details on sources of autofluorescence or
9.5.18 Supplement 6
Current Protocols in Cytometry
A
B FACS-Gus fluorescence
Free fluorescein in the substrate stock is generally not a problem because very pure FDG and FDGlcU are commercially available (Molecular Probes). However, it is possible that old stocks of substrate can accumulate some hydrolyzed product. To test this, simply load cells with little or no endogenous activity in the presence of high concentrations of inhibitor (PETG for β-gal or 1,4-SL for Gus). The initial contribution to fluorescence from free fluorescein can be assessed by monitoring the cells
Relative cell number
endogenous hydrolases, or on how to reduce their effects on the assay, see above discussion of background fluorescence. Autofluorescence should be determined on cells not loaded with substrate. Autofluorescence compensation (Roederer and Murphy, 1986; Alberti et al., 1987) should be done so as to yield maximum sensitivity for the FACS-Gal assay. This is especially important with cells that have significant autofluorescence, such as fibroblasts or macrophages.
800 600 400 200 0
0
200 400 600 800 Relative fluorescein fluorescence (flow cytometry)
0
1 2 3 4 5 Relative Gus activity/ cell
FACS-Gal fluorescence
C 800 600 400 200 0 0
500
1000
1500
2000
Number of β- gal molecules per cell
Figure 9.5.6 Calibration of fluorescein fluorescence from flow cytometry with absolute hydrolase activity. (A) Subsets of cells from a hydrolase-transfected population are sorted according to their fluorescence. A few such “slices” are shown for a cell line expressing β-glucuronidase. This panel was collected with linear amplification. The sorted cells were then assayed for enzymatic activity using the methylumbelliferone assay. (B) The linear correlation of fluorescein fluorescence and methylumbelliferone fluorescence is shown for the Gus reporter. The line indicates the linear least-squares regression. There is a linear correlation between cell enzymatic content and fluorescein fluorescence. Below a certain level of enzyme content, no fluorescein fluorescence is generated. There is no obvious explanation for this lag in the amount of enzyme required before FDGIcU substrate hydrolysis occurs. (C) The same calibration performed on cells expressing lacZ. Interestingly, there is a strong nonlinear correlation between fluorescein fluorescence and enzyme content. This nonlinear dependence is described in more detail elsewhere (Fiering et al., 1991; Roederer et al., 1991).
Studies of Cell Function
9.5.19 Current Protocols in Cytometry
Supplement 6
over time, e.g., a couple of hours. Endogenous activity, if present, will result in increasing fluorescence as a function of time. The contaminating fluorescence is that found by extrapolating to time zero and comparing to samples that have been mock-loaded with substrate. Endogenous activity can be determined on a parental cell line that is reporter gene negative. As discussed above, chloroquine can be used to lessen the contribution of endogenous activity and increase the sensitivity of the assays. In general, the most accurate quantitation of β-gal activity is obtained by measuring the rate of fluorescence increase after quenching. This rate can be easily determined by removing aliquots of the quenched sample at various time points (e.g., 2, 5, 10, and 30 min) and adding them to tubes containing PETG. For flow cytometric analysis, it is probably best to use the median fluorescence for each time point, primarily because the median is less sensitive to cells that have exhausted the FDG or FDGlcU substrates.
Anticipated Results
Reporters of Gene Expression: Enzymatic Assays
As an example, a calibration of the FACSGal and FACS-Gus assays is presented in Figure 9.5.6. Calibration allows the correlation between fluorescence as measured by flow cytometry and the absolute enzyme content per cell, and is achieved by combining the sorting capabilities of the cytometer with the independent methylumbelliferone assay as described in the Alternate Protocol. Fibroblasts loaded with the relevant fluorescein substrate were processed for flow cytometric analysis. Several populations of cells were sorted on the basis of fluorescein fluorescence, such that each fraction included cells of a small range of fluorescence, covering the entire fluorescence range from dim to bright. Each sorted population was lysed and assayed by the methylumbelliferone assay. In Figure 9.5.6, the mean fluorescein fluorescence of each sorted sample is plotted against the enzyme content per cell for that sample. Such an approach should be used to verify the proper execution of flow cytometry–based reporter gene assays. To generate a standard curve, known amounts of purified Gus or β-gal enzyme are titrated into separate wells on the 96-well microtiter plate into which cells were sorted. Control enzymes and sorted cells are assayed simultaneously for conversion of MUG/MUGlcU to methylumbelliferone. After the reaction is stopped, the level of methylumbelliferone fluorescence generated in the control wells is used
to generate a standard curve, from which the number of reporter enzyme molecules in each sorted well can then be determined. As shown in Figure 9.5.6, there is a monotonic relationship between cellular enzyme content and the fluorescence measured by the flow cytometric assay. However, the precise relationships between enzyme content and activity are different for β-gal and Gus, neither following the prediction of Michaelis-Menten kinetics; future biochemical studies may elucidate the underlying mechanisms accounting for these unusual kinetics.
Time Considerations For the hydrolase assays, the incubation for fluorescence development can take anywhere from 15 min to several hours, depending on the desired sensitivity of the assay.
Literature Cited Abe, K., Hashiyama, M., Macgregor, G., and Yamamura, K. 1996. Purification of primordial germ cells from TNAPβ-geo mouse embryos using FACS-gal. Dev. Biol. 180:468-472. Alberti, S., Parks, D.R., and Herzenberg, L.A. 1987. A single laser method for subtraction of cell autofluorescence in flow cytometry. Cytometry 8:114-119. Bierer, B.E., Mattila, P.S., Standaert, R.F., Herzenberg, L.A., Burakoff, S.J., Crabtree, G., and Schreiber, S.L. 1990. Two distinct signal transmission pathways in T lymphocytes are inhibited by complexes formed between an immunophilin and either FK506 or rapamycin. Proc. Natl. Acad. Sci. U.S.A. 87:9231-9235. Brenner, D.G., Lin-Chao, S., and Cohen, S.N. 1989. Analysis of mammalian cell genetic regulation in situ by using retrovirus-derived “portable exons” carrying the Escherichia coli lacZ gene. Proc. Natl. Acad. Sci. U.S.A. 86:5517-5521. Brombacher, F., Schafer, T., Weissenstein, U., Tschopp, C., Andersen, E., Burki, K., and Baumann, G. 1994. IL-2 promoter-driven lacZ expression as a monitoring tool for IL-2 expression in primary T cells of transgenic mice. Int. Immunol. 6:189-197. Chun, K.T., Bar-Nun, S., and Simoni, R.D. 1990. The regulated degradation of 3-hydroxy-3methylglutaryl-CoA reductase requires a shortlived protein and occurs in the endoplasmic reticulum. J. Biol. Chem. 265:22004-22010. Emilie, D., Peuchmaur, M., Barad, M., Jouin, H., Maillot, M.C., Couez, D., Nicolas, J.F., and Malissen, B. 1989. Visualizing interleukin-2 gene expression at the single cell level. Eur. J. Immunol. 19:1619-1624. Fiering, S.N. 1990. A FACS-Based System for Gene Expression Studies that Uses E. coli lacZ as a Combination Reporter Gene/Selectable Marker. Doctoral thesis, Stanford University.
9.5.20 Supplement 6
Current Protocols in Cytometry
Fiering, S., Northrop, J.P., Nolan, G.P., Mattila, P.S., Crabtree, G.R., and Herzenberg, L.A. 1990. Single cell assay of a transcription factor reveals a threshold in transcription activated by signals emanating from the T-cell antigen receptor. Genes & Dev. 4:1823-1834. Fiering, S., Roederer, M., Nolan, G.P., Micklem, D.R., Parks, D.R., and Herzenberg, L.A. 1991. Improved FACS-Gal: Flow cytometric analysis and sorting of viable eukaryotic cells expressing reporter gene constructs. Cytometry 12:291-301. Floch, V., Le Bolch, G., Audrezet, M.P., Yaouanc, J.J., Clement, J.C., des Abbayes, H., Mercier, B., Abgrall, J.F., and Ferec, C. 1997. Cationic phosphonolipids as non viral vectors for DNA transfection in hematopoietic cell lines and CD34+ cells. Blood Cells Mol. Dis. 23:69-87. Floch, V., Audrezet, M.P., Guillaume, C., Gobin, E., Le Bolch, G., Clement, J.C., Yaouanc, J.J., des Abbayes, H., Mercier, B., Leroy, J.P., Abgrall, J.F., and Ferec, C. 1998. Transgene expression kinetics after transfection with cationic phosphonolipids in hematopoietic non adherent cells. Biochim. Biophys. Acta 1371:53-70. Gojo, S., Kitamura, S., Germeraad, W.T., Yoshida, Y., Niwaya, K., and Kawachi, K. 1996. Ex vivo gene transfer into myocardium using replicationdefective retrovirus. Cell Transplant. 5:S81-S84. Heinzel, S.S., Krysan, P.J., Tran, C.T., and Calos, M.P. 1991. Autonomous DNA replication in human cells is affected by the size and the source of the DNA. Mol. Cell Biol. 11:2263-2272. Jefferson, R.A. 1989. The GUS reporter gene system. Nature 342:837-838. Jefferson, R.A., Burgess, S.M., and Hirsh, D. 1986. β-Glucuronidase from Escherichia coli as a gene-fusion marker. Proc. Natl. Acad. Sci. U.S.A. 83:8447-8451. Jefferson, R.A., Kavanagh, T.A., and Bevan, M.W. 1987. GUS fusions: β-Glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO J. 6:3901-3907. Karttunen, J., Sanderson, S., and Shastri, N. 1992. Detection of rare antigen-presenting cells by the lacZ T cell activation assay suggests an expression cloning strategy for T cell antigens. Proc. Natl. Acad. Sci. U.S.A. 89:6020-6024. Kerr, W.G., Nolan, G.P., Serafini, A.T., and Herzenberg, L.A. 1989. Transcriptionally defective retroviruses containing lacZ for the in situ detection of endogenous genes and developmentally regulated chromatin. Cold Spring Harbor Symp. Quant. Biol. 54:767-776. Kerr, W.G., Nolan, G.P., Johnsen, J.B., and Herzenberg, L.A. 1991. In situ detection of stage-specific genes and enhancers in B cell differentiation via gene-search retroviruses. Adv. Exp. Med. Biol. 292:187-200. Ko, J.S.H., Nakauchi, H., and Takahashi, N. 1990. The dose dependence of glucocorticoid-inducible gene expression results from changes in the number of transcriptionally active templates. EMBO J. 9:2835-2842.
Krasnow, M.A., Cumberledge, S., Manning, G., Herzenberg, L.A., and Nolan, G.P. 1991. Whole animal cell sorting of Drosophila embryos. Science 251:81-85. Kruger, A., Schirrmacher, V., and von Hoegen, P. 1994. Scattered micrometastases visualized at the single-cell level: Detection and re-isolation of lacZ-labeled metastasized lymphoma cells. Int. J. Cancer 58:275-284. Lorincz, M., Roederer, M., Herzenberg, L.A., and Nolan, G.P. 1996. A FACS-based mammalian reporter gene system utilizing the Escherichia coli gus gene. Cytometry 24:321-329. Mansour, S.L., Thomas, K.R., Deng, C., and Capecchi, M.R. 1990. Introduction of a lacZ reporter gene into the mouse int-2 locus by homologous recombination. Proc. Natl. Acad. Sci. U.S.A. 87:7688-7692. Mattila, P.S., Ullman, K.S., Fiering, S., Emmel, E.A., McCutcheon, M., Crabtree, G.R., and Herzenberg, L.A. 1990. The actions of cyclosporinA and FK506 suggest a novel step in the activation of T-lymphocytes. EMBO J. 9:4425-4433. Milan, D. and Nicolas, J.F. 1991. Activator-dependent and activator-independent defective recombinant retroviruses from bovine leukemia virus. J. Virol. 65:1938-1945. Nir, R., Yisraeli, Y., Lamed, R., and Sahar, E. 1990. Flow-cytometry sorting of viable bacteria and yeasts according to β-galactosidase activity. Appl. Environ. Microbiol. 56:3861-3866. Nolan, G.P., Fiering, S., Nicolas, J.F., and Herzenberg, L.A. 1988. Fluorescence-activated cell analysis and sorting of viable mammalian cells b ased o n β-D-galactosidase activity after transduction of Escherichia coli lacZ. Proc. Natl. Acad. Sci. U.S.A. 85:2603-2607. Norton, P.A. and Coffin, J.M. 1985. Bacterial βgalactosidase as a marker of Rous sarcoma virus gene expression and replication. Mol. Cell. Biol. 5:281-290. Paigen, K. 1989. Mammalian β-glucuronidase: Genetics, molecular biology, and cell biology. Prog. Nucleic Acid Res. Mol. Biol. 37:155-205. Reddy, S., DeGregori, J., von Melchner, H., and Ruley, H.E. 1991. Retrovirus promoter-trap vector to induce lacZ gene fusions in mammalian cells. J. Virol. 65:1507-1515. Rocancourt, D., Bonnerot, C., Jouin, H., Emerman, M., and Nicolas, J.F. 1990. Activation of a βgalactosidase recombinant provirus: Application to titration of human immunodeficiency virus (HIV) and HIV-infected cells. J. Virol. 64:26602668. Roederer, M. and Murphy, R.F. 1986. Cell-by-cell autofluorescence correction for low signal-tonoise systems: Application to epidermal growth factor endocytosis by 3T3 fibroblasts. Cytometry 7:558-565. Roederer, M., Mays, R.W., and Murphy, R.F. 1989. Effect of confluence on endocytosis by 3T3 fibroblasts: Increased rate of pinocytosis and ac-
Studies of Cell Function
9.5.21 Current Protocols in Cytometry
Supplement 6
cumulation of residual bodies. Eur. J. Cell Biol. 48:37-44. Roederer, M., Staal, F.J.T., Raju, P.A., Ela, S.W., Herzenberg, L.A., and Herzenberg, L.A. 1990. Cytokine-stimulated human immunodeficiency virus replication is inhibited by N-acetyl-L-cysteine. Proc. Acad. Natl. Sci. U.S.A. 87:48844888. Roederer, M., Fiering, S., and Herzenberg, L.A. 1991. FACS-Gal: Flow cytometric analysis and sorting of cells expressing reporter gene constructs. Methods: A Companion to Methods Enzymol. 2:248-260. Russo-Marie, F., Roederer, M., Sager, B., Herzenberg, L.A., and Kaiser, D. 1993. β-Galactosidase activity in single differentiating bacterial cells. Proc. Natl. Acad. Sci. U.S.A. 90:8194-8198. Saalmuller, A. and Mettenleiter, T.C. 1993. Rapid identification and quantitation of cells infected by recombinant herpesvirus (pseudorabies virus) using a fluorescence-based β-galactosidase assay and flow cytometry. J. Virol. Methods 44:99-108. Savatier, N., Rocancourt, D., Bonnerot, C., and Nicolas, J.F. 1989. A novel system for screening antiretroviral agents. J. Virol. Methods 26:229236. Schachtschabel, U., Pavlinkova, G., Lou, D., and Kohler, H. 1996. Antibody-mediated gene delivery for B-cell lymphoma in vitro. Cancer Gene Ther. 3:365-372. Staal, F.J.T., Roederer, M., Herzenberg, L.A., and Herzenberg, L.A. 1990. Intracellular thiols regulate activation of nuclear factor κB and transcription of human immunodeficiency virus. Proc. Natl. Acad. Sci. U.S.A. 87:9943-9947. Strair, R.K., Towle, M., and Smith, B.R. 1990. Retroviral mediated gene transfer into bone marrow progenitor cells: Use of β-galactosidase as a selectable marker. Nucl. Acids Res. 18:47594762.
Yancopoulos, G.D., Nolan, G.P., Pollock, R., Prockop, S., Li, S.C., Herzenberg, L.A., and Alt, F.W. 1990. A novel fluorescence-based system for assaying and separating live cells according to VDJ recombinase activity. Mol. Cell. Biol. 10:1697-1704. Zenke, M., Steinlein, P., Wagner, E., Cotten, M., Beug, H., and Birnstiel, M.L. 1990. Receptormediated endocytosis of transferrin-polycation conjugates: An efficient way to introduce DNA into hematopoietic cells. Proc. Natl. Acad. Sci. U.S.A. 87:3655-3659. Zlokarnik, G., Negulescu, P.A., Knapp, T.E., Mere, L., Burres, N., Feng, L., Whitney, M., Roemer, K., and Tsien, R.Y. 1998. Quantitation of transcription and clonal selection of single living cells with β-lactamase as reporter. Science 279:84-88.
Key References Fiering et al., 1991. See above. Significant new additions to the original protocol and a more quantitative (and biochemical) characterization of the assay. Lorincz et al., 1996. See above. First detailing of the FACS-Gus system for mammalian cells. Nolan et al., 1988. See above. The first description of a flow cytometry–based assay for reporter gene activity.
Contributed by Matt Lorincz Fred Hutchinson Cancer Research Center Seattle, Washington Mario Roederer Stanford University Stanford, California
Weintraub, H., Soriano, P., and Zhuang, Y. 1994. The helix-loop-helix gene E2A is required for B cell formation. Cell 79:875-884.
Reporters of Gene Expression: Enzymatic Assays
9.5.22 Supplement 6
Current Protocols in Cytometry
Estimation of Membrane Potential by Flow Cytometry
UNIT 9.6
The procedures discussed in this unit estimate membrane potential (∆Ψ) by monitoring the distribution of fluorescent lipophilic dyes, such as the cationic cyanine series and the anionic oxonols, between cells and the suspending medium. Lipophilicity enables indicator molecules to pass freely through the lipid portion of the membrane; provided the concentration of the indicator is substantially lower than the concentrations of the ions that establish the potential, the concentration gradient of the indicator is determined by the potential difference across the membrane. Basic Protocol 1 describes flow cytometric measurement using any one of a number of dyes; see Background Information for more detailed discussion of these dyes along with a listing of pertinent spectral information. Basic Protocol 2 provides a more accurate and precise ratiometric method for measurement of membrane potential in bacteria. Test stimuli for work with eukaryotic cells can be provided by the potassium ionophore valinomycin, which hyperpolarizes cells in the low-potassium, high-sodium media in which experiments are typically performed, and by the relatively nonselective ionophore gramicidin, which depolarizes cells. It should be noted that whereas gramicidin can be used to depolarize cells following treatment with valinomycin, application of these ionophores in the reverse order does not reverse the depolarization first induced by gramicidin. It is recommended that control samples using each of these ionophores be run with each assay (see Anticipated Results).Valinomycin can be used to hyperpolarize bacteria, although the change from control is typically smaller than in the case of eukaryotic cells, since bacterial cytoplasmic membrane potentials are typically relatively high. Gramicidin may lyse bacterial cells; the proton ionophore carbonyl cyanide m-chlorophenylhydrazone (CCCP) is a more reliable agent for producing depolarized controls. Timing, temperature, cell concentration, and protein concentration in the medium are all critical (see Commentary); however, a considerable range of any of these variables is tolerable provided all samples in an experiment are treated consistently. The protocols require expertise in basic techniques for cell culture and handling (APPENDIX 3B) and for flow cytometry. ESTIMATION OF ∆Ψ IN EUKARYOTIC CELLS OR BACTERIA
BASIC PROTOCOL 1
Materials Cell culture Hanks’ balanced salt solution (HBSS) with calcium and magnesium but without phenol red (e.g., Life Technologies or APPENDIX 2A) Dimethyl sulfoxide (DMSO) 1 mM gramicidin D (Sigma) in DMSO (store at 4°C to 25°C; stable at least 1 month) 1 mM valinomycin (Sigma) in DMSO (store at 4°C to 25°C; stable at least 1 month) Dye working solution (see recipe): 20 µM DiIC1(3) or DiIC1(5); or 10 µM DiOC5(3) or DiOC6(3); or 20 µM DiBAC4(3) or DiSBAC2(3) 12 × 75–mm tubes Additional reagents and equipment for cell culture and handling (APPENDIX 3B) Studies of Cell Function Contributed by Howard M. Shapiro Current Protocols in Cytometry (2004) 9.6.1-9.6.12 Copyright ©2004 by John Wiley & Sons, Inc.
9.6.1 Supplement 28
1. Pellet cells and resuspend in HBSS at a concentration of ∼106 cells/ml. 2. Divide cell suspension into 1-ml aliquots in 12 × 75–mm tubes, including at least one tube to serve as an untreated control, one to serve as a depolarized control, one to serve as a hyperpolarized control, and others as required by the conditions of the experiment (e.g., different stimuli or different concentrations of stimulus). 3. Add 10 µl DMSO to the untreated control tube, 10 µl of 1 mM gramicidin D to the depolarized control tube, and 10 µl of 1 mM valinomycin to the hyperpolarized control tube. Treat the experimental tubes with stimulus as required. 4. To each aliquot, add 5 µl dye working solution. Incubate 5 min at room temperature, preferably in the dark. Final dye concentrations in cell suspension are 50 nM for DiOC5(3) and DiOC6(3); 100 nM for DiIC1(3) and DiIC1(5); and 100 nM for DiBAC4(3) and DiSBAC2(3). Excitation and Emission Wavelengths ( λ) for Cyanine and Oxonol Dyes
Table 9.6.1
Dye Cyanine dyes DiOC2(3), DiOC5(3), DiOC6(3) DiIC1(3) DiIC1(5) Oxonol dyes DiBAC4(3) DiSBAC2(3)
Excitation (maximum λ in nm; laser source)
Emission (maximum λ in nm; filters)
∼490; 488-nm argon ion or solid state
505; 520-530-nm bandpass (also 610-nm band-pass for DiOC2(3) ratiometric method) 565; 575-585-nm band-pass
∼540; 488-nm argon ion or solid state; 532-nm YAG ∼640; 633-nm He-Ne or 635-640-nm diode ∼490; 488-nm argon ion or solid state ∼535; 488-nm argon ion or solid state; 532-nm YAG
∼660; 660-680-nm band-pass or 665-nm long-pass ∼515; 520-530-nm band-pass ∼560; 575-585-nm band-pass
20,000 CCRF- CEM cells
255
untreated
valinomycin
Cell number
gramicidin
0 0
Estimation of Membrane Potential by Flow Cytometry
DiIC1 (3) fluorescence
255
Figure 9.6.1 DiCl1(3) fluorescence histograms of untreated (DMSO control) CCRF-CEM cells, cells hyperpolarized with valinomycin, and cells depolarized with gramicidin.
9.6.2 Supplement 28
Current Protocols in Cytometry
5. Analyze cells in the flow cytometer, using excitation and emission wavelengths as indicated in Table 9.6.1, and triggering on the forward scatter signal. 6. Analyze data comparing the mean, median, or mode of fluorescence histograms from experimental samples with the equivalent data from untreated, depolarized, and/or hyperpolarized controls. Because the staining method is an equilibrium method, and because membrane potential may change rapidly, it is essential that incubation times of cell samples with dye and stimuli, and the delay between sample preparation and analysis, be kept constant. In an example illustrated here, CCRF-CEM cell samples were prepared at intervals such that each aliquot was incubated with DMSO, valinomycin, or gramicidin for 5 min prior to addition of DiIC1(3), and introduced into the flow cytometer after an additional 5 min incubation with dye. The DiIC1(3) fluorescence histograms obtained are shown in Figure 9.6.1. Similar histograms are shown in Figure 9.6.2 for DMSO- and valinomycin-treated cells using bis(dibutylthiobarbituric acid) trimethine oxonol [DiTBAC4(3)], a close analog of DiSBAC2(3) with the same optical properties. Compare the direction of the change in ∆Ψ using a cationic versus an anionic probe. Alternatively, when measurements of time as a parameter are possible, dye fluorescence may be plotted against time following the addition of stimuli. When this is done, it is desirable to be able to identify the position of the population mean or median at each time point. An example of this type of analysis, illustrating the response of polymorphonuclear leukocytes to stimulation with phorbol myristyl acetate (PMA) using DiOC5(3) as the membrane potential probe, appears in Figure 9.6.3.
MEASUREMENT OF ∆Ψ IN BACTERIA USING DiOC2(3) This method is more accurate and precise than that described in Basic Protocol 1, but is not applicable to eurokaryotic cells. See Background Information for details.
BASIC PROTOCOL 2
Materials Bacterial culture (typically 1 × 108 to 1 × 109 cells/ml) Mueller-Hinton broth (Life Technologies) with 50 mg/liter Ca2+ (MHBc) Dimethyl sulfoxide (DMSO)
20,000 CCRF- CEM cells
511
untreated
Cell number
valinomycin
0 0
DiTBAC4 (3) fluorescence
255
Figure 9.6.2 DiTBAC4(3) fluorescence histograms of untreated (DMSO control) CCRF-CEM cells and cells hyperpolarized with valinomycin. DiTBAC4(3) is a close analog of DiSBAC2(3) that has the same optical properties but is not presently commercially available.
Studies of Cell Function
9.6.3 Current Protocols in Cytometry
Supplement 28
PMA added
Figure 9.6.3 Time course of depolarization of polymorphonuclear leukocytes following treatment with phorbol myristyl acetate (PMA); ∆Ψ is estimated from DiOC5(3) fluorescence. Figure kindly provided by Dr. J. Paul Robinson, Purdue University, and reprinted with permission of ASM Press from Robinson et al. (1997).
Green fluorescence
1024
512
0 0
1200
2400
Time (sec)
2 mM valinomycin (Sigma) in DMSO (store at 4°C to 25°C; stable at least 1 month) 2 mM CCCP (Sigma) in DMSO (store at 4°C; stable for at least 1 month) Dye working solution: 3 mM DiOC2(3) in DMSO (store at 4°C; stable for at least 1 month) 12 × 75–mm tubes 1. Divide bacterial suspension into 1-ml aliquots in 12 × 75–mm tubes, including at least one tube to serve as a control and others as required by the conditions of the experiment (e.g., different stimuli or different concentrations of stimulus). For work with bacteria, “stimuli” are frequently antibiotics, in which case aliquots are incubated for relatively long time periods (minutes to hours), preferably with shaking to insure adequate oxygenation.
2. Dilute bacteria in 1 ml MHBc to a target concentration of 1 × 106 to 1 × 107 cells/ml. 3. Prepare controls. a. For the untreated control, add 7.5 µl DMSO and keep sample 5 min at room temperature before adding dye. b. For the hyperpolarized control, add 2.5 µl of 2 mM valinomycin (final concentration 5 µM) and keep sample 5 min at room temperature before adding dye. c. For the depolarized control, add 7.5 µl of 2 mM CCCP (final concentration 15 µM) and keep sample 5 min at room temperature before adding dye. 4. Add 10 µl dye working solution [3 mM DiOC2(3); final concentration 30 µM] and keep sample 5 min at room temperature. Estimation of Membrane Potential by Flow Cytometry
9.6.4 Supplement 28
Current Protocols in Cytometry
A Red fluorescence
104
103
untreated
102 + CCCP 101
100 100
101
102
103
104
Green fluorescence
B 650 600 untreated
+ CCCP
550 500
Cell count
450 400 350 300 250 200 150 100 50 0 100
101
102
103
104
DiOC2(3) red/green fluorescence
Figure 9.6.4 Measurement of membrane potential in Staphylococcus aureus using a ratiometric method. (A) Dot plot of green versus red fluorescence of DiOC2(3) in a control culture (black dots) and a culture depolarized with CCCP (gray dots). (B) Histograms (logarithmic scale) of red/green fluorescence in a control culture (black) and a culture depolarized with CCCP (gray); the cells represented are the same cells shown in (A).
5. Analyze cells in the flow cytometer using 488-nm excitation from an argon-ion or solid-state laser, triggering on the forward- or side-scatter signal, and measuring DiOC2(3) green fluorescence through a 525- to 530-nm band-pass filter with ∼20-nm bandwidth and red fluorescence through a 610-nm band-pass filter with ∼20-nm bandwidth. Membrane potential is proportional to the ratio of [DiOC2(3) red fluorescence] to [DiOC2(3) green fluorescence]. If high-resolution linear data are available from the flow cytometer, this ratio may be calculated directly by division and multiplied by an appropriate scaling constant to place calculated values on the display scale used. In instruments on which data are collected using logarithmic amplifiers, a quantity proportional to the red/green fluorescence ratio is calculated by adding a constant to the red fluorescence channel value and subtracting the green fluorescence channel value (Novo et al., 1999).
Studies of Cell Function
9.6.5 Current Protocols in Cytometry
Supplement 28
The addition of a constant value is necessary to keep values of the calculated parameter on the same scale as is used for the raw fluorescence measurements. For a 256-channel logarithmic scale, with 64 channels per decade, a constant value of 96 is convenient; the calculated parameter, which serves as a measure of ∆Ψ, then represents the log of (103/2 × [red fluorescence/green fluorescence]). Dot plots of green versus red fluorescence of DiOC2(3) in a control culture and a depolarized culture are shown in Figure 9.6.4A; histograms (logarithmic scale) of the red/green fluorescence ratio in the same cells are shown in Figure 9.6.4B. Measurements of ∆Ψ using DiOC2(3) may be calibrated by controlled application of valinomycin in the presence of different external potassium ion concentrations (Novo et al., 1999). The red/green fluorescence ratio is measured for cells in a range of buffers containing 5 ìM valinomycin and various concentrations of potassium; the concentration of sodium ion is adjusted to keep the combined molarity of potassium and sodium at 300 mM.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Dye stock solutions Prepare stock solutions of the following dyes (as needed) at the given concentration in dimethyl sulfoxide (DMSO): Cyanine dyes: Hexamethylindocarbocyanine iodide [DiIC1(3)], 1 mM Hexamethylindodicarbocyanine iodide [DiIC1(5)] 1 mM Diethyloxacarbocyanine [DiOC2(3)], 3 mM Dipentyloxacarbocyanine [DiOC5(3)], 1 mM Dihexyloxacarbocyanine [DiOC6(3)], 1 mM Oxonol dyes: Bis(1,3-dibutylbarbituric acid) trimethine oxonol [DiBAC4(3)], 1 mM Bis(1,3-diethylthiobarbituric acid) trimethine oxonol [DiSBAC2(3)], 1 mM Store all solutions in the dark at 4° to 25°C (stable at least 1 month). Dyes listed above are all available from Molecular Probes. DMSO is available from Sigma.
Dye working solutions Before use, dilute dye stock solutions (see recipe) in dimethyl sulfoxide (DMSO) as follows: 10 µl DiIC1(3) or DiIC1(5) in 490 µl DMSO (20 µM final) 10 µl DiOC5(3) or DiOC6(3) in 990 µl DMSO (10 µM final) 10 µl DiBAC4(3) or DiSBAC2(3), in 490 µl DMSO (20 µM final) Keep working solutions ≤1 week in the dark at 4° to 25°C 3 mM DiOC2(3) stock solution is used as working solution. If DMSO level in cells is a concern, working solutions can be made in ethanol (see Critical Parameters). It is best to make working solutions fresh.
COMMENTARY Background Information
Estimation of Membrane Potential by Flow Cytometry
Origin and determination of ∆Ψ In most living cells, differences between interior and exterior concentrations of ions such as sodium, potassium, and chloride generate an electrical potential across the cytoplasmic membrane. Membrane potential (∆Ψ), like cy-
toplasmic calcium ion concentration and distribution, and intracellular pH, may change early in the course of surface receptor–mediated cell activation processes related to the development, differentiated function, and pathology of a large number of cell types, and these changes in ionic environment may play a role in trans-
9.6.6 Supplement 28
Current Protocols in Cytometry
membrane signaling in response to cell surface ligand-receptor interactions. It is therefore frequently of scientific interest to estimate ∆Ψ in individual cells. This can be done by flow or static cytometry using any of a number of fluorescent probes. The current state of the art allows fairly precise determination of ∆Ψ in terms of population averages. However, for the identification of significant heterogeneity within populations, and for the detection of changes over time, the accuracy and precision of ∆Ψ measurements on a cellby-cell basis are considerably lower than those of, for example, DNA content measurements. The methods presented here generally yield reasonably accurate estimates of ∆Ψ. Determination of membrane potential is done using fluorescently labeled cationic or anionic lipophilic probes. Known as distributional dyes, these can pass easily across the membrane until a ∆Ψ-dependent equilibrium is reached. The distribution of intracellular and extracellular dye is measured by flow cytometry, and is then used to estimate ∆Ψ. Once cells have been equilibrated with an indicator cation, electrical depolarization of the cells, i.e., a reduction in ∆Ψ, will cause release of indicator from cells into the medium, whereas hyperpolarization, i.e., an increase in ∆Ψ, will make cells take up additional indicator from the medium. The reverse is seen with an indicator anion. The indicator distribution will not adequately represent the new value of ∆Ψ until equilibrium has again been reached, which requires periods ranging from a few seconds to several minutes. Thus, although distributional probes may be suitable for detection of slow changes in ∆Ψ, they cannot be used to monitor the faster changes that occur during the propagation of action potentials in tissues such as nerve and muscle. Other dyes, which undergo rapid conformational and spectral changes in response to changes in the electrical field within the membrane, can be used to detect action potentials in single cells, but these probes are generally unusable for flow cytometry because of the relatively small changes in emission intensity involved. The resting potential across the cytoplasmic membrane of mammalian cells is usually approximated by the Goldman equation:
∆Ψ =
P [K + ]i + PNa [Na + ]i + PCl [Cl− ]o RT ln K F PK [K + ]o + PNa [Na + ]o + PCl [Cl − ]i
where ∆Ψ is the membrane potential, R is the gas constant (8.314510 J/mol K), T is the temperature in Kelvin (K), F is the Faraday constant (9.6485309 × 104 C/mol), [X]i is the concentration of X ions inside the cell, [X]o is the concentration of X ions outside the cell, and PX is the permeability of the membrane to X ions. For a univalent lipophilic cationic indicator species C+, the ratio of concentrations inside and outside the cell is given by the Nernst equation: [C+]i/[C+]o = e−F∆Ψ/RT, where the notation is the same as above. For example, a potential difference ∆Ψ of 61 mV at 37°C would result in a [C+]i/[C+]o ratio equal to 10, and a potential difference of 122 mV in a ratio of 100, based on the Nernst equation. This distribution is closely approximated for relatively water-soluble dyes such as tetramethylrhodamine methyl and ethyl esters (Ehrenberg et al., 1988); however, more hydrophobic dyes, such as the cyanines, tend to yield higher ratios of [C+]i/[C+]o than would be predicted by the Nernst equation, with the ratio increasing with decreasing water solubility of a homologous series of dyes (Sims et al., 1974). Cyanine versus oxonol dyes Estimation of ∆Ψ has been successfully demonstrated using cationic cyanine dyes (Shapiro et al., 1979) and anionic oxonol dyes (Shapiro, 1982, 2000, 2003). Cyanine dyes were originally developed as sensitizers for photographic film, which accounts for the availability of members of the series with absorption maxima ranging from the ultraviolet to the infrared. The structure of a typical symmetric cyanine dye is shown in Figure 9.6.5. The shorthand notation originally developed by Sims et al. (1974) gives the formula for this dye as DiYCn + 1(2m + 1). Dipentyl- and dihexyloxacarbocyanine, DiOC5(3) and DiOC6(3), are among the membrane potential probes most widely used in cytometry because they can be excited by the 488-nm laser source most commonly used in fluorescence flow cytometers, and are readily detected through the 520- to 530-nm band-pass filters commonly used for fluorescein. Hexamethylindocarbocyanine, DiIC1(3), can also be excited by the 488-nm laser, and is detected with the 575- to 585-nm band-pass filters used for phycoerythrin. Hexamethylindodicarbocyanine, DiIC1(5), is excited at higher wavelengths (633-635 nm), and is detected by the same filters used to detect allophycocyanin or the covalent label Cy5, which also contains the indodicarbocyanine
Studies of Cell Function
9.6.7 Current Protocols in Cytometry
Supplement 28
A Y
Y CH2 (CHCH)m N+
N+ CH3 (CH2 )n
(CH2 )n CH3
B CH3 (CH2 )3
H3 C
N
N CH
O
CH
CH
O
N (CH2 )3
(CH2 )3
O
OH
N O
CH3
(CH2 )3
O H3 C
Figure 9.6.5 Structure of cyanine and oxonol dyes. (A)The symmetric cyanine dyes are formulated as DiYCn+1(2m + 1) in the nomenclature of Sims et al. (1974). When Y is O or S, the substituent is oxygen or sulfur. When Y is I, the substituent is C(CH3)2. (B)Structure of the oxonol dye bis(1,3-dibutylbarbituric acid) trimethine oxonol, DiBAC4(3).
Estimation of Membrane Potential by Flow Cytometry
chromophore. Although DiOC5(3) and DiOC6(3) are commonly used when ∆Ψ is the only parameter being measured by fluorescence, DiIC1(3) and DiIC1(5) yield essentially identical results, and can more easily be incorporated into multicolor staining protocols, e.g., in combination with fluorescein-labeled antibodies. For details on specific wavelengths for all of the dyes in this protocol, see Table 9.6.1. The basic principle of cytometric membrane potential estimation using cyanine dyes requires that dye concentrations be kept low enough so that the fluorescence measured from a cell varies directly and monotonically with the amount of cell-associated dye. For DiOC5(3) and DiOC6(3), a working concentration ≤50 nM is desirable; for DiIC1(3) and DiIC1(5), 100 nM is suitable. Cyanine dyes present several problems as indicators of ∆Ψ. The cationic cyanine dyes penetrate cells and distribute into mitochondria as well as the cytosol; because there is a potential gradient across the mitochondrial membrane, typically of ≥100 mV, the dye concentration in cells is affected by mitochondrial as well as cytoplasmic membrane potential. Cyanine dyes may be toxic to cells. They are also pumped out by the action of glycoprotein efflux pumps associated with the multiply drug-resistant phenotype (Kessel et al., 1991). Because oxonol dyes are negatively charged, the intracellular concentrations of these dyes remain lower than the extracellular concentrations; as
a result, the toxicity of oxonols is lower than that of cyanines on a molar basis. Oxonol dyes do not accumulate in organelles, which makes interpreting results more straightforward. Oxonols are also essentially unaffected by efflux pumps, and are preferred by some workers on that account. However, oxonol dyes typically require higher fluorescence gain than cyanines, and whereas cyanines do not strongly stain dead cells with damaged membranes, oxonols do bind to the internal constituents of such cells, making the elimination of dead cells by gating somewhat more complex than with cyanines. The more commonly used oxonols include bis(1,3-dibutylbarbituric acid) trimethine oxonol and bis(1,3-diethylthiobarbituric acid) trimethine oxonol. The shorthand descriptions—DiBAC4(3) and DiSBAC2(3) or DiTBAC2(3)—uniquely characterize each dye and may therefore be used routinely without danger of confusion. DiBAC4(3) has frequently been described in the literature as “bis-oxonol”; this nomenclature should be avoided because it could apply equally to either of these oxonols and to other analogous compounds. The structure of DiBAC4(3) is shown in Figure 9.6.5. DiBAC4(3) can be excited/measured using the same filters as for fluorescein. DiSBAC2(3) is detected using a phycoerythrin filter. Why calibration of ∆Ψ measurements in eukaryotic cells is impractical Addition of valinomycin increases potassium permeability (PK in the Goldman equa-
9.6.8 Supplement 28
Current Protocols in Cytometry
tion) to the point at which ∆Ψ is determined almost entirely by the transmembrane [K+] gradient. If [K+]o is low, as is the case in plasma or extracellular fluid and in most physiologic buffer solutions and media, valinomycin addition hyperpolarizes cells. However, if [K+]o is high, valinomycin depolarizes cells, and if [K+]o = [K+]i, valinomycin leaves ∆Ψ unchanged. Membrane potential estimation using cyanine dyes was originally performed on cell suspensions in cuvettes (Sims et al., 1974), using high enough dye concentrations so that most of the fluorescence of intracellular dye was quenched. Under these circumstances, it was customary to construct a calibration curve using fluorescence measurements of valinomycin-treated cells at known values of [K+]o. The broad fluorescence distributions of cyanine and oxonol dyes in samples of ionophore-treated cells, in which all cells should be at the same ∆Ψ, suggest that factors other than membrane potential exert a substantial influence on dye fluorescence. As it is theoretically the concentration, not the total amount, of intracellular dye that provides an estimate of ∆Ψ, dividing the measured fluorescence value by the cell volume should yield a more accurate ∆Ψ value. However, although division by a parameter value representative of cell size (e.g., extinction) produces a parameter with a somewhat narrower distribution than the fluorescence distribution, the variance still remains quite large, and it is essentially impossible to calibrate ∆Ψ values to the nearest few millivolts. More to the point, although it is relatively easy to detect shifts in ∆Ψ affecting all or a large fraction of cells in a population, it is also essentially impossible to detect small subpopulations even when they undergo substantial changes in ∆Ψ in response to a stimulus. Ratiometric methods may increase measurement precision The precision of membrane potential measurements may be significantly increased by the use of ratiometric dyes such as those used for calcium (UNIT 9.8) and pH (UNIT 9.3) measure ments. One such probe, 1-(3-sulfonatopropyl)4-[(β-[ 2-(di-n-butylamino)-6-naphthyl]vinyl] pyridinium betaine (di-4-ANEPPS; Montana et al., 1989), is electrochromic—i.e., its spectral characteristics change with the electric field within the membrane. However, the ratio of red fluorescence emission intensities of di-4ANEPPS with green and blue-violet excitation, which provides the estimate of ∆Ψ, changes by only ∼10% for a 90-mV change in ∆Ψ. Al-
though the probe has been used successfully in image cytometry, it is unlikely that flow cytometry would be able to resolve changes of 10 to 20 mV. Gonzalez and Tsien (1997) introduced another ratiometric method that specifically senses the fast potential response of oxonol molecules in the membrane lipid bilayer using fluorescence resonance energy transfer (FRET; UNIT 1.12) between the oxonol dye and a fluorescently labeled lectin or phospholipid, allowing fairly precise measurement of cytoplasmic ∆Ψ. Changes in fluorescence ratio exceed 50 per cent for a 100-mV change in ∆Ψ, and responses can be detected in fractions of a millisecond in static cytometers; the technique is also applicable to bulk fluorometry and flow cytometry. However, there is some question as to its suitability for use in bacteria, because the number of oxonol molecules bound within a single bacterial membrane is quite small. Estimation of cytoplasmic membrane potential in bacteria: A ratiometric method is preferable Membrane potential is generated and maintained by the energy metabolism in both aerobic and anaerobic microorganisms. As in eukaryotic cells, a breach of the bacterial cytoplasmic membrane will reduce ∆Ψ to zero; thus, cells that would be classified as dead on the basis of a loss of membrane integrity can also be distinguished from live cells with active metabolism on the basis of ∆Ψ values. Additionally, because bacteria have no organelles, measurements of ∆Ψ cannot be confused by gradients across other membranes. A number of workers have used flow cytometric estimation of ∆Ψ with rhodamine 123, cyanine, or oxonol dyes to detect and count viable bacteria and to rapidly determine bacterial antibiotic susceptibilities (Shapiro, 1988; Kaprelyants and Kell, 1992; Ordóñez and Wehman, 1993; Mason et al., 1994; Jepras et al., 1995; López-Amorós et al., 1995). The flow cytometric procedures first used with bacteria were essentially the same as those for estimation of ∆Ψ in eukaryotic cells, with two significant differences. In their native state, the outer membranes of gram-negative bacteria are impermeable to lipophilic materials, including the dyes and ionophores used in this protocol. When gram-negative organisms are analyzed, the outer membrane must be permeabilized by the addition of 1 to 5 mM EDTA or EGTA to the medium. Also, as ∆Ψ in both aerobic and anaerobic bacteria is determined primarily by
Studies of Cell Function
9.6.9 Current Protocols in Cytometry
Supplement 28
the proton gradient across the membrane, it is common practice to use a proton ionophore such as carbonyl cyanide chlorophenyl hydrazone (CCCP) instead of gramicidin as a standard depolarizing stimulus. In bacteria, as in eukaryotic cells, histograms of cyanine or oxonol dye fluorescence in untreated control cultures and depolarized cultures typically overlap, and it is impossible to assign values of ∆Ψ based on simple fluorescence measurements. The ratiometric method described in this unit can be calibrated, and provides much better discrimination between energized and depolarized bacteria (Fig. 9.6.4) than can otherwise be obtained. The dye used, DiOC2(3), shifts fluorescence emission from green to red at concentrations much higher than were previously used for estimation of ∆Ψ . In bacteria equilibrated with 30 µ M dye, green fluorescence is dependent on the size of the organism (or aggregate of organisms), whereas red fluorescence is both size and potential dependent. As can be seen from Figure 9.6.4A, red and green fluorescence measurements are highly correlated in both control and depolarized cultures, and the distributions of the red/green ratio from depolarized and untreated control cells show virtually no overlap (the control culture contains a small number of depolarized, presumably nonviable, cells). Accurate and precise ∆Ψ measurement in bacteria may improve understanding of mechanisms of action of antimicrobial agents (Mason et al., 1995; Novo et al., 2000; Shapiro, 2003; Silverman et al., 2003), aiding in the development of new drugs to combat the increasing number of microbial pathogens that have developed resistance to existing antimicrobials.
Estimation of Membrane Potential by Flow Cytometry
Mitochondrial membrane potential estimation with rhodamine 123, JC-1, and other dyes In general, the spectra of cyanine and oxonol dyes change with concentration. Although the dyes typically form nonfluorescent dimers and polymers, many also exhibit fluorescence emission shifts. In the membrane potential estimation methods described above, dye remains in equilibrium with cells during the measurement. As has already been mentioned, cationic dyes such as the cyanines penetrate cells, and as there is typically a potential gradient of ≥100 mV between the cytosol and the mitochondrial interior, the dyes distribute into mitochondria at even higher concentrations than are present in the cytosol. If cells with physiologically intact
mitochondria are equilibrated with relatively high concentrations of cationic dye, and then washed, dye will be retained in the mitochondria; if mitochondrial function is compromised, eliminating the potential gradient across the mitochondrial membrane, dye will not be retained, and mitochondrial fluorescence or the lack thereof can therefore provide an indicator of mitochondrial integrity. The dye first widely used for studies of mitochondria is rhodamine 123 (Chen, 1989). This dye has an excitation maximum of ∼488 nm and an emission maximum at ∼530 nm; it is substantially less hydrophobic than the cyanines. Cells are usually loaded with rhodamine 123 at dye concentrations of ∼10 µM, and washed prior to microscopic or cytometric observation. Substitution of cyanine dyes for rhodamine 123 yields similar results, although there is generally somewhat more cytoplasmic retention of cyanines due to their hydrophobicity. In principle, estimates of ∆Ψ should be obtainable from fluorescence emission ratios of dyes which undergo spectral shifts with changes in concentration. The cyanine dye 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolylcarbocyanine iodide (JC-1) forms fluorescent aggregates that, when excited at 490 nm, emit maximally at 590 nm, whereas the monomeric dye emits maximally at 527 nm. The fluorescence emission ratio was proposed as a quantitative probe of mitochondrial membrane potential by Smiley et al. (1991). In practice, JC-1 aggregates in solution, takes much longer to equilibrate to stable fluorescence values than do dyes such as DiOC5(3) and DiOC6(3), and produces broad distributions of fluorescence ratio. However, although JC-1 cannot reliably be used for quantitative ratiometric measurements, the shift from predominantly red fluorescence in cells with energized mitochondria to predominantly green fluorescence in cells that have lost mitochondrial membrane potential makes it useful for assessment of mitochondrial function (UNIT 9.14). Cyanines, rhodamine 123, and other cationic dyes are actively extruded from cells by the action of glycoprotein efflux pumps (Kessel et al., 1991), and changes in cellular fluorescence caused by changes in efflux pump activity can be (and have been) misinterpreted as membrane potential changes. This is less likely to happen when cytoplasmic membrane potential is being estimated than when mitochondria are being studied because in the former case, a substantial amount of extracellular dye remains
9.6.10 Supplement 28
Current Protocols in Cytometry
in equilibrium with the cells. Mitochondrial measurements using rhodamine 123 and JC-1 are now in reasonably wide use as a criterion of apoptosis. It has been reported that, when used at very low concentrations (~1 nM), DiOC6(3) can provide a quantitative measurement of mitochondrial membrane potential (Rottenberg and Wu, 1998).
Critical Parameters and Troubleshooting Solvents Cyanine and oxonol dyes, gramicidin, valinomycin, and other ionophores are hydrophobic, and will tend to plate out of aqueous solutions onto the surfaces of glass and plastic test tubes and onto the tubing in flow cytometers. Most of these compounds are highly soluble in DMSO, which, because it does not readily evaporate, is a convenient vehicle for preparation of stock and working solutions. In the protocols given here, the amount of DMSO added to cell suspensions is sufficiently small to keep the overall concentration at ≤2% (v/v). If there is some concern about perturbation of cellular properties by this concentration of DMSO, working solutions may be made up in ethanol. It should also be emphasized that because DMSO is denser than water, dyes and ionophores added to cell suspensions will tend to sink to the bottom. Thus, vortexing or brisk agitation is recommended to insure proper mixing. Protein concentrations in cell suspensions Equilibration of dyes, ionophores, and low concentrations of test stimuli tends to occur most rapidly (≤5 min) in cell suspensions without added protein. However, when the objective of an experiment is to determine the effect on ∆Ψ of a preparation containing a substantial amount of protein, it is desirable to add a protein such as bovine serum albumin to all samples, in order to maintain protein concentrations nearly equal in treated and control samples. If this is not done, binding of dye to protein in the suspension and resulting loss from cells may result in an apparent depolarization. Tubing As stated above, the dyes and ionophores used in this protocol stick to cytometer tubing. It is therefore important to preequilibrate the tubing before use to prevent baseline shift. Likewise, tubing should be thoroughly cleaned after use by flushing the instrument with 25% chlorine bleach (1 to 2 ml should
suffice). If this is not done, residual dye may migrate from tubing into cell samples subsequently introduced into the instrument, which may produce the appearance of immunofluorescent staining in unlabeled cells. In some cases—e.g., when a gramicidin-treated sample is run as a depolarized control—ionophore in the sample may stick to the cytometer tubing and subsequently bind to cells in a later sample, producing a real change in ∆Ψ whose source is likely to be misinterpreted. When there is a possibility that this may happen, it may be prudent to flush with bleach between samples. Because bleach itself can readily kill cells, causing irreversible loss of ∆Ψ, sufficient water or saline to remove the bleach should be run through the system before the next sample is analyzed.
Anticipated Results This procedure provides an estimation of membrane potential (∆Ψ) as a function of the distribution of fluorescent lipophilic dye between cells and surrounding medium. In general, it is useful for demonstrating differences in ∆Ψ within or between populations rather than for assigning specific values of ∆Ψ (see Background Information section on calibration). The use of depolarizing (gramicidin) and hyperpolarizing (valinomycin) controls in addition to untreated controls facilitates interpretation of data. The cyanine dye fluorescence distributions measured from cells treated with depolarizing stimuli are shifted toward lower fluorescence intensity values, as compared to untreated controls, whereas fluorescence distributions from cells treated with hyperpolarizing stimuli are shifted toward higher values. Oxonol fluorescence distributions shift in the reverse direction. Because it is unlikely, on physiologic and pharmacologic grounds, that any stimulus will produce greater depolarization than gramicidin (CCCP in bacteria) or greater hyperpolarization than valinomycin, observation of a fluorescence distribution outside the range defined by the depolarized and hyperpolarized controls should raise the suspicion that something has gone wrong.
Time Considerations Membrane potential estimation by flow cytometry is meaningful only when the cells under study remain viable and the equilibrium between cells and dye is unperturbed by factors other than the stimuli applied, at least until the cells have passed the observation point of the
Studies of Cell Function
9.6.11 Current Protocols in Cytometry
Supplement 28
flow cytometer. Timing is particularly critical when the membrane potential change to be observed lasts for a few minutes or less, in which case kinetic measurements incorporating time as a measurement parameter represent the only realistic approach to obtaining consistent data. Even when stimuli produce longerlasting effects, it is important to maintain a relatively constant duration of incubation of cells with dyes and stimuli from sample to sample. Analysis of a single sample of eukaryotic cells or bacteria typically requires ∼10 min from addition of dyes to completion of data collection.
Literature Cited Chen, L.B. 1989. Fluorescent labeling of mitochondria. Methods Cell Biol. 29:103-123. Ehrenberg, B., Montana, V., Wei, M.D., Wuskell, J.P., and Loew, L.M. 1988. Membrane potential can be determined in individual cells from the Nernstian distribution of cationic dyes. Biophys. J. 53:785-794.
Novo, D., Perlmutter, N.G., Hunt, R.H., and Shapiro, H.M. 2000. Multiparameter flow cytometric analysis of antibiotic effects on membrane potential, membrane permeability, and bacterial counts of Staphylococcus aureus and Micrococcus luteus. Antimicrob. Agents Chemother. 44:827-834. Ordóñez, J.V. and Wehman, N.M. 1993. Rapid flow cytometric antibiotic susceptibility assay for Staphylococcus aureus. Cytometry 14:811-818. Robinson, J.P., Carter, W.O., and Narayanan, P.K. 1997. Functional assays by flow cytometry. In ASM Manual of Clinical Laboratory Immunology, 5th ed. (N.R. Rose, E.C. de Macario, J.P. Folds, H.C. Lane, and R.M. Nakamura, eds.) pp. 245-254. ASM Press, Washington, DC. Rottenberg, H. and Wu, S. 1998. Quantitative assay by flow cytometry of the mitochondrial membrane potential in intact cells. Biochim. Biophys. Acta 1404:393-404.
Gonzalez, J.E. and Tsien, R.Y. 1997. Improved indicators of cell membrane potential that use fluorescence resonance energy transfer. Chem. Biol. 4:269-277.
Shapiro, H.M. 1982. Cytological Assay Procedure. U.S. Patent No. 4,343,982, issued Aug. 10, 1982.
Jepras, R.I., Carter, J., Pearson, S.C., Paul, F.E., and Wilkinson, M.J. 1995. Development of a robust flow cytometric assay for determining numbers of viable bacteria. Appl. Environ. Microbiol. 61:2696-2701.
Shapiro, H.M. 2000. Membrane potential estimation by flow cytometry. Methods 21:271-276.
Kaprelyants, A.S. and Kell, D.B. 1992. Rapid assessment of bacterial viability and vitality by rhodamine 123 and flow cytometry. J. Appl. Bacteriol. 72:410-422. Kessel, D., Beck, W.T., Kukuruga, D., and Schulz, V. 1991. Characterization of multidrug resistance by fluorescent dyes. Cancer Res. 51:46654670. López-Amorós, R., Comas, J., and Vives-Rego, J. 1995. Flow cytometric assessment of Escherichia coli and Salmonella typhimurium starvation-survival in seawater using rhodamine 123, propidium iodide, and oxonol. Appl. Environ. Microbiol. 61:2521-2526. Mason, D.J., Allman, R., Stark, J.M., and Lloyd, D. 1994. Rapid estimation of bacterial antibiotic susceptibility with flow cytometry. J. Microsc. 176:8-16. Mason, D.J., Power, E.G.M., Talsania, H., Phillips, I., and Gant, V.A. 1995. Antibacterial action of ciprofloxacin. Antimicrob. Agents Chemother. 39:2752-2758.
Estimation of Membrane Potential by Flow Cytometry
Novo, D., Perlmutter, N.G., Hunt, R.H., and Shapiro, H.M. 1999. Accurate flow cytometric membrane potential measurement in bacteria using diethyloxacarbocyanine and a ratiometric technique. Cytometry 35:55-63.
Montana, V., Farkas, D.L., and Loew, L.M. 1989. Dual-wavelength ratiometric fluorescence measurements of membrane potential. Biochemistry 28:4536-4539.
Shapiro, H.M. 1988. Practical Flow Cytometry, 2nd ed., pp. 296-298. Alan R. Liss, New York.
Shapiro, H.M. 2003. Practical Flow Cytometry, 4th ed., pp. 385-402, 519-522. John Wiley & Sons, Hoboken, N.J. Shapiro, H.M., Natale, P.J., and Kamentsky, L.A. 1979. Estimation of membrane potentials of individual lymphocytes by flow cytometry. Proc. Natl. Acad. Sci. U.S.A. 76:5728-5730. Silverman, J.A., Perlmutter, N.G., and Shapiro, H.M. 2003. Correlation of daptomycin bactericidal activity and membrane depolarization in Staphylococcus aureus. Antimicrob. Agents Chemother. 47:2538-2544. Sims, P.J., Waggoner, A.S., Wang, C.H., and Hoffman, J.F. 1974. Studies on the mechanism by which cyanine dyes measure membrane potential in red blood cells and phosphatidylcholine vesicles. Biochemistry 13:3315-3330. Smiley, S.T., Reers, M., Mottola-Hartshorn, C., Lin, M., Chen, A., Smith, T.W., Steele, G.D. Jr., and Chen, L.B. 1991. Intracellular heterogeneity in mitochondrial membrane potentials revealed by a J-aggregate-forming lipophilic cation JC-1. Proc. Natl. Acad. Sci. U.S.A. 88:3671-3675.
Contributed by Howard M. Shapiro Howard M. Shapiro, M.D., P.C. West Newton, Massachusetts
9.6.12 Supplement 28
Current Protocols in Cytometry
Oxidative Metabolism
UNIT 9.7
This unit will demonstrate how measurement of oxidative products can be made in single cells using flow cytometry. Such measurements may be performed to determine the degree of activation of cells, to determine if cells are capable of responding to stimulation, or to evaluate the ability of a test substance to stimulate readily activatable cell populations. The key difference between the flow cytometry methods presented and nonflow methods is the ability to measure the oxidative products of single cells, and further to measure these products within the cell itself, not in the extracellular milieu. The general approach is to use either purified cell populations or cell mixtures (e.g., white blood cells), load them with a suitable probe, stimulate them, and monitor the resultant change in fluorescence. One method (see Basic Protocol) involves measuring H2O2 production in the oxidative burst of the human neutrophil, using the well-known probe 2′,7′-dichlorofluorescin diacetate (also known as 2′,7′-dichlorodihydrofluorescin; DCFH-DA) as the indicator. A second method (see Alternate Protocol 1) uses hydroethidine (HE) to measure superoxide anions (O2−) in the oxidative burst; these two methods may also be combined (see Alternate Protocol 2) for simultaneous measurements. Yet another method (see Alternate Protocol 3) is to use dihydrorhodamine 123 (DHR) for oxidative burst H2O2 measurements. Each probe is capable of extracting different information from the cells, such as the nature of the reactive species, the kinetics of activation and response, or the site of product formation. When accurate quantitation of H2O2 production is needed, actual molar equivalents can be correlated to fluorescence channels (see Support Protocol 1). It is unnecessary to perform these careful calibrations if simple qualitative changes are being sought. When performing these cell function assays, it is very important that cell populations not be too old. Restrict studies to within 6 hr of the time blood samples are taken. MEASUREMENT OF H2O2 PRODUCTION IN THE OXIDATIVE BURST OF HUMAN NEUTROPHILS USING DCFH-DA
BASIC PROTOCOL
This protocol describes the use of 2′,7′-dichlorofluorescin diacetate (DCFH-DA) to measure the production of H2O2, which oxidizes the nonfluorescent probe to a fluorescent one that is detected by the flow cytometer. DCFH-DA added to the cell suspension is taken up by the cells and is trapped there upon hydrolysis by cellular esterases. This process takes ∼5 to 10 min for human neutrophils. The protocol allows a full 15 min at 37°C to ensure complete hydrolysis. The cells are then stimulated by a known agent such as phorbol myristate acetate (PMA) or by some test substance. A control population must be left unstimulated as nonspecific oxidation always takes place and must be monitored via a nonstimulated control. After a suitable period of stimulation, generally 30 min, a final measurement establishes the maximum stimulation of the cells. An accurate kinetic rate for activation of the respiratory burst can be determined by monitoring a single sample continuously over the same 30-min time period. Materials Vacutainer tubes treated with preservative-free heparin (see Support Protocol 2) Ammonium chloride lysing solution (working solution; APPENDIX 2A) Hanks balanced salt solution (HBSS; APPENDIX 2A) 20 mM 2′,7′-dichlorofluorescin diacetate (DCFH-DA, from Molecular Probes; prepare fresh in 100% ethanol and keep on ice, protected from light) 10 µg/ml phorbol myristate acetate (PMA) working solution I (see recipe) Studies of Cell Function Contributed by J. Paul Robinson Current Protocols in Cytometry (1997) 9.7.1-9.7.14 Copyright © 1997 by John Wiley & Sons, Inc.
9.7.1 Supplement 2
50-ml centrifuge tubes, sterile 12 × 75–mm polyethylene tube and caps Prepare a white blood cell suspension 1. Obtain 10 ml fresh blood in a Vacutainer tube treated with preservative-free heparin. Place 3 ml blood in a sterile 50-ml centrifuge tube and add 47 ml lysing solution. The heparin used as anticoagulant should be preservative free, because preservative materials that are normally present can interfere with functional assays. Support Protocol 2 explains how to prepare a Vacutainer tube with preservative-free heparin.
2. Place the tubes on a bench rotor for 10 min at 30 rpm, room temperature, then centrifuge them 10 min at 250 × g, 4°C. 3. Carefully remove tubes, decant supernatant gently, and place tubes in rack. Cap each tube, hold upright, and drag the base of the tube across a test tube rack to resuspend the pelleted cells. It is vital to resuspend cell pellet before adding any significant volume of liquid. Otherwise, cells will clump and be very difficult to assay by flow cytometry. The action of drawing the tubes across the rack should be firm and fast, with the tube maintained in an upright position. This resuspension step may need to be repeated several times.
4. Resuspend the cells in 10 ml HBSS. Centrifuge 10 min at 200 × g, 4°C. 5. Decant supernatant, cap tube, and resuspend the cell pellet as in step 3. 6. Count the cells using a hemacytometer or a Coulter counter (see APPENDIX 3A for details on both methods) and resuspend in an appropriate volume of HBSS at a concentration of 1–2 × 106/ml. Stain the cells 7. Place 5 ml cell suspension (5–10 × 106 cells) in a 50-ml centrifuge tube and add 5 µl of 20 mM DCFH-DA (20 µM final). Gently mix the suspension and place the tube in a 37°C water bath for 15 min. This procedure is generally termed “cell loading.” The final concentration of ethanol from the dye is only 0.1%, which is far below the recommended 1% maximum. If it becomes necessary to exceed this concentration, it is recommended that an ethanol control be run with every series of tubes. It is important that the entire population be loaded in bulk to eliminate the possibility of different dye concentrations in test and control cell populations. If the assay cannot be performed immediately, the cells can be kept on ice for 2 to 3 hr before loading. Keeping loaded cells on ice is also possible but less desirable as dye may leak from the cells. Cells should be warmed to 37°C for 5 min before running the assay.
Prepare the cytometer 8. While the cells are incubating, prepare all other necessary reagents, label and arrange an appropriate number of 12 × 75–mm polyethylene test tubes, and set up the cytometer for the standard measurement of linear FITC fluorescence (excitation at 488 nm, emission collected at 525 ± 20 nm). 9. Run a sample of fluorescent beads and set the high voltage on the green photomultiplier tube to the setting determined previously for standardization of the assay (see Support Protocol 1).
Oxidative Metabolism
Perform the assay 10. Remove the loaded cells from the water bath, mix gently, and aliquot 250 µl cell suspension into each 12 × 75–mm tube. Set up at least one control cell tube (use of duplicate controls is optimal).
9.7.2 Supplement 2
Current Protocols in Cytometry
FALS
neutrophils
monocytes
lymphocytes
Figure 9.7.1 Gating of neutrophils from a suspension of white blood cells. Cells in a white cell suspension fall into distinct clusters on a scatter histogram because of differences in size (FALS = forward-angle light scatter) and granularity (90° scatter). The large granular neutrophils are found at the upper right. A gate is chosen around this population to exclude monocytes and lymphocytes. Physical isolation or purification of the neutrophils is generally unnecessary.
90 scatter
11. Stimulate the first experimental tube with 2.5 µl of 10 µg/ml PMA working solution I (100 ng/ml final). As soon as PMA has been added, run the tube on the flow cytometer to ensure a true zero value. Then proceed to individually stimulate and run each experimental tube, and to run control tubes without addition of PMA. Careful timing is necessary to ensure identical conditions. Stagger addition of activation reagents at fixed intervals (approximately the time required to measure each sample on the flow cytometer) and maintain the same time interval for subsequent measurements. This ensures that the timing is identical for each sample.
12. Gate on the neutrophil population (see Fig. 9.7.1). Measure green fluorescence for at least 5,000 to 10,000 gated cells from each tube. If samples this size cannot be obtained, the assay can still be run with a minimum of 50,000 cells/tube, collecting as few as 500 to 1000 cells; collection times will be longer because of the low cell density. It is necessary to collect these suggested numbers of cells to obtain an adequate representative sample of the population.
MEASUREMENT OF SUPEROXIDE (O2− ) PRODUCTION IN NEUTROPHILS USING HYDROETHIDINE
ALTERNATE PROTOCOL 1
This protocol is used for the detection of superoxide anions (O2−) in cell populations. The probe is hydroethidine (HE), which is freely permeable to the cell. Upon oxidation, ethidium bromide forms and eventually binds to nucleic acids. Fluorescence emission is measured at 620 nm, well above the 525-nm emission wavelength for DCF. HE is relatively specific for O2−, as demonstrated by oxidation in a cell-free assay system using potassium superoxide. Additional Materials (also see Basic Protocol) Cells in culture (optional) 10 mM hydroethidine (HE, from Polysciences; mol. wt. 315; prepare fresh in DMSO and keep on ice, protected from light) 1a. For human neutrophils: Prepare cells as for DCFH-DA procedure (see Basic Protocol, steps 1 to 6). 1b. For cells in culture: Wash cells thoroughly one to three times to ensure all traces of previous culture medium are removed. Resuspend in HBSS at a concentration of 2 × 106 cells/ml.
Studies of Cell Function
9.7.3 Current Protocols in Cytometry
Supplement 2
2. Add 1 µl of 10 mM HE per ml cell suspension (10 µM final) and incubate 5 min at 37°C. Neutrophils load HE very rapidly. Endothelial cells may require longer incubation (up to 45 min). Test cells to ensure that the dye concentration is adequate.
3. Set up the flow cytometer with excitation at 488 nm and emission collected using a 610-nm absorbance long-pass filter. The emission peak of the HE oxidation product, ethidium bromide (EB), is ∼630 nm. A significant amount of fluorescence is emitted above 575 nm.
4. Run a sample of fluorescent beads and set the high voltage on the collection photomultiplier tube to the setting determined previously for standardization of the assay (see Support Protocol 1). Perform the calibration for this assay as described in Support Protocol 1, using ethidium bromide in place of reagent DCF as the calibrating fluorochrome.
5. Perform the assay (see Basic Protocol, steps 10 to 12), collecting EB fluorescence. Establish appropriate gates for flow cytometric analysis, as this is dependent upon the cell type being used. ALTERNATE PROTOCOL 2
SIMULTANEOUS MEASUREMENT OF H2O2 AND O2− IN NEUTROPHILS This protocol combines the above two protocols in a single assay. This technique has the advantage of differentiating between H2O2 and O2− but still provides quantitatively valuable information from the assay system. Because DCF and EB have quite different emission frequencies, there is very little fluorescence overlap. Additional Materials (also see Basic Protocol) 10 mM hydroethidine (HE, Polysciences; MW 315) in DMSO (prepare fresh and keep on ice, protected from light) 1. Prepare cell suspension (see Basic Protocol, steps 1 to 6). 2. Load the DCFH-DA probe first (see Basic Protocol, step 7, for procedure, but use a 10-min incubation). Load one extra tube for DCFH-DA only. It will be necessary to load and stimulate a sample of cells with individual fluorochromes to establish appropriate fluorescence compensation. Although small, the spectal overlap may not be negligible.
3. Load the HE into the same cells (see Alternate Protocol 1, step 2). Load one additional tube for HE only. 4. Proceed with cytometer preparation and the assay (see Basic Protocol, steps 8 to 12) with the following alterations to the basic procedure: excite fluorescence at 488 nm, measure DCF emission at 525 ± 20 nm, and measure EB emission above 610 nm.
Oxidative Metabolism
9.7.4 Supplement 2
Current Protocols in Cytometry
REACTIVE OXYGEN METABOLISM USING DIHYDRORHODAMINE 123 Dihydrorhodamine 123 (DHR) is an uncharged, nonfluorescent dye that is intracellularly oxidized to the highly fluorescent rhodamine 123 during the oxidative burst. The advantage of DHR over DCFH-DA is substantially improved sensitivity, particularly with weak activators such as formyl-methionyl-leucyl-phenylalanine (fMLP).
ALTERNATE PROTOCOL 3
Additional Materials (also see Basic Protocol) 50 mM dihydrorhodamine 123 (DHR; see recipe) 3 mM propidium iodide (PI) solution (see recipe) 100 µM formyl-methionyl-leucyl-phenylalanine (fMLP) working solution (see recipe) 1. Prepare cell suspension as for DCFH-DA procedure (see Basic Protocol, steps 1 to 6). 2. Bulk load the appropriate volume of suspension with 1 µl of 50 mM DHR per ml cells (50 µM final). Incubate 12 min at 37°C. As with DCFH-DA, it is important to load cells in bulk if at all possible, to ensure that the dye concentration is the same for all samples.
3. Add 20 µl of 3 mM PI (60 µM final) per ml cell suspension and incubate another 3 min at 37°C. PI is included for discrimination of live versus dead cells.
4. Set up the flow cytometer with excitation at 488 nm. Use a 520 ± 20–nm bandpass filter for DHR emission (same as for FITC) and a 550-nm long-pass filter for PI emission. 5. Remove the loaded cells from the water bath, mix gently, and place 250-µl aliquots of cell suspension into 12 × 75–nm tubes. Set up at least one control cell tube (use of duplicate controls is optimal). 6. Stimulate the first experimental tube with 2.5 µl of 100 µM fMLP working solution (1 µM final). As soon as fMLP has been added, run the tube on the flow cytometer to ensure a true zero value. Then proceed to individually stimulate and run each experimental tube, and to run control tubes without addition of fMLP. Careful timing is necessary to ensure identical conditions. Stagger addition of activation reagents at fixed intervals (approximately the time required to measure each sample on the flow cytometer) and maintain the same time interval for subsequent measurements. This ensures that the timing is identical for each sample.
7. Gate on the neutrophil population (see Fig. 9.7.1). Measure linear green fluorescence for at least 5,000 to 10,000 gated cells from each tube. If samples this size cannot be obtained, the assay can still be run with a minimum of 50,000 cells/tube, collecting as few as 500 to 1000 cells; collection times will be longer because of the low cell density. It is necessary to collect these suggested numbers of cells to obtain an adequate representative sample of the population.
Studies of Cell Function
9.7.5 Current Protocols in Cytometry
Supplement 2
SUPPORT PROTOCOL 1
CALIBRATION OF THE FLOW CYTOMETER FOR H2O2 MEASUREMENTS Calibration of the flow cytometer allows direct correlation between the fluorescence channel number and the amount of H2O2 formed per cell, so that one can calculate the actual H2O2 production. Cells are loaded with a range of known concentrations of DCFH-DA and maximally stimulated with PMA. Fluorescence of the DCF produced is measured in two ways: on the flow cytometer and in a spectrofluorometer. Actual DCF concentrations in the spectrofluorometer samples can be determined from the calibration curve of reagent DCF and correlated with the fluorescence channel numbers from the flow cytometer. A fluorescent bead is run simultaneously on the flow cytometer so that the settings can be reproduced. Thereafter, the same cytometer is set up exactly as for the calibration, and the fluorescence channel number is directly related to the H2O2 per cell. The calibration is unnecessary for simple qualitative comparison of functional responses of different cell preparations. Additional Materials (also see Basic Protocol) Absolute ethanol 200 ng/ml phorbol myristate acetate (PMA) working solution II (see recipe) PBS-gel (see Support Protocol 3) Reagent-grade dichlorofluorescein (DCF; Sigma) Sonicator 10- to 20-ml polycarbonate centrifuge tubes Statistical software for correlation Prepare cell suspension 1. Collect a large volume of blood, as much as 50 ml if desired (see Basic Protocol, step 1). A significant number of neutrophils is required.
2. Prepare a white blood cell suspension (see Basic Protocol, steps 2 to 6; adjust for the increased blood volume). 3. Make up five working solutions of DCFH-DA in absolute ethanol (or HE in DMSO) at the following concentrations: 0.1 mM, 1 mM, 5 mM, 10 mM, and 20 mM. These solutions will be used at 1 ìl/ml (1/1000 dilution) to load the cell preparations for each dose response.
Make preliminary determination of flow cytometry settings 4. Load 0.5 ml cells with 0.5 µl of 20 mM DCFH-DA (20 µM final) for 15 min at 37°C. 5. Put 100 µl of loaded cells into each of two 12 × 75–mm tubes. Add 100 µl of 200 ng/ml PMA working solution II (100 ng/ml final) to one tube and 100 µl PBS-gel to the other. PMA working solution II is made with PBS-gel to supply the cells with adequate nourishment when the suspension is diluted by the addition of the stimulant. This level of PMA provides maximal stimulation for the cells, which should respond with the maximal possible increase in fluorescence.
6. Incubate the tubes 20 min in a covered 37°C water bath before running on the flow cytometer. Set the photomultiplier tube voltage such that both stimulated and unstimulated cells are at acceptable levels of fluorescence. Oxidative Metabolism
Unstimulated cells should have an on-scale histogram. Stimulated cells can be from five to twenty times brighter, so a log scale may be necessary to keep them on-scale as well.
9.7.6 Supplement 2
Current Protocols in Cytometry
7. Obtain several sets of calibration beads and run them to identify two different beads that fall within the range of the two cell samples. The voltage, gain, and channel position of the beads form the calibrated setup for this instrument. In this manner the flow cytometer can be set at a fixed voltage level for all subsequent measurements.
8. Place 3.5 ml cells at 2 × 106 cells/ml in each of five 50-ml tubes. Label one for each of the dye concentrations in step 3. Stain and stimulate the cells 9. Load each 50-ml tube with 1 µl of the corresponding dye solution and incubate 15 min in a 37°C water bath. 10. For each of the five loaded samples, place a 1-ml aliquot of loaded cells into each of three 12 × 75–mm test tubes and label with the dye concentration and the word “cyto.” Have ready a second triplicate set of empty tubes marked with the dye concentrations and “spectro,” and place these on ice. 11. Return the 12 × 75–mm tubes to the 37° water bath and add 1 ml of 200 ng/ml PMA working solution II (equal volumes of cells and stimulant). Adding an equal volume of PMA dilutes the cells. Therefore, the PMA is made up in PBS-gel to provide them with adequate nourishment. The final PMA concentration of 100 ng/ml ensures maximal stimulation of cells.
12. Incubate the set of tubes 45 min in a 37°C water bath. Run DCF calibration standards on the spectrofluorometer 13. While the cells are incubating, prepare a set of calibration standards using reagentgrade DCF. Prepare at least 4 ml (or a sufficient volume for triplicates of the spectrofluorometer cuvette) of each of the following concentrations in absolute ethanol: 0.1 µM, 0.5 µM, 1 µM, 5 µM, 10 µM, 50 µM, 100 µM, 1 mM, and 5 mM. Reagent DCF is a fluorescent compound identical to that produced from the oxidation of DCFH in the cell.
14. Measure standards in the spectrofluorometer using the following settings: slit = 1, excitation = 488 nm, and emission = 525 nm. Make triplicate measurements for each standard and plot the mean of the concentration versus fluorescence intensity. Disposable plastic cuvettes are recommended for the spectrofluorometry.
15. At the end of the incubation period immediately place all “cyto” tubes from step 12 in an ice bath to stop the reaction; this may take 2 to 3 min but will not affect the assay. Using disposable pipet tips and moving from low to high concentration, pipet 500 µl PMA-activated cell suspension as quickly as possible into the matching tubes labeled “spectro” for each concentration. Keep all tubes on ice. Prepare cell lysates for the spectrofluorometer 16. For each “spectro” tube to be run on the spectrofluorometer, lyse all cells in the suspension by sonicating in a bench sonicator at 30% of maximum power. IMPORTANT NOTE: It is necessary to perform steps 16 to 20 in rapid succession. If personnel allow, do the spectrofluorometry and flow cytometry work simultaneously as described below. Test a cell suspension with the available sonicator to determine the optimal setting. Have tubes on ice for the sonication to prevent heating the suspension. Total disruption of cells will probably require sonication for at least 30 sec. It may be necessary to repeat the sonication. In that case, sonicate all tubes once, then repeat the entire process.
Studies of Cell Function
9.7.7 Current Protocols in Cytometry
Supplement 2
17. Place lysates into polycarbonate centrifuge tubes and centrifuge 50 min at 20,000 × g, 4°C. Run flow cytometry samples 18. While lysates are being centrifuged, collect the fluorescence histograms for each “cyto” sample by flow cytometry using the settings already determined in steps 4 to 7. It is vital to run the flow samples as soon as possible on the flow cytometer. If adjustments in settings are required to run a particular sample, change only the gain, never the high voltage. The gain on a photomultiplier tube is a linear function, whereas the voltage is not. When the gain is changed, the fluorescence mean channel number must be divided by the ratio of gains to make it comparable to measurement with the original gain. For example, if the gain is halved, the mean channel number must be doubled.
Run lysates on the spectrofluorometer 19. Very carefully remove supernatants from the centrifuged lysates (from step 17) into the set of prelabeled “spectro” tubes on ice. 20. With the same settings used for the DCF calibration curve, measure the fluorescence of each tube and note the results. SUPPORT PROTOCOL 2
PREPARATION OF VACUTAINER TUBES WITH PRESERVATIVE-FREE HEPARIN Commercially available blood collection tubes contain preservatives in the anticoagulant. These preservatives can influence cells and interfere with functional assays. This protocol describes how to prepare tubes containing preservative-free heparin. Heparin is preferable to either EDTA or ACD for use in functional assays. NOTE: This preparation must be done in a biological safety cabinet under aseptic conditions. Materials 16 × 100–mm red-top Vacutainer tubes 2-ml bottle of 1000 U/ml preservative-free heparin (LyphoMed) Alcohol wipes 1.0-ml allergist syringes 1. Clean tops of 16 × 100–mm Vacutainer tubes and heparin bottle with alcohol wipes. 2. Draw up 1.0 ml heparin in a syringe. Watch out for bubbles. 3. Carefully insert the needle through the top of a tube and inject 0.1 ml heparin. Take care not to lose vacuum in the collection tubes. Work quickly and maintain back pressure on the syringe plunger because the vacuum will draw in the heparin.
4. Withdraw the needle and repeat injection with the next tube. Continue until the desired number of tubes is reached. 5. Label tubes as heparinized and store up to 1 month at 4°C. Be sure that the labeling is clear so that the tubes are not mistaken for “serum” tubes without anticoagulant. The tubes contain 100 U heparin, for a final concentration of 10 U/ml when filled to a 10-ml capacity. Adjust amount of heparin injected for different volumes. When using these tubes for blood collection, always have extras available in case vacuum has been lost during addition of heparin. Oxidative Metabolism
9.7.8 Supplement 2
Current Protocols in Cytometry
PREPARATION OF PBS-GEL The PBS-gel used in many cell function assays is designed to be a support medium for cells, providing protein, an energy source, and physiological conditions. A gelatin concentrate prepared and diluted in PBS as needed will significantly reduce preparation time for the medium. This protocol gives instructions for 100 ml of 10× concentrate, enough to make ∼8 liters PBS-gel. The aliquots can be stored at 4°C for up to 1 year.
SUPPORT PROTOCOL 3
Materials Disodium EDTA Glucose Gelatin (Difco) PBS (APPENDIX 2A) Prepare concentrate 1. On a heated magnetic stirrer, bring 80 ml distilled, deionized water to 45° to 55°C. Slowly stir in 10.0 g gelatin, 7.604 g disodium EDTA, and 9.00 g glucose. Continue stirring until all reagents are in solution. 2. Dilute to 100 ml with water, continuing to stir and to keep warm. 3. Pipet 1.2-ml aliquots into capped 1.5-ml microcentrifuge tubes, label, and store at 4°C. The specified volume ensures that a full 1.0 ml can be removed from the aliquot.
Prepare PBS-Gel 4. Before use, warm an aliquot of concentrate to ∼45°C. Shake to mix contents and pipet 1.0 ml concentrate into 99 ml warm PBS. Mix and adjust pH to 7.4 with 1 N NaOH. Buffer should appear homogeneous. If clumps are visible, discard and make again. Dispose of any leftover PBS-gel at the end of the day.
REAGENTS AND SOLUTIONS Use distilled, deionized water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Dihydrorhodamine 123 (DHR), 50 mM Dissolve DHR (mol. wt. 346; Molecular Probes) at 17.3 mg/ml in dimethylformamide (DMF). Keep on ice, protected from light. Do not store; for best results, DHR should be freshly made. CAUTION: DMF will attack many plastics. Use a glass container instead.
Formyl-methionyl-leucyl-phenylalanine (fMLP), 100 ìM Stock solution: Dissolve fMLP (mol. wt. 436.7; Sigma) at 4.37 mg/ml (10 mM) in DMSO and store in 10-µl aliquots ≤12 months at −20°C. Working solution: Thaw an aliquot and add 990 µl HBSS (APPENDIX 2A; 100 µM fMLP final). Discard any unused portion. Phorbol myristate acetate (PMA) working solutions I and II Stock solution: Dissolve PMA (mol. wt. 616.84; LC Labs) at 2 mg/ml in DMSO and store in 10-µl aliquots ≤12 months at −20°C. Working solution I: Thaw an aliquot and dilute to 10 µg/ml with PBS. Discard any unused portion. Working solution II: Thaw an aliquot and dilute to 200 ng/ml in PBS-gel (see Support Protocol 3). Discard any unused portion.
Studies of Cell Function
9.7.9 Current Protocols in Cytometry
Supplement 2
Propidium iodide (PI) solution, 3 mM Prepare 2 mg/ml (3 mM) propidium iodide in PBS (APPENDIX 2A). Store wrapped in foil up to 1 month at 4°C. COMMENTARY Background Information
Oxidative Metabolism
Neutrophil defense mechanisms include both oxidative and nonoxidative processes. Clinical syndromes have been described in which a selective depletion of one major component of one or more pathways exists. However, on the whole neutrophils activate many of their defense mechanisms concurrently; thus, small selective defects may not be readily recognized. This occurs in myeloperoxidase (MPO) deficiency, which is characterized by a reduced bactericidal rate for neutrophils, although ultimate killing capacity remains normal. A relatively simple assay system that will provide information on the resting and stimulated nature of highly activatable cells such as neutrophils is quite useful (Robinson et al., 1983). Several activation pathways are possible, and many sites can be blocked or inhibited during these activation processes. Flow cytometry is uniquely suited for measurement of oxidative burst in neutrophils. The major advantages of flow cytometry over more conventional techniques are that relatively small volumes of blood are required for the assays, and the results are objective and quantifiable. Additionally, it is not necessary to purify the cell populations, as must be done for some other bulk methods (see Fig. 9.7.1). The procedure takes less time from venipuncture to assay, a major advantage because the cells are readily activatable. Another valuable advantage of flow cytometry is that measurements are taken on individual cells and provide very accurate results. Importantly, it is possible to measure the production of oxidants within the cell, as the fluorescent indicator is inside the cell itself. Such assays can be performed with as few as 20,000 cells per tube (Loesche et al., 1988); however, the degree of difficulty also increases under these conditions. Measuring the oxidative burst in blood by flow cytometry is a very practical application of that technology, particularly when the blood must be taken from neonates or small children. Before the development of the dichlorofluorescein (DCF) assay by flow cytometry (Bass et al., 1983; Duque et al., 1985) accurate measures of intracellular H2O2 were not possible;
only bulk determinations could be made. Basic Protocol 1 is a further development of the work of Bass and colleagues. The assay depends upon the incorporation of 2′,7′-dichlorofluorescin diacetate (DCFH-DA) into the hydrophobic lipid regions of the cell. Hydrolytic enzymes cleave the acetate moieties, leaving the nonfluorescent molecule 2′,7′-dichlorofluorescin (DCFH), which is trapped within the cell due to its polarity. Upon cell activation, NADPH oxidase catalyzes the reduction of O2 to O2−, which is then further reduced to H2O2. H2O2 and peroxidases are able to oxidize the trapped DCFH to 2′,7′-dichlorofluorescein (DCF), which is highly fluorescent at 530 nm. The green fluorescence produced is proportional to the amount of H2O2 generated. Alternatives to DCFH-DA include dihydrorhodamine 123 (DHR), an uncharged, nonfluorescent derivative of the laser dye rhodamine 123, and hydroethidine (HE). DHR can be substituted for DCFH-DA for measurement of H202, and exhibits very similar emission properties (Rothe et al., 1988). The advantages of DHR include as much as three-fold greater sensitivity (Rothe et al., 1988) and less susceptibility to leakage from the cell, which is a major problem for DCFH-DA and must be carefully controlled in the experimental design. HE is used to measure O2−, and can be used on its own or in combination with DCFH-DA (Rothe and Valet, 1990). HE is able to permeate the cell membrane easily. It is then oxidized to ethidium bromide, which is trapped in the nucleus by interchalation into DNA. Because the amount of fluorescent dye (DCF) formed is proportional to the cellular oxidant production, the green fluorescence intensity becomes a measure of the oxidants produced by the cells. Cells, however, respond differently to different activation substances, so it is possible to replace the PMA in the Basic Protocol with formyl peptides (e.g., fMLP), TNF-α, IFN-γ, IgG, bacteria, yeast, or yeast products (e.g., zymosan). Each substance will activate the cells in a unique manner, so a complete analysis of the metabolic processes can be made. While this protocol is designed primarily to study neutrophils, other cell types may be used.
9.7.10 Supplement 2
Current Protocols in Cytometry
Fluorescence/cell (cytometer)
500
100
10 10
100
1000
pmol DCF/million cells (spectrofluorometer)
Figure 9.7.2 Correlation of mean fluorescence per cell by flow cytometry with mean concentration of reagent DCF per 5 million cells by spectrophotometry. Neutrophils loaded with varying concentrations of DCFH-DA are stimulated with PMA and divided into aliquots for analysis on the flow cytometer and on the spectrofluorometer (as cell lysate). Spectrofluorometric values are determined from a standard curve generated, such as this, using reagent DCF.
These include but may not be restricted to monocytes or macrophages, lymphocytes, endothelial cells, keratinocytes, and chondrocytes. Protocols for each cell type must be modified to allow for differences in cellular esterase content, cell size, activity, lifespan, and other factors. For certain studies it may be necessary or desirable to calculate the exact amount of H2O2 produced. In those instances, the flow cytometer must be calibrated (see Support Protocol 1) so that all measurements from the flow assay are directly correlated and quantitatively accurate. This is accomplished by establishing a correlation between the fluorescence of maximally stimulated cells on the flow cytometer and that of reagent DCF on the spectrofluorometer. Cells are run on the flow cytometer to determine fluorescence channel number and also on the spectrofluorometer after lysing by sonication, to measure the DCF produced (see Fig. 9.7.2). Finally, a correlation coefficient is calculated between the flow cytometer and spectrofluorometer setting, and the equation describing this relationship links the flow channel
number to the amount of H2O2 produced. Because all measurements on the flow cytometer are on a per cell basis, the calculation becomes a single-cell calculation when the total number of cells is taken into consideration. Each time an assay is run, the cytometer is set up exactly as described for the calibration, and the channel number for the fluorescence becomes directly related to the amount of H2O2 per cell. The calibration needs to be performed on each flow cytometer used for the assays; recalibration is necessary only when photomultiplier tubes or amplifiers are changed.
Critical Parameters One of the most important considerations in performing these functional assays is maintenance of an unstimulated, resting cell suspension. Therefore, keep cell suspensions on ice except when actual measurements are being made. Further, load with indicator probe (e.g., DCFH-DA) only the volume of cells necessary for each assay. Keep the bulk cell suspension on ice at all times and load the next set of cells 15 min before the last assay is complete. Always use fresh sterile plasticware for cell and
Studies of Cell Function
9.7.11 Current Protocols in Cytometry
Supplement 2
reagent preparation. Bacterial contamination will not be a problem because the time span is short; the biggest problem normally encountered is the accidental introduction of activator to the cell population by a contaminated pipet tip. For the most accurate calculations, collect linear green fluorescence. It may be necessary to collect the signal on two different amplifiers with gains set independently, or even on a logarithmic amplifier. The expected increase in fluorescence is a factor of ten- to twenty-fold over unstimulated levels. One technique for maintaining temperature control over an extended time period is simply to keep all tubes in a 37°C water bath, removing the tubes only to put them onto the flow cytometer. Prepare the samples in tubes that fit on the flow cytometer so that they can be taken straight from the rack to the instrument. If several measurements are to be made on a single tube over time, make sure the volume of cells is sufficient. Some cytometers remove a fixed amount from each tube, whereas others do not. In either case, modify the volume of cells to be adequate for several measurements. For example, 300 µl would be a minimum needed for four measurements. When running an assay for 60 to 90 min, it is vital that the cytometer be maintained in a well-calibrated condition throughout. Therefore, it may not be possible to run other samples on the instrument while this assay is proceeding. Live cells must be maintained in good metabolic condition for functional studies. This is crucial for neutrophils, which are very shortlived cells. The presence of glucose in the incubation buffers is necessary, as is maintaining the cell suspension on ice to prevent clumping and activation. Either clumped or activated cells can ruin the assay (see Troubleshooting).
Troubleshooting
Oxidative Metabolism
Several problems may be encountered with these assays. Cell activation. Neutrophils are delicate cells to maintain in suspension. They can easily activate each other, or they can be activated by small quantities of enzymes or H2O2 in the extracellular milieu. Once activated, they can begin oxidation of the fluorescent probe before the assay is started. Very bright background fluorescence will result, with only a small increase in experimentally stimulated cells over control.
Cell clumping. Activated cells will also clump, making them impossible to run through the flow cytometer. Satisfactory, though inelegant, methods exist for rescuing valuable cells from this condition. Physical removal of small clumps with a pipet may forestall further clumping. Small clumps beget large clumps! If there are many clumps and a large number of cells, filter the entire suspension through several layers of sterile surgical gauze. Recounting the cell suspension is a small price to pay for saving the experiment. Very bright control cells. On occasion the entire cell population appears to become activated, resulting in high fluorescence intensity even in unstimulated cells. Essentially nothing can be done in this situation except to start over from the beginning with a new blood sample. Use new sterile test tubes and never reuse these for cell isolation or preparation. Highly fluorescent controls can also result from an oxidized fluorescent probe solution. It is recommended that the probe be made up daily to minimize this problem. Cell maintenance over long periods. Very long assays lasting more than a few hours after cell isolation may prove problematic. First, unstimulated cells won’t last forever. Thus, poor functional activity may be simply a result of cell age. Second, indicator dye can leak out through cell membranes with time. It is preferable to keep cells on ice unloaded, incubating with dye just before the assay is run, rather than holding loaded cells for an extended period.
Anticipated Results In a mixed white cell population the most reactive cells will usually be the granulocytes. Normally, these cells produce a respiratory burst response marked by a five- to twenty-fold increase in fluorescence intensity. With linear amplification (which is preferable), highly stimulated populations may build up in the highest channel of the histogram, as illustrated in Figure 9.7.3. However, the magnitude of this change is often dependent on the stimulant chosen and its concentration. For example, PMA in the range of 8 ng/ml provides halfmaximal stimulation, whereas 100 ng/ml will always provide maximal stimulation. fMLP is a much less potent activator, and even 10−6 M may not produce a noticeable response in granulocytes loaded with DCFH-DA. In this situation, the DHR probe provides a more sensitive evaluation where at least a four-fold response can be expected.
9.7.12 Supplement 2
Current Protocols in Cytometry
Time Considerations
Literature Cited
Two hours should be the maximum time for preparing all cell suspensions. The activation studies should take an additional hour. A wellorganized routine will keep these times to a minimum and the cells in prime condition. Depending upon the type of cytometer and the nature of the data analysis available, adequate data analysis may take several hours.
Bass, D.A., Parce, J.W., De Chatelet, L.R., Szejda, P., Seeds, M.C., and Thomas, M. 1983. Flow cytometric studies of oxidative product formation by neutrophils: A graded response to membrane stimulation. J. Immunol. 130:1910-1917. Duque, R.E., Robinson, J.P., Hudson, J.L., Till, G.O., and Ward, P.A. 1985. Fed. Proc. (abstr.). Loesche, W.J., Robinson, J.P., Flynn, M., Hudson, J.L., and Duque, R.E. 1988. Infect. Immun. 56:156-160.
Cell number
Fluo resc enc e
A
0
600
1200
1800
Time (sec)
Fluo resc enc e
Cell number
B
0
600 1200 Time (sec)
1800
Figure 9.7.3 Stimulation of neutrophils by PMA over time. (A) Unstimulated cells, which show no increase in DCF fluorescence over 30 min. (B) Activated cells (stimulated with 50 ng/ml PMA), which show a rapid 10-fold increase in fluorescence following an initial time lag. Data were collected by sampling the same set of tubes multiple times over a 30-min period.
Studies of Cell Function
9.7.13 Current Protocols in Cytometry
Supplement 2
Robinson, J.P., Bruner, L.H., Bassøe, C.F., Hudson, J.L., Ward, P.A., and Pham, S.H. 1983. Measurement of intracellular fluorescence on human monocytes relative to oxidative metabolism. J. Leukocyte Biol. 43:304-310. Rothe, G. and Valet, G. 1990. Flow cytometric analysis of respiratory burst activity in phagocytes with hydroethidine and 2′,7′-dichlorofluorescin. J. Leukocyte Biol. 47:440-448.
Rothe, G., Oser, A., and Valet, G. 1988. Dihydrorhodamine 123: A new flow cytometric indicator for respiratory burst activity in neutrophil granulocytes. Naturwissenschaften 75:354-355.
Contributed by J. Paul Robinson Purdue University Cytometry Laboratories West Lafayette, Indiana
Oxidative Metabolism
9.7.14 Supplement 2
Current Protocols in Cytometry
Measurement of Intracellular Calcium Ions by Flow Cytometry
UNIT 9.8
The flow cytometer can be used to measure various functional parameters that are of increasing interest to immunologists. The recent development of a number of new fluorescent probes makes it possible to measure the concentrations of various intracellular free ions in single living cells. Among these ions are calcium, magnesium, sodium, potassium, and hydrogen (pH). Most earlier techniques for measuring cellular activation parameters determined the mean value for a population of cells, which did not permit optimal resolution of the responses. The flow cytometer is particularly useful for this purpose because it can measure ion concentrations in large numbers of single cells and thereby allows ion concentration to be correlated with other parameters such as immunophenotype and cell cycle stage. In many cases, there is marked heterogeneity in the changes that occur, sometimes even in populations of cells that were previously thought to be homogeneous. A limitation of flow cytometry, however, is that it does not permit kinetic resolution of certain complex kinetic responses such as cellular oscillatory responses (Osipchuk and Cahalan, 1992; Allbritton and Meyer, 1993). This requires video microscopy with digital image analysis, a technique that is complementary to flow cytometry for the study of various parameters of cell activation. This unit describes flow cytometric protocols using the dyes indo-1 (see Basic Protocol), fluo-3 (see Alternate Protocol 1), and Fura Red (see Alternate Protocol 2) to measure intracellular calcium concentration. Also presented are the use of a detergent to facilitate dye loading (see Alternate Protocol 3) and the use of calcium buffers to calibrate a flow cytometric calcium assay (see Support Protocol). A spectrofluorometer method for the measurement of intracellular Ca2+ can be found in June et al. (1995). USE OF INDO-1 AND FLOW CYTOMETRY TO MEASURE CELLULAR CALCIUM CONCENTRATION
BASIC PROTOCOL
In this protocol, intracellular ionized calcium concentration ([Ca2+]i) is measured using indo-1 dye and ratiometric analysis. Most commercially available flow cytometers can be used to perform this assay, provided the instrument is capable of UV illumination; however, note that the Becton Dickinson FACScan and Coulter Profile and XL cannot be used. In addition, on many instruments, cells may be electronically sorted based on a particular calcium response; the sorted cells can be cultured for later analysis. The protocol can be divided into three stages: preparation of cells to be analyzed, setup of the flow cytometer, and data analysis and display (UNITS 10.3–10.6). The protocol requires expertise in basic flow cytometric techniques. Materials Murine splenic lymphocytes or human peripheral blood lymphocytes (UNIT 5.1) Cell loading medium (see recipe) 100 mM probenecid (see recipe) 2 mg/ml indo-1 pentaacetoxymethyl ester (indo-1; see recipe) 1 mg/ml ionomycin (see recipe) Dimethyl sulfoxide (DMSO; Sigma) or 10% (v/v) bleach in water Saline (0.85% [(w/v)] NaCl) or PBS (APPENDIX 2A) 12 × 75–mm polypropylene tubes (Falcon) Beckman TJ-6 rotor (or equivalent) 30° or 37°C water bath Studies of Cell Function Contributed by Carl H. June, Ryo Abe, and Peter S. Rabinovitch Current Protocols in Cytometry (1997) 9.8.1–9.8.19 Copyright © 1997 by John Wiley & Sons, Inc.
9.8.1 Supplement 2
Flow cytometer with UV light source and heated sample chamber (e.g., Becton Dickinson, Coulter, or Ortho), and software for kinetic and ratiometric analysis (Phoenix Flow Systems) Load cells with indo-1 1. Collect lymphocytes in 12 × 75–mm polypropylene tubes and centrifuge 6 min at 180 × g (950 rpm in Beckman TJ-6), room temperature. Resuspend pellet in cell loading medium at 106 to 107 cells/ml. Use murine splenic lymphocytes or human peripheral blood lymphocytes in initial experiments because they are easily and reliably loaded. Later, when the other aspects of the technique have been validated on the flow cytometer, both adherent and nonadherent cells can be analyzed. Use polypropylene tubes to minimize loss of cells.
2. Optional: Add 100 mM probenecid (4 mM final). Probenecid may improve cellular loading by minimizing leakage of indo-1 and cell-to-cell variation in dye content; sulfinpyrazone will also work (see Critical Parameters and Troubleshooting section on cellular response).
3. Add 2 mg/ml indo-1 to 2 µg/ml final. Incubate 30 min at 30° or 37°C. The cells are loaded with the membrane-permeant form (pentaacetoxymethyl ester) of indo-1 (indo-1). Cellular esterases cleave the AM moiety, resulting in the trapping of the highly charged indo-1 in the cells. Typically, ∼20% of the dye becomes trapped and concentrated within the cell. Optimal conditions for loading must be empirically determined for each cell type. Rates of loading of indo-1 vary between cell types, especially as a consequence of variations in intracellular esterase activity. More rapid loading rates are seen in platelets and monocytes than in lymphocytes, and in growing cell lines rather than resting cells. More uniform loading is often observed if pluronic F-127 is included together with indo-1 (see Alternate Protocol 3). In addition, incubation at 30°C can aid loading of cells that tend to compartmentalize the dye. An optional step is to stain an aliquot of indo-1-loaded cells for simultaneous immunofluorescence analysis (UNIT 6.2). This is done by treating the cells with saturating amounts of azide-free FITC- or phycoerythrin (PE)-conjugated antibody—as determined by titration experiments (UNIT 4.1) or by manufacturer’s recommendations—and incubating 20 min at 22°C. Incubation may be done at 4°C to minimize antigen modulation, but after chilling, cells may require an extended equilibration time at 37°C to return to physiologic functioning for the calcium assay. In cases where antigen modulation is a problem, PE-conjugated antibodies are preferred, because unlike FITC, PE can still be detected after cellular internalization.
4. Centrifuge cells 6 min at 180 × g, room temperature. Gently resuspend (do not vortex) cells in cell loading medium at the desired cell concentration (∼3 × 106 cells/ml). Store cells at room temperature and protect from light until analysis. Leakage and compartmentalization of indo-1 is accelerated if cells are stored at 37°C. It is often preferable to let cells rest for ∼15 min before starting analysis. This presumably allows complete conversion of the calcium-insensitive form of indo-1 ester into the calcium-sensitive (charged) form of indo-1 and enhances cell uptake of additional equilibrating calcium to compensate for indo-1-bound calcium. Measurement of Intracellular Calcium Ions by Flow Cytometry
Set up the flow cytometer 5. Set up and adjust flow cytometer. Use a violet bandpass filter centered at 395 ± 10 nm and a blue bandpass filter centered at 500 ± 15 nm or a green bandpass filter centered at 525 ± 15 nm (see Critical Parameters and Troubleshooting).
9.8.2 Supplement 2
Current Protocols in Cytometry
6. Set light scatter gates and optimize photomultiplier tube gain settings by placing the mean blue fluorescence in the upper half of the histogram channels and the violet fluorescence in the lower half of the histogram channels. Use linear rather than logarithmic amplification, and gate out dead cells with light scatter and violet fluorescence windows (see Critical Parameters and Troubleshooting). For kinetic experiments such as this, the cytometer should be able to display time as a parameter. Violet and blue fluorescence can be plotted versus time; however, the violet/blue indo-1 ratio is optimally observed by allowing the cytometer to calculate this ratio, which can then be plotted as a function of time (see Anticipated Results and Figure 9.8.2).
7. Check instrument setup and cellular loading by treating ∼1 × 105 cells (in cell loading medium) with 1 mg/ml ionomycin at 1 to 2 µg/ml final. An immediate response in 100% of cells should occur. Carefully remove any ionomycin that may adhere to the tubing by flushing the lines 1 min with DMSO or 10% bleach, then 1 min with cell loading medium. It is critical that each experiment include a determination of R, the ratio of violet/blue fluorescence of resting cells, and Rmax, the ratio of violet/blue fluorescence of cells after stimulation by the calcium-ionophore ionomycin. If the instrument is properly aligned and the cells loaded adequately, the ratio of Rmax to R is 6:9 and >99% of the cells respond (see Critical Parameters and Troubleshooting). Use fluorescence microscopy to verify quality of indo-1 loading; compartmentalization of indo-1 is indicated by the presence of punctate dots of fluorescence.
8. To calibrate the indo-1 fluorescence ratio to the concentration of ionized calcium ([Ca2+]i), suspend cells in a series of calcium/EGTA buffers and ionomycin (see Support Protocol for details). Analyze indo-1-loaded cells 9. Warm an aliquot of indo-1-loaded cells 5 to 10 min at 37°C before analysis. Use ∼5 × 105 cells per 10-min assay. 10. Analyze cells at 37°C in cell loading medium. Use saline or PBS for the sheath fluid. The rate of cell analysis can vary. Commonly, cells are analyzed at 200 to 300 cells/sec. A higher flow rate may be required when sorting cells or when analyzing a rare event, such as a calcium signal in a small subset of cells identified by monoclonal antibody staining (see Critical Parameters and Troubleshooting).
11. Optional (for instruments with dual-beam illumination): Set regions for simultaneous immunofluorescence analysis. Combining the use of FITC and PE with indo-1 analysis allows determination of [Ca2+]i in complex immunophenotypic subsets. Limiting the analysis of indo-1 fluorescence to windows of FITC versus PE fluorescence allows information relating to each identifiable cellular subset to be derived from a single sample. On instruments without provision for analysis of four separate fluorescence wavelengths, both the FITC signal and the signal from calcium-free indo-1 can be detected with the same filter element. Note that the FITC signal must be delayed from the long-wavelength indo-1 signal derived from the UV laser.
12. If desired, sort cells on the basis of [Ca2+]i responses. Sorting on the basis of indo-1 fluorescence can be an important tool for selection and identification of genetic variants in biochemical pathways leading to Ca2+ mobilization and cell growth and differentiation (see Background Information).
Studies of Cell Function
9.8.3 Current Protocols in Cytometry
Supplement 2
ALTERNATE PROTOCOL 1
USE OF FLUO-3 AND FLOW CYTOMETRY TO MEASURE CELLULAR CALCIUM CONCENTRATION Fluo-3 is a fluorescein-based calcium probe developed by Minta et al. (1989). Use of this probe allows flow cytometric measurement of calcium on instruments that are not equipped with a UV light source. Additional Materials (also see Basic Protocol) 10 mg/ml fluo-3 acetoxymethyl ester (fluo-3 AM; see recipe) Load cells with fluo-3 AM 1. Collect lymphocytes in 12 × 75–mm polypropylene tube and centrifuge 6 min at 180 × g (950 rpm in Beckman TJ-6), room temperature. Resuspend pellet in cell loading medium at 106 to 107 cells/ml. 2. Optional: Add 100 mM probenecid to 4 mM final. Probenecid may improve cellular loading and cell-to-cell variation in dye content by minimizing leakage of fluo-3 AM; sulfinpyrazone will also work (see Critical Parameters and Troubleshooting section on cellular response).
3. Add 10 mg/ml fluo-3 AM at 3 to 4 µg/ml final. Incubate 30 min at 30° or 37°C. For additional comments related to loading, see Basic Protocol, step 3. Note that when performing simultaneous immunofluorescence analysis with fluo-3, only PE-conjugated antibodies can be used.
4. Centrifuge cells 6 min at 180 × g, room temperature. Gently resuspend pellet in cell loading medium at the desired cell concentration (~3 × 106/ml). Store cells at room temperature and protect from light until analysis. Leakage and compartmentalization of fluo-3 is accelerated if cells are stored at 37°C. It is often preferable to let cells rest ∼15 min before analysis. This presumably allows complete conversion of the calcium-insensitive form to the calcium-sensitive form of the dye, and enhances cell uptake of additional equilibrating calcium to compensate for dye-bound calcium.
Set up the flow cytometer 5. Excite fluorescence with the 488-nm line of an argon ion laser. Collect fluorescence at 525 nm using linear amplification. The instrument must be equipped with a means to analyze cells that are maintained at 37°C.
Analyze fluo-3-loaded cells 6. Analyze cells by gating on forward and right-angle light scatter. Exclude dead cells by gating out cells without fluo-3 fluorescence. 7. Determine adequacy of cellular loading and instrumental alignment by treating a sample of cells with 1 mg/ml ionomycin at 2 µg/ml final (see Basic Protocol, step 7). If <100% of cells respond to the calcium ionophore, examine the cells with a fluorescence microscope to check for compartmentalization of fluo-3.
8. Remove residual ionomycin from the instrument by flushing the line 1 min with DMSO or 10% bleach and 1 min with cell loading medium.
Measurement of Intracellular Calcium Ions by Flow Cytometry
9. Analyze results by determining the mean response versus time and the fraction of responding cells versus time. It is more difficult to determine the percentage of responding cells with fluo-3 than with indo-1. This is because with fluo-3 there is much more heterogeneity in the distribution of fluorescence values of unstimulated cells, due to the contribution of both cell size (fluo-3
9.8.4 Supplement 2
Current Protocols in Cytometry
content) and cell calcium concentration to the signal. Consequently, there is relatively poor separation between responding cells and nonresponding cells. Software approaches for such analysis from kinetic data have been previously described (Rabinovitch et al., 1992). Improved resolution of responses can be obtained by analysis of the ratio of fluo-3 to Fura Red fluorescence (see Alternate Protocol 2).
SIMULTANEOUS USE OF FLUO-3 AND FURA RED FLUORESCENCE RATIOS FOR FLOW CYTOMETRIC CALCIUM MEASUREMENT
ALTERNATE PROTOCOL 2
With this protocol, a high-sensitivity, low-noise assay for calcium concentration changes can be performed using dyes that are excited at 488 nm (Lipp and Niggli, 1993; Novak and Rabinovitch, 1994). Previously, the indo-1-based ratiometric assay using UV illumination was required to detect small subsets of responding cells using flow cytometric assays. Combined cellular loading with two nonratiometric calcium probes, Fura Red and fluo-3, now permits a sensitive ratiometric assay using visual illumination. Additional Materials (also see Basic Protocol) 10 mg/ml fluo-3 acetoxymethyl ester (fluo-3 AM; see recipe) 10 mg/ml Fura Red acetoxymethyl ester (Fura Red AM; see recipe) R-phycoerythrin (PE)–labeled antibody (optional) Load cells with fluo-3 AM and Fura Red AM 1. Collect cells in 12 × 75–mm polypropylene tube and centrifuge 6 min at 180 × g (950 rpm in Beckman TJ-6), room temperature. Resuspend pellet in cell loading medium at 106 to 107 cells/ml. 2. Optional: Add 100 mM probenecid to 4 mM final. Probenecid may improve cellular loading and cell-to-cell variation in dye content by minimizing leakage of fluo-3 AM; sulfinpyrazone will also work (see Critical Parameters and Troubleshooting section on cellular response).
3. Add 10 mg/ml fluo-3 AM at 4 µg/ml final and 10 mg/ml Fura Red AM at 10 µg/ml final. Incubate 30 min at 30°C or 37°C. The amounts of fluo-3 AM and Fura Red AM may require adjustment; the goal is to achieve simultaneous loading with balanced emissions from the two probes. Fura Red has relatively weak fluorescence, so ∼2.5 to 3 times more Fura Red AM than fluo-3 AM should be used (Lipp and Niggli, 1993; Schild et al., 1994). For additional information related to loading, see Basic Protocol, step 3. Note that when performing simultaneous immunofluorescence analysis with fluo-3 and Fura Red, only PE-conjugated antibodies can be used.
4. Centrifuge cells 6 min at 180 × g. Gently resuspend pellet in cell loading medium at the desired cell concentration (∼3 × 106 cells/ml). Store cells at room temperature and protect from light until analysis. Leakage and compartmentalization of the probes is accelerated if cells are stored at 37°C. It is often preferable to let cells rest ∼15 min before analysis. This presumably allows complete conversion of the calcium-insensitive form to the calcium-sensitive form of the dye, and enhances cell uptake of additional equilibrating calcium to compensate for dye-bound calcium.
5. Optional: Stain cells with R-phycoerythrin (PE)–labeled antibody.
Studies of Cell Function
9.8.5 Current Protocols in Cytometry
Supplement 2
Set up flow cytometer 6. Excite fluorescence with the 488-nm line of an argon ion laser. Collect fluo-3 fluorescence at 515 to 535 nm and Fura Red emission at 665 to 685 nm using linear amplification. The instrument must be equipped with a means to analyze cells that are maintained at 37°C.
Analyze fluo-3- and Fura Red–loaded cells 7. Analyze cells by gating on forward and right angle light scatter. Exclude dead cells by gating out cells without fluo-3 fluorescence. Collect fluo-3 fluorescence versus time and Fura Red fluorescence versus time. Use software to analyze the fluo-3/Fura Red ratio versus time in a fashion similar to indo-1 data analysis (see Basic Protocol, step 6, and see Anticipated Results). There is spectral overlap between the Fura Red AM and PE fluorescence emissions; spectral compensation is required to separate the signals.
8. Determine adequacy of cellular loading and instrumental alignment by treating a sample of cells with 1 mg/ml ionomycin at 2 µg/ml final. If <100% of cells respond to the calcium ionophore, examine the cells with a fluorescence microscope to check for compartmentalization of fluo-3 or Fura Red.
9. Remove residual ionomycin from the instrument by flushing the line 1 min with DMSO or 10% bleach and 1 min with cell loading medium. 10. Analyze results by determining the mean response versus time and the fraction of responding cells versus time. ALTERNATE PROTOCOL 3
USE OF PLURONIC DETERGENT F-127 TO LOAD CELLS WITH INDO-1 AM, FLUO-3 AM, OR FLUO-3 AM AND FURA RED AM FOR FLOW CYTOMETRIC CALCIUM MEASUREMENT When using cell lines or cells other than lymphocytes, it is not uncommon to have difficulty with proper loading, as evidenced by poor shifts after cellular stimulation. Poor loading is commonly due to compartmentalization and incomplete hydrolysis of the dye ester. Use of pluronic detergent may sometimes circumvent this difficulty (Poenie et al., 1986; Lanza et al., 1987; Vandenberghe and Ceuppens, 1990), and with some cell types causes a spectacular improvement in signaling. In addition, there is less cell-to-cell heterogeneity in indo-1 and fluo-3 AM uptake after loading in the presence of pluronic F-127. To load cells in the presence of pluronic F–127, proceed with Basic Protocol, Alternate Protocol 1, or Alternate Protocol 2 with the changes indicated below. Additional Materials (also see Basic Protocol) 50-µg vial Ca2+ probe (Molecular Probes): indo-1 pentaacetoxymethyl ester (indo-1 AM), fluo-3 acetoxymethyl ester (fluo-3 AM), or Fura Red acetoxymethyl ester (Fura Red AM) 20% (w/v) pluronic F-127 in DMSO (Molecular Probes; store indefinitely at −20°C; if highly viscous, warm prior to use); warm at 37°C until dissolved FBS (heat inactivated 1 hr at 56°C)
Measurement of Intracellular Calcium Ions by Flow Cytometry
1. After resuspending cells, prepare probe/pluronic mixture: To a 50-µg vial of Ca2+ probe, add 25 µl (for indo-1 or fluo-3) or 10 µl (for Fura Red) of 20% pluronic F-127 and 113 µl FBS. Mix and allow to dissolve 5 min.
9.8.6 Supplement 2
Current Protocols in Cytometry
2. Add 5 to 8 µl (indo-1 or fluo-3) or 12.5 to 20 µl (Fura Red) of the probe/pluronic mixture from step 1 to 1 ml cells (instead of probe alone) and incubate 30 min at 30° or 37°C. The final concentrations of the probes and pluronic F-127 are 1.8 to 2.9 ìM (indo-1), 1.7 to 2.8 ìM (fluo-3), 5.3 to 8.3 ìM (Fura Red), and 0.02% (F-127). Slightly more or less probe mixture may be required, depending on the probe batch and cell type.
USE OF CALCIUM/EGTA BUFFERS TO CALIBRATE FLOW CYTOMETRIC CALCIUM MEASUREMENTS
SUPPORT PROTOCOL
Calibration of flow cytometric calcium measurements may be performed using either a spectrofluorometer (Rabinovitch et al., 1986; June et al., 1995; also see Commentary) or by the use of the protocol described below, which employs a series of precisely prepared calcium buffer solutions to perform an in situ calibration in cells. The protocol (Chused et al., 1987; Li et al., 1987) can be carried out with the aid of commercially prepared calcium buffer kits. It is not possible to directly prepare a solution that contains calcium in a concentration similar to that found inside living cells due to the contamination of laboratory water by calcium (generally micromolar amounts). In addition, it is not possible to prepare solutions with sufficient precision using gravimetric methods because EGTA contains variable amounts of water (Miller and Smith, 1984). Therefore, buffers of known free calcium concentration are useful for experiments in which cells are maintained at a particular calcium concentration and for standardization of indo-1 fluorescence ratio changes. To prepare solutions of known calcium concentration, prepare a binary buffer solution consisting of buffer B (an equimolar solution of calcium and a calcium chelator) and buffer A (identical to buffer B except that it lacks calcium). Either EGTA or BAPTA is used as the chelator because they both have high selectivity for calcium over magnesium, an ion that is ∼104 times more abundant than calcium in the cytoplasm, and therefore can be used to control calcium when present at physiological concentrations. To obtain a buffer with a desired calcium concentration, the experimental conditions (temperature, pH, magnesium concentration, and ionic strength) are entered into a set of equations to determine the necessary ratio of buffers A snd B. Additional Materials (also see Basic Protocol) 100 mM K2H2EGTA (Molecular Probes) 100 mM K2CaEGTA (Molecular Probes) Poisoned Dulbecco’s phosphate-buffered saline (DPBS; see recipe) Prepare calcium/EGTA buffers 1. Add 10 ml of 100 mM K2H2EGTA to 90 ml poisoned DPBS (buffer A). Add 10 ml of 100 mM K2CaEGTA to 90 ml poisoned DPBS (buffer B). Use plasticware rather than glassware in the preparation of the calcium buffers, as glassware may be a source of calcium contamination.
2. Prepare buffer 2 by mixing 40 ml of buffer A plus 10 ml buffer B. 3. Prepare buffer 3 by mixing 36 ml of buffer 2 plus 10 ml buffer B. 4. Prepare buffer 4 by mixing 32 ml of buffer 3 plus 10 ml buffer B. 5. Prepare buffer 5 by mixing 28 ml of buffer 4 plus 12 ml buffer B. 6. Prepare buffer 6 by mixing 24 ml of buffer 5 plus 14 ml buffer B. 7. Prepare buffer 7 by mixing 20 ml of buffer 6 plus 18 ml buffer B. The above steps result in the preparation of a series of eight buffers with increasing calcium concentration from nanomolar to micromolar. See Table 9.8.1 for free calcium concentra-
Studies of Cell Function
9.8.7 Current Protocols in Cytometry
Supplement 2
Table 9.8.1
Free Calcium Concentration in Calibration Buffersa
Buffer
Total [Ca2+] (mM)
[Ca2+]i (no Mg2+)
[Ca2+]i ([Mg2+] = 1 mM)
A 2 3 4 5 6 7 B
0 2.00 3.74 5.23 6.67 7.89 8.89 10
0 27 nM 65 nM 120 nM 218 nM 410 nM 876 nM ~33 µM
0 31 nM 74 nM 135 nM 246 nM 462 nM 988 nM ~35 µM
aAt 37°C and pH 7.20.
tions of the buffers at 37°C and pH 7.20. Free calcium concentrations are given for either 0 mM or 1 mM magnesium (an approximate intracellular concentration). To obtain other buffers of desired free calcium concentrations, determine the amount of total calcium and EGTA required and mix buffer A and buffer B in the necessary ratio. See the protocol provided by Molecular Probes for background information, and for details of calculating the calcium concentration in the buffer solutions (Haugland, 1992). The complexity of these calculations makes use of a computer highly desirable. A program that will determine ionized calcium and magnesium concentrations as a function of pH, total calcium and magnesium concentrations, temperature, and ionic strength is available from the authors. EGTA is useful for buffers between 10−8 M and 10−5 M ionized calcium (pCa of 8 to 5). Nitriloacetic acid (NTA) or dinitro BAPTA (Pethig et al., 1989) should be used to prepare buffers of pCa 4 to 5. Gravimetric methods may be used to prepare calcium solutions of pCa ≤3. Control of pH is especially important when using calcium/EGTA buffers. For example, changing the pH from 7.4 to 7.1 can result in a change in ionized calcium concentration from ∼0.1 ìM to ∼0.4 ìM. EGTA is not specific for calcium; it binds many metals more avidly. Aluminum and lanthanum are occasionally added to cells to study GTP-binding proteins and calcium channels. BAPTA, a calcium buffer that retains selectivity over magnesium, may be used instead of EGTA. BAPTA has much less pH sensitivity than EGTA (Tsien, 1980).
8. If desired, check buffers by measuring the actual free calcium concentration with the salts of indo-1 AM or fluo-3 AM as described in the spectrofluorometer protocol. The dye esters are not calcium sensitive, so the salts must be used for this measurement.
Perform in situ calibration assay using flow cytometer 9. Load mouse or human lymphocytes with indo-1 AM, fluo-3 AM, or fluo-3 AM plus Fura Red AM in buffer A containing poisons (see step 3 of Basic Protocol, Alternate Protocol 1, and Alternate Protocol 2, respectively). 10. Wash cells once with cell loading medium and resuspend aliquots of 5 × 105 cells in 1 ml in each of the eight buffers. Take care to assure that there is no serum carryover; serum will buffer calcium, and any factor that affects the pH—temperature, viscosity, or ionic strength of the buffer—will change the calcium concentration of the buffers. Measurement of Intracellular Calcium Ions by Flow Cytometry
11. Incubate cells 90 min at 37°C to permit equilibration of cells. The cells are “clamped” in the presence of metabolic inhibitors and ionophores, so that the calcium concentration of the buffer should be the same as that of the cell interior. See
9.8.8 Supplement 2
Current Protocols in Cytometry
1200
Indo-1 ratio (channel number)
1000
800
600
400
200
0 1
10
100
1000
10000
Ca2+ concentration (nM)
Figure 9.8.1 Example of in situ calibration of indo-1 ratio shifts. Human T cells were loaded with indo-1 and electrochemical gradients disabled as described in the Support Protocol. The cells were suspended in a series of calcium buffers ranging from 6 nM to 22 µM and steady-state indo-1 fluorescence ratios were determined. The ratio of the 22-µM sample was off-scale and was above channel 1024.
Chused et al. (1987), Thomas and Delaville (1991), and Negulescu and Machen (1990) for details. A 1- to 2-hr incubation is required to achieve equilibrium.
12. Analyze cells to determine the steady-state fluorescence ratio (for indo-1- or fluo-3 plus Fura Red–loaded cells) or absolute fluorescence (for fluo-3-loaded cells) of cells in each buffer solution. Some dead cells should be apparent, reflecting the cellular poisons.
13. Plot peak ratio channel (for indo-1- or fluo-3 plus Fura Red–loaded cells) or channel (for fluo-3-loaded cells) versus calcium concentration of each buffer. See Figure 9.8.1 for an example and see Background Information for further details. The free calcium concentration of each buffer is dependent on the pH. If the pH is not 7.20, the numbers will differ from those given in the Molecular Probes protocol, and the calcium concentration must be calculated as described above.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Cell loading medium HBSS (APPENDIX 2A) containing: 1 mM calcium 1 mM magnesium 1% (v/v) FBS (heat-inactivated 1 hr at 56°C) or 0.5% (w/v) BSA Store ≤1 month at 4°C
Studies of Cell Function
9.8.9 Current Protocols in Cytometry
Supplement 5
DPBS (Dulbecco’s phosphate-buffered saline) Per liter: 0.20 g KCl (2 mM final) 0.20 g KH2PO4 (1.5 mM final) 8 g NaCl (137 mM final) 2.16 g Na2HPO4⋅7H2O (8 mM final) 20 mM HEPES, pH 7.20 This buffer contains no calcium or magnesium.
Fluo-3 AM (acetoxymethyl ester), 10 mg/ml Dissolve fluo-3 AM ester (mol. wt. 1130; Molecular Probes) at 10 mg/ml in DMSO and store ≤6 months in a desiccator in the dark at −20°C. Fura Red AM (acetoxymethyl ester), 10 mg/ml Dissolve Fura Red AM ester (mol. wt. 945; Molecular Probes) at 10 mg/ml in DMSO and store ≤6 months in a desiccator in the dark at −20°C. Indo-1 AM (pentaacetoxymethyl ester), 2 mg/ml Dissolve indo-1 AM ester (Molecular Probes) at 2 mg/ml in DMSO and store ≤6 months in a desiccator in the dark at −20°C. The molecular weight of indo-1 AM ester is 1009.9 g/mol; a 2 mg/ml stock is almost exactly 2 mM.
Ionomycin, 1 mg/ml Dissolve at 1 mg/ml in DMSO or 100% ethanol. Store ≤1 year at −20°C. Use of DMSO minimizes evaporation of solvent and consequent concentration of ionomycin. Ionomycin is available from Calbiochem.
Poisoned DPBS DPBS (see recipe) containing: 3 µg/ml ionomycin (from 1 mg/ml stock in DMSO; see recipe) 2 µg/ml nigericin (from 10 mg/ml stock in methanol) 10 µM carbonyl cyanide m-chlorophenylhydrazone (CCCP; from 1 mM stock in DMSO) 40 mM 2-deoxyglucose (from 1 M stock in water) 60 mM sodium azide (from 3 M stock in water) Store ≤1 month at 4°C Probenecid, 100 mM Mix quantities of probenecid (Sigma) and water to yield a 100 mM solution. Add 1 M NaOH until probenecid dissolves (pH should be 9 to 10). Store ≤1 month at room temperature. Probenecid is relatively insoluble in aqueous solution unless alkalinized.
COMMENTARY Background Information
Measurement of Intracellular Calcium Ions by Flow Cytometry
In their resting state, eukaryotic cells maintain an internal calcium ion concentration that is far below that of the extracellular environment. Ionized calcium has an important role as a mediator of transmembrane signal transduction, and elevations in intracellular ionized calcium concentration ([Ca2+]i) regulate diverse cellular processes. Thus, measurement of [Ca2+]i
in living cells is of considerable interest to investigators in immunology and cell biology. The mechanism by which the calcium concentration increases in cells involves a series of complex biochemical reactions whose precise details remain speculative. It is not simply the result of opening of channels that permits calcium to be transported down its electrochemical gradient. Binding of an agonist to its specific membrane receptor causes activation of phos-
9.8.10 Supplement 5
Current Protocols in Cytometry
pholipase C, which in many cases requires intervening activation of a kinase or a guanine nucleotide–binding protein. Phospholipase C causes hydrolysis of a membrane phospholipid, phosphatidylinositol 4,5-biphosphate (PIP2), which yields a water-soluble product, inositol 1,4,5-triphosphate (IP3) and a lipid, 1,2-diacylglycerol (DAG). IP3 then causes the release of calcium from intracellular stores, while DAG and calcium are required to activate most isoforms of protein kinase C. Thus, a single agonist can result in production of at least two second messengers, making the polyphosphatidyl inositide pathway a bifurcating system (Berridge, 1993). Calcium is therefore a third messenger that controls numerous cellular processes. Although it appears clear in most cases that the initial elevation of ionized calcium is due to the release of intracellular calcium stores, little is known about the regulation of the influx of calcium from extracellular sources that is necessary to sustain the response. Until early in the last decade, measurement of [Ca2+]i was restricted to large invertebrate cells where the use of microelectrodes was possible. Subsequently, bioluminescent indicators such as aequorin, a calcium-activated photoprotein, have been described, but were limited in their application by the necessity of loading cells by microinjection or other forms of membrane disruption (Blinks et al., 1982; Cobbold and Rink, 1987). More recently, Tsien et al. (1982) developed quin2, making it possible for the first time to measure [Ca2+]i in virtually any population of cells. Unfortunately, quin2 has a relatively low extinction coefficient and quantum yield, and this, in conjunction with the fact that it does not have useful ratioing properties, made detection of calcium responses in single cells impractical. With the invention of fura-2 and indo-1 (Grynkiewicz et al., 1985), representing a second family of dyes, it became possible to measure [Ca2+]i in single cells of almost any type. Fura-2 is best suited for applications involving fluorescence microscopy with digital image analysis, while indo-1, due to its unique fluorescence properties, is best suited for flow cytometry. With indo-1, there is a 6-fold increase in signal when cells change from basal levels of [Ca2+]i to saturating amounts of [Ca2+]i. In addition, the ratioed signal from indo-1 is independent of cell size or brightness. Cells loaded with indo-1 exhibit a 3-fold range in brightness, and after ratioing at the proper wavelengths, the cell-to-cell variation in the signal from resting cells has a coefficient of
variation of only 5% to 10% (Rabinovitch et al., 1986). Thus, with indo-1 it is possible to discriminate responses of small subpopulations of cells within a larger population of nonresponding cells. Flow cytometric cell conjugate assays have also been developed so that the calcium signals that occur during antigen presentation or target cell recognition can be studied using indo-1 (Abe et al., 1992; Alexander et al., 1992; Van Graft et al., 1993). Another advantage of indo-1 is the sensitivity of its response. Results of artificial mixing experiments with Jurkat T leukemia and K562 myeloid leukemia cell lines indicated that subpopulations of cells with variant [Ca2+]i comprising <1% of total cells could be accurately identified (Rabinovitch et al., 1986). Although toxicity of the dye must be considered, indo-1 has been shown not to affect reproductive viability of human T cells or murine B cells (Rabinovitch, 1986; Chused et al., 1987). This allows the sorting and subsequent propagation of lymphoid and nonlymphoid cells based on calcium responses (Goldsmith and Weiss, 1987; Liddle et al., 1992). Low-illumination intensity should be used to minimize photodamage to the cells. For fluo-3 and quin2, [Ca2+]i determination is sensitive to cell size and changes in intracellular dye concentration as well as to [Ca2+]i. In spectrofluorometer-based assays, it is necessary to calibrate at the end of each individual assay by determining the fluorescence intensity of the dye at zero and saturating [Ca2+]i. For nonratiometric calcium probes such as these, the following equation (Tsien et al., 1982) is used: nanomolar [Ca 2+ ]i = Kd ×
( F − Fmin ) ( Fmax − F )
whereKd represents the effective dissociation constant for calcium-bound fluo-3, Fmax represents the maximum fluorescence (ionophoretreated cells, or detergent-permeabilized cells), Fmin is the fluorescence of the probe in the absence of calcium, and F is the fluorescence of the test sample. In contrast, use of the [Ca2+]dependent shift in indo-1 emission wavelength allows the ratio of fluorescence intensities of the dye at the two wavelengths to be used to calculate nanomolar [Ca2+] using the following equation for the fluorescence ratio: nanomolar [Ca 2+ ]i = Kd ×
( R − Rmin )Sf2 ( Rmax − R)Sb2
where Kd is the effective dissociation constant (250 nM at 37°C, pH 7.05); R, Rmin, and Rmax
Studies of Cell Function
9.8.11 Current Protocols in Cytometry
Supplement 2
Measurement of Intracellular Calcium Ions by Flow Cytometry
are the fluorescence intensity ratios of violet/blue fluorescence at resting, zero, and saturating [Ca2+]i, respectively; and Sf2/Sb2 is the ratio of the fluorescence intensity at long (blue or green) wavelengths of the calcium-free and bound dye, respectively (Grynkiewicz et al., 1985). See Thomas and Delaville (1991) for information concerning subtraction of background signals and other details of calibration. The Kd varies as a function of temperature, pH, ionic strength, and viscosity of the cytosol. The Kd of indo-1, fluo-3, and fura-2 are known to vary considerably. For example, the Kd of fura-2 is 135 nM at 20°C, pH 7.05, and 224 nM at 37°C, pH 7.15 (Grynkiewicz et al., 1985). The Kd of fluo-3 is 400 nM at 22°C, and 864 nM at 37°C (Merritt et al., 1990). Thus, considerable caution must be exercised when nonstandard conditions are used. The term Sf2/Sb2 is a constant that varies depending on the interference filters used, and thus will vary from instrument to instrument and with differing configurations on the same instrument. For ratiometric calcium measurements with indo-1, because [Ca2+]i is independent of total intracellular dye concentration and instrumental variation in efficiency of excitation or emission detection, it is not necessary to measure the fluorescence of the dye in the calcium-free and saturated states for each individual assay, as is the case with spectrofluorometric-based assays. In principle, it is sufficient to calibrate the instrument once, and only R is measured for each subsequent analysis. The support protocol describes the calibration of fluorescence ratio to [Ca2+]i, yielding a regression curve that relates indo-1 violet/blue ratio or the fluo-3/Fura Red green/red ratio to [Ca2+]i. Although the protocol does not require access to a spectrofluorometer, preparation of a series of calcium buffers is required, a process now considerably simplified by the commercial availability of the buffers. These buffers must yield accurate and reproducible free calcium concentrations. A computer program is required to calculate [Ca2+]i for EGTA- or BAPTA-based calcium buffers. The free calcium concentration is a function of pH, temperature, and ionic strength, as well as the total concentrations of Mg2+, Ca2+, and EGTA (Blinks et al., 1982). A DOS program for IBMcompatible computers is available from the authors. The viscosity of cell cytosol also affects measurements (Poenie, 1990). For calibration of flow cytometric calcium measurements, the strategy proposed by Chused and coworkers (Chused et al., 1987) is to treat cells
with the calcium ionophore ionomycin and a cocktail of metabolic poisons in order to collapse the calcium gradient to zero, so that [Ca2+]i = [Ca2+]o. With this nondisruptive calibration approach, it is not necessary to determine the constants of the fluorescence ratio equation because the calibration is based upon a regression formula that relates R to ionomycin-treated cells suspended in a series of precisely prepared calcium buffers. A limitation of this approach is that it cannot be used to quantitate calcium concentrations that are less than those found in resting cells because the condition [Ca2+]i = [Ca2+]o can only be met for values above the resting calcium concentration. Surprisingly, in some cell types, it may not be possible to use calcium ionophores to collapse the calcium gradient across the plasma membrane, as marked heterogeneity in the response is observed (Li et al., 1987; Ishida and Chused, 1988). In addition, cell membrane polarity can affect the response of lymphocytes to calcium ionophores; there appear to be marked differences between T and B lymphocytes in this respect (Ishida and Chused, 1988; Gelfand et al., 1989; Gelfand and Cheung, 1990).
Critical Parameters and Troubleshooting General experimental conditions Cell number. The number of cells required per assay depends on the number of experiments planned and the expected proportion of responding cells. Experiments with cells having a homogeneous response need ∼105 cells per 10-min assay to quantify the response of a major population, while 106 cells per assay are needed if quantitation of the response of a minor population is desired. Medium. The choice of cell culture medium can be dictated primarily by the metabolic requirements of the cells, subject only to the presence of millimolar concentrations of calcium (to enable agonist-stimulated calcium influx) and reasonable pH buffering. The use of serum and phenol red does not impair the detection of fluorescence signals in flow cytometry, although these components affect assays run on a spectrofluorometer. Setup. The choice of emission filters is dictated by the spectral characteristics of the shift in indo-1 emission upon binding to calcium. An increase in [Ca2+]i is detected with indo-1 as an increase in the ratio of lower- to higher-wavelength emission. Some commercially available
9.8.12 Supplement 2
Current Protocols in Cytometry
bandpass filters are typically centered on the violet peak emission of the calcium-bound indo-1 dye (410 nm) and on the blue emission peak of calcium-free indo-1 dye (485 nm). These filters include wavelengths nearer the isobestic point that do not exhibit as large a dependence upon calcium binding. A larger dynamic range in the ratio of wavelengths is obtained if blue emission below 485 nm is not collected and the blue emission bandpass is set above 485 nm, or even into green emission, above 500 nm. Similarly, the violet bandpass filter should be chosen to minimize the collection of wavelengths above 400 nm. Thus, given that fluorescence emission intensity is not limiting, improved ratio shifts are obtained with the use of emission filters that are centered away from peak emission. When analyzing shifts in cellular calcium, increases above baseline do occur, but significant decreases are rare. When [Ca2+]i increases, indo-1 violet fluorescence increases and blue fluorescence decreases. In order to maintain the fluorescence shifts on scale, the violet setting is initially biased to the low end of the scale while the blue setting is biased towards the upper end of the scale. Indo-1 is a vital dye, and therefore dead cells can be efficiently excluded from analysis. It is best to set the windows to exclude dead cell fluorescence (which should be rare because the indo-1 loading procedure is nontoxic for lymphocytes) on the violet fluorescence parameter because indo-1 fluorescence shifts during increases in [Ca2+]i. If the blue parameter were used, then the decreasing fluorescence levels of responding cells could enter the dead-cell range and be excluded from analysis. Temperature. It is critical that cell analysis not be done at room temperature because the generation of cellular [Ca2+]i responses is highly temperature dependent. Thus, the sample chamber must be maintained at 37°C, and the time that cells spend in tubing in transit to the interrogation point must be kept minimal (<10 sec); if transit time is longer, the sample tubing must also be warmed. The agonist is introduced into the sample by quickly ceasing flow, removing the sample container, adding agonist, restarting flow, and injecting the new sample into the flow cell. With practice, this procedure can be completed in <15 sec. If the analysis of the onset of more rapid [Ca2+]i transients is required, then it is necessary to use one of the various stop-flow injection methods available.
Probe. The parameters considered above for indo-1 apply to fluo-3 and Fura Red assays as well. The purity and quality of the probe may vary. In particular, the authors have encountered bad lots of indo-1 and fluo-3. Fluo-3 has actually improved with “age,” as the calciumsensitive increase in fluorescence of current batches is now much better than for the original batches. The protocol describing the combined use of fluo-3 and Fura Red permits, for the first time, sensitive ratiometric calcium measurements using visible light excitation. This feature should make this technique useful; however, the investigator should be aware that calibration is especially difficult with this assay, as the two probes may leak out of the cells at different rates, so that the signal of the cells immediately after loading may vary as a consequence of time. As with all new assays, the limitations will only become fully apparent with widespread use. Instrument function Poor cellular response may be due to suboptimal instrument function. To evaluate the instrument, the cells should be stimulated with the calcium ionophore ionomycin, using the amount required to induce a maximal change in the Rmax to R ratio. Ionomycin is protein bound, and the amount required varies depending on the choice of medium. Generally, 1 to 2 µg/ml is sufficient. Because ionomycin is toxic, if too much is added, the cells become permeable and lose fluorescence in both violet and blue regions. The expected response to ionomycin is a loss of blue fluorescence and a gain in violet fluorescence. If the Rmax to R ratio increases by ∼6-fold (depending on the filter set used), the instrument is functioning properly. If the increase is less than expected, obtain and test an independent preparation of cells, such as murine thymocytes or human peripheral blood lymphocytes (aliquots of cryopreserved cells are convenient for this purpose). Note, however, that there appear to be species-dependent differences in toxicity to ionomycin; for instance, murine lymphocytes are more sensitive than human lymphocytes. If these test cells load properly and also respond poorly, check the instrument alignment, the bandpass filters, and the dichroic mirror. The violet or blue signals may be improperly focused, or there may possibly be interference from a second laser. The source of the problem can be pinpointed by separately analyzing the blue and violet signals after iono-
Studies of Cell Function
9.8.13 Current Protocols in Cytometry
Supplement 2
phore treatment. The violet signal should increase ∼3-fold and the blue signal should decrease ∼2-fold. If one signal fails to change appropriately, the cause can be quickly determined by changing the photomultiplier tubes and/or the fiber optic cables. On some instruments it may be difficult to display the ratio of fluorescence, so that increases in [Ca2+]i are depicted as increasing ratio values. In particular, the analog ratio circuits of some Coulter instruments are limited in their range of acceptable inputs. Thus the ratio of blue/violet fluorescence is displayed, resulting in a counterintuitive decrease in ratio shift with increases in [Ca2+]i. Some instruments may have nonlinearity in signal amplification or may introduce errors into the calculation of the indo-1 AM ratio. It is important that no artifactual offset be introduced in the ratio by either analog or digital calculation. Test the calculation by altering the excitation power over a broad range during analysis of loaded cells. A correctly calculated ratio will not show dependence upon excitation intensity. Loading cells with a broad range of indo-1 AM or fluo-3 AM and Fura Red AM concentrations should also result in a constant value for the violet/blue or green/red ratios (Rabinovitch et al., 1986; Novak and Rabinovitch, 1994).
Measurement of Intracellular Calcium Ions by Flow Cytometry
Cellular response If there is a poor cellular response with adequate instrument function, then the problem is with the cells. The cells must be loaded with sufficient probe to be easily detected. This should be checked independently with fluorescence microscopy. If the cells are too dim or bright, or if the dye is compartmentalized, the ability to detect calcium signals will be impaired. Compartmentalization can be minimized by loading cells and storing cells at lower temperatures (Malgaroli et al., 1987); the authors routinely use 30°C rather than 37°C. For unknown reasons, the calcium signaling of B cells, but not of T cells, is particularly sensitive to overloading with indo-1 AM (Rabinovitch et al., 1986; Chused et al., 1987). The cells must be suspended in medium that contains calcium; without it, responses are blunted. In the simultaneous analysis of [Ca2+]i and immunofluorescence, the antibody probe can alter the cellular [Ca2+]i. It is becoming increasingly clear that binding of monoclonal antibodies to cell-surface proteins can alter [Ca2+]i, even when these proteins are not recognized as part of a signal transducing pathway. For exam-
ple, antibody binding to CD4 will reduce CD3mediated [Ca2+]i signals. If the anti-CD4 is cross-linked to the CD3 complex, as with a goat anti-mouse antibody, the CD3 signals are augmented. Therefore, a reciprocal staining strategy using negative selection should be used whenever possible so that the cellular subpopulation of interest is unlabeled while undesired cell subsets are identified by antibody staining. For example, the CD4+ subset in peripheral blood lymphocytes (PBL) may be identified by staining with a combination of CD8, CD20, and CD11b antibodies. Different antibodies that bind to the same molecule can vary greatly in their ability to cause calcium signaling, so the choice of antibody is important. When staining cells with monoclonal antibodies for functional studies, be certain that the antibodies are azide-free, so that metabolic processes are uninhibited. Commercial antibody preparations may require dialysis before use. Several problems may be encountered in loading cells with calcium probes. These include dye concentrations, effects of DMSO and decomposition of the dye, cellular compartmentalization, incomplete hydrolysis of the dye esters, quenching by heavy metals, and secretion or leakage of the dye. Dye concentration. The amount of dye ester required for loading depends on the cell concentration, the number and type of cells to be loaded, and the protein content of the loading buffer. Cells loaded with the correct amount of dye are those loaded with the lowest concentration that causes detectable cellular fluorescence with the photomultiplier tube gain settings in midrange. On a properly adjusted flow cytometer with efficient fluorescence detection, this corresponds, for example, to an intracellular indo-1 AM concentration of 5 to 20 µM; see Thomas and Delaville (1991) for details on the determination of intracellular probe concentration. In practice, optimal loading conditions are determined by loading aliquots of cells with several different concentrations of dye. After a 30-min incubation, 10 µl of each cell suspension is removed and added to 0.5 ml HBSS. The fluorescence intensity of this mixture is measured on the flow cytometer, and cell batches that are too dim or too bright are discarded. It is critical that cells not be overloaded. A brighter signal is not always better! DMSO. Acetoxymethyl esters of indicator dyes are usually dissolved in DMSO as a 1 to 5 mM stock solution. It is important that the
9.8.14 Supplement 2
Current Protocols in Cytometry
final concentration of DMSO not exceed 0.5% during the loading process. Excess DMSO has a number of biologic effects, including effects on calcium homeostasis (Lupu-Meiri et al., 1993). Dyes should be reconstituted only with DMSO that has been dried over 4-Å molecular sieves. In the authors’ experience, dyes will often decompose after 6 weeks of storage if reconstituted with DMSO that was not dried prior to use. To avoid these problems, either reconstitute dye with anhydrous DMSO or use nondried DMSO and discard reconstituted dye at 6 weeks. Alternatively, purchase small aliquots and reconstitute on a weekly basis with nondried DMSO. (Molecular Probes sells 50µg aliquots of a lyophilized oil that is stable for >1 year.) Compartmentalization. Analysis of [Ca2+]i using indicator dyes is predicated upon achieving uniform distribution of the dye within the cytoplasm. In several cell types, the related dye fura-2 has been reported to be compartmentalized within organelles (Di Virgilio et al., 1990). In bovine aortic endothelial cells, fura-2 has been reported to be localized to mitochondria; however, under those conditions, indo-1 remained diffusely cytoplasmic (Steinberg et al., 1987). Neutrophils, monocytes, and some cell lines (rather than primary cells) are more susceptible to compartmentalization. In general, loading can be improved by incubation at 30°C because compartmentalization is enhanced by prolonged incubation of cells at 37°C. The cellular distribution of the probe should be examined microscopically to confirm the expected loading behavior of the dye in each new application. In experiments where it is necessary to incubate loaded cells at 37°C with various treatments, control cells incubated under identical conditions should be included. If, for a particular cell type loaded with a probe, the magnitude of change between R and Rmax is in good agreement with the values predicted from spectral curves of the dye in a cell-free buffer, it is unlikely that the dye is in a compartment inaccessible to cytoplasmic Ca2+, in a form unresponsive to [Ca2+]i (e.g., still esterified), or in a cytoplasmic environment where the spectral properties of the dye are altered (e.g., low pH). With regard to the second condition, it has been proposed that because indo-1 fluorescence (but not that of the indo-1 AM ester) is quenched in the presence of millimolar concentrations of Mn2+, Mn2+ in the presence of ionomycin can be used as a further test of complete hydrolysis of the indo-1 AM ester
within cells (Luckhoff, 1986). This approach does not work for fluo-3. Incomplete hydrolysis. Within some cell types, hydrolysis of the dyes to their charged forms may be incomplete (Luckhoff, 1986). The fluorescence of the ester has little spectral dependence upon changes in Ca2+, so the presence of this dye form could lead to false estimates of [Ca2+]i. Again, results of calibration experiments are helpful in excluding this possibility. In some circumstances, the use of pluronic F-127, a nonionic, high-molecularweight surfactant, may aid in loading indo-1 AM into cells that are otherwise difficult to load (see Alternate Protocol 3). Several approaches have been suggested for calibration of intracellular indo-1 signals in cases of difficult cellular loading (Scanlon et al., 1987; Owen and Shuler, 1989). Heavy metal ions. Heavy metals such as zinc may be concentrated in some cell lines, and these metals may cause spurious signals to be reported by the calcium indicator. The use of the membrane-permeant form of the heavy metal chelator diethylenetriaminepentaacetic acid (TPEN) for cell lines that contain increased amounts of heavy metals has been described (Arslan et al., 1985). TPEN binds heavy metals but not calcium or magnesium. Leakage or secretion. Probenecid and sulfinpyrazone, blockers of organic anion transport, may be useful in cells that actively secrete calcium-sensitive dyes or for minimizing cellto-cell variation in dye content (Di Virgilio et al., 1990; Vandenberghe and Ceuppens, 1990; Baus et al., 1994). Cell viability. Under typical conditions, the baseline ratio for indo-1 or fluo-3/Fura Red should show little variation (<3%) from sample to sample. Some cell lines may have altered mean values of resting [Ca2+]i; this is often due to a subpopulation of cells with elevated [Ca2+]i. It may result from impaired cell viability or may be due to the presence of cells in certain phases of the cell cycle. When the baseline is not stable from sample to sample, be certain that the cells have been equilibrated ≥5 min at 37°C before analysis, the agonist from the previous experiment has been completely removed from the sample line, and the cells are healthy. Calibration difficulties An approach for standardizing flow cytometric calcium measurements is presented in this unit (see Support Protocol). It is critical that the cocktail of cellular poisons be titrated to
Studies of Cell Function
9.8.15 Current Protocols in Cytometry
Supplement 2
Measurement of Intracellular Calcium Ions by Flow Cytometry
abolish the ability of the cell to defend its cellular calcium concentration. Some dead cells should be apparent during the analysis, particularly in the buffers containing higher calcium. Load the cells in low-calcium buffer; take care not to expose the cells to medium containing calcium once they have been disabled by poisons, as it is not possible to return the cells to low calcium levels with the buffers. This is presumably due to the presence of highaffinity calcium binding sites within the cells. There are many potential pitfalls in calcium calibration (Merritt et al., 1990; Roe et al., 1990; Williams and Fay, 1990; Thomas and Delaville, 1991; Negulescu and Machen, 1990). For details concerning the use of calcium buffers, see Haugland (1992). It is worth noting that EGTA, although displaying an inconvenient pH-dependent calcium affinity, is stable on prolonged storage, and BAPTA, although useful for its pH independence, can decompose on storage over a few months (Oiki et al., 1994). The behavior of calcium probes in different cell types can vary (Arslan et al., 1985; Baus et al., 1994). In addition, the choice of pH buffers can affect calcium homeostasis (Ganz et al., 1990). The affinity of calcium chelators and probes increases with decreasing ionic strength (Miller and Smith, 1984; Oiki et al., 1994, and references therein). Caged calcium chelators have been developed that release bound calcium after photo stimulation. These compounds present intriguing experimental possibilities, and may affect calibration with indo-1 (Hadley et al., 1993). Finally, biologic differences in responses to agonists have been observed in cells, depending on cell density and the state of attachment to substrates (Szollosi et al., 1991). Thus, extreme care must be used not to extrapolate results from one cell type or experimental condition to another (Malgaroli et al., 1987). Alternative calibration approaches have been proposed. Determining Rmax, Rmin, and Sf2/Sb2 for calibration of indo-1 ratios allows direct determination of the constants in the equation that relates indo-1 fluorescence to [Ca2+]i (Rabinovitch et al., 1986). A novel method employing measurements of spectral shifts has also been proposed for calcium calibration (Kachel et al., 1990). Regardless of the choice of calibration technique, it is strongly recommended that for each batch of cells analyzed, ordinary quality control should include a determination of the value of Rmax/R as described in the Basic Protocol.
Anticipated Results The most elementary form of display of the indo-1 fluorescence on the flow cytometer is as a bivariate plot of violet versus blue signals. In this case, the increase in ratio seen with increased [Ca2+]i will be observed as a rotation around the axis through the origin. This method of analysis is cumbersome, however, and does not account for time as a primary variable. Fortunately, all commercial flow cytometers have some provision for a direct calculation of the fluorescence ratio of violet/blue itself, either by analog circuitry or by digital computation. Plotted as a histogram of the ratio values, unperturbed cell populations show narrow distributions of ratio, even when cellular loading is very heterogeneous, and thus coefficients of variation of indo-1 or fluo-3 and Fura Red ratios of <10% are not uncommon. The effects of perturbation of [Ca2+]i by agonists can be noted by sequential analysis and histogram storage, with comparisons in the ratio distributions being made during subsequent data analysis. A more informative and elegant display is obtained by a bivariate plot of ratio versus time. Bivariate data can be viewed in real time or subsequently plotted as a dot plot in which greater intensity indicates larger numbers of cells (see Fig. 9.8.2A). For quantitative comparison between different cell treatments and quicker grasp of data trends, it is often useful to distill the bivariate data into a single descriptive parameter versus time. Calculation of the mean y-axis value for each x-axis time interval (see Fig. 9.8.2B) allows presentation of the data as mean ratio versus time, or after calibration, as the mean [Ca2+]i of the population versus time. Although this presentation yields much of the information of interest, data relating to heterogeneity of the [Ca2+]i response is lost. For example, note the complexity of the response of lymphocytes after stimulation through the T cell receptor/CD3 complex. More than 95% of the cells respond within 10 min after stimulation. The response can be divided into two phases: an initial rapid response of high magnitude that occurs in ∼15% of cells, followed by a sustained response that occurs in nearly all cells. Note that the time of the peak mean response occurs at ∼3.5 min (see Fig. 9.8.2B) while the time of the peak response occurs earlier in a subset of cells, at 3 min (see Fig. 9.8.2A). A linear amplification of indo-1 blue and violet fluorescence intensity makes determination of the ratio straightforward. However, if cellular indo-1 loading is extremely heteroge-
9.8.16 Supplement 2
Current Protocols in Cytometry
100
A
Indo -1 ratio
80
60
40
20
Mea n ratio
0 40 35 30 25 20 15 10 5 0
B
0
2
6
4
8
10
Time
Figure 9.8.2 Effects of T cell receptor stimulation on CD4 cell ionized calcium concentration ([Ca2+]i). Peripheral blood lymphocytes were loaded with indo-1 AM and stained with PE-anti-CD8. The cells were maintained at 37°C; after obtaining a baseline for ∼1 min, anti-CD3 MAb was added during the gap in analysis. (A) The indo-1 ratio of 395 nm/500 nm fluorescence emission was calculated and the value for each cell displayed on the y axis versus time on the x axis. The results are displayed as a dot plot on a 100 × 100 pixel grid, where the number of cells per pixel is represented by increasing shades of gray. Changes in [Ca2+]i in the CD4 subset of T cells are depicted by setting electronic gates on the indo-1 fluorescence derived from the PE-negative cells. (B) The mean response of the data from panel A plotted versus time.
neous, violet and blue emission intensities should be converted to log values in order to observe a broader range of cellular fluorescence. In this case, the logarithm of the ratio is calculated by subtracting the log of the blue signal from the log of the violet signal (Rabinovitch et al., 1986).
Time Considerations
Allow ∼1 hr for instrument alignment and warm-up. Cellular loading takes ∼30 min. Experimental time is labor-intensive and varies widely. For many kinetic experiments, the cells will be analyzed continuously for a period of 5
or 10 min. In 3 hr, ∼15 to 20 samples can be tested. Carefully evaluate the results “on the fly,” as the results from early samples may suggest subsequent conditions to test. Allow 1 hr for data reduction and analysis.
Literature Cited Abe, R., Ishida, Y., Yui, K., Katsumata, M., and Chused, T.M. 1992. T cell receptor–mediated recognition of self-ligand induces signaling in immature thymocytes before negative selection. J. Exp. Med. 176:459-468. Alexander, R.B., Bolton, E.S., Koenig, S., Jones, G.M., Topalian, S.L., June, C.H., and Rosenberg, S.A. 1992. Detection of antigen-specific T lym-
Studies of Cell Function
9.8.17 Current Protocols in Cytometry
Supplement 2
phocytes by determination of intracellular calcium concentration using flow cytometry. J. Immunol. Methods 148:131-141. Allbritton, N.L. and Meyer, T. 1993. Localized calcium spikes and propagating calcium waves. Cell Calcium 14:691-697. Arslan, P., Di Virgilio, F., Beltrame, M., Tsien, R.Y., and Pozzan, T. 1985. Cytosolic Ca2+ homeostasis in Ehrlich and Yoshida carcinomas. A new, membrane-permeant chelator of heavy metals reveals that these ascites tumor cell lines have normal cytosolic free Ca2+. J. Biol. Chem. 260:2719-2725. Baus, E., Urbain, J., Leo, O., and Andris, F. 1994. Flow cytometric measurement of calcium influx in murine T cell hybrids using Fluo-3 and an organic-anion transport inhibitor. J. Immunol. Methods 173:41-47. Berridge, M.J. 1993. Inositol trisphosphate and calcium signalling. Nature 361:315-325. Blinks, J.R., Wier, W.G., Hess, P., and Prendergast, F.G. 1982. Measurement of Ca2+ concentrations in living cells. Prog. Biophys. Mol. Biol. 40:1114. Chused, T.M., Wilson, H.A., Greenblatt, D., Ishida, Y.L., Edison, J., Tsien, R.Y., and Finkelman, F.D. 1987. Flow cytometric analysis of murine splenic B lymphocyte cytosolic free calcium response to anti-IgM and anti-IgD. Cytometry 8:396-404. Cobbold, P.H. and Rink, T.J. 1987. Fluorescence and bioluminescence measurement of cytoplasmic free calcium. Biochem. J. 248:313-328. Di Virgilio, F., Steinberg, T.H., and Silverstein, S.C. 1990. Inhibition of Fura-2 sequestration and secretion with organic anion transport blockers. Cell Calcium 11:57-62. Ganz, M.B., Rasmussen, J., Bollag, W.B., and Rasmussen, H. 1990. Effect of buffer systems and pHi on the measurement of [Ca2+]i with fura 2. FASEB J. 4:1638-1644. Gelfand, E.W. and Cheung, R.K. 1990. Dissociation of unidirectional influx of external Ca2+ and release from internal stores in activated human T lymphocytes. Eur. J. Immunol. 20:1237-1241. Gelfand, E.W., MacDougall, S.L., Cheung, R.K., and Grinstein, S. 1989. Independent regulation of Ca2+ entry and release from internal stores in activated B cells. J. Exp. Med. 170:315-320. Goldsmith, M.A. and Weiss, A. 1987. Isolation and characterization of a T-lymphocyte somatic mutant with altered signal transduction by the antigen receptor. Proc. Natl. Acad. Sci. U.S.A. 84:6879-6883. Grynkiewicz, G., Poenie, M., and Tsien, R.Y. 1985. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J. Biol. Chem. 260:3440-3450. Measurement of Intracellular Calcium Ions by Flow Cytometry
Hadley, R.W., Kirby, M.S., Lederer, W.J., and Kao, J.P. 1993. Does the use of DM-nitrophen, nitr-5, or diazo-2 interfere with the measurement of Indo-1 fluorescence? Biophys. J. 65:2537-2546.
Haugland, R.P. 1992. Calcium indicators, chelators and ionophores. In Handbook of Fluorescent Probes and Research Chemicals (K.D. Larison, ed.) pp. 113-128. Molecular Probes, Inc., Eugene, Ore. Ishida, Y. and Chused, T.M. 1988. Heterogeneity of lymphocyte calcium metabolism is caused by T cell-specific calcium-sensitive potassium channel and sensitivity of the calcium ATPase pump to membrane potential. J. Exp. Med. 168:839-852. June, C.H., Abe, R., and Rabinovitch, P.S. 1995. Measurement of intracellular ions by flow cytometry. In Current Protocols in Immunology (J.E. Cooligan, A.M. Kruisbeek, D.H. Margulies, E.M. Shevach, and Strober, W. eds.) pp. 5.5.6-5.5.7. John Wiley & Sons, New York. Kachel, V., Kempski, O., Peters, J., and Schodel, F. 1990. A method for calibration of flow cytometric wavelength shift fluorescence measurements. Cytometry 11:913-915. Lanza, F., Beretz, A., Kubina, M., and Cazenave, J.P. 1987. Increased aggregation and secretion responses of human platelets when loaded with the calcium fluorescent probes quin2 and fura-2. Thromb. Haemost. 58:737-743. Li, Q., Altschuld, R.A., and Stokes, B.T. 1987. Quantitation of intracellular free calcium in single adult cardiomyocytes by fura-2 fluorescence microscopy: Calibration of fura-2 ratios. Biochem. Biophys. Res. Commun. 147:120-126. Liddle, R.A., Misukonis, M.A., Pacy, L., and Balber, A.E. 1992. Cholecystokinin cells purified by fluorescence-activated cell sorting respond to monitor peptide with an increase in intracellular calcium. Proc. Natl. Acad. Sci. U.S.A. 89:51475151. Lipp, P. and Niggli, E. 1993. Ratiometric confocal Ca2+-measurements with visible wavelength indicators in isolated cardiac myocytes. Cell Calcium 14:359-372. Luckhoff, A. 1986. Measuring cytosolic free calcium concentration in endothelial cells with indo-1: The pitfall of using the ratio of two fluorescence intensities recorded at different wavelengths. Cell Calcium 7:233-248. Lupu-Meiri, M., Beit-Or, A., Christensen, S.B., and Oron, Y. 1993. Calcium entry in Xenopus oocytes: Effects of inositol trisphosphate, thapsigargin and DMSO. Cell Calcium 14:101-110. Malgaroli, A., Milani, D., Meldolesi, J., and Pozzan, T. 1987. Fura-2 measurement of cytosolic free Ca2+ in monolayers and suspensions of various types of animal cells. J. Cell Biol. 105:21452155. Merritt, J.E., McCarthy, S.A., Davies, M.P., and Moores, K.E. 1990. Use of fluo-3 to measure cytosolic Ca2+ in platelets and neutrophils. Loading cells with the dye, calibration of traces, measurements in the presence of plasma, and buffering of cytosolic Ca2+. Biochem. J. 269:513-519.
9.8.18 Supplement 2
Current Protocols in Cytometry
Miller, D.J. and Smith, G.L. 1984. EGTA purity and the buffering of calcium ions in physiological solutions. Am. J. Physiol. 246:C160-C166. Minta, A., Kao, J.P., and Tsien, R.Y. 1989. Fluorescent indicators for cytosolic calcium based on rhodamine and fluorescein chromophores. J. Biol. Chem. 264:8171-8178. Negulescu, P.A. and Machen, T.E. 1990. Intracellular ion activities and membrane transport in parietal cells measured with fluorescent dyes. Methods Enzymol. 192:38-81. Novak, E.J. and Rabinovitch, P.S. 1994. Improved sensitivity in intracellular ionized calcium measurement using fluo-3/fura red fluorescence ratios. Cytometry 17:135. Oiki, S., Yamamoto, T., and Okada, Y. 1994. Apparent stability constants and purity of Ca-chelating agents evaluated using Ca-selective electrodes by the double-log optimization method. Cell Calcium 15:209-216. Osipchuk, Y. and Cahalan, M. 1992. Cell-to-cell spread of calcium signals mediated by ATP receptors in mast cells. Nature 359:241-244. Owen, C.S. and Shuler, R.L. 1989. Spectral evidence for noncalcium interactions of intracellular Indo-1. Biochem. Biophys. Res. Commun. 163:328-333. Pethig, R., Kuhn, M., Payne, R., Adler, E., Chen, T.H., and Jaffe, L.F. 1989. On the dissociation constants of BAPTA-type calcium buffers. Cell Calcium 10:491-498. Poenie, M. 1990. Alteration of intracellular fura-2 fluorescence by viscosity: A simple correction. Cell Calcium 11:85-91. Poenie, M., Alderton, J., Steinhardt, R., and Tsien, R. 1986. Calcium rises abruptly and briefly throughout the cell at the onset of anaphase. Science 233:886-889. Rabinovitch, P.S., June, C.H., Grossmann, A., and Ledbetter, J.A. 1986. Heterogeneity among T cells in intracellular free calcium responses after mitogen stimulation with PHA or anti-CD3. Simultaneous use of indo-1 and immunofluorescence with flow cytometry. J. Immunol. 137:952961. Rabinovitch, P.S., June, C.H., and Kavanagh, T.J. 1992. Measurements of cell physiology: Ionized calcium, pH and glutathione. In Clinical Flow Cytometry: Principles and Applications (K.D. Bauer, R.E. Duque, and T.V. Shankey, eds.) pp. 505-534. Williams and Wilkins, Baltimore.
cator dyes Fluo-3 and Fura-red. Cell Calcium 15:341-348. Steinberg, S.F., Bilezikian, J.P., and Al-Awqati, Q. 1987. Fura-2 fluorescence is localized to mitochondria in endothelial cells. Am. J. Physiol. 253:C744-C747. Szollosi, J., Feuerstein, B.G., Hyun, W.C., Das, M.K., and Marton, L.J. 1991. Attachment of A172 human glioblastoma cells affects calcium signalling: A comparison of image cytometry, flow cytometry, and spectrofluorometry. Cytometry 12:707-716. Thomas, A.P. and Delaville, F. 1991. The use of fluorescent indicators for measurements of cytosolic-free calcium concentration in cell populations and single cells. In Cellular Calcium: A Practical Approach (J.G. McCormack and P.H. Cobbold, eds.) pp. 1-54. Oxford University Press, New York. Tsien, R.Y. 1980. New calcium indicators and buffers with high selectivity against magnesium and protons: Design, synthesis, and properties of prototype structures. Biochemistry 19:23962404. Tsien, R.Y., Pozzan, T., and Rink, T.J. 1982. Calcium homeostasis in intact lymphocytes: Cytoplasmic free calcium monitored with a new, intracellularly trapped fluorescent indicator. J. Cell Biol. 94:325-334. Vandenberghe, P.A. and Ceuppens, J.L. 1990. Flow cytometric measurement of cytoplasmic free calcium in human peripheral blood T lymphocytes with fluo-3, a new fluorescent calcium indicator. J. Immunol. Methods 127:197-204. Van Graft, M., Kraan, Y.M., Segers, I.M., Radosevic, K., De Grooth, B.G., and Greve, J. 1993. Flow cytometric measurement of [Ca2+]i and pHi in conjugated natural killer cells and K562 target cells during the cytotoxic process. Cytometry 14:257-264. Williams, D.A. and Fay, F.S. 1990. Intracellular calibration of the fluorescent calcium indicator Fura-2. Cell Calcium 11:75-83.
Key References Haugland, 1992. See above. Provides a wealth of information on calcium probes, calcium buffers, and calcium ionophores. Negulescu and Machen, 1990. See above. Thomas and Delaville, 1991. See above.
Roe, M.W., Lemasters, J.J., and Herman, B. 1990. Assessment of fura-2 for measurements of cytosolic free calcium. Cell Calcium 11:63-72.
Two excellent overviews of calibration strategies and methods.
Scanlon, M., Williams, D.A., and Fay, F.S. 1987. A Ca2+-insensitive form of fura-2 associated with polymorphonuclear leukocytes. Assessment and accurate Ca2+ measurement. J. Biol. Chem. 262:6308-6312.
Contributed by Carl H. June and Ryo Abe Naval Medical Research Institute Bethesda, Maryland
Schild, D., Jung, A., and Schultens, H.A. 1994. Localization of calcium entry through calcium channels in olfactory receptor neurones using a laser scanning microscope and the calcium indi-
Peter S. Rabinovitch University of Washington Seattle, Washington
Studies of Cell Function
9.8.19 Current Protocols in Cytometry
Supplement 2
Intracellular Cytokines
UNIT 9.9
Cytokines are very important in the regulation of many cellular systems. The usual methods for measuring cytokines are based on the detection of these molecules following release by the cell into the environment. Traditionally, cytokine assays were biological assays using cytokine-sensitive cell lines, enzyme-linked immunosorbant assays (ELISAs), or radioimmunoassays. Although all of these are sensitive and quantitative, they are bulk assays and measure the average amount of cytokines/cell number/unit volume. Little information can be obtained about individual cells using such techniques. Often, it is useful or necessary to determine the number or percentage of cells actually producing a specific cytokine, or to know the phenotype of the cell producing a cytokine, or to identify a subset or cell type based on the type of cytokines it produces. Intracellular detection of cytokines by flow cytometry following specific staining of cytokines with antibodies allows for maximal information about the cells producing the cytokine(s) and at least semiquantitation or relative quantitation. The amount of information that can be obtained is essentially limited only by the number of fluorescence parameters that can be detected. The principal protocol described here was developed for detecting cytokines produced by T cells (see Basic Protocol). It not only provides information on the type of cytokines being produced but also helps to phenotypically identify the specific cells producing them. In the case of T cells, CD4+ or CD8+ cells can be subdivided into TH1/TH2 helper cells or TC1/TC2 suppressor cells, respectively. This procedure can be useful in examining the response of cells to a variety of agonists, the immune function in various disease states, and the level of lymphocyte activation or suppression. In addition, it can be used to detect a variety of cytokines in many cell types, although the methodology must be carefully evaluated for each cytokine or cell type of interest. Support protocols are included for the stimulation of T cells with allogeneic cells (see Support Protocol 1), and for the performance of controls for labeling specificity (see Support Protocol 2). MEASUREMENT OF INTRACELLULAR INTERFERON GAMMA AND INTERLEUKIN 4 IN T LYMPHOCYTES
BASIC PROTOCOL
This protocol measures interferon gamma (IFN-γ) and interleukin 4 (IL-4) in T lymphocytes. Additionally, cell-surface markers can be labeled, if desired. The procedure has been successfully used with human and rat whole blood and with rat and mouse spleen cells, and is routinely performed using four antibody reagents labeled with different fluorochromes. Although it is recommended as a general procedure for many cells, the optimal conditions must be empirically determined for each cell type. Because of the wide availability of fluorochromes with different excitation and emission wavelengths, almost any flow cytometer is capable of performing these assays. This protocol describes cytokine labeling procedures for use with flow cytometry. The analysis of intracellular cytokines by flow cytometry is no different from any other analysis of similar cell types in which an equivalent number of fluorochromes are used. See UNIT 6.2 for details on the proper analysis of cells labeled with multiple antibodies. Materials Rat, mouse, or human whole blood or rat or mouse spleen cells (Kruisbeek, 1993) Heparin-containing Vacutainer tubes (UNIT 9.7) or syringe containing sodium heparin at 20 U/ml blood Cell culture medium (see recipe) 10 µg/ml PMA (see recipe) Contributed by George F. Babcock Current Protocols in Cytometry (2004) 9.9.1-9.9.11 Copyright ©2004 by John Wiley & Sons, Inc.
Studies of Cell Function
9.9.1 Supplement 28
250 µg/ml A23187 (see recipe) 1 mM monensin (see recipe) or 100 µg/ml brefeldin A (see recipe) Wash buffer (see recipe), ice cold 20 µg/ml murine gamma globulin (see recipe) ice cold Fluorochrome-labeled antibodies (see recipe; anti-cytokine, anti–cellular antigen, and/or isotypic controls), or kit for intracellular cytokine staining (see recipe) 1% and 4% buffered fixative (see recipe) Permeabilizing solution (one of the following): 0.1% saponin (see recipe), FACS permeabilizing solution (Becton Dickinson Immunocytometry), Fix and Perm (Caltag Labs), Cytofix/Cytoperm (PharMingen), Cell Permeabilization Kit (Harlan Bioproducts for Science), Cytoperm (Serotech), or Permeafix (Ortho) 12 × 75–mm polystyrene or polypropylene snap-capped tubes Centrifuge (Beckman GS-6R with swinging bucket rotor, or equivalent) CO2 incubator at 37°C Additional reagents and equipment for stimulation with allogeneic cells (see Support Protocol 1) and for flow cytometry (UNIT 6.2) Prepare and stimulate samples 1a. For blood samples: Collect blood in heparin-containing Vacutainer tubes (UNIT 9.7) or in syringe containing 20 U sodium heparin/ml blood. 1b. For spleen samples: Prepare spleen cells and suspend at a final concentration of 107 cells/ml. See Kruisbeck (1993) for details of spleen cell preparation.
2. Divide into 100-µl aliquots in 12 × 75–mm polystyrene or polypropylene tubes. Add 100 µl cell culture medium to each tube. The tubes are divided into two or three groups. The negative control (group A) is left untreated. The positive sample (group B) is stimulated with PMA and A23187 (see step 3). If desired, a third group (C) is stimulated with antigen (allogeneic cells; see Support Protocol 1). In this case, group C is the positive sample and group B acts as the positive control. The final number of tubes to be prepared is determined by the number of cytokines/antigens and controls to be examined.
3. Optional: Stimulate cells in group C tubes by addition of allogeneic cells (see Support Protocol 1). 4. Add 2.5 µl of 10 µg/ml PMA and 1.0 µl of 250 µg/ml A23187 to the tubes in group B and mix. Alternatively, add 1 ìl of 1 mg/ml ionomycin in place of A23187.
5. Add 3 µl of 1 mM monensin or 10 µl of 100 µg/ml brefeldin A to the tubes in groups A, B, and C. Mix well. This is an important step because these agents prevent release of the cytokines from the cells. Although some investigators have a strong preference between these compounds, either can be used in this protocol. Monensin is suggested first because it is less expensive.
6. Incubate all tubes 4 hr (human cells) or 5 hr (rat cells) in 5% CO2 at 37°C. The optimal time must be determined for each cytokine with each cell type. Optimal times range from 1 to 12 hr. Intracellular Cytokines
7. Add 3 ml ice-cold wash buffer and centrifuge cells 5 min at 250 × g (∼1200 rpm in a GS-6R with swinging bucket rotor), 4°C. Decant supernatant.
9.9.2 Supplement 28
Current Protocols in Cytometry
Label samples for cell-surface antigens 8. Gently resuspend cell pellet in 0.5 ml of 20 µg/ml ice-cold murine gamma globulin. Incubate cells 15 min at 4°C, and gently agitate the cell suspension. Do not wash; leave the gamma globulin in the samples. Steps 8 through 10 can be omitted if cell-surface antibody staining is not required.
9. Add the appropriate concentration of fluorochrome-labeled cell-surface antibodies or isotypic controls to the appropriate tubes. Mix tubes and incubate 30 min in the dark at 4°C. Each antibody should be titered before use, but a concentration of 1 to 2 ìg of antibody/sample is usually sufficient.
10. Add 3 ml ice-cold wash buffer. Centrifuge cells 5 min at 250 × g, 4°C, and decant supernatant. Repeat. 11. Add 500 µl of 4% buffered fixative to each tube and incubate 20 min in the dark at 4°C. Do not incubate longer than 12 hr.
12. Add 3 ml ice-cold wash buffer. Centrifuge cells 5 min at 250 × g, 4°C, and decant supernatant. Perform cytokine labeling 13. Add 1.0 ml permeabilizing solution and gently resuspend the pellet. Incubate the tubes 12 min in the dark at room temperature, mixing several times during the incubation period. This is intended to be a one-step red cell lysis/permeabilization process. However, solutions such as FACS permeabilizing solution require the use of a separate lysis solution before the fixation step. If this or similar permeabilization solutions are used, follow manufacturer’s instructions.
14. Add 3 ml ice-cold permeabilizing solution. Centrifuge cells 5 min at 250 × g, 4°C, and decant supernatant. Repeat. 15. Resuspend the pellet in 0.25 ml of 20 µg/ml ice-cold murine gamma globulin plus 0.25 ml permeabilizing solution, and incubate 30 min in the dark, 4°C. 16. Add anti-cytokine antibodies or isotypic controls to the appropriate tubes and incubate 30 min at 4°C in the dark. Each antibody should be titered before use, but a concentration of 1 to 2 ìg antibody/sample is usually sufficient. It is recommended that separate controls be run with unlabeled antibody and with an excess amount of the appropriate purified cytokine (see Support Protocol 2).
17. Add 3 ml ice-cold wash buffer. Centrifuge cells 5 min at 250 × g, 4°C, and decant supernatant. Repeat wash procedure two more times. 18. Resuspend cells in 0.5 ml of 1% buffered fixative and store at 4°C in the dark until assayed. The cells are now ready to be assayed by flow cytometry (UNIT 6.2). It is also prudent, at this point, to check cell labeling by fluorescence microscopy (see Chapter 2).
Studies of Cell Function
9.9.3 Current Protocols in Cytometry
Supplement 28
SUPPORT PROTOCOL 1
STIMULATION OF T CELLS WITH ALLOANTIGEN When antigen is used as the stimulator, different investigators have reported cytokine responses that range from undetectable above background (without PMA/ionophore) to strong. Relatively strong responses to allogeneic cells have been observed. This protocol for stimulation of T cells with allogeneic cells is presented as a model for antigen-driven stimulation. Allogeneic lymphocytes can be obtained from a specific (single) donor of interest, or a pool of lymphocytes can be used. A pool from 20 different individuals provides sufficient diversity for most applications. Additional Materials (also see Basic Protocol) Allogeneic lymphocytes 500 µg/ml mitomycin C (optional; prepare in DPBS and store up to 2 weeks in the dark at 4°C) Dulbecco’s phosphate-buffered saline (DPBS, e.g. Life Technologies; prepare according to manufacturer’s instructions), ice cold 5% CO2/95% air incubator at 37°C 1. Suspend allogeneic lymphocytes (stimulator cells) in cell culture medium at 3 × 106 cells/ml for γ-irradiation (step 2a), or at 3 × 107 cells/ml for mitomycin C treatment (step 2b). 2a. For γ-irradiation: Inactivate stimulator cells using 30 Gy γ-irradiation. This step is necessary to prevent activation and division of the stimulator cells. Failure to inactivate stimulator cells will result in an artificially elevated cytokine response.
2b. For mitomycin C treatment: Inactivate stimulator cells by mixing 50 µl of 500 µg/ml mitomycin C per 1 ml cells (final 25 µg/ml) and incubating 45 min at 37°C. Add 3 ml ice-cold DPBS and centrifuge 5 min at 250 × g, 4°C. Decant supernatant. Repeat wash twice and resuspend final pellet in cell culture medium at 3 × 106 cells/ml. This step is necessary to prevent activation and division of the stimulator cells. Failure to inactivate stimulator cells will result in an artificially elevated cytokine response.
3. Add 100 µl of stimulator cells to 100 µl of responder cells (group C; see Basic Protocol, steps 2 and 3). 4. Mix cells and incubate 15 to 18 hr in a 5% CO2/95% air atmosphere at 37°C. The cells are now ready for the remainder of the labeling procedure (see Basic Protocol, steps 5 through 18). SUPPORT PROTOCOL 2
TESTING SPECIFICITY OF CYTOKINE STAINING It is important to determine whether the anti-cytokine reactivity obtained is actually specific for the cytokine of interest or represents nonspecific binding. Isotype controls should always be run in parallel. At regular intervals (e.g, when a new test is set up, when the source of cells changes, or when a specific antibody is changed) it is suggested that blocking controls be performed with excess cytokine and with excess unlabeled antibody. Materials (also see Basic Protocol) Purified cytokine Unconjugated anti-cytokine antibody
Intracellular Cytokines
9.9.4 Supplement 28
Current Protocols in Cytometry
Perform blocking control with excess cytokine 1a. Set up separate 12 × 75–mm tubes for the controls, and follow the Basic Protocol through the second addition of murine gamma globulin (see Basic Protocol, steps 1 through 15). 2a. In another clean set of 12 × 75–mm tubes, mix the predetermined amount of each fluorochrome-labeled anti-cytokine antibody (see Basic Protocol, step 16) with an excess amount of the specific purified cytokine. Incubate the antibody/cytokine mixture 30 min in the dark at 4°C. This amount must be determined for each cytokine, but several ìg are usually sufficient.
3a. Add the antibody/cytokine mixture to the cells (see Basic Protocol, step 16) and continue with the rest of the procedure (see Basic Protocol, steps 17 and 18). Any binding detected upon analysis can be presumed to be nonspecific and should be considered background.
Perform blocking control with excess unlabeled antibody 1b. Set up separate 12 × 75–mm tubes for the controls, and follow the Basic Protocol through the second addition of murine gamma globulin (see Basic Protocol, steps 1 through 15). 2b. Add a ten-fold excess of identical but unconjugated antibody to each control tube. Incubate 20 min at 4°C. This can be done with the isotypic controls as well as with the anti-cytokine antibodies.
3b. Add the predetermined amount of fluorochrome-labeled anti-cytokine antibody or isotypic control antibody to each tube (see Basic Protocol, step 16), and continue with the rest of the procedure (see Basic Protocol, steps 17 and 18). Any binding detected upon analysis can be presumed to be nonspecific and should be considered background.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
A23187, 250 ìg/ml Stock solution: Prepare at 5 mg/ml in dimethylsulfoxide (DMSO). Divide into 25-µl aliquots and store at −70°C; stable for at least one year. Working solution: Prepare by diluting stock solution 1:20 (final 250 µg/ml) in fresh cell culture medium (see recipe). Use immediately and discard any unused working solution. Brefeldin A, 100 ìg/ml Stock solution: Prepare at 10 mg/ml in methanol. Store at −70°C; stable for at least 1 year. Working solution: Prepare by diluting stock solution 1:100 (final 100 µg/ml) in cell culture medium (see recipe) and use immediately. Discard any unused working solution.
continued
Studies of Cell Function
9.9.5 Current Protocols in Cytometry
Supplement 28
Buffered fixative, 1% and 4% To prepare 500 ml of fixative, heat and stir ∼400 ml H2O to 70° to 80°C in a water bath. Add 5.35 g sodium cacodylate (cacodylic acid), 3.8 g sodium chloride, and 5.0 g (for 1%) or 20 g (4%) reagent-grade paraformaldehyde. Stir 10 to 15 min or until dissolved. Allow solution to cool while continuing to stir. Adjust total volume to 500 ml, and adjust pH to 7.2 to 7.4 with NaOH. Store up to 2 weeks at 4°C. Cell culture medium 480 ml RPMI 1640 480 ml Click’s EHAA medium 10 ml 100× nonessential amino acid solution (Life Technologies) 0.11 g sodium pyruvate 10 ml insulin, tranferrin, selenium supplement (ITS; Sigma) 3.3 µl 2-mercaptoethanol 0.6 g L-glutamine 5.0 g BSA 5.0 g catalase 10 ml human serum or 5 ml mouse or rat serum H2O to 1 liter Adjust pH to 7.4 Filter sterilize with a 0.22-µm filter and store up to two weeks at 4°C Fluorochrome-labeled antibodies A number of different antibodies from several suppliers have been used in these procedures. All antibodies listed below have been found to work satisfactorily and no particular one has been found to not work. However, antibodies or antibodyfluorochrome conjugates that are not on this list should be tested and optimized prior to experimental use. Anti-cytokine antibodies: Anti–rat IL-4, FITC-conjugated (R & D) Anti–mouse IFN-γ, PE- or FITC-conjugated (Pharmingen) Anti–rat IL-4, PE-conjugated (PharMingen) Anti–human IL-4, PE-conjugated (Becton Dickinson Immunocytometry) Anti–human IFN-γ, FITC-conjugated (Becton Dickinson Immunocytometry) Antibodies used to stain intracellular receptor: Anti-CD14, FITC-conjugated (Coulter) Antibodies used to stain intracellular proteins: Anti–heat shock protein 72, unlabeled (StressGen Biotechnologies); use with goat anti–mouse IgG, conjugated to PE or FITC (Kirkegaard and Perry) Antibodies used to stain cell-surface antigens: Almost any antibody (from any supplier) that has been used for flow cytometry can be used, including antibodies against CD3, CD4, CD8, CD11b, and CD14
continued Intracellular Cytokines
9.9.6 Supplement 28
Current Protocols in Cytometry
Kits for intracellular cytokine staining During the past several years, intracellular cytokine staining has become widely used and kits are now available for a number of cytokines. Although laboratory-tolaboratory variation in the way this technique is performed can be observed in the literature, the differences are relative minor, far less so than was apparent five or so years ago. Many companies now produce kits for the detection of intracellular cytokines, including, among others, Becton Dickinson Biosciences, BioLegend, Biozol, eBioscience, and R & D Systems (see Internet Resources) . Included among the cytokines in these kits are: Human: IL-1, IL-2, IL-3, IL-4, IL-6, IL-8, IL-10, IL-11, IL-12, IL-13, IFN, TNF, GM-CSF, GRO, IP-10, MCP-1, MCP-3, MIG, MIP-1, RANTES Mouse: IL-2, IL-3, IL-4, IL-6, IL-10IFN, TNF, GM-CSF, MCP-1 Rat: IL-4, IL-10, GM-CSF. The availability of a large number of anti-cytokine antibodies, anti–intracellular antigen antibodies, and isotypic antibodies, as well as flow cytometry instruments that can analyze four or more colors, makes multicolor immunofluorescent staining for both intracellular cytokines and cell surface antigens a powerful technique. The wide variety of fluorochrome labels makes it increasingly easy to find multiple combinations of anti–cell surface and anti-cytokine antibodies. Some of the fluorochromes found to be useful for intracellular staining include Alexa Fluor 488 and 647, FITC, PE, PerCP, Cy5, PE-Cy5, and PE-Cy5.5.
Monensin, 1 mM Stock solution: Prepare at 5% (w/v) in methanol. To facilitate solubilization of monensin, place in a 40° to 43°C water bath. Store at 4°C; stable for at least 4 months. Working solution: Dilute to 1 mM in cell culture medium (see recipe). Use within 1 hr and discard any unused working solution. Murine gamma globulin, 20 ìg/ml Stock solution: Prepare by dissolving mouse gamma globulin in wash buffer (see recipe) at a concentration of 1 mg/ml. Filter sterilize with a 0.22-µm filter. Store at 4°C; stable for at least 2 months. Working solution: To prepare, dilute stock 1:50 in wash buffer (see recipe). Discard after each use. Phorbol 12-myristate 13-acetate (PMA), 10 ìg/ml Stock solution: Prepare at 1 mg/ml in DMSO. Store at −70°C; stable for at least 1 year. Working solution: Prepare by diluting stock 1:100 (final 10 µg/ml) in cell culture medium (see recipe) just before use. Discard any unused working solution. Saponin, 0.1% (w/v) To prepare 100 ml of solution, add 0.1 g of saponin and 2.38 g N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES) to commercial Dulbecco’s phosphate-buffered solution (DPBS, e.g., Life Technologies; prepared according to manufacturer’s instructions), and bring the final volume to 100 ml. Filter sterilize with a 0.22-µm filter, and store at 4°C; stable for at least 1 month. Wash buffer Prepare commercial Dulbecco’s phosphate-buffered saline (DPBS, e.g., Life Technologies) according to manufacturer’s instructions. Add 0.5% (w/v) BSA and 0.1% (w/v) NaN3. Adjust pH with HCl or NaOH to 7.4 and filter sterilize with a 0.22-µm filter. Store at 4°C; stable for at least 1 month. Studies of Cell Function
9.9.7 Current Protocols in Cytometry
Supplement 28
A
B
29
Cell number
Cell number
21
14
7
35
0 0.1
1
100 10 IL-4-FITC
0.1
1000
D
1
10
100
1000
IFN - γ - R E D 6 1 3 s t r e p t a v i d i n
IFN-γ -RED 613
1000
IFN-γ -RED 613
53
17
0
C
71
100
10
1
0.1 0.1
1
E
100 10 IL-4-FITC
1000
0.1
1
10
100
1000
IL-4-FITC
IL-4-FITC
1000
100
10
1 0.1 0.1
1
10
100
1000
CD4-ECD
Intracellular Cytokines
Figure 9.9.1 Typical histograms of lymphocytes labeled with anti-CD3, anti-CD4, anti-IFN-γ, and anti-IL-4 antibodies. The lymphocyte populations were gated using 90° and forward-angle light scattering. The cells were further gated on CD3 and CD4. (A-D) CD3+CD4+ rat lymphocytes stained with FITC-conjugated anti-IL-4 and biotinylated anti-IFN-γ plus streptavidin–RED 613. Single-parameter histograms show IL-4-stained cells (A) and IFN-γ-stained cells (B). Two-parameter histograms show a population of cells producing IFN-γ and almost no IL-4 (C; note heterogeneity in IFN-γ staining pattern), and another population of cells producing mostly IL-4 and little IFN-γ (D; note intensity of IL-4 staining). (E) CD3+CD4+ human lymphocytes stained with ECD-conjugated anti-CD4 and PE-conjugated anti-IL-4. A two-parameter histogram displays a subpopulation of lymphocytes producing IL-4.
9.9.8 Supplement 28
Current Protocols in Cytometry
COMMENTARY Background Information Comparing intracellular staining for cytokines with other methods of measuring cytokines is similar to any such comparison of techniques—each method has its pros and cons. Intracellular cytokine staining provides a significant addition to other procedures and technologies already widely used in many laboratories. A complete profile of cytokine response can be obtained when intracellular cytokine staining is used in combination with bulk cytokine measurements—e.g., ELISA—to quantify the amount of cytokine production and with molecular methods—e.g., northern blotting or polymerase chain reaction (PCR)—to determine gene regulation. Perhaps more importantly, intracellular cytokine staining and flow cytometry allow cytokine biology to be analyzed at the single-cell level. Although the frequency of cytokine-producing cells can also be determined using the ELISPOT assay (Czerkinsky et al., 1983), significantly more information can be obtained using the methodology described here. Cells can be phenotyped, analyzed for multiple markers, examined for activation markers and/or the presence of specific receptors, and analyzed for other cytoplasmic antigens in addition to cytokines. Intracellular cytokine detection by flow cytometry also offers rapid analysis of a large number of individual cells. All these features indicate that intracellular cytokine detection can add important information that can be useful to the cell biologist examining cell activation, as well as to the immunologist interested in cytokine production and phenotypic changes or shifts during disease processes. It should be noted that this method can be applied to cells other than leukocytes. The basic procedures remain the same, with minor adjustments depending upon the cell type and the reagents. If the desired antibody/fluorochrome combination is not commercially available, it is quite easy to make directly labeled antibodies using one of the many available kits. Excellent labeling results have been obtained with kits from Molecular Probes (Alexa 350, 488, and 647) and Prozyme (PE and APC). Molecular Probes offers kits using conventional labeling procedures as well as the newer Zenon labeling technology. Both are reliable methods, but the Zenon technology is exceptionally easy and convenient. See Internet Resources for Web sites of Molecular Probes and Prozyme.
The principle of intracellular cytokine staining is actually quite simple. The cells of interest, usually lymphocytes, are stimulated with phorbol myristate acetate (PMA) and a Ca2+ ionophore (ionomycin or A23187) to increase cytokine production. The cells are activated in the presence of monensin or brefeldin A, which prevents intracellular trafficking and thus release of the cytokine from the Golgi complex. They are then fixed, permeabilized, stained for the cytokine(s) of choice, and analyzed by flow cytometry (Frede et al., 1997). With appropriate modification, this method can be used for molecules other than cytokines. For example, a modified version has been used to detect intracellular CD11b, CD14, and inducible heat shock protein 72 (Rodeberg et al., 1997). The level or quantity of the specific cytokine is often of critical importance when examining cellular interactions or disease states. A commonly asked question relates to quantification of intracellular methods versus extracellular methods such as ELISA and Cellular Bead Array (CBA; BD Biosciences). Several studies have addressed this question, including reports by Schuerwegh et al. (2003) and Elson et al., (1995). In general, authors have found good correlation between intracellular staining and other methods. Although intracellular staining detects production of the cytokine, but not secretion, most authors report a close correlation between the two.
Critical Parameters and Troubleshooting Cell quality. One of the most important aspects of this procedure is obtaining quality cell preparations to analyze. This is normally not a problem with leukocytes obtained from healthy individuals; however, leukocytes from treated or diseased animals may be an entirely different matter. It has frequently been observed that leukocytes from animals or humans treated with immunosuppressive drugs are very sensitive to the effects of cell-lysing reagents. The cells have an altered morphology and appear damaged. This is usually detected in the forward-versus-side light-scattering patterns. The area of debris increases and both the forward and side scattering signals decrease in the regions where the cells appear. Consequently, the percentage of cytokine-positive cells is usually very low and the background staining may become very high. This problem can often be alleviated by reducing the time of cell lysis
Studies of Cell Function
9.9.9 Current Protocols in Cytometry
Supplement 28
Intracellular Cytokines
and/or permeabilization. If cells are obtained from animals or humans undergoing any type of treatment, or if a portion of the cells are treated in vitro with agents not used on the control cells, the light-scattering patterns need to be observed carefully. A reduction in light scattering and/or an increase in debris is suggestive of possible problems. Cell-surface staining and fixation. Cells are stained for surface antigens (i.e., phenotyped) before fixation to prevent destruction of antigens by the fixative. The fixation then crosslinks the antibodies to the cell surface. It is very important to fix the cells after cytokine induction but before the permeabilization step. This step purportedly prevents the loss of the cytokines from the cells once the membrane is permeabilized. However, the period of fixation is critical. If cells are overfixed, the cytokines or cell surface antigens may be overly crosslinked and may not react with the antibodies. In some cases, any fixation may destroy reactivity. This is a difficult problem that may not be completely solvable, although screening different clones of antibodies that detect different epitopes on the same molecule may help in some cases. Lysis and permeabilization. Many reagents are commercially available for lysis of red cells and permeabilization of the cell membrane. Any of these reagents can probably be used under the appropriate conditions. However, saponin has the advantage of both lysing red cells and permeabilizing the membrane in one step. Its effects are also easily modulated by altering the period of incubation. Several of the commercial products have difficulty lysing red cells from animals and humans that have either undergone treatment with immunosuppressive drugs or experienced severe trauma. This is also true of blood that has been refrigerated. In these cases, aggressive lysing treatment is needed, which may damage or destroy the leukocytes (see cell quality, above). Time. The strength of the cytokine response in activated lymphocytes is dependent upon time. Unfortunately, the optimal time for production of many of the cytokines varies. If more than one cytokine is to be measured in the same cells, the time period required for activation must be maximized for the detection of each cytokine. Fluorochrome. Optimal detection of the cytokine in the lowest concentration can be enhanced by use of the proper fluorochrome. Phycoerythrin (PE) generally gives the highest
signal-to-noise ratio (S/N) and should be used for the cytokine that has the lowest concentration, if possible. Tricolor (PE/Cy5) also has a high S/N and is good if weak signals are expected. It is often assumed that large molecules such as PE or the tandem reagents cannot get into cells. However, it has been the author’s experience, as well as that of other investigators, that these reagents enter the cells as readily as the smaller dyes. Cell activation. The activation of T cells is most commonly achieved using PMA and either ionomycin or A23187. These compounds give a fairly strong response which is quite reproducible. The response to more physiological antigens appears more variable, ranging from fairly strong to none. The author has consistently found that the T cell production of IFN-γ and IL-4 (as markers for TH1 and TH2, respectively) in response to allogeneic lymphocytes is consistently strong. However, the responding cells (cytokine producers) must undergo an extended stimulation period (15 hr) before the addition of monensin or brefeldin A.
Anticipated Results The results obtained by these techniques are much like those from multicolor immunophenotyping techniques. As with any measurement of antigen-antibody interaction by flow cytometry, the density of the antigen determines the strength of the fluorescence under optimal conditions. When measuring intracellular cytokines or other molecules, the cell-to-cell variability, in terms of fluorescence, can be significant. This should not automatically be considered experimental error; it probably represents true variation in the level of cytokine production. When cells are stained for IL-4 and IFN-γ together, IL-4 is often the dimmer of the two. It is suggested that the procedure be optimized for this cytokine. Examples of typical intracellular cytokine data are shown in Figure 9.9.1.
Time Considerations The measurement of intracellular cytokine is a long procedure. Expect to spend 1 hr preparing the cells, 4 to 20 hr stimulating the cells, and 4 to 5 hr preparing and staining the activated cells. Generally, analysis of samples is slow because of a limited number of cells and the fairly long setup associated with four-color compensation. Allow an average of 5 min/tube in the panel.
9.9.10 Supplement 28
Current Protocols in Cytometry
Literature Cited Czerkinsky, C., Nilsson, L., Nygren, H., Ouchterlony, Ö., and Tarkowski, A. 1983. A solid phase enzyme-linked immunospot (ELISPOT) assay for enumeration for specific antibody-secreting cells. J. Immunol. Methods 65:109-121. Elston, L., Nutman, T., Metcalfe, D., and Prussin, C. 1995. Flow cytometric analysis for cytokine production identifies T helper 1 and T helper 2 and T helper 0 cells within the human CD4+ CD27– lymphocyte population. J. Immunol. 154:42944301. Frede, S.E., Valente, J., Alexander, J.W., and Babcock, G.F. 1997. The relationship of blood transfusion and immunosuppression with the Th1/Th2 paradigm. Transplant. Proc. 29:11531154. Kruisbeek, A.M. 1993. Isolation of mouse mononuclear cells. In Current Protocols in Immunology (J.E. Coligan, A.M. Kruisbeek, D.H. Margulies, E.M. Shevach, and W. Strober, eds.) pp. 3.1.23.1.5. John Wiley & Sons, New York.
Prussin, C. 1997. Cytokine flow cytometry: Understanding cytokine biology at the single-cell level. J. Clin. Immunol. 17:195-204. An excellent review and troubleshooting guide. Sander, B., Andersson, J., and Andersson, U. 1991. Assessment of cytokines by immunofluorescence and the paraformaldehyde-saponin procedure. Immunol. Rev. 119:65-93. First description of the use of cytometry to measure intracellular cytokines. Vikingson, A., Pederson, K., and Muller, D. 1994. Enumeration of IFN-γ producing lymphocytes by flow cytometry and correlation with quantitative measurement of IFN-γ. J. Immunol. Methods 173:219-228. Explanation of protocol and comparison of intracellular staining with other methods.
Internet Resources http://www.bdbiosciences.com
Rodeberg, D.E., Morris R.E., and Babcock, G.F. 1997. Azurophilic granules of human neutrophils contain CD14. Infect. Immun. 65:47474753.
Web site of Becton Dickinson Biosciences
Schuerwegh, A., DeClerck, L., Bridts, C., and Stevens, W. 2003. Comparison of intracellular cytokine production with extracellular cytokine levels using two flow cytometric techniques. Cytometry 55B:52-58.
http://www.biozol.com
http://www.biolegend.com Web site of BioLegend.
Web site of Biozol (Eching, Germany). http://www.eBioscience.com Web site of eBioscience.
Key References Levy, A.E., Alexander, J.W., and Babcock, G.F. 1997. A strategy for generating consistent long term donor specific tolerance to solid organ allografts. Transplant. Immunol. 5:83-88.
http://rndsystems.com Web site of R & D Systems. http://www.probes.com
Application of this protocol for detection of TH1 and TH2 cells.
Web site of Molecular Probes.
Prussin, C. and Metcalfe, D. 1995. Detection of intracytoplasmic cytokine using flow cytometry and directly conjugated anti-cytokine antibodies. J. Immunol. Methods 188:117-128.
Web site of Prozyme.
Description of basic method and various fluorochromes that can be used.
http://www.prozyme.com
Contributed by George F. Babcock University of Cincinnati College of Medicine and Shriners Burns Institute Cincinnati, Ohio
Studies of Cell Function
9.9.11 Current Protocols in Cytometry
Supplement 28
Assays of Natural Killer (NK) Cell Ligation to Target Cells The formation of a conjugate between an effector lymphocyte and a target cell is the first step in cellular cytotoxicity reactions. This protocol, previously known as the “binding assay,” measures the ability of different lymphoid cell populations to adhere to a tumor target cell. In particular, it is used to monitor the formation of conjugates in vitro between natural killer (NK) cells and susceptible target cells (e.g., K562 cells for human NK, YAC cells for mouse NK) based on side-scatter signals, and it allows discrimination between effector and target cells based on detection of effector cell–specific antigens using fluorochrome-conjugated monoclonal antibodies.
UNIT 9.10
BASIC PROTOCOL
Most commercially available flow cytometers can be used to perform this assay. This protocol requires expertise in basic flow cytometric techniques (Chapter 1, UNITS 5.1 & 6.2). Materials Carbonyl iron (Sigma) Heparinized human blood (UNIT 5.1) Phosphate-buffered saline (PBS), pH 7.4 (APPENDIX 2A) Ficoll-Hypaque (Seromed) or Percoll (Sigma; S = 1.077) RPMI-10 medium (see recipe) K562 chronic myeloid leukemia cells grown in suspension in RPMI-10 to logarithmic phase Propidium iodide (PI) stock solution (see recipe) Fluorochrome-conjugated monoclonal antibodies (MAbs; e.g., PE-conjugated anti-CD16, FITC-conjugated anti-CD3, and PerCP-conjugated anti-CD8) 37°C water bath 15- and 50-ml polypropylene centrifuge tubes (Falcon) Beckman GS-15R centrifuge (or equivalent) Flow cytometer with 488-nm light source (e.g., Becton Dickinson, Coulter, or Ortho) Software for multiparametric analysis Additional reagents and equipment for removing monocytes from blood (UNIT 5.1), cell culture (APPENDIX 3B), and counting cells (APPENDIX 3A) NOTE: All solutions and equipment coming into contact with cells must be sterile and proper sterile technique should be used accordingly. NOTE: All culture incubations are performed in a humidified 37°C, 5% CO2 incubator unless otherwise specified. Prepare effector cells 1. In order to deplete monocytes, add carbonyl iron to a final concentration of 10 mg/ml to heparinized human blood, and resuspend the mixture with a 10-ml pipet. Incubate 1 hr at 37°C, gently agitating the tube every 10 min. Heparinized blood (10 ml) or buffy coat cells (50 ml) can be used as a source of effector cells. Monocytes phagocytose iron particles so they become heavier than other cells. The majority of the monocyte population will be at the bottom of the tube after gradient separation. Studies of Cell Function Contributed by Filippo Renò, Stefano Papa, Marco Vitale, and Loris Zamai Current Protocols in Cytometry (1998) 9.10.1-9.10.8 Copyright © 1998 by John Wiley & Sons, Inc.
9.10.1 Supplement 4
2. Dilute the blood cell suspension 1:1 with PBS, pH 7.4. Dilute buffy coat material 1:5 with PBS, pH 7.4.
3. Layer 10 ml diluted cells on a 3-ml Ficoll or Percoll cushion in a 15-ml polypropylene centrifuge tube. Centrifuge 20 min at 700 × g, room temperature. 4. Using a pipet, gently remove 3 to 5 ml of plasma, then collect the white ring of mononuclear cells. (UNIT 5.1). The mononuclear cells are found in the interface between Ficoll or Percoll cushion and the plasma.
5. Wash the mononuclear cells twice by resuspending the cells in RPMI-10 and centrifuging 5 min at 400 × g, room temperature. The mononuclear fraction is a heterologous population containing B and T lymphocytes and NK cells.
6. Resuspend the pellet in 10 ml RPMI-10 and transfer to a 75-cm2 flask. Incubate 30 min at 37°C to deplete residual monocytes by adhesion to the plastic flask (UNIT 5.1). At this point, it is a good idea to immunophenotype the cells to determine the distribution of lymphocyte subpopulations, e.g., CD16+, CD 56+, CD16+8+, and CD57+.
Prepare target cells 7. Harvest logarithmic phase culture of K562 target cells. If the cultures are harvested during the logarithmic phase, there are a minimum of dead target cells.
8. Count both peripheral blood lymphocytes (PBL; step 6) and K562 cells using a hemocytometer (APPENDIX 3A). Alternatively, the cells can be counted using a Neubauer chamber. It is important to assess the number of dead cells in the target population before starting each experiment (UNIT 9.2). Determine the percentage of dead cells by trypan blue exclusion (APPENDIX 3A). In order to obtain good results, the proportion of dead cells should not exceed 5%, not only in this ligation protocol but also in the killing protocol.
9. Mix effector and target cells in an effector-to-target (E:T) ratio of 1:1 and check their proportion using the cytometer. Manual counting can lead to an error in the relative proportion between target and effector cells. A good method to check the real E:T ratio is to draw two different gates in the scatter contour plot as shown in Figure 9.10.1. Using this plot it is also possible to check for the presence of a residual monocytic population that can alter the assay result.
Form conjugates 10. Transfer a total of 2 × 105 cells to each 15-ml tube. Add PI stock solution to a final concentration of 0.5 µg PI/ml. CAUTION: Propidium iodide is potentially hazardous: it is known to be a tumorigenic agent. E:T ratios of 2:1 and 4:1 could also be used in order to enhance the relative number of conjugates; higher E:T ratios will not affect the determination, but they will allow a more precise definition of the level of conjugation for each lymphocyte subpopulation. Assays of Natural Killer (NK) Cell Ligation to Target Cells
Propidium iodide is used in this case for detecting dead cells. Dead cells will appear bright orange-red in the fluorescence diagram and can easily be gated out.
9.10.2 Supplement 4
Current Protocols in Cytometry
1000
K562 cell
Side scatter
800
600
400 lymphocytes 200
0 0
200
400
600
800
1000
Forward scatter
Figure 9.10.1 Scatter-contour plot of lymphocytes and K562 cells mixed in a 1:1 ratio. Note the difference in size between the lymphocytes and the big tumor cells. Using this graph, it is possible to verify both the ratio between effector and target cells and the presence of monocytes, which occupy the intermediate area and which could be confused with small target cells. The arrow indicates the region where monocytes are found if they are not depleted.
11. Centrifuge the tube 7 min at 250 × g, room temperature. Incubate 10 min in a 37°C water bath to promote conjugate formation. Then incubate 30 min at 4°C to prevent lytic activation. It is preferable to use a water bath with mild agitation because that will help the sample reach working temperature in a shorter period of time. Times must be strictly adhered to because activation of the lytic process can reduce the number of bound effectors due to their detachment from the target cells just after the release of lysosomal products.
Stain samples 12. Gently resuspend samples. Add fluorochrome-conjugated MAbs to a concentration of 20 µg/ml per 106 effector cells and incubate 20 min at 4°C. At the same time incubate 106 mononuclear cells with the same combination of MAbs for instrument standardization. Incubate 105 target cells alone to check for autofluorescence. For the experiment described here, only two MAbs (PE-conjugated anti-CD16 for NK cells and FITC-conjugated anti-CD3 for T lymphocytes) are used, to simplify the description. The MAbs should be directly labeled with fluorochromes such as FITC, PE, PerCP, PECy5, or APC (UNIT 6.2). The MAbs should recognize antigens that are not present on the target cell surface and do not interfere with the specific binding of effector cells. Experiments can also be performed to investigate the presence of surface antigens that are present in the binding site (Papa et al., 1994; Zamai et al., 1994).
13. Wash the cells once with PBS.
Studies of Cell Function
9.10.3 Current Protocols in Cytometry
Supplement 4
104
103 PE anti- CD16
unbound
bound
102 bound K562
101
unbound 100 100
101
102
103
104
FITC anti- CD3
Figure 9.10.2 Fluorescence dot-plot of lymphocytes identified with two different monoclonal antibodies, FITC-conjugated anti-CD3 and PE-conjugated anti-CD16, and autofluorescent K562 cells. It is possible to identify CD3+ (T-lymphocytes) and CD16+ (NK) cells conjugated with target cells.
Perform flow cytometry 14. Analyze the target cell population alone. In the scatter plot (see Fig. 9.10.1), position live K562 cells in the upper right quadrant of the plot to allow good resolution of the lymphocytes when the two populations are analyzed together. If the K562 population is in another position, correct the amplification of the scatter signals. The position of K562 cells in the fluorescence cytogram is essential to allow discrimination between bound and unbound effector cells. K562 tumor cells are larger than lymphocytes, and monocytes are between the two in size. In a fluorescence cytogram, K562 cells should have autofluorescence values similar to the ones shown in Figures 9.10.2 and 9.10.3. Dead cells also show higher autofluorescence values and can compromise the results if the starting percentage of dead cells is >5% to 10%.
15. Analyze effector cells alone, using a scatter plot to check their position and to check for the presence of residual monocytes. 16. Using a fluorescence cytogram, compensate the fluorescence of labeled effectors. Simultaneous analysis of two subpopulations can be performed using lymphocytes labeled with two MAbs that recognize antigens not coexpressed by the same subpopulations. The MAbs must be conjugated with different fluorochromes (e.g., FITC-conjugated anti-CD4 versus PE-conjugated anti-CD8 or FITC-conjugated anti-CD3 versus PE-conjugated anti-CD16).
17. Analyze a 1:1 mixture of target and effector cells (see Fig. 9.10.2). Assays of Natural Killer (NK) Cell Ligation to Target Cells
18. Draw a gate (R1) around all the population positive for the antigen(s) of interest. In Figure 9.10.2, only the CD16-positive cells are gated.
9.10.4 Supplement 4
Current Protocols in Cytometry
19. Draw another gate (R2) around the bound cells inside the previous gate. If the fluorescence compensation is carried out correctly and the target cell viability is 90% to 95%, bound cells are recognized as those acquiring bound target cell fluorescence (see Fig. 9.10.2).
20. Acquire ≥50,000 total events. Divide the number of cells in R2 (bound) by the number of cells in R1 (bound + unbound) to obtain the percentage of bound cells belonging to the effector subpopulations being examined. If the effector subpopulations are poorly represented, acquire only that population using gate R1 as a “live” gate so only cells positive for that antigen will be acquired.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Propidium iodide solution Dissolve propidium iodide to a final concentration of 10 µg/ml in RPMI 1640. Store up to 6 months at 4°C. CAUTION: Propidium iodide is a potentially hazardous agent; therefore always use all necessary precautions when handling it, especially while weighing the powder.
RPMI-10 RPMI 1640 medium containing: 10% (v/v) fetal bovine serum (FBS; heat inactivated 30 min at 56°C; APPENDIX 2A) 100 U/ml penicillin 0.1 mg/ml streptomycin sulfate Store at 4°C COMMENTARY Background Information During the last decade a number of optical microscopy– or flow cytometry–based methods have been developed for quantitative determination of the lytic activity of NK/CTL cells against specific targets (Kimberley et al., 1986; Papa et al., 1988). Other methods, meanwhile, were developed for recognition and quantitative/qualitative evaluation of cells forming conjugates with tumor targets (Segal and Stephany, 1984; Storkus et al., 1986; Vitale et al., 1992). These techniques are characterized by being single-cell assays; they are more reliable and controllable than total cell assays such as the 51Cr release assay (Albright and Albright, 1983) and the FDA assay (Blomberg et al., 1996). Moreover, they are based on the ability to distinguish effector cells, living and dead target cells, and conjugates. This ability to distinguish the different cells is based on differences between scattering signals produced by effector and target cells (Fig. 9.10.1), autofluorescence displayed by different cell types in flow analysis as seen for K562 cells (Fig. 9.10.2), and fluorescence acquired after bind-
ing of specific fluorochrome-conjugated monoclonal antibodies to different populations of effector cells (Fig. 9.10.2 and 9.10.3). Measurement of light scatter is advantageous because it avoids any possible interference with lytic activity due to previous sample preparation. The relationship that links forward scatter to cellular dimension through the diffraction pattern is well known. However, the forward-scatter signal is also affected by the relative refractive index (Mullaney and Dean, 1970), which undergoes large variation during cell death, thus inducing a constant lowering of the forward-scatter signal and allowing detection of cell death on the cytogram. On the other hand, side scatter has been shown to depend mainly on cytoplasmic granularity and consequently on cytoplasmic vacuolization during cell death (Vitale et al., 1989). Therefore, the morphological changes that take place after killing produce significant modifications of cellular light-scattering properties. The main requirement of these techniques is the ability to clearly discriminate effector cells from target cells and conjugates. While
Studies of Cell Function
9.10.5 Current Protocols in Cytometry
Supplement 4
104
unbound PE anti-CD16
103
R1
bound R2
102
101
100 100
101
102
103
104
FITC anti-CD3
Figure 9.10.3 Fluorescence dot-plot showing gating methods for calculating the proportion of NK cells bound to the target cells. A gate is drawn for CD16+ cells (bound and unbound, R1 + R2). Another gate is drawn only around the bound portion (R2). The first gated region contains 1905 of 50,000 cells (3.81%); there are 1003 bound cells (2.81% of the total and 52.65% of the gated cells). The abscissa mean value for unbound cells is 23.2 and for bound cells it is 35.4
Assays of Natural Killer (NK) Cell Ligation to Target Cells
this can be obtained on a scattering matrix for evaluation of cytotoxicity, for evaluation of conjugates this discrimination must be made on the fluorescence matrix, making this method more flexible for the use of different target cells. Furthermore, this binding assay permits identification of different effector cells attached to the targets (the conjugates) based on effector cell labeling with different directly conjugated monoclonal antibodies. Different monoclonal antibodies are identified based on the conjugated fluorochrome, so this technique resembles a typical dual- or triple-fluorescence analysis as in a normal flow phenotype analysis. These methods have been set up primarily in order to investigate non-MHC-restricted cytotoxicity (Herberman et al., 1986). They are not restricted to this application but can be extended to antibody-dependent cellular cytotoxicity (ADCC) or other adhesion-based mechanisms. The authors’ experience demonstrates the usefulness of these methods with different target cells (data not shown). The only limitations are that the methods can be used to monitor only reactions that occur in suspension, and the effector compartment (lympho-
cyte in this case) must be clearly distinguishable on the basis of scatter signals.
Critical Parameters and Troubleshooting This method requires an accurate and complete depletion of monocytes. Monocytes can alter the assay results because their autofluorescence can overlap the K562 tumor cell autofluorescence and their size is similar to that of small target cells, so they may be counted as lymphocytes. However, monocyte contamination is easily monitored: monocytes are present in a specific area between lymphocytes and K562 cells in a scatter plot of a mixed sample (see Fig. 9.10.1), and they do not possess any of the antigens used to recognize lymphocyte subpopulations. Monocyte contamination can be relatively low in a 1:1 mixture of effector and target cells, but it becomes larger when the effector-to-target ratio is increased to 2:1 or 4:1 for better resolution of bound cells. This protocol is also preparatory for the single-cell cytotoxicity assay where effector to target cell ratios are 6.25:1 to 25:1 and the monocyte contamination can be relatively high.
9.10.6 Supplement 4
Current Protocols in Cytometry
Regarding targets, when cell lines are employed, it is necessary to work with cells in the logarithmic phase of growth to reduce the number of spontaneous dead cells present in the samples. Nonviable cells affect both assays: the SCCA, for the high level of dead cells present in controls, and the binding assay, for the increased autofluorescence of dead cells. This last problem can be avoided by gating on the scattering cytogram (Fig. 9.10.1 and 9.10.2). As for other flow techniques, these methods need an accurate setting of the cytometer because of the use of all parameters, including scatter, for analysis. Moreover, when these samples are run on analyzers (such as FACScan), amplification of scatter signals must be corrected. For setting the proper signal amplification at E:T = 1:1, the mixed sample must be run in order to optimize the cytogram position of effector cells and target cells (Fig. 9.10.1). Another critical parameter is fluorescence compensation. It is preferable to proceed with a dual-parameter compensation setting done using directly labelled cells (for example, lymphocytes labelled with PE-conjugated antiCD8 and FITC-conjugated anti-CD4) from the mononuclear samples under investigation. In this way, the sample can be used both for setting the proper signal amplification for the scatter and for amplification and compensation of the fluorescence channels (PE for red and FITC for green). Compensation should be employed in order to perform accurate calculations. Many problems can be due to the type of cytometer used and its optic and fluidic systems. For this assay, the set of filters employed is composed of a 488-nm bandpass filter on the orthogonal scatter, a 530 ± 15-nm bandpass filter on the green channel, and a 585 ± 30-nm bandpass filter on the red channel. The optics and hydrodynamics of analyzers such as FACScan, Ortho Absolute, Coulter Profile, and Coulter XL are perfectly optimized to perform these assays. If a sorter is used to perform this assay, it must be equipped with a nozzle tip of diameter ≥100 µm to avoid disruption of conjugates when they pass through the orifice (Vitale et al., 1989). Sample analysis must be performed at a slow flow rate with a sheath pressure ∼10 lb/in.2 in order to obtain a good orientation of conjugates within the jet. In the case of triple staining for effector cells, the setting of the third fluorescence channel
should be obtained according to the manufacturer’s instructions. Whereas analyzers are fixed machines, sorters can be adjusted with proper filters for different fluorescent probes; for example, a duallaser instrument will need a 660 ± 20-nm bandpass filter for APC and Tricolor and a 630 ± 20-nm bandpass filter for RED 613. MAbs for these experiments must be carefully selected. Antigens recognized by the MAbs cannot be expressed in the target cell membrane or on the monocytes. If the antigen is expressed by monocytes, the monocyte population must be completely depleted. The antigens recognized by the MAbs cannot be coexpressed on the same effector subpopulations. If the antigens are coexpressed, triple staining should be done using two other MAbs—one FITC-conjugated and the other PE-conjugated—to distinguish the two subpopulations and their binding to the target cells. The presence of the shared antigen in both populations is then detected using an argon laser and a third MAb labeled with a fluorochrome, such as PerCP and Red 670, that has an emission >600 nm. Different MAbs belonging to the same cluster should be pretested to avoid using antibodies that increase or inhibit effector-to-target cell binding. For example, when using cells that express the Fc receptor, F(ab′)2 MAbs must be used to avoid antibody-mediated binding. Finally, the brighter the fluorescence of the MAb, the better the measurement of binding.
Anticipated Results The behavior of effector and target cells can be detected in flow cytometry on the basis of scatter signals. This method has an advantage because of the relative difference in signal intensity between lymphocytes and target cells (K562). Lymphocytes are morphologically one of the simplest cells in the organism, consisting of a nucleus and a little cytoplasm, and their signal is restricted to a specific area in both forward and side scatter. In contrast, all cells derived from peripheral tumors and cell lines are much larger and present a large cytoplasmic compartment enriched with granules; this enhances their side scatter and autofluorescence signals (see Fig. 9.10.1). In the binding assay, autofluorescence of target cells plays a major role in detecting conjugates and assessing the type of effector cell present. This assay was designed to identify the lymphoid subpopulation able to bind to targets in a multiparametric analysis, including small subsets of lymphocytes which can only be distinguished by dou-
Studies of Cell Function
9.10.7 Current Protocols in Cytometry
Supplement 4
ble staining with directly labeled monoclonal antibodies. The discovery of new 488 nm–excitable fluorochromes, such as RED 613, that emit in the far-red region of the spectrum has given a further boost to the application of this technique by allowing the use of MAbs in triple-staining experiments to discriminate between three different effector populations in a single experiment. Because the targets are clearly distinguishable on the fluorescence cytogram (Fig. 9.10.2), a two- or three-color analysis is best when effector analysis is based on two antigens that do not overlap (Fig. 9.10.2). A third antigen could be one that is partially represented in the subset defined by the first two monoclonal antibodies. The third antigen could also be one that displays a dim fluorescence only just overlapping the autofluorescence region of the target cells. Evaluation of binding values can be obtained directly from the fluorescence cytogram for the first two subsets by gating the whole subset population (bound and unbound lymphocytes; Fig. 9.10.2) then extracting the value of the bound ones (Fig. 9.10.3).
Time Considerations
Lymphocyte sample preparation takes ∼2 hr. Target cell preparation takes only a few minutes. The whole binding process takes just over an hour. Immunophenotyping takes ∼1 hr. Flow cytometry analysis takes ∼5 min per sample.
Kimberley, M.G., Chapman, G., Marks, R., and Perry, R. 1986. A fluorescence NK assay using flow cytometry. J. Immunol. Methods 86:7-13. Mullaney, P.F. and Dean, P.N. 1970. The small angle light scattering of biological cells. Theoretical considerations. Biophys. J. 10:764-773. Papa, S., Vitale, M., Mariani, A.R., Roda, P., Facchini, A., and Manzoli, F.A. 1988. Natural killer function in flow cytometry. I. Evaluation of NK lytic activity on K562 cell line. J. Immunol. Methods 107:73-78. Papa, S., Gregorini, A., Pascucci, E., Bartolucci, M., Rocchi, M.B.L., and Valentini, M. 1994. Inhibition of NK binding to K562 cells induced by MAb saturation of adhesion molecules on target membrane. Eur. J. Histochem. 38:83-90. Segal, D.M. and Stephany, D.A. 1984. The measurement of specific cell-cell interactions by dual-parameter flow cytometry. Cytometry 5:169-174. Storkus, W.S., Balber, A.E., and Dawson, J.R. 1986. Quantitation and sorting of vitally stained natural killer cell-target cell conjugates by dual beam flow cytometry. Cytometry 7:163-169. Vitale, M., Rizzoli, R., Mariani, A.R., Neri, L.M., Facchini, A., and Papa, S. 1989. Evaluation of NK-to-target cell binding and evidence for T cell conjugates by flow cytometry. Cytotechnology 2:59-62. Vitale, M., Zamai, L., Papa, S., Mazzotti, G., Facchini, A., Monti, G., and Manzoli, F.A. 1992. Natural killer function in flow cytometry. III. Surface marker determination of K562-conjugated lymphocytes by dual laser flow cytometry. J. Immunol. Methods. 149:189-196. Zamai, L., Rana, R., Mazzotti, G., Centurione, L., Di Pietro, R., and Vitali, M. 1994. Lymphocyte binding to K562 cells: Effect of target cell irradiation and correlation with ICAM-1 and LFA-3 expression. Eur. J. Histochem. 38-53-60.
Literature Cited Albright, J.W. and Albright, J.F. 1983. Age-associated impairment of murine natural killer activity. Proc. Natl. Acad. Sci. U.S.A. 80:6371-6375. Blomberg, K., Hautala, R., Lövgren, J., Mukkala, V.-M., Lindquist, C., and Akerman, K. 1996. Time-resolved fluorometric assay for natural killer activity using target cells labelled with a fluorescence enhancing ligand. J. Immunol. Methods 193:199-206. Herberman, R.B., Reinolds, C.W., and Ortaldo, J.R. 1986. Mechanism of cytotoxicity by natural killer (NK) cells. Annu. Rev. Immunol. 4:651-664.
Contributed by Filippo Renò and Stefano Papa University of Urbino Urbino, Italy Marco Vitale University of Brescia Brescia, Italy Loris Zamai University of Bologna Bologna, Italy
Assays of Natural Killer (NK) Cell Ligation to Target Cells
9.10.8 Supplement 4
Current Protocols in Cytometry
Flow Cytometric Analysis of Cell Division by Dye Dilution
UNIT 9.11
As cells of the immune system undergo considerable expansion and differentiation during both ontogeny and immune responses, the ability to determine the division history of cell populations undergoing proliferation is particularly useful in the study of immunological phenomena. There are a number of techniques available for determining cell division both in vivo and in vitro; however, most (such as incorporation of BrdU) have the drawback of only being able to identify cells that have undergone a limited number of cell divisions. Other methods exist that can quantify overall division (e.g., incorporation of tritiated thymidine) but tell nothing about the division history of individual cells. The technique described in this unit uses the intracellular fluorescent label carboxyfluorescein diacetate succinimidyl ester (CFDA-SE or CFSE) to tag proliferating cells. Covalently bound CFSE is divided equally between daughter cells, allowing discrimination of successive rounds of cell division. The technique is applicable to in vitro cell division, as well as to in vivo division of adoptively transferred cells, and can resolve up to eight successive generations using flow cytometry. The stain is long lived, allowing cells to be analyzed for several months after transfer. CFSE is fluorescein-derived and has the same spectral characteristics as fluorescein, allowing monoclonal antibodies conjugated to phycoerythrin (PE) or other compatible fluorochromes to be used to immunophenotype the dividing cells. Since its introduction in 1994, this technique has become the method of choice for analyzing in vitro and in vivo division of lymphocytes and other hematopoietic cells. The original publication (Lyons and Parish, 1994) had been cited over 420 times by mid-2003, with many more citing secondary sources. DETERMINATION OF CELL DIVISION USING CARBOXYFLUORESCEIN DIACETATE SUCCINIMIDYL ESTER (CFDA-SE or CFSE) The cell-permeant fluorescein-based dye CFSE covalently attaches to cytoplasmic components of cells, resulting in uniform bright fluorescence. Upon cell division, the dye is distributed equally between daughter cells, allowing the resolution of up to eight cycles of cell division by flow cytometry. This technique has been useful in determining division-related phenotypic and functional changes during differentiation of B cells, T cells, and hematopoietic precursor cells.
BASIC PROTOCOL
Materials Single-cell suspension of cells of interest (e.g., lymphoid cells, cultured cell line) PBS (APPENDIX 2A)/0.1% (w/v) BSA 5 mM 5-(and -6)-carboxyfluorescein diacetate succinimidyl ester (CFDA-SE or CFSE; see recipe) RPMI 1640/10% (v/v) FBS Culture medium or injection medium appropriate for the experiment Antibodies for immunophenotyping (optional) Flow cytometer with 488-nm argon laser, or multi-laser instrument if fluorochromes not excited at 488 nm are being used Additional reagents and equipment for cell culture and harvesting (APPENDIX 3B), and for immunophenotyping (UNIT 6.2)
Studies of Cell Function Contributed by A. Bruce Lyons and Kathleen V. Doherty Current Protocols in Cytometry (2004) 9.11.1-9.11.10 Copyright © 2004 by John Wiley & Sons, Inc.
9.11.1 Supplement 27
1. Resuspend cells of interest in PBS/0.1% BSA at a final concentration of 5 × 107 cells/ml. It is important that cells be well suspended and not aggregated, to ensure uniform labeling with CFSE. The quantity of cells labeled will depend on the experiment. Starting numbers for in vitro experiments are typically from 105 to 106 cells per ml, depending on the duration of culture. In contrast, between 1 × 107 and 5 × 107 cells are injected per mouse for adoptive transfers. Inclusion of protein (BSA) improves viability (see Critical Parameters and Troubleshooting).
2. Add 2 µl of 5 mM CFSE per milliliter cells (final 10 µM) in a tube that is ≥6× the volume of cells. Incubate 10 min at 37°C. Alternatively, dilute 5 mM CFSE solution to 20 ìM in PBS/0.1% BSA and add immediately to an equal volume of 2× concentrated cell suspension to improve uniformity of staining (see Critical Parameters and Troubleshooting). These conditions provide adequate staining intensity for tracking division over 2 to 14 days (Fig. 9.11.1). Because the starting intensity is essentially linear with respect to CFSE concentration (Fig. 9.11.2), staining can be manipulated if necessary (see Background Information).
3. Quench staining by adding 5 vol ice-cold RPMI 1640/10% FBS and incubating 5 min on ice. The high protein concentration in this step ensures that unbound CFSE is mopped up by free amines, stopping further staining.
4. Wash cells three times in the culture medium to be used, or, for in vivo experiments, in the injecting medium. This step ensures that CFSE bound to protein in the supernatant is removed, preventing any subsequent uptake into bystander cells.
Percent starting fluorescence
100
75
50
25
0
0
10
20
30
40
50
60
Days postinjection
Flow Cytometric Analysis of Cell Division by Dye Dilution
Figure 9.11.1 Decay of CFSE fluorescence intensity of nondivided cells. CFSE-labeled murine splenic lymphocytes (2 × 107) were injected intravenously. At time intervals shown, spleens were removed and analyzed by flow cytometry. The fluorescence intensity of nondividing cells was determined and expressed as a percentage of starting CFSE fluorescence intensity.
9.11.2 Supplement 27
Current Protocols in Cytometry
2000
Fluorescence intensity
1500
1000
500
0 0
2.5
5 7.5 ( CFSE µM)
10
12.5
Figure 9.11.2 Linearity of staining with respect to CFSE concentration. Murine splenic lymphocytes were stained using the standard protocol with a series of final CFSE concentrations. Fluorescence intensity (arbitrary units) plotted against CFSE concentration shows that staining is linear with respect to CFSE concentration.
200
CD4+
control
Events
CD8+
0 100
101
102 103 CFSE fluorescence
104
Figure 9.11.3 In vitro proliferation of T cells. Murine splenic T cells purified by flow cytometry were stained with CFSE and cultured alone or in the presence of immobilized anti-CD3 antibody and anti-CD28. After 3 days of culture, cells were labeled with PE-conjugated anti-CD4 or anti-CD8 antibodies before analysis. Histograms show CFSE fluorescence of undivided control cells, CD4+ T cells, and CD8+ T cells. Note that CFSE staining in conjunction with immunophenotyping allows comparison of the kinetics of proliferation in different populations. Studies of Cell Function
9.11.3 Current Protocols in Cytometry
Supplement 27
Anti-CD45R-PE fluorescence (orange)
104
103
102
101
10 0
100
101 102 103 CFSE fluorescence (green)
104
Figure 9.11.4 In vivo B cell division in absence of T cell division after injection of splenic lymphocytes. Murine splenic lymphocytes (2 × 107) labeled with CFSE were injected intravenously. After 14 days, a cell suspension of the recipient’s spleen was labeled with anti-CD45R-PE to enable discrimination between B and T cells. Events with green fluorescence above the autofluorescence background were collected, ensuring that only CFSE-positive events were analyzed (∼0.5% of total events in this experiment). A total of 5000 CFSE-positive events were collected, and anti-CD45R staining revealed cell division in the B, but not T, cell compartment.
5. Set up in vitro cell cultures under appropriate conditions, or adoptively transfer cells. These steps will be determined entirely by the aim of the experiment. See Commentary for a number of references to published work looking at in vitro and in vivo behavior of a number of lymphohematopoietic cell types. Adoptive transfer of lymphocytes is usually performed by intravenous injection via the tail vein in rats and mice (see Fulcher et al., 1996; Lyons, 1997).
6. Harvest cells and stain with antibodies for immunophenotyping (UNIT appropriate.
6.2)
where
Antibodies labeled with phycoerythrin (PE), PerCP, or a tandem dye such as PE/Cy5 are recommended for use with CFSE labeling (see Critical Parameters and Troubleshooting). See Figure 9.11.3 for an example of a proliferating culture of T cells.
7. Set up a flow cytometer for excitation at 488 nm. Collect green fluorescence (CFSE) with a 525-nm band-pass filter. Collect immunophenotyping signals with a 575-nm band-pass filter for orange fluorescence (for PE) or a 675-nm band-pass filter for red fluorescence (for PerCP or PE/Cy5).
Flow Cytometric Analysis of Cell Division by Dye Dilution
At early time points, green-channel fluorescence intensity may be very bright and high compensation levels may be needed to remove contaminating signal in the other channels, particularly the orange channel. For the FACscan (Becton Dickinson), increasing the green-channel gain will decrease the amount of orange-green compensation required. If compensation proves difficult, the starting CFSE staining intensity may have to be lowered. Once the instrument is set up, the same settings can be used for the duration of the experiment.
9.11.4 Supplement 27
Current Protocols in Cytometry
In the case of adoptively transferred cells, it may be appropriate to collect two groups of files (one to allow enumeration of CFSE-positive cells within the sample; the other to collect CFSE-positive events only) for separate analysis of the transferred population, given that CFSE-positive events may represent <0.5% of total. See Figure 9.11.4 for an example of B cell proliferation after adoptive transfer.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
5-(and -6)-carboxyfluorescein diacetate succinimidyl ester (CFDA-SE or CFSE), 5 mM Dissolve CFSE (mol. wt. 557, Molecular Probes) in dimethyl sulfoxide (DMSO) to a final concentration of 5 mM (2.785 mg/ml). Use gentle pipetting to help CFSE into solution. Divide into convenient (e.g., 50-µl) aliquots and store up to 1 year at −20°C under desiccating conditions, protected from light. COMMENTARY Background Information Both the hematopoietic and immune systems are characterized by a requirement for a large amount of cell division, coupled with differentiation to specialized effector cells. As these systems consist predominantly of free cells, the study of their lineage relationships, cellular division, and differentiation has been difficult. Crude, bulk measures of division and differentiation do not allow the subtleties of such complex systems to be explored. The intracellular dye CFSE is cell permeant and remains nonfluorescent until cellular esterases cleave carboxyl groups from the molecule, rendering it both nonpermeant and fluorescent. In addition, the succinimidyl moiety covalently attaches to amine groups, and thereby contributes to the longevity of staining. During cell division, CFSE is distributed equally between daughter cells. This has been validated by simultaneous quenching of Hoechst dye fluorescence by BrdU to determine cell division (Lyons and Parish, 1994). Depending on the intensity of the starting fluorescence, around eight discrete division cycles can be determined using this standard protocol in hematopoietic cells. The number of divisions which can be followed is limited only by the autofluorescence level of unlabeled cells and the uniformity in size of the labeled cell population. As a dividing cell population approaches this autofluorescence level, the division peaks start to compress, preventing resolution beyond about eight cycles (Hodgkin et al., 1996). The standard protocol is used for tracking division over a period of 2 to 14 days, and is suitable for most in vitro applications, as well
as for in vivo transfers over a similar time frame. On a flow cytometer set up for detecting a bright fluorescein-conjugated antibody, undivided cells will have a fluorescence intensity in the fourth decade on a logarithmic scale. For longer-term tracking, more intense staining may be required. The optimal starting fluorescence intensity will depend primarily on the length of time cells will be tracked and on the inherent resilience of the population of choice to the toxicity associated with intense CFSE staining. In the first day or two after staining, fluorescence intensity drops by about 60% in the absence of division (Fig. 9.11.1), as the more labile constituents binding CFSE are catabolized; thereafter, the intensity remains relatively stable for several weeks to months. This initial decrease does not affect the ability to follow division, as such losses are proportional regardless of cell division. The starting intensity of fluorescence obtained is essentially linear with respect to CFSE concentration (Figure 9.11.2) and duration of incubation with CFSE; therefore, the level of staining may be manipulated to suit the experiment. The CFSE method has been successfully used to determine division-related rules of immunoglobulin isotype switching by murine B lymphocytes (Hodgkin et al., 1996; Hasbold et al., 1998) and in vitro proliferative behavior of T lymphocytes (Wells et al., 1997; Fig. 9.11.3), B lymphocytes (Kindler and Zubler, 1997), and human hematopoietic precursors (Nordon et al., 1997). It can be used to model in vivo phenomena, such as the alloresponse to major histocompatibility complex mismatch (Fig. 9.11.5; Table 9.11.1). The technique has also
Studies of Cell Function
9.11.5 Current Protocols in Cytometry
Supplement 27
More recently, the CFSE cell-division technique has been combined with immunophenotypic analysis of nuclear proteins (Hasbold and Hodgkin, 2000). Cultured cells are stripped of the outer membrane and cytosol before being probed with specific antibodies against nuclear components. The nuclear preparations of proliferating cells show remarkably sharp peaks after division. The alteration to nuclear factor levels can therefore be tracked, allowing examination of potential transcription regulators as cells divide and differentiate. The use of CFSE to track division has not been limited to eukaryotic cells, with at least one group extending its use to the examination of bacterial cell division (Ueckert et. al., 1997).
been used to track the division behavior of cells adoptively transferred in vivo (Lyons and Parish, 1994; Fulcher et al., 1996; Lyons, 1997; Fig. 9.11.4). As an internal control, it is possible to co-inject populations that have been labeled with different intensities of CFSE to allow comparison of survival and division behavior with a control population (Fulcher et al., 1996). This exploits the linearity of staining with respect to concentration of CFSE and duration of staining time (e.g., a four-fold increase in CFSE concentration during the staining procedure gives a four-fold increase in final fluorescence intensity; see Fig. 9.11.2). This approach is especially useful if the more heavily stained population does not undergo division.
70
A
60 50 40 30
Events (counts)
20 10 0 70
B 60
7 6
0
5
50
4 3
40
2
30
1
20 10 0 100
Flow Cytometric Analysis of Cell Division by Dye Dilution
101
102 CFSE fluorescence
103
104
Figure 9.11.5 In vitro alloresponse (mixed lymphocyte response). Murine splenic lymphocytes from BALB/c mice were stained with CFSE and cultured alone (A) or at a 16:1 ratio with C57/Bl6 irradiated, nonlabeled splenocytes (B). After 7 days culture, cells were stained with anti-CD45R-PE, and T cells (CD45R−) were analyzed by flow cytometry. Note division of a proportion of T lymphocytes cultured with allogeneic stimulators, and no division when cultured alone. Numbered peaks correspond to successive cell divisions at day 7. The number of events in each peak can be determined, allowing the frequency of responders in the starting population to be calculated (Table 9.11.1).
9.11.6 Supplement 27
Current Protocols in Cytometry
Table 9.11.1
Peak 0 1 2 3 4 5 6 7
In Vitro Alloresponse (Mixed Lymphocyte Response)a
Mean Ratio fluorescence
Events
Cumulative divisions
1124 513 248 116 59 32 15 8
3533 158 281 358 320 436 718 628
244 330 + 158 = 488 378 + 281 = 659 398 + 358 = 756 476 + 320 = 796 516 + 436 = 952 314 + 718 = 1032 0 + 628 = 628
1.00 0.46 0.22 0.10 0.053 0.029 0.013 0.007
aTable generated from data in Figure 9.11.5B. Note that the fluorescence intensity of peaks
closely follows the predicted serial dilution with each division. The number of events in a given division cycle is halved to calculate the number of division events occurring in the preceding cycle. In the example in bold, 1032 events in the sixth cycle have arisen from 516 divisions in the fifth cycle. This calculation, performed iteratively, allows estimation of the starting proportion of responding cells. In this experiment, T cell response is 244/(3533 + 244), or 6.5%, which is within published estimates of alloresponding precursor percentages obtained using limiting dilution analysis.
Recently, an orange-red-emitting dye with characteristics similar to those of CFSE has been developed (SNARF-1 carboxylic acid, acetate, succinimidyl ester; Molecular Probes) which may be useful where fluorescent probes incompatible with CFSE are required. This reagent does not appear to be excited by 488nm argon lasers as effectively as CFSE, giving a lower intensity of staining, and may be more sensitive to pH variation. Other authors have described the use of the cell membrane–inserting dye PKH-26 to track cell division (Ashley et al., 1993). This dye may be used where a red fluorescent probe is desirable. This reagent is available with a computer program that can be used to deconvolute data to determine numbers of cells in each division cycle, and therefore analyze cellular kinetics (Sigma, Cell Census Plus). This type of analysis is particularly valuable when the starting population of cells exhibits a broad range of staining intensity, which can obscure discrete division peaks. Such computer analysis may also be useful in the interpretation of CFSE data. Some investigators routinely use programs such as ModFit (Verity Software; see Internet Resources), which has the ability to calculate precursor frequency and proliferation indices from listmode data. A program called CFSE Modeler (ScienceSpeak; see Internet Resources) has been specifically designed for analyzing such data, but at the present time it is available only for Macintosh computers. A fine example of how such analyses may be used to develop mathematical models of T cell activa-
tion and differentiation has recently been published (Gett and Hodgkin, 2000). Successive cell cycles can also be determined using a BrdU/Hoechst/ethidium bromide method (Kubbies et al., 1992). The merit of that method is that it can be used with a broad spectrum of cell types, as it is not affected by dye loading variability or toxicity (see Critical Parameters and Troubleshooting). The disadvantages are that only three successive cycles can be followed, that BrdU must be present at relatively high concentrations during the entire period of cell division, and that viable cells cannot be recovered for further functional studies.
Critical Parameters and Troubleshooting A number of recent reviews covering the wide range of applications of the CFSE technique are very useful to those considering applying this analysis to their particular experimental system (Lyons, 1999, 2000; Lyons et al., 2001; Parish and Warren, 2001). In particular, a series of nine review articles examining specific applications was published in the December 1999 issue of the journal Immunology and Cell Biology (Vol. 77, issue 6). Cell viability As with all experiments involving live cells, maintenance of viability is crucial. A number of cell types may be quite sensitive and easily driven into apoptosis. This has been reported to be more of a problem for recently activated
Studies of Cell Function
9.11.7 Current Protocols in Cytometry
Supplement 27
cells. It is important to harvest cells quickly and maintain them on ice before use. Follow standard biohazard precautions when human cells are used. The concentration of protein (0.1% w/v BSA) included in the staining step improves the viability obtained. This is particularly important when using cells sensitive to being driven into apoptosis, such as B lymphocytes. Lower levels of cell aggregation are seen in the presence of protein, which may contribute to the improved viability. This low level of protein does not significantly lower the intensity of CFSE staining obtained. If cell viability is lower than acceptable, a number of approaches may be helpful. Cells can be stained with CFSE on ice or at room temperature instead of at 37°C. In some cases, the lower temperatures result in a marked improvement of cell viability with little lowering of the final intensity of staining. The authors speculate that a similar amount of CFSE is taken up by cells, but activation of esterases only occurs once cells are warmed. This slows down the process of covalent coupling of CFSE until cells are in culture medium, perhaps resulting in less stress to the cells. If the problem persists, it may be necessary to titrate the CFSE stain to determine a level at which acceptable cell survival is obtained, while maintaining sufficient CFSE fluorescence intensity to enable the resolution of an appropriate number of cell divisions. Increasing the amount of protein by adding FBS to the staining procedure may help maintain viability. Because the addition of protein also tends to quench CFSE fluorescence, appropriate protein levels will need to be determined by titration. However, in most cases, the standard protocol results in viabilities exceeding 90%, as measured by exclusion of DNA-staining dyes. It is usually wise, however, to monitor growth and/or function of both CFSE-stained and unstained cells (using standard techniques such as the measurement of tritiated thymidine uptake) to control for any detrimental effects of high levels of CFSE staining.
Flow Cytometric Analysis of Cell Division by Dye Dilution
Uniformity of staining The uniformity of staining will determine the resolution of cell division peaks obtained. Published reports have examined only hematopoietic cells of relatively uniform size. Some anecdotal evidence suggests that human cells give somewhat less uniform peaks than cells from inbred murine strains. Cell suspensions must be mixed promptly after addition of CFSE. Addition of a prediluted 2× CFSE stock
solution (20 µM in PBS/0.1% BSA) to an equal volume of a 2× concentrated cell suspension may improve uniformity of staining. The dye should be diluted immediately before addition to the cell suspension. This may also improve cell viability, as addition of DMSO directly to cell suspensions can sometimes be detrimental. Level of staining It is important to choose an appropriate level of staining for the type of experiment being performed (also see Background Information). Very heavy staining often results in suboptimal cell viability, as well as in making compensation between detecting channels difficult or impossible at early time points after staining. In contrast, understaining limits the number of division cycles that can be resolved. Generally, it is very useful to titrate CFSE staining levels to determine optimal levels for a particular cell type. Immunophenotyping Cells direct from culture, or suspensions made from various lymphoid organs after adoptive transfer of CFSE-stained cells, can be probed with monoclonal antibodies that are detected using appropriate fluorochromes for the cytometer setup being used. A single 488nm argon-ion laser will allow the simultaneous detection of CFSE, an orange fluorochrome such as phycoerythrin (PE), and a red fluorochrome such as PerCP or a tandem dye such as PE/Cy5. This allows determination of division behavior in a subset of cells, or monitoring of phenotypic changes during division. More complex analyses can be performed using a dual- or triple-laser instrument coupled with appropriate fluorochromes and detectors. In general, for analysis of in vitro cultured cells, approximately 106 cells are required for each marker combination. In the case of in vivo transfer experiments, CFSE-positive cells can be rare, often 0.5% to 0.05% of total; therefore, 107 or more cells may need to be stained for each marker combination to obtain sufficient events for analysis, especially in populations of complex immunophenotype. Resolution of peaks In some cases, broad division peaks may be encountered, making accurate identification of division cycles difficult. This is normally due to a heterogeneously sized starting population. In some cases this can be overcome by presorting by flow cytometry. After CFSE staining, a population of cells can be sorted so their fluo-
9.11.8 Supplement 27
Current Protocols in Cytometry
rescence intensity is within a 40-channel interval on a 1024-channel scale (Nordon et al., 1997). This ensures high-resolution tracking of cell division. However, as this will in part result in selection of cells based on volume or size, it may be necessary to establish experimental validity. Some investigators have sorted two 40-channel intervals on either side of the mean fluorescence intensity. Alternatively, data can be analyzed using a two-parameter plot of forward or side scatter versus CFSE fluorescence, which can often aid in the resolution of heterogeneously sized starting populations. Commercially available computer-based deconvolution programs may also be helpful and are referred to in more detail elsewhere in this unit.
earlier time points may give more valuable kinetic information. In vivo cell transfers can be tracked for days to months, depending on the experiment. The length of time that transferred cells survive, as well as their division behavior, determines the duration of tracking. Preparation for flow cytometry requires ∼2 to 3 hours, depending on the complexity of staining with antibodies. For analysis of CFSEpositive events after in vivo transfer of labeled cells, larger numbers of cells need to be prepared. Accordingly, flow analysis time may be substantially greater for in vivo transfer experiments than for in vitro cultures, as CFSE-labeled events can be relatively rare.
Anticipated Results
Literature Cited
Using the standard protocol under optimal conditions, approximately eight division cycles can be identified before autofluorescence obscures division peaks. With a single 488-nm argon-ion laser cytometer, using antibodies conjugated with phycoerythrin and PerCP or tandem PE/Cy5 dyes, it is possible to obtain information on the expression of two phenotypic markers concurrently with cell division cycle. A more sophisticated dual- or triple-laser instrument will allow a substantially more complex immunophenotypic study coupled with cell division determination. Cells labeled to a high intensity with CFSE can be tracked in vivo for several months (Lyons, 1997). The behavior of the cell population under study will determine how long cells can be followed. Rapid cell decay and/or division will limit the time that cells can be tracked.
Ashley, D.M., Bol, S.J., Waugh, C., and Kannourakis, G. 1993. A novel approach to the measurement of different in vitro leukaemic cell growth parameters: The use of PKH GL fluorescent probes. Leukemia Res. 17:873-882.
Time Considerations The preparation of murine or human cell suspensions requires 30 to 90 min, depending on the source and on the quantity required. Staining with CFSE takes 30 to 45 min, including washes to remove free CFSE. Typically, cell cultures are from 1 to 7 days duration, depending on the kinetics of division responses. In many cases, division is extremely rapid over a short interval. For example, B cell cultures stimulated via CD40 and cytokines start dividing around day two, and have undergone up to eight cycles by day four (Hodgkin et al., 1996). The choice of time points for in vitro proliferation studies needs to be carefully considered. The peak proliferation time determined by tritiated thymidine uptake may not be appropriate for the CFSE method. In contrast,
Fulcher, D.A., Lyons, A.B., Korn, S., Cook, M.C., Koleda, C., Parish, C., Fazekas de St. Groth, B., and Basten, A. 1996. The fate of self-reactive B-cells depends primarily on the degree of antigen receptor engagement and availability of Tcell help. J. Exp. Med. 183:2313-2328. Gett, A.V. and Hodgkin, P.D. 2000. A cellular calculus for signal integration by T cells. Nature Immunol. 240:75-244. Hasbold, J. and Hodgkin, P.D. 2000. Flow cytometric cell division tracking using nuclei. Cytometry 40:230-237. Hasbold, J., Lyons, A.B., Kehry, M.R., and Hodgkin, P.D. 1998. Cell division number regulates IgG1 and IgE switching of B cells following stimulation by CD40L and IL-4. Eur. J. Immunol. 28:1040-1051. Hodgkin, P.D., Lee, J.H., and Lyons, A.B. 1996. B cell differentiation and isotype switching is related to division cycle number. J. Exp. Med. 184:277-281. Kindler, V. and Zubler, R.H. 1997. Memory, but not naive, peripheral blood B lymphocytes differentiate into Ig-secreting cells after CD40 ligation and costimulation with IL-4 and the differentiation factors IL-2, IL-10, and IL-3. J. Immunol. 159:2085-2090. Kubbies, M., Goller, B., and Van Bockstaele, D.R. 1992. Improved BrdU-Hoechst bivariate cell kinetic analysis by helium-cadmium single laser excitation. Cytometry 13:782-786. Lyons, A.B. 1997. Pertussis toxin pretreatment alters the in vivo division behaviour and survival of B lymphocytes, after intravenous transfer. Immunol. Cell Biol. 75:7-12. Studies of Cell Function
9.11.9 Current Protocols in Cytometry
Supplement 27
Lyons, A.B. 1999. Divided we stand: Tracking cell proliferation with carboxyfluoresceindiacetate succinimidyl ester. Immunol. Cell Biol. 75:509515. Lyons, A.B. 2000. Analysing cell division in vivo and in vitro using flow cytometric measurement of CFSE dye dilution. J. Immunol. Methods 243:147-154. Lyons, A.B. and Parish, C.R. 1994. Determination of lymphocyte division by flow cytometry. J. Immunol. Methods 171:131-137. Lyons, A.B., Hasbold, J. and Hodgkin, P.D. 2001. Flow cytometric analysis of cell division history using dilution of carboxyfluoresceindiacetate succinimidyl ester, a stably integrated fluorescent probe. Meth. in Cell Biol.. 63:375-398. Nordon, R.E., Ginsberg, S.S., and Eaves, C.J. 1997. High-resolution cell division tracking demonstrates the Flt3-ligand-dependence of human marrow CD34+CD38− cell production in vitro. Br. J. Haematol. 98:528-539. Parish, C.R. and Warren, H.S. 2001. Use of the intracellular fluorescent dye CFSE to monitor lymphocyte migration and proliferation. In Current Protocols in Immunology (J.E. Coligan, A.M. Kruisbeek, D.H. Margulies, E.M. Shevach, and W. Strober, eds.) pp 4.9.1-4.9.10. John Wiley & Sons, New York.
Internet Resources http://www.probes.com The web site of Molecular Probes, the supplier of CFSE/CFDA-SE. This site contains much useful information on the application of the CFSE technique, and information on the orange-red emitting alternative dye SNARF-1 carboxylic acid, acetate, succinimidyl ester. http://www.vsh.com The web site of Verity Software House. ModFit software is used by many researchers to analyze proliferation data; it is available for both PC and Macintosh computers. http://www.sciencespeak.com The web site of ScienceSpeak, supplier of CFSE Modeler, which is available for Macintosh computers.
Contributed by A. Bruce Lyons Institute of Medical and Veterinary Science Adelaide, South Australia Kathleen V. Doherty Australian Red Cross Blood Service Adelaide, South Australia
Wells, A.D., Gudmundsdottir, H., and Turka, L.A. 1997. Following the fate of individual T cells throughout activation and clonal expansion. Signals from T cell receptor and CD28 differentially regulate the induction and duration of a proliferative response. J. Clin. Invest. 100:3173-3183.
Flow Cytometric Analysis of Cell Division by Dye Dilution
9.11.10 Supplement 27
Current Protocols in Cytometry
Reporters of Gene Expression: Autofluorescent Proteins
UNIT 9.12
Since the first successful demonstration in 1994, green fluorescent protein (GFP) and its derivatives have become arguably the most popular reporters of heterologous gene expression (Chalfie et al., 1994). This “green revolution” (Stearns, 1995) may be attributed to: (1) the convenience of avoiding cell manipulations, staining, or substrate loading prior to direct detection; (2) the widespread availability of convenient filter sets that facilitate multiparameter flow cytometric or microscopic analyses; (3) the successful production of many autofluorescent fusion proteins allowing real-time monitoring and subcellular localization in live cells; and (4) the exceptional physical properties of GFP and its derivatives, which permit autocatalytic fluorescence, without cofactors, in a variety of organisms from prokaryotes to eukaryotes without apparent toxicity. GFP, and its red-shifted variants in particular, are convenient reporters in many cytometric applications (Ropp et al., 1995, 1996). As such, the purpose of this unit is to provide sufficient background and methodologies so the reader may apply this relatively new technology to experimental systems of interest. Two basic protocols are provided. Basic Protocol 1 utilizes red-shifted GFP as a reporter of successful gene transfer by retroviral infection. Using this straightforward and time-efficient assay, transduction conditions may be optimized and cell subsets may be sorted based on gene expression. With only minor modification, the protocol may be applied whenever conditions for introduction of DNA into mammalian cells require examination through detection of gene expression. In Basic Protocol 2, red-shifted GFP is used in bivariate analysis for studies of cell cycle. Correlation of GFP expression levels with DNA content, especially when GFP is fused to a protein of interest, opens diverse opportunities for investigators. Not only can the population be divided into GFP-positive and GFP-negative subsets to reveal effects on cell cycle, but dosage effects within the GFP-positive subset may also be examined. It is important that this unit be read in conjunction with UNIT 9.5 (“Reporters of Gene Expression: Enzymatic Assays”), which supplies further background. STRATEGIC PLANNING Many issues surrounding the use of GFP and its derivatives are the same as those for other reporter genes (UNIT 9.5). The first issue is to identify the aims and questions underpinning the project, since GFP is not always the most appropriate reporter. For example, GFP and its derivatives are not as readily detected as enzymatic reporters, which may provide much greater sensitivity on a per-molecule basis compared to wild-type GFP (Niswender et al., 1995). Also, the half-life of GFP in cells is usually ≥24 hr, making its use as a direct reporter of active-transcription kinetics less attractive than alternatives. For this reason, fusion proteins containing codon-optimized mutants of GFP may lead to the artificial stabilization and false concentration of such proteins in subcellular compartments. However, even this problem may now be avoidable with the use of unstable or destabilized GFP variants. Questions that may be well addressed using cytometric analysis of GFP, and, in particular, commercially available red-shifted variants of GFP fused to proteins of interest, include the following. 1. What is the efficiency of gene transfer into specific cells and how can the conditions of transfer be optimized? 2. What is the relative level of expression in different organisms, tissues, cell lines, or lineages? Contributed by John E.J. Rasko Current Protocols in Cytometry (1999) 9.12.1-9.12.16 Copyright © 1999 by John Wiley & Sons, Inc.
Studies of Cell Function
9.12.1 Supplement 7
Table 9.12.1
Selected Variants of Native GFP
GenBank accession no. Absorbance or plasmid maximum (nm)
Emission maximum (nm)
Wild-type GFP
U17997 M62654
395 (and weaker at 478)
508
Chalfie et al. (1994) Prasher et al. (1992)
S65Ta (red-shifted) Y66Ha (blue) T203Ya (yellow-green)
U50963 U19276 pEBFP pQBI50-BFP pEYFP pGFPtpz
489
511
382 or 380b 387 514
448 or 440b 450 527
Zolotukhin et al. (1996) Cubitt et al. (1995) Heim and Tsien (1996) Palm et al. (1997) Heim and Tsien (1996) Ormö et al. (1996)
Y66Wa (blue-green)
pECFP
433 and 453c
475 and 501c
Variant
Reference
Heim and Tsien (1996) Ormö et al. (1996)
aAmino acids altered at the position indicated, using single-letter code. Additional alterations in amino acid residues may be found in the references
indicated. bDifferent authors cite minor differences in maxima. cExcitation at both wavelengths is feasable.
3. How is the relative level of expression altered following manipulation of cis- or trans-acting elements? 4. Can cell subsets expressing different amounts of GFP be identified and selected to reveal altered expression of other proteins of interest in multiparameter analyses? 5. Does overexpression cause perturbations in cell viability, turnover, apoptosis, passage through cell cycle, or, indeed, any other indicator of cell physiology for which a cytometric assay is available? 6. Can analysis of fluorescence resonance energy transfer between a donor and acceptor pair of autofluorescent proteins be used to report specific cellular events? Which Autofluorescent Protein to Use? After wild-type GFP (called gfp10) was cloned, a number of laboratories experienced disappointing results as a consequence of their inability to detect its relatively poor fluorescence in comparison to FITC using the same filter sets (Prasher et al., 1992; Chalfie et al., 1994). Factors contributing to initial failures included inhibitory sequences in the 5′-untranslated region (Chalfie et al., 1994), cryptic splice regions recognized by plants (Reichel et al., 1996; Haseloff et al., 1997), a poor Kozak consensus sequence in the translation-initiation region, lower translational efficiency due to the presence of codons found infrequently in highly expressed mammalian genes (Haas et al., 1996), delayed chromophore formation due to the need for post-translational oxidation (Heim et al., 1994; Pines, 1995), and the use of filter sets optimized for FITC rather than specific GFP-optimized filters.
Reporters of Gene Expression: Autofluorescent Proteins
Many of the problems surrounding the expression and detection of GFP have been surmounted. Changes in amino acids in the region of the chromophore have created many new autofluorescent proteins with altered spectral properties (Table 9.12.1; also see Delagrave et al., 1995; Ehrig et al., 1995; Heim et al., 1995; Cormack et al., 1996; Crameri et al., 1996; Heim and Tsien, 1996). These spectral variants of GFP offer a lavish array of excitation and emission choices and are tempting to use in multiparameter analyses. However, the most dependable and widely used autofluorescent molecules detected using flow cytometry of mammalian cells are the red-shifted variants of GFP (Table 9.12.2) and
9.12.2 Supplement 7
Current Protocols in Cytometry
Table 9.12.2
Commercially Available, Improved, Red-Shifted GFP Plasmidsa
Plasmid
Company
Product Line
pEGFP PQBI25 pGreen Lantern pGFPemd-c
Clontech Quantum Biotechnology, CPG Life Technologies Packard Instrument
Living Colors AFPs Green Lantern CytoGem
aAny of these red-shifted GFP variants may be detected simultaneously with blue-fluorescing mutant GFP
(Y66H) on the flow cytometer using a 475-nm dichroic long-pass filter to separate green and blue fluorescence signals. Dual-laser excitation (at 407 nm and 488 nm) and use of laser-line-blocking filters is optimal (Anderson et al., 1996; Ropp et al., 1996). Simultaneous detection of red-shifted GFP variants and yellow-green fluorescent protein (T203Y) has been achieved using 488-nm excitation and a dichroic short-pass 525-nm filter to direct detection using 510/20 and 550/30 emission filters.
the blue fluorescent proteins (BFPs). For experiments in bacteria or yeast, optimized red-shifted GFP (Kahana et al., 1995) and UV-excited GFP are available (Crameri et al., 1996). The choice of which autofluorescent molecule to use in a particular system may be influenced by many factors. Clearly, if the autofluorescent reporter is being used in bivariate analysis alongside an antibody which has been conjugated to a fluorochrome, then the signals must be distinguishable. Red-shifted GFPs cannot readily be used in conjunction with FITC, as they share similar spectra. Even yellow-green variants of GFP may be difficult to resolve from FITC or red-shifted GFP signals in the flow cytometer equipped with standard filters, as a consequence of their overlapping spectra (Lybarger et al., 1998). Another factor in choosing which GFP variant to use is the potential for induction of mutations in cells exposed to UV irradiation. Avoid BFPs if repeated interrogation of live cells is planned. The faster photobleaching of BFPs may be an additional reason to prefer the more photostable, red-shifted GFPs (Heim and Tsien, 1996; Patterson et al., 1997). For experiments involving quantitation of gene expression, red-shifted GFPs are preferred over wild-type GFP, as the former do not exhibit a concentration-dependent reduction in chromophore absorption around 488 nm (Ward and Bokman, 1982). Generally the red-shifted GFPs offer much greater sensitivity as reporters of gene expression as compared to wild-type GFP. However, detection limits of any autofluorescent protein may be constrained by the autofluorescence intrinsic to a specific cell. Reduction of cellular autofluorescence by growing cells in medium lacking flavoproteins has been reported (Benson et al., 1979; Zylka and Schnapp, 1996). For any given cell type, different GFP variants may express at higher levels above background, and some testing may be warranted if the way has not been previously paved by other authors. In light of the frequent use of GFP fusion proteins targeted to various subcellular regions and organelles, the effect of pH on fluorescence should be recognized. Organelles in both the endocytic and exocytic pathways may be highly acidified, with pH as low as 4.6 (Mellman et al., 1986). Whereas wild-type GFP loses fluorescence only when pH is below 4 or above 12 (Bokman and Ward, 1981), red-shifted variants are more sensitive to pH, with highest fluorescence retained between pH 7.0 and 11.5 (Patterson et al., 1997). This sensitivity of red-shifted variants may be used to advantage when local pH is noninvasively reported by targeting GFPs to specific subcellular regions (Kneen et al., 1998; Llopis et al., 1998). Vector Use and Construction A common application of GFP or its variants is as a rapid and convenient positive control for successful gene expression—e.g., to compare different transfection protocols. In this
Studies of Cell Function
9.12.3 Current Protocols in Cytometry
Supplement 7
case, any one of the burgeoning number of commercially available expression vectors (see Table 9.12.2) may be all that is required. However, beyond straightforward optimization experiments, the use of GFP and variants as reporters in alternative vectors or as fusion proteins will require manipulation of the plasmids themselves. As detailed subcloning strategies are beyond the scope of this unit, the reader is referred to Chapter 9 in Ausubel et al. (1998). Many dozens of publications attest to the utility of GFP fusion proteins. GFP and/or its variants have been targeted to the nucleus (Davis et al., 1995), mitochondria (Rizzuto et al., 1996), peroxisome (Kalish et al., 1996), Golgi apparatus (Cole et al., 1996), endoplasmic reticulum (Haseloff et al., 1997), and plasma membrane (Moriyoshi et al., 1996). By concentrating GFP in a specified region, higher detection sensitivity may be achieved (Niswender et al., 1995; Haseloff et al., 1997). Furthermore, construction of vectors that create in-frame fusions to GFP at the amino- or carboxy-terminus of the chimeric molecule have, surprisingly, led to proteins that apparently retain their normal functions (Flach et al., 1994; Marshall et al., 1995; Moores et al., 1996). A number of commercially available vectors have been designed with convenient restriction sites in the three reading frames to facilitate construction of fusion proteins (e.g., from Clontech, PharMingen, and Quantum Biotechnologies). Still, many fusion proteins have been successfully created using short linkers, PCR-introduced restriction sites, overlap-extension PCR (Horton et al., 1993), or a blunt-ended PCR approach (Lo et al., 1998). GFP fusion proteins have become indispensable tools for the study of diverse intracellular processes, including protein dynamics, subcellular localization, and cell cycle. GFP variants have been included as reporters in experiments using fluorescence resonance energy transfer (FRET), fluorescence recovery after photobleaching (FRAP; Ellenberg et al., 1997; Subramanian and Meyer, 1997), and fluorescence loss in photobleaching (FLIP; Cole et al., 1996). FRET occurs when radiative energy (not photons) is transferred from a short wavelength (donor) to a longer wavelength (acceptor). The energy transfer is highly sensitive to orientation and distance between the two fluorophores (dos Remedios and Moens, 1995). GFP variants linked covalently through a protein that changes conformation in response to its molecular environment may be used as FRET partners (Mitra et al., 1996). For example, apoptosis has been monitored in cells that express a plasmid containing GFP linked via a caspase recognition site to BFP (Xu et al., 1998). Changes in intracellular calcium concentration have been observed in cells expressing cyan-GFP-labeled calmodulin and yellow-GFP-labeled calmodulin-binding peptide (Miyawaki et al., 1997; Tsien and Miyawaki, 1998). Many more protein-protein interactions will likely be amenable to FRET approaches, opening the application of high-throughput screening protocols to both in vivo and in vitro systems. Targeting of fusion molecules to subcellular regions for FRET will further expand experimental opportunities (Selvin, 1995). BASIC PROTOCOL 1
Reporters of Gene Expression: Autofluorescent Proteins
USE OF RED-SHIFTED GFP TO QUANTITATE AND SELECT RETROVIRUS-TRANSDUCED CELLS BY FLOW CYTOMETRY This protocol describes the use of a red-shifted mutant of GFP to report, quantify, and select cells following successful retroviral transduction. The method as presented makes use of the EGFP variant of GFP (Clontech), although other red-shifted mutants (Table 9.12.2) will perform similarly. This method (after step 5) may also be used to optimize conditions for introduction of heterologous genes using transfection, electroporation, or infection into cells of interest. The principle of the assay is that once cells are infected, expression of red-shifted GFP should be readily detected above background in live cells so that quantitation and selection of positives may be performed.
9.12.4 Supplement 7
Current Protocols in Cytometry
Materials Adherent murine NIH/3T3 cells (ATCC #CRL 1658) Culture medium: e.g., complete DMEM medium containing 10% FBS (APPENDIX 2A) 4 mg/ml polybrene in PBS (1000× stock; filter through 0.2-µm filter and store up to several months at 4°C; see APPENDIX 2A for PBS) Virus-containing medium (fresh or thawed from storage at –70°C): in this example pLNCG vector (Rasko et al., 1997), which contains EGFP in the pLNCX vector (Miller and Rosman, 1989); see Cepko (1995) for general procedure Wash solution: PBS (APPENDIX 2A) containing 2% (v/v) FBS (APPENDIX 2A) 1 mg/ml propidium iodide (1000× stock; store up to several months at 4°C) 60-mm tissue culture dishes 12 × 75–mm round-bottom polystyrene tubes (clear plastic facilitates verification of cell pellets) Refrigerated centrifuge Flow cytometer or sorter equipped with 488-nm argon-ion laser Additional reagents and equipment for cell culture and trypsinization of cells (APPENDIX 3B) NOTE: All incubations are performed in a humidified 37°C, 5% CO2 incubator unless otherwise specified. 1. Plate adherent NIH/3T3 cells in 60-mm tissue culture dishes containing DMEM/10% FBS at 5 × 105 cells/plate (prepare enough plates for assay and controls). Begin incubation (see APPENDIX 3B for cell culture technique). Typically, cells should be plated to achieve ∼50% confluency on the day of transduction to ensure active division during exposure to virus. For optimal transduction of nonadherent cells, protocols employing cocultivation with irradiated packaging lines (Pear, 1996), “spinoculation” (alternatively called spin infection; also described in Pear, 1996), or coating of petri dish with, e.g., fibronectin fragments (Hanenberg et al., 1996) are often preferred.
Day 0: transduce cells 2. Immediately before use, prepare transduction medium (2 ml for each dish to be transduced) consisting of polybrene (4 µg/ml final concentration added from 1000× stock), virus-containing medium (comprising ≤50% for the final transduction medium volume), and culture medium to make up the remaining volume. To minimize toxicity to cells, the virus-containing medium should contribute no more than half the final transduction medium volume.
3. Replace culture medium with transduction medium and return dishes to incubator for 3 to 18 hr. Increasing incubation time may improve transduction efficiency, with a consequent increase in cell toxicity. “No-virus” negative controls should include polybrene. It is convenient to commence transduction in the late afternoon and remove the transduction medium next morning. Infection of some cell types at 32°C has been reported to improve transduction efficiency (Kotani et al., 1994).
Day 1: incubate cells 4. Replace transduction medium with culture medium and return dishes to incubator. The duration of incubation following exposure to virus depends on the time interval required for EGFP expression to reach a peak. Although the first EGFP-expressing cells may be detected within 6 hr of transduction (earlier than wild-type GFP), an accurate estimate of the total percentage of cells transduced is best quantified no earlier than 48 hr
Studies of Cell Function
9.12.5 Current Protocols in Cytometry
Supplement 7
B 104 Red fluorescence (>600 nm)
Red fluorescence (>600 nm)
A 104 103 102 101 100
101 100
C 104
D 104 Red fluorescence (>600 nm)
15 nm)
102
Red fluorescence (>600 nm)
Green fluorescence (530
103
103 102 101 100
E Green fluorescence (530
15 nm)
Green fluorescence (530
15 nm)
103 102 101 100
15 nm)
E 2850
Green fluorescence (530
F 104 M1
Counts
Counts
103
R2
102 101
0 100
Reporters of Gene Expression: Autofluorescent Proteins
101 102 103 Green fluorescence (530 15 nm)
104
100 100
101 102 103 Green fluorescence (530 15 nm)
104
Figure 9.12.1 Application of optimal regions to assess the positive population and at the same time minimize the contribution of autofluorescent cells will be amply rewarded (Anderson et al., 1996). The first four panels show flow cytometric analyses using dot plots of NIH/3T3 control cells (A and B) or the same cells transduced with a EGFP-containing vector (C and D) with red/green compensation set to 0% (A and C) or 15.0% (B and D). The four populations (EGFP–PI–, EGFP+PI–, EGFP–PI+, and EGFP+PI+) are more clearly resolved in (D) as compared to (C). Note that despite adequate spectral compensation for most cells shown in (D), some very bright cells expressing EGFP remain falsely “PI-positive” (in the green overflow channel and therefore not compensatable). This is important, since indifferent application of gating in only the red channel to exclude PI+ cells would inadvertently remove true EGFP+ events. The histogram (E) and dot plot (F) represent the same data file shown in (D) with gating to exclude true PI+ cells. Both M1 and R2 were placed to give 0.01% GFP+ events from control data (B). However, the percentage EGFP+ in R2 is higher (28.93%) compared to M1 (27.18%) due to the failure of the single-parameter analysis to include true EGFP+ cells present in the left-most area of R2. By applying a two-parameter analysis (F), autofluorescent events occurring in the red channel may be excluded, whereas the one-parameter analysis (E) could fail to detect a population of dull EGFP+ events. This effect will be more pronounced in samples in which the true EGFP+ cells exhibit a lower fluorescence intensity.
9.12.6 Supplement 7
Current Protocols in Cytometry
after exposure to virus-containing medium (specifics will depend on the copy number, promoter, site of integration, and cell line used).
Day 3: assay for GFP expression in transduced cells by flow cytometry 5. Trypsinize cells (APPENDIX 3B) and transfer cells at 0.5–1.0 × 106 cells per 12 × 75–mm polystyrene tube. Centrifuge 5 min at 300 × g, 4°C, then remove supernatant. 6. Resuspend cells in 0.5 ml of 4°C wash solution containing PI (1 µg/ml added from 1000× stock) prior to analysis. Cells are best analyzed immediately, although a delay of up to 3 hr may be tolerated if the cells are stored in the dark at 4°C or on ice.
7. Set up flow cytometer. Detection of EGFP by flow cytometry is performed using standard settings for FITC and PI. Typically a 15-mW argon laser exciting at 488 nm is used in combination with 530 ± 15–nm band-pass and >600-nm long-pass emission filters (depending on cytometer sensitivity). Neutral-density filtering may be used if adequate spectral compensation of maximum GFP fluorescence cannot be achieved (see below).
8. Optimize EGFP fluorescence detection by applying compensation for spectral overlap. Acquire data. After voltage gains have been set to place signals optimally within the GFP and PI channels, fluorescence compensation should be carefully selected to minimize signal from the EGFP contributing to the red channel (see discussion of autofluorescence in GFPtransfected cells in the Critical Parameters section of UNIT 9.5; also see Fig. 9.12.1). If very bright GFP-positive cells remain apparently “PI-positive” despite optimized compensation and use of neutral-density filtering, replacing PI with 7-aminoactinomycin D (7-AAD) as a marker of dead cells may improve results (see Critical Parameters and Troubleshooting in this unit).
9. Analyze data to optimally discriminate negative and positive populations using “no-virus” control plates for reference as shown in Figure 9.12.1. Quantitation of live cells expressing EGFP may be indicated as: (1) a percentage of cells with fluorescence above a specified threshold (e.g., by setting a region in which ≤0.01% of live “no-virus” control cells are positive), (2) a ratio between median or mean fluorescence intensities from histograms, or (3) a median or mean of the fluorescence difference, after subtracting the control from the experimental histogram. The valid use of these methods will depend on the actual data.
10. Optional: Apply live gates to sort populations of interest. It is useful to apply a gate (e.g., on pulse width of forward light scatter) using the pulse-processing option to minimize doublets (technique will vary depending upon the instrument). Sterile sorted cells should be returned to culture medium and then to the incubator as soon as possible.
BIVARIATE ANALYSIS OF RED-SHIFTED GFP FOR CELL-CYCLE STUDIES This protocol describes the simultaneous detection of red-shifted GFP expression and DNA content revealed by propidium iodide (PI). Initially, the protocol may be used to address the question of whether the level of GFP expression per se may alter the distribution of cells within specific phases of the cell cycle in particular cell types. Once it has been established that GFP is not toxic, highly informative experiments may subsequently be performed in which expression of genes of interest fused to the GFP gene may be correlated with the cell cycle.
BASIC PROTOCOL 2
Studies of Cell Function
9.12.7 Current Protocols in Cytometry
Supplement 7
Cells are fixed in a solution that does not destroy EGFP autofluorescence, rinsed with PBS, and stained with PI in a solution containing detergent and RNase. Data are acquired on the flow cytometer using the same filter sets described in Basic Protocol 1. Materials EGFP-expressing cells Paraformaldehyde solution (see recipe), 4°C Wash solution: PBS (APPENDIX 2A) containing 2% (v/v) FBS (APPENDIX 2A) PI cell-cycle staining solution (see recipe) 12 × 75–mm round-bottom polystyrene tubes (clear plastic facilitates verification of cell pellets) Refrigerated centrifuge Flow cytometer or sorter equipped with a 488-nm argon-ion laser Additional reagents and equipment for cell culture and trypsinization of cells (APPENDIX 3B) 1a. For adherent cells: Trypsinize ∼1 × 106 cells expressing EGFP alone or as a fusion protein (see APPENDIX 3B for trypsinization technique) and transfer into 12 × 75–mm tubes on ice. 1b. For cells in suspension culture: Directly harvest ∼1× 106 cells into 12 × 75–mm tubes on ice. 2. Centrifuge cells 5 min at 300 × g, 4°C. To prevent artifactual changes in cell cycle, do not pause between this step and step 5.
3. Remove the supernatant (preferably by aspiration) and then agitate the tube to disperse the cells into the residual medium. If the pellet is not resuspended, clumping and ineffective permeabilization of cells may result.
4. Resuspend cells in 1 ml of 1% paraformaldehyde solution at 4°C, added dropwise while vortexing. Incubate 20 min on ice. 5. Centrifuge cells 5 min at 300 × g, 4°C, decant supernatant, then thoroughly resuspend cells in 4°C wash solution and leave on ice for 10 min. At this point cells are fixed and may be stored at 0° to 4° for several weeks without loss of GFP fluorescence.
6. Centrifuge cells 5 min at 300 × g, 4°C, decant supernatant, then thoroughly resuspend cells in 0.25 ml wash solution. 7. Add 0.25 ml PI cell-cycle staining solution and incubate 30 min at 37°C in the dark. Cells may be analyzed immediately after incubation or stored at 4°C in the dark prior to analysis.
8. Perform flow cytometry.
Reporters of Gene Expression: Autofluorescent Proteins
Detection of EGFP is performed with standard settings for FITC presented on a logarithmic scale. Detection of PI for cell cycle is performed with standard emission detection at red wavelengths (e.g., >600 nm long-pass filter) presented on a linear scale. Flow rates should be low. Many operators syringe cells several times through a 21-G needle to minimize clumping. Acquisition of a large number of events will facilitate precise subset analysis within EGFP expression levels. Pulse-area and -width parameters in the red channel should be acquired
9.12.8 Supplement 7
Current Protocols in Cytometry
to facilitate exclusion of doublets by gating during analysis (technique will vary depending upon the instrument) . Careful attention to compensation settings must be made so that cells expressing GFP in specific phases of cell cycle (e.g., G1 peak) are parallel to the GFP axis. If this is not achieved, then the coefficient of variation in the G1 peak will be suboptimal for analysis of the two-parameter histogram (see Fig. 9.12.2).
9. Perform bivariate analysis with appropriate software to deconvolute and quantify cell cycle phases in relation to EGFP expression (see Fig. 9.12.2 and UNIT 7.9).
400
0 450
103
R5
102
R4
0
R3
600
R2
101
50 100 150 200 Red fluorescence (>600 nm)
0 250
Counts
Green fluorescence (530 15 nm)
104
0
0 250 Red fluorescence (>600 nm)
Figure 9.12.2 Bivariate analysis of nuclear-targeted EGFP-fusion protein and cell cycle. The density plot at left shows pulse area from PI signal on the linear x axis and EGFP fluorescence on the logarithmic y axis. Doublet signals have been minimized by gating on pulse-width versus pulse-area parameters. Application of parallelogram-shaped regions R2 to R5 permits cell-cycle analysis of subsets based on EGFP expression shown in histograms at right.
Studies of Cell Function
9.12.9 Current Protocols in Cytometry
Supplement 7
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions see APPENDIX 2A; for suppliers see SUPPLIERS APPENDIX.
Paraformaldehyde solution Wearing a face mask and gloves, add 0.5 g paraformaldehyde (Sigma) to 45 ml water. Add 4 µl of 10 N NaOH and warm to 65°C. When all solid has dissolved (usually ≤30 min) add 5 ml 10× PBS (see APPENDIX 2A for 1× recipe), and confirm pH is ∼7.4. Filter solution through a 0.20-µm filter to remove any debris. This solution is best prepared on day of use.
PI cell-cycle staining solution 1 ml 1 mg/ml propidium iodide (PI) 9 ml phosphate-buffered saline (PBS; APPENDIX 2A) 10 µl Triton X-100 (Sigma; 0.1% v/v final) 1 mg DNase-free RNase (Boehringer Mannheim) Store up to 1 month at 4°C in the dark COMMENTARY Background Information
Reporters of Gene Expression: Autofluorescent Proteins
General considerations In recent years green fluorescent protein (GFP), cloned from the bioluminescent jellyfish Aequorea victoria, has become a widely applied reporter of gene expression and protein localization. The existence of green fluorescent proteins had been suggested around 1970 when discrepancies between the in vivo and in vitro bioluminescent spectra of coelenterates implied an energy-transfer phenomenon from the initial calcium-activated phosphoprotein (Morin and Hastings, 1971a,b; Wampler et al., 1971). It is now known that activated aequorin in the jellyfish excites endogenous GFP, leading to emission of green light. However neither aequorin nor any other factor specific to jellyfish is required for this phenomenon, as evidenced by the fact that purified native jellyfish GFP will produce green light in vitro following direct excitation with blue light. Indeed, there are no known additional enzymes, substrates, or cofactors required for this process other than aerobic conditions (Heim et al., 1994; Inouye and Tsuji, 1994a). Furthermore the active chromophore which emits green light is a cyclic tripeptide suspended in the middle of a rigid cylindrical structure for which the term “β-can” has been coined (Ormö et al., 1996; Yang et al., 1996). The protection afforded to the chromophore by the β-can may explain why GFP expressed in a variety of in vitro conditions exhibits the same fluorescent spectrum. Specifically, wild-type GFP absorbs blue light maximally at 395 nm and 478 nm and emits
green light at 508 nm (see Table 9.12.1). Although less efficient absorption occurs at 478 nm, the resulting green fluorescence is more photostable than emission resulting from excitation at 395 nm. The cloning and heterologous expression of wild-type GFP (Prasher et al., 1992; Chalfie et al., 1994; Inouye and Tsuji, 1994b) facilitated studies by academic and industrial investigators who showed that modification of its amino acid sequence (particularly in the tripeptide chromophore S65 Y66 G67) could alter its fluorescent properties (see Table 9.12.1). Not only were the extinction coefficient and excitation and emission spectra altered in GFP variants, but mutations were also introduced to improve protein stability, translational efficiency, and conformational maturation of the chromophore (Heim et al., 1995; Anderson et al., 1996; Cormack et al., 1996; Brejc et al., 1997). These highly expressed GFP variants offered reporters of much greater sensitivity for microscopists and cytometrists. Still, the fluorescence quantum yields of wild-type GFP and red-shifted variants are very similar (Patterson et al., 1997). Thus the wild-type GFP of only 238 amino acid residues (mol. wt. 27 kDa) has spawned a large number of derivative molecules with applicability in a number of experimental systems. Utility of GFP and its variants as reporters of gene expression Since making its début as a reporter of gene expression in bacteria and worms (Chalfie et al., 1994), GFP and its variants have been ex-
9.12.10 Supplement 7
Current Protocols in Cytometry
pressed in a remarkably diverse range of organisms. The creation of transgenic GFP models in plants (Baulcombe et al., 1995), yeast (Flach et al., 1994), mosquito (Higgs et al., 1996), Drosophila (Wang and Hazelrigg, 1994; Barthmaier and Fyrberg, 1995), zebrafish (Amsterdam et al., 1995), Dictyostelium discoideum (Maniak et al., 1995), Xenopus (Tannahill et al., 1995), and mouse (Ikawa et al., 1995) attests to their utility. Furthermore, expression of redshifted GFP in retroviruses used in hemopoietic reconstitution assays (Persons et al., 1997; Rasko et al., 1997) and many other examples has enabled tracking of mature cells of different phenotypes which arose from common precursors. Additionally, GFP has been used successfully as a reporter of viral transduction in a number of systems (Cheng et al., 1996; Higgs et al., 1996; Levy et al., 1996), although one notable failure was reported due to possible toxicity of GFP in a retrovirus packaging cell line (Hanazono et al., 1997). Transduction rates reported by red-shifted GFPs parallel results using enzymatic and antibiotic-resistance markers with equal linearity. One important benefit seen with red-shifted GFP is the apparent absence of “pseudotransduction” (Liu et al., 1996; Alexander et al., 1997), an artifact seen when enzymatic reporter proteins—such as alkaline phosphatase (AP) or β-galactosidase— are incorporated into virions (J. Rasko, unpub. observ.). Utility of GFP as a selectable marker GFP and its variants are very convenient selectable markers. Selection of GFP-expressing cells by sorting avoids the use of potentially toxic drug selections, enables rapid selection at multiple time points, allows simultaneous selection of both positive-expressing and negative cells, and provides opportunities for multiparameter selection (see UNIT 9.5 for further discussion). However, the ability to detect a fluorescent signal from GFP in a given cell is greatly influenced by the activity of the promoter used in the expression vector. Although there are advantages in using constructs that contain an internal ribosomal entry site (IRES) to express GFP from a single transcript containing a gene of interest, reduced expression of the reporter may compromise sensitivity (Zhou et al., 1998). Bicistronic vectors that produce separate transcripts from different promoters may be helpful in such situations (Miller and Rosman, 1989).
Critical Parameters and Troubleshooting Failure to account for the autofluorescence of cells may lead to poor signal above background or underestimation of true positives. Discrimination of endogenous autofluorescence in cells versus signal from GFP and its variants may be achieved using the approaches described in Figure 9.12.1 and UNIT 9.5. A major contribution of autofluorescence in cells comes from the fluorescent spectra of riboflavin and flavinoids, which are similar to those of wildtype and red-shifted GFPs. Punctate autofluorescence may arise from NADH associated with mitochondria (Aubin, 1979); however, this autofluorescence may be reduced by exciting at 488 nm (Niswender et al., 1995). As always, the higher expression of GFP variants will likely improve fluorescent signal above endogenous autofluorescent background. Similar issues concerning autofluorescence are relevant to investigators performing microscopy. Since cellular autofluorescence often extends into the yellow part of the spectrum, use of a long-pass or yellow band-pass emission filter will often facilitate visual discrimination between red-shifted GFP fluorescence (green) and cellular autofluorescence (yellow). Cells should be viewed in colorless medium (lacking phenol red), such as PBS, to improve visual detection and discrimination. When red-shifted GFP variants are markedly overexpressed in cells, it may be difficult to achieve adequate fluorescent compensation of the green (EGFP) channel’s contribution to the red (PI) channel. Some investigators choose not to compensate for spectral overlap and analyze their data without the benefits fluorescent compensation offers. If neutral-density filtering does not overcome the problem (see Fig. 9.12.1), then substituting 7-AAD (5 µg/ml final) for PI may provide a satisfactory result. In experiments combining EGFP and 7-AAD (or PI), a 575-nm band-pass filter can be employed for detection of autofluorescence. Optimal detection of 7-AAD may be performed using a third PMT combined with a 650-nm long-pass filter.
Anticipated Results Typical results obtained from experiments using Basic Protocol 1 and Basic Protocol 2 are shown in Figure 9.12.1 and Figure 9.12.2, respectively. The signal obtained from GFP-expressing cells will vary considerably with choice of vector, cells, experimental conditions, detection system, and GFP variant. Al-
Studies of Cell Function
9.12.11 Current Protocols in Cytometry
Supplement 7
though both protocols describe assay of cells following short-term culture, GFP expression is stable for months in live mammalian fibroblasts. The percentage of red-shifted GFPexpressing human fibroblasts does not change over a 1-month observation period following retroviral transduction—attesting to a lack of selective pressure due to constitutive overexpression of EGFP.
Reporters of Gene Expression: Autofluorescent Proteins
Sensitivity and quantitation Many factors affect the ability to detect expression of GFP and variants above background (for discussion of autofluorescence, see Critical Parameters and Troubleshooting). The level of GFP fluorescence has not yet been shown by independent assay to correlate linearly with the number of GFP molecules in a cell. Formally, therefore, only relative quantitation of GFP protein should be reported. It remains possible that a nonfluorescent or nondetectable fraction of GFP is sequestered in cells—which could be problematical in experiments using fusion proteins. Piston and colleagues have presented clear data showing that 1 µM wild-type GFP expressed in the cytoplasm of mammalian HeLa cells is detectable above background using optimal imaging equipment (Niswender et al., 1995). Use of red-shifted and optimized GFP variants should offer detection sensitivities 5to 10-fold greater than wild-type GFP (Patterson et al., 1997). Approximately 10,000 EGFP molecules are required for detection in the cytoplasm of a single cell. Fluorometric assays of lysed GFP-expressing cells may be used to quantify the amount of GFP present by reference to a standard curve of recombinant GFP established at the same time. Fusion proteins containing GFP variants are now frequently used as markers to track expression of proteins. Fluorescently tagged proteins may be detected with sensitivity beyond that achieved by typical antibody techniques, with the added advantage that the cells remain intact (Wang and Hazelrigg, 1994). The problems of nonspecific antibody binding are also avoided when GFP fusion proteins are used. Successful expression and detection of GFP fusions have been reported in many dozens of publications to date. Nevertheless, it is contingent upon the investigator to demonstrate that the fused protein has not altered significantly from its native function (Flach et al., 1994; Marshall et al., 1995; Moores et al., 1996). Partly for this reason, it is common to create separate fusions of GFP to either the amino or carboxy terminus
of the gene to facilitate timely analysis of both proteins. Also, it is possible that production of a fusion protein could lead to loss of detectable fluorescence of the autofluorescent (e.g., redshifted GFP) partner. Independent verification of GFP expression in sorted cells or in cells thought to contain a GFP fusion protein may be performed by immunoblotting using commercially available antibodies. Using a fusion protein containing both GFP and the enzymatic reporter chloramphenicol acetyl transferase, a correlation has been demonstrated between GFP fluorescence and fusion-protein levels assessed by the enzymatic assay (Albano et al., 1998). Effects of fixation Fixation of cells prior to cytometric analysis is frequently required—e.g., to allow access to intracellular epitopes or because the sample may contain pathogenic microorganisms. Many fixatives will alter the fluorescent behavior of chromophores. Although GFP and its variants are resistant to fluorescence quenching (Chalfie et al., 1994), minor changes in cellular fluorescence after fixation should be controlled for. Paraformaldehyde fixation, used in Basic Protocol 2, increases the intrinsic fluorescence of cells (see Critical Parameters and Troubleshooting). If sorted GFP-expressing cells are to be viewed on slides using an epifluorescence microscope, use of nail polish as a sealant should be avoided (rubber cement will work well instead). Specimens may be stored for weeks at 4°C in the dark after thorough fixation with fresh paraformaldehyde and subsequent removal of fixative by washings. However, cellular autofluorescence will increase over time despite optimum fixation and storage as described—thereby compromising the GFP signal above background. In Basic Protocol 2, paraformaldehyde fixation may result in altered stoichiometry of PI binding to DNA, which can lead to poor resolution of cell cycle phases. Repeated washing of cells with PBS containing Triton X-100 (0.1% v/v) can modify DNA cross-linkage and improve resolution. Also, washing in ice-cold ethanol (70% v/v) after fixation may be helpful prior to staining with PI. Direct fixation of cells with ethanol (70% v/v) or methanol will fail due to presumed leakage of GFP from cells (absolute ethanol will cause sample dehydration and consequent loss of fluorescence). However if a GFP fusion protein is targeted to a subcellular region, such as the plasma membrane, then ethanol fixation (UNIT 7.9) should
9.12.12 Supplement 7
Current Protocols in Cytometry
produce excellent results (Wang et al., 1996; Kalejta et al., 1997). Finally, if problems persist in the form of poor CVs and a reduced G2/G1, then replacing PI with DAPI (which is less sensitive to chromatin structure and avoids use of RNase, but which will require a UV light source) may bring success.
Time Considerations Quantitative assays involving GFP and variants are time-efficient and convenient when applied to flow cytometry. As few as ten minutes may elapse from the time cells are harvested, placed in wash solution containing PI, and measured in the flow cytometer. By avoiding incubation times and centrifugation steps required by antibody staining or enzymatic techniques, use of GFP variants provides greater flexibility in experimental design. For example, cells obtained ex vivo may be expeditiously assayed to avoid potentially spurious changes that might occur after their removal had other detection methods been used. The GFP field has advanced rapidly over the last few years. However all the improvements, substantial as they are, have been based on modifications of the original GFP isolated from Aequorea victoria. There are perhaps 20 other GFPs in nature that are as yet uncloned (Cormier et al., 1974). It remains to be seen whether natural selection or selection in vitro will provide the investigator with greater opportunities.
Literature Cited Albano, C.R., Randers-Eichhorn, L., Bentley, W.E., and Rao, G. 1998. Green fluorescent protein as a real time quantitative reporter of heterologous protein production. Biotechnol. Prog. 14:351354. Alexander, I.E., Russell, D.W., and Miller, A.D. 1997. Transfer of contaminants in adeno-associated virus vector stocks can mimic transduction and lead to artifactual results. Hum. Gene Ther. 8:1911-1920. Amsterdam, A., Lin, S., and Hopkins, N. 1995. The Aequorea victoria green fluorescent protein can be used as a reporter in live zebrafish embryos. Dev. Biol. 171:123-129. Anderson, M.T., Tjioe, I.M., Lorincz, M.C., Parks, D.R., Herzenberg, L.A., Nolan, G.P., and Herzenberg, L.A. 1996. Simultaneous fluorescenceactivated cell sorter analysis of two distinct transcriptional elements within a single cell using engineered green fluorescent proteins. Proc. Natl. Acad. Sci. U.S.A. 93:8508-85011. Aubin, J.E. 1979. Autofluorescence of viable cultured mammalian cells. J. Histochem. Cytochem. 27:36-43.
Ausubel, F.A., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., and Struhl, K. (eds.). 1998. Current Protocols in Molecular Biology. John Wiley & Sons, New York. Barthmaier, P. and Fyrberg, E. 1995. Monitoring development and pathology of Drosophila indirect flight muscles using green fluorescent protein. Dev. Biol. 169:770-774. Baulcombe, D.C., Chapman, S., and Santa Cruz, S. 1995. Jellyfish green fluorescent protein as a reporter for virus infections. Plant J. 7:10451053. Benson, R.C., Meyer, R.A., Zaruba, M.E., and McKhann, G.M. 1979. Cellular autofluorescence—is it due to flavins? J. Histochem. Cytochem. 27:44-48. Bokman, S.H. and Ward, W.W. 1981. Renaturation of Aequorea green fluorescent protein. Biochem. Biophys. Res. Commun. 101:1372-1380. Brejc, K., Sixma, T.K., Kitts, P.A., Kain, S.R., Tsien, R.Y., Ormo, M., and Remington, S.J. 1997. Structural basis for dual excitation and photoisomerization of the Aequorea victoria green fluorescent protein. Proc. Natl. Acad. Sci. U.S.A. 94:2306-2311. Cepko, C. 1995. Large-scale preparation and concentration of retrovirus stocks. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 9.12.19.12.6. John Wiley & Sons, New York. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W.W., and Prasher, D.C. 1994. Green fluorescent protein as a marker for gene expression. Science 263:802805. Cheng, L., Fu, J., Tsukamoto, A., and Hawley, R.G. 1996. Use of green fluorescent protein variants to monitor gene transfer and expression in mammalian cells. Nature Biotechnol. 14:606-609. Cole, N.B., Smith, C.L., Sciaky, N., Terasaki, M., Edidin, M., and Lippincott-Schwartz, J. 1996. Diffusional mobility of Golgi proteins in membranes of living cells. Science 273:797-801. Cormack, B.P., Valdivia, R.H., and Falkow, S. 1996. FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173:33-38. Cormier, M.J., Hori, K., and Anderson, J.M. 1974. Bioluminescence in coelenterates. Biochim. Biophys. Acta 346:137-164. Crameri, A., Whitehorn, E.A., Tate, E., and Stemmer, W.P. 1996. Improved green fluorescent protein by molecular evolution using DNA shuffling. Nature Biotechnol. 14:315-319. Cubitt, A.B., Heim, R., Adams, S.R., Boyd, A.E., Gross, L.A., and Tsien, R.Y. 1995. Understanding, improving and using green fluorescent proteins. Trends Biochem. Sci. 20:448-455. Davis, I., Girdham, C.H., and O’Farrell, P.H. 1995. A nuclear GFP that marks nuclei in living Drosophila embryos; Maternal supply overcomes a delay in the appearance of zygotic fluorescence. Dev. Biol. 170:726-729.
Studies of Cell Function
9.12.13 Current Protocols in Cytometry
Supplement 7
Delagrave, S., Hawtin, R.E., Silva, C.M., Yang, M.M., and Youvan, D.C. 1995. Red-shifted excitation mutants of the green fluorescent protein. Bio/Technology 13:151-154.
Horton, R.M., Ho, S.N., Pullen, J.K., Hunt, H.D., Cai, Z., and Pease, L.R. 1993. Gene splicing by overlap extension. Methods Enzymol. 217:270279.
dos Remedios, C.G. and Moens, P.D. 1995. Fluorescence resonance energy transfer spectroscopy is a reliable “ruler” for measuring structural changes in proteins: Dispelling the problem of the unknown orientation factor. J. Struct. Biol. 115:175-185.
Ikawa, M., Kominami, K., Yoshimura, Y., Tanaka, K., Nishimune, Y., and Okabe, M. 1995. A rapid and non-invasive selection of transgenic embryos before implantation using green fluorescent protein (GFP). FEBS Lett. 375:125-128.
Ehrig, T., O’Kane, D.J., and Prendergast, F.G. 1995. Green-fluorescent protein mutants with altered fluorescence excitation spectra. FEBS Lett. 367:163-166. Ellenberg, J., Siggia, E.D., Moreira, J.E., Smith, C.L., Presley, J.F., Worman, H.J., and LippincottSchwartz, J. 1997. Nuclear membrane dynamics and reassembly in living cells: Targeting of an inner nuclear membrane protein in interphase and mitosis. J. Cell Biol. 138:1193-206.
Inouye, S. and Tsuji, F.I. 1994b. Aequorea green fluorescent protein. Expression of the gene and fluorescence characteristics of the recombinant protein. FEBS Lett. 341:277-280. Kahana, J.A., Schnapp, B.J., and Silver, P.A. 1995. Kinetics of spindle pole body separation in budding yeast. Proc. Natl. Acad. Sci. U.S.A. 92:9707-9711.
Flach, J., Bossie, M., Vogel, J., Corbett, A., Jinks, T., Willins, D.A. and Silver, P.A. 1994. A yeast RNA-binding protein shuttles between the nucleus and the cytoplasm. Mol. Cell. Biol. 14:8399-8407.
Kalejta, R.F., Shenk, T., and Beavis, A.J. 1997. Use of a membrane-localized green fluorescent protein allows simultaneous identification of transfected cells and cell cycle analysis by flow cytometry. Cytometry 29:286-291.
Haas, J., Park, E.C., and Seed, B. 1996. Codon usage limitation in the expression of HIV-1 envelope glycoprotein. Curr. Biol. 6:315-324.
Kalish, J.E., Keller, G.A., Morrell, J.C., Mihalik, S.J., Smith, B., Cregg, J.M., and Gould, S.J. 1996. Characterization of a novel component of the peroxisomal protein import apparatus using fluorescent peroxisomal proteins. EMBO J. 15:3275-3285.
Hanazono, Y., Yu, J.M., Dunbar, C.E., and Emmons, R.V. 1997. Green fluorescent protein retroviral vectors: Low titer and high recombination frequency suggest a selective disadvantage. Hum. Gene Ther. 8:1313-1319. Hanenberg, H., Xiao, X.L., Dilloo, D., Hashino, K., Kato, I., and Williams, D.A. 1996. Colocalization of retrovirus and target cells on specific fibronectin fragments increases genetic transduction of mammalian cells. Nat. Med. 2:876-882. Haseloff, J., Siemering, K.R., Prasher, D.C., and Hodge, S. 1997. Removal of a cryptic intron and subcellular localization of green fluorescent protein are required to mark transgenic Arabidopsis plants brightly. Proc. Natl. Acad. Sci. U.S.A. 94:2122-2127. Heim, R. and Tsien, R.Y. 1996. Engineering green fluorescent protein for improved brightness, longer wavelengths and fluorescence resonance energy transfer. Curr. Biol. 6:178-182. Heim, R., Prasher, D.C., and Tsien, R.Y. 1994. Wavelength mutations and posttranslational autoxidation of green fluorescent protein. Proc. Natl. Acad. Sci. U.S.A. 91:12501-12504. Heim, R., Cubitt, A.B., and Tsien, R.Y. 1995. Improved green fluorescence. Nature 373:663-664.
Reporters of Gene Expression: Autofluorescent Proteins
Inouye, S. and Tsuji, F.I. 1994a. Evidence for redox forms of the Aequorea green fluorescent protein. FEBS Lett. 351:211-214.
Higgs, S., Traul, D., Davis, B.S., Kamrud, K.I., Wilcox, C.L., and Beaty, B.J. 1996. Green fluorescent protein expressed in living mosquitoes— without the requirement of transformation. Biotechniques 21:660-664.
Kneen, M., Farinas, J., Li, Y., and Verkman, A.S. 1998. Green fluorescent protein as a noninvasive intracellular pH indicator. Biophys. J. 74:15911599. Kotani, H., Newton, P.B.I., Zhang, S., Chiang, Y.L., Otto, E., Weaver, L., Blaese, R.M., Anderson, W.F., and McGarrity, G.J. 1994. Improved methods of retroviral vector transduction and production for gene therapy. Hum. Gene Ther. 5:19-28. Levy, J.P., Muldoon, R.R., Zolotukhin, S., and Link, C.J., Jr. 1996. Retroviral transfer and expression of a humanized, red-shifted green fluorescent protein gene into human tumor cells. Nature Biotechnol. 14:610-614. Liu, M.L., Winther, B.L., and Kay, M.A. 1996. Pseudotransduction of hepatocytes by using concentrated pseudotyped vesicular stomatitis virus G glycoprotein (VSV-G)-Moloney murine leukemia virus-derived retrovirus vectors: Comparison of VSV-G and amphotropic vectors for hepatic gene transfer. J. Virol. 70:2497-2502. Llopis, J., McCaffery, J.M., Miyawaki, A., Farquhar, M.G., and Tsien, R.Y. 1998. Measurement of cytosolic, mitochondrial, and Golgi pH in single living cells with green fluorescent proteins. Proc. Natl. Acad. Sci. U.S.A. 95:6803-6808. Lo, W., Rodgers, W., and Hughes, T. 1998. Making genes green: Creating green fluorescent protein (GFP) fusions with blunt-end PCR products. Biotechniques 25:94-98.
9.12.14 Supplement 7
Current Protocols in Cytometry
Lybarger, L., Dempsey, D., Patterson, G.H., Piston, D.W., Kain, S.R., and Chervenak, R. 1998. Dualcolor flow cytometric detection of fluorescent proteins using single-laser (488-nm) excitation. Cytometry 31:147-152. Maniak, M., Rauchenberger, R., Albrecht, R., Murphy, J., and Gerisch, G. 1995. Coronin involved in phagocytosis: Dynamics of particle-induced relocalization visualized by a green fluorescent protein tag. Cell 83:915-924. Marshall, J., Molloy, R., Moss, G.W., Howe, J.R., and Hughes, T.E. 1995. The jellyfish green fluorescent protein: A new tool for studying ion channel expression and function. Neuron 14:211-215. Mellman, I., Fuchs, R., and Helenius, A. 1986. Acidification of the endocytic and exocytic pathways. Ann. Rev. Biochem. 55:663-700. Miller, A.D. and Rosman, G.J. 1989. Improved retroviral vectors for gene transfer and expression. Biotechniques 7:980-990. Mitra, R.D., Silva, C.M., and Youvan, D.C. 1996. Fluorescence resonance energy transfer between blue-emitting and red-shifted excitation derivatives of the green fluorescent protein. Gene 173:13-17. Miyawaki, A., Llopis, J., Heim, R., McCaffery, J.M., Adams, J.A., Ikura, M., and Tsien, R.Y. 1997. Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature 388:882-887. Moores, S.L., Sabry, J.H., and Spudich, J.A. 1996. Myosin dynamics in live Dictyostelium cells. Proc. Natl. Acad. Sci. U.S.A. 93:443-446. Morin, J.G. and Hastings, J.W. 1971a. Biochemistry of the bioluminescence of colonial hydroids and other coelenterates. J. Cell. Physiol. 77:305-312. Morin, J.G. and Hastings, J.W. 1971b. Energy transfer in a bioluminescent system. J. Cell. Physiol. 77:313-318. Moriyoshi, K., Richards, L.J., Akazawa, C., O’Leary, D.D. and Nakanishi, S. 1996. Labeling neural cells using adenoviral gene transfer of membrane-targeted GFP. Neuron 16:255-260. Niswender, K.D., Blackman, S.M., Rohde, L., Magnuson, M.A., and Piston, D.W. 1995. Quantitative imaging of green fluorescent protein in cultured cells: Comparison of microscopic techniques, use in fusion proteins and detection limits. J. Microsc. 180:109-116. Ormö, M., Cubitt, A.B., Kallio, K., Gross, L.A., Tsien, R.Y., and Remington, S.J. 1996. Crystal structure of the Aequorea victoria green fluorescent protein. Science 273:1392-1395. Palm, G.J., Zdanov, A., Gaitanaris, G.A., Stauber, R., Pavlakis, G.N., and Wlodawer, A. 1997. The structural basis for spectral variations in green fluorescent protein. Nature Struct. Biol. 4:361365. Patterson, G.H., Knobel, S.M., Sharif, W.D., Kain, S.R., and Piston, D.W. 1997. Use of the green fluorescent protein and its mutants in quantita-
tive fluorescence microscopy. Biophys. J. 73:2782-2790. Pear, W. 1996. Transient transfection methods for preparation of high-titer retroviral supernatants. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 9.11.1-9.11.18. John Wiley & Sons, New York. Persons, D.A., Allay, J.A., Allay, E.R., Smeyne, R.J., Ashmun, R.A., Sorrentino, B.P., and Nienhuis, A.W. 1997. Retroviral-mediated transfer of the green fluorescent protein gene into murine hematopoietic cells facilitates scoring and selection of transduced progenitors in vitro and identification of genetically modified cells in vivo. Blood 90:1777-1786. Pines, J. 1995. GFP in mammalian cells. Trends Genet. 11:326-327. Prasher, D.C., Eckenrode, V.K., Ward, W.W., Prendergast, F.G., and Cormier, M.J. 1992. Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111:229-233. Rasko, J.E.J., Kiem, H.-P., Morris, J.C., Gottschalk, R.J., Peterson, L.J., Andrews, R.G., and Miller, A.D. 1997. Efficient transduction of hemopoietic progenitors using green fluorescent protein (GFP)–containing retroviral vectors with vious pseudotypes. Blood 90:118a. Reichel, C., Mathur, J., Eckes, P., Langenkemper, K., Koncz, C., Schell, J., Reiss, B., and Maas, C. 1996. Enhanced green fluorescence by the expression of an Aequorea victoria green fluorescent protein mutant in mono- and dicotyledonous plant cells. Proc. Natl. Acad. Sci. U.S.A. 93:5888-93. Rizzuto, R., Brini, M., De Giorgi, F., Rossi, R., Heim, R., Tsien, R.Y., and Pozzan, T. 1996. Double labelling of subcellular structures with organelle-targeted GFP mutants in vivo. Curr. Biol. 6:183-188. Ropp, J.D., Donahue, C.J., Wolfgang-Kimball, D., Hooley, J.J., Chin, J.Y., Hoffman, R.A., Cuthbertson, R.A., and Bauer, K.D. 1995. Aequorea green fluorescent protein analysis by flow cytometry. Cytometry 21:309-317. Ropp, J.D., Donahue, C.J., Wolfgang-Kimball, D., Hooley, J.J., Chin, J.Y., Cuthbertson, R.A., and Bauer, K.D. 1996. Aequorea green fluorescent protein: Simultaneous analysis of wild-type and blue-fluorescing mutant by flow cytometry. Cytometry 24:284-288. Selvin, P.R. 1995. Fluorescence resonance energy transfer. Methods Enzymol. 246:300-334. Stearns, T. 1995. Green fluorescent protein: The green revolution. Curr. Biol. 5:262-264. Subramanian, K. and Meyer, T. 1997. Calcium-induced restructuring of nuclear envelope and endoplasmic reticulum calcium stores. Cell 89:963-971. Tannahill, D., Bray, S., and Harris, W.A. 1995. A Drosophila E(spl) gene is “neurogenic” in
Studies of Cell Function
9.12.15 Current Protocols in Cytometry
Supplement 7
Xenopus: A green fluorescent protein study. Dev. Biol. 168:694-697. Tsien, R.Y. and Miyawaki, A. 1998. Seeing the machinery of live cells. Science 280:1954-1955. Wampler, J.E., Hori, K., Lee, J.W., and Cormier, M.J. 1971. Structured bioluminescence: Two emitters during both the in vitro and the in vivo bioluminescence of the sea pansy, Renilla. Biochemistry 10:2903-2909. Wang, S. and Hazelrigg, T. 1994. Implications for bcd mRNA localization from spatial distribution of exu protein in Drosophila oogenesis. Nature 369:400-403. Wang, D.S., Miller, R., Shaw, R., and Shaw, G. 1996. The pleckstrin homology domain of human beta I sigma II spectrin is targeted to the plasma membrane in vivo. Biochem. Biophys. Res. Comm. 225:420-426. Ward, W.W. and Bokman, S.H. 1982. Reversible denaturation of Aequorea green-fluorescent protein: Physical separation and characterization of the renatured protein. Biochemistry 21:45354540. Xu, X., Gerard, A.L., Huang, B.C., Anderson, D.C., Payan, D.G., and Luo, Y. 1998. Detection of programmed cell death using fluorescence energy transfer. Nucl. Acids Res. 26:2034-2035. Yang, F., Moss, L.G., and Phillips, G.N., Jr. 1996. The molecular structure of green fluorescent protein. Nature Biotechnol. 14:1246-1251. Zhou, Y., Aran, J., Gottesman, M.M., and Pastan, I. 1998. Co-expression of human adenosine deaminase and multidrug resistance using a bicistronic retroviral vector. Hum. Gene Ther. 9:287-293. Zolotukhin, S., Potter, M., Hauswirth, W.W., Guy, J., and Muzyczka, N. 1996. A “humanized” green fluorescent protein cDNA adapted for high-level expression in mammalian cells. J. Virol. 70:4646-4654. Zylka, M.J. and Schnapp, B.J. 1996. Optimized filter set and viewing conditions for the S65T mutant of GFP in living cells. Biotechniques 21:220-1, 224-6.
Key Reference Chalfie, M. and Kain, S. (eds). 1998. Green Fluorescent Proteins: Proteins, Properties, Applications and Protocols. John Wiley & Sons, New York. Provides detailed background for the use of GFP in many biological systems.
Reporters of Gene Expression: Autofluorescent Proteins
Internet Resources http://www.bio.net:80/hypermail/FLUORESCENTPROTEINS/ A World Wide Web newsgroup offering broad discussions covering many aspects of autofluorescent proteins and an opportunity to post questions as well as search by text words. http://www.image1.com/products/metagfp/ mutation.html A table containing an extensive list of mutant GFPs, with commentary. http://www.aurorabio.com/ Aurora Biosciences Corporation Web site. http://www.clontech.com/clontech/Catalog/Gene Expression/GFPintro.html Web page for Living Colors Fluorescent Protein. http://www.invitrogen.com/expressions/697-6.html Web page for pTracer vectors, which incorporate GFP gene. http://www2.lifetech.com:80/catalog/techline/ molecular_biology/product_description/ gl1vedes.html Web page for pGreen Lantern 1 reporter vector. http://www.packardinst.com/cgi-bin/hazel.exe? action=SERVE&item=/prod_serv/fluor.htm Web page for CytoGem GFP reporter vectors. http://www.pharmingen.com/ PharMingen Web site. http://www.qbi.com/Products/autofluorescent.asp Web page for Quantum Biosciences Autofluorescent Proteins (AFPs). http://www.omegafilters.com/products/ fluorescence/xf500/index.html Filter sets optimized for use with GFPs, made by Omega Optics. http://www.chroma.com/ Filter sets optimized for use with GFPs, made by Chroma Technology.
Contributed by John E.J. Rasko Centenary Institute of Cancer Medicine and Cell Biology Sydney, Australia
Dr. Rasko wishes to acknowledge the support by Fellowship DRG081 from the Cancer Research Fund of the Damon Runyon–Walter Winchell Foundation for studies at the Fred Hutchinson Cancer Research Center, Seattle, Washington. He also wishes to thank Dr. Andreas Strasser, Dr. Francis Battye, and Dr. Chung L. Li for his early education in flow cytometry.
9.12.16 Supplement 7
Current Protocols in Cytometry
In Vitro Invasion Assays: Phagocytosis of the Extracellular Matrix
UNIT 9.13
Phagocytosis is usually thought of as an immune cell function. However, the ability of tumor cells to phagocytose extracellular matrix (ECM) proteins is correlated with their invasive capacity. This unit contains the protocols required to perform a flow cytometry– based assay that allows rapid quantitative assessment of phagocytosis. Flow cytometry is used to measure the ability of cancer cells to degrade and internalize fluorescently labeled ECM molecules. A Support Protocol describes the coupling of FITC to gelatin, although any fluorophore could be substituted. Fluorescently labeled gelatin can then be crosslinked to form a solid planar matrix or, alternatively, small subcellular-sized beads. Whereas the planar matrix must be degraded by the cells to facilitate uptake (Basic Protocol), small beads can be taken up in a manner that is independent of proteolysis and requires only binding and engulfment of particles (Alternate Protocol). Additionally, the Basic Protocol describes how to culture and recover cells from fluorescent substrata for analysis. MEASUREMENT OF THE PHAGOCYTIC ACTIVITY OF CANCER CELLS ON A CROSS-LINKED FITC-CONJUGATED GELATIN MATRIX FITC-gelatin is used as a reporter molecule to measure the ability of cells to phagocytose partially proteolyzed FITC-gelatin matrix. The protocol describes how to cross-link FITC-gelatin to create an ECM support for cell growth. Cells plated onto the FITC-gelatin proteolyze the matrix and then phagocytose partially digested fragments. At the end of the incubation time, cells are released from the matrix by trypsinization. This protocol assumes familiarity with both standard tissue culture techniques (APPENDIX 3B) and flow cytometry measurements.
BASIC PROTOCOL
Materials FITC-conjugated gelatin (see Support Protocol) Unlabeled gelatin/sucrose solution (see recipe) 0.5% (v/v) glutaraldehyde/PBS solution (see recipe) PBS (APPENDIX 2A): ice cold, nonsterile, as well as room temperature, sterile 70% (v/v) ethanol, ice cold Serum-free cell growth medium with antibiotics: Improved minimal essential medium (IMEM; Life Technologies) with 1× penicillin/streptomycin (from 100× stock; Life Technologies) Cells of interest Trypsin/EDTA solution (Life Technologies) Complete cell growth medium: IMEM with 10% (v/v) heat-inactivated FBS (APPENDIX 2A) 3.0% (v/v) formaldehyde/PBS solution (see recipe) 24-well tissue culture dishes, prechilled to 4°C Filter-top cell strainer tubes (Falcon) Flow cytometer equipped with 488-nm argon-laser excitation and 545-nm (530 ± 30–nm) band-pass emission filter Prepare plates 1. Warm an aliquot of FITC-conjugated gelatin and an aliquot of unlabeled gelatin/sucrose solution to 37°C. Coat the base of each well of a prechilled 24-well tissue culture dish with ∼100 µl of either unlabeled gelatin or FITC-gelatin. Remove any excess solution immediately and incubate the plate 15 min at 4°C (on ice). Contributed by Emma T. Bowden, Susette Mueller, and Peter J. Coopman Current Protocols in Cytometry (2000) 9.13.1-9.13.8 Copyright © 2000 by John Wiley & Sons, Inc.
Studies of Cell Function
9.13.1 Supplement 12
Cells plated onto unlabeled gelatin will provide a background reading of the autofluorescence of cells in this assay. The authors suggest performing test and control samples in triplicate.
2. Pipet 1 ml ice-cold 0.5% glutaraldehyde/PBS solution down the wall of each well, taking care not to disturb the gelatin coating at the base of the well. Incubate 15 min on ice. The importance of using freshly made glutaraldehyde solutions cannot be overemphasized. This cross-linking step determines the density of the gelatin matrix and must be standardized between experiments. Both the concentration of the glutaraldehyde and the incubation time on gelatin will affect the density of the matrix produced. This may affect the ability of cells to degrade the matrix and hence affect cell behavior in the assay.
3. Carefully remove glutaraldehyde and wash the films gently with 1.0 ml/well ice-cold nonsterile PBS. 4. Add 1 ml ice-cold 70% ethanol to each well to sterilize the films. Incubate at room temperature for 15 min. After this, all manipulations should be performed in a laminar flow hood to ensure sterile conditions for tissue culture.
5. Gently wash films with 1 ml/well sterile PBS, three times, 5 min each, at room temperature. 6. Add 1 ml/well serum-free cell growth medium with antibiotics and incubate 1 hr at 37°C or overnight at 4°C. This step is required to block free aldehyde groups. Sterile 0.3 M glycine can be substituted for the medium.
Label cells by proteolysis and phagocytosis of matrix 7. Remove serum-free medium and seed 100,000 to 200,000 cells/well in 1 ml serumfree medium with antibiotics. Allow the cells to attach and spread for up to 72 hr at 37°C under standard cell growth conditions. The optimal cell number must be determined for the cell type. There must be sufficient cells to read by flow cytometry; however, increasing the cell density to confluence will inhibit phagocytosis. The optimal time for exposure to FITC-gelatin must also be determined. Although cells will begin to degrade the matrix within 30 min of plating, an incubation of 6 to 18 hr usually provides cells that have internalized sufficient matrix to measure differences between cell lines and/or to assess treatments that enhance or inhibit phagocytic activity.
8. Remove medium and wash cells 5 min with 1 ml/well sterile PBS at room temperature. Harvest and fix cells 9. Release cells from matrix by incubating in 200 µl/well trypsin/EDTA solution 5 min at room temperature. Release of cells from the matrix can easily be assessed by light microscopy. If cells are very slow to trypsinize, the plate can be incubated at 37°C. Trypsin also removes FITC-gelatin from the cell surface, so that only intracellular label is detected by flow cytometry. In Vitro Invasion Assays: Phagocytosis of the Extracellular Matrix
10. Add 1 ml complete cell growth medium to each well and transfer each cell suspension to a 1-ml microcentrifuge tube. Isolate cells by centrifuging 5 min at 200 × g, 4°C. 11. Gently resuspend cells in 1.0 ml sterile PBS and repeat centrifugation.
9.13.2 Supplement 12
Current Protocols in Cytometry
12. Resuspend cell pellet in 50 µl of 3.0% formaldehyde/PBS solution and incubate 5 min at room temperature. Adjust final volume to 500 µl with sterile PBS. This fixation step is introduced for operator safety during flow cytometry analysis.
Analyze cells 13. Pass cells through a filter-top cell strainer tube just prior to flow cytometric analysis to remove cell clumps. 14. Analyze fluorescence of the cells in a flow cytometer using 488-nm excitation and collecting emission at 545 nm (530 ± 30 nm). Directly compare the fluorescence of cells plated on unlabeled gelatin to the fluorescence of those plated on FITC-gelatin. The authors usually identify a population of cells from a control sample on unlabeled gelatin to give a baseline of fluorescence. All other FITC-gelatin samples should fall above this value. The fluorescence can be expressed as fold over background (i.e., FITC-gelatin fold over unlabeled gelatin) or molecules of equivalent soluble fluorochrome (MESF; see Anticipated Results and UNIT 1.3).
MEASUREMENT OF PHAGOCYTIC ACTIVITY OF CANCER CELLS ON CROSS-LINKED FITC-CONJUGATED GELATIN BEADS
ALTERNATE PROTOCOL
This protocol, which describes the use of FITC-gelatin as a reporter molecule to measure the phagocytic activity of adherent cancer cell lines, allows phagocytosis to be separated from matrix degradation. The phagocytosis measured by the Basic Protocol requires protease activity to begin degradation of the matrix before any material is phagocytosed. In contrast, this assay measures just the internalization of subcellular (<10-µm) FITCgelatin beads. Beads are prepared by sonication of a warm gelatin solution in the presence of 1-butanol. This allows the formation of small droplets of gelatin, which are then stabilized as spheres by chilling and cross-linking. Additional Materials (also see Basic Protocol) Gelatin type A from porcine skin, 300 bloom (50 to 100 kDa; Sigma) Water-saturated 1-butanol (see recipe) 5% (v/v) glutaraldehyde/PBS solution (see recipe) 50°C water bath Kontes Micro Ultrasonic Cell Disruptor (probe type) 12-well tissue culture dishes Prepare FITC-gelatin beads 1. Prewarm a 500-µl aliquot of FITC-conjugated gelatin to 50°C. Add sufficient gelatin to bring the final concentration (w/v) to 5%. Incubate at 50°C until the additional gelatin is fully dissolved. The starting concentration of gelatin is ∼2%.
2. Add 0.15 ml FITC-gelatin to 0.5 ml water-saturated 1-butanol. 3. Vortex 10 sec, sonicate 10 sec at no. 30 setting, and then incubate 15 min on ice. 4. Add 0.8 ml 5% glutaraldehyde/PBS solution at 4°C, vortex 10 sec, and incubate 1 hr on ice. 5. Centrifuge 1 min, 4°C, at maximum speed in a microcentrifuge to pellet the gelatin. Remove the supernatant and sonicate the bead pellet in 1 ml ice-cold sterile PBS. Repeat centrifugation and sonication. Studies of Cell Function
9.13.3 Current Protocols in Cytometry
Supplement 12
6. Remove large beads (>20 µm) by centrifuging 5 sec at 330 × g, 4°C. The 10-µm beads should remain in the supernatant. 7. Centrifuge 30 sec at 330 × g, 4°C, and isolate 10-µm beads from supernatant. 8. Wash 10-µm beads three times in 1.0 ml ice-cold, sterile PBS. Quench in 1 ml serum-free growth medium with antibiotics as described for gelatin films (see Basic Protocol, step 6). After this step, beads can be stored for several weeks at −20°C in a 1:1 (v/v) mixture of serum-free medium and glycerol. However, if this method of storage is used, beads must be washed 4 to 5 times in 1.0 ml serum-free medium before use.
Label and analyze cells 9. Seed cells at 70,000 to 100,000 cells/well in a 12-well tissue culture dish in the presence of 100,000 to 200,000 beads. Use serum-free cell growth medium with antibiotics. 10. Allow cells to phagocytose the beads for up to 72 hr at 37°C under standard cell growth conditions. 11. Release cells from the plate and analyze by flow cytometry as described (see Basic Protocol, steps 9 to 14). SUPPORT PROTOCOL
PREPARATION OF FITC-CONJUGATED GELATIN MATRIX Fluorescein is a small inorganic molecule that can easily be coupled to the amino groups of a range of proteins. This protocol uses the reactive group of fluorescein isothiocyanate (FITC) to couple this fluorophore to gelatin. However, other ECM molecules, such as Matrigel (Collaborative Biomedical Products), laminin, or fibronectin, can also be successfully used. For a discussion of these types of coupling reactions using fluorescein and other reactive fluorophores, refer to UNIT 4.2. CAUTION: FITC is hazardous; follow appropriate precautions during handling and disposal. Materials Low-salt conjugation buffer (see recipe) Fluorescein 5-isothiocyanate (FITC) isomer I (Sigma) Gelatin type A from porcine skin, 300 bloom (50 to 100 kDa; Sigma) High-salt conjugation buffer (see recipe) PBS (APPENDIX 2A), nonsterile, 37°C Sucrose Dialysis tubing (MWCO 6000 to 8000; volume/length = 3.3 ml/cm) Spectrophotometer (e.g., Beckman DU-65) 1. Warm 200 ml low-salt conjugation buffer to 37°C and add 6 mg FITC. 2. Dissolve 0.5 g gelatin in 25 ml high-salt conjugation buffer at 37°C. Gelatin is a heterogeneous mixture of water-soluble, acid-hydrolyzed collagen fragments. The molecular weight of 300-bloom gelatin is 50 to 100 kDa.
In Vitro Invasion Assays: Phagocytosis of the Extracellular Matrix
3. Wash dialysis tubing inside and out with sterile distilled water. Rinse with high-salt conjugation buffer.
9.13.4 Supplement 12
Current Protocols in Cytometry
4. Dialyze gelatin solution against FITC/low-salt conjugation buffer for ≥90 min in complete darkness at 37°C with stirring. FITC is susceptible to bleaching under constant exposure to light. The required darkness can be achieved simply by wrapping the dialysis beaker in aluminum foil.
5. Replace FITC/low-salt conjugation buffer with a large volume (500 ml) of prewarmed (37°C) PBS and continue dialysis for 2 to 3 days with at least two buffer changes per day. FITC-gelatin requires extensive dialysis against PBS.
6. Carefully remove FITC-gelatin from the dialysis tubing and add sucrose to give a final concentration of 2% (w/v). 7. Divide into 500-µl aliquots and store for up to 3 months at 4°C. 8. Measure OD at 280 nm and 493 nm. 9. Estimate the FITC-to-protein ratio according to the calculations outlined in UNIT 4.2. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Formaldehyde/PBS solution, 3.0% (v/v) Dilute a 10-ml vial of 16% formaldehyde stock solution (EM Science) into ice-cold PBS (APPENDIX 2A) to a final concentration of 3.0% (v/v). Store up to 1 week at 4°C. The total volume prepared will depend on the number of samples being tested; one 24-well plate of samples requires 1200 ìl. The 10-ml vials of 16% formaldehyde (under nitrogen) should be stored at 4°C. Once opened, the formaldehyde can be transferred to a screw-top tube and kept for ∼1 week at 4°C.
Glutaraldehyde/PBS solution, 0.5% and 5% (v/v) Dilute an entire 10-ml vial of 8% glutaraldehyde (EM Science) into ice-cold PBS (APPENDIX 2A) to a final concentration of 0.5% or 5% (v/v), as needed. Store at 4°C for up to 1 week. The 10-ml stock vials of 8% glutaraldehyde (under nitrogen) should be stored at 4°C.
High-salt conjugation buffer 19.07 g Na2B4O7 (mol. wt. 381.4; 50 mM final) 2.3 g NaCl (mol. wt. 58.44; 40 mM final) 700 ml water Adjust pH to 9.3 with NaOH Bring final volume to 1 liter Check final pH and store up to 3 months at room temperature Low-salt conjugation buffer Prepare as for high-salt conjugation buffer (see recipe) but omit NaCl. Unlabeled gelatin/sucrose solution Dissolve 0.2 g gelatin and 0.2 g sucrose in 10 ml PBS (APPENDIX 2A) at 37°C. Divide into 1-ml aliquots and store up to 3 months at −20°C. Water-saturated 1-butanol Vortex a 50:50 (w/w) mixture of distilled water and 1-butanol. Allow the two phases to separate and use the upper phase. Store up to 1 month at 4°C.
Studies of Cell Function
9.13.5 Current Protocols in Cytometry
Supplement 12
COMMENTARY Background Information
In Vitro Invasion Assays: Phagocytosis of the Extracellular Matrix
A number of cell types from the immune system, including polymorphonuclear granulocytes, monocytes, and macrophages, can recognize the presence of microorganisms or tissue debris usually associated with infection, inflammation, and wound repair (Brown, 1995). These particles are bound by cell-surface receptors, engulfed, internalized, and then degraded during a process called phagocytosis. This phenomenon has also been detected in invasive tumor cells including human breast cancer cell lines (Montcourrier et al., 1994; Coopman et al., 1996), human epitheloid cervix carcinoma cells (Van Peteghem et al., 1980), and rat glioma cells (Bjerknes et al., 1987). Tumor-cell phagocytosis has several component activities, including ECM proteolysis and removal of partially degraded matrix from the cell surface (Ruoslahti, 1992; Montgomery et al., 1994). This behavior is highly relevant to tumor progression (Stetler-Stevenson et al., 1993; Lochter and Bissell, 1995; MacDougall and Matrisian, 1995; Coopman et al., 1998). This unit provides protocols for two assays to separate proteolysis and internalization versus internalization only. Serine and metalloproteinase inhibitors have been used to determine that extracellular matrix degradation is an essential component of tumor-cell phagocytosis (Coopman et al., 1998). In addition, phagocytosis can be modulated by perturbation of specific integrin heterodimers (Coopman et al., 1996). Finally, sensitivity to cytochalasin D confirms that tumorcell phagocytosis is mediated by the actin cytoskeleton (Coopman et al., 1998). These assays are fast and reproducible. They utilize endogenous-type substrates as compared to the more common latex-, polystyrene-, or dextran-bead–based assays. They have been used as relatively rapid-throughput, low-cost methods to compare cell lines and to test the efficacy of treatments designed to stimulate or inhibit invasion. Even carcinoma cell lines that are not very invasive in other assays, such as MCF-7 breast cancer cells, show detectable activity in these assays (Coopman et al., 1998). The advantage of these assays is that they are usually performed in the presence of serumfree medium. This is in direct contrast to several invasion assays that have an absolute requirement for serum (Imamura et al., 1991; Itoh et al., 1999). They may also be performed in the presence of serum-containing medium; how-
ever, the authors have found that the presence of serum can reduce the rate of phagocytosis (unpublished data). The reasons for this are unknown. Another advantage to these assays as a measure of invasive potential is that all the tumor cell lines tested to date demonstrate phagocytic activity to some degree. Therefore, many types of measurement are possible, for example, comparisons of the response of invasive versus noninvasive cell lines to drug treatments.
Critical Parameters and Troubleshooting For a discussion of the critical parameters for fluorochrome labeling of proteins, see UNIT 4.2. Once prepared, FITC-gelatin is extremely stable under the recommended storage conditions. The authors have demonstrated that MDA-MB-231 cells will phagocytose FITClabeled collagen type I, Matrigel, or crosslinked gelatin (Coopman et al., 1998). However, gelatin is more commonly used because it is easier to handle and less expensive than the other substrates listed. Preparation of 24-well plates with gelatin and FITC-gelatin is relatively easy, but careful attention should be paid when the films are washed, as vigorous pipetting can disrupt them. The authors recommend pipetting solutions down the side of the well to avoid this problem. Gelatin phagocytosis is both cell-density and time dependent. For example, FITC-gelatin uptake in the model cell line MDA-MB-231 is inhibited when the cells are plated at >100,000 to 200,000 cells per well (24-well plate; Coopman et al., 1998). Also, for this cell line, the optimum time for phagocytosis assays is 6 to 24 hr (Coopman et al., 1998). After 24 hr, increased incubation of cells with FITCgelatin does not give a proportional increase in FITC-gelatin uptake by cells. The data obtained for MDA-MB-231 can be used as a guideline to design assays for other cell lines. Also, the authors always include this invasive breast carcinoma cell line as a positive control within each experiment performed. This provides an internal standard for each experiment and allows comparison between experiments performed on different days. When comparing the phagocytic capacity of relatively noninvasive carcinoma cell lines, the use of longer time points (i.e., 24 hr) will maximize the detection of differences. For highly invasive cell lines, 6 hr is usually sufficient.
9.13.6 Supplement 12
Current Protocols in Cytometry
300
gelatin FITC-gelatin
Cell number
250
FITC-gelatin + integrinperturbing peptide
200
150
100
50
0 100
101
102
103
104
Fluorescence intensity
Figure 9.13.1 Representative data from a phagocytosis experiment. MDA-MB-231 cells were incubated for 18 hr on unlabeled gelatin, FITC-gelatin, or FITC-gelatin in the presence of an integrin-perturbing agent (Coopman et al., 1996). Fluorescence intensity was measured in 10,000 cells from each sample.
Anticipated Results Data are standardized by calibration using fluorescent reference standards, and are expressed as mean MESF units (absolute calibrated fluorescence intensity), as fold over background (relative phagocytosis of FITC-labeled matrix as compared to unlabeled matrix), or as a percentage of stimulation or inhibition of phagocytosis as compared to controls in the absence of treatments. For a cell line such as MDA-MB-231, the fold over background is ∼4. Although this may seem low, the high degree of reproducibility means that these data are consistently significant. Also, this rate can be routinely stimulated to 12- or 14-fold over gelatin background by modulating integrin activity (Coopman et al., 1996). Results for a typical experiment are shown in Figure 9.13.1.
Time Considerations The protocol for preparation of FITC-gelatin takes several days. Preparation of beads takes ∼1 hr. However, both can be prepared in advance and stored for several days in serumfree growth medium plus antibiotics at 4°C or in 5% glycerol at −20°C. The preparation of coated plates for the phagocytosis assay can be accomplished in <2 hr. Incubation of cells on gelatin is usually between 3 and 72 hr. Cell harvesting to the point
of reading the samples by flow cytometry also takes ∼1 hr. Thus, the whole assay in the Basic Protocol can be done in a day. However, the assay is flexible and includes several points at which overnight incubation steps can be introduced. For example, the FITC-gelatin-coated plate can be made the day before it is required and left to quench overnight at 4°C. The preparation of FITC-gelatin-coated beads can be accomplished in 2 to 3 hr. Once they are made, the assay conditions are equivalent to those performed on coated plates; thus, assays performed using beads are subject to the same time condiderations.
Literature Cited Bjerknes, R., Bjerkvig, R., and Laerum, O.D. 1987. Phagocytic capacity of normal and malignant rat glial cells in culture. J. Natl. Cancer Inst. 2:279288. Brown, E.J. 1995. Phagocytosis. Bioessays 2:109-117. Coopman, P.J., Thomas, D.M., Gehlsen, K.R., and Mueller, S.C. 1996. Integrin α3β1 participates in the phagocytosis of extracellular matrix molecules by human breast cancer cells. Mol. Biol. Cell 11:1789-1804. Coopman, P.J., Do, M.T.H., Thompson, E.W., and Mueller, S.C. 1998. Phagocytosis of cross-linked gelatin matrix by human breast carcinoma cells correlates with their invasive capacity. Clin. Cancer Res. 4:507-515.
Studies of Cell Function
9.13.7 Current Protocols in Cytometry
Supplement 12
Imamura, F., Horai, T., Mukai, M., Shinkai, K., and Akedo, H. 1991. Serum requirement for in vitro invasion by tumor cells. Jpn. J. Cancer Res. 5:493-496.
Ruoslahti, E. 1992. The Walter Herbert Lecture. Control of cell motility and tumour invasion by extracellular matrix interactions. Br. J. Cancer 2:239-242.
Itoh, K., Yoshioka, K., Akedo, H., Uehata, M., Ishizaki, T., and Narumiya, S. 1999. An essential part for Rho-associated kinase in the transcellular invasion of tumor cells. Nature Med. 2:221-225.
Stetler-Stevenson, W.G., Aznavoorian, S., and Liotta, L.A. 1993. Tumor cell interactions with the extracellular matrix during invasion and metastasis. Annu. Rev. Cell Biol. 9:541-573.
Lochter, A. and Bissell, M.J. 1995. Involvement of extracellular matrix constituents in breast cancer. Semin. Cancer Biol. 3:165-173.
Van Peteghem, M.C., Mareel, M.M., and De Bruyne, G.K. 1980. Phagocytic capacity of invasive malignant cells in three-dimensional culture. Virchows Arch. B Cell. Pathol. Mol. Pathol. 2:193-204.
MacDougall, J.R. and Matrisian, L.M. 1995. Contributions of tumor and stromal matrix metalloproteinases to tumor progression, invasion and metastasis. Cancer Metast. Rev. 4:351-362. Montcourrier, P., Mangeat, P.H., Valembois, C., Salazar, G., Sahuquet, A., Duperray, C., and Rochefort, H. 1994. Characterization of very acidic phagosomes in breast cancer cells and their association with invasion. J. Cell Sci. 107:2381-2391.
Contributed by Emma T. Bowden and Susette Mueller Georgetown University Medical Center Washington, D.C.
Montgomery, A.M.P., Reisfeld, R.A., and Cheresh, D.A. 1994. Integrin αvβ3 rescues melanoma cells from apoptosis in three-dimensional dermal collagen. Proc. Natl. Acad. Sci. U.S.A. 91:88568860.
Peter J. Coopman Montpellier University II Montpellier, France
In Vitro Invasion Assays: Phagocytosis of the Extracellular Matrix
9.13.8 Supplement 12
Current Protocols in Cytometry
Flow Cytometric Analysis of Mitochondrial Membrane Potential Using JC-1 The role of mitochondria in several biological processes is receiving growing attention. In particular, there is interest in, and some controversy over, the possible impairment of the functionality of this organelle during the process of programmed cell death/apoptosis (Cossarizza et al., 1995, 1999; Skulachev, 1996). According to some authors, the collapse in mitochondrial membrane potential (∆Ψ) is one of the early events, if not the cause, of apoptosis (Kroemer et al., 1997). Others suggest that functional mitochondria are crucial for apoptosis to occur. Accurate measurements of ∆Ψ, the analysis of ATP production, and morphological observations indicate good functionality of the cell and of these organelles during the first phases of apoptosis (Cossarizza et al., 1994; Tiso et al., 1995; Richter et al., 1996). Several techniques are used to investigate the role of this organelle, including classical biochemical or molecular biology methods, but flow cytometry appears to play the most important role. However, researchers are utilizing various fluorescent dyes, some of which are not completely adequate to the analysis of ∆Ψ. This unit presents a protocol based on staining with 5,5′,6,6′-tetrachloro-1,1′,3,3′tetraethylbenzimidazolylcarbocyanine iodide (JC-1) and describes the advantages and disadvantages of this and certain other probes.
UNIT 9.14
BASIC PROTOCOL
This protocol presents one method for staining cells with JC-1 and for subsequent analysis of the cells by flow cytometry. The JC-1 dye undergoes a reversible change in fluorescence emission from green to greenish orange as ∆Ψ increases. Measurements made using this dye can be both qualitative, based on the shift in fluorescence emission, and quantitative, based on the intensity of fluorescence emission. Cells with high ∆Ψ will form JC-1 aggregates and fluoresce red; those with low ∆Ψ will contain monomeric JC-1 and fluoresce green (Reers et al., 1991). CAUTION: DMSO enhances the uptake of dye through the skin. Wearing nitrile gloves and eye protection during all stages of sample handling is strongly recommended. Medical advice should be sought if dye or dye solution is inhaled or ingested. For disposal, all staining solutions should be poured through a funnel with a filter containing activated charcoal. When the solution passing through becomes fluorescent, the filter should be incinerated and a fresh filter installed. Filtrate should be disposed of according to institutional guidelines; if it is nonfluorescent, it can usually be poured down the sink. Materials Cells in suspension (e.g., human peripheral blood lymphocytes, monocytes, or human tumor cell lines such as HL-60, U937, K562; APPENDIX 3B) Complete cell culture medium (e.g., complete RPMI with 10% FBS; APPENDIX 2A) Depolarizing drug: e.g., 0.1 mM valinomycin in DMSO or 0.25 mM carbonyl cyanide m-(trifluoromethoxy)phenylhydrazone (FCCP) in DMSO 1 mg/ml JC-1 stock solution in dimethyl sulfoxide (DMSO); store in small aliquots at −20°C for up to 1 year PBS (APPENDIX 2A) Flow cytometer equipped with a 488-nm excitation light source and band-pass filters centered around 525 and 590 nm
Studies of Cell Function Contributed by Andrea Cossarizza and Stefano Salvioli Current Protocols in Cytometry (2000) 9.14.1-9.14.7 Copyright © 2000 by John Wiley & Sons, Inc.
9.14.1 Supplement 13
104
B
104
A
103 102
100
101
102
103
100
101
102
103
valinomycin
101 100
JC-1 orange fluorescence
control
104 100 101 JC-1 green fluorescence
102
103
104
Figure 9.14.1 HL-60 cells, (A) control and (B) depolarized (treated with 100 nM valinomycin), stained with 2.5 µg/ml JC-1. Note the shift to the bottom and to the right by cells with depolarized mitochondria.
Stain cells with JC-1 1. Harvest ≥2 × 105 cells and adjust total volume to 1 ml with prewarmed (37°C) fresh complete cell culture medium. Prepare a positive control sample, in which mitochondria of all cells have been depolarized with drugs able to collapse ∆Ψ. Stain and analyze control and experimental samples in parallel. This protocol is intended for cells such as lymphocytes and monocytes, or for cell lines such as HL-60 or U937. Other cell types may require different staining conditions. Nevertheless, all the other cells used by the authors so far (human fibroblasts, keratinocytes, sperm cells, tumor cell lines of different origins, murine hepatocytes, thymocytes and astrocytes, hemocytes from invertebrates) and also isolated mitochondria responded quite well to this staining procedure. Adjustments to the protocol described below may often prove negligible. Typically, the K+ ionophore valinomycin (≥100 nM) or carbonyl cyanide m-(trifluoromethoxy)phenylhydrazone (FCCP, 250 nM) can be used. This treatment will immediately result in a dramatic change of the fluorescence distribution to that of the depolarized state (Fig. 9.14.1), and is needed to adjust the electronic fluorescence compensation.
2. Thaw an aliquot of 1 mg/ml JC-1 stock solution and add 2.5 µl to 1 ml cell suspension (final 2.5 µg/ml), while vortexing. Vortex cell suspension until dye is well dissolved and a uniform red-violet color is obtained. JC-1 tends to form aggregates when added to aqueous media. To avoid this, add the probe while vortexing. Do not refreeze unused stock.
3. Incubate samples 10 min in the dark at 37°C.
Analysis of Mitochondrial Membrane Potential Using JC-1
During the staining procedure, all reagents must be kept at room temperature and carefully checked for pH (7.4), since mitochondrial ∆Ψ is very sensitive to temperature and pH. The staining procedure must be performed under dimmed light and incubation must be in the dark, because of the light sensitivity of JC-1.
4. Wash cell suspension by adding 2 ml PBS and centrifuging 5 min at 500 × g, room temperature. Discard supernatant and repeat wash.
9.14.2 Supplement 13
Current Protocols in Cytometry
104
B
104
A control 103
103
valinomycin
100
102
101
102
103
101
101
monocytes
100
102
lymphocytes
100
JC-1 orange fluorescence
lymphocytes
104 100 101 JC-1 green fluorescence
monocytes
102
103
104
Figure 9.14.2 After incubation with 2.5 µg/ml JC-1 at 37°C, peripheral blood mononulear cells usually show a single population of monocytes, while here lymphocytes give rise to two separate peaks. This could indicate the presence of a functional heterogeneity within lymphocytes (A). Such a phenomenon, however, is not always present. Treating cells with valinomycin results in a loss of orange signal either from lymphocytes or from monocytes (B).
5. Resuspend cells in 0.3 ml PBS. Set up flow cytometer 6. Adjust a flow cytometer equipped with a 488-nm light source using standard procedures. For detection of JC-1 dye, use a band-pass filter centered around 525 nm (green fluorescence) and a band-pass filter centered around 590 nm (orange fluorescence). Use logarithmic signal amplification with typical green-orange electronic signal compensation near 4% and orange-green electronic signal compensation around 10%. Analyze JC-1 stained cells 7. Acquire data from experimental samples in listmode format, using logarithmic signal amplification for the green and the orange channels. Do not acquire data at ≥250 to 300 signals/second. Cells with high ∆Ψ form J-aggregates and therefore show high red fluorescence. In cells with low ∆Ψ, JC-1 exists in its monomeric form, thus showing only green fluorescence. Usually, the green fluorescence of depolarized cells is a little bit higher than that of polarized cells because of the presence of higher amounts of JC-1 monomers. When the sample contains a heterogeneous cell population, it is possible to see different fluorescence patterns due to the variable content in membranes and mitochondria of cell subpopulations. This is typically the case of peripheral blood mononuclear cells (PBMC), formed by lymphocytes and monocytes (Fig. 9.14.2), the former being smaller and having less mitochondrial content than the latter. As a result, the fluorescence pattern of JC-1 from such samples shows at least two distinct peaks, one corresponding to lymphocytes, and another, brighter in both green and orange fluorescence, corresponding to monocytes. In some cases, especially after incubation at 37°C, lymphocytes give rise to two separate peaks, indicating a consistent functional heterogeneity.
8. If desired, isolate cells by flow sorting on the basis of ∆Ψ level.
Studies of Cell Function
9.14.3 Current Protocols in Cytometry
Supplement 13
COMMENTARY Background Information
Analysis of Mitochondrial Membrane Potential Using JC-1
Intracellular energy produced by the mitochondrial respiratory chain is stored as an electrochemical gradient consisting of a transmembrane electrical potential (∆Ψ) of ∼180 to 200 mV (negative inside), and a proton gradient of ∼1 unit. This energy drives the synthesis of ATP. ∆Ψ has been studied with a variety of membrane-permeable lipophilic cations, which are accumulated by living cells, organelles, and liposomes exhibiting a negative interior membrane potential. These compounds, which exhibit either optical or fluorescence activity after accumulation into energized systems, include probes such as 3,3′-dihexyloxadicarbocyanine iodide [DiOC6(3)], nonylacridine orange (NAO), safranine O, rhodamine 123, as well as r adiolabeled probes (e.g., [3H]methyltriphenylphosphonium) and unlabeled probes used with specific electrodes (e.g., tetraphenylphosphonium ion, TPP+). These systems have several possible disadvantages, including (1) the time required to achieve equilibrium distribution of a mitochondrial membrane probe; (2) the degree of passive (nonspecific) binding of probes to a membrane component; (3) the toxic effects of probes on mitochondrial functional integrity; (4) timeconsuming sampling procedures (preparation of isolated organelles or other samples); (5) interference from light-scattering changes and from absorption changes of mitochondrial components; and (6) requirement for large amounts of biological materials. Examples of problems with nonspecific binding include NAO, which binds to cardiolipin (Maftah et al., 1989); rhodamine 123, which has several energy-independent binding sites (Lopez-Mediavilla et al., 1989); and DiOC6(3), which has been widely used for studies on ∆Ψ, despite its high capacity to bind other membranes than those of mitochondria and its low sensitivity to agents capable of depolarizing such organelles (Terasaki et al., 1984; Jenssen et al., 1986; Salvioli et al., 1997). A TPP electrode affords an easy and precise tool to measure ∆Ψ due to the low interference between bound TPP+ and the membrane and to the lack of electrode response to species other than TPP+. However, this method requires large amounts of biological samples, and uptake of this lipophilic cation by intact mammalian cells is indeed a slow process.
Rhodamine 123 was the first probe used to to assess ∆Ψ in living cells (Johnson et al., 1980; Darzynkiewicz et al., 1981; Goldstein and Korczack, 1981; Collins and Foster, 1983; Nadakavukaren et al., 1985). Typically, the signal coming from cells whose mitochondria have a low potential has to be much lower than that of control samples, and, in a classical histogram, depolarized populations should go to the left. However, the shift in rhodamine 123 fluorescence is small, and the controls and treated cell distributions are not always separate. The operator must thus decide “by eye” where the population of cells with depolarized mitochondria begins (Fig. 9.14.3). Rhodamine 123 binding to mitochondria is also difficult to calculate when the cell has mitochondrial heterogeneity due, for example, to a high number of mature or immature organelles, as occurs in a continuously growing cell line. Moreover, different mitochondrion binding sites exist for rhodamine 123, i.e., sites that are freely accessible whatever the energy status of the mitochondria, and sites that are hidden in the energized state and freely accessible in the de-energized form of the organelles. This situation has been attributed to different maturative states of the organelles (Lopez-Mediavilla et al., 1989). Thus, organelles in a single cell can have different rhodamine 123 binding sites and different fluorescence emissions. Consequently, it is very difficult to ascertain whether mitochondria bind rhodamine 123 in an energydependent or energy-independent manner, and it is possible to obtain a false lack of depolarization in cells treated with agents such as valinomycin. Additionally, some tumor cells expel this dye rapidly because of the high activity of their multidrug resistance pump. Therefore, the authors suggest that rhodamine 123 has both poor sensitivity and poor specificity (Salvioli et al., 1997). DiOC6(3) was first used for the analysis of plasma membrane potential (Jenssen et al., 1986). In fact, no formal demonstration of the specificity and sensitivity of this probe exists in intact cells; however, the first observations in cells, relative to doses of 200 nM, indicated a specific localization in the endoplasmic reticulum (Terasaki et al., 1984). Used at 40 nM, DiOC6(3) was thought to detect ∆ψ changes in intact cells (Petit et al., 1995). In the authors’ hands, DiOC6(3) reacted properly
9.14.4 Supplement 13
Current Protocols in Cytometry
150 Events
control
0
valinomycin
100
101
102
103
104
Rhodamine 123 fluorescence
128
Figure 9.14.3 HL-60 cells treated with 100 nM valinomycin and stained with 10 µg/ml rhodamine 123. Note the (false) lack of depolarization in cells treated with valinomycin.
Events
control
0
valinomycin
100
101
102
103
104
DiOC6(3) fluorescence
Figure 9.14.4 HL-60 cells treated with 100 nM valinomycin and stained with 40 nM DiOC6(3). Note the (false) lack of depolarization in cells treated with valinomycin.
when U937 cells were treated with FCCP, but such behavior was not observed in cells treated with valinomycin (Salvioli et al., 1997). The same was observed with other cell lines, such as HL-60 (Fig. 9.14.4), and many other cell types. Moreover, when cells are kept in the presence of plasma membrane–depolarizing agents such as ouabain or high doses of extracellular K+, a consistent decrease in
DiOC6(3) fluorescence occurs, indicating a consistent sensitivity of the probe for the plasma membrane potential. A review of fluorescent dyes currently available for flow cytometric studies of mitochondria can be found in UNIT 9.4. To overcome the drawbacks of the dyes decribed above, the authors have developed a new technique to detect variations in ∆Ψ at the
Studies of Cell Function
9.14.5 Current Protocols in Cytometry
Supplement 13
single-cell or single-organelle level using the lipophilic cation JC-1 (Cossarizza et al., 1993, 1996). The fluorescence emission of the JC-1 dye changes depending on ∆Ψ, due to the reversible formation of JC-1 aggregates upon membrane polarization, which causes shifts in emitted light from ∼530 nm (i.e., emission of JC-1 monomeric form) to ∼590 nm (i.e., emission of J-aggregates) when excited at 488 nm. The color of the dye changes reversibly from green to greenish orange as the mitochondrial membrane becomes more polarized (Hada et al., 1977; Reers et al., 1991; Smiley et al., 1991). Aggregate formation begins at potential values on the order of ∼80 to 100 mV, and reaches its maximum at ∼200 mV. The main advantage of the use of JC-1 is that it can be both qualitative, by considering the shift from green to orange fluorescence emission, and quantitative, by measuring the absolute values of green and orange fluorescence emission. A disadvantage is that its fluorescence emission spans the green, yellow, and part of the red wavelengths of the spectrum, and therefore it is highly unlikely that JC-1 can be used in combination with other fluorochromes.
Critical Parameters and Troubleshooting
Stock solutions of JC-1 are stable at −20°C, but deteriorate upon several rounds of thawing and refreezing. Since ∆Ψ depends on temperature, it is paramount to stain cells at 37°C.
Anticipated Results
Changes in ∆Ψ can occur in a variety of physiopathological conditions and may depend on the cell model that is used. Studies on apoptosis are quite complex, as many types of triggers for cell death exist and have different intracellular targets. To test the sensitivity of JC-1 fluorescence toward changes in ∆Ψ, the authors treated HL-60 cells with valinomycin and subsequently performed cell staining and flow cytometry as described in the protocol (Fig. 9.14.1). Figure 9.14.2 shows the complex patterns of JC-1 fluorescence obtained with a sample of peripheral blood mononuclear cells.
Time Considerations
Analysis of Mitochondrial Membrane Potential Using JC-1
In general, the staining described requires <1 hour, including washing of the cell suspension. Data acquisition takes a few minutes per sample.
Literature Cited Collins, J.M. and Foster, K.A. 1983. Differentiation of promyelocytic (HL-60) cells into mature granulocytes: Mitochondrial-specific rhodamine 123 fluorescence. J. Cell Biol. 96:94-99. Cossarizza, A., Baccarani Contri, M., Kalashnikova, G., and Franceschi, C. 1993. A new method for the cytofluorimetric analysis of mitochondrial membrane potential using the J-aggregate forming lipophilic cation 5,5′,6,6′-tetrachloro1,1′,3,3′-tetraethylbenzimidazolcarbocyanine iodide (JC-1). Biochem. Biophys. Res. Commun. 197:40-45. Cossarizza, A., Kalashnikova, G., Grassilli, E., Chiappelli, F., Salvioli, S., Capri, M., Barbieri, D., Troiano, L., Monti, D., and Franceschi, C. 1994. Mitochondrial modifications during rat thymocyte apoptosis: A study at the single cell level. Exp. Cell Res. 214:323-330. Cossarizza, A., Salvioli, S., Franceschini, M.G., Kalashnikova, G., Barbieri, D., Monti, D., Grassilli, E., Tropea, F., Troiano, L., and Franceschi, C. 1995. Mitochondria and apoptosis: A cytofluorimetric approach. Fund. Clin. Immunol. 3:67-68. Cossarizza, A., Ceccarelli, D., and Masini, A. 1996. Functional heterogeneity of isolated mitochondrial population revealed by cytofluorimetric analysis at the single organelle level. Exp. Cell Res. 222:84-94. Cossarizza, A., Mussini, C., Borghi, V., Mongiardo, N., Nuzzo, C., Pedrazzi, J., Benatti, F., Moretti, L., Pinti, M., Franceschi, C., and De Rienzo, B. 1999. Apoptotic features of peripheral blood mononuclear granulocytes and monocytes during primary, acute HIV infection. Exp. Cell Res. 247:304-311. Darzynkiewicz, Z., Staiano-Coico, L., and Melamed, M.R. 1981. Increased mitochondrial uptake of rhodamine 123 during lymphocyte stimulation. Proc. Natl. Acad. Sci. U.S.A. 78:2383-2387. Goldstein, S. and Korczack, L.B. 1981. Status of mitochondria in living human fibroblasts during growth and senescence in vitro: Use of the laser dye rhodamine 123. J. Cell Biol. 91:392-398. Hada, H., Honda, C., and Tanemura, H. 1977. Spectroscopic study on the J-aggregate of cyanine dyes. I. Spectral changes of UV bands concerned with J-aggregate formation. Photogr. Sci. Eng. 21:83-91. Jenssen, H.-L., Redmann, K., and Mix, E. 1986. Flow cytometric estimation of transmembrane potential of macrophages—A comparison with microelectrode measurements. Cytometry 7:339-346. Johnson, L.V., Walsh, M.L., and Chen, L.B. 1980. Localization of mitochondria in living cells with rhodamine 123. Proc. Natl. Acad. Sci. U.S.A. 77:990-994.
9.14.6 Supplement 13
Current Protocols in Cytometry
Kroemer, G., Zamzani, N., and Susin, S.A. 1997. Mitochondrial control of apoptosis. Immunol. Today 18:44-51. Lopez-Mediavilla, C., Orfao, A., Gonzales, M., and Medina, J.M. 1989. Identification by flow cytometry of two distinct rhodamine-123-stained mitochondrial populations in rat liver. FEBS Lett. 254:115-120. Maftah, A., Petit, J.M., Ratinaud, M.H., and Julien, R. 1989. 10-N-Nonyl-acridine orange: A fluorescent probe which stains mitochondria independently of their energetic state. Biochem. Biophys. Res. Commun. 164:185-190. Nadakavukaren, K.K., Nadakavukaren, J.J., and Chen, L.B. 1985. Incresased rhodamine 123 uptake by carcinoma cells. Cancer Res. 45:60936099. Petit, P.X., Lecoeur, H., Zorn, E., Dauguet, C., Mignotte, B., and Gougeon, M.-L. 1995. Alterations in mitochondrial structure and function are early events of dexamethasone-induced thymocyte apoptosis. J. Cell Biol. 130:157-167. Reers, M., Smith, T.W., and Chen, L.B. 1991. J-aggregate formation of a carbocyanine as a quantitative fluorescent indicator of membrane potential. Biochemistry 30:4480-4486. Richter, C., Schweizer, M., Cossarizza, A., and Franceschi, C. 1996. Control of apoptosis by the cellular ATP level. FEBS Lett. 378:107-110.
rhodamine 123, is a reliable fluorescent probe to assess ∆Ψ in intact cells. Implications for studies on mitochondrial functionality during apoptosis. FEBS Lett. 411:77-82. Skulachev, V.P. 1996. Why are mitochondria involved in apoptosis? Permeability transition pores and apoptosis as selective mechanisms to eliminate superoxide-producing mitochondria and cell. FEBS Lett. 397:7-10. Smiley, S.T., Reers, M., Mottola-Hartshorn, C., Lin, M., Chen, A., Smith, T.W., Steele, G.D., and Chen, L.B. 1991. Intracellular heterogeneity in mitochondrial membrane potential revealed by a J-aggregate-forming lipophilic cation JC-1. Proc. Natl. Acad. Sci. U.S.A. 88:3671-3675. Terasaki, M., Song, J., Wong, J.R., Weiss, M.J., and Chen, B.L. 1984. Localization of endoplasmic reticulum in living and glutaraldehyde-fixed cells with fluorescent dyes. Cell 38:101-108. Tiso, M., Gangemi, R., Bargellesi-Severi, A., Pizzolitto, S., Fabbi, M., and Risso, A. 1995. Spontaneous apoptosis in human thymocytes. Am. J. Pathol. 147:434-444.
Contributed by Andrea Cossarizza and Stefano Salvioli University of Modena Modena, Italy
Salvioli, S., Ardizzoni, A., Franceschi, C., and Cossarizza, A. 1997. JC-1, but not DiOC6(3) or
Studies of Cell Function
9.14.7 Current Protocols in Cytometry
Supplement 13
Multiparameter Analysis of Physiological Changes in Apoptosis
UNIT 9.15
Changes in mitochondrial parameters leading to cells with compromised mitochondrial function are a hallmark of apoptosis. Flow cytometric assays of mitochondrial parameters allow one to detect these changes and to quantitate the number of compromised cells. These assays involve staining the cells with fluorescent dyes that monitor specific features of the mitochondrion such as membrane potential, mitochondrial electron transport, oxidation of mitochondrial cardiolipin, and altered cellular redox status. Since the fluorescence of each individual dye covers only part of the visible spectrum, it is possible to combine several dyes in a single assay. In this way, it becomes possible to perform multiparameter assays of physiological changes during apoptosis and to investigate whether or not certain changes occur simultaneously in the same cells. Three protocols are presented here: a combined assay for NADPH levels and mitochondrial membrane potential (see Basic Protocol 1), a protocol for cell cycle stage–specific apoptosis (see Basic Protocol 2), and a protocol for cellular thiol and mitochondrial cardiolipin levels (see Basic Protocol 3). Samples can be split into three aliquots for analysis by each of the three protocols. COMBINED ASSAY FOR NADPH LEVELS AND MITOCHONDRIAL MEMBRANE POTENTIAL
BASIC PROTOCOL 1
This protocol takes advantage of the fact that NADPH emits blue fluorescence upon excitation with ultraviolet light. This is combined with the green and red fluorescence of MitoTracker Green and CMXRosamine, respectively. This latter combination allows better resolution of apoptotic and “normal” cells than does staining with CMXRosamine alone. CAUTION: DMSO and dye solutions are potentially toxic to humans. Use (nitrile) gloves and wear eye protection at all stages of handling. Seek medical advice if dye or dye solutions are ingested or inhaled. The dyes mentioned are for in vitro use only; do not administer either externally or internally. All staining solutions should be poured through a funnel with a filter containing activated charcoal in a fume hood; the dye-free solution may be disposed of in the sink. When the passing solution becomes fluorescent, the filter should be incinerated or disposed of according to applicable rules for environmental hygiene and a fresh filter should be installed. NOTE: To maintain cells in a physiologically optimal state and to avoid photochemical damage to the stained cells, all incubations are performed at 37°C with subdued light in fresh culture medium. All solutions should be prewarmed to 37°C, prior to analysis, to conserve cell physiology and function. NOTE: All protocols described in this unit have been performed on cultured animal cells; limited data exist regarding the use of these methods for plant cells and in yeast. Expertise is assumed for basic techniques in flow cytometry as well as in cell culture and harvesting of both suspension cultures and adherent cells. Materials Cultured cells (APPENDIX 3B) Culture medium (optimal for cell-type), 37°C 200 µM stock solutions of MitoTracker Green FM and CMXRosamine in DMSO; store at −20°C in the dark Contributed by Martin Poot Current Protocols in Cytometry (2000) 9.15.1-9.15.7 Copyright © 2000 by John Wiley & Sons, Inc.
Studies of Cell Function
9.15.1 Supplement 14
15-ml screw-capped centrifuge tubes 37°C water bath with cover (e.g., a sheet of aluminum) 12 × 75–mm polypropylene tubes Flow cytometer with either a mercury arc lamp or two argon lasers (tuned to ultraviolet and to 488 nm) as excitation source Computer for data collection and processing Stain cells 1. Harvest cultured cells by standard procedures in 15-ml scew-capped centrifuge tubes and centrifuge 5 min at 200 × g, room temperature. Resuspend the cell pellet at a density of 0.5–1.0 × 106 cells per ml in prewarmed cell culture medium. Incubate cell suspensions in a 37°C water bath with cover ≥5 min. Since the functional state of the mitochondria is to be monitored, it is advisable to keep cell suspensions at their optimal temperature (i.e., 37°C) and to allow them to recover for at least 5 min after harvesting.
2. Thaw 200 µM stock solutions of MitoTracker Green FM and CMXRosamine in DMSO at room temperature, keeping them protected from light (e.g., in a drawer). IMPORTANT NOTE: Dye solutions decompose rapidly if exposed to light.
3. Transfer 1-ml aliquots of prewarmed cell suspension (step 1) to 12 × 75–mm polypropylene tubes. Add 1 µl each of the 200 µM MitoTracker Green FM and CMXRosamine stock solutions. Mix immediately by vortexing briefly at maximum speed. Incubate for 15 to 30 min in a 37°C water bath in the dark or under subdued light. After staining, store tubes with cell suspensions in a melting-ice bath until use (up to 1 hr). Final dye concentrations in the range of 100 to 200 nM are recommended, because at higher concentrations nonmitochondrial staining may occur.
Set up and optimize the flow cytometer 4. To excite samples stained with MitoTracker Green FM and CMXRosamine, tune the argon-ion laser to the 488-nm line. Use a band-pass filter centered at ∼530 nm to collect fluorescence from MitoTracker Green FM and a 630-nm long-pass filter for CMXRosamine. (Due to the wide variation in cellular contents of mitochondria, it is advisable to use logarithmic signal amplification for the signal channels collecting mitochondria-related fluorescence.) Carefully resuspend the cell sample by gently pipetting up and down a few times immediately before analysis. Collect data from at least 10,000 cells, preferably from 20,000 or more. Ultraviolet (360 nm)-excited blue autofluorescence, collected with a 450-nm-centered band-pass filter, is proportional to cellular NAD(P)H content (see Critical Parameters and Troubleshooting). If quantitative evaluation of NAD(P)H content is required, an untreated control sample has to be included; the fluorescence of this sample is set as 100%. A zero NAD(P)H value can be generated by analyzing a sample of cells fixed in 70% ethanol, which is devoid of NAD(P)H.
Multiparameter Analysis of Physiological Changes in Apoptosis
Cells stained with MitoTracker Green FM show maximal emission at 516 nm. Cells stained with CMXRosamine show maximal absorption at 594 nm and maximal emission at 608 nm; they also exhibit significant absorption in the UV region of the spectrum and may be excitable with a mercury arc lamp. During the staining period cells tend to clump; therefore, to obtain meaningful data on a per-cell basis, it is essential to disaggregate cells by pipetting vigorously immediately before analysis.
9.15.2 Supplement 14
Current Protocols in Cytometry
ASSAY FOR CELL CYCLE STAGE–SPECIFIC APOPTOSIS This protocol combines the ultraviolet-excited blue fluorescence of Hoechst 33342 dye, which resolves cells according to the G1, S, and G2 stages of the cell cycle, with the green and red fluorescence of MitoTracker Green FM and CMXRosamine.
BASIC PROTOCOL 2
Materials Cells in suspension (see Basic Protocol 1, step 1) 200 µM stock solutions of MitoTracker Green FM and CMXRosamine in DMSO; store at −20°C in the dark 1 mM stock solution of Hoechst 33342 dye in double-distilled water; store at 4°C in the dark (do not freeze) 12 × 75–mm polypropylene tubes 37°C water bath with cover (e.g., a sheet of aluminum) Flow cytometer with either a mercury arc lamp or two argon lasers (tuned to ultraviolet and to 488 nm) as excitation source Computer for data collection and processing NOTE: Do not use phosphate-containing buffers, since Hoechst dyes precipitate with phosphates. Stain cells 1. Harvest cells and thaw 200 µM stock solutions of MitoTracker Green FM and CMXRosamine (see Basic Protocol 1, steps 1 and 2). 2. Transfer 1-ml aliquots of prewarmed cell suspension (step 1) to 12 × 75–mm polypropylene tubes. Add 20 µl 1 mM Hoechst 33342 and 1 µl each of 200 µM MitoTracker Green FM and CMXRosamine dye stock solutions. Mix immediately by vortexing briefly at maximal speed. Incubate 30 min in a 37°C water bath in the dark or under subdued light. After staining, put tubes with cell suspensions in a melting-ice bath until use. Smaller volumes can be used; however, final MitoTracker Green FM and CMXRosamine dye concentrations in the range of 100 to 200 nM are recommended, because at higher concentrations nonmitochondrial staining may occur. A final Hoechst 33342 dye concentration of 20 ìM to saturate cellular DNA is recommended, as some cell types may have multidrug resistance–like membrane-resident dye-pumping mechanisms, which will lower the actual dye concentration inside the cell. When using a novel cell type it may be useful to titrate the Hoechst 33342 dye concentration to determine optimal staining conditions.
Set up and optimize the flow cytometer 3. Set up and optimize the flow cytometer (see Basic Protocol 1). Collect fluorescence from the MitoTracker Green FM and CMXRosamine dyes with the same filters as described above (see Basic Protocol 1, step 4). Ultraviolet (360 nm)-excited blue fluorescence from the Hoechst 33342 dye, collected with a 450-nm-centered band-pass filter, is proportional to cellular DNA content (see Critical Parameters and Troubleshooting).
Studies of Cell Function
9.15.3 Current Protocols in Cytometry
Supplement 14
BASIC PROTOCOL 3
COMBINED ASSAY FOR CELLULAR REDUCED THIOL AND CARDIOLIPIN LEVELS In this protocol the ultraviolet-excited blue fluorescence of monobromobimane, for the detection of cellular reduced thiol, is combined with the green and red fluorescence of nonyl acridine orange (NAO), for detection of cardiolipin. Materials Cells in suspension (see Basic Protocol 1, step 1) 10 mM stock solution of monobromobimane in absolute ethanol; store at 4°C in the dark 1 mM stock solution of nonyl acridine orange (NAO) in absolute ethanol; store at 4°C in the dark 12 × 75–mm polypropylene tubes 37°C water bath with cover (e.g., a sheet of aluminum) Flow cytometer with either a mercury arc lamp or two argon lasers (tuned to ultraviolet and to 488 nm) as excitation source Computer for data collection and processing 1. Harvest and suspend cells (see Basic Protocol 1, step 1). 2. Transfer 1-ml aliquots of prewarmed cell suspension into 12 × 75–mm polypropylene tubes. Add 5 µl 10 mM monobromobimane and 1 µl 1 mM NAO stock solutions in absolute ethanol. Mix immediately by briefly vortexing at maximal speed. Incubate 30 min at 37°C in the dark or under subdued light. After staining, put tubes with cell suspensions in a melting-ice bath until use. 3. Set up and optimize the flow cytometer as described (see Basic Protocol 1). Collect green and red fluorescence from the NAO dye with the same filters as described above (see Basic Protocol 1, step 4). Ultraviolet (360 nm)-excited blue fluorescence (collected with a 450-nm-centered bandpass filter) from the monobromobimane dye is proportional to cellular reduced thiol content. The red fluorescence of NAO-stained cells is proportional to mitochondrial cardiolipin content (see Critical Parameters and Troubleshooting).
COMMENTARY Background Information
Multiparameter Analysis of Physiological Changes in Apoptosis
Key events taking place during apoptosis include loss of mitochondrial membrane potential, decrease in mitochondrial electron transport, oxidation of mitochondrial cardiolipin, and altered cellular redox status (Green and Kroemer, 1998; Green and Reed, 1998). After staining cells with fluorescent dyes such as JC-1, CMXRosamine, and DiOC6(3), changes in mitochondrial parameters were demonstrated by flow cytometry (Cossarizza et al., 1994; Macho et al., 1996; Backway et al., 1997). Initially it was not clear whether the changes observed by flow cytometry were independent of each other or components of one or several pathways of apoptosis. By staining cells simultaneously with dyes for several biochemical parameters that exhibit distinct fluo-
rescence emission ranges, Backway et al. (1997) demonstrated in leukemic blast cells that loss of mitochondrial membrane potential took place after glutathione was depleted. This loss of mitochondrial membrane potential preceded intracellular formation of reactive oxygen. By performing multiparameter flow cytometry, the authors were able to map a “death sequence.” Poot and Pierce (1999) further expanded this concept and showed that camptothecin treatment of human lymphoblastoid cells led to the formation of a cell subpopulation that underwent a simultaneous decrease in mitochondrial membrane potential, NADH level, and oxidative turnover. Cells stained with NAO show strong fluorescence in the green, yellow, and red regions of the spectrum; therefore, NAO cannot be
9.15.4 Supplement 14
Current Protocols in Cytometry
A
B 1000 “normal”
G1
CMXRosamine fluorescence
“normal” 100 S
10
G2
C
“apoptotic”
G1
“apoptotic”
sub-G1 S
1
G2 0.1 0.1
1
10
100
Mito Tracker Green FM fluorescence
1000
DNA content (Hoechst 33342)
Figure 9.15.1 Cytogram of MitoTracker Green FM and CMXRosamine fluorescence (A) and cell cycle histograms of “normal” (B) and “apoptotic” (C) cells from a culture of human lymphoblastoid cells that were treated with 1 µg/ml mitomycin C 8 hr at 37°C, harvested, and stained according to Basic Protocol 2. Apoptotic cells show a lower CMXRosamine and a higher MitoTracker Green FM fluorescence. The two right hand panels show the cell cycle distributions, based on the simultaneous Hoechst 33342 fluorescence, of the “normal” (B) and “apoptotic” (C) cells.
combined with any other dye that emits in the red or far-red part of the spectrum. The red fluorescence of NAO is proportional to the mitochondrial cardiolipin content (Gallet et al., 1995); however, during apoptosis the ratio of red to green fluorescence of NAO appears to decrease (Pierce et al., unpub. observ.).
Critical Parameters and Troubleshooting Dye combinations In addition to precautions to be taken when performing assays of viable cell function (described in UNIT 9.4), one has to be aware of a specific problem that may occur when several dyes are used in combination, i.e., their fluorescence emission spectra may overlap. In case quantitative information is required (i.e., the amount of fluorescence has to reflect a biochemical quantity), samples stained with a single dye should be included for use in setting fluorescence compensation. The ultraviolet-excitable blue autofluorescence of unstained cells reflects the mitochondrial NAD(P)H level (Thorell, 1983). This level
of NAD(P)H proved in itself to be a useful apoptosis-related parameter (Poot and Pierce, 1999). After reaction with reduced thiols, monobromobimane generates the same blue emission as NAD(P)H (Poot et al., 1986); therefore, measurements of NAD(P)H and cellular reduced thiol levels after staining with monobromobimane cannot be performed in the same sample. If the level of cellular reduced thiol is to be derived after monobromobimane staining of cells, an unstained control sample, to measure ultraviolet-excitable blue autofluorescence, has to be included. The green, yellow, and red fluorescence of NAO is very strong; therefore, it cannot be combined with any other dye that emits in the red or far-red part of the spectrum. The Hoechst 33342 dye generates significant amounts of green and yellow fluorescence, which may interfere with dyes emitting in this spectral region. To prevent Hoechst fluorescence from “bleeding” into the emission channel of another dye, one should excite Hoechst 33342 with a time-delayed (second) laser. Studies of Cell Function
9.15.5 Current Protocols in Cytometry
Supplement 14
1000
“apoptotic”
NAO fluorescence
100
“normal”
10
1
0.1
0.1
1
10 MBB fluorescence
100
1000
Figure 9.15.2 Cytogram of a combined assay for cellular reduced thiol (MBBr fluorescence) and cardiolipin (NAO fluorescence) levels (see Basic Protocol 3). Signals representing “normal” and “apoptotic” cells are labeled.
1000
CMXRosamine/MitoTracker Green FM fluorescence
100
“apoptotic”
“normal” 10
1
0.1 0.1
1
10
100
1000
NAD(P)H
Multiparameter Analysis of Physiological Changes in Apoptosis
Figure 9.15.3 Cytogram of a combined assay for NAD(P)H levels (UV-excited blue autofluorescence) and mitochondrial membrane potential (CMXRosamine/MitoTracker Green FM fluorescence) (see Basic Protocol 1). Signals representing “normal” and “apoptotic” cells are labeled.
9.15.6 Supplement 14
Current Protocols in Cytometry
Anticipated Results
Literature Cited
Figure 9.15.1 displays a typical result from an assay for cell cycle stage–specific apoptosis. Human lymphoblastoid cells were treated with 1 µg/ml mitomycin C 8 hr at 37°C, harvested, and stained according to the assay for cell cycle stage–specific apoptosis described in this unit (see Basic Protocol 2). Apoptotic cells show a lower CMXRosamine and a higher MitoTracker Green FM fluorescence (Fig. 9.15.1, panel A). The two right-hand panels show the cell cycle distributions, based on gated Hoechst 33342 fluorescence, of the “normal” (Fig. 9.15.1, panel B) and “apoptotic” (Fig. 9.15.1, panel C) cells. Note the sub-G1 and the high level of S-phase cells in the apoptotic cell subpopulation. If the G1 phase of the “normal” cells appears as a broad peak, a higher concentration of Hoechst 33342 dye or a longer incubation time should be used. Figure 9.15.2 shows a typical result from a combined assay for cellular reduced thiol (MBBr fluorescence) and cardiolipin (NAO fluorescence) levels. In Figure 9.15.3, a typical result from a combined assay for NAD(P)H levels (UV-excited blue autofluorescence) and m itocho nd rial membrane potential (CMXRosamine/MitoTracker Green FM fluorescence) is shown. Signals representing the “normal” and the “apoptotic” cells are indicated in both figures.
Backway, K.L., McCulloch, E.A., Chow, S., and Hedley, D.W. 1997. Relationships between the mitochondrial permeability transition and oxidative stress during ara-C toxicity. Cancer Res. 57:2446-2451.
Time Considerations For warm-up and laser alignment of the flow cytometer 1 hr should be allowed. The time needed to harvest, stain, and analyze cells varies according to the number of samples to be analyzed.
Cossarizza, A., Kalashnikova, G., Grassilli, E., Chiappelli, F., Salvioli, S., Capri, M., Barbieri, D., Troiano, L., Monti, D., and Franceschi, C. 1994. Mitochondrial modifications during rat thymocyte apoptosis: A study at the single cell level. Exp. Cell Res. 214:323-330. Gallet, P.F., Maftah, A., Petit, J.M., Denis-Gay, M., and Julien, R. 1995. Direct cardiolipin assay in yeast using the red fluorescence emission of 10-N-nonyl acridine orange. Eur. J. Biochem. 228:113-119. Green, D. and Kroemer, G. 1998. The central executioners of apoptosis: Caspases or mitochondria? Trends Cell. Biol. 8:267-271. Green, D.R. and Reed, J.C. 1998. Mitochondria and apoptosis. Science 281:1309-1312. Macho, A., Decaudin, D., Castedo, M., Hirsch, T., Susin, S.A., Zamzami, N., and Kroemer, G. 1996. Chloromethyl-X-Rosamine is an aldehyde-fixable potential-sensitive fluorochrome for the detection of early apoptosis. Cytometry 25:333-340. Poot, M. and Pierce, R.H. 1999. Detection of changes in mitochondrial function during apoptosis by simultaneous staining with multiple fluorescent dyes and correlated multiparameter flow cytometry. Cytometry 35:311-317. Poot, M., Verkerk, A., Koster, J.F., and Jongkind, J.F. 1986. De novo synthesis of glutathione in human fibroblasts during in vitro aging and in some metabolic diseases as measured by a flow cytometric method. Biochim. Biophys. Acta 883:580-584. Thorell, B. 1983. Flow-cytometric monitoring of intracellular flavins simultaneously with NAD(P)H levels. Cytometry 4:61-65.
Contributed by Martin Poot University of Washington Seattle, Washington
Studies of Cell Function
9.15.7 Current Protocols in Cytometry
Supplement 14
Signal Transduction During Natural Killer Cell Activation ated signaling pathways that modulate NK functions.
Natural killer (NK) cells are a subpopulation of lymphocytes bearing a CD16+, CD3−, sIg− phenotype that can mediate cytotoxicity of certain tumor cells, virus-infected cells, and normal cells. In addition to their cytotoxic potential, NK cells secrete a variety of cytokines and chemokines that can modulate the function, growth, and differentiation of other immune cells. These different responses are initiated by the interaction of specific NK surface receptors with defined soluble or cell-associated ligands. There are several different types of receptors on the NK cell surface, including “triggering” receptors (e.g., FcγRIIIA, NKR-P1, and 2B4), adhesion molecules, cytokine receptors, and MHC-recognizing killer-cell inhibitory receptors (Leibson, 1997; Long and Wagtmann, 1997; Brumbaugh et al., 1998; Lanier, 1998; Yokoyama, 1998). The functional response of an NK cell is the result of the integration of signals transduced by these different types of receptors. Some of these signaling pathways are similar to those in other lymphoid cells, but there are also unique features employed by NK cells. This overview will focus on receptor-initi-
ζζ
Grb2 Sos Shc
Lck
Fcγ Receptor-Initiated Signaling FcγRIIIA receptor The FcγRIIIA is a multimeric receptor complex expressed on NK cells (Fig. 9.16.1). The low-affinity, ligand-binding α chain (CD16) is
α
γ γ
P P P P
P P P
?
Alternative modes of killing by NK cells involve antibody-dependent cellular cytotoxicity (ADCC) and natural cytotoxicity. These two forms of cellular cytotoxicity are initiated by different receptors and possess both common and distinct signal transduction pathways (Leibson, 1997; Long and Wagtmann, 1997; Brumbaugh et al., 1998; Lanier, 1998; Yokoyama, 1998). ADCC is triggered by ligation of the low-affinity Fc receptor FcγRIIIA (CD16) with the Fc portion of antibodies bound to cell-associated antigens. In contrast, the pathways that regulate natural cytotoxicity are less well defined and appear to involve a variety of different receptors.
LAT P
PLC-γ
SLP76
Syk
LAT
ACTIVATION SIGNALS
ZAP-70
Ras
UNIT 9.16
ERK
Rho-family of G proteins
DAG
Ca+2
PKC
p85 p110 PI 3K
Vav
IP3
calcineurin P
NFAT
NFAT
granule exocytosis, ADCC, transcriptional regulation
Figure 9.16.1 Signaling pathways activated after FcγRIII stimulation in NK cells. Ligation of FcγRIII induces the phosphorylation and activation of proximal protein tyrosine kinases. Once activated, these kinases, together with adaptor molecules, promote the assembly of signaling complexes that regulate specific NK cell functions. Contributed by Claudia C.S. Chini and Paul J. Leibson Current Protocols in Cytometry (2000) 9.16.1-9.16.13 Copyright © 2000 by John Wiley & Sons, Inc.
Studies of Cell Function
9.16.1 Supplement 14
an integral type I single membrane–spanning glycoprotein that belongs to the Ig superfamily (Perussia, 1998). In the transmembrane domain there is a conserved stretch of 8 amino acids that is necessary for the association of the α chain with two other subunits that are required for surface expression of the receptor and for signal transduction. These two associated chains are disulfide-linked homo- or heterodimers of TCRζ and FcεRIγ (Anderson et al., 1989; Lanier et al., 1989; Orloff et al., 1990; Kurosaki et al., 1991). Although these associated chains do not possess intrinsic enzymatic activity, they do contain immunoreceptor tyrosine-based activation motifs (ITAMs) resembling those found in other multisubunit immune-recognition receptors like the T cell receptor (TCR), B cell receptor (BCR), and other Fc receptor complexes (FcR) (Cambier, 1995; Qian and Weiss, 1997). Phosphorylation of the conserved YXXL sequence in the ITAM provides docking sites for protein tyrosine kinases (PTKs) and other signaling molecules containing Src homology 2 (SH2) domains (Songyang et al., 1993). Binding of the SH2 domain of the PTK to ITAMs leads to phosphorylation of multiple substrates that are essential for the downstream signaling and the cytotoxic function of NK cells (Einspahr et al., 1991; O’Shea et al., 1991, 1992; Vivier et al., 1991). Blocking tyrosine phosphorylation with tyrosine kinase inhibitors prevents the generation of ADCC (Einspahr et al., 1991; O’Shea et al., 1992).
Signal Transduction During Natural Killer Cell Activation
Protein tyrosine kinases Two distinct families of PTK appear to be important for NK signaling: the Src and Syk families. NK cells express many Src family members, including Lck, Lyn, Fyn, and Yes (Eiseman and Bolen, 1990). Src family members contain a myristylation sequence at the amino-terminus that allows their interaction with the inner leaflet of the plasma membrane. They also contain SH2 and SH3 domains that bind phosphotyrosines and proline-rich sequences, respectively. These PTKs are activated after aggregation of the multisubunit receptors and subsequently tyrosine phosphorylate ITAMs in the receptor complex (Tamir and Cambier, 1998). ITAM phosphorylation serves to recruit numerous SH2-containing signaling proteins including the two Syk-family PTK members, ZAP-70 and Syk, which are both expressed by NK cells (Vivier et al., 1993; Stahls et al., 1994). Unlike Src-family members, the Syk-family PTKs do not contain myristylation sites or SH3 domains, but do have
two tandem amino-terminal SH2 domains which interact with the phosphorylated ITAMs (Chu et al., 1998). This interaction leads to the phosphorylation and activation of numerous downstream signaling molecules. Several studies have indicated that Lck, ZAP, and Syk undergo tyrosine phosphorylation and association with the FcγRIIIA-receptor complex following receptor cross-linking (Cone et al., 1993; Pignata et al., 1993; Salcedo et al., 1993; Vivier et al., 1993; Stahls et al., 1994). In addition, it was demonstrated that Lck overexpression in NK cells enhances the tyrosine phosphorylation and catalytic activity of ZAP-70 and Syk (Ting et al., 1995). This suggests a model in which activation of a Src family member (Lck) regulates downstream Syk-family PTKs, as observed in TCR, BCR, and FcεRI signaling (Weiss and Littman, 1994). However, NK cells from Lck-deficient mice can generate ADCC (Wen et al., 1995), indicating that Lck is not necessary for NK cell-mediated killing. One explanation may be that Syk family members can be activated and mediate killing in a Src-family PTK-independent manner (Zoller et al., 1997; Williams et al., 1998). Recent data support the hypothesis that Syk, but not ZAP-70, is preferentially involved in the signaling pathways that modulate cytotoxicity. First, overexpression of wild-type Syk, but not ZAP-70, enhances NK cell–mediated killing, and only a kinase-inactive Syk could decrease cell-mediated killing (Brumbaugh et al., 1997). Second, NK cells from humans and mice lacking ZAP-70 can mediate ADCC (Chan et al., 1994; Elder et al., 1994; Negishi et al., 1995). Finally, it is known that Syk, but not ZAP-70, can induce tyrosine phosphorylation of ITAM-containing receptor subunits in the absence of a Src-family member (Zoller et al., 1997; Williams et al., 1998). Taken together, these data suggest that Syk plays an important role in human NK-cell activation. PLC-γ One of the most critical molecules activated downstream of Syk-family kinases is phospholipase C-γ (PLC-γ). PLC-γ is an enzyme that following its activation by tyrosine phosphorylation cleaves membrane phosphoinositides into inositol 1,4,5-trisphosphate (IP3) and sn-1,2-diacylglycerol (DAG). These second messengers mediate release of intracellular calcium and activation of protein kinase C (PKC), respectively. Both PLC-γ1 and PLC-γ2 are activated after FcγRIIIA cross-linking (Azzoni et
9.16.2 Supplement 14
Current Protocols in Cytometry
al., 1992; Ting et al., 1992), and this activation is dependent on tyrosine phosphorylation (Ting et al., 1991). Indeed, elevations in intracellular calcium are required for granule release during NK killing (Windebank et al., 1988; Cassarella et al., 1989). Overexpression of Lck increases PLC-γ phosphorylation (Ting et al., 1995), and chimeras of CD16:ZAP and CD16:Syk cause calcium mobilization (Kolanus et al., 1993), suggesting potential roles for these kinases in the signaling pathway that leads to PLC-γ activation. The role of PKC in ADCC is not so clear. Addition of pharmacological stimulators of PKC, like phorbol esters, together with calcium ionophores induces granule release (Atkinson et al., 1990). However, pharmacological inhibition of PKC has no effect on ADCC (Bonnema et al., 1994a). These data suggest that although there is a PKC-dependent pathway that can mediate granule release in NK cells, FcγRIIIA-initiated secretion is regulated by a PKC-independent pathway (Bonnema et al., 1994a). PI-3 kinase Another downstream effector of FcγRIIIA cross-linking is phosphoinositol-3 (PI-3) kinase. PI-3 kinase is a heterodimer that phosphorylates phosphoinositides at the D-3 position of the inositol ring. These phosphorylated inositides play important roles in multiple signal transduction pathways (Scharenberg and Kinet, 1998). Ligation of the FcγRIII on NK cells rapidly activates PI-3 kinase, increasing its association with phosphotyrosine-containing proteins (Bonnema et al., 1994a; Kanakaraj et al., 1994). Inhibition of PI-3 kinase activity by the inhibitor wortmannin blocks granule release and ADCC in NK cells (Bonnema et al., 1994a), suggesting that FcγRIII-dependent killing occurs through a PI-3 kinase–dependent pathway. Adapter proteins Although most investigations have focused on the role of protein kinases in NK-cell cytotoxicity, there are several studies that have investigated the function of adaptor molecules in regulation of NK-cell activation. Adaptor molecules are proteins without known enzymatic activity that play a critical role in signal transduction by mediating protein-protein interactions (Peterson et al., 1998). FcγRIII crosslinking induces the tyrosine phosphorylation of several adaptor molecules that are possible regulators of NK cell–mediated cytotoxicity. One of these proteins, p52Shc, binds the adap-
tor protein Grb2 and may be involved in activation of the downstream Ras pathway (Ricciarda et al., 1996). A direct role for Ras activation in NK cytotoxicity has not been demonstr ated, but the MAPK/extracellular signal-regulatory kinase 2 (ERK2) pathway, which is dependent on Ras activation in most cell types, appears to play an important role in the development of cellular cytotoxicity. In many cell types, the MAPK pathway controls transcriptional events that regulate functions such as proliferation and differentiation. Indeed, recent studies indicate that the ERK pathway is activated after FcγRIII cross-linking (Trotta et al., 1996; Milella et al., 1997), and that this pathway is important for transcription control of TNF-α and c-fos following FcγR stimulation (Trotta et al., 1996). However, distinct from its ability to control transcription factor activation, recent work suggests that ERK may play a role in the development of ADCC since an inhibitor of the MAPK pathway, PD098059, and overexpression of a dominant-negative MAPK blocked ADCC (Trotta et al., 1998). Roles for the adaptor proteins LAT and SLP76 in ADCC have also been proposed. LAT (linker for the activation of T cells) is a 36-kDa protein (p36) that is phosphorylated following TCR cross-linking in T cells (Zhang et al., 1998) and FcγRIII cross-linking in NK cells (Jevremovic et al., 1999). In T cells, LAT appears to interact with Grb2, SLP-76, Vav, and the p85 regulatory subunit of PI-3K, and is required for PLC-γ phosphorylation and activation following TCR cross-linking (Zhang et al, 1998). In NK cells, LAT was shown to be required for FcγRIII-dependent phosphorylation of PLC-γ, and overexpression of LAT leads to an increase in ADCC (Jevremovic et al., 1999). SLP-76, an adaptor originally identified as a Grb2 binding protein in T cells (Jackman et al., 1995), is also phosphorylated after FcγRIII cross-linking (Binstadt et al., 1998). T cell lines deficient in SLP-76 have impaired activation of both PLC-γ and Ras pathways (Yablonski et al., 1998), and SLP-76 overexpression in cytotoxic T lymphocytes enhances cell-mediated killing (Binstadt et al., 1998). However, the functional significance of SLP-76 in NK signaling and cytotoxicity remains to be determined. Vav-Rac Another potential regulator of FcγRIII-mediated killing is Vav. Vav is a guanine nucleotide exchange factor for Rac, a member of the Rho-
Studies of Cell Function
9.16.3 Current Protocols in Cytometry
Supplement 14
family of GTP-binding proteins (G proteins) (Crespo et al., 1997) that is involved in regulation of the actin cytoskeleton. Indeed, Vav has been reported to be phosphorylated upon FcγRIII ligation (Xu and Chong, 1996; Billadeau et al., 1998), and recent data show that Vav and Rac regulate granule exocytosis and ADCC (Billadeau et al., 1998), demonstrating a role for Rho-family proteins in NK cytotoxicity. In addition to low-molecular-weight GTPbinding proteins, heterotrimeric G-proteins may also play a role in ADCC (Whalen et al., 1992; Maghazachi et al., 1996). NFAT Although it is known that FcγRIII crosslinking activates many molecules and signaling pathways, less has been described about the mechanisms by which these pathways can result in transcriptional control of a variety of genes. One transcription factor, which was shown to be activated in NK cells after FcγRIII cross-linking, is nuclear factor of activated T cells (NFAT; Aramburu et al., 1995). Cytoplasmic NFAT undergoes dephosphorylation by the phosphatase calcineurin. This dephosphorylation leads to its activation and nuclear migration. Cyclosporin A (CsA) inhibits calcineurin activity, thus inhibiting NFAT activation (Crabtree, 1999). In T cells, NFAT regulates the expression of a number of cytokine genes, including IL-2. In NK cells, NFAT activity can regulate the transcription of GM-CSF and TNF-α (Aramburu et al., 1995). Interestingly, although the transcriptional control of the IFNγ promoter is CsA sensitive there are no NFAT binding sites in the IFN-γ promoter, suggesting that another CsA-sensitive transcription factor may be involved in its regulation (Penix et al., 1993).
Signaling During Natural Cytotoxicity
Signal Transduction During Natural Killer Cell Activation
NK cells can kill a variety of cells infected with pathogens without prior sensitization and in the absence of antibody. This process is called natural cytotoxicity and the NK receptors and signal transduction pathways that mediate this form of killing are less well characterized than the pathways that modulate ADCC. Attempts to identify a single triggering receptor have been unsuccessful, and the emerging picture suggests that a variety of activating receptors and coreceptors on the NK cell surface modulate natural cytotoxicity (Leibson, 1997; Brumbaugh et al., 1998). Many of the intracellular signaling molecules that were shown to be involved in ADCC are also important for
natural cytotoxicity. Natural cytotoxicity, like ADCC, utilizes rapid PTK activation (Einspahr et al., 1991; O’Shea et al., 1992) and PLC-γ release of membrane phosphoinositides (Seaman et al., 1987; Chow et al., 1988; Steele and Brahmi, 1988), and increases in intracellular calcium (Chow et al., 1988). Molecules such as Syk (Brumbaugh et al., 1997), Vav (Billadeau et al., 1998), LAT (Jevremovic et al., 1999), and ERK (Trotta et al., 1998; Wei et al., 1998) also appear to play critical roles in the development of natural cytotoxicity. On the other hand, in contrast to ADCC, natural killing of targets can occur through a PKC-dependent, but PI-3K-independent, mechanism (Bonnema et al., 1994a). Adhesion molecules such as β integrins can also modulate cytotoxicity, but it is less clear if they are involved in triggering killing, providing costimulation, or primarily facilitating adhesion with target cells. The cytoplasmic domain of these proteins interacts with cytoskeletal components, and aggregation of these receptors results in activation of PTKs. These activated PTKs tyrosine phosphorylate multiple substrates, which leads to the assembly of cytoskeletal proteins in focal adhesions (Newton et al., 1997). NK cells express many adhesion molecules that modulate NK function (Helander and Timonen, 1998). α4β1 and α5β1 integrins are expressed by NK cells and mediate their binding to fibronectin (Gismondi et al., 1995). Cross-linking of β1 integrins on NK cells induces increases in intracellular calcium (Palmieri et al., 1995), tyrosine phosphorylation of proteins (Gismondi et al., 1995), activation of the Ras/MAPK pathway (Mainiero et al., 1998), and enhancement in NK-cell cytotoxic activity (Palmieri et al., 1995), suggesting that β1 integrins may function as activating receptors. The β2 integrin LFA-1 interacts with the intercellular adhesion molecule ICAM-2 on the target cells, and this interaction can also play an important role in the induction of target lysis. Indeed, LFA-1-ICAM-2 interactions can induce increases in intracellular calcium, granule release, and cytokine secretion from NK cells (Melero et al., 1993; Sugie et al., 1995). Target cell lysis can be blocked by antibodies directed against LFA-1 as well as against ICAM-2 (Helander et al., 1996), indicating an important role for these molecules in NK cytotoxicity. Several other receptors have been described as potential triggers of NK cell–mediated killing, including NKR-P1, NKp44, NKp46, Lag3, 2B4, DNAM-1, and certain MHC-recognizing
9.16.4 Supplement 14
Current Protocols in Cytometry
receptors. NKR-P1 is a member of the C-type lectin family and is expressed as disulfidelinked homodimers. The NKR-P1 molecule associates with the FcRγ subunit (Arase et al., 1997), and signal transduction by this receptor appears to have similarities with signaling initiated by other multisubunit immune recognition receptors, including PTK and PLC-γ activation (Campbell and Giorda, 1997; Cerny et al., 1997). NKR-P1 cross-linking leads to PI turnover, calcium mobilization, NK cytotoxicity, granule exocytosis, and IFN-γ production (Giorda et al., 1990; Ryan et al., 1991, 1995; Lanier et al., 1994; Arase et al., 1996; Cifone et al., 1997). Antibodies for this receptor are able to initiate NK cytotoxicity in a redirected killing assay. Furthermore, NKR-P1-deficient NK cells are defective in killing certain targets (Ryan et al., 1995), indicating that this receptor has a functional role in regulating the development of certain forms of cellular cytotoxicity. NKp46 and NKp44 are two other recently identified receptors that can initiate NK cell cytotoxicity (Pessino et al., 1998; Vitale et al., 1998). While NKp46 couples to the intracellular transduction machinery through association with CD3ζ , NKp44 appears to associate with DAP12 (Vitale et al., 1998), a 16-kDa membrane receptor that contains a single ITAM. LAG3 is a member of the Ig superfamily and was originally identified as a CD4-related molecule. Inactivation of the LAG3 gene in mice does not result in any abnormal T-cell phenotype, but NK cells from these mice demonstrate reduced cytotoxicity against some types of tumor cell lines (Miyazaki et al., 1996) suggesting a role for this receptor in certain forms of natural cytotoxicity. 2B4 belongs to the Ig superfamily and is found on the surface of all NK and T cells (Schatzle et al., 1999). It can mediate cytotoxicity and recently its ligand was identified as CD48 (Latchman et al., 1998). 2B4 was originally identified as an activating molecule, because its cross-linking results in increased cytotoxicity, γ-interferon secretion, and granule exocytosis (Garni-Wagner et al., 1993). Later, it was shown that 2B4 receptors have stimulatory and inhibitory forms (Schatzle et al., 1999). Cross-linking of DNAM-1, a type I integral membrane protein, also leads to its tyrosine phosphorylation and development of cellular cytotoxicity in a reverse ADCC (Shibuya et al., 1996). Its function appears to be PKC-dependent (Shibuya et al., 1998). For most of these receptors, identification of specific ligands and full characterization
of the regulatory signal transduction pathways still remain to be determined. Another group of activating receptors present in NK cells are the MHC-recognizing activating receptors. Previous studies have suggested an inverse correlation between the expression levels of surface MHC class I molecules on target cells and their susceptibility to killing by NK cells. While exploring the mechanism of NK recognition of “missing self,” a number of MHC-recognizing receptors were described that belong to the C-type lectin family (Ly-49 and CD94/NKG2) and the immunoglobulin superfamily (KARs and KIRs; Leibson, 1995; Binstadt et al., 1997; Burshtyn and Long, 1997; Lanier, 1997; Vely and Vivier, 1997; Bensussan, 1998). Both families contain receptors that can either activate or inhibit killing. The human killer cell–activating receptors (KARs) have extracellular domains homologous to the killer cell–inhibitory receptors (KIRs), which are monomeric type I glycoproteins with 2 or 3 Ig-like domains in their extracellular region. In contrast to KIRs, their short cytoplasmic tails do not contain the immunoreceptor tyrosine-based inhibitory motifs (ITIMs) that initiate KIR inhibitory function (Burshtyn and Long, 1997; Vely and Vivier, 1997). They do, however, contain a single lysine residue in their transmembrane region similar to those present in other activating receptors (e.g., TCR αβ, FcγRIIIα, and FcεRIα). This residue is required for assembly of the ligand-binding subunits with ITAM-containing signaling subunits (Biassoni et al., 1996; Bottino et al., 1996). Recently, it was shown that KARs noncovalently associate with DAP12 (Lanier et al., 1998a). Upon the cross-linking of KAR, tyrosine residues in the ITAM of DAP12 become tyrosine phosphorylated and recruit ZAP-70 or Syk molecules, resulting in cellular activation (Lanier et al., 1998a). The CD94/NKG2 complex is a heterodimer composed of two disulfide-bonded glycoproteins of the C-type lectin family (Lopez-Botet et al., 1998). Initially, it was demonstrated that cross-linking of CD94 results in positive or negative signals depending on the NK clone used (Perez-Villar et al., 1995; Brumbaugh et al., 1996). Later, it was shown that CD94 associates with members of the NKG2 family (Lazetic et al., 1996), which contains stimulatory (NKG2-C and -E) and inhibitory (NKG2A and -B) forms (Houchins et al., 1997). CD94 is an invariant subunit and is required for surface expression of NKG2. NKG2-C and -E lack an ITIM and cannot mediate inhibitory signals.
Studies of Cell Function
9.16.5 Current Protocols in Cytometry
Supplement 14
These activating forms of NKG2 also contain a charged amino acid in their transmembrane domain, suggesting a potential interaction with other signaling subunits (Lopez-Botet et al., 1998). In fact, like KARs, activating forms of CD94/NKG2 associate with DAP12, and ligation of this complex results in the tyrosine phosphorylation of DAP12 (Lanier et al., 1998b). After CD94/NKG2-C stimulation there is activation of several PTKs, including Lck, ZAP, and Syk, followed by tyrosine phosphorylation of PLC-γ, and an increase in intracellular calcium (Brumbaugh et al., 1996; Lazetic et al., 1996; Lanier et al., 1998b). Ly49, a group of type II lectin-like MHC-recognizing receptors present in mice, also have stimulatory (Ly49D and H) and inhibitory forms (Ly49A, B, C, E, and F; Mason et al., 1996; Burshtyn and Long, 1997). Like the two activating receptors described above, Ly49D associates with DAP12 to transmit positive signals to NK cells (McVicar et al., 1998; Smith et al., 1998). Ligation of Ly49D results in phosphorylation of several substrates, including PLC-γ, Cbl, ERK, and Syk (McVicar et al., 1998). In addition, signaling through Ly49D results in intracellular calcium mobilization that can be blocked by expression of dominant-negative Syk (McVicar et al., 1998). The emerging picture suggests that DAP12 is a common subunit employed by several types of MHC-recognizing activating receptors in order to transduce positive signals in NK cells.
Cytokine-Initiated Enhancement of NK Cell Activation
Signal Transduction During Natural Killer Cell Activation
NK cells express a variety of cytokine receptors that modulate NK cell effector functions such as proliferation, survival, and cytokine production (Perussia, 1991). Cytokines such as IL-2, IL-12, IL-15, and IFN-α enhance NK cytotoxicity (Bonnema et al., 1994b; Carson et al., 1994; Welsh, 1984), but the signaling pathways involved in their effects in NK cells have not been fully elucidated. The receptors for the specific cytokines indicated above lack intrinsic tyrosine-kinase activity, but ligand binding, and subsequent clustering of these receptors, leads to the phosphorylation and activation of members of the Janus (JAK) family of tyrosine kinases. Activation of JAK kinases is responsible for induction of several signaling cascades including the phosphorylation and activation of the family of “signal transducers and activators of transcription” (STATs). Once activated, STATs translocate to the nucleus, bind to specific DNA sequences, and thereby
regulate gene transcription (Ihle et al., 1998). In NK cells, it was shown that IL-2 activates Jak-1 and Jak-3, whereas IL-12 activates Jak-2 and Tyk-2 (Bacon et al., 1995a). IL-2, IL-7, and IL-15 are able to activate several members of the STAT family (Yu et al., 1998; Wang et al., 1999), but in contrast, IL-12 activates mainly STAT4 (Bacon et al., 1995b). Interestingly, although mice with disruption of the STAT4 gene are viable, the enhancement of NK cytolytic function normally observed by IL-12 treatment of NK cells is defective (Bacon et al., 1995b). Thus, it appears that the differential responses of NK cells to specific cytokines is regulated by the activation of distinct subsets of the JAK and STAT families.
INHIBITION OF NK CELL ACTIVATION The capacity of NK cells to mediate killing is determined by the balance between positive and negative regulatory events. Normal MHC class I–bearing cells are often relatively resistant to NK cell–mediated killing because of their ability to ligate NK-cell inhibitory receptors. These receptors recognize MHC class I complexes on target cells and this ligation inhibits NK cell activation, cytotoxicity, and cytokine secretion (reviewed in Leibson, 1995; Binstadt et al., 1997; Burshtyn and Long, 1997; Lanier, 1997; Vely and Vivier, 1997; Bensussan, 1998). Malignant or virus-infected cells with fewer MHC class I complexes on their surface may fail to engage inhibitory receptors and thus become more susceptible to killing by NK cells. As already mentioned, NK cell–inhibitory MHC-recognizing receptors are divided in two families of receptors that belong to either the C-type lectin family (CD94/NKG2 and Ly49) or immunoglobulin superfamily (human KIRs p58 and p70). These inhibitory receptors are polymorphic, and receptors in each family differ in their specificity for subgroups of MHC class I molecules. These receptors are clonally distributed on populations of NK cells, and each NK cell appears to have at least one type of MHC-recognizing inhibitory receptor (reviewed in Leibson, 1995; Binstadt et al., 1997; Burshtyn and Long, 1997; Lanier, 1997; Vely and Vivier, 1997; Bensussan, 1998). Although the different types of inhibitory receptors differ in their extracellular region, they have similar cytoplasmic domains, suggesting a common mechanism of signal transduction (Fig. 9.16.2). One important similarity in their cytoplasmic domain is the presence of one or
9.16.6 Supplement 14
Current Protocols in Cytometry
KIR
Src-family PTK
P P
SHP-1
inhibits “triggering” receptor-induced tyrosine kinase activation by dephosphorylating molecules like SLP76, LAT, and ? others
blocks NK cell activation Figure 9.16.2 Inhibitory signaling from MHC-recognizing receptors. Ligation of MHC-recognizing receptors results in the phosphorylation of tyrosine residues in the cytoplasmic ITIM by Src-family PTK. The tyrosine phosphatase SHP-1 is then recruited to the phosphorylated ITIM. SHP-1 dephosphorylates molecules associated with activating receptors, thus promoting inhibition of NK cell activation.
two copies of a specific ITIM, containing the sequence [I,V]xYxx[L,V]. The amino acids at the −2 position upstream of the tyrosine in the ITIM differentiate this sequence from the ITAMs of the multisubunit immune-recognition receptors. Cross-linking of the MHC-recognizing inhibitory receptors induces phosphorylation of the tyrosine residue in the ITIM, and this phosphorylation is essential for transduction of the inhibitory signal (Burshtyn et al., 1996). It has been proposed that Src-family tyrosine kinases, such as Lck, play a central role in the phosphorylation of ITIMs (Binstadt et al., 1996; Burshtyn et al., 1996), since ITIM phosphorylation does not occur following KIR cross-linking in the Lck-deficient cell line JCAM1. Furthermore, overexpression of wildtype, but not catalytically inactive, Lck in these cells restores KIR phosphorylation (Binstadt et al., 1996). Taken together, these data demonstrate a proximal requirement for Src-family PTKs in mediating KIR inhibitory signals. Phosphorylation of the MHC-recognizing inhibitory receptors results in the recruitment of the tyrosine phosphatases SHP-1 and SHP-2 (Binstadt et al., 1996; Burshtyn et al., 1996; Campbell et al., 1996; Fry et al., 1996; Olcese et al., 1996; Nakamura et al., 1997). SHP-1 is involved in down modulating signaling from several families of leukocyte receptors, including antigen, cytokine, and growth-factor receptors. It contains two amino-terminal SH2 domains and a carboxy-terminal catalytic domain
(Yi et al., 1992; Neel, 1997). Cross-linking of different types of MHC-recognizing inhibitory receptors results in association of the carboxyterminal domain of SHP-1 with the tyrosinedphosphorylated ITIM of the receptor (reviewed in Binstadt et al., 1997; Burshtyn and Long, 1997). The importance of SHP-1 in KIR signaling is highlighted by the observation that overexpression of a catalytically inactive form of SHP-1 (acting as a dominant-negative) can reverse the inhibitory effect of KIR ligation in both ADCC and natural cytotoxicity (Binstadt et al., 1996; Burshtyn et al., 1996). Interestingly, coligation of KIRs or inhibitory forms of CD94/NKG2 along with FcγRIII blocks early events in the activation pathways including the phosphorylation of the ζ -chain ITAM, PLC-γ, and ZAP-70, and also the mobilization of intracellular calcium (Kaufman et al., 1995; Binstadt et al., 1996; Brumbaugh et al., 1996). One possibility is that SHP-1 dephosphorylates and blocks one of the proximal tyrosine kinases involved in NK-cell activation. However, a direct effect of SHP-1 on these molecules has not been demonstrated. On the other hand, LAT and SLP-76 have been implicated as potential SHP-1 targets. It has been shown that following KIR recognition of class I ligands, PLC-γ is phosphorylated, but does not associate with tyrosine-phosphorylated LAT (Valiante et al., 1996). In fact, SHP-1 has been shown to dephosphorylate LAT, suggesting a potential model where dephosphorylation of LAT by
Studies of Cell Function
9.16.7 Current Protocols in Cytometry
Supplement 14
SHP-1 could prevent LAT from interacting with PLC-γ, blocking the production of PLCγ-generated second messengers (Valiante et al., 1996). Recent data show that SLP-76 is another direct substrate of SHP-1 (Binstadt et al., 1998). SLP-76 dephosphorylation by SHP-1 may be involved in negatively regulating cellular cytotoxicity, since it was observed that SLP-76 has a positive effect in cellular cytotoxicity (Binstadt et al., 1998). It will be important to determine if LAT and SLP-76 are the selective substrates of SHP-1, or if there are other signaling molecules that upon dephosphorylation by SHP-1 play a role in the inhibition of NK function.
SUMMARY During the past few years our understanding of transmembrane signaling during NK-cell activation and the generation of cytotoxic response has greatly expanded. The discovery of novel triggering and inhibitory receptors and their characterization has shown us the complexity of these processes. In order to understand how NK cells coordinate these different activating and inhibitory signals, further characterization of these receptors and their associated signal transduction pathways will be needed. A more complete picture of the roles of different signaling pathways in NK-cell activation will also be important for the design of therapeutic approaches to selectively enhance or suppress NK-cell activation.
LITERATURE CITED Anderson, P., Caligiuri, M., Ritz, J., and Schlossman, S.F. 1989. CD3-negative natural killer cells express ζ TCR as part of a novel molecular complex. Nature 341:159-162. Aramburu, J., Azzoni, L., Rao, A., and Perussia, B. 1995. Activation and expression of the nuclear factors of activated T cells, NFATp and NFATc, in human natural killer cells: Regulation upon CD16 ligand binding. J. Exp. Med. 182:801-810. Arase, H., Arase, N., and Saito, T. 1996. Interferon γ production by natural killer (NK) cells and NK1.1 T+ cells upon NKR-P1 cross-linking. J. Exp. Med. 183:2391-2396. Arase, N., Arase, H., Park, S.Y., Ohno, H., Ra, C., and Saito, T. 1997. Association with FcRγ is essential for activation signal through NKR-P1 (CD161) in natural killer (NK) cells and NK1.1+ T cells. J. Exp. Med. 186:1957-1963.
Signal Transduction During Natural Killer Cell Activation
Atkinson, E.A., Gerrard, J.M., Hildes, G.E., and Greenberg, A.H. 1990. Studies of the mechanism of natural killer (NK) degranulation and cytotoxicity. J. Leukoc. Biol. 47:39-48.
Azzoni, L., Kamoun, M., Salcedo, T.W., Kanakaraj, P., and Perussia, B. 1992. Stimulation of Fc gamma RIIIA results in phospholipase Cgamma 1 tyrosine phosphorylation and p56lck activation. J. Exp. Med. 176:1745-1750. Bacon, C.M., McVicar, D.W., Ortaldo, J.R., Rees, R.C., O’Shea, J.J., and Johnston, J.A. 1995a. Interleukin 12 (IL-12) induces tyrosine phosphorylation of JAK2 and TYK2: Differential use of Janus family tyrosine kinases by IL-2 and IL-12. J. Exp. Med. 181:399-404. Bacon, C.M., Petricoin, E.F. III, Ortaldo, J.R., Rees, R.C., Larner, A.C., Johnston, J.A, and O’Shea, J.J. 1995b. Interleukin 12 induces tyrosine phosphorylation and activation of STAT4 in human lymphocytes. Proc. Natl. Acad. Sci. U.S.A. 92:7307-7311. Bensussan, A. 1998. The human natural killer lymphocyte receptors for MHC class I molecules. Eur. Cytok. Netw. 9:577-583. Biassoni, R., Cantoni, C., Falco, M., Verdiani, S., Bottino, C., Vitale, M., Conte, R., Poggi, A., Moretta, A., and Moretta, L. 1996. The human leukocyte antigen (HLA)-C-specific “activatory” or “inhibitory” natural killer receptors display highly homologous extracellular domains, but differ in their transmembrane and intracytoplasmic portions. J. Exp. Med. 183:645-650. Billadeau, D.D., Brumbaugh, K.M., Dick, C.J., Schoon, R.A., Bustelo, X.R., and Leibson, P.J. 1998. The Vav-Rac1 pathway in cytotoxic lymphocytes regulates the generation of cell-mediated killing. J. Exp. Med. 188:549-559. Binstadt, B.A., Brumbaugh, K.M., Dick, C.J., Scharenberg, A.M., Williams, B.L., Colonna, M., Lanier, L.L., Kinet, J.-P., Abraham, R.T., and Leibson, P.J. 1996. Sequential involvement of Lck and SHP-1 with MHC-recognizing receptors on NK cells inhibits FcR-initiated tyrosine kinase activation. Immunity 5:629-638. Binstadt, B.A., Brumbaugh, K.M., and Leibson, P.J. 1997. Signal transduction by human NK cell MHC-recognizing receptors. Immunol. Rev. 155:197-203. Binstadt, B.A., Billadeau, D.D., Jevremovic, D., Williams, B.L., Fang, N., Yi, T., Koretzky, G.A., Abraham, R.T., and Leibson, P.J. 1998. SLP-76 is a direct substrate of SHP-1 recruited to killer cell inhibitory receptors. J. Biol. Chem. 273:27518-27523. Bonnema, J.D., Karnitz, L.M., Schoon, R.A., Abraham, R.T., and Leibson, P.J. 1994a. Fc receptor stimulation of phosphatidylinositol 3-kinase in natural killer cells is associated with protein kinase C-independent granule release and cell-mediated cytotoxicity. J. Exp. Med. 180:1427-1435. Bonnema, J.D., Rivlin, K.A., Ting, A.T., Schoon, R.A., Abraham, R.T., and Leibson, P.J. 1994b. Cytokine-enhanced NK cell-mediated cytotoxicity. Positive modulatory effects of IL-2 and IL-12 on stimulus-dependent granule exocytosis. J. Immunol. 152:2098-2104.
9.16.8 Supplement 14
Current Protocols in Cytometry
Bottino, C., Sivori, S., Vitale, M., Cantoni, C., Falco, M., Pende, D., Morelli, L., Augugliaro, R., Semenzato, G., Biassoni, R., Moretta, L., and Moretta, L. 1996. A novel surface molecule homologous to the p58/p50 family of receptors is selectively expressed on a subset of human natural killer cells and induces both triggering of cell functions and proliferation. Eur. J. Immunol. 26:1816-1824. Brumbaugh, K.M., Perez-Villar, J.J., Dick, C.J., Schoon, R.A., Lopez-Botet, M., and Leibson, P.J. 1996. Clonotypic differences in signaling from CD94 (kp43) on NK cells lead to divergent cellular responses. J. Immunol. 157:2804-2812. Brumbaugh, K.M., Binstadt, B.A., Billadeau, D.D., Schoon, R.A., Dick, C.J., Ten, R.M., and Leibson, P.J. 1997. Functional role for syk tyrosine kinase in natural killer cell-mediated natural cytotoxicity. J. Exp. Med. 186:1-9. Brumbaugh, K.M., Binstadt, B.A., and Leibson, P. 1998. Signal transduction during NK cell activation: Balancing opposing forces. Curr. Top. Microbiol. Immunol. 230:103-122. Burshtyn, D.N. and Long, E. 1997. Regulation through inhibitory receptors: Lessons from natural killer cells. Trends Cell Biol. 7:473-479. Burshtyn, D.N., Scharenberg, A.M., Wagtmann, N., Rajagopalan, S., Berrada, K., Yi, T., Kinet, J.-P., and Long, E.O. 1996. Recruitment of tyrosine phosphatase HCP by the killer cell inhibitory receptor. Immunity 4:77-85. Cambier, J.C. 1995. Antigen and Fc receptor signaling. The awesome power of the immunoreceptor tyrosine-based activation motif (ITAM). J. Immunol. 155:3281-3285. Campbell, K.S. and Giorda, R. 1997. The cytoplasmic domain of rat NKR-P1 receptor interacts with the N-terminal domain of p56lck via cysteine residues. Eur. J. Immunol. 27:72-77. Campbell, K.S., Dessing, M., Lopez-Botet, M., Cella, M., and Colonna, M. 1996. Tyrosine phosphorylation of a human killer inhibitory receptor recruits protein tyrosine phosphatase 1C. J. Exp. Med. 184:93-100. Carson, W.E., Giri, J.G., Lindemann, M.J., Linett, M.L., Ahdieh, M., Paxton, R., Anderson, D., Eisenmann, J., Grabstein, K., and Caligiuri, M.A. 1994. Interleukin (IL) 15 is a novel cytokine that activates human natural killer cells via components of the IL-2 receptor. J. Exp. Med. 180:1395-1403. Cassarella, M.A., Anegon, L., Guturi, M.C., Griskey, P., Trinchieri, G., and Perussia, B. 1989. FcγR (CD16) interaction with ligand induces Ca+2 mobilization and phosphoinositide turnover in human natural killer cells. Role of Ca+2 in FcγR (CD16)-induced transcription and expression of lymphokine genes. J. Exp. Med. 169:549-567. Cerny, J., Fiserova, A., Horvath, O., Bezouska, K., Pospisil, M., and Hoejsi, V. 1997. Association of human NK cell surface receptors NKR-P1 and CD94 with Src-family protein kinases. Immunogenetics 46:231-236.
Chan, A.C., Kadlecek, T.A., Elder, M.E., Filipovich, A.H., Kuo, W.-L., Iwashima, M., Parslow, T.G., and Weiss, A. 1994. ZAP-70 deficiency in an autosomal recessive form of severe combined immunodeficiency. Science 264:1599-1601. Chow, S.C., Ng, J., Nordstedt, C., Fredholm, B.B., and Jondal, M. 1988. Phosphoinositide breakdown and evidence for protein kinase C involvement during human NK killing. Cell. Immunol. 114:96-103. Chu, D.H., Morita, C.T., and Weiss, A. 1998. The Syk family of protein tyrosine kinases in T-cell activation and development. Immunol. Rev. 165:167-180. Cifone, M.G., Roncaioli, P., Cironi, L., Festuccia, C., Meccia, A., D’Alo, S., Botti, D., and Santoni, A. 1997. NKR-P1A stimulation of arachidonategenerating enzymes in rat NK cells is associated with granule release and cytotoxic activity. J. Immunol. 159:309-317. Cone, J.C., Lu, Y., Trevillyan, J.M., Bjorndahl, J.M., and Phillips, C.A. 1993. Association of the p56lck protein tyrosine kinase with the FcγRIIIA/CD16 complex in human natural killer cells. Eur. J. Immunol. 23:2488-2497. Crabtree, G.R. 1999. Generic signals and specific outcomes: signaling through Ca+2, calcineurin, and NF-AT. Cell 96:611-614. Crespo, P., Schuebel, K.E., Ostrom, A.A., Gutkind, J.S., and Bustelo, X.R. 1997. Phosphotyrosinedependent activation of Rac-1 GDP/GTP exchange by the Vav proto-oncogene product. Nature 375:169-172. Einspahr, K.J., Abraham, R.T., Binstadt, B.A., Uehara, Y., and Leibson, P.J. 1991. Tyrosine phosphorylation provides an early and requisite signal for the activation of natural killer cell cytotoxic function. Proc. Natl. Acad. Sci. U.S.A. 88:6279-6283. Eiseman, E. and Bolen, J.B. 1990. Src-related tyrosine protein kinases as signaling components in hematopoietic cells. Cancer Cells 2:303-310. Elder, M.E., Lin, D., Clever, J., Chan, A.C., Hope, T.J., Weiss, A., and Parslow, T.G. 1994. Human severe combined immunodeficiency due to a defect in ZAP-70, a T cell tyrosine kinase. Science 264:1596-1598. Fry, A.M., Lanier, L.L., and Weiss, A. 1996. Phosphotyrosines in the killer cell inhibitory receptor motif of NKB1 are required for negative signaling and for association with protein tyrosine phosphatase 1C. J. Exp. Med. 184:295-300. Garni-Wagner, B.A., Purohit, A., Mathew, P.A., Bennett, M., and Kumar, V. 1993. A novel function-associated molecule related to non-MHCrestricted cytotoxicity mediated by activated natural killer cells and T cells. J. Immunol. 151:60-70. Giorda, R., Rudert, W.A., Vavassori, C., Chambers, W.H., Hiserodt, J.C., and Trucco, M. 1990. NKR-P1, a signal transduction molecule on natural killer cells. Science 249:1298-1300.
Studies of Cell Function
9.16.9 Current Protocols in Cytometry
Supplement 14
Gismondi, A., Milella, M., Palmieri, G., Piccoli, M., Frati, L., and Santoni, A. 1995. Stimulation of protein tyrosine phosphorylation by interaction of NK cells with fibronectin via alpha 4 beta 1 and alpha 5 beta 1. J. Immunol. 154:3128-3137.
Lanier, L.L., Corliss, B.C., Wu, J., Leong, C., and Phillips, J.H. 1998a. Immunoreceptor DAP12 bearing a tyrosine-based activation motif is involved in activating NK cells. Nature 391:703707.
Helander, T.S. and Timonen, T. 1998. Adhesion in NK cell function. Curr. Top. Microbiol. Immunol. 230:89-99.
Lanier, L.L., Corliss, B., Wu, J., and Phillips, J.H. 1998b. Association of DAP12 with activating CD94/NKG2C NK cell receptors. Immunity 8:693-701.
Helander, T.S., Carpen, O., Turunen, O., Kovanen, P.E., Vaheri, A., and Timonen, T. 1996. ICAM-2 redistributed by ezrin as targets for natural killer cells. Nature 382:265-268.
Latchman, Y., McKay, P.F., and Reiser, H. 1998. Identification of the 2B4 molecule as a counter receptor for CD48. J. Immunol. 161:5809-5812.
Houchins, J.P., Lanier, L.L., Niemie, E.C., Phillips, J.H., and Ryan, J.C. 1997. Natural killer cell cytolytic activity is inhibited by NKG2-A and activated by NKG2-C. J. Immunol. 158:36033609.
Lazetic, S., Chang, C., Houchins, J.P., Lanier, L.L., and Phillips, J.H. 1996. Human natural killer receptors involved in MHC class I recognition are disulfide-linked heterodimers of CD94 and NKG2 subunits. J. Immunol. 157:4741-4745.
Ihle, J.N., Thierfelder, W., Teglund, S., Stravapodis, D., Wang, D., Feng, J., and Parganas, E. 1998. Signaling by the cytokine receptor superfamily. Ann. N.Y. Acad. Sci. 865:1-9.
Leibson, P.J. 1995. MHC-recognizing receptors: They’re not for T cells anymore. Immunity 3:5-8.
Jackman, J.K., Motto, D.G., Sun, Q., Tanemoto, M., Turck, C., Peltz, G., Koretzky, G., and Findell, P. 1995. Molecular cloning of SLP-76, a 76-kDa tyrosine phosphoprotein associated with Grb2 in T cells. J. Biol. Chem. 270:7029-7032. Jevremovic, D., Billadeau, D.D., Schoon, R.A., Dick, C.J., Irvin, B.J., Zhang, W., Samelson, L.E., Abraham, R.T., and Leibson, P.J. 1999. A role for the adaptor protein LAT in human NK cell-mediated cytotoxicity. J. Immun ol. 162:2453-2456. Kanakaraj, P., Duckworth, B., Azzoni, L., Kamoun, M., Cantley, L.C., and Perussia, B. 1994. Phosphatidylinositol-3 kinase activation induced upon FcγIIIA-ligand interaction. J. Exp. Med. 179:551-558. Kaufman, D.S., Schoon, R.A., Robertson, M.J., and Leibson, P.J. 1995. Inhibition of selective signaling events in natural killer cells recognizing major histocompatibility complex class I. Proc. Natl. Acad. Sci. U.S.A. 92:6484-6488. Kolanus, W., Romeo, C., and Seed, B. 1993. T cell activation by clustered tyrosine kinases. Cell 74:171-183. Kurosaki, T., Gander, I., and Ravetch, J.V. 1991. A subunit common to an IgG Fc receptor and the T-cell receptor mediates assembly through different interactions. Proc. Natl. Acad. Sci. U.S.A. 88:3837-3841. Lanier, L.L. 1997. Natural killer cells: from no receptors to too many. Immunity 6:371-378. Lanier, L.L. 1998. NK cell receptors. Annu. Rev. Immunol. 16:359-393. Lanier, L.L., Yu, G., and Phillips, J.H. 1989. Co-association of CD3ζ with a receptor (CD16) for IgG Fc on human natural killer cells. Nature 342:803-805. Signal Transduction During Natural Killer Cell Activation
Lanier, L.L., Chang, C., and Phillips, J.H. 1994. Human NKR-P1A. A disulfide-linked homodimer of the C-type lectin superfamily expressed by a subset of NK and T lymphocytes. J. Immunol. 153:2417-2428.
Leibson, P.J. 1997. Signal transduction during natural killer cell activation: Inside the mind of a killer. Immunity 6:655-661. Long, E.O. and Wagtmann, N. 1997. Natural killer cell receptors. Curr. Opin. Immunol. 9:344-350. Lopez-Botet, M., Carretero, M., Bellon, T., Perezvillar, J.J., Llano, M., and Navarro, F. 1998. The CD94/NKG2 C-type lectin receptor complex. Curr. Top. Microbiol. Immunol. 230:41-52. Maghazachi, A.A., Al-Aoukaty, A., Naper, C., Torgersen, K.L., and Rolstad, B. 1996. Preferential involvement of Go and Gz proteins in mediating rat natural killer cell lysis of allogenic and tumor target cells. J. Immunol. 157:5308-5314. Mainiero, F., Gismondi, A., Soriani, A., Cippitelli, M., Palmieri, G., Jacobelli, J., Piccoli, M., Frati, L., and Santoni, A. 1998. Integrin-mediated rasextracellular regulated kinase (ERK) signaling regulates interferon gamma production in human natural killer cells. J. Exp. Med. 188:1267-1275. Mason, L.H., Anderson, S.K., Yokoyama, W.M., Smith, H.R.C., Winkler-Pickett, R., and Ortaldo, J.R. 1996. The Ly-49D receptor activates murine natural killer cells. J. Exp. Med. 184:2119-2128. McVicar, D.W., Taylor, L.S., Gosselin, P., WilletteBrown, J., Mikhael, A.I., Gehlen, R.L., Nakamura, M.C., Linnemeyer, P., Seaman, W.E., Anderson, S.K., Ortaldo, J.R., and Mason, L.H. 1998. DAP12-mediated signal transduction in natural killer cells. A dominant role for Syk protein tyrosine kinase. J. Bio l. Ch em. 273:32934-32942. Melero, I., Balboa, M.A., Alonso, J.L., Yague, E., Pivel, J.P., Sanchez-Madrid, F., and Lopez-Botet, M. 1993. Signaling through the LFA-1 leukocyte integrin actively regulates intracellular adhesion and tumor necrosis factor-α production in natural killer cells. Eur. J. Immunol. 23:1859-1865. Milella, M., Gismondi, A., Roncaioli, P., Bisogno, L., Palmieri, G., Frati, L., Cifone, M.G., and Santoni, A. 1997. CD16 cross-linking induces both secretory and extracellular signal-regulated kinase (ERK)-dependent cytosolic phospholipase A2 (PLA)2 activity in human natural killer
9.16.10 Supplement 14
Current Protocols in Cytometry
cells. Involvement of ERK, but not PLA2, in CD16-triggered granule exocytosis. J. Immunol. 158:3148-3154.
Perussia, B. 1991. Lymphokine-activated killer cells, natural killer cells and cytokines. Curr. Opin. Immunol. 3:49-55.
Miyazaki, T., Dierich, A., Benoist, C., and Mathis, D. 1996. Independent modes of natural killing distinguished in mice lacking Lag3. Science 272:405-408.
Perussia, B. 1998. Fc receptors on natural killer cells. Curr. Top. Microbiol. Immunol. 230:63-88.
Nakamura, M.C., Niemi, E.C., Fisher, M.J., Shulz, L.D., Seaman, W.E., and Ryan, J.C. 1997. Mouse Ly-49A interrupts early signaling events in natural killer cell cytotoxicity and functionally associates with the SHP-1 tyrosine phosphatase. J. Exp. Med. 185:673-684.
Pessino, A., Sivori, S., Bottino, C., Malaspina, A., Morelli, L., Moretta, L., Biassoni, R., and Moretta, A. 1998. Molecular cloning of NKp46: A novel member of the immunoglobulin superfamily involved in triggering of natural cytotoxicity. J. Exp. Med. 188:953-960.
Neel, B.H. 1997. Role of phosphatases in lymphocyte activation. Curr. Opin. Immunol. 9:405-420.
Peterson, E.J., Clements, J.L., Fang, N., and Koretzky, G.A. 1998. Adaptor proteins in lymphocyte antigen-receptor signaling. Curr. Opin. Immunol. 10:337-344.
Negishi, I., Motoyama, N., Nakayama, K., Senju, S., Hatakeyama, S., Zhang, Q., Chan, A.C., and Loh, D.Y. 1995. Essential role for ZAP-70 in both positive and negative selection of thymocytes. Nature 376:435-438.
Pignata, C., Prasad, K.V.S., Robertson, M.K., Levine, H., Rudd, C.E., and Ritz, J. 1993. FcγRIIIA-mediated signaling involves src family lck in human natural killer cells. J. Immunol. 151:6794-6800.
Newton, R.A., Thiel, M., and Hogg, N. 1997. Signaling mechanisms and the activation of leukocyte integrins. J. Leukoc. Biol. 61:422-426.
Qian, D. and Weiss, A. 1997. T cell antigen receptor signal transduction. Curr. Opin. Cell Biol. 9:205212.
Olcese, L., Lang, P., Vely, F., Cambiaggi, A., Marguet, D., Blery, M., Hippen, K.L., Biassoni, R., Moretta, A., Moretta, L., Cambier, J.C., and Vivier, E. 1996. Human and mouse killer-cell inhibitory receptors recruit PTP1C and PTP1D protein tyrosine phosphatases. J. Immunol. 156:4531-4534.
Ricciarda, G., Palmieri, G., Piccoli, M., Frati, L., and Santoni, A. 1996. CD16-mediated p21ras activation is associated with Shc and p36 tyrosine phosphorylation and their binding with Grb2 in human natural killer cells. J. Exp. Med. 183:179186.
Orloff, D.G., Ra, C., Frank, S.J., Klausner, R.D., and Kinet, J.-P. 1990. Family of disulphide-linked dimers containing the ζ and η chains of the T-cell receptor and the γ chain of Fc receptors. Nature 347:189-191. O’Shea, J.J., Weissman, A.M., Kennedy, I.C.S., and Ortaldo, J.R. 1991. Engagement of the natural killer cell IgG Fc receptor results in tyrosine phosphorylation of the ζ chain. Proc. Natl. Acad. Sci. U.S.A. 88:350-354. O’Shea, J.J., McVicar, D.W., Kuhns, D.B., and Ortaldo, J.R. 1992. A role for protein tyrosine kinase activity in natural cytotoxicity as well as antibody-dependent cellular cytotoxicity. J. Immunol. 148:2497-2502. Palmieri, G., Serra, A., De Maria, R., Gismondi, A., Milella, M., Piccoli, M., Frati, L., and Santoni, A. 1995. Cross-linking of alpha 4 beta 1 and alpha 5 beta 1 fibronectin receptors enhances natural killer cytotoxic activity. J. Immunol. 155:5314-5322. Penix, L., Weaver, W.M., Pang, Y., Young, H.A., and Wilson, C.B. 1993. Two essential regulatory elements in the human interferon gamma promoter confer activation specific expression in T cells. J. Exp. Med. 178:1483-1496. Perez-Villar, J.J., Melero, I., Rodriguez, A., Carretero, M., Aramburu, J., Sivori, S., Orengo, A.M., Moretta, A., and Lopez-Botet, M. 1995. Functional ambivalence of the Kp43 (CD94) NK cell-associated surface antigen. J. Immunol. 154:5779-5788.
Ryan, J.C., Niemi, E.C., Goldfien, R.D., Hiserodt, J.C., and Seaman, W.E. 1991. NKR-P1, and activating molecule on rat natural killer cells, stimulates phosphoinositide turnover and a rise in intracellular calcium. J. Immunol. 147:32443250. Ryan, J.C., Niemi, E.C., Nakamura, M.C., and Seaman, W.E. 1995. NKR-P1A is a target-specific receptor that activates natural killer cell cytotoxicity. J. Exp. Med. 181:1911-1915. Salcedo, T.W., Kurosaki, T., Kanakaraj, P., Ravetch, J.V., and Perussia, B. 1993. Physical and functional association of p56lck with FcγRIIIA (CD16) in natural killer cells. J. Exp. Med. 277:1475-1480. Scharenberg, A.M. and Kinet, J.-P. 1998. PtdIns3,4,5-P3: A regulatory nexus between tyrosine kinases and sustained calcium signals. Cell 94:58. Schatzle, J.D., Sheu, S., Stepp, S., Mathew, P.A., Bennett, M., and Kumar, V. 1999. Characterization of inhibitory and stimulatory forms of the murine natural killer receptor 2B4. Proc. Natl. Acad. Sci. U.S.A. 96:3870-3875. Seaman, W.E., Eriksson, E., Dobrow, R., and Imboden, J.B. 1987. Inositol trisphosphate is generated by a natural killer cell tumor line in response to target cells or to cross-linked monoclonal antibody OX-34: possible signaling role for the OX-34 determinant during activation by target cells. Proc. Natl. Acad. Sci. U.S.A. 84:4239-4243. Studies of Cell Function
9.16.11 Current Protocols in Cytometry
Supplement 14
Shibuya, A., Campbell, D., Hannum, C., Yssel, H., Franz-Bacon, K., McClanahan, T., Kitamura, T., Nicholl, J., Sutherland, G.R., Lanier, L.L., and Phillips, J.H. 1996. DNAM-1, a novel adhesion molecule involved in the cytolytic function of T lymphocytes. Immunity 4:573-581.
Trotta, R., Puorro, K.A., Paroli, M., Azzoni, L., Abebe, B., Eisenlohr, L.C., and Perussia, B. 1998. Dependence of both spontaneous and antibody-dependent, granule exocytosis-mediated NK cytotoxicity on extracellular signal-regulated kinases. J. Immunol. 152:6648-6656.
Shibuya, A., Lanier, L.L., and Phillips, J.H. 1998. Protein kinase C is involved in the regulation of both signaling and adhesion mediated by DNAX accessory molecule-1 receptor. J. Immunol. 161:1671-1676.
Valiante, N.M., Phillips, J.H., Lanier, L.L., and Parham, P. 1996. Killer cell inhibitory receptor recognition of human leukocyte antigen (HLA) class I blocks formation of a pp36/PLC-γ signaling complex in human natural killer (NK) cells. J. Exp. Med. 184:2243-2250.
Smith, K.M., Wu, J., Bakker, A.B., Phillips, J.H., and Lanier, L.L. 1998. Ly-49D and Ly-49H associate with mouse DAP12 and form activating receptors. J. Immunol. 161:7-10. Songyang, Z., Shoelson, S.E., Chaudhuri, M., Gish, G., Pawson, T., Haser, W.G., King, F., Roberts, T., Rafnofsky, S., Lechleider, R.J., Neel, B.G., Birge, R.B., Fajardo, J.E., Chou, M.M., Hanafusa, H., Schaffhausen, B., and Cantley, L.C. 1993. SH2 domains recognize specific phosphopeptide sequences. Cell 72:767-778. Stahls, A., Liwszyc, G.E., Couture, C., Mustelin, T., and Andersson, L.C. 1994. Triggering of human natural killer cells through CD16 induces tyrosine phosphorylation of the p72syk kinase. Eur. J. Immunol. 24:2491-2496. Steele, T.A. and Brahmi, Z. 1988. Phosphatidylinositol metabolism accompanies early activation events in tumor target cell–stimulated human natural killer cells. Cell. Immunol. 112:402-413. Sugie, K., Minami, Y., Kawakami, T., and Uchida, A. 1995. Stimulation of NK cells via leukocyte function associated antigan (LFA)-1. J. Immunol. 154:1691-1698. Tamir, I. and Cambier, J.C. 1998. Antigen receptor signaling: Integration of protein tyrosine kinase functions. Oncogene 17:1353-1364. Ting, A.T., Einspahr, K.J., Abraham, R.T., and Leibson, P.J. 1991. Fcγ receptor signal transduction in natural killer cells. Coupling to phospholipase C via a G protein-independent, but tyrosine kinase-dependent pathway. J. Immunol. 147:31223127. Ting, A.T., Karnitz, L.M., Schoon, R.A., Abraham, R.T., and Leibson, P.J. 1992. Fcγ receptor activation induces the tyrosine phosphorylation of both phospholipase C (PLC)-γ1 and PLC-γ2 in natural killer cells. J. Exp. Med. 176:1751-1755. Ting, A.T., Dick, C.J., Schoon, R.A., Karnitz, L.M., Abraham, R.T., and Leibson, P.J. 1995. Interaction between lck and syk family tyrosine kinases in Fcγ receptor-initiated activation of natural killer cells. J. Biol. Chem. 270:16415-16421. Trotta, R., Kanakaraj, P., and Perussia, B. 1996. FcγR-dependent MAP kinase activation in leukocytes: A common signal transduction event necessary for expression of TNF-α and early activation genes. J. Exp. Med. 184:1027-1035. Signal Transduction During Natural Killer Cell Activation
Vely, F. and Vivier, E. 1997. Conservation of structural features reveals the existence of a large family of inhibitory cell surface receptors and noninhibitory/activatory counterparts. J. Immunol. 159:2075-2077. Vitale, M., Bottino, C., Sivori, S., Sanseverino, L., Castriconi, R., Marcenaro, E., Augugliaro, R., Moretta, L., and Moretta, A. 1998. NKp44, a novel triggering surface molecule specifically expressed by activated natural killer cells, is involved in non-major histocompatibility complex-restricted tumor cell lysis. J. Exp. Med. 187:2065-2072. Vivier, E., Morin, P., O’Brien, C., Druker, B., Schlossman, S.F., and Anderson, P. 1991. Tyrosine phosphorylation of the FcγRIII(CD16): ζ complex in human natural killer cells. Induction by antibody-dependent cytotoxicity but not by natural killing. J. Immunol. 146:206-210. Vivier, E., da Silva, A.J., Ackerly, M., Levine, H., Rudd, C.E., and Anderson, P. 1993. Association of a 70-kDa tyrosine phosphoprotein with the CD16:ζ :γ complex expressed in human natural killer cells. Eur. J. Immunol. 23:1872-1876. Wang, K.S., Ritz, J., and Frank, D.A. 1999. IL-2 induces STAT4 activation in primary NK cells and NK cell lines, but not T cells. J. Immunol. 162:299-304. Wei, S., Gamero, A.M., Liu, J.H., Daulton, A.A., Valkov, N.I., Trapani, J.A., Larner, A.C., Weber, M.J., and Djeu, J.Y. 1998. Control of lytic function by mitogen-activated protein kinase/extracellular regulatory kinase 2 (ERK2) in a human natural killer cell line: Identification of perforin and granzyme B mobilization by functional ERK2. J. Exp. Med. 187:1763-1765. Weiss, A. and Littman, D.R. 1994. Signal transduction by lymphocyte antigen receptors. Cell 76:263-274. Welsh, R.M. 1984. Natural killer cells and interferon. Crit. Rev. Immunol. 5:55-93. Wen, T., Zhang, L., Kung, S.K.P., Molina, T.J., Miller, R.G., and Mak, T.W. 1995. Allo-skin graft rejection, tumor rejection and natural killer activity in mice lacking p56lck. Nature 357:161164. Whalen, M.M., Doshi, R.N., and Bankhurst, A.D. 1992. Effects of pertussis toxin treatment in natural killer function. Immunology 75:402-407.
9.16.12 Supplement 14
Current Protocols in Cytometry
Williams, B.L., Schreiber, K.L., Zhang, W., Wange, R.L., Samelson, L.E., Leibson, P.J., and Abraham, R.T. 1998. Genetic evidence for differential coupling of syk family kinases to the T-cell receptor: Reconstitution studies in a ZAP-70-deficient Jurkat T cell line. Mol. Cell. Biol. 18:13881399. Windebank, K.P., Abraham, R.T., Powis, G., Olsen, R.A., Barna, T.J., and Leibson, P.J. 1988. Signal transduction during natural killer cell activation: Inositol generation and regulation by cAMP. J. Immunol. 141:3951-3957. Xu, X. and Chong, A.S.-F. 1996. Vav in natural killer cells is tyrosine phosphorylated upon cross-linking of FcγRIIIA and is constitutively associated with a serine/threonine kinase. Biochem. J. 318:527-532. Yablonski, D., Kuhne, M.R., Kadlecek, T., and Weiss, A. 1998. Uncoupling of nonreceptor tyrosine kinases from PLC-γ1 in an SLP-76-deficient T cell. Science 281:413-415. Yi, T., Cleveland, J.L., and Ihle, J.N. 1992. Protein tyrosine phosphatase containing SH2 domains: Characterization, preferential expression in hematopoietic cells, and localization to human chromosome 12p12-p13. Mol. Cell. Biol. 12:836-846.
Yokoyama, W.M. 1998. Natural killer cell receptors. Curr. Opin. Immunol. 10:298-305. Yu, C.R., Young, H.A., and Ortaldo, J.R. 1998. Characterization of cytokine differential induction of STAT complexes in primary human T and NK cells. J. Leukoc. Biol. 64:245-258. Zhang, W., Sloan-Lancaster, J., Kitchen, J., Trible, R.P., and Samelson, L.E. 1998. LAT: The ZAP70 tyrosine kinase substrate that links T cell receptor to cellular activation. Cell 92:83-92. Zoller, K.E., MacNeil, I.A., and Brugge, J.S. 1997. Protein tyrosine kinases syk and ZAP-70 display distinct requirements for src family kinases in immune response receptor signal transduction. J. Immunol. 158:1650-1659.
Contributed by Claudia C.S. Chini and Paul J. Leibson Mayo Clinic and Foundation Rochester, Minnesota
Studies of Cell Function
9.16.13 Current Protocols in Cytometry
Supplement 14
Assessment of Surface Markers and Functionality of Dendritic Cells (DCs)
UNIT 9.17
Dendritic cells (DCs) are a complex group of mainly bone marrow–derived cells that have been found to play an important role in the afferent branch of the immune response (Steinman, 1991). However, DCs represent only a minute subpopulation of peripheral blood mononuclear cells, as well as of bulk cellular populations of the lung, intestine, genitourinary tissue, and lymphoid tissue. DCs have also been found in the epidermis, dermis, and mucous membranes and constitute ∼2% of the total cellular population of the human epidermis (Katz et al., 1985). The so-called Langerhans cells (LCs) are skin-derived DCs that migrate to regional lymphoid organs after takeup of antigen and undergo an activation/maturation step. Thereafter, LCs interact with and activate T cells. Because of their significant capability to take up, process, and present soluble antigens to responder cells in the lymphoid tissues in the context of the restricted MHC pathway, LCs have been considered one of the most important elements in the afferent arm of the immune response (Katz et al., 1985; Steinman, 1991; Cella et al., 1997; Banchereau and Steinman, 1998). This unit presents protocols for flow cytometric assessment of both surface markers (see Basic Protocol 1), and DC functionality, including antigen uptake (see Basic Protocol 2), cytokine-modulated cell division (see Basic Protocol 3), in vitro cytotoxicity (see Basic Protocol 4), and in vivo cytotoxicity (see Basic Protocol 5). A support protocol describes the generation of DCs and LCs from peripheral blood mononuclear cells (PBMCs). IMMUNOPHENOTYPING OF DENDRITIC CELLS Surface antigens on DCs have been studied by diverse flow cytometric approaches. The panel of monoclonal antibodies used for DC immunophenotyping may be either unlabeled (indirect method) or labeled with FITC, PE, APC, or PerCP (direct method). The negative controls should be isotype-matched FITC-, PE-, PerCP-, or APC-labeled unrelated antibodies. The isotype controls are used to set up markers in the plots. Single-color or multicolor immunofluorescence analysis can be performed. The latter requires determination of appropriate compensation for each color (UNIT 1.14). Sample histograms from a single-color assay can be overlaid with isotype control histograms as shown in Figure 9.17.1. The overlay representation has the advantage of providing visual assessment of the degree of similarity or difference between the DC marker and the isotype control. Density plots of mAb reactivity versus morphology, either size (FS) or granularity (SS), are very useful in defining subsets of DCs as shown in Figure 9.17.2.
BASIC PROTOCOL 1
With the indirect method, cells are harvested and incubated with unlabeled monoclonal antibody from either cell culture supernatant or ascites fluid. After washing, they are labeled with a fluorochrome-conjugated second antibody, washed again, and fixed for cytometric analysis. In the direct method, cells are harvested and incubated with one or more fluorochrome-labeled monoclonal antibodies. They are then washed and fixed for cytometric analysis. A panel of mAbs with potential reactivity against DCs should be used (Hock et al., 1994; Jiang et al., 1995; Bender et al., 1996). The author suggests using the panel of mAbs submitted to the dendritic cell section of the 7th International Workshop and Conference on Human Leukocyte Differentiation Antigens, which defined the CD nomenclature for some DC markers (Hart et al., 2001; Nuñez and Filgueira, 2001).
Studies of Cell Function Contributed by Rafael Nuñez Current Protocols in Cytometry (2001) 9.17.1-9.17.15 Copyright © 2001 by John Wiley & Sons, Inc.
9.17.1 Supplement 17
control 80
0
0
200
400
600
800
1000
0
0
FL1H (FITC) 36
70
CD4
No. of cells
No. of cells
IgG
200
400
600
800
FL1H (FITC)
200
1000
400 600
800 1000
MHC CLASS II
90
No. of cells
No. of cells
No. of cells
200
0
FL1H (FITC)
50
CD45
0
0
400 600 800 1000
FL1H (FITC) B220/CD45R
0
Mac1/CD11b
No. of cells
70
sample
0
0 0
200 400 600 800 1000
FL1H (FITC)
0
200
400 600
800 1000
FL1H (FITC)
Figure 9.17.1 Immunophenotyping of DCs by flow cytometry. A mouse DC line was immunophenotyped with a panel of monoclonal antibodies. Histograms of the samples are overlaid with those of the isotype control.
Materials Dendritic cells (DCs; see Support Protocol) Complete medium for DC culture (see recipe) Unlabeled monoclonal antibodies (mAbs) and/or FITC-, PE-, PerCP-, and APC-labeled mAbs: CD1a, CD1b/c, BG6, HP-F1 (CD85i), BU10, RFD-1, CMRF-44, 7H5 (CD85a), ZM3.8 (CD85j) , 55K-2 (fascin), MMR1.16, MMR190.BB3 (CD206), L25, CMRF-56, RFD-7, MR15-2 (CD205), DCGM-4 (CD207), TPD153, 42D1 (CD85f), DEC-205 (CD205), MMRI-4 (CD205), DC-LAMP (CD208), AZN-D1 (CD209), AZN-D2 (CD209), CMRF-75, CMRF-82, CD11c, CD80, CD86, and HLA-DR Phosphate-buffered saline (PBS; APPENDIX 2A) or stain buffer (Pharmingen) FITC-labeled anti-mouse Ig (for the indirect method; Pharmingen) Cellfix (Becton Dickinson)
Assessment of Surface Markers and Functionality of Dendritic Cells (DCs)
5-ml round-bottom tubes (Falcon) Centrifuge, 4°C Flow cytometer with 488-nm excitation (e.g., FACSCalibur; Becton Dickinson) and 530 ± 15 nm band-pass filter for collecting green fluorescence Software for analysis (e.g., CellQuest, Becton Dickinson; FlowJo, Flow Jo; or WinMDI) Additional reagents and equipment for counting cells in a hemacytometer (APPENDIX 3A)
9.17.2 Supplement 17
Current Protocols in Cytometry
B
104
101
102 103 FL1-Height
104
iDC
100
102 103 FL1-Height 4.005
104
101
102 103 FL1-Height 44.020
104
100
102 103 FL1-Height
102 103 FL1-Height
F
104
100
101
102 103 FL1-Height
104
28.004
0
101
6.007
104
0
101
100
FSC-Height 200 400 600 800 1000
101
D
0
0
FSC-Height 200 400 600 800 1000
20.021
100
100
100
101
102 103 FL1-Height
104
30.006
FSC-Height 200 400 600 800 1000
102 103 FL1-Height
FSC-Height 200 400 600 800 1000
0
FSC-Height 200 400 600 800 1000
104
FSC-Height 200 400 600 800 1000
101
E
100
102 103 FL1-Height 43.019
FSC-Height 200 400 600 800 1000
100
101
0
0
FSC-Height 200 400 600 800 1000
19.020
100
48.024
0
104
FSC-Height 200 400 600 800 1000
C
102 103 FL1-Height
0
101
24a.025
0
FSC-Height 200 400 600 800 1000
0
100
25.001
0
FSC-Height 200 400 600 800 1000
1.002
FSC-Height 200 400 600 800 1000
A
101
102 103 FL1-Height
LC mDC
104
100
iDC
101
102 103 FL1-Height
104
LC mDC
Figure 9.17.2 Representative mAb reactivity against DCs (iDCs and LC-mDCs) at 7 days of culture.
Stain cells 1. Harvest dendritic cells (DCs), resuspend in 10 ml complete medium for DC culture, and count. Adjust concentration to 106 cells/ml. Perform at least three independent sets of experiments.
2. Place 100 µl DC cell suspension in 5-ml round-bottom tubes. 3a. For indirect method (single-color staining): Add 10 µl mAb from cell culture supernatant or 1 µl mAb from ascites. 3b. For direct method (multicolor staining): Add 1 µl of each FITC-, PE-, PerCP-, and APC-labeled mAb. 4. Put tubes on ice, cover with aluminum foil, and incubate 30 min. 5. Wash cells with 2 ml PBS or stain buffer. Centrifuge 10 min at 300 × g, 4°C, and discard supernatant. For direct method, proceed to step 8. For indirect method, repeat wash step and continue to step 6. 6. Stain cells by adding 10 µl diluted FITC-labeled anti-mouse Ig to the pellet and gently mix. Dilute antibody 1:100 in PBS. Store aliquots ≤1 year at −20°C.
7. Incubate 30 min on ice. 8. Wash one more time as described in step 5 and fix with 200 µl Cellfix.
Studies of Cell Function
9.17.3 Current Protocols in Cytometry
Supplement 17
Table 9.17.1 Data Aquisition Parameters for a FACSCalibur Flow Cytometer
Detector FS (488 nm) SS (488 nm) FL1 (green) for FITC (530 nm) FL2 (orange) for PE (585 nm) FL3 (red) for PerCP (670 nm) FL4 (red) for APC (661 nm)
Voltage
AmpGain
Mode
E00 350 582 600 650 800
1 1 1 1 1 1
Lin Lin Log Log Log Log
Analyze on flow cytometer 9. First measure control populations, unstained or stained only with the secondary antibody (2° Ab). Perform all measurements within a single day and under similar equipment settings in order to avoid instrument variation. Secondary Ab only–stained cells serve for determining the markers for the negative population in the histograms and for setting up the quadrant borders in the density plots. At least 99% of the negative cells should be located on the left side of the histogram markers, or if density plots are used, 99% of the negative cells should be located in the lower left quadrant (negative).
10a. For single-color staining: Collect forward scatter (FS) at 488 nm, side scatter (SS) at 488 nm, and log green fluorescence (FITC) at 530 nm. Acquire a minimum of 104 cells per sample, and store data as listmode files. 10b. For multicolor staining: Collect FS at 488 nm, SS at 488 nm, green fluorescence for FITC at 530 nm, orange fluorescence for PE at 585 nm, red fluorescence for PerCP at 670 nm, and red fluorescence for APC at 661 nm. For an example of data acquisition parameters on a FACSCalibur flow cytometer see Table 9.17.1. Set up appropriate compensation for each color. Typical values on the FACSCalibur are green (FL1) − 1.5% orange (FL2), orange (FL2) − 25% green (FL1), orange (FL2) − 0% red 1 (FL3), red 1 (FL3) − 15.4% orange (FL2), red 1 (FL3) − 25% red 2 (FL4), red 2 (FL4) − 1.8% red 1 (FL3).
11. Set up a histogram of forward light scatter (FS) versus orthogonal scatter (SS) and draw a gate around the DCs. 12. Analyze the gated cells by drawing histogram plots displaying the fluorescent reactivity collected at 530 nm (green channel) and density plots displaying the fluorescent reactivity collected at 530 nm (green channel) and FS (488 nm). 13. Analyze data with software for analysis like CellQuest, FlowJo, or WinMDI. Perform statistical analysis (mean, standard deviation) and graphics. BASIC PROTOCOL 2
Assessment of Surface Markers and Functionality of Dendritic Cells (DCs)
DETERMINATION OF ANTIGEN UPTAKE BY DENDRITIC CELLS Antigen uptake, one of the most important functions of DCs, can also be assayed by flow cytometry. DCs are pulsed with an antigen conjugated to fluorescein and then the DC-antigen mixture is incubated at 37°C for various lengths of time. Control cells incubated with the same amount of antigen are kept on ice for the same period of time. Antigen uptake is stopped by adding ice-cold FACS buffer, which inhibits cell metabolism. Subsequently, the cells are extensively washed and fixed with 2% formaldehyde
9.17.4 Supplement 17
Current Protocols in Cytometry
10 minutes
80
66%
96%
MB
MFI 384
No. of cells
Time 0
30 minutes
80
MB
MFI 452
80
No. of cells
14%
MB
MB 0
MFI 275
0
200
200 400 600 800 1000 FL1-H(OVA-FITC)
0
0
200
400 600 800 FL1-H(OVA-FITC)
1000
120 minutes
90
81%
MB
85%
MFI 401
No. of cells
0
1000
60 minutes
90 0
400 600 800 FL1-H(OVA-FITC)
MB
MFI 409
MB 0
0
200
400 600 800 FL1-H(OVA-FITC)
1000
0
0
200
400 600 800 FL1-H(OVA-FITC)
1000
Figure 9.17.3 Antigen uptake determined by flow cytometry. Human DCs were pulsed with antigen conjugated with fluorescein (OVA-FITC) and incubated at 37°C for varying lengths of time. Samples were evaluated at 0, 10, 30, 60, and 120 min. Histograms of the samples are overlaid with those of the control.
before cytometric analysis. Data collected on antigen binding of control cells kept on ice throughout the entire experiment are used as a control. The overlay representation complemented with the assessment of the percent of positive cells and the measurement of mean fluorescence intensity (MFI) of the sample provide a clear determination of the DC uptake capability as shown in Figure 9.17.3 (Nuñez et al., 1998). Materials Dendritic cells (DCs; see Support Protocol) OVA-FITC antigen (Molecular Probes); store in aliquots ≤1 year at −20°C FACS buffer (see recipe) Fixative (e.g., Cellfix, Becton Dickinson; or PBS [APPENDIX 2A] containing 2% formaldehyde) 5-ml round-bottom tubes (Falcon) 37°C, 5% CO2 incubator Flow cytometer (e.g., FACSCalibur; Becton Dickinson) Additional reagents and equipment for counting cells in a hemacytometer (APPENDIX 3A) 1. Harvest dendritic cells, resuspend in 10 ml complete medium, and count. Adjust concentration to 106 cells/ml. 2. Place 100 µl cell suspension in 5-ml round-bottom tubes and incubate 10 min on ice.
Studies of Cell Function
9.17.5 Current Protocols in Cytometry
Supplement 17
3. Prepare graded doses of OVA-FITC antigen and add to cells (final concentration 0.001 to 1 mg/ml). Prepare controls with the same doses. Leave controls on ice; do not incubate at 37°C. 4. Incubate DC-antigen mixture in a 37°C, 5% CO2 incubator, protected from light. 5. Evaluate samples at 0, 10, 30, 60, and 120 min. 6. Stop antigen uptake by adding ice-cold 2 ml FACS buffer. The FACS buffer contains PBS supplemented with 5% FCS and 0.01% sodium azide that inhibit cell metabolism.
7. Wash cells with FACS buffer three times by gently centrifuging 10 min at 300 × g, 4°C. 8. Fix cells with 200 µl Cellfix or with PBS containing 2% formaldehyde. 9. Run on flow cytometer. Collect log green fluorescence (FITC) at 530 ± 15 nm. BASIC PROTOCOL 3
ASSESSMENT OF DENDRITIC CELL DIVISION UNDER CYTOKINE MODULATION The stable intracytoplasmic dye 5-(and-6)-carboxyfluorescein diacetate succinimidyl ester (CFSE) is used as a protein-binding dye for measuring and tracking cell division (UNIT 9.11). This method allows identification of cell progeny and analysis of the division history of individual cells that have undergone multiple rounds of division. This kinetic process is seen during the flow cytometric evaluation as discrete peaks with a progressive reduction of CFSE fluorescence intensity (Kamau et al., 2000). DC proliferation in the presence and absence of cytokines was determined by CFSE staining. Results from each treatment and time interval were grouped and the density plots show the decrease in CSFE fluorescence intensity in subsequent days (Figure 9.17.4). Materials Dendritic cells (DCs) Complete medium (see recipe) rhGM-CSF (Novartis) rhIL-4 (Genzyme) rhTGF-β1 (R&D Systems; optional) PBS (APPENDIX 2A) 5-(and-6)-carboxyfluorescein diacetate succinimidyl ester (CFSE; Molecular Probes) 25-cm2 tissue culture flasks 5-ml round-bottom tubes (Falcon) 37°C, 5% CO2 incubator Flow cytometer (e.g., FACSCalibur, Becton Dickinson) Additional reagents and equipment for counting cells in a hemacytometer (APPENDIX 3A) 1. Culture DCs in 25-cm2 tissue culture flasks with complete DC medium containing 250 ng/ml rhGM-CSF and 100 ng/ml rhIL-4, with or without 10 ng/ml rhTGF-β1. Store aliquots of rhGM-CSF, rhIL-4, and rhTGF-β1 ≤1 year at −70°C.
Assessment of Surface Markers and Functionality of Dendritic Cells (DCs)
2. Transfer to 5-ml round-bottom tubes and wash DCs two times with PBS by centrifuging 10 min at 300 × g, 4°C. 3. Resuspend DCs in 2 ml PBS. Count cells and adjust concentration to 5 × 107 cells/ml.
9.17.6 Supplement 17
Current Protocols in Cytometry
1000 0
0
100 101 102 103 104 CFSE fluorescence
100
101 102 103 104 CFSE fluorescence
idc0hoursa250.009
0
0
FSC-Height
1000
idc36hours.005
FSC-Height
1000
idc60hoursa250.013
0
dc-lc0hoursa250.010
FSC-Height
1000
dc-lc36hoursa250.011
FSC-Height
1000 FSC-Height 1000
idc96hoursa250.013
0
0
100 101 102 103 104 CFSE fluorescence
FSC-Height
1000
101 102 103 104 CFSE fluorescence
FSC-Height
1000
idc162hoursa250.014
FSC-Height
dc-lc60hoursa250.012
0
0
100
100 101 102 103 104 CFSE fluorescence
iDC
dc-lc96hoursa250.012
FSC-Height
1000
1000 0
DC-LC
FSC-Height
dc-LC162hoursa250a.016
100 101 102 103 104 CFSE fluorescence
100 101 102 103 104 CFSE fluorescence
100 101 102 103 104 CFSE fluorescence
100 101 102 103 104 CFSE fluorescence
162 hours
96 hours
60 hours
36 hours
100 101 102 103 104 CFSE fluorescence
0 hours
Figure 9.17.4 Human DC proliferation in the presence and absence of cytokines, determined by CFSE staining. Results from each treatment and time interval were grouped and the density plots show the decrease in CFSE fluorescence intensity in subsequent days. Five panels of density plots (upper row) show the CFSE fluorescence for DCs under the effect of GM-CSF, IL-4, and TGF-β1, while the lower row shows the panels of DC treated with GM-CSF, IL-4, but not TGF-β1. The different panels for the DCs treated with TGF-β1 (upper row) show a series of discrete peaks exhibiting a progressive decrease in CFSE fluorescence, at different time intervals, a feature suggestive of cell division. In contrast, the lower row shows that the DCs have not decreased in fluorescence after 36 hr. In addition, treatment with TGF-β1 yields a population with high forward scatter. Overall, these results suggest that DCs treated with TGF-β1 divided and differentiated into at least two subsets, whereas the untreated DCs remained in a steady stage of differentiation (Nuñez and Filgueira, 2001).
4. Add 2.8 µg/ml CFSE and incubate cells 10 min in a 37°C, 5% CO2 incubator, mixing 3 to 4 times. 5. Add several volumes of ice-cold complete medium (up to a maximum volume of 4 ml) in order to stop CFSE binding. 6. Centrifuge 10 min at 300 × g, 4°C. Resuspend cell pellet with fresh complete medium. 7. Adjust cell density to 1 × 106 cells/ml and continue to cultivate in 25-cm2 tissue culture flasks in a 37°C, 5% CO2 incubator. 8. Determine the CFSE fluorescence signal by flow cytometry immediately after staining, and at 36, 60, 96, and 162 hr post-staining. Collect log green fluorescence at 530 ± 15 nm. ASSESSMENT OF IN VITRO CYTOTOXICITY BY FLOW CYTOMETRY The two-color assay for in vitro CTL uses DiOC18(3) (green fluorescence) as a membrane stain for target cells, and PI (red fluorescence) as a nuclear stain to identify damaged or dead target cells. The unstained effector cells can be gated out on the basis of their lower forward and side scatter characteristics. The labeling and the CTL procedure are done using the LIVE/DEAD Cell-Mediated Cytotoxicity Kit from Molecular Probes. Target and effector cells are mixed and cultured 4 hr. Then, PI is added to the cultures and incubated 30 min. The cultures are washed and analyzed on the flow cytometer.
BASIC PROTOCOL 4
Studies of Cell Function
9.17.7 Current Protocols in Cytometry
Supplement 17
101
102
104
100
104 103 102
0
101
103
Propidium iodide fluorescence
104
103
4a.068
101
102
103
104
100
101
102
103
25a.038
5a.069
1a.065
100
101
102
103
104
100
102
103
104
0
101
Propidium iodide fluorescence
103 102
0
101
Propidium iodide fluorescence
103 102 101
104
DIOC18(3) fluorescence
104
DIOC18(3) fluorescence
104
DIOC18(3) fluorescence
0
Peptide +
102
0
100
Propidium iodide fluorescence
8a.072
101
Propidium iodide fluorescence
103 102 101
0
Peptide −
Propidium iodide fluorescence
104
28a.041
101
102
103
104
100
101
102
103
DIOC18(3) fluorescence
DIOC18(3) fluorescence
DIOC18(3) fluorescence
Mock
HSV-1 gB
β gal
104
Assessment of Anti β gal CTL activity by flow cytometry
Figure 9.17.5 Flow cytometric assessment of in vitro CTL activity. The DiOC18(3)-stained live target cells are located in the lower right quadrant. The dead targets are located in the upper left quadrant. The dying cells are located in the upper right quadrant. There was ∼15% of antigen-specific CTL activity after immunization of mice with DCs transduced with amplicon(s) compared with the mock mice and between the loaded (peptide +) and unloaded (peptide −) targets. A mouse was immunized with DCs transduced with amplicon β gal and the targets were either MC57 loaded with a β gal peptide containing a CTL epitope (β gal) or MC57 loaded with a peptide containing a CTL epitope against gB (HSV-1 gB). The target negative control was MC57 alone (peptide −). A mock-immunized mouse was also assayed (Mock).
This flow cytometric approach was used for assaying the CTL activity generated in mice immunized with DCs transduced with amplicon β gal (Fig. 9.17.5). Materials Antigens Target cells PBS (APPENDIX 2A) LIVE/DEAD Cell-Mediated Cytotoxicity Kit (Molecular Probes) 3 mM DiOC18(3) stock solution in DMSO 50 µg/ml propidium iodide (PI) stock solution in PBS Complete medium supplemented with 10% (w/v) FCS (see recipe), ice-cold Effector cells (purified T or spleen cells) (Kruisbeek, 2000)
Assessment of Surface Markers and Functionality of Dendritic Cells (DCs)
37°C, 5% CO2 incubator 5-ml round-bottom tubes (Falcon) Flow cytometer (e.g., FACSCalibur; Becton Dickinson) Additional reagents and equipment for counting cells in a hemacytometer (APPENDIX 3A)
9.17.8 Supplement 17
Current Protocols in Cytometry
Prepare cells 1. Load antigens into target cells either by culturing 2 hr with peptides or by infecting with virus in a 37°C, 5% CO2 incubator. If using virus antigens, infect cells 24 hr prior to the assay.
2. Wash target cells two times with PBS and centrifuge 5 min at 300 × g, 4°C. 3. Resuspend target cells in 0.1 ml PBS. Count and adjust concentration with PBS to 5 × 107 cells/ml. Stain target cells 4. Add 1 µl (weak stain) or 10 µl (strong stain) DiOC18(3). 5. Incubate cells 10 to 20 min in a 37°C, 5% CO2 incubator, mixing 3 to 4 times. 6. Add several volumes of ice-cold complete medium supplemented with 10% FCS to stop DiOC18(3) staining. 7. Wash target cell two times with 2 ml complete medium by centrifuging 5 min at 300 × g, 4°C. 8. Resuspend cell pellets with 2 ml fresh complete medium containing 10% FCS. Prepare and add effector cells 9. Prepare effector cells (purified T or spleen cells) according to Kruisbeek (2000). 10. Mix target and effector cells in 1:100, 1:33, 1:11, and 1:3 ratios. 11. Culture the mixture in 24-well plates 4 to 6 hr in a 37°C, 5% CO2 incubator. 12. Add 10 µg/ml PI per well and incubate 30 min in a 37°C, 5% CO2 incubator. 13. Transfer the mixture to 5-ml round-bottom tubes and protect the tubes from direct light. Run sample on flow cytometer 14. Determine the fluorescence signal by flow cytometry immediately after staining. 15. Collect log green fluorescence [DiOC18(3)] at 530 ± 15 nm, log orange or red fluorescence (PI) at 585 ± 15 nm or above 590 nm. 16. Set up appropriate compensation for each color. For example, green-8.9% orange, orange-25% green. Compensation with PI collected at higher wavelengths (red) will be less.
ASSESSMENT OF IN VIVO CYTOTOXICITY BY FLOW CYTOMETRY Mice are immunized with DCs transduced with amplicon carrying the gene for the rabies glycoprotein and boosted with a commercial vaccine 4 days prior to being infused with a mixture of 2.5 × 107 target cells consisting of uninfected targets, DiOC18(3)-labeled at a low concentration, and target cells infected with the rabies virus, labeled at a high concentration. After 24 hr, mice are bled and cells are evaluated by flow cytometry (Fig. 9.17.6). It is expected that the mice primed by DCs transduced with the amplicon rabies can generate a CTL activity that destroys targets infected with the rabies virus while leaving the uninfected targets.
BASIC PROTOCOL 5
Studies of Cell Function
9.17.9 Current Protocols in Cytometry
Supplement 17
104
100
104
14.16 4.03 36.29
100
4.53 3.82 19.81
DC vaccine −
103
104
100
104
100.00 84.25 8.58
5.94 4.11 27.66
3.89 3.85 19.11
DC vaccine +
104
103
104
10 0
101
102 FL1-H
103
104
100.00 85.88 8.06
9.27 4.02 71.05
100
101
102 FL1-H
Sample ID: 5ac Gated Events: 3515
Marker %Gated Mean Median All M1 M2
103
40
Counts
30
100
Sample ID: 3ac Gated Events: 5447 Median
102 FL1-H
50
50
103
101
5ac.005
20
Counts 102 FL1-H
Marker %Gated Mean All M1 M2
102 FL1-H
10
101
Sample ID: 1ac Gated Events: 8050 Median
101
40
50 40 30
Counts 103
200 400 600 800 1000
FSC-H
100
3ac.003
10
102 FL1-H
Marker %Gated Mean 100.00 62.59 31.87
0
FSC-H
0
104
0
101
Sample ID: 1ac Gated Events: 5721
All M1 M2
103
20
40 30
Counts
20 10 0
100
102 FL1-H
1ac.002
50
1ac.001
101
30
103
5ac.005
20
102 FL1-H
200 400 600 800 1000
200 400 600 800 1000
FSC-H 101
3ac.003
0
100
1ac.002
0
0
FSC-H
200 400 600 800 1000
1ac.001
Marker %Gated Mean Median
3.82 3.79 20.54
DC vaccine +
All M1 M2
100.00 86.34 11.01
13.85 4.79 87.90
4.87 4.66 18.60
DC vaccine +
Assessment of in vivo CTL activity by flow cytometry
Figure 9.17.6 Assessment of in vivo CTL activity by flow cytometry. Three mice immunized with DCs cleared out most of the infected targets while the control did not. Mice were immunized with DCs transduced with rabies amplicon (DC vaccine +) or mock immunized (DC vaccine −). Three months later, all mice were boosted with a commercial rabies vaccine. Four days after boosting, mice were infused with a mixture of target cells consisting of uninfected targets labeled with DiOC18(3) at low concentration and target cells infected with rabies virus and labeled at high concentration. Mice were bled 24 hr later, and cells were evaluated by flow cytometry. Upper panels show the density plots of cells located in the gate for targets. Middle panels shown the histograms of the gated cells with marker for the peaks (M1 is the marker for the peak with low concentration of the label and M2 is the marker for the peak with high concentration of the label). Lower panels show statistics for the cells included into the markers.
Assessment of Surface Markers and Functionality of Dendritic Cells (DCs)
Materials Antigen Target cells of interest Virus (optional) PBS (APPENDIX 2A) LIVE/DEAD Cell-Mediated Cytotoxicity Kit (Molecular Probes) 3 mM DiOC18(3) stock solution in DMSO Complete medium (see recipe), 4°C or on ice Mice Lysis solution (Becton Dickinson) Anticoagulant (e.g., citrate, EDTA) Cellfix (Becton Dickinson) 37°C, 5% CO2 incubator Sterile 1.8-ml microcentrifuge tubes (Eppendorf)
9.17.10 Supplement 17
Current Protocols in Cytometry
5-ml round-bottom tubes (Falcon) Flow cytometer (e.g., FACSCalibur, Becton Dickinson) Additional reagents and equipment for counting cells in a hemacytometer (APPENDIX 3A) Prepare cells 1. Load antigens into target cells either by infection with virus or by transfection. Set parallel cultures with uninfected target cells. If using a virus, infect cells 24 hr prior to the assay.
2. Wash infected or loaded target cells two times with PBS by centrifuging 5 min at 300 × g, 4°C. 3. Resuspend target cells in 0.1 ml PBS without FCS. Count and adjust concentration to 5 × 107 cells/ml. Stain cells 4. Add 1 µl DiOC18(3) stock solution to uninfected target cells (weak stain) and 10 µl to infected target cells (strong stain). 5. Incubate cells 10 to 20 min in a 37°C, 5% CO2 incubator, with intermittent mixing 3 to 4 times. 6. Add several volumes of ice-cold medium supplemented with 10% FCS to stop DiOC18(3) staining. 7. Wash target cells two times by centrifuging 5 min at 300 × g, 4°C. 8. Resuspend cell pellets with fresh complete medium. Count and adjust concentration to 5 × 107 cells/ml. Infect mice and collect blood 9. Mix equal volumes of targets (infected and noninfected) and inject 0.5 ml (5 × 107 cells/ml) mixed target cells i.v. into mice. 10. After 24 hr, prepare sterile 1.8-ml microcentrifuge tubes with a mixture of 200 µl lysis solution and 10 µl anticoagulant. Collect 200 µl blood into each tube. 11. Mix and incubate 30 min at room temperature. 12. Wash two times with PBS without FCS by centrifuging 5 min at 300 × g, 4°C. 13. Resuspend pellet in 200 µl Cellfix and transfer to 5-ml round-bottom tubes; protect the tubes from direct light. 14. Determine the fluorescence signal of the target cells by flow cytometry. GENERATION OF DENDRITIC CELLS (DCs) DC populations can be generated from precursors obtained from healthy donors. Two types of DCs, dermal DC-like cells (iDCs) and DC-like Langerhans cells (LC-mDCs), are generated by culturing adherent PBMCs 7 days with a mixture of cytokines like rhGM-CSF plus rhIL-4 or rhGM-CSF plus rhIL-4 plus rhTGF-β1 (Sallusto and Lanzavecchia, 1994; Geissmann et al., 1998). The DC (iDC and LC-mDC) populations can be harvested and evaluated at 7, 14, 21, and 28 days.
SUPPORT PROTOCOL
Studies of Cell Function
9.17.11 Current Protocols in Cytometry
Supplement 17
Materials Adherent PBMCs obtained following the protocol for isolation of PBMCs and the supporting protocol for isolation of adherent cells (Kanof et al., 1996) Complete medium for DC culture (see recipe) rhGM-CSF (Novartis). Store aliquots ≤1 year at −70°C rhIL-4 (Genzyme). Store aliquots ≤1 year at −70°C rhTGF-β1 (R & D Systems). Store aliquots ≤1 year at −70°C 25-cm2 tissue culture flasks (Corning) 37°C, 5% CO2 incubator Additional reagents and equipment for counting cells in a hemacytometer (APPENDIX 3A) 1a. To generate dermal DC-like cells (iDCs): In 25-cm2 tissue culture flasks, culture adherent PBMCs derived from a normal healthy donor with complete medium supplemented with 250 ng/ml rhGM-CSF and 100 ng/ml rhIL-4 in a 37°C, 5% CO2 incubator. 1b. To generate DC-like Langerhans cells (LC-mDCs): In 25-cm2 tissue culture flasks, culture adherent PBMCs derived from a normal healthy donor with complete medium supplemented with 250 ng/ml rhGM-CSF, 100 ng/ml rhIL-4, and 10 ng/ml rhTGF-β1 in a 37°C, 5% CO2 incubator. 2. Harvest and evaluate cultures at 7, 14, 21, and 28 days. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Complete medium for DC culture RPMI-1640 with 25 mM HEPES buffer and L-glutamine (Life Technologies), supplemented with 10% (v/v) heat-inactivated FBS (APPENDIX 2A), 1 ml antibiotic/antimycotic (Life Technologies)/500 ml medium. Sterilize by filtration through a 0.22-µm filter. Store ≤2 months at 4°C. FACS buffer PBS (APPENDIX 2A) supplemented with 5% (v/v) heat-inactivated FCS and 0.01% (w/v) sodium azide. Store ≤1 month at 4°C. COMMENTARY Background Information
Assessment of Surface Markers and Functionality of Dendritic Cells (DCs)
Dendritic cells (DCs) are a complex group of mainly bone marrow–derived cells that play an important role in the afferent branch of the immune response (Steinman, 1991). However, DCs represent only a minute subpopulation of the peripheral blood mononuclear cells (PBMCs), as well as of bulk cellular populations of the lung, intestine, genitourinary tissue, and lymphoid tissue. DCs have also been found in the epidermis, dermis, and mucous membranes and constitute ∼2% of the total cellular population of the human epidermis (Katz et al., 1985; Cella et al., 1997). The so-called Langerhans cells (LCs) are skin-derived DCs that
migrate to the regional lymphoid organs where, after antigen takeup, they undergo an activation/maturation step. Thereafter, LCs interact with and activate T cells. Because of their significant capability to take up, process, and present soluble antigens to responder cells in the lymphoid tissues in the context of the restricted MHC pathway, LCs have been considered one of the most important elements in the afferent arm of the immune response (Katz et al., 1985; Steinman, 1991; Cella et al., 1997; Banchereau and Steinman, 1998). Recently, successful efforts to generate DCs from freshly isolated monocytes derived from PBMCs or from CD34 blood precursors using
9.17.12 Supplement 17
Current Protocols in Cytometry
GM-CSF and IL-4, as well as GM-CSF and/or TNF, have produced differentiated DCs, but with only a limited life span in culture. However, until recently, most of the in vitro experiments involving LCs have been conducted either with LC-containing epidermal cell suspensions or with suspensions with varying degrees of LC enrichment, but not with pure monoclonal cell populations. In many cases, the results obtained may have been influenced, to some extent, by the presence of contaminating cells such as keratinocytes and melanocytes in epidermal suspensions, or lymphocytes and possibly other type of cells in PBMC-derived DCs (Hock et al., 1994; Sallusto and Lanzavecchia, 1994; Jiang et al., 1995; Bender et al., 1996). Therefore, an approach has been developed to generate LCs from isolated monocytes. This unit describes the approach for generating DCs and LCs from blood monocytes (Geissmann et al., 1998). The characterization of surface markers on human DCs has been a very difficult and elusive task because of the lack of appropriate reagents with high specificity for DC identification (Banchereau and Steinman, 1998). However, some molecules, whose genes recently have been cloned and sequenced (e.g., CD83, DEC205), have been found to be strongly associated with DCs (Jiang et al., 1995; Bender et al., 1996). Additionally, a panel of monoclonal antibodies (e.g., CMRF-44) that recognize molecules on DCs has been raised (Hock et al., 1994). Therefore, there is a growing need for grouping and establishing a common and comprehensive nomenclature for such DC-associated molecules, as well as clarifying and defining the lineage(s) of DCs and the existing DC subsets by using a panel of mAbs (Hart et al., 2001; Nuñez and Filgueira, 2001). These developments have prompted flow cytometric approaches to evaluate the reactivity of mAbs against DC populations.
Critical Parameters and Troubleshooting The viability of the cells, the freshness of the reagents, and the strict follow-up of the protocols are critical elements in obtaining satisfactory results. It is mandatory to keep all the fluorescent reagents protected from light. In addition, during and after staining, keep the samples (tubes) covered with aluminum foil until immediately before cytometric measurements.
Anticipated Results The protocols for immunophenotyping allow determination of (1) the existence of subsets within the iDC and LC-mDC populations; (2) the kinetics of antigen expression on DCs at diverse intervals of time; and (3) the identification of specific markers for DC subsets. The protocol for flow cytometric determination of antigen uptake provides an approach for clear determination of the DC capability to incorporate and process soluble antigens. The simultaneous evaluation on iDCs and LC-mDCs of forward scatter and mAb reactivity at four time points demonstrates the presence of significant variability within mAb reactivities of the iDC and LC-mDC populations (Fig. 9.17.2). At day 7, several mAbs displayed strong reactivity and grouped together (Fig. 9.17.2C, E). In addition, there was another group of mAbs with an intermediate level of reactivity (Fig. 9.17.2D, F). However, some mAbs showed no reactivity at day 7 (Fig. 9.17.2A) compared to the negative control (Fig. 9.17.2B). Variations in the reactivity of the mAbs at diverse time points were found. Moreover, at day 7 mAb HP-F1 was negative for iDCs and positive for LC-mDCs, thus discriminating iDCs from LC-mDCs. Within the panel analyzed at day 7, there are mAbs that show reactivity only against iDCs with high FS (mAb TPD153) or against a fraction of iDCs (mAbs DC-LAMP and 55K-2). These results strongly suggest the existence of subsets within the iDC and LC-mDC populations (Nuñez and Filgueira, 2001; Nuñez et al., 2001). A single-color assay allows the presentation of sample histograms overlaid with those of the isotype control. The overlay representation has the advantage of providing a visual assessment of the degree of similarity or difference between the marker on the DCs and the isotype control (Nuñez and Filgueira, 2001). Further immunophenotyping assays on human DCs and immortalized DCs demonstrated the presence of significant diversity within the TNFα-treated and IFNγ-treated populations compared with the reactivity of untreated HDCL1. Thus, the results suggest the existence of diverse stages in the maturation process of DCs passing from immature (untreated) to mature (TNFα-treated) and to differentiation towards apoptosis (IFNγ-treated; Nuñez et al., 2001). Additional studies aiming to perform further characterization of surface antigens on DCs have been conducted by diverse approaches, including multicolor flow cytometry. The panel
Studies of Cell Function
9.17.13 Current Protocols in Cytometry
Supplement 17
Assessment of Surface Markers and Functionality of Dendritic Cells (DCs)
of monoclonal antibodies used for multicolor immunophenotyping were PE, PerCP, APC, or FITC conjugated. The negative controls for the analysis were isotype-matched FITC-, PE-, PerCP-, or APC-labeled unrelated antibodies. The controls provided markers for the plots. The assays were performed after appropriate color compensations were done for each marker (Nuñez and Filgueira, 2001). The protocol for CFSE-labeled DCs permitted tracking of DC populations that have undergone a different number of cell divisions, shown in Figure 9.17.4. The protocol for DiOC18(3) and PI staining is effective for simultaneously visualizing the living and dead populations of target cells after mixing with effector cells generated in mice primed with DCs. A significant increase in dead (PI-stained) and dying [PI- and DIOC18(3)stained] target cells after exposure to effector cells indicates a clear susceptibility of the target cells. In contrast, if the proportion of dead and dying target cells is not significant, the DCs have neither primed nor influenced the immune response of the host (Fig. 9.17.5). Thus, the study of Nuñez et al. (2001) demonstrated that DiOC18(3) when used in combination with PI is effective for simultaneously visualizing both the living and dead populations of target cells before and after priming the host with DCs. Protocols for flow cytometric assessment of in vitro and in vivo CTL activity by T cells triggered after priming with DCs are also described. The assay for in vitro CTL uses a two-color cytometric approach with DiOC18(3) as a membrane stain for the target cells, and PI as a nuclear stain of damaged or dead target cells. DiOC18(3) stains green and is detected at 530 nm (FL1). PI stains red and is detected at 585 nm (FL2) and >650 nm (FL3). The DiOC18(3)-stained live target cells are located in the lower right quadrant. The dead targets are located in the upper left quadrant. The dying cells are located in the upper right quadrant. Thus, this flow cytometric approach was used for assaying the CTL activity generated in mice immunized with DC transduced with amplicon β gal. It was found that there was a negligible (<15% of antigen-specific lysis) CTL activity after immunization of mice with DCs transduced with amplicon(s) compared to the mock mice and between the loaded (peptide +) and unloaded (peptide −) targets (Nuñez, 2001). Furthermore, an additional flow cytometric approach was also used for assaying the in vivo CTL activity generated in mice immunized
with amplicon-transduced DCs. The amplicons carried the gene for the rabies glycoprotein. The DC-primed mice were boosted with a commercial vaccine 4 days prior to the infusion of a mixture of 105 target cells. Two groups of target cells were used and labeled with DiOC18(3). Moreover, uninfected targets were labeled at low concentrations while target cells infected with the rabies virus were labeled at high concentrations. After 24 hr, mice were bled and cells were evaluated by flow cytometry. It was expected that the mice primed by DCs transduced with the amplicon rabies could generate a CTL activity that destroys the targets infected with the rabies virus while the uninfected targets could remain. It was found that the three mice immunized with DCs cleared out most of the infected targets while the control did not.
Time Considerations The staining procedures are quite simple and do not require extensive incubation times. Most of the incubations require <30 min. However, washing procedures should be enforced in order to keep very low background during the staining. The staining method with DiOC18(3) and PI has the advantage of being rapid (<30 min) and the cells do not require extra processing prior to staining.
Literature Cited Banchereau, J. and Steinman, R.M. 1998. Dendritic cells and the control of immunity. Nature 392:245-252. Bender, A., Sapp, M., Schuler, G., Steinman, R.M., and Bhardwaj, N. 1996. Improved methods for the generation of dendritic cells from nonproliferating progenitors in human blood. J. Immunol. Methods 196:121-135. Cella, M., Sallusto, F., and Lanzavecchia, A. 1997. Origin, maturation and antigen presenting function of dendritic cells. Curr. Opin. Immunol. 9:10-16. Geissmann, F., Prost, C., Monnet, J.-P., Dy, M., Brousse, N., and Hermine, O. 1998. Transforming growth factor beta1, in the presence of granulocyte/macrophage colony-stimulating factor and interleukin 4, induces differentiation of human peripheral blood monocytes into dendritic Langerhans cells. J. Exp. Med. 187:961-966. Hart, D.N.J., Clark, G.J., MacDonald, K., Kato, M., Vuckovic, S., Lopez, A., Wykes, M., and Munster, D. 2001. 7th Leucocyte Differentiation Antigen Workshop, DC section summary (D. Mason, ed.). Leukocyte Typing VII, Oxford University Press, Oxford. In press.
9.17.14 Supplement 17
Current Protocols in Cytometry
Hock, B.D, Sterling, G.C, Daniel, P.B., and Hart, D.N. 1994. Characterization of CMRF-44, a novel monoclonal antibody to an activation antigen expressed by allostimulatory cells within peripheral blood, including dendritic cells. Immunology 83:573-581. Jiang, W., Swiggard, W.J., Heufler, C., Peng, M., Mirza, A., Steinman, R.M., and Nussenzweig, M.C. 1995. The receptor DEC-205 expressed by dendritic cells and thymic epithelial cells is involved in antigen presentation. Nature 375:151155. Kamau, S., Hurtado, M., Müller-Doblies, U., Grimm, F., and Nuñez, R. 2000. Flow cytometric assessment of allopurinol susceptibility in Leishmania infantum promastigotes. Cytometry 40:353-360. Kanof, M.E., Smith, P.D., and Zola, H. 1996. Preparation of human mononuclear cell populations and subpopulations. In Current Protocols in Immunology (J. Coligan, A. Kruisbeek, D. Margulies, E. Shevach, and W. Strober, eds.) pp. 7.1.1-7.1.7. John Wiley & Sons, New York. Katz, S.I., Cooper, K.D., Iijima, M., and Tsuchida, T. 1985. The role of Langerhans cells in antigen presentation. J. Invest. Dermatol. 85:96-98. Kruisbeek, A.M. 2000. Isolation and fractionation of mononuclear cell populations. In Current Protocols in Immunology (J.E. Coligan, A.M. Kruisbeek, D.H. Margulies, E.M. Shevach, and W. Strober, eds.) pp. 3.1.2-3.1.5. John Wiley & Sons, New York.
Nuñez, R., Filgueira, L., and Nuñez, C. 2001. Flow cytometric assessment on human dendritic cell lines (HDCL) of monoclonal antibody (mAb) reactivities and determination of surface antigen expression by cytokine modulation. In Leukocyte Typing VII (D. Mason, ed.). Oxford University Press. Oxford. In press. Sallusto, F. and Lanzavecchia, A. 1994. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor alpha. J. Exp. Med. 179:1109-1118. Steinman, R.M. 1991. The dendritic cell system and its role in immunogenicity. Annu. Rev. Immunol. 9:271-296.
Key References Geissmann et al., 1998. See above. Initial reference for generating LCs from monocytes. Hart et al., 2001. See above. Describes the new CD markers assigned for dendritic cells, as well as an additional set of markers identified in dendritic cells that did not fulfill the requirements for assignment of a CD. Nuñez, 2001. See above. Contains a chapter on flow cytometric evaluation of DCs. Nuñez and Filgueira, 2001. See above.
Nuñez R. 2001. Flow Cytometry for Research Scientists: Principles and Applications. Horizon Scientific Press, Norfolk, England.
Displays the mAbs used in the DC workshop and the procedures for generating monocyte-derived DCs and LCs.
Nuñez, R. and Filgueira, L. 2001. Flow cytometry assessment of monoclonal antibody (mAb) reactivities against dendritic cells (DC). In Leukocyte Typing VII (D. Mason, ed.). Oxford University Press, Oxford. In press.
Contributed by Rafael Nuñez Memorial Sloan-Kettering Cancer Center New York, New York
Nuñez, R., Sanchez, M., Filgueira, L., Wild, P., and Nuñez, C. 1998. Characterization of two human dendritic cell lines that express CD1a, take-up, process and present soluble antigens and induce MLR. Immunol. Lett. 61:33-43.
Studies of Cell Function
9.17.15 Current Protocols in Cytometry
Supplement 17
Stem Cell Identification and Sorting Using the Hoechst 33342 Side Population (SP)
UNIT 9.18
This protocol describes the use of the fluorescent DNA-binding dye Hoechst 33342 to identify and purify murine hematopoietic stem cells, the so-called side population (SP). The purification is based on the differential efflux of Hoechst dye relative to other bone marrow cells (Goodell et al., 1996) and was originally established for murine hematopoietic stem cells (HSCs) on normal C57Bl/6 bone marrow (NBM). Therefore, the Basic Protocol is specific for murine bone marrow cells, although conditions for additional tissues and species are noted (Goodell et al., 1997; Jackson et al., 1999). However, the author strongly suggests that initial experiments be performed using murine bone marrow exactly as described in order to establish the procedure in one’s laboratory and to definitively identify the side population (SP) on the flow cytometer. To assist this effort, the Support Protocol outlines a procedure for extraction of murine bone marrow. STAINING AND ANALYSIS OF STEM CELL SIDE POPULATION IN MURINE BONE MARROW
BASIC PROTOCOL
The ability to discriminate Hoechst-SP cells is based on the differential efflux of Hoechst 33342 by a multi-drug-like transporter (see Commentary). This is an active biological process. Therefore, optimal resolution of the profile is obtained with great attention to the staining conditions. The Hoechst concentration, cell concentration, staining time, and staining temperature are all critical. Likewise, when the staining process is complete, the cells should be maintained at 4°C in order to inhibit further dye efflux. Samples for staining and sorting must be absolutely fresh and the protocol must be carried out without interruption. If one adheres rigorously to the protocol, SP cells should be easy to find. Optimal Hoechst-staining protocols vary only slightly for multiple species. Staining for 90 min is optimal for mouse SP cells, whether from bone marrow or muscle, whereas 120 min is optimal for human, rhesus, and swine cells. In other respects, the protocol is exactly as described. The first time cells other than mouse bone marrow are stained, it is advisable to check the staining conditions by staining aliquots of cells for 60, 90, and 120 min, using the same Hoechst concentration as for mouse. Materials DMEM+ (see recipe) Mouse bone marrow cells prepared according to Support Protocol, or bone marrow cells from another species, or cells from non-bone marrow tissue 1 mg/ml Hoechst 33342 (see recipe) Hanks buffered salt solution+ (HBSS+, see recipe), on ice 200 µg/ml propidium iodide in PBS (see recipe; optional) 5 mM verapamil in 95% ethanol (optional; Sigma) Circulating water bath at exactly 37°C 250-ml or 50-ml polypropylene centrifuge tubes (Corning) Flow cytometer with UV laser (usually high-power argon tuned to 351- to 364-nm excitation) and standard 488-nm laser 450/20 band-pass (BP) filter 675-nm edge filter long-pass (EFLP; Omega Optical) 610-nm dichroic mirror short-pass (DMSP) filter Additional reagents and equipment for counting cells (APPENDIX 3A) Studies of Cell Function Contributed by Margaret A. Goodell Current Protocols in Cytometry (2002) 9.18.1-9.18.11 Copyright © 2002 by John Wiley & Sons, Inc.
9.18.1 Supplement 19
Prepare cells 1. Check the circulating water bath with a thermometer to ensure that the temperature is precisely 37°C. Prewarm DMEM+ while preparing the bone marrow (for mouse bone marrow, see Support Protocol). For non-bone marrow tissues, extract cells using proteases such as trypsin and collagen. Count and resuspend at 106 cells/ml.
2. Count the nucleated cells as carefully and accurately as possible, excluding red blood cells (RBCs; APPENDIX 3A). The RBCs are saucer shaped and smaller than the rest of the cells. Under the microscope they sometimes appear to have a black dot in the middle, which is related to the central depression. They can comprise up to 20% of the marrow cells. If they cannot be comfortably distinguished or excluded by eye, lyse the RBCs using one of many published protocols. Under the microscope the marrow should appear as a nice single-cell suspension with no clumps. Marrow from 2 femurs and 2 tibias of a C57Bl/6 mouse yields an average of 5 × 107 nucleated cells. This number varies from strain to strain. If necessary, magnetic enrichments may be performed prior to Hoechst staining.
3. Centrifuge bone marrow 5 min at 500 × g, 4°C. Resuspend at 106 cells/ml in prewarmed DMEM+ and mix well by gently inverting the 250-ml polypropylene centrifuge tube. It is most convenient to stain large volumes of marrow in 250-ml polypropylene tubes. Alternatively, 50-ml tubes can be used. The tubes must be polypropylene, as many cells stick to polystyrene. Human bone marrow samples are sometimes received late in the day. If preparation and sorting cannot be performed that same day, the author recommends that the marrow be kept overnight at 4°C prior to Hoechst staining. Such storage should only be undertaken when it is unavoidable and should never be a common practice. Cell viability is best when non-ficoll-separated bone marrow is kept at 4°C. (Note that mouse marrow does not need to be separated on a ficoll gradient.) In the morning, ficoll separated or nonseparated bone marrow may be warmed to 37°C, resuspended at 106 cells/ml, and stained with Hoechst as described. Plating the bone marrow on tissue-culture plastic and leaving it in the incubator overnight is not recommended, as some stem cells may be lost on the plastic. Optional: To confirm that the correct cells are identified on the flow cytometer, reserve an aliquot of cells and block the efflux of Hoechst in the side population of this aliquot by adding 50 ìM verapamil to these cells before incubation in the Hoechst staining solution. Confirm the absence of cells in the SP gating region of this aliquot. Verapamil and other MDR inhibitors decrease the speed of dye extrusion.
Stain with Hoechst 33342 4. Add 1 mg/ml Hoechst 33342 to a final concentration of 5 µg/ml (a 200-fold dilution of the stock), cap the tube, and mix by gentle inversion. 5. Incubate tubes exactly 90 min in the 37°C water bath. Make sure that the water level is sufficiently high to ensure that cell temperature is maintained at 37°C. Mix tubes several times during the incubation. Because of the sensitivity of the staining to temperature, it is preferable to have a dedicated water bath. Do not use a water bath that is constantly fluctuating in temperature due to heavy use.
Stem Cell Identification and Sorting Using Hoechst 33342 Side Population
6. After 90 min, centrifuge cells 5 min at 500 × g, 4°C and resuspend in ice-cold HBSS+. If the samples are not going to be further stained with antibodies (see below), the cold HBSS+ should contain 2 µg/ml propidium iodide (PI) for dead cell discrimination. All further manipulations must be performed at 4°C to inhibit efflux of Hoechst dye from the cells.
9.18.2 Supplement 19
Current Protocols in Cytometry
Addition of PI is not required to see the SP cells, but will help. Hoechst is somewhat toxic to the bone marrow, and the PI will allow exclusion of dead cells from the profile (see Fig. 9.18.1).
7. At this point, samples may be run directly on the cytometer or further stained with antibodies to confirm the identity of the population, followed by resuspension of cells in cold HBSS+ containing 2 µg/ml PI. Magnetic enrichments may also be employed at this stage if the entire procedure is carried out at 4°C. The mouse SP population seems fairly homogeneous with respect to cell surface markers. Of SP cells, ∼85% will be Sca-1+, c-Kit+, CD45+, and lineage marker negative. The author recommends staining with two antibodies, one that positively stains most SP cells (Sca-1 or c-Kit) and one that does not stain SP cells but stains a large fraction of the bone marrow (e.g., Gr-1 or B220, all available from PharMingen). A suggested combination is Sca-1-PE and Gr-1-FITC. Figure 9.18.2 shows typical staining of whole marrow and SP cells with these markers. Human SP cells can be stained with CD34 and some other marker.
Set up flow cytometer 8. Excite Hoechst 33342 at 351 to 364 nm and collect blue fluorescence with a 450/20 band-pass (BP) filter and red fluorescence with a 675-nm edge filter long-pass (EFLP). Use a 610-nm dichroic mirror short-pass (DMSP) to separate the emission wavelengths. Propidium iodide (PI) fluorescence excited at 351 to 364 nm is also measured through the 675 EFLP; the PI red signal will be much brighter than Hoechst (see Fig. 9.18.1). Hoechst blue is the standard analysis wavelength for Hoechst 33342 DNA content analysis. Although other filter sets/combinations can work sufficiently well, the author has found these to give the best results.
Run on the flow cytometer 9. Place Hoechst-stained cells on the cytometer. If possible, keep cold by the use of a chilling apparatus. It is not necessary to establish live gates on forward versus side scatter parameters. First, display the histogram of Hoechst red (x-axis) versus blue (y-axis) fluorescence. With the detectors in linear mode, adjust the voltages so that the red blood cells are seen in the lower left corner and the dead cells (stained with PI and very bright) line up on a vertical line to the far right (Fig. 9.18.1). The bulk of the rest of the cells can be centered. It should be possible to identify a major G0-G1 population with S-G2/M cells going off to the upper right corner. 10. Once a profile similar to that shown in Figure 9.18.1 can be seen, draw a live gate to exclude the red and dead cells. Then, collect 50,000 to 100,000 events within this live gate in order to identify the SP region definitively. The SP region should appear as shown in Figure 9.18.1. The prevalence is low, ∼0.05% of whole bone marrow in the mouse and even lower in human samples (0.01% to 0.03% of ficolled marrow).
11. Once the SP region has been correctly identified, confirmed by antibody staining and/or comparison to cells treated with verapamil (see step 3, annotation), cells may be sorted using standard procedures and precautions, including sterilization of tubing and sheath fluid if the sorted cells are to be cultured or transplanted.
Studies of Cell Function
9.18.3 Current Protocols in Cytometry
Supplement 19
10000
B.
1000
4000
3000
SS
Hoechst blue
A.
100
2000
1000
10
0
1 1
10
100
1000
0
10000
1000
Hoechst red
Hoechst blue
D.
10000
1000
red blood cells and debris
100
4000
3000
4000
2000
0
1 1
10
100
1000
0
10000
1000
F.
10000
1000
2000
FS
PI positives: dead cells 4000
3000 SP ~0.03%
100
SS
Hoechst blue
3000
4000
1000
Hoechst red
E.
4000
3000
10
SP
3000
FS
SS
C. S-G2/M G0-G1
2000
2000
1000
10
0
1 1
10
100
1000
Hoechst red
10000
0
1000
2000
FS
Figure 9.18.1 Hoechst 33342 staining of whole murine bone marrow. Whole bone marrow from C57Bl/6 mice is shown stained with Hoechst alone (A,B) and Hoechst with propidium iodide (PI; C,D,E,F). In A and B, Hoechst staining reveals multiple populations on both Hoechst fluorescence (A) and forward (FS) versus side scatter (SS; B). With PI staining (C,D,E,F), a line of brightly stained PI-positive cells appears against the far right side of the Hoechst profile and the other populations are somewhat simplified. Note there are two main G0-G1 populations (C), each with an associated S-G2/M tail. The SP is also visible, and red cells and debris collect in the lower left corner. E and F are gated for live nucleated cells as shown on the Hoechst plot (E). The 100,000 events collected in the live gate shown allow the SP to be better defined and clean up the FS/SS view. When cells within this live gate are displayed on a plot of FS versus SS (F), the profile is substantially cleaner. The SP as indicated represents ∼0.03% of the live-gated cells in bone marrow.
Stem Cell Identification and Sorting Using Hoechst 33342 Side Population
9.18.4 Supplement 19
Current Protocols in Cytometry
0.94
0.091
B.
1000
1000
100
100
10
1
1.49
97.5 1
C.
10
100
1000
4000
1
3.72
43.5 1
D.
0.62
10
10000
3000
10
100 1000 Sca-1-PE
10000 0.52
10000
1.03
1000
2000
100 0.03 Gr-1-FITC
UV 405/30
10000 52.1
Gr-1-FITC
A. 10000
1000
0
10
1 0
1000
2000 3000 UV 670/40
4000
17 1
81.4 10
100 1000 SCA-1-PE
10000
Figure 9.18.2 Confirmation of the SP phenotype in murine (C57Bl/6) bone marrow. Whole murine bone marrow is shown unstained (A) and stained with an anti-granulocyte antibody, Gr-1, and an antibody against the stem cell marker, Sca-1 (B,D). Whole bone marrow has a low percentage of Sca-1+Gr-1− cells (∼3.7% in this experiment, B). When the SP cells are gated (C) and displayed with these same markers, ∼80% of the SP cells are Sca-1+ and Gr-1−. This demonstrates the relatively high uniformity of this population and also confirms the presence of the expected stem cell phenotype.
PREPARATION OF MOUSE BONE MARROW The author highly recommends using mouse bone marrow for the first attempt at Hoechst staining. This short protocol describes bone marrow cell extraction.
SUPPORT PROTOCOL
Materials Mouse (the author uses C57Bl/6, but other strains will work) HBSS+ (see recipe) 70% ethanol in spray bottle Rugged and delicate scissors Small forceps 10-cm petri dishes 10-ml syringes 18- and 27-G needles 50-ml tubes 70-µm filter or mesh 1. Sacrifice the mouse according to accepted institutional protocol, lay the body on its back, and spray the abdomen with 70% ethanol to sterilize. 2. Make a horizontal (leg-to-leg) incision with the rugged scissors through the skin of the abdomen a little lower than the level of the hips.
Studies of Cell Function
9.18.5 Current Protocols in Cytometry
Supplement 19
3. From the incision, pull the skin up and down simultaneously—the skin should come off over the legs. Grab the knees with the forceps and pull out of the skin until the legs are exposed completely. 4. Remove the tibias by cutting through the knees with delicate scissors. Take as much muscle off the bone as possible. Use the tendons where they connect at the ankle to remove all the muscle in one motion and take the foot off at the same time. Get the bone as clean as possible. Place the tibias in a 10-cm petri dish containing ∼5 ml HBSS+, kept on ice. 5. Clean the muscle off the femur, using the tendons at the knee while the bone is still attached to the mouse. Cut the femurs off the mouse at the hips. This is where the majority of marrow is, so try to get most of the femur. If muscle is still attached to the femur, try to cut as much off as possible before putting it into the petri dish with the other bones. The muscle is removed to keep the marrow from sticking to it and to make the bone clearly visible.
6. Having 4 bones in the petri dish (from one mouse), take a 10-ml syringe with a 27-G needle and fill with HBSS+. With forceps, hold a bone vertically over a fresh petri dish, insert the needle into the end of the bone, and expel liquid through the bone to push marrow onto the plate. If the opening is too small for the needle, trim the end of the bone slightly. In both femurs, move the needle around a little to ensure expulsion of as much of the marrow as possible. Also, flip the bone upside down and expel medium through the other end as well. If 10 ml is not sufficient to clean all 4 bones, put a little more medium into syringe to finish. Bones will become white when the marrow is expelled.
7. Replace the 27-G needle with an 18-G needle. Aspirate the bone marrow up and down in the petri dish four or five times to break the chunks of marrow into single cells. Avoid putting air in the syringe, as air bubbles kill cells. Finally, expel the marrow into a 50-ml tube. If desired, filter through a 70-µm filter or mesh at this point to make sure that no bone or other chunks remain. Count the nucleated cells, as described in the Basic Protocol. An average yield from an 8-week-old C57Bl/6 mouse is ∼5 × 107 nucleated cells.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
DMEM+ Dulbecco’s modified Eagle’s medium (Life Technologies) containing: 2% (v/v) fetal calf serum 10 mM HEPES (Life Technologies) Store up to 4 weeks, 4°C
Stem Cell Identification and Sorting Using Hoechst 33342 Side Population
HBSS+ Hanks’ balanced salt solution (Life Technologies) containing: 2% (v/v) fetal calf serum 10 mM HEPES Store up to 4 weeks, 4°C
9.18.6 Supplement 19
Current Protocols in Cytometry
Hoechst 33342 (also known as bis-benzimide) Resuspend at 1 mg/ml Hoechst 33342 in water, filter sterilize, and freeze at −20°C in 1-ml aliquots. Can be obtained from Sigma. Since Hoechst is not costly, there is no reason to reuse old dye.
Propidium iodide Stock solution: 10 mg/ml propidium iodide (Sigma) in water; freeze up to 12 months at −20°C. Working solution: dilute stock 1:50 in PBS (APPENDIX 2A) to 200 µg/ml; cover with aluminum foil and store up to 2 months at 4°C. Final concentration in samples should be 2 µg/ml. COMMENTARY Background Information The Hoechst-SP cell purification is one of several strategies for stem cell purification. Some of these are based solely on cell surface markers (Spangrude et al., 1988; Uchida and Weissman, 1992; Jordan et al., 1995); others use some combination of surface markers and other properties such as low fluorescent dye uptake (e.g., rhodamine 123 or Hoechst 33342; Pallavicini et al., 1985; Spangrude and Johnson, 1990; Wolf et al., 1993; Goodell et al., 1996). Previous studies using Hoechst 33342 for stem cell enrichment viewed Hoechst fluorescence at only one wavelength (blue; Pallavicini et al., 1985). Interestingly, this group sorted the lowest G0 population, which undoubtedly included the SP population. Without the use of two wavelengths as in this protocol, however, neither the multiple populations nor the SP was clearly resolved. The DNA-binding Hoechst 33342 dye is normally used to identify cells in G0-G1 versus S-G2/M. The author fortuitously observed the complex staining populations shown in Figure 9.18.1. A literature search found a similar observation of multiple Hoechst populations in the thymus using different emission wavelengths, discussed below (Watson et al., 1985). By dissecting the multiple populations and transplanting them into animals, the author discovered that only the SP population had stem cell activity (Goodell et al., 1996). This population was also strikingly homogeneous with respect to expression of other cell surface markers previously identified by others on stem cells (Fig. 9.18.2). The author also showed that the population exhibited low Hoechst 33342 fluorescence due to active efflux of the dye, likely by a multi-drug resistance (MDR)-like mechanism. There was a direct correlation between dye efflux and stem cell capacity, such that the cells at the lowest tip of the SP (with the highest
dye efflux and least amount of dye) exhibited the highest stem cell activity over the longest period of time (Goodell et al., 1997). The main advantage of the SP purification for stem cells is its potential application to stem cells of other species (Goodell et al., 1997; Storms et al., 2000) or tissues (Gussoni et al., 1999; Jackson et al., 1999). But to date, in vivo repopulating activity, the gold standard for defining stem cell activity, has been shown only for murine bone marrow (Goodell et al., 1996) and muscle-derived cells (Gussoni et al., 1999). Compared to other surface marker-based strategies for purification of murine HSCs, it may be simpler to identify the SP than other more subtly defined populations (Morrison and Weissman, 1994), and Hoechst dye is considerably less costly than antibodies (although, the best SP stem cell purifications are obtained using the dye in conjunction with at least one antibody, such as Sca-1, to exclude nonhematopoietic stem cells from the SP as shown in Fig. 9.18.2). The main disadvantage of the SP method is the high sensitivity to staining conditions and the consequent variability in staining that some researchers experience. In addition, the prevalence of the population in human bone marrow is low (0.01% to 0.03%), thus yields can be frustratingly small. Also, the flow cytometry equipment needed (UV laser, preferably with an additional laser) is not ubiquitous and the cytometer setup is somewhat unusual, so some effort can be incurred in simply establishing the procedure in laboratories that do not perform it routinely. Finally, since Hoechst is a DNAbinding dye, there is some toxicity associated with its use, particularly at high dye concentrations. This toxicity is associated with its DNA binding capacity and ability to inhibit DNA topoisomerase I, thereby causing DNA strand breaks (Chen et al., 1993). The author has not
Studies of Cell Function
9.18.7 Current Protocols in Cytometry
Supplement 19
seen toxicity associated with cells in the SP population, likely due to efficient efflux of the dye, but this has never been thoroughly explored.
Stem Cell Identification and Sorting Using Hoechst 33342 Side Population
Mechanisms of Hoechst staining The author believes that the SP is resolved in this method because of the following processes. The Hoechst 33342 molecule diffuses through cellular membranes at a limited rate. This process starts as soon as the whole sample is placed in Hoechst-containing medium. As time goes by, the amount of dye bound to DNA inside each nucleus increases and the concentration of free dye in the medium decreases, until an equilibrium is reached between diffusion into the cells and diffusion/transport out of the cells. Consequently, this equilibrium is affected by the concentration of both dye and cells; lowering the dye concentration or increasing the cell concentration in the medium proportionally reduces the number of dye molecules available to incorporate into each cell and can potentially prevent visualization of the SP cells. Meanwhile, as the dye diffuses into the cells and its concentration drops in the medium, some of it is being actively pumped out of the cells by transmembrane glycoproteins with MDR-like activity. Probably, most cells will have some members of the multi-drug resistance family in their membranes, but stem cells may possess either different members, or higher amounts or higher activity of those constituents. The SP cells, thus, stand out as a discrete population when the speed of the active efflux of the dye is greater than the speed of diffusion of Hoechst into the cells. In other words, the SP is able to extrude the dye against a concentration gradient, when compared to the rest of the bone marrow cells. Incubating the cells with MDR inhibitors decreases the speed of dye extrusion, as shown by blocking efflux with verapamil or other inhibitors (Goodell et al., 1996). Likewise, increasing the concentration of the dye or decreasing the concentration of the cells can eliminate the SP. When dye efflux is inhibited in stem cells, dye content reflects DNA content, and the dye fluorescence can be used to separate cells on the basis of cell cycle status (Goodell et al., 1996). There are many members of the MDR family, but to date, those involved in Hoechst efflux in stem cells have not been conclusively identified. Mice in which both the MDR1a and MDR1b genes have been knocked out have nonexistent rhodamine 123 efflux (Schinkel et
al., 1997) but normal Hoechst 33342 efflux (Goodell, unpub. observ.). Therefore, there must be additional transporter(s) involved. One good candidate is the MDR-related protein called BCRP, as reported recently by Zhou et al. (2001). If this transporter is confirmed to be expressed exclusively on stem cells, it could allow for purification of SP cells based on antibodies against the transporter. What is normally effluxed by the Hoechst dye transporter? The author has speculated that it is something that otherwise causes the cells to differentiate, but this remains to be elucidated. The complex Hoechst emission pattern As can be seen from Figure 9.18.1, the dye emission pattern in murine bone marrow is complex, with multiple populations being revealed by dual-wavelength analysis. One can see the SP population and multiple G0-G1 and S-G2/M populations (Fig. 9.18.1). What accounts for this complex pattern? Red Hoechst fluorescence. The peak fluorescence emission of Hoechst is ∼450 nm (blue). It is very unusual for a fluorochrome to have significant fluorescence >200 nm higher than the peak, such as that seen ∼650 nm (red). Many have asked why the red Hoechst signal is even seen; no flow-cytometry textbook documents Hoechst emission out this far. The author does not think that this represents a separate emission peak for Hoechst fluorescence, but rather the fact that Hoechst stains cells very brightly. Therefore, significant fluorescence is still detected this far out because the overall quantity of signal is so great. However, although a signal is easily detectable, it is not very bright in the red wavelengths, relative to the blue. For this reason it helps to have a red-sensitive PMT for Hoechst-red detection. Note that the red signal is not propidium iodide. PI-positive cells are even brighter than these Hoechst-red cells and can be seen lining up at the far right of the profile (see Fig. 9.18.1). Of course, if a 488-nm laser is also running, these PI-positive cells will appear in the PE and PI channels until they are gated out on the basis of the Hoechst profile. Multiple G0-G1 peaks/populations. Multiple Hoechst-staining populations were first noted by Watson et al. (1985) in experiments designed to look at DNA content in chick thymocytes. Also noting a strong dye concentration-dependent and staining time–dependent effect, they observed two main populations and one minor population when looking at two Hoechst emission wavelengths, green (515 to
9.18.8 Supplement 19
Current Protocols in Cytometry
560 nm) and purple (390 to 440 nm). They proposed that the populations were due to the presence of at least two binding sites for Hoechst in DNA and different spectral properties of the dye in the two different sites. Loken (1980) also found, using nonsaturating concentrations of dye, that T and B cells could be distinguished solely on the basis of Hoechst dye staining using only one wavelength, which was undoubtedly the same phenomenon less clearly revealed. This idea of two binding sites/spectra is consistent with studies from other laboratories using both Hoechst 33342 and a closely related dye, Hoechst 33258 (which cannot traverse the plasma membrane in living cells). The Hoechst dyes bind AT base pairs in the minor groove (Sriram et al., 1992). Higher dye concentrations reveal secondary binding sites, which may be lower affinity sites or additional sites perhaps uncovered by extensive binding to primary sites that changes the conformation of the DNA/chromatin. These secondary sites confer different spectral properties to the dye (Stokke and Steen, 1986). Studies of dye binding to supercoiled versus relaxed plasmid DNA demonstrated different binding modes and emission spectra that were also affected by dye concentration (Sandhu et al., 1985). Accordingly, it was proposed that one could use the ratio of dye fluorescence at two wavelengths to probe chromatin structure (Steen and Stokke, 1986). Different cell types may have substantial differences in the availability or conformation of specific regions of chromatin to dye binding, which could contribute a great deal to the presence of the multiple populations seen in murine bone marrow. Indeed, when whole bone marrow is co-stained with antibodies against B cells (B220), the lower G0-G1/S-G2/M population is predominantly stained, while the granulocytes (visualized with the Gr-1 antibody) are predominantly found in the upper G0-G1 population. In summary, three major properties of Hoechst 33342 contribute to the ability to distinguish stem cells using this dye. First, the dye is a DNA-binding dye that can be used to measure DNA content in living cells. Second, Hoechst has at least two binding modes that result in different spectral properties, allowing multiple populations to be resolved when fluorescence emission is viewed at two wavelengths simultaneously. Finally, identification of the side population (SP) stem cells based on dye efflux mediated by a multi-drug-resistance-like transporter is possible.
Critical Parameters As discussed above, both the staining time and dye and cell concentrations are of paramount importance. Because Hoechst efflux is an active process, all further manipulation of the cells and staining must be done at 4°C to avoid further efflux of Hoechst from the cells. The author’s original studies were performed under the conditions described in the protocol, and deviations from these may give aberrant results. Watson et al. (1985) also describe a heavy time and Hoechst concentration dependence of the appearance of multiple Hoechst populations as discussed above. An edifying experiment is to titrate both staining time and Hoechst concentration for the cell population of interest in order to observe the change in populations and resolution of the SP. Other tips for optimal resolution of the multiple Hoechst populations Since analysis of the Hoechst dye is performed in linear mode, good coefficients of variation (CVs) of the flow cytometer are critical. The author performs alignments in linear mode with particles that have a very tight distribution (e.g., Flow-Check beads from Coulter). Furthermore, on BD machines, the UV laser is used in the “first” position for optimal CVs; this has the added benefit of allowing thresholding on DNA (Hoechst blue), and thus, red blood cells are irrelevant. However, this is not necessary. After the cells are put on, the laser alignment can be further optimized to tighten the populations. In keeping with having good CVs, the sample differential pressure must be as low as possible. Preferably, the maximum sample differential pressure is calibrated with the alignment particle. In other words, if the CV for the alignment particle is 3% with a low differential pressure, then determine the maximum differential pressure that will still give good percent CVs and do not ever exceed that pressure. Finally, a relatively high power on the UV laser gives the best CVs, e.g., 50 to 100 mW generally give the best Hoechst signal on jetin-air instruments. Less power will suffice, but the populations may not be as clearly resolved. Likewise, sensitive red detectors (photomultiplier tubes, or PMTs) are recommended—use the most sensitive ones available for detecting Hoechst red emission.
Troubleshooting In many years of experience with different flow cytometry laboratories, a constant theme
Studies of Cell Function
9.18.9 Current Protocols in Cytometry
Supplement 19
Stem Cell Identification and Sorting Using Hoechst 33342 Side Population
has been that if a protocol does not work, the experimenter blames the flow cytometer operator, and the operator blames the experimenter. While helping numerous researchers establish the SP sorting techniques in their laboratories, the author has found that improvements can be made on both sides. Neither the flow cytometer set up nor the staining procedure is intuitive, and both require skill and sensitivity to optimally resolve the multiple Hoechst populations, including the SP. The first suggestion for troubleshooting is to use murine bone marrow and stain according to the given specifications. Because the protocol is so sensitive to Hoechst concentration and staining time (and temperature), check all these parameters rigorously. Remake the Hoechst dye stock, check the dilution of the stock to working concentration, check the temperature of the water bath with a thermometer, ensure that the cells are resuspended at 106 cells/ml, and take care to adhere to the staining time. Using two times the dye for half the time will not substitute! Use antibody staining as a quality control, as shown in Figure 9.18.2. Carefully check the concentration of cells that are brought to the flow cytometer operator (provide the preferred concentration). If the cell concentration is too low, a higher pressure will be needed to run the cells, and resolution will be compromised. If murine bone marrow works, use a murine bone marrow control every time other tissues and species are stained for the first few experiments for a consistent good positive control—then one will know whether or not to blame the operator. If the problems appear to be instrument-related, start by trying to get optimal CVs with a high-uniformity bead product such as FlowCheck beads from Coulter. Some lasers give lower UV power (e.g., the Enterprise), which can reduce the CVs. Good laser alignment becomes paramount in these situations. Check the sensitivity of the PMTs on the laser. Some are more sensitive than others, and therefore will detect red fluorescence better. PMTs can be swapped into the appropriate position, if necessary, to improve sensitivity. Keep sample differentials low. Overall, consider that DNA content is being analyzed, as in the use of PI for cell-cycle analysis, so careful instrument set-up will yield the best results. At last resort, call the local service representative for an instrument tune-up.
Anticipated Results With good quality staining of mouse bone marrow, ∼0.05% of bone marrow cells (exclusive of the red and dead cells) should be in the SP population. Between 60% and 90% (usually ∼85%) of these should be Sca-1+, CD45+, cKit+, CD34−/low, and lineage-marker negative (Goodell et al., 1996, 1997). If only 60% of the SP cells are Sca-1+, a more conservative definition of the SP population (lower down on the tail) should result in a higher proportion of Sca-1+ cells (and therefore stem cells). When sorting stem cells, one can obtain ∼5000 cells per C57Bl/6 mouse (6- to 8-week-old).
Time Considerations Typically, the author prepares bone marrow from 10 mice with a yield of ∼50,000 stem cells after a full day’s work. On a high-speed flow cytometer (e.g., Cytomation or FacsVantage with Turbosort sorting from 25,000 to 60,000 cells/sec), one can sort bone marrow from all 10 mice in ∼5 hr with no pre-enrichment. If working with a slower instrument or if limited by time (or the cost of hourly flow time), magnetic enrichment for murine stem cells using Sca-1 or c-Kit prior to SP sorting is recommended. This offers an enrichment of ∼10-fold, and a concomitant reduction in flow cytometer time.
Literature Cited Chen, A.Y., Yu, C., Gatto, B., and Liu, L.F. 1993. DNA minor groove-binding ligands: A different class of mammalian DNA topoisomerase I inhibitors. Proc. Natl. Acad. Sci. U.S.A. 90:81318135. Goodell, M.A., Brose, K., Paradis, G., Conner, A.S., and Mulligan, R.C. 1996. Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J. Exp. Med. 183:1797-1806. Goodell, M.A., Rosenzweig, M., Kim, H., Marks, D.F., DeMaria, M., Paradis, G., Grupp, S.A., Sieff, C.A., Mulligan, R.C., and Johnson, R.P. 1997. Dye efflux studies suggest that hematopoietic stem cells expressing low or undetectable levels of CD34 antigen exist in multiple species. Nat. Med. 3:1337-1345. Gussoni, E., Soneoka, Y., Strickland, C.D., Buzney, E.A., Khan, M.K., Flint, A.F., Kunkel, L.M., and Mulligan, R.C. 1999. Dystrophin expression in the mdx mouse restored by stem cell transplantation. Nature 401:390-394. Jackson, K.A., Mi, T., and Goodell, M.A. 1999. Hematopoietic potential of stem cells isolated from murine skeletal muscle. Proc. Natl. Acad. Sci. U.S.A. 96:14482-14486.
9.18.10 Supplement 19
Current Protocols in Cytometry
J o rda n, C.T., A stle , C.M., Zawadzki, J., Mackarehtschian, K., Lemischka, I.R., and Harrison, D.E. 1995. Long-term repopulating abilities of enriched fetal liver stem cells measured by competitive repopulation. Exp. Hematol. 23:1011-1015. Loken, M.R. 1980. Separation of viable T and B lymphocytes using a cytochemical stain, Hoechst 33342. J. Histochem. Cytochem. 28:3639. Morrison, S. and Weissman, I. 1994. The long-term repopulating subset of hematopoietic stem cells is deterministic and isolatable by phenotype. Immunity 1:661-673. Pallavicini, M.G., Summers, L.J., Dean, P.N., and Gray, J.W. 1985. Enrichment of murine hemopoietic clonogenic cells by multivariate analyses and sorting. Exp. Hematol. 13:11731181. Sandhu, L.C., Warters, R.L., and Dethlefsen, L.A. 1985. Fluorescence studies of Hoechst 33342 with supercoiled and relaxed plasmid pBR322 DNA. Cytometry 6:191-194. Schinkel, A., Mayer, U., Wagenaar, E., Mol, C., van Deemter, L., Smit, J., van der Valk, M., Voordouw, A., Spits, H., van Tellingen, O., Zijlmans, J., Fibbe, W., and Borst, P. 1997. Normal viability and altered pharmacokinetics in mice lacking mdr1-type (drug-transporting) P-glycoproteins. Proc. Natl. Acad. Sci. U.S.A. 94:4028-4033. Spangrude, G.J. and Johnson, G.R. 1990. Resting and activated subsets of mouse multipotent hematopoietic stem cells. Proc. Natl. Acad. Sci. U.S.A. 87:7433-7437. Spangrude, G.J., Heimfeld, S., and Weissman, I.L. 1988. Purification and characterization of mouse hematopoietic stem cells. Science 241:58-62. Sriram, M., van der Marel, G.A., Roelen, H.L., van Boom, J.H., and Wang, A.H. 1992. Conformation of B-DNA containing O6-ethyl-G-C base pairs stabilized by minor groove binding drugs: Mole cula r s truc ture of d(CGC[e6G]AATTCGCG complexed with Hoechst 33258 or Hoechst 33342. EMBO J. 11:225-232.
Steen, H.B. and Stokke, T. 1986. Fluorescence spectra of cells stained with a DNA-specific dye, measured by flow cytometry. Cytometry 7:104106. Stokke, T. and Steen, H.B. 1986. Binding of Hoechst 33258 to chromatin in situ. Cytometry 7:227234. Storms, R.W., Goodell, M.A., Fisher, A., Mulligan, R.C., and Smith, C. 2000. Hoechst dye efflux reveals a novel CD7(+)CD34(−) lymphoid progenitor in human umbilical cord blood. Blood 96:2125-2133. Uchida, N. and Weissman, I.L. 1992. Searching for hematopoietic stem cells: Evidence that Thy1.1lo Lin− Sca-1+ cells are the only stem cells in C57BL/Ka-Thy-1.1 bone marrow. J. Exp. Med. 175:175-184. Watson, J.V., Nakeff, A., Chambers, S.H., and Smith, P.J. 1985. Flow cytometric fluorescence emission spectrum analysis of Hoechst-33342stained DNA in chicken thymocytes. Cytometry 6:310-315. Wolf, N.S., Kone, A., Priestley, G.V., and Bartelmez, S.H. 1993. In vivo and in vitro characterization of long-term repopulating primitive hematopoietic cells isolated by sequenctional Hoechst 33342-rhodamine 123 FACS selection. Exp. Hematol. 21:614-622. Zhou, S., Schuetz, J.D., Bunting, K.D., Colapietro, A.M., Sampath, J., Morris, J.J., Lagutina, I., Grosveld, G.C., Osawa, M., Nakauchi, H., and Sorrentino, B.P. 2001. The ABC transporter Bcrp1/ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat. Med. 7:10281034.
Contributed by Margaret A. Goodell Baylor College of Medicine Houston, Texas
Stewart Connor was a stellar flow cytometrist who, with patience and humor, taught me flow cytometry and helped discover SP cells. I thank Glenn Paradis (MIT) for years of productive and fun work together. I also thank Carlos Ramos for helpful discussions and the current Baylor College of Medicine/Texas Children’s Hospital team, Mike Cubbage and Brian Newsom, for their tireless support of my laboratory.
Studies of Cell Function
9.18.11 Current Protocols in Cytometry
Supplement 19
Assessment of Phagocyte Functions by Flow Cytometry
UNIT 9.19
Phagocytes are crucial to effective defense against infections. Neutrophilic polymorphonuclear leukocytes (PMNs), and monocytes and macrophages (MΦs), are crucial for cellular defense against infections. Phagocyte functions encompass phagocytosis (attachment and internalization), intra- and extracellular digestion of targets, oxidative burst, and chemotaxis. Accurate measurement of phagocytosis in one cell population requires discrimination from other populations by light scatter or fluorescent staining. Phagocytosis of most microorganisms requires serum factors (opsonins). The major opsonins are antibodies, complement factors, and fibronectin. Flow cytometry is applied to both routine and experimental quantification of phagocyte functions as well as to the quantitation of opsonin and antigen concentrations and activities. Bacteria, zymosan particles, and microspheres are used as targets for different experimental and clinical purposes. The basic protocols in this unit outline a four-step tandem procedure for evaluating all the above-mentioned phagocyte functions: attachment and internalization (see Basic Protocol 1), processing of bacteria and zymosan particles (see Basic Protocol 2), phagocytosis and oxidative burst (see Basic Protocol 3), and chemotaxis (see Basic Protocol 4). PHAGOCYTOSIS (ATTACHMENT AND INTERNALIZATION) AND PHAGOSOMAL pH
BASIC PROTOCOL 1
This protocol is a two-step procedure for measurement of total phagocytosis (attachment and internalization) and phagosomal pH. FITC-stained bacteria and zymosan particles are used as targets, which are mixed with opsonins and leukocytes under controlled mixing and temperature conditions. The first step concerns measurements of percentage phagocytosis, phagocyte fluorescence, and the loss of targets from the suspension, followed by calculation of the phagocytic index. In the second step, trypan blue is added after phagocytosis, and percentage phagocytosis and phagocyte fluorescence are recorded. The numbers of attached and internalized targets are calculated from the phagocytic index, the fluorescence of the phagocytes before and after trypan blue quenching, and the fluorescence of targets. The difference in the fluorescence between the internalized and the free, extracellular targets is taken as an indirect measure of phagosomal pH using a standard curve of pH versus FITC fluorescence. The intensity of the phagocyte FITC fluorescence depends on complex intracellular mechanisms that have to be taken into account when evaluating phagosomal pH (see Commentary). Materials Targets: Bacteria S. aureus Cowan III (NCTC 8532; see Support Protocol 4) Zymosan particles (see Support Protocol 3 and/or 5) Polychromatic red fluorescence (PC red)-dyed microspheres, 1-µm diameter (Fluoresbrite Plain Microspheres, Polysciences) coated with outer membrane vesicles from meningococci (OMV-beads; see Support Protocol 7) Sørnes’s buffer (see recipe) Opsonins (see Support Protocol 9) Dihydrorhodamine 123 (DHR; see recipe) White blood cell suspension (see Support Protocol 1) Dulbecco’s phosphate-buffered saline (DPBS; see recipe) containing 0.02% (w/v) EDTA, ice cold Contributed by Carl-Fredrik Bassøe Current Protocols in Cytometry (2002) 9.19.1-9.19.22 Copyright © 2002 by John Wiley & Sons, Inc.
Studies of Cell Function
9.19.1 Supplement 21
Trypan blue Vindeløv’s high-salt solution (see recipe) Fluorescent beads (e.g., DNA-Check, Coulter) Flow cytometer with 488-nm excitation and filters for the detection of green fluorescence (R123, 505 to 545 nm), orange-red fluorescence (PC red beads, 560 to 590 nm), and red fluorescence (CD14-PE-Cy5, 660 to 700 nm) Non-pyrogenic 96-well microtiter plates, sterile 12 × 75–mm tubes suitable for use on the flow cytometer Prepare targets 1. Thaw selected targets. Calculate target concentration by flow cytometry using a known target volume (VT), and a known volume (VL) and concentration (CL) of leukocytes as internal reference, and note percent leukocytes (PL) and percent targets (PT). Target concentration = (PT/PL) × (CLVL/VT). Targets are prepared and kept for several years. See specific support protocols for target preparation.
2. Adjust bacteria and zymosan targets to 2.5 × 108 and 1.25 × 108 per milliliter in Sørnes’s buffer, respectively. Targets are adjusted to a useful concentration relative to the amount of added FITC, and the applied quenching. The initial target/phagocyte (PMN + MΦ) ratio (R) is 20:1 for bacteria and 10:1 for zymosan particles.
Incubate cells 3. Add 20 µl each of targets, diluted opsonins, and DHR working solution to each well of 96-well microtiter plates, and adjust the final incubation mixture to 80 µl per well by adding 20 µl Sørnes’s buffer. Use an additional duplicate plate (termed QUICKMIX) for controls. Add all the mentioned ingredients, except diluted opsonins, to the QUICK-MIX plate. 4. Incubate 7.5 min at 37°C. 5. Add 20 µl white blood cell suspension to each well. 6. Incubate suspensions 7.5 min for phagocytosis, and 30 min for processing (see Basic Protocol 2) under controlled mixing at 37°C. 7. Stop mixing, and immediately add 200 µl ice-cold DPBS containing 0.02% EDTA to each well using the same sequence of additions as for the cells in order to equalize phagocytosis time for all the wells. Add 20 µl diluted opsonins to the QUICK-MIX plate. 8. Label three 12 × 75–mm tubes for each well. Add 400 µl DPBS with 0.02% EDTA to one set, 400 µl of 2 mg/ml trypan blue in DPBS with 0.02% EDTA to the second set, and 400 µl Vindeløv’s high salt-solution to the third set. The first set of tubes is used for determination of percentage phagocytosis, phagocyte fluorescence, target loss, and phagocytic index. The second set of tubes is used for discrimination between attachment and internalization. The third set of tubes is reserved for processing in Basic Protocol 2. Assessment of Phagocyte Functions by Flow Cytometry
9. Mix the contents of each well in the 96-well plate thoroughly using a pipet (cells and targets sediment at different rates). For each well, transfer 100 µl suspension to each of the three labeled tubes (step 8) corresponding to that well. Keep the tubes on ice and in the dark until flow cytometry measurement.
9.19.2 Supplement 21
Current Protocols in Cytometry
A 1023 L
Forward scatter
M
0 0
1023 Side scatter
B
Forward scatter
1023 N1 N2
N3 N4 0 100 101 102 FITC fluorescence
103
100 101 102 FITC fluorescence
103
C
Forward scatter
1023 O1 O2
O3 O4 0
Figure 9.19.1 The monocytes (M) and neutrophils (L) in the FS versus SS cytogram of leukocytes phagocytosing bacteria or zymosan particles are gated to make separate FS versus FITC fluorescence cytograms of monocytes and neutrophils (A). Percentage statistics and mean FITC fluorescence values are recorded in the latter cytograms (B). The third cytogram (C) of FS versus FITC fluorescence gives the relative counts of phagocytes and free targets. The phagocytic index is calculated from the latter relative counts.
Studies of Cell Function
9.19.3 Current Protocols in Cytometry
Supplement 21
10. Mix the contents of each tube thoroughly (cells and targets sediment at different rates) just prior to running on the flow cytometer. Acquire data 11. Calibrate flow cytometer with fluorescent beads (e.g., DNA-Check, Coulter). Use electronic color compensation (UNIT 1.14) to minimize the spectral emission overlaps between fluorochromes. Coincidence rate of targets and leukocytes in the flow cell should be 1% to 2% (<5%).
12. Run the first set of tubes (DPBS with 0.02% EDTA only). Collect forward scatter (FS), side scatter (SS), and yellow-green FITC fluorescence (FITC; Fig. 9.19.1). Create ungated FS versus FITC cytogram (total cytogram). 13. Set region about noise. For the phagocytosis cytogram, make gated FS versus FITC from cytogram by subtracting the noise region to create phagocytosis cytogram. 14. Create quadrant statistics in phagocytosis cytogram. Place the vertical limit close to the left of the targets. Place the horizontal close beneath the lymphocytes. Record quadrant statistics: region 1 contains non-phagocytes (N), region 2 contains phagocytes (F), and region 4 contains targets (T). 15. Record percentage phagocytosis: P% = 100 × F%/(F% + N%). 16. Calculate average phagocytic index: IP = R – (T%/F%). R is the initial target/phagocyte (PMN + MΦ) ratio.
17. Note the mean fluorescence of the free, extracellular targets (FTe). Create PMN and MΦ cytograms 18. Make gated FS versus FITC cytograms by gating on PMN and MΦ, respectively, in the FS versus SS cytogram. Create quadrant statistics by placing the vertical and horizontal limits as described in step 14. 19. Record quadrant statistics: region 1 contains non-phagocytosing PMNs or MΦs (N); region 2 contains phagocytosing PMNs or MΦs (F). 20. Calculate PMN percentage phagocytosis: PCPMN% = 100 × FCPMN%/(FCPMN% + NCPMN%). 21. Calculate MΦ percentage phagocytosis: PCMΦ% = 100 × FCMΦ%/(FCMΦ% + NCMΦ%). 22. Record the mean fluorescence of the phagocytosing PMNs (FCPMN) and the mean fluorescence of the phagocytosing MΦs (FCMΦ). Determine attachment and internalization 23. Run the second set of tubes (trypan blue–containing samples) with the same flow cytometer settings and record PMN percentage phagocytosis (PQPMN%). 24. Calculate percentage of PMNs with only adherent targets: Assessment of Phagocyte Functions by Flow Cytometry
PAPMN% = PCPMN% – PQPMN%. 25. Record MΦ percentage phagocytosis (PQMΦ%).
9.19.4 Supplement 21
Current Protocols in Cytometry
26. Calculate percentage of MΦs with only adherent targets: PAMΦ% = PCMΦ% – PQMΦ%. 27. Record the mean fluorescence of the phagocytosing PMN (FQPMN) and the mean fluorescence of the phagocytosing MΦs (FQMΦ). 28. Calculate the mean number of targets attached to each PMN: IAPMN = (FCPMN – FQPMN)/FTe. 29. Calculate the mean number of targets attached to each MΦ: IAMΦ = (FCMΦ – FQMΦ)/FTe. 30. Calculate the mean fluorescence of the targets internalized by the PMNs: FTiPMN = FQPMN/(IPMN × FTe). 31. Read the phagosomal pH off the pH versus fluorescence standard curve. Establish a standard curve by measuring target fluorescence in suspensions with pH 4 to 10 at 0.5-pH step intervals. PROCESSING OF BACTERIA AND ZYMOSAN PARTICLES Phagocytes are added to Vindeløv’s high-salt solution containing the detergent NP40. In a single step, phagocytes are lysed, most of the targets are liberated, and some adhere to
BASIC PROTOCOL 2
1000 digestion
EB fluorescence
Nc − Z
Nc
100
10
1 Z .1 .1
1
10 100 FITC fluorescence
1000
Figure 9.19.2 Cytogram for the measurement of digestion. Cytogram of phagocytosing leukocytes after the dissolution of the leukocyte cytoplasm in Vindeløv’s solution. Region Nc (EB-fluorescent nuclei) contains mainly lymphocyte nuclei. Region Nc + Z (EB- and FITC-fluorescent nuclei) contains monocyte and neutrophil nuclei with 1 to 3 attached targets. Region Z contains free targets (bacteria or zymosan particles). The mean EB and FITC fluorescence values are recorded from region Z.
Studies of Cell Function
9.19.5 Current Protocols in Cytometry
Supplement 21
the cell nuclei. Target DNA is stained with ethidium bromide (EB), and RNA is dissolved by RNase. Flow cytometry measurements on FS, SS, and FITC and EB fluorescence can be made on individual leukocyte nuclei and leukocyte nuclei with a few adherent targets, as well as on the liberated, free targets and those that remained extracellularly during phagocytosis. Materials Cells in Vindeløv’s high-salt solution from Basic Protocol 1, step 9 Flow cytometer with 488-nm excitation and filter set for detection of green (FITC) and red (EB) fluorescence
A CD14-PE-Cy5 fluorescence
103
102 B 101
A
100
0
1023 Side scatter
B
R123 fluorescence
103
D1D2
102
101
100
D3D4 100
101
102
103
PE-labeled OMV-beads
Assessment of Phagocyte Functions by Flow Cytometry
Figure 9.19.3 After the phagocytosis of OMV-beads, monocytes and neutrophils cannot be discriminated by FS versus SS cytogram. Therefore, monocytes and neutrophils are stained with PE-Cy5-labeled antiCD14 monoclonal antibody, and discriminated by PE-Cy5 fluorescence and SS (A). Monocytes are positive for CD14 and are in region B, whereas, neutrophils are negative or weakly CD14-positive, and located in region A. Monocytes and neutrophils are gated to separate cytograms (B) for measurement of phagocytosis (PE-fluorescence of OMV-beads) and oxidative burst (R123 fluorescence). The mean PE and R123 fluorescence of double-positive cells are recorded from region D2, and that of single-positive cells from regions D1 and D4.
9.19.6 Supplement 21
Current Protocols in Cytometry
Acquire data For bacteria targets 1a. Collect yellow-green FITC fluorescence and red EB fluorescence, and make log FITC versus log EB cytogram. Place electronic gating region about individual cell nuclei (Fig. 9.19.2, region Nc), cell nuclei with adherent targets (Fig. 9.19.2, region Nc + Z), and free and liberated target cell nuclei (Fig. 9.19.2, region Z). 2a. Record mean fluorescence values on region Z in Figure 9.19.2. 3a. Compare mean green FITC and red EB fluorescence of QUICK-MIX with that of PHS and PHS-IN. For zymosan particles 1b. Make appropriate regions and collect information as in steps 1a to 3a (new settings are required due to the higher FITC and EB fluorescence intensities of the zymosan particles). 2b. Gate on region Z in Figure 9.19.2 into FS versus SS cytogram. 3b. Record mean FS and SS from the targets, and compare mean FS and SS of QUICKMIX (see below) with that of HYP and HYP-IN. PHAGOCYTOSIS AND OXIDATIVE BURST This protocol is used for the concurrent measurement of phagocytosis and oxidative burst. Various microspheres can be used as targets (see Commentary). In this case, phagocytosis is monitored by polychromatic red fluorescent (PC red) microspheres coated with outer membrane vesicles from meningococci (OMV-beads). The oxidative burst is quantified by the conversion of dihydrorhodamine123 (DHR) to green fluorescent rhodamine 123 (R123). PMNs and MΦs are discriminated using an orange fluorescent monoclonal antiCD14 antibody. Data are simultaneously collected on targets, PMNs, and MΦs.
BASIC PROTOCOL 3
Additional Materials (also see Basic Protocol 1) CD14-PE-Cy5 monoclonal antibody Desired microspheres (see Support Protocol 7), adjusted to a concentration of 2.5 × 108 /ml in Sørnes’s buffer (see recipe); the initial microsphere/phagocyte (PMN + MΦ) ratio (R) is 20:1. Incubate cells 1. Follow Basic Protocol 1, steps 3 to 7. 2. Prepare microtiter plates in advance by adding 5 µl undiluted monoclonal antibody to the desired number of wells. Add 100 µl cell suspension to each well, mix the suspension gently, and incubate 1 hr in the dark on ice. 3. Mix the contents of each well thoroughly using a pipet and transfer to a 12 × 75–mm tube containing 400 µl DPBS with 0.02% EDTA and proceed with flow cytometry. Acquire data 4. Create cytogram of CD14 versus SS. Set separate regions about noise and small particles, monocytes, and neutrophils. Also create a histogram on red fluorescence (targets and phagocytes) gated from debris and small particles. Collect statistics on free targets, and record mean target red fluorescence. Studies of Cell Function
9.19.7 Current Protocols in Cytometry
Supplement 21
5. Create two cytograms to monitor phagocytosis and oxidative burst by monocytes and neutrophils by placing quadrant statistics as described in Basic Protocol 1 (Fig. 9.19.3). Record the percentages in all regions. 6. Percentage phagocytosis P% = percentage of counts in region 2 plus region 4. Percentage oxidative burst PO% = percentage of counts in region 1 plus region 2. Percentage of inactive cells = percentage from region 3. Percentage of non-phagocytosing cells producing oxidative burst = percentage from region 1. Percentage of phagocytosing cells without oxidative burst = percentage from region 4. 7. Record the mean green fluorescence in region 1 (oxidative burst alone), green and red fluorescence in region 2 (oxidative burst and phagocytosis), and red fluorescence in region 4 (phagocytosis only). 8. For region 2 in cytograms from both monocytes and neutrophils, calculate phagocytic index IPMN (IMΦ) = mean red fluorescence of phagocytes in region 2 divided by the red fluorescence of the free targets. 9. For region 2, calculate neutrophil oxidative ratio OPMN = mean green fluorescence of phagocytes in region 2 divided by IPMN, and monocyte oxidative ratio OMΦ = mean green fluorescence of phagocytes in region 2 divided by IMΦ. BASIC PROTOCOL 4
CHEMOTAXIS Measurement of chemotaxis is based on the selective, active, and stimulated motion of neutrophils across a 3-µm filter. The source population serves as a reference, and non-stimulated cells as a control. Two or more chemotactic stimuli, zymosan-activated serum (ZAS) and formyl-methionyl-leucyl-phenylalanine (fMLP), are routinely used. Materials White blood cell suspension (see Support Protocol 1) Sørnes’s buffer (see recipe) DPBS (see recipe) containing 0.2% (w/v) EDTA 5% zymosan-activated serum (ZAS; see Support Protocol 3) Paraformaldehyde (optional) 12 × 75–mm tubes 24-well transwell chemotaxis plates (6.5-mm diameter, 3-µm pore size; Corning) 37°C, 5% CO2 incubator Prepare cells 1. Centrifuge white blood cell suspension 5 min at 350 × g, 20°C, and resuspend in Sørnes’s buffer to a final concentration of 5 × 106 leukocytes/ml. 2. For reference sample, add 2 ml DPBS containing 0.2% EDTA, 100 µl cell suspension, and 500 µl Sørnes’s buffer to each of two 12 × 75–mm tubes. These tubes contain the same volume as the final dilution in the lower well, and are the 100% cell count.
3. Add 600 µl Sørnes’s buffer to the lower chamber of two wells of the 24-well transwell microtiter plate (control). 4. Add 600 µl of 5% ZAS (or other stimulus) in Sørnes’s buffer to the lower chamber of two wells of the 24-well transwell microtiter plate (control). Assessment of Phagocyte Functions by Flow Cytometry
5. Add 100 µl cell suspension (5 × 106 leukocytes/ml) to the corresponding upper chambers of the 24-well transwell microtiter plate.
9.19.8 Supplement 21
Current Protocols in Cytometry
6. Incubate 30 min in a 37°C, 5% CO2 incubator. After the incubation, immediately wash upper and lower chambers each with 2 ml ice-cold DPBS containing 0.2% EDTA, and transfer the leukocytes to 12 × 75–mm tubes. Keep the cell suspensions on ice and proceed with flow cytometry, or fix cells in 1.0% final paraformaldehyde for later investigation. Acquire data 7. Record total and differential counts of lymphocytes, MΦs, and PMNs using FS versus SS and appropriate gates from reference sample and the lower chamber. 8. Calculate spontaneous (random) motion = 100 × PMN count in lower chamber without stimulus divided by PMN count in reference suspension. 9. Calculate chemotaxis = 100 × PMN count in lower chamber with stimulus divided by PMN count in reference suspension. PREPARATION OF WHITE BLOOD CELL SUSPENSION This protocol describes the isolation of a white blood cell suspension from peripheral blood. The suspension provides starting material for the basic protocols.
SUPPORT PROTOCOL 1
Materials Lysing solution (see recipe) Dulbecco’s phosphate buffered saline with glucose and BSA (DPBS-GA; see recipe) Vacutainer tubes containing 100 U preservative-free heparin (UNIT 9.7) 50-ml centrifuge tubes, sterile Additional materials for blood collection 1. Collect peripheral blood sample in the Vacutainer tube containing heparin. 2. In a 50-ml centrifuge tube, dilute heparinized blood 1:10 (v/v) in lysing solution, mix gently, and leave 10 min at room temperature. 3. Centrifuge 5 min at 350 × g, room temperature, resuspend the pellet in 20 ml washing solution, and centrifuge again 5 min at 350 × g, room temperature. 4. If lysis is incomplete, resuspend the pellet 5 min in NH4Cl solution (final 8 mg/ml), pH 7.4, centrifuge 5 min at 350 × g, room temperature, and resuspend in 20 ml DPBS-GA. 5. Centrifuge 5 min at 350 × g, room temperature, and resuspend pellet in 1 ml DPBS-GA per 10 ml of starting heparinized blood. 6. Adjust the concentration to 1.25 × 107 white cells/ml in DPBS-GA.
Studies of Cell Function
9.19.9 Current Protocols in Cytometry
Supplement 21
SUPPORT PROTOCOL 2
PREPARATION OF SERUM FROM PERIPHERAL BLOOD Blood is collected in serum tubes with no anticoagulant and is allowed to stand undisturbed until it clots. The clot is removed and the remaining liquid is centrifuged to remove any residual erythrocytes and debris. The resulting serum is divided into aliquots and frozen. Materials Freshly extracted peripheral blood Red-top collection tubes (no anticoagulant) 50-ml centrifuge tubes Additional equipment for venipuncture 1. Place tube of blood in a rack and allow to sit undisturbed until a clot forms. Blood must be collected without an anticoagulant. If a clot does not start to form within 15 min, initiate clotting by inserting a wooden applicator stick into the tube. Then begin with step 1.
2. Using a wooden applicator stick, gently loosen the clot and remove from the tube. Do not break up clot.
3. Transfer serum to 50-ml centrifuge tube and centrifuge 10 min at 2700 × g, 4°C. 4. Save the supernatant. 5. Combine the supernatants from at least 7 different donors, divide into 500-µl aliquots or other desired volume, and store in appropriate containers at –20°C. Do not subject sera to repeated freeze-thaw cycles. SUPPORT PROTOCOL 3
ZYMOSAN-ACTIVATED SERUM (ZAS) AND PREOPSONIZED ZYMOSAN PARTICLES ZAS is used as one stimulus for cell motility in the chemotaxis assay. Preopsonized zymosan particles are used in the assay for complement-mediated phagocytosis. ZAS and complement fragment C3b- and C3bi-opsonized zymosan particles are prepared simultaneously. Materials Zymosan A particles (Sigma) DPBS (see recipe) Fresh human serum (see Support Protocol 2) Sørnes’s buffer (see recipe) 15-ml plastic centrifuge tube 37°C incubator with rotator 1. Add 100 mg zymosan A particles and then 10 ml DPBS to a 15-ml plastic centrifuge tube and shake vigorously. 2. Centrifuge 10 min at 2000 × g, 20°C. 3. Pour off supernatant and add 12.5 ml fresh human serum. Shake and rotate solution 1 hr at 37°C. 4. Centrifuge 10 min at 2000 × g, 20°C, decant serum and label as ZAS, and store up to 5 years at –80°C.
Assessment of Phagocyte Functions by Flow Cytometry
5. Resuspend zymosan A pellet in 10 ml Sørnes’s buffer. Use immediately or store as preopsonized zymosan particles in 1-ml aliquots up to 5 years at –80°C.
9.19.10 Supplement 21
Current Protocols in Cytometry
FITC-LABELING OF BACTERIA FITC-labeled bacteria are used to investigate adhesion of bacteria to leukocytes, and antibody-dependent and antibody-plus-complement-stimulated phagocytosis.
SUPPORT PROTOCOL 4
Materials Fluorescein isothiocyanate (FITC) 96% and 70% ethanol Heart infusion broth (HIB; Difco) S. aureus Cowan III (NCTC 8532 O/N) Blood agar plates (e.g., Becton Dickinson) 0.9% (w/v) NaCl Sørnes’s buffer (see recipe) Spectrophotometer 1. Weigh out 10 mg FITC, and add to 0.1 ml of 96% ethanol. Dilute the FITC-ethanol solution to 100 ml in heart infusion broth (HIB). 2. Grow S. aureus Cowan III on blood agar, and add bacteria to ∼2 ml HIB. Mix until the suspension is free from clumps. Warm 100 ml HIB to 37°C and add the 2-ml S. aureus suspension. Read the OD using a spectrophotometer (the OD of the suspension should be 0.05 to 0.1 at 620 nm). Incubate at 37°C until OD = 1.0 (∼2.5 hr). 3. Pellet the bacteria by centrifuging 10 min at 2000 × g, room temperature. Resuspend bacteria in ∼2 ml of 0.9% NaCl. Mix until there are no visible bacteria clumps. 4. Dilute bacteria in ∼90 ml ice-cold 70% ethanol to OD=1.0. Let the bacteria fix 30 min at 0°C. Repeat step 3 three times at 4°C. 5. Dilute the bacteria in two 50-ml tubes each containing 30 ml FITC solution (see step 1). Incubate 30 min at 37°C with continuous mixing. 6. Repeat step 3 three times. Resuspend bacteria in Sørnes’s buffer, and adjust to OD = 1. Divide suspensions into 0.5-ml aliquots and store FITC-stained bacteria up to 5 years at –80°C. FITC-LABELING OF ZYMOSAN PARTICLES Phagocytosis is mediated by complement fragments, mainly C3b and C3bi. Zymosan particles bind complement factor C3 that is converted into C3b and C3bi, which remain covalently bound on the particles’ surface. FITC-labeled zymosan particles are used to measure complement-mediated phagocytosis, and intracellular processing mediated by complement.
SUPPORT PROTOCOL 5
Additional Materials (also see Support Protocol 4) Zymosan A particles DPBS (see recipe) 10-ml glass tubes 0.2-µm filter End-over-end rotator, 37°C 1. Add 100 mg zymosan A particles to each of ten 10-ml glass tubes. 2. Make FITC solution: Place 10 mg FITC in a 10-ml glass tube. Add 0.5 ml of 96% ethanol to dissolve the FITC. Add 9.5 ml DPBS at 37°C to a final 1 mg/ml FITC stock solution. Dilute with DPBS at 37°C to a working solution containing 0.05 mg/ml FITC.
Studies of Cell Function
9.19.11 Current Protocols in Cytometry
Supplement 21
3. Filter sterilize through a 0.2-µm filter. 4. Add 10 ml of 0.05 mg/ml FITC solution to each glass tube with zymosan particles. Incubate 30 min at 37°C with end-over-end rotation. 5. Centrifuge 10 min at 2000 × g, room temperature, and resuspend the pellet in DPBS. Repeat three times. 6. Centrifuge 10 min at 2000 × g, 20°C, and resuspend the pellet in 200 ml Sørnes’s buffer. Store 0.5-ml aliquots protected from light up to 5 years at –80°C. SUPPORT PROTOCOL 6
DILUTIONS OF FITC-LABELED ZYMOSAN A PARTICLES This protocol is used for the adjustment of FITC concentration, in case the FITC-stained particles are not quenched by trypan blue. Additional Materials (also see Support Protocol 5) Trypan blue 15-ml polypropylene tubes 1. Add 100 mg zymosan A particles and 2 ml DPBS to a 15-ml polypropylene tube. Mix with a pipet or ultrasound (30 sec) until the suspension is free from clumps. Dilute to 10 ml with DPBS. Centrifuge 10 min at 2000 × g, room temperature, and discard the supernatant. 2. Dissolve 0.25 mg FITC in 0.1 ml of 96% ethanol and immediately dilute in 1 ml DPBS at 37°C. Wait until all FITC is solubilized. Pass through a 0.22-µm filter. Make the following concentrations of FITC in DPBS: 0.25, 0.1, 0.05, and 0.025 mg/ml. 3. Resuspend zymosan A pellet in a little FITC solution and mix until free from clumps. Thereafter, add 10 ml FITC solution per 100 mg zymosan particles. Incubate with end-over-end rotation 30 min at 37°C. 4. Centrifuge 10 min at 2000 × g, room temperature. Decant and resuspend in 10 ml room temperature 0.9% NaCl. Repeat this step three times. 5. Resuspend the pellet in 10 ml DBPS. Count zymosan particles using flow cytometry (see Support Protocol 8) and dilute to 5 × 108 particles/ml. 6. Run suspension on flow cytometer and place the modal FITC fluorescence in channel ∼100. 7. Add 100 µl FITC-labeled zymosan suspension to 400 µl trypan blue suspension. Run on flow cytometer. Use the solution with the highest FITC concentration that is 100% effectively quenched. 8. Store the selected FITC-stained zymosan population in 1-ml aliquots up to 5 years at –80°C.
Assessment of Phagocyte Functions by Flow Cytometry
9.19.12 Supplement 21
Current Protocols in Cytometry
ANTIGEN COATING OF POLYSTYRENE BEADS In principle, beads are incubated with an excess of antigen overnight at room temperature (20°C). Unreacted sites on the beads are blocked with 2% BSA. Antigen-coated beads are suspended in a storage buffer, and kept protected from daylight at 4°C.
SUPPORT PROTOCOL 7
OMV from N. meningitidis strain 44/76 (B:15:P1.7,16) is prepared as for vaccine production (Lehmann, 1997). Materials Fluoresbrite PC red 1.0-µm microspheres (Polysciences) 0.1 M borate buffer, pH 8.5 (0.1 M boric acid; Polysciences) OMV-beads with and without 2% (w/v) bovine serum albumin, endotoxin-free (BSA; Roche Diagnostics) Sodium phosphate storage buffer (Polysciences) End-over-end rotator 1. Centrifuge 500 µl Fluoresbrite PC red microspheres (4.55 × 1010 microspheres/ml) 5 min at 15,600 × g, room temperature. Resuspend pellet in 550 µl of 0.1 M borate buffer, pH 8.5. 2. Add 450 µl of a 0.94 mg/ml (saturating) OMV suspension. Mix by end-over-end rotation 20 hr in the dark at room temperature. 3. Centrifuge OMV-beads 5 min at 15,600 × g, room temperature, and resuspend them in OMV suspension containing 2% endotoxin-free BSA. Incubate with end-over-end rotation 20 hr at 20°C. 4. Centrifuge OMV-beads with blocked free surface sites 5 min at 15,600 × g, room temperature, and resuspend in sodium phosphate storage buffer. Store up to 1 year at 4°C. COUNTING TARGETS For the quantification of small particles, leukocytes of a known concentration are used as reference. Mix known volumes of leukocytes and beads. Discriminate leukocytes, debris, and fluorescent beads by forward and side scatter. Place electronic windows around targets and leukocytes and record the relative count of beads and leukocytes. Calculate the concentration of beads by multiplying the concentration of leukocytes by the relative bead/leukocyte ratio.
SUPPORT PROTOCOL 8
OPSONINS The phagocytosis of many pathogenic microorganisms and fungi requires serum opsonins. Complement is a natural opsonin for many microorganisms and fungi. Complement factor 3 (C3) binds to the surface of many non-capsulated microorganisms and zymosan particles. C3 is converted to C3b and C3bi that remain attached to the targets, and bind to complement receptors on phagocytes. For example, C3bi binds to complement receptor 3(CD11b). Hypo- and agammaglobulinemic sera are useful sources of complement. Due to the low antibody concentration, they are used in conjunction with zymosan particles to monitor complement-mediated phagocytosis.
SUPPORT PROTOCOL 9
Infections normally induce the synthesis of antibodies. In the acute phase (day 0 to 2), the concentration of specific antibodies against the infectious agent is low (acute-phase serum). After 2 to 4 weeks, the concentration of specific antibodies is high (convalescent serum). Antibodies bind to the microorganisms by the Fab part, and link them to the phagocytes through the Fc part. The Fc part of IgG promotes phagocytosis by binding to Fc-receptors FcγRII (CD32) and FcγRIII (CD16) on neutrophils, and FcγRII (CD32) on
Studies of Cell Function
9.19.13 Current Protocols in Cytometry
Supplement 21
monocytes. Convalescent serum is used as a source of specific antibodies. Acute-phase serum serves as control. Materials Healthy human serum Hypogammaglobulinemic serum Acute-phase and convalescent serum Sørnes’s buffer (see recipe) For a pooled serum from healthy controls 1a. Collect serum (see Support Protocol 2) from healthy individuals, and ensure that CH50 is within the normal reference range. 2a. Mix equal volumes of serum from each of seven healthy individuals and label as pooled human serum (PHS). Heat-inactivate a portion (step 3a) and store the remainder in 0.5-ml aliquots up to 5 years at –80°C. 3a. Inactivate complement by heating 30 min at 56°C. Label as PHS-IN and immediately store in 0.5-ml aliquots up to 5 years at –80°C. 4a. Thaw shortly before use. For hypogammaglobulinemic serum 1b. Collect hypogammaglobulinemic serum (HYP) from agammaglobulinemic patients before immunoglobulin substitution. Heat-inactivate a portion (step 2b) and store the remainder in 0.5-ml aliquots up to 5 years at –80°C. 2b. Inactivate complement by heating 30 min at 56°C. Label as HYP-IN. Immediately after preparation store in 0.5-ml aliquots up to 5 years at –80°C. 3b. Thaw shortly before use. For acute-phase and convalescent serum 1c. For antigen- and antibody-dependent phagocytosis: collect acute-phase serum (APS) from patient on admission, and convalescent serum (REC), 6 weeks later. Heat-inactivate a portion (step 2c) and store the remainder in 0.5-ml aliquots at –80°C. In the example presented here, sera collected at admission and 6 weeks later from a patient with meningococcal C:15:P1.7,16 disease are used.
2c. Inactivate complement by heating 30 min at 56°C. Label as REC-IN. Immediately after preparation store in 0.5-ml aliquots up to 5 years at –80°C. 3c. Thaw shortly before use. 4c. Dilute 0.5 ml of each serum (APS, REC) 1:4 in Sørnes’s buffer. OMV is obtained from serogroup B meningococci vaccination strain 44/76, B:15:P1.7.16. OMV-suspension consists of 50 ìg protein and 3.5 ìg lipopolysaccharide per ml suspension.
Assessment of Phagocyte Functions by Flow Cytometry
9.19.14 Supplement 21
Current Protocols in Cytometry
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Dihydrorhodamine (DHR) 123 Stock solution: 10 mg/ml in dimethylsulfoxide (DMSO). Store protected from light ≤3 weeks at –80°C. Working solution: Dilute 1:1000 in Sørnes’s buffer (see recipe) before use. Dulbecco’s phosphate buffered saline (DPBS) Make 1 liter: 8 g NaCl 1.44 g Na2HPO4.2H2O 0.2 g KCl 0.2 g KH2PO4 Adjust to pH 7.4 Store ≤2 months at 4°C Dulbecco’s phosphate buffered saline with glucose and BSA (DPBS-GA) 100 ml DPBS (see recipe) 0.5 g BSA 0.1 g glucose Adjust to pH 7.4 Make fresh daily Lysing solution Make 1 liter: 8 g NH4Cl 0.8 g NaHCO3 0.88 g EDTA.2H2O pH ≈ 6.8 Store ≤2 months at 4°C Sørnes’s buffer 100 ml DPBS-GA (see recipe) 13.2 mg CaCl2.2H2O 12.1 mg MgSO4.7H2O Adjust to pH 7.4 Make fresh daily Vindeløv’s high-salt solution Make 1 liter: 0.01 M glycine 0.7507 g NaOH 9.8575 mg ethidium bromide (2.5 × 10-5 M) 1 ml Nonidet P-40 (0.1%, v/v) 17.532 g NaCl (0.3 M) pH should be 10 Store up to 1 year at 4°C. Just before use, add 700 U/liter ribonuclease from 100 U/ml stock (see recipe). Ribonuclease Stock solution 1 mg/ml = 100 U/ml Dissolve in 50 mM Tris⋅Cl, pH 8 (APPENDIX 2A). Store in 0.5-ml aliquots at –80°C. Studies of Cell Function
9.19.15 Current Protocols in Cytometry
Supplement 21
COMMENTARY Background Information General considerations Many specific disorders of phagocyte functions are known, such as chronic granulomatous disease, CD11b/CD18 deficiency, lazy leukocyte, and Chediak-Higashi syndrome. These are hereditary syndromes that manifest themselves in childhood. However, little is known concerning phagocyte functions in common, acquired clinical disorders of adults and the elderly. Surprisingly little has been done on leukemia, myelodysplasia, and bone marrow failure. For example, up to the year 2000, only 14 investigations of phagocyte functions in leukemia were available. Therefore, it is difficult to advise on the selection of useful methods in acquired immunodeficiencies related to phagocyte functions. Several methods are available for the monitoring of phagocyte functions. The present unit aims at in-depth investigations of phagocyte functions that reflect the tandem attack and processing of pathogenic microorganisms. Basic Protocols 1, 3, and 4 can be applied as stand-alone procedures. Digestion cannot be monitored without knowing that phagocytosis has taken place to a sufficient degree to monitor intracellular processing. Therefore, if Basic Protocol 2 is applied, Basic Protocol 1 must be used in conjunction. All four basic protocols can be used together on a routine basis. The protocols are fine-tuned to minimize the workload. For example, 20 µl of each reagent is added to all microwells to minimize preparation time, and the same buffer (DPBS) is used for cell preparations, dye and serum dilutions, and all incubations. All investigations are done in duplicate. Reproducibility is excellent. Day-to-day and interindividual variations with leukocytes from healthy individuals are small.
Assessment of Phagocyte Functions by Flow Cytometry
Less laborious methods Simpler methods are available, but these rely on assumptions that require careful controls. Technical details that might seem trivial are quantitatively important. The most important caveats are addressed below. Time-consuming isolation of leukocytes has been a major concern in working with flow cytometry and phagocytosis. Whole-blood methods have the advantage of being the least time consuming, and are useful in screening. The effect of serum opsonins cannot be dis-
criminated from cellular defects. If a deviation from the expected values is observed, it is impossible to know whether the unexpected results are caused by serum opsonins or leukocytes. In such instances, the present methods are needed as controls. It should also be noted that diminished phagocyte function can be compensated by effects of serum factors. Even when the overall results are within the reference range, whole-blood assays may miss important deviations from the reference range. Targets The purpose of the investigation determines the choice of target. Since sera and targets vary, there is no universal calibration method for phagocytosis. Zymosan particles In brief, zymosan particles are phagocytosed by complement, mainly C3bi, and are selected to monitor serum C3 opsonic activity and complement receptor activity (mainly CD11b/CD18). In HYP, 95% to 100% of PMNs and MΦs phagocytose zymosan particles. With HYP-IN, neither PMNs nor MΦs phagocytose zymosan particles. Two-color investigations using directly conjugated antibodies show that all the PMNs and MΦs from the same suspensions carry complement receptor 1 (CR1, CD35) and complement receptor 3 (CR3, CD11b/CD18). Both live and dead microorganisms can be used. S. aureus is used to monitor serum opsonins (mainly IgG), Fcγ-receptors, and the cooperation between IgG and complement. PHS and PHS-IN are used for combined IgG-complement phagocytosis and IgG-mediated phagocytosis, respectively. Assessment of phagocytosis using microorganisms is influenced by properties other than antigenic properties of microorganisms. Intracellular sorting in MΦs, e.g., differs between various strains of the same microbial species. Microorganisms containing green fluorescent protein (GFP) are widely used for monitoring phagocytosis. Some technical problems have not been addressed. For example, the killing and digestion of GFP by powerful phagocyte enzymes is usually not taken into account. Caution is required with interpretation of the results of fluorescence measurements.
9.19.16 Supplement 21
Current Protocols in Cytometry
Microspheres Coated beads have been used in phagocytosis assays for a long time, but the use of bead-associated antigens in the quantification of phagocytosis is of recent date (Bassøe et al., 2000). The advantage of using fluorescent beads lies in the ability to quantify the phagocytic and oxidative responses in mixed cell populations. Microspheres can be seen as surrogate microorganisms, and are useful for the study of specific opsonins and antigens. Various types of coated beads have been used in phagocytosis assays (reviewed in Bassøe et al., 2000): latexIgG, latex-IgG-C3, microspheres-C3b, microspheres-iC3b, beads bearing ligands for CD11b, latex-lysozyme, and OMV-beads. Any of these coated beads might be used in combined assays for monocyte and neutrophil functions. OMV-beads mimic N. meningitidis. Phagocytosis and oxidative burst mediated by OMV-beads require the presence of specific antimeningococcal antibodies and the appropriate antigen. REC and APS are used for combined IgG-complement phagocytosis and IgGmediated phagocytosis, respectively. The same principle can be applied to other antigens that can be attached to polystyrene microspheres. Attachment and internalization Various agents have been used to discriminate attached from internalized particles. The enzyme lysostaphin, which dissolves free and attached staphylococci, does not penetrate neutrophils, and does not digest internalized bacteria. Crystal violet is used to quench attached targets, but diffuses into phagocytes and quenches intracellular targets too. For this reason, crystal violet should be abandoned and replaced with trypan blue, which does not enter phagocytes. Attachment The percentage of phagocytosing PMNs and MΦs is of the same order of magnitude before and after trypan blue quenching. Thus, PMNs and MΦs do not carry targets only on the surface. As a rule, 0.5 to 2.0 targets are attached to the outer side of the plasma membrane. Internalization The remaining targets associated with PMNs or MΦs are internalized. These results are obtained whether phagocytosis is sustained by PHS or complement is inactivated (PHSIN).
Opsonins There is ample proof that phagocytosis of zymosan particles, many strains of S. aureus, N. meningitidis, and many other bacteria species does not occur in the absence of opsonins. The importance of attaching opsonins to suitable antigens is underestimated. Passive adsorption of opsonins to microspheres results in feeble phagocytosis. Target/phagocyte ratios >50:1 and incubation times >30 min are required to observe phagocytosis. Under these conditions, percentage phagocytosis is only ≤15%, and the phagocytic index ∼1 to 3. This low rate of phagocytosis is probably due to the attachment of the Fcγ-part of the IgG molecules to the microsphere surface, making them unavailable to Fcγ receptors on the leukocyte surface. Antigens Opsonins bind to antigens attached to the surface of targets. C3bi and the Fcγ-fragment of IgG extend into the medium, making them easily available for complement receptors and Fcγ-R, respectively. Phagocytic index Flow cytometric measurements of the phagocytic index have adapted the principles introduced by Leishman (1927) and Maaløe (1946). Leishman’s method is based on counting the number of targets associated with each cell. This is equivalent to measuring phagocyte fluorescence along the fluorescence dimension of the correspondingly labeled targets. Maaløe’s method depends on counting the number of free targets before and after phagocytosis. Leishman’s and Maaløe’s methods have advantages and disadvantages. (1) Leishman’s approach is applicable to fluorescent microspheres. The stability of the fluorescence is due to incorporation of dye in the solid polymer. The fluorescence intensity is unaffected by enzymes, toxic radicals, and pH. Thus, the mean phagocytic index can be easily and accurately calculated by dividing the mean fluorescence of each phagocyte subpopulation by the mean fluorescence of the free microspheres. When only one subpopulation of phagocytes is present, the phagocytic index calculated from microsphere fluorescence closely correlates with the index obtained from the loss of particles from the suspension (Lehmann et al., 1997). (2) Maaløe’s approach overcomes the problem that the phagocytic index cannot be quantified by the FITC fluorescence of the phago-
Studies of Cell Function
9.19.17 Current Protocols in Cytometry
Supplement 21
cytes containing FITC-labeled targets (Bassøe et al., 1983). Although that approach adds some technical inconveniences (see Troubleshooting), it is mandatory for accurate measurements.
Critical Parameters Fluorescence measurements Phagocyte FITC fluorescence has been used as a measure of phagocytosis. This is a simple approach that gives an overall impression of the combined fluorescence of attached and internalized targets. The phagosomal environment profoundly affects target fluorescence. Some fluorescence changes persist after the targets are liberated from the phagocytes (Bassøe, 1984; Hurst et al., 1984). In order to overcome the influence of phagosomal pH on fluorescent labels like FITC, phagocytes have been fixed and permeabilized (Cantinieaux et al., 1989). However, FITC molecules may be chemically modified by the oxidative burst and other intracellular chemical reactions (Hurst et al., 1984). Fluorescent molecules attached to digestible targets may be released from the targets by phagosomal enzymes (Bassøe, 1984). Permeabilization may liberate digested, small fluorescent amino acids, peptides, and glycosaminoglycans from the cells. However, digestion can barely be measured with <15-min incubation, and only a few fluorescent molecules are expected to be released with shorter incubation times. Fluorescence measurements may underestimate phagocytic indices. In general, the estimation of the phagocytic index from the fluorescence of digestible targets or pH-sensitive fluorescent probes is discouraged. QUICK-MIX controls The fluorescence of FITC molecules is influenced by serum factors other than electrolytes. Proteins, for example, may diminish the fluorescence intensity of FITC. The controls overcoming these problems are nicknamed QUICK-MIX. QUICK-MIX controls are not subject to phagocytosis, but all the other incubation steps are followed. There is only one exception. QUICK-MIX is not required for trypan blue–quenched samples of zymosan particle phagocytosis incubated in HYPO-IN since the leukocytes in these samples do not phagocytose at all. Assessment of Phagocyte Functions by Flow Cytometry
Phagosomal pH Phagocyte FITC fluorescence depends on quenching, digestion and release of florescent molecules (amino acids, peptides, proteins, and glucosaminoglycans) from stained targets, electrolyte composition and concentrations, and pH. Clearly, the estimation of phagosomal pH using FITC fluorescence should be interpreted with care. However, little digestion of phagocytosed material takes place during the first 15 min of incubation (Bassøe, 1984). At 15 min, the modal FITC fluorescences of liberated and free FITC-labeled particles are similar (Bassøe, 1984). At 7.5 min, the fluorescence intensity of liberated, FITC-stained targets is similar to that of the free FITC-labeled particles. Thus, within this short incubation time, the changes in FITC fluorescence of phagocytosed targets are reversible. FITC fluorescence is a sensitive pH indicator (Bassøe, 1984). Accordingly, with incubation times <15 min, total phagocyte fluorescence can be used as a (semiquantitative) measure of phagosomal pH. Mixing The frequency of hits between targets and phagocytes influences the rate of phagocytosis, and the phagocytosis index is profoundly affected by mixing (Bassøe, 2000). Several mixing procedures have been used: periodical stirring, end-over-end rotation, shaking, and automated microtiter plate mixers. Periodical stirring and shaking are undefined terms and should be avoided. End-overend rotation is used in flow cytometry phagocytosis assays to secure optimal mixing. The disadvantage is the large volume and high cell numbers (∼5 × 106 phagocytes) required. In order to miniaturize phagocyte assays, and at the same time standardize mixing, use 96-well microwell plates and standard mixing. Considering its importance, mixing of targets and leukocytes is taken on too easily. Vast differences are reported in assays with corresponding concentrations and activities of targets, phagocytes, and opsonins, but differences in mixing (see Bassøe, 2000). Some commercially available mixers are easy to use, the rate of mixing can be controlled, and temperature is the same in all microwells. Other mixers are more laborious to use and temperature is not reproducible from one well to the next. The two following shakers work well and are recommended:
9.19.18 Supplement 21
Current Protocols in Cytometry
(1) IKA-Minishaker MS 1 (IKA-Works) is not temperature controlled, and was placed in a incubator at 37°C. (2) iEMS incubator/shaker (Thermo Labsystems) contains different shaking levels and microtiter plates are temperature controlled.
MΦs also after phagocytosis. An exception is cells from patients with paroxysmal nocturnal hemoglobinuria (PNH), who lack CD14. For these patients, phagocytosing MΦs cannot be discriminated from PMNs by staining for CD14.
Incubation time The rate of phagocytosis is astonishingly high. Non-lymphocytes (PMNs + MΦs) phagocytose ∼80 S. aureus bacteria each within 15 min, and engulf ∼6 to 9 OMV-beads in 7.5 min. The phagocytic index increases with increasing incubation time, but if the incubation time is prolonged beyond the point when all targets are phagocytosed, a diminished rate of phagocytosis may be missed. When microspheres are used, the phagocytic index can be accurately calculated from the fluorescence values immediately from the start of incubation. A short incubation time, e.g., 5 min, might seem useful because at that point the rate of phagocytosis is maximal. Short incubation times pose problems for Maaløe’s method. With short incubation times, only a small fraction of the bacteria and zymosan particles are phagocytosed, leaving only a small difference between the initial and final counts of free targets. Such a small difference between target/phagocyte ratios is hard to measure by Maaløe’s method, and is quantitatively unreliable. In the author’s hands, the minimum incubation time with initial target/phagocyte ratio of 20:1 is 7.5 min. When target/phagocyte ratio is 10:1, the differences in the initial ratios and those following phagocytosis become prominent after 7.5 min.
Digestion The measurement of digestion relies on fluorescence of liberated targets. Therefore, target concentrations and incubation times must be chosen so that most of the targets are internalized. Two to three targets adhere to cell nuclei after exposure to the detergent NP40 in the Vindeløv’s solution. Because the number of adherent particles is unknown, the fluorescence of these targets cannot be used to measure digestion.
Signal collection A defined collection time minimizes the drift in fluorochrome distribution and fluorescence intensity between samples. Collect data for 30 sec. Signals are typically recorded from 1500 neutrophils, 100 to 200 monocytes, and 24,000 microspheres. Sample sequence In order to utterly reduce possible drift, analyze the contents of two parallel microwell series in opposite sequence. Discrimination of PMNs from MΦs The phagocytosis of microspheres increases the SS of MΦs. Therefore, phagocytosing MΦs cannot be discriminated from phagocytosing PMNs by light scatter alone. Staining with antiCD14-PE-Cy5 discriminates PMNs from
Troubleshooting Hemolysis Lysis of red blood cells is mandatory since red blood cells interfere with measurements by blocking the leukocyte access to targets. If the pelleted leukocyte suspension is red after the first NH4Cl lysis, lysis should be repeated. Cell clumping Earlier methods relied on Hank’s buffered salt solution supplemented with albumin, or other buffers. Buffers containing Ca2+ and Mg2+ promote leukocyte aggregation. Cell clumping is overcome by washing leukocytes in DPBS-GA. Note that Ca2+ and Mg2+ are omitted from the solution. Cells can be kept in DPBS-GA on ice for a few hours without affecting phagocyte functions. Cell clumping may occur during phagocytosis. If so, the number of aggregates can be measured using FS versus SS. The percentage of cell clumps should be the same before and after phagocytosis. Results should be discarded if the fraction of doublets and clumps before and after phagocytosis is not of the same order of magnitude. Adhesion to the sample flow lines in the flow cytometer Adhesion of targets and leukocytes to the lines of the sample flow system is minimized by diluting the suspensions in EDTA and by keeping the samples on ice until the measurement. Microspheres, E. coli, S. typhi, Str. epidermidis, S. aureus, and zymosan particles do not adhere. However, some bacteria, e.g.,
Studies of Cell Function
9.19.19 Current Protocols in Cytometry
Supplement 21
meningococci, adhere to sample flow lines. This causes a loss of targets as they pass through the fluidics system, and decreases the measured target/phagocyte ratio. The phenomenon is observed as a target/phagocyte ratio higher after phagocytosis than before. Percentage phagocytosis can still be measured accurately, but not the phagocytic index. The problem can be lessened and possibly eliminated by diluting the samples to avoid coincidence, and increasing the sample flow rate. Controls using microscopic counts are required. Spontaneous phagocytosis Some albumin batches are contaminated, in particular, with endotoxin that can elicit phagocytosis and oxidative burst, and affect intracellular processing. To avoid this problem, purchase a small amount of albumin, record the batch number, and test the albumin in the absence of serum to assess unwarranted phagocytosis and oxidative burst. Compare phagocyte FITC fluorescence after the phagocytosis of FITC-labeled targets in the absence and presence of opsonins. If the batch works as expected, purchase and use this albumin batch. Spontaneous oxidative burst DHR in solution is slowly converted to R123. If control neutrophils that have not been phagocytosing microspheres stain green, indicating R123 fluorescence, this staining is probably due to the conversion of DHR to R123 before phagocytosis. Make a fresh, new DHR solution.
Assessment of Phagocyte Functions by Flow Cytometry
Trypan blue quenching Several reports in the literature concern inability to quench FITC fluorescence with trypan blue. Different batches of trypan blue— even from the same company source—have been reported to vary in quenching ability. This problem is related not to trypan blue, but to the relative concentration of trypan blue to the intensity of target FITC staining. Therefore, the staining procedure is standardized using a defined optical density of the bacteria to be FITC stained. The ability of trypan blue to quench target FITC fluorescence is easily monitored by flow cytometry. If effective quenching of the targets cannot be performed, targets should be stained less intensively. The FITC staining protocol given above allows effective quenching of FITC-labeled bacteria and zymosan particles by trypan blue. Another problem with trypan blue is clumps. Trypan blue clumps should be first
removed by a crude filter followed by sterile filtration (0.22-µm pore size). Phagocytic index This problem regards the differential count of leukocytes and targets. When the target/phagocyte ratio is >20:1, the relative counts of leukocytes and targets become inaccurate. For example, with target%/leukocyte%, one may have a 96%:4% ratio of 24, a 95%:5% ratio of 19, and a 94%:6% ratio of 15.7. Thus, a small measured difference in the percentage of targets from 96% to 94% shifts the (apparent) target load from 24:1 to 15.7:1. This change in target load profoundly affects phagocyte fluorescence. Accuracy is increased by lowering the ratio of targets to phagocytes. Therefore, the determination of the concentration of targets should be based not on a single measurement, but repeated measurements of target/leukocyte ratios using several target dilutions. Using a low target/phagocyte ratio diminishes the strain on the phagocytes, and increases the danger of statistical type 1 error. With small targets such as bacteria and microspheres of similar size, a target/phagocyte ratio of 20:1 is useful for many purposes. No phagocytosis Proteins and other chemicals can adhere to various tubes used for reagent storage. Use polypropylene tubes that do not absorb essential ingredients such as opsonins.
Anticipated Results Percentage phagocytosis The measurement of the percentage of phagocytosing leukocytes was introduced by Hamburger in 1927, and has since been a parameter of choice (Bassøe et al., 1983). Quadrant statistics must be set with care to monitor percentage phagocytosis. Usually, phagocytes are at least five times more fluorescent than free targets. Non-phagocytosing leukocytes are less fluorescent than individual free targets. Therefore, when the vertical quad-stat line is set just above the fluorescence of the targets, phagocytes are clearly discriminated from nonphagocytes. Since non-phagocytosing leukocytes do not fluoresce at all, the discriminator monitors an all-or-none phagocytosis response. The all-or-none discriminator is useful for the monitoring of receptor deficiencies and lack of receptor function. About 96±3% of peripheral blood neutrophils phagocytose S. aureus, zymosan particles, and OMV-beads and in
9.19.20 Supplement 21
Current Protocols in Cytometry
HYP, PHS, and REC, respectively. Percentage phagocytosis drops to ∼70% in REC-IN with OMV-beads as target and is 0% to 5% in HYPIN and with zymosan particles as target. Percentage phagocytosis of S. aureus depends on the PHS. It varies between 20% and 70% in PHS-IN, but is stable for a long time for any one PHS batch. This implies that complement and IgG alone sustain phagocytosis by ∼96±3% and 70% of the peripheral blood neutrophils. Percentage phagocytosis by MΦs is ∼70% to 75% with S. aureus, zymosan particles, and OMV-beads in HYP, PHS, and REC, respectively. The value drops to the order of 5%, 10% to 50%, and 20% in HYP-IN, PHS-IN, and APS, respectively. These results are obtained in spite of the regular observation that all PMNs from the same suspensions carry FcγRIII (CD16) and FcγRII (CD32), and all the MΦs carry CD32 and FcγRI (CD64). Under the same conditions none of the PMNs carries CD64, and MΦs do not express CD16. Thus, there is an expected discrepancy between receptor expression and the ability to phagocytose by the same receptors. Phagocyte subpopulations One advantage of using fluorescent microspheres is their low CV of ∼1.0, as compared to the CV of FITC-labeled bacteria or zymosan particles, which is ∼30% to 40%. The latter high CV makes impossible the accurate discrimination of subpopulations of phagocytes differing in fluorescence, whereas such a discrimination between cell subpopulations is possible with microspheres. When leukocytes and OMV-beads are incubated with inactive serum, the neutrophils do not associate with OMV-beads. In REC, nearly all neutrophils are phagocytosing. The percentage oxidative phagocytosis and the phagocytic index vary within narrow limits. The difference between percentage phagocytosis and percentage oxidative burst, i.e., percentage in region 4 minus percentage in region 1, is of the order 0.6%. The fraction of non-phagocytosing neutrophils is ∼1.5%. The fluorescence of the OMV-beads is stable for several years. In APS, monocytes are mainly localized to region 1. In REC, the majority of monocytes are observed in region 2. Very few MΦs are resting, i.e., in region 3. With APS and in the presence of DHR, PMNs do not display green fluorescence. Phagocytosis is triggered in REC, and a single population is observed.
In the absence of DHR, monocytes do not fluoresce in the green. When incubated with DHR 7.5 min at 37°C, monocytes become green fluorescent even in the absence of serum, OMV-beads, mixing, and antiCD14. The percentage of monocytes in region 1 is 89±9%. AntiCD14 does not influence the monocyte R123 fluorescence. The oxidative ratio is a derived parameter that informs on the conversion of DHR to R123 per ingested bead. MΦs start with a higher R123 fluorescence than do neutrophils. With increasing incubation time, the monocyte oxidative ratio declines, whereas that of PMNs increases. After 15 min, the values are of the same order of magnitude in both cell types. Chemotaxis The controls contain lymphocytes, monocytes, and neutrophils in proportion to the peripheral blood counts. Following spontaneous migration and chemotaxis, only neutrophils are observed in the FS versus SS cytograms. Expected spontaneous migration is ∼30% to 50%, and chemotaxis 60% to 90%. Chemotaxis values depend on the concentration and activity of ZAS and other chemotactic agents (IL-8, fMLP, LTB4, and others).
Time Considerations The whole procedure, i.e., Basic Protocols 1 to 4, is a demanding 1-day workload. Four parallel samples can be investigated in 1 day, but it is advisable to select those parts of the individual protocols that are suited for the purpose at hand. In Basic Protocol 1, e.g., a suitable bacteria strain can be selected for study, perhaps from a patient, and zymosan particles and microspheres can be ignored. Likewise, complement receptors and serum complement opsonic activity can easily and rapidly be investigated by using zymosan particles alone. Microspheres are useful for the investigation of specific antigens, opsonins, and respiratory burst. However, oxidative burst can also be studied using unstained, non-fluorescent microorganisms and zymosan particles. After the addition of the sample to Vindeløv’s solution, the EB and FITC staining of targets are stable at least 2 to 3 days in the dark at 4°C. Flow cytometric investigations can therefore be delayed.
Studies of Cell Function
9.19.21 Current Protocols in Cytometry
Supplement 21
Literature Cited Bassøe, C.-F., Laerum, O.D., Glette, J., Hopen, G., Haneberg, B., and Solberg, C.O. 1983. Simultaneous measurement of phagocytosis and phagosomal pH by flow cytometry: Role of polymorphonuclear neutrophilic leukocyte granules in phagosome acidification. Cytometry 4:254-262. Bassøe, C.-F. 1984. Processing of Staphylococcus aureus and zymosan particles by human leukocytes measured by flow cytometry. Cytometry 5:86-91. Bassøe, C.-F. 2000. Flow cytometric quantitation of phagocytosis in acute myelogenous leukemia. Acta Haematol. 102:163-171. Bassøe, C.-F., Smith, I., Sørnes, S., Halstensen, A., and Lehmann, A.K. 2000. Concurrent measurement of antigen- and antibody-dependent oxidative burst and phagocytosis in monocytes and neutrophils. Meth. Enzymol. 21:203-220.
Cantinieaux, B., Hariga, C., Courtoy, P., Hupin, J., and Fondu, P. 1989. Staphylococcus aureus phagocytosis. A new cytofluorometric method using FITC and paraformaldehyde. J. Immunol. Methods 121:203-208. Hurst, J.K., Albrich, J.M., Green, T.R., Rosen, H., and Klebanoff, S. 1984. Myeloperoxidase-dependent fluorescein chlorination by stimulated neutrophils. J. Biol. Chem. 259:4812-4821. Lehmann, A.K., Halstensen, A., Holst, J., and Bassøe, C.-F. 1997. Functional assays for evaluation of serogroup B meningococcal structures as mediators of human opsonophagocytosis. J. Immunol. Methods 200:55-68.
Contributed by Carl-Fredrik Bassøe Haukeland University Hospital Bergen, Norway
Assessment of Phagocyte Functions by Flow Cytometry
9.19.22 Supplement 21
Current Protocols in Cytometry
Flow Cytometric Analysis of Calcium Mobilization in Whole-Blood Platelets Flow cytometry (FCM) has been shown to be a promising technique for the analysis of platelet phenotype and function, including platelet activation; however, the technical issues and choice of adequate biological parameters are still controversial. Most FCM work on platelet function is performed on fixed whole-blood or purified platelet preparations, which precludes the possibility of either studying the earliest biochemical events involved in platelet activation or performing such studies in the physiological milieu.
UNIT 9.20
BASIC PROTOCOL
FCM provides a convenient method to evaluate platelet activation by following the kinetics of intracellular free Ca2+, using sensitive fluorescent indicators that can be loaded into intact cells. Moreover, in the clinical setting, whole-blood techniques have obvious advantages to avoid artifactual platelet activation and allow the maintenance of nearphysiological conditions. This unit describes a fast and sensitive flow cytometric procedure using the Ca2+-sensitive dye fluo-3 AM and the platelet-specific antibody CD41-PE to determine the kinetics of intracellular Ca2+ mobilization in whole-blood platelets with minimal manipulation of the samples. This technique may be applied to reveal fast and transient increases in cytosolic calcium upon platelet stimulation with the agonists ADP and thrombin. In brief, whole blood is collected and diluted in modified Tyrode’s buffer. Platelets are loaded with the calcium indicator fluo-3 acetomethyl ester (fluo-3 AM), stained with phycoerythrin (PE)-conjugated monoclonal antibody against the platelet-specific constitutive antigen CD41 (GPIIb/IIIa complex), and if thrombin is being used as the agonist, treated with the tetrapeptide GPRP to inhibit fibrin polymerization. The double-stained whole blood is then stimulated with platelet agonists and the real-time changes in fluo-3 fluorescence are analyzed. This protocol provides a simple and sensitive tool to assess, in vitro, the time course and intensity of signal-transduction responses to different platelet agonists under near-physiological conditions, and should be broadly applicable to studies of platelet reactivity. Materials Rested donor free from antiplatelet treatment 135 mM tribasic sodium citrate, pH 7.4 Modified Tyrode’s buffer (see recipe) 1 mM fluo-3 AM (see recipe) CD41-PE monoclonal antibody (Immunotech) 5 mg/ml glycyl-L-prolyl-L-arginyl-L-proline (GPRP; see recipe); for thrombin activation studies Agonist (e.g., human α-thrombin, ADP; see recipe) 19-G needle 12 × 75–mm polypropylene tubes Flow cytometer with 488-nm excitation and band-pass filters centered at 525 (green) and 575 nm (orange) NOTE: Use new or carefully washed polypropylene tubes in all steps to avoid platelet activation on glass or polystyrene surfaces. Studies of Cell Function Contributed by Maria-do-Céu Monteiro, Maria-José Gonçalves, Filipe Sansonetty, and José-Enrique O’Connor
9.20.1
Current Protocols in Cytometry (2003) 9.20.1-9.20.8 Copyright © 2003 by John Wiley & Sons, Inc.
Supplement 24
NOTE: The reader is encouraged to read UNIT 6.10 for more detail on platelet handling and UNIT 9.8 for data analysis. 1. Perform a blood draw from a rested donor (15 to 30 min) without tourniquet from an antecubital vein using a 19-G needle in order to avoid artifactual platelet activation. Discard the first 2.5 ml and collect the next 5 ml of whole blood (WB) directly into a 12 × 75–mm polypropylene tube containing 0.5 ml of 135 mM tribasic sodium citrate to prevent coagulation. Mix gently by inversion. In general, platelet functional assays must be performed after an overnight fast. Confirm that normal donors have not received any antiplatelet treatment, such as aspirin, within the preceding 11 days.
Load platelets with fluo-3 AM 2. Dilute anticoagulated WB 1:10 in modified Tyrode’s buffer (see recipe) and distribute in 500-µl aliquots. Dilution of whole blood must be done within 1 hr.
3. To each 500-µl aliquot, add 2.5 µl of 1 mM fluo-3 AM (5 µM final). Incubate 15 min at 37°C in the dark. Immunostain platelets with CD41-PE monoclonal antibody 4. In fresh tubes, mix 5 µl CD41-PE monoclonal antibody and 25 µl WB loaded with fluo-3 AM. CD41-PE is used to identify the platelet population in WB (Fig. 9.20.1).
5. Optional: If thrombin or calcium ionophores are the agonists which will be used or the assay is performed in calcium-containing buffer, add 5 µl of 5 mg/ml glycyl-Lprolyl-L-arginyl-L-proline (GPRP; 2.5 mM final). GPRP is used to inhibit fibrin polymerization.
6. Incubate 15 min at room temperature in the dark. 7. Dilute with 1 ml modified Tyrode’s buffer. From each diluted sample, 500 ìl will be used to determine kinetic variations of intracellular Ca2+.
Collect and analyze data 8. Set up the flow cytometer to collect forward-angle light scatter (FS), side-angle light scatter (SS), green fluorescence (525 nm; i.e., fluo-3 AM), and orange fluorescence (575 nm; i.e., PE) intensities. Set a stop-time condition at 300 sec. 9. Define correlated histograms and listmode data for logarithmically amplified signals (FS, SS, fluo-3, and PE). 10. For each plot, define rectangular analysis regions over the axis, each covering the whole length of the fluorescence axis, to determine mean responses versus time (Fig. 9.20.1). 11. Run 500 µl of WB double stained with fluo-3 and anti-CD41-PE (step 7) as a baseline control tube. Calcium Mobilization in Whole-Blood Platelets
12. Adjust cytosettings and define analytical gates based upon logarithmic FS and SS signals, and CD41+ events to visualize and select the platelet population (Fig. 9.20.1). Stop the tube and save the adjustments for further analysis in each individual subject.
9.20.2 Supplement 24
Current Protocols in Cytometry
C
1000
1244
1000
1000
1 .1
1
10
100
1000
.1
1
1
2
3
4
1000
1
2
3
4
A
0
B
C
D
75
E
150
Time
F
225
300
CD41-PE
0
.1
.1
1
1
1
(120 × 120)
100 10
CD41-PE
(120 × 120)
100 10
100
F 1000
E
10
CD41-PE
100
D
1000
CD41-PE
1000
Side scatter
Fluo-3 fluorescence
4
(128 × 128)
100
3
10
10
2
.1
c
0 1
10
Side scatter
933 622
Count
1
311
10
Forward scatter
.1 .1
1
100
B
100
A
.1
1
10
100
Fluo-3 fluorescence
1000
.1
1
10
100
1000
Fluo-3 fluorescence
Figure 9.20.1 Criteria for identifying and gating the platelet population in whole-blood samples. (A) Light-scatter properties of whole blood (WB). Single platelets (arrow) exhibit low forward and side scatter in log-amplified dot plots. (B) Histogram of a whole-blood sample immunostained with CD41-PE. Region within cursor C may be used to gate the platelet population. Notice that depending on the experimental conditions, CD41+ events may include single platelets, aggregated platelets, and platelets coincident with or bound to erythrocytes and leukocytes. (C) Log-amplified dot plot showing morphological properties of platelet subpopulations positive for CD41. Single platelets can be specifically gated as indicated by the arrow. Panel D shows the procedure and time course of a typical kinetic assay of intracellular Ca2+ mobilization in a normal donor, while panels E and F show the confirmation of the specificity of intracellular Ca2+ mobilization following activation of whole blood with platelet agonists. Specifically, the collection of baseline fluorescence for 20 sec with a pause (arrow) for addition of agonist (125 µM ADP), after which acquisition is resumed, is shown. The vertical lines crossing the plot are analysis regions defined on the time axis (1,024 channels = 300 sec) and covering the full length of the fluorescence axis. The data correspond to a single-platelet population according to the gating strategy shown in panel C. Intracellular Ca2+ in CD41+ and CD41− events in whole blood, which is either unstimulated or stimulated for 20 sec with 1 U/ml thrombin, are shown in panels E and F, respectively. Only CD41+ events show a significant increase in fluo-3 fluorescence.
13. Run 500 µl of sample (step 7) and acquire baseline fluo-3 fluorescence in the platelet population for ∼20 sec. Briefly pause sample acquisition and quickly add 25 µl agonist at appropriate concentration. Resume acquisition immediately, and record changes in log green fluorescence. Careful timing is absolutely necessary to ensure identical conditions. The addition of agonist must be performed in the minimal possible time and the same time interval must be maintained for subsequent measurements. Studies of Cell Function
9.20.3 Current Protocols in Cytometry
Supplement 24
14a. For ADP-treated samples: Run a blank tube (i.e., unstained WB; step 7) between successive samples in order to remove residual agonist from the sample line. 14b. For thrombin-treated samples: Flush sample lines with 0.1% bleach to ensure that all protein is removed, followed by unstained diluted whole blood. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
ADP sodium salt, 2 mM Prepare a stock solution by dissolving sodium adenosine 5′-diphosphate (ADP; 427.2 g/mol; Sigma) in water to a concentration of 2 mM. Store up to 6 months in convenient aliquots at −20°C. The day of the experiment, dilute in modified Tyrode’s buffer (see recipe) to 100 µM (1 mM). Fluo-3 AM, 1 mM Dissolve fluo-3 acetoxymethyl ester (fluo-3 AM; 1130 g/mol; Molecular Probes) at 1 mM in dimethylsulfoxide (DMSO). Store up to 6 months at −20°C Dye may be purchased in 50-ìg aliquots to avoid aliquoting or refreezing.
Human α-thrombin Prepare a stock solution by dissolving 250 U human α-thrombin (Sigma) in 1 ml water. Store in convenient aliquots up to 6 months at −20°C. The day of the experiment, dilute in modified Tyrode’s buffer (see recipe) to 0.1 to 20 U/ml. GPRP, 5 mg/ml Dissolve GPRP (Sigma) in sterile water at 5 mg/ml. Store in convenient aliquots up to 6 months at −20°C. Modified Tyrode’s buffer 8.006 g NaCl (137 mM final) 0.209 g KCl (2.8 mM final) 0.203 g MgCl2⋅6H2O (1 mM final) 1.008 g NaHCO3 (12 mM final) 0.071 g Na2HPO4⋅2 H2O (0.4 mM final) 2.383 g HEPES (10 mM final) Adjust pH to 7.4 with 0.1 N NaOH Adjust volume to 1 liter with water Pass solution through a 0.20-µm filter Store up to 1 month at 4°C Just before use, add 0.991 g glucose (5.5 mM final) and 3.5 g BSA (0.35% final). Discard after use.
COMMENTARY Background Information
Calcium Mobilization in Whole-Blood Platelets
Platelet activation with most, if not all, agonists results in a rise of free Ca2+, which appears to be due to both influx across the plasma membrane and release from internal stores, mainly the dense tubular system (Johnson et al., 1985; Sage et al., 1989; Salzman and Ware, 1989; Ware et al., 1996). The increase of cytosolic free Ca2+ is a key early
event which follows platelet stimulation and precedes several activation responses, including shape change, aggregation, secretion, and expression of procoagulant activity (Rink et al., 1982; Salzman and Ware, 1989; Weiss and Lages 1997). Conversely, many platelet functions require a minimal threshold level of intracellular Ca2+ for their occurrence (Salzman and Ware, 1989).
9.20.4 Supplement 24
Current Protocols in Cytometry
Several studies using fluorescent Ca2+-sensitive indicators and FCM in samples of washed or gel-filtered platelets have been reported (Davies et al., 1988, 1989; Jennings et al., 1989; Merrit et al., 1990; Fijnheer et al., 1991; Kyle et al., 1995), although the whole-blood approach has obvious advantages. In the protocol presented in this unit, sample manipulation is minimal, in contrast to other methods involving separation procedures that cause artifactual platelet activation and eventual loss of platelet subpopulations. Whole-blood platelets are analyzed in a more physiologically representative milieu, where erythrocytes and leukocytes are present and may affect platelet responses to activation (Santos et al., 1986; Larsen et al., 1989; Marcus et al., 1995). In this experimental condition, platelet population can be clearly and easily differentiated from other blood cells on the basis of their light-scatter profile and their labeling by a PE-conjugated monoclonal antibody against a specific antigen of the platelet surface. Work in the authors’ laboratory using the proposed protocol (Monteiro et al., 1999) agrees with previous studies by other groups (Davies et al., 1989; Jennings et al., 1989; Sage et al., 1989; Salzman and Ware, 1989). These reports showed a very fast and transient rise in intracellular free Ca2+ in platelets stimulated with thrombin or ADP, with the time course and intensity of the response being heterogeneous, but dependent on the type and dose of the agonist. Heterogeneity of platelet response to agonists is an important aspect to be considered. The kinetic assay presented here quantifies real-time intracellular Ca2+ mobilization in single platelets, thus allowing one to detect and compare platelet subpopulations differing in their response, including transient changes.
Critical Parameters and Troubleshooting In general, platelet functional assays are susceptible to some variability due to traumatic venous puncture, use of tourniquet, delay in sample process, and manipulation. Thus, these factors must be carefully controlled to maintain platelets in a resting state. The clearest limitations of this technique results from the difficulty in calibrating [Ca2+]i, for reasons similar to those described by other authors in other experimental models (Jennings et al., 1989). Such calibration in the intact cells demands assumptions that cannot be easily
proved, namely that hydrolysis of fluo-3 AM ester to the acid form is complete, that microviscosity and cytoplasm refractive index do not affect the Kd of fluo-3, and that the ionophore used in [Ca2+]i measurement equilibrates totally with all intracellular Ca2+ stores. However, if the objective of the experiment is to compare response levels, it is possible to normalize the results by determining the fluorescence ratio (Rfluo-3) between fluo-3 fluorescence in platelets stimulated by agonist with increased free Ca2+ and fluo-3 fluorescence of unstimulated platelets—i.e., Rfluo-3 = (agonist-stimulated platelet mean fluorescence)/(unstimulated platelet mean fluorescence). In this way, differences in basal values of fluo-3 fluorescence are minimized and different results can be compared adequately (Figs. 9.20.2 and 9.20.3). Fluorochrome concentration should be titrated to establish the minimum that allows sensitive detection of free Ca2+ changes. Besides other possible intrinsic toxic effects, fluo3 is a Ca2+ chelator and for that reason, an increase of fluo-3 concentration inside the cell may decrease the availability of this second messenger, thus interfering with functional responses that might be intended to be simultaneously evaluated. In such nonsaturating concentrations, fluo-3 accumulation varies according to the relationship of platelet/dye concentration. When whole-blood samples are used, another difficulty is to control the exact relationship of platelet concentration/dye concentration, since leukocytes can incorporate the probe and the absolute cell count is variable among individuals, even in donors having normal counts. The approach used consisted of titrating the dye concentration for a given whole-blood dilution in individuals with normal blood cell count. Although incubation of whole blood with fluo-3 AM results in the loading of platelets as well as other blood cells, appropriate gating conditions and the use of platelet-specific agonists ensure that the mobilization of intracellular Ca2+ is detected only in single platelets, and, conversely, that platelets are the only cells responding to the agonist (see Fig 9.20.1C). Platelet identification in whole blood is achieved using a PE-conjugated monoclonal antibody against a specific platelet glycoprotein, such as CD41, CD61, or CD42b. Some antibodies against platelet surface glycoproteins may induce activation, thus eventually changing intracellular Ca2+. Another critical aspect of the whole-blood technique is to avoid fibrin polymerization,
Studies of Cell Function
9.20.5 Current Protocols in Cytometry
Supplement 24
Fluorescence ratio
A
B
C
5
5
5
0.005 U/mL thr
U/mL thr 0.05
D 0.5 U/mL thr
5
4
4
4
4
3
3
3
3
2
2
2
2
1
1
1
1
0
0
50 100 150
E 5
0
0
50 100 150
5
0
50 100 150
G
F 5 µM ADP
0
µM ADP 12.5
5
0
1 U/mL thr
0
H µM ADP 25
5
4
4
4
4
3
3
3
3
2
2
2
2
1
1
1
1
0
0
50 100 150
0
0
50 100 150
0
0
50 100 150
50 100 150
0
50 µM ADP
0
50 100 150
Time
Figure 9.20.2 Examples of concentration-response relationships in Ca2+ mobilization in whole-blood platelets stimulated with thrombin (thr), a strong agonist (panels A to D), or ADP, a weak agonist (panels E to F). The experimental conditions are as described in this unit. Data points in each plot represent mean values (n = 5) of fluo-3 fluorescence ratios (Rfluo-3 = agonist-stimulated platelet mean fluorescence/unstimulated platelet mean baseline fluorescence) for each analysis region defined in the time axis of the corresponding histogram. The data correspond to a single-platelet population, according to the gating strategy shown in Figure 9.20.1C. Time scale is in seconds.
especially if the agonist is thrombin. This methodological difficulty can be solved by using the tetrapeptide GPRP, initially described by Michelson (1991). Since the buffer used in this protocol is essentially Ca2+ free, Ca2+ responses are the consequence of mobilization from internal stores. To analyze Ca2+ influx, a buffer containing Ca2+ must be used and in this case GPRP must be added prior to the first dilution of whole blood. GPRP binds to and blocks the fibrinogen and fibrin polymerization sites (Achyuthan et al., 1986).
Anticipated Results Calcium Mobilization in Whole-Blood Platelets
The applicability of this method to the study of signal transduction in platelets has been assessed on activation induced by agonists differing both in the mechanisms of signal
transduction and the intensity and duration of Ca2+ mobilization (Monteiro et al., 1999). Thus, ADP and thrombin (Fig. 9.20.2) may be used as weak or strong agonists, respectively (Benett and Kolodziej, 1992). Under these experimental conditions, the time- and concentration-dependence of platelet response to the agonists, as well as the effect of conditions affecting platelet function, may be determined quantitatively, as shown in Figures 9.20.2 and 9.20.3. Thrombin induces a fast and dose-dependent elevation in free intracellular Ca2+, attaining a value higher than four times the basal level. Thereafter, Ca2+ decreases to almost the resting level. ADP also induces fast and transient increases in cytosolic free Ca2+ level, although of lesser intensity than those induced by thrombin.
9.20.6 Supplement 24
Current Protocols in Cytometry
A
B 12.5 µM ADP
Fluorescence ratio
2
0.5 U/mL thrombin
6 5 4 3
1 2 before after ticlopidine
before after ticlopidine
1
0
0 0
100
200
0
100
200
Time
Figure 9.20.3 Application of the protocol in a clinical setting, as shown by the effect of ticlopidine, an antiplatelet drug, on the mobilization of intracellular Ca2+ in platelets activated with 12.5 µM ADP (panel A) or 0.5 U/ml thrombin (panel B). Ticlopidine was administered orally to healthy volunteers (39.6 ± 6.9 year old) at a dose of 250 mg twice a day for 4 days. The experimental conditions are as described in the unit. Data points in each plot represent mean values (n=5) of fluo-3 fluorescence ratios (Rfluo-3 = agonist-stimulated platelet mean fluorescence/unstimulated platelet mean baseline fluorescence) for each analysis region defined in the time axis of the corresponding histogram. Time scale is in seconds.
Time Considerations Since time delay in sample processing is a limitation in this technique, one should always set up the necessary tubes and reagents before obtaining the blood sample. Loading whole blood with fluo-3 AM takes 15 min and staining with CD41-PE also takes ∼15 min. For each kinetic assay, cells should be analyzed continuously for a period of 3 to 5 min.
Literature Cited Achyuthan, K.E., Dobson, J.V., and Greenberg, C.S. 1986. Gly-Pro-Arg-Pro modifies the glutamine residues in alpha- gamma-chains of fibrinogen: Inhibition of transglutaminase cross-linking. Biochem. Biophys. Acta 872:261-268. Bennett, J.S. and Kolodziej, M.A. 1992. Disorders of platelet function. Dis. Mon. 38:577-631. Davies, T.A., Drotts, D., Weil, G.J., and Simons, E.R. 1988. Flow cytometric measurements of cytoplasmic calcium changes in human platelets. Cytometry 9:138-142. Davies, T.A., Drotts, D.L., Weil, G.J., and Simons, E.R. 1989. Cytoplasmatic Ca2+ is necessary for thrombin-induced platelet activation. J. Biol. Chem. 264:19600-19606.
Fijnheer, R., Homburg, C.H.E., Hooibrink, B., Boomgaard, M.N., Korte, D., and Ross, D. 1991. Loss of thrombin-induced Ca2+ mobilization in a subpopulation of platelets during storage. Thromb. Haemost. 66:350-354. Jennings, L.K., Dockter, M.E., Wall, C.D., Fox, C.F., and Kennedy, D.M. 1989. Calcium mobilization in human platelets using indo-1 and flow cytometry. Blood 74:2674-2680. Johnson, P.C., Ware, J.A., and Cliveden, P.B. 1985. Measurement of ionized calcium in blood platelets with photoprotein aequorin. Comparison with quin-2. J. Biol. Chem. 260:2069-2076. Kyle, P.M., Jackson, M.C., Buckley, D.C., Swiet, M., and Redman, C.W.G. 1995. Platelet intracellular free calcium response to arginine vasopressin is similar in preeclampsia and normal pregnancy. Am. J. Obstet. Gynecol. 172:654660. Larsen, E., Celi, A., Gilbert, G.E., Furie, B.C., Erban, J.K., Bonfanti, R., Wagner, D.D., and Furie, B. 1989. PADGEM protein: A receptor that mediates the interaction of activated platelets with neutrophils and monocytes. Cell 59:305-312. Marcus, A.J., Safier, L.B., Broekman, J.B., Islam, N., Fliesssbach, J.H., Hajjar, C.A., Kaminski, W.E., Jendraschak, E., Silverstein, R.L., and Schacky, C. 1995. Thrombosis and inflammation as multicellular processes: Significance of cellcell interactions. Thromb. Haemost. 74:213-217.
Studies of Cell Function
9.20.7 Current Protocols in Cytometry
Supplement 24
Merrit, J.E., McCarthy, S.A., Davies, M.P.A., and Moores, K.E. 1990. Use of fluo-3 to measure cytosolic Ca2+ in platelets and neutrophils. Biochem. J. 269:513-519.
Santos, M.T., Vallés, J., Aznar, J., and Requejo, J.L.P. 1986. Role of red blood cells in early stages of platelet activation by collagen. Thromb. Haemost. 56:376-381.
Michelson, A.D., Ellis, P.A., Barnard, M.R., Matic, G.B., Viles, A.F., and Kestin, A.S. 1991. Downregulation of platelet surface glycoprotein Ib-IX complex in whole-blood stimulated by thrombin, ADP or an in vivo wound. Blood 77:770-779.
Ware, J.A., Johnson, P.C., Smith, M., and Salzman, E.W. 1996. Effect of common agonists on cytoplasmatic ionized calcium concentration in platelets. J. Clin. Invest. 77:878-886.
Modderman, P.W., Huisman, H.G., van Mourik, J.A., von dem Borne, A.E. 1988. A monoclonal antibody to the human platelet glycoprotein IIb/IIIa complex induces platelet activation. Thromb. Haemost. 60:68-74. Monteiro, M.C., Sansonetty, F., Gonçalves, M.J., and O’Connor, J.E. 1999. Flow cytometric kinetic assay of calcium mobilization in wholeblood platelets using fluo-3 and CD41. Cytometry 35:302-310. Rink, T., Smith, S., and Tsien, R. 1982. Cytoplasmic free Ca2+ threshold and Ca2+-independent activation for shape-change and secretion. FEBS Lett. 148:21-26. Sage, S.O., Merritt, J.E., Hallam, T.J., and Rink, T.J. 1989. Receptor-mediated calcium entry in fura2-loaded human platelets stimulated with ADP and thrombin. Dual wavelength studies with Mn2+. Biochem. J. 258:923-926. Salzman E.W. and Ware, J.A. 1989. Ionized calcium as an intracellular messenger in blood platelets. Prog. Haemost. Thromb. 9:177-202.
Weiss, H.J. and Lages, B. 1997. Platelet prothrombinase activity and intracellular calcium response in patients with storage pool deficiency, Glycoprotein IIb-IIIa deficiency or impaired platelet coagulant activity: A comparison with Scott Syndrome. Blood 89:1599-1611.
Contributed by Maria-do-Céu Monteiro and Maria-José Gonçalves Instituto Politécnico de Saúde-Norte Paredes, Portugal Filipe Sansonetty Escola de Ciências de Saúde, Universidade do Minho Braga, Portugal IPATIMUP Porto, Portugal José-Enrique O’Connor Centro de Citometría y Citómica Universidad de Valencia Valencia, Spain
The authors are indebted to Mr. Alexandre Salvador for his excellent technical support. The authors also wish to acknowledge grants from PRODEP-Departamento de Ensino Superior and Izasa, S.A.
Calcium Mobilization in Whole-Blood Platelets
9.20.8 Supplement 24
Current Protocols in Cytometry
Flow Cytometric Analysis of Cytokine Responses in Stimulated Whole Blood: Simultaneous Quantitation of TNF-α-Secreting Cells and Soluble Cytokines
UNIT 9.21
The structural and functional integrity of the immune system in vertebrates is of crucial importance for their defense against infections and tumors as well as for the generation of self-tolerance. The multiple cellular and soluble elements that comprise the immune system combine in a single, coordinated response to carry out these functions. Cytokines are directly involved in the generation of such a coordinated response. In fact, the sequential production of many different cytokines is an early event occurring immediately after the activation of immune cells in response to inflammation or to a specific antigen. Coordinated secretion of cytokines—together with the expression and secretion of their receptors—acts as a modulator/regulator of the immune response by causing, among other events, activation or inhibition of effector cells. Typically, groups of cytokines are simultaneously secreted, and the soluble amounts of each secreted cytokine and the presence or absence of their receptors form a complex network of finely regulated interactions, which will result in different types of specific effector responses by the immune system. These typically include T cell–mediated antibody production and/or T cell cytotoxicity. Thus, evaluation of the pattern of cytokine production by a given cell population is often used to ascertain the functional status of the immune system. Several different methods are currently available to measure blood cell responses related to cytokine production in either a semi-quantitative or quantitative “bulk” way. The most commonly used techniques include assessment of mRNA transcripts by northern blot, RT-PCR, or in situ hybridization, and evaluation of protein secretion using different immunoassays (ELISA, western blot, or immunocytochemistry). Despite their high sensitivity, these approaches do not enable simultaneous quantitation of soluble cytokines and specific identification and characterization of the secreting cells. In order to link cytokine secretion to the activated cell populations that produce them, new alternative approaches have been developed to identify cytokine production at the single-cell level. Flow cytometry procedures for intracellular detection of cytokines are described in UNIT 9.9. Additional approaches aimed at the evaluation of cytokine production at the single-cell level include the ELISPOT and antibody matrix–based flow cytometric assays. Although these methods enable determination of which cells are actually producing a specific cytokine, they typically inhibit protein secretion or prevent diffusion into the extracellular medium, thereby altering the natural cytokine network. In addition, these approaches do not allow for the quantitative evaluation of the overall amount of cytokines produced. This unit describes a new flow cytometry–based method that allows for the simultaneous detection of cytokine-secreting cells and quantitation of multiple soluble cytokines. SIMULTANEOUS DETECTION OF CELL ACTIVATION AND QUANTITATION OF THE SECRETED CYTOKINES This Basic Protocol is performed using whole-blood samples—which assure the presence of all autologous blood cells and serum-soluble components—in order to produce the minimum alteration of the cellular microenvironment while cell activation and cytokine production in response to one or more stimuli are being evaluated.
BASIC PROTOCOL
Studies of Cell Function
Contributed by Arancha Rodríguez-Caballero, Andrés García-Montero, Clara Bueno, and Alberto Orfao
9.21.1
Current Protocols in Cytometry (2003) 9.21.1-9.21.21 Copyright © 2003 by John Wiley & Sons, Inc.
Supplement 25
Tumor necrosis factor α (TNF-α) expression was selected to identify those cells actively secreting cytokines since TNF-α is one of the most ubiquitous and earliest released cytokines. TNF-α-secreting cells are specifically detected through the use of a specific inhibitor of TNF-α-converting enzyme (TACE), which under normal conditions is responsible for the shedding of TNF-α from the membrane of those cells producing it. This synthetic matrix metalloprotease inhibitor does not affect the release of cytokines other than TNF-α, and the doses used still enable secretion of between 5% and 10% of all TNF-α produced by the stimulated cells. This assures maintenance of the natural cell-cytokine network and makes it possible to simultaneously quantify, in the same measurement, the cytokines secreted by using a specific, highly sensitive, flow cytometry–based bead immunoassay. Materials Human peripheral whole blood Vacutainer tubes containing heparin (Becton Dickinson) Cell culture medium (see recipe) 1 µg/ml phorbol myristate acetate (PMA; see recipe) 50 µg/ml ionomycin stock solution (see recipe) 1% (v/v) DMSO in PBS (APPENDIX 2A for PBS) 100% ethanol 2 mM TACE inhibitor (see recipe) Antibody-coated bead solutions: coated with anti-IFN-γ, anti-TNF-α, anti-IL-2, anti-IL-4, anti-IL-5, and anti-IL-10 MAbs (e.g., BD Biosciences) Fluorochrome-labeled monoclonal antibodies: Anti human TNF-α-PE (e.g., Mab 11, BD Biosciences) Anti human CD45-PE-Cy5 (e.g., J33 MAb, Immunotech) Anti human CD3-FITC (e.g., HIT3a, BD Biosciences) Anti human CD8-APC (e.g., B9.11, Immunotech) Mix of anti-cytokine (IFN-γ, TNF-α, IL-2, IL-4, IL-5, and IL-10) PE-conjugated MAbs (e.g., BD-CBA kit, BD Biosciences) Recombinant IFN-γ, TNF-α, IL-2, IL-4, IL-5, IL-10 cytokine standards (see recipe) and appropriate assay diluent (e.g., BD Biosciences) PBS (APPENDIX 2A) Flow cytometer calibration beads (unstained, FITC-, PE-, PE-Cy5-, and APC-stained BD CaliBRITE beads, BD Biosciences) and cytometer setup beads (including FITC- and PE-positive control detectors; BD-CBA kit, BD Biosciences) 12 × 75–mm round-bottomed polystyrene tubes (Falcon, BD) Humidified 37°C, 5% CO2 cell incubator Flow cytometer equipped with two lasers (emitting at 488 nm and at 630 nm) and with a filter set for detection of fluorescence emission at 530 ± 30, 585 ± 42, 661 ± 16, and ≥670 nm and appropriate software for instrument setup, data acquisition, and analysis Spreadsheet database software (e.g., Microsoft Excel) NOTE: It is important that all volumetric measurements be very precise. See UNIT 6.4 for reverse pipetting technique.
Analysis of Cytokine Responses in Stimulated Whole Blood
9.21.2 Supplement 25
Current Protocols in Cytometry
Prepare and stimulate blood samples 1. Collect peripheral blood (PB) samples in vacutainer tubes containing heparin (anticoagulant). 2. Dilute PB samples 1:5 (v/v) in cell culture medium and gently mix. Dilution of blood in cell culture medium can be as little as 1:1 (v/v) to increase the concentration of target cells if they are expected to be activated at very low frequencies/numbers.
3. Label two separate 12 × 75–mm polystyrene tubes as “stimulated” and “non-stimulated” and pipet 250 µl diluted PB into each one. 4. To the tube labeled “stimulated,” add 6.25 µl of 1 µg/ml PMA solution and 5 µl of 50 µg/ml ionomycin stock solution. The combination of PMA and ionomycin is used to produce a generic non-specific activation of T cells. For sample stimulation, concentrations of 25 ng/ml PMA plus 1 ìg/ml ionomycin are typically used.
5. To the tube labeled “non-stimulated,” add 6.25 µl of 1% DMSO and 5 µl of 100% ethanol. If the whole blood has been properly stored at room temperature under sterile conditions, PB lymphocytes can be successfully stimulated as long as 24 hr after collection, but it is better to keep a shorter period of time between sample draw and stimulation.
6. Add 2.5 µl of 2 mM TACE inhibitor solution to each tube (final concentration 20 µM). 7. Incubate the two tubes 3 hr in a 37°C, 5% CO2 cell incubator. In order to obtain optimal stimulation, incubate the tubes in an upright or slightly tilted position and gently mix every hour to enhance oxygenation of the cells.
8. Harvest PB samples from the cell incubator and store them at 4°C until step 15. Stain sample 9. Mix the antibody-coated bead solutions (IFN-γ, TNF-α, IL-2, IL-4, IL-5, and IL-10) in a roller mixer or by gently vortexing for 15 to 30 sec. 10. Label ten 12 × 75–mm polystyrene tubes with the following concentrations: 5000 pg/ml, 2500 pg/ml, 1250 pg/ml, 625 pg/ml, 312.5 pg/ml, 156 pg/ml, 80 pg/ml, 40 pg/ml, 20 pg/ml, and 0 pg/ml (standards). 11. Label two additional 12 × 75–mm polystyrene tubes with “stimulated” and “nonstimulated” (sample). 12. Pipet 120 µl of each bead solution (10 µl for each standard and sample tube) into one tube and gently mix. Typically, 10 µl bead solution contains ∼6000 beads.
13. Into each standard tube and each sample tube, pipet 50 µl of the mixed antibodycoated bead solution plus 50 µl of the mixture of phycoerythrin (PE)-labeled monoclonal antibodies against IFN-γ, TNF-α, IL-2, IL-4, IL-5, and IL-10. 14. Add 50 µl of the appropriate recombinant IFN-γ, TNF-α, IL-2, IL-4, IL-5, IL-10 cytokine standard dilution to the corresponding standard tube. Studies of Cell Function
9.21.3 Current Protocols in Cytometry
Supplement 25
15. Add 50 µl stimulated PB sample and 50 µl non-stimulated sample to the tubes labeled stimulated and non-stimulated, respectively. It is important to measure the 50 ìl of the capture beads mixture, the detection antibody mixture, and either sample or standard precisely.
16. Incubate all tubes 2 hr at room temperature protected from direct light exposure. 17. Fifteen min prior to the end of the incubation period add to the stimulated and non-stimulated sample tubes 10 µl anti-CD45-PE-Cy5, 10 µl anti-CD3-FITC, and 5 µl anti-CD8-APC monoclonal antibodies to stain the leukocytes, total T lymphocytes, and both CD8+ T cells and CD8+ NK cells, respectively. Simultaneously, add 3 µl anti-TNF-α-PE monoclonal antibody to stain TNF-α-secreting leukocytes at saturating conditions. 18. At the end of the incubation period, add 1 ml PBS and wash the cells by centrifuging 5 min at 200 × g, room temperature, to remove antibody excess; carefully aspirate the supernatant and discard it. Add 300 µl PBS to each pellet. Set up instrument 19. Perform calibration of the flow cytometer according to the instructions of the manufacturer. Check the flow system and perform compensation of the instrument for a lyse/no-wash mode with the appropriate mixture of calibration beads (unstained and FITC-, PE-, PE-Cy5-, and APC-stained CaliBRITE beads). Calibration typically includes sequential adjustment of side light scatter (SS) and forward light scatter (FS) followed by fluorescence compensation; this latter step is performed to avoid intrinsic spectral overlap of the different fluorochromes used, so that no fluorochrome emission into an inappropriate detector will be detected (UNIT 1.14). For this purpose, calibration beads are typically used following the recommendations of the manufacturer. For establishing and maintaining system linearity, see UNIT 1.4.
20. Set the SS and FS amplifiers to log mode and decrease the SS PMT voltage by 100 from the value obtained in step 19. Final values of the SS detector in a FACScalibur instrument will typically be between 300 and 350.
21. Prepare a tube with 50 µl cytometer setup beads diluted in 450 µl PBS and run the beads in the flow cytometer at a rate of at least 400 events/sec. In this tube, only two bead subsets with distinct red fluorescence intensities are included: one to establish the upper and the other to establish the lower bead-associated red fluorescence levels.
22. Adjust the voltage of the red fluorescence detector so that the median value of the brightest bead population shows a fluorescence intensity of ∼5000 (arbitrary relative linear fluorescence units scaled from 1 to 10,000). 23. Finely adjust the voltages of the FITC (green fluorescence) and PE (orange fluorescence) detectors so that the median of both the green fluorescence and orange fluorescence of the cytometer setup beads is ∼2.0 to 2.5 (arbitrary relative linear fluorescence units scaled from 1 to 10,000; Fig. 9.21.1A,B).
Analysis of Cytokine Responses in Stimulated Whole Blood
24. Prepare two tubes with a mixture of 50 µl cytometer setup beads and 50 µl FITCpositive control detector in one tube, and a mixture of 50 µl cytometer setup beads and 50 µl PE-positive control detector in the second. 25. Incubate these tubes 30 min at room temperature protected from direct light.
9.21.4 Supplement 25
Current Protocols in Cytometry
A
B
Unstained Setup Beads
103 102 101 100 102 101 103 Green fluorescence
101
FITC-stained Setup Beads
100
D
102 101 103 Orange fluorescence
104
FITC-stained Setup Beads 104
Orange fluorescence
Red fluorescence
102
104
104 103 102 101 100
103 102 101 100
100
E
102 101 103 Green fluorescence
104
100
F
PE-stained Setup Beads
102 101 103 Green fluorescence
104
PE-stained Setup Beads 104
Orange fluorescence
104 Orange fluorescence
103
100 100
C
Unstained Setup Beads 104
Red fluorescence
Red fluorescence
104
103 102 101 100
103 102 101 100
100
102 101 103 Green fluorescence
104
100
102 101 103 Red fluorescence
104
Figure 9.21.1 Calibration of the flow cytometer. Panels A and B show the populations of setup beads, which establishes the upper bead-associated red fluorescence level at ∼5000 (arbitrary relative linear fluorescence units scaled from 1 to 10,000). In this tube containing unstained setup beads, mean signals of both green fluorescence (FITC; A) and orange fluorescence (PE; B) must be adjusted to ∼2.0 to 2.5 (arbitrary relative linear fluorescence units scaled from 1 to 10,000). Panels C and D show the subpopulation of setup beads with low red fluorescence that emits green fluorescence once stained with the FITC-positive control detector. The red fluorescence emission of the FITC-stained beads is dim and identical to that of the FITC-negative bead population (C).The mean signal of the FITC-stained beads in the PE-associated orange fluorescence detector must be identical to the FITC-negative bead population (D). Panels E and F show the orange fluorescence of the subpopulation of setup beads once stained with the PE-positive control detector. Only the PE-stained bead population is detected by the orange fluorescence detector and its emission does not overlap into the green fluorescence detector (E). The red fluorescence emission of the PE-stained beads is dim and identical to that of the PE-negative bead population (F).
Studies of Cell Function
9.21.5 Current Protocols in Cytometry
Supplement 25
26. Add 400 µl PBS to each tube and run the first tube in the flow cytometer at a rate of at least 400 events/sec. Once stained with the appropriate FITC-positive control detector, one of the two populations of cytometer setup beads becomes FITC positive while the other one does not.
27. While running the mixture of cytometer setup beads and FITC-positive control detector adjust the compensation orange minus percent green so that the fluorescence emission of the FITC-stained bead population appears only in the FITC detector and its mean signal in the PE detector is identical to that of the FITC-negative bead population (see Fig. 9.21.1C,D). 28. Run the second tube (mixture of cytometer setup beads and PE-positive control detector) in the flow cytometer at a rate of at least 400 events/sec. Once stained with the appropriate PE-positive control detector, one of the two populations of cytometer setup beads becomes PE positive while the other one does not.
29. While running this bead mixture, adjust the compensation green minus percent orange fluorescence value so that the PE-stained bead population appears only in the PE detector and its mean signal in the FITC detector is identical to that of the PE-negative bead population (see Fig. 9.21.1E). 30. With the same beads running, adjust the compensation red minus percent orange so that the PE-stained bead population appears clearly only in the PE detector and its signals in the red fluorescence detector are dimly positive and identical to those of the PE-negative bead population (see Fig. 9.21.1F). 31. Set the threshold parameter of the red fluorescence detector at ∼350 to 400 (out of a scale from 0 to 10,000 arbitrary relative linear fluorescence units). This should allow detection of all bead populations (labeled with fluorochromes emitting red fluorescence detected on the red fluorescence detector) as well as all CD45-PE-Cy5+ leukocyte subpopulations. At the same time, debris, red cells, and platelets are excluded, as shown in Figure 9.21.3A.
32. Save the optimized instrument settings for later retrieval. Instrument setup should be carried out prior to the assay during sample stimulation or incubation of the stimulated sample with the cytokine-capturing bead mixture. Instrument setup must be checked daily (see UNIT 6.1) and re-done whenever the performance of the instrument does not fulfill the acceptance criteria (see UNIT 6.1).
Acquire data 33. Open the data acquisition software program and set acquisition mode. 34. Retrieve the optimized instrument settings and set them. 35. Set the number of events to be counted for the acquisition of beads from the tubes containing cytokine standards. Information on a minimum of 300 beads should be acquired for each of the six different types of capture beads; all bead populations will appear in the R1 region shown in Figures 9.21.1A, 9.21.2A, and 9.21.3A.
36. Set the flow rate between 150 and 400 events/sec. Analysis of Cytokine Responses in Stimulated Whole Blood
37. Run and acquire data for the negative control tube (0 pg/ml of any of the selected cytokines: IFN-γ, TNF-α, IL-2, IL-4, IL-5, and IL-10) before any other recombinant standard tubes.
9.21.6 Supplement 25
Current Protocols in Cytometry
A
B 104
104 103
R1
SS
SS
103 102 101
102 101
100
100 100
101
102 FS
103
104
100
102 101 103 Red fluorescence
104
Orange fluorescence
C 80 pg/ml
0 pg/ml
625 pg/ml
5000 pg/ml
Red fluorescence
D
IL-4
104
PE-MFI
103 102 101 100 100
A = 0.16
101 102 103 Concentration (pg/ml)
104
y = ((A − D)/(1 + (x/C)^B)) + D B = 2.27 C = 3.45 D = 5.76 R^2 = 0.9993
Figure 9.21.2 Identification of cytokine-capturing bead populations, analysis of cytokine standards, and construction of a cytokine standard curve. Panel A shows the tight FS/SS characteristics of the bead cluster (R1). Panel B shows the six different cytokine-capturing bead populations (from left to right: IFN-γ, TNF-α, IL-10, IL-5, IL-4, and IL-2) separated according to their discrete red fluorescence intensities. Panel C shows four orange fluorescence (PE) versus red fluorescence bivariate dot plots in which the increased cytokine concentrations are directly correlated with the increased mean orange fluorescence intensity channel (MFI). Panel D shows a typical logarithmic calibration curve constructed with several concentrations (0, 20, 40, 80, 156, 312.5, 625, 1250, 2500, and 5000 pg/ml) of recombinant IL-4 in which the MFI of the anti-IL4-PE-conjugated detector antibody for each standard tube is plotted versus the respective concentration of the cytokine.
Studies of Cell Function
9.21.7 Current Protocols in Cytometry
Supplement 25
A
B
Non-Gated 104
Non-Gated 104 R2
R1
102
R3
101
101
102 FS
103
104
Gated on R3 and R4 (= G1)
102 100 101 104 103 FL3/bead and CD45-PECy5 red fluorescence
TOTAL LYMPHOCYTES 103 R5 102 101 R6 100
Gated on (G1 and R6) or (G1 and R5)
D TNF-α-PE orange fluorescence
104
104 TOTAL T-CELLS 103 102 101 100
102 100 101 103 104 CD3-FITC green fluorescence Gated on (G1 and R6) or (G1 and R5)
F
104
Gated on R1 and R2 104
103 102 101 100 101
102
103
104
CD8-APC deep-red fluorescence
IL-2
IFNγ
103 102
TNFα IL-4
101 IL-10
100 100
Analysis of Cytokine Responses in Stimulated Whole Blood
102 100 101 103 104 CD8-APC deep-red fluorescence
Cytokine detectors PE orange fluorescence
TNF-α-PE orange fluorescence
E
R4
100 100
CD8-APC deep-red fluorescence
102 101
100
C
Threshold
103 SS
SS
103
100
101
102
IL-5 103
104
Bead red fluorescence
Figure 9.21.3 Simultaneous identification of peripheral blood leukocytes and cytokine-capturing beads. The different FS/SS properties of beads and cells allow one to distinguish capture beads (R1) from cells (panel A). The red fluorescence acquisition threshold allows simultaneous detection of the six different bead populations (R2) as well as all CD45-PE-Cy5+ leukocyte cell populations (B). Panel C shows a CD3-FITC green fluorescence versus CD8-APC deep red fluorescence bivariate dot plot in which only the lymphocyte population, gated simultaneously on R3 (A) and R4 (B), is included. In panel C, R5 shows the CD3+/CD8+ and R6 the CD3+/CD8− T lymphocyte subsets, respectively. Panels D and E show TNF-α-PE orange fluorescence versus CD8-APC deep red fluorescence bivariate dot plots in which only the two major lymphocyte populations, CD3+/CD8− (dark dots included in G1 and R6) and CD3+/CD8+ (grey events included in G1 and R5), are displayed. Panel D shows an example of a non-stimulated sample in which a threshold line in the TNF-α-PE-associated orange fluorescence is established to include all TNF-α-negative cells. Panel E shows a PMA plus ionomycin–stimulated sample in which the TNF-α-secreting lymphocytes are clearly distinguished by their increased fluorescence emission in the PE orange fluorescence detector. Panel F shows a PMA plus ionomycin–stimulated sample in which quantifiable soluble amounts of IFN-α, TNF-α, IL-4, and IL-2 are detected.
9.21.8 Supplement 25
Current Protocols in Cytometry
38. Sequentially run and acquire data for each of the other standard tubes. Acquisition should be in order from the lowest to the highest concentrations of cytokine standards.
39. Once acquisition of the last standard tube is complete, change the number of events to be acquired to at least 30,000, including both capture beads (region R1 from Fig. 9.21.2 and Figure 9.21.3) and CD45+ cells (leukocytes). If the threshold is too low, many of the acquired events might correspond neither to beads nor to CD45+ cells. In order to include all beads or leukocytes, this threshold should not be too high (see Fig. 9.21.3B).
40. Set the flow rate between 100 and 400 events/sec and acquire data for the non-stimulated sample. 41. Acquire data for the stimulated sample. NOTE: Since no erythrocyte-lysing steps were used, data acquisition must be performed at a low flow rate in order to avoid as much as possible artifacts caused by coincidence of a red cell with either a bead or a leukocyte during analysis.
Construct standard calibration curves for soluble cytokine measurements 42. Read the file containing the data from the negative control (0 pg/ml) standard tube. All standard tubes contain only a mixture of the six different types of cytokine-capturing beads (IFN-γ, TNF-α, IL-2, IL-4, IL-5, and IL-10) without cells.
43. Select two bivariate dot plot representations of red fluorescence versus SS and red fluorescence versus orange fluorescence in which all six different bead populations are displayed as shown in Figure 9.21.2B,C. 44. Identify each of the six bead populations by drawing electronic regions around each of the six clusters of events showing identical SS but distinct and increasing red fluorescence intensity as shown in Figure 9.21.2B. 45. Calculate the mean PE-associated orange fluorescence intensity (MFI; arbitrary relative linear units scaled from 0 to 10,000) for each population of cytokine-capturing beads as identified in step 44. 46. Perform an identical analysis (steps 42 to 45) for each one of the other standard tubes (20 pg/ml, 40 pg/ml, 80 pg/ml, 156 pg/ml, 312.5 pg/ml, 625 pg/ml, 1250 pg/ml, 2500 pg/ml, and 5000 pg/ml). At this stage, there will be a series of ten known cytokine concentrations (0 to 5000 pg/ml) associated with ten PE-associated MFI values, for each one of the six different cytokinecapturing beads.
47. Using a spreadsheet data base (e.g., Excel), construct six logarithmic/logarithmic fitting curves (one for each cytokine) by plotting, for each standard tube, the PE-associated MFI obtained for that particular cytokine-capturing bead population versus the concentration of the respective cytokine in that standard tube, as shown in Figure 9.21.2D. Alternatively, automatic software programs (BD-CBA software) can be used to perform all steps described from the analysis of the cytokine standard files to the construction of the logarithmic fitting curves (steps 45 to 47).
Studies of Cell Function
9.21.9 Current Protocols in Cytometry
Supplement 25
Analyze data for non-stimulated sample 48. Read the file containing the data from the non-stimulated sample. In those files corresponding to the non-stimulated and stimulated sample, information on beads and cells was simultaneously acquired.
49. Draw two bivariate dot plots of FS versus SS and SS versus CD45-PE-Cy5 red fluorescence. 50. In the FS versus SS bivariate dot plot, select the bead population as the tightest cluster of events with a relatively high SS and low FS (R1 in Fig. 9.21.3A); further gate the whole bead population in the SS versus CD45-PE-Cy5 bivariate dot plot (R2) as shown in Figure 9.21.3B. In this latter SS versus CD45-PE-Cy5 dot plot, six different clusters of beads/events become apparent, each corresponding to a different population of beads.
51. Identify the total lymphocyte population in the FS versus SS and the SS versus CD45-PE-Cy5 red fluorescence bivariate dot plots by drawing two electronic regions (R3 and R4 in Fig. 9.21.3A,B) around the clusters of events with low FS and low SS (R3 in Fig. 9.21.3A) and around the cluster of events with low SS/high CD45 (R4 in Fig. 9.21.3B), respectively. 52. Draw a new green fluorescence (CD3-FITC) versus deep red fluorescence (CD8APC) bivariate dot plot in which only those events fulfilling the criteria to be in R3 and R4 (Gate 1) are included (Fig. 9.21.3C). Place T cell activation threshold 53. Identify the two major T cell subsets by drawing two different electronic regions (R5 and R6) in the CD3-FITC versus CD8-APC bivariate dot plot, one including the CD3+/CD8+ events (R5 in Fig. 9.21.3C) and the other the CD3+/CD8− events (R6 in Fig. 9.21.3C). 54. Draw a TNF-α-PE orange fluorescence versus CD8-APC deep-red fluorescence bivariate dot plot in which only the total CD3+ T cells are included (Fig. 9.21.3D); in this plot, use different colors to identify CD8+/CD3+ events (Gate 1 and R5) and CD8−/CD3+ T cells (Gate1 and R6) as displayed in Figure 9.21.3D. 55. Place a threshold line to include ≥99% of the negative cells in the orange fluorescence axis as shown in Figure 9.21.3D and save it for later retrieval. See UNIT 6.2 for more detail on the proper analysis of lymphoid cells labeled with multiple antibodies by placing arbitrary markers for positivity in negative controls.
Calculate basal cytokine amounts 56. Draw a new red fluorescence versus orange fluorescence bivariate dot plot in which only events corresponding to the six different bead populations (R1) are displayed as shown in Figure 9.21.3F. 57. Calculate the PE MFI for each of the six different bead populations as described in steps 44 and 45.
Analysis of Cytokine Responses in Stimulated Whole Blood
58. Extrapolate these MFI values to the standard calibration curves constructed previously in step 47 and calculate the soluble amounts present in the non-stimulated sample for each of the six cytokines analyzed (IFN-γ, TNF-α, IL-2, IL-4, IL-5, and IL-10) in pg/ml.
9.21.10 Supplement 25
Current Protocols in Cytometry
Analyze data for stimulated sample 59. Read the file containing data from the stimulated sample. 60. Identify and gate on those events corresponding to the total lymphocyte population and the different T cell subsets as described above in steps 51 to 53 (see Fig. 9.21.3A,B,C). 61. Select a TNF-α-PE orange fluorescence versus CD8-APC deep red fluorescence bivariate dot plot in which only the gated CD3+/CD8+ (G1 and R5) and CD3+/CD8− (G1 and R6) cells are included as shown in Figure 9.21.3E. 62. Retrieve the threshold line placed in the non-stimulated sample file (step 55) and calculate the proportion of cells with TNF-α-PE orange fluorescence values above this threshold for both the CD3+/CD8+ and CD3+/CD8− T cell subsets as illustrated in Fig. 9.21.3E. NOTE: Events above the threshold line would correspond to cells actively producing TNF-α (TNF-α+ cells).
63. Within the TNF-α+/CD3+/CD8+ and the TNF-α+/CD3+/CD8− cell populations, calculate the TNF-α-PE MFI values. Within TNF-α+ T cells, the MFI of TNF-α-PE orange fluorescence will vary depending on the amount of TNF-α produced per T cell.
Calculate cytokine production 64. Select the six different bead populations as described in step 50 and represent them in a red fluorescence versus orange fluorescence bivariate dot plot as described in step 56. 65. Calculate the PE MFI values for each cytokine-capturing bead population (see Fig. 9.21.3F) as described above in step 57. 66. Extrapolate these PE MFI values to the standard calibration curve corresponding to each cytokine and calculate the absolute amount of soluble cytokine (pg/ml) present in the stimulated sample. To calculate the specific amount of each soluble cytokine secreted in response to the stimuli, levels found for each cytokine in the non-stimulated sample (step 58) should be subtracted from the amounts found for the same cytokine in the stimulated sample. Due to the inhibitory effect of the TACE inhibitor, soluble amounts of TNF-α detected in this assay would represent only ∼5% to 10% of all TNF-α produced; the other 90% to 95% is retained on the surface membrane of the secreting cells.
MEASUREMENT OF GENERIC INFLAMMATORY RESPONSE INDUCED BY LIPOPOLYSACCHARIDE
ALTERNATE PROTOCOL
This alternate protocol describes a method to evaluate the functional status of the immune system in response to inflammatory stimuli by simultaneous measurement of monocyte activation and quantitation of IL-1β, IL-6, IL-8, IL-10, IL-12, and TNF-α production. Additional Materials (also see Basic Protocol) 2.5 µg/ml bacterial endotoxin lipopolysaccharide (LPS; see recipe) Beads coated with anti-TNF-α, anti-IL-1β, anti-IL-6, anti-IL-8, anti-IL-10, and anti-IL-12 MAbs (e.g., BD Biosciences) Fluorochrome-labeled monoclonal antibodies: Anti human CD14-FITC (e.g., MφP9, BD Biosciences) Studies of Cell Function
9.21.11 Current Protocols in Cytometry
Supplement 25
Mix of anti-cytokine (TNF-α, IL-1β, IL-6, IL-8, IL-10, and IL-12) PE-conjugated MAbs (e.g., BD-CB kit, BD Biosciences) Recombinant TNF-α, IL-1β, IL-6, IL-8, IL-10, and IL-12 cytokine standards (e.g., BD Biosciences; see recipe) Stimulate cytokine production by peripheral blood monocytes with LPS 1. Follow Basic Protocol, steps 1 to 3. 2. To the stimulated tube, add 10 µl of 2.5 µg/ml LPS stock solution. 3. To the non-stimulated tube, add 10 µl PBS (stimulus vehicle). If the whole blood has been properly stored at room temperature under sterile conditions, PB lymphocytes can be successfully stimulated as long as 24 hr after collection, but it is better to maintain a shorter period between sample draw and stimulation.
4. Add 2.5 µl TACE inhibitor solution to each tube (final concentration 20 µM). 5. Incubate the two tubes 4 hr in a humidified 37°C, 5% CO2 cell incubator. In order to obtain optimal stimulation, incubate the tubes in an upright or slightly tilted position and gently mix every hour in order to enhance oxygenation of the cells.
6. Harvest PB samples from the cell incubator and store them at 4°C until performing step 13. Stain sample 7. Mix the antibody-coated bead solutions (TNF-α, IL-1β, IL-6, IL-8, IL-10, and IL-12) in a roller mixer or by gently vortexing 15 to 30 sec. 8. Label ten 12 × 75–mm polystyrene tubes with the following concentrations: 5000 pg/ml, 2500 pg/ml, 1250 pg/ml, 625 pg/ml, 312.5 pg/ml, 156 pg/ml, 80 pg/ml, 40 pg/ml, 20 pg/ml, and 0 pg/ml (standards). 9. Label two additional 12 × 75–mm polystyrene tubes as “stimulated” and “non-stimulated” (sample.) 10. Pipet 120 µl of each bead solution (10 µl for each standard and sample tube) into one tube and gently mix. Typically, 10 ìl bead solution contains ∼6000 beads.
11. Into each standard tube and sample tube, pipet 50 µl of the mixed antibody-coated bead solution plus 50 µl of the mixture of PE-labeled monoclonal antibodies against TNF-α, IL-1β, IL-6, IL-8, IL-10, and IL-12. 12. Add 50 µl of the appropriate cytokine standard dilution to the corresponding standard tube. 13. Add 50 µl stimulated sample and 50 µl non-stimulated sample to the tubes labeled stimulated and non-stimulated, respectively. 14. Incubate all tubes 2 hr at room temperature protected from direct light exposure.
Analysis of Cytokine Responses in Stimulated Whole Blood
15. Fifteen min prior to the end of the incubation period, add to the stimulated and non-stimulated sample tubes 10 µl anti-CD45-PE-Cy5 and 10 µl anti-CD14-FITC monoclonal antibodies to stain the leukocytes and monocytes, respectively. Simultaneously add 3 µl anti-TNF-α-PE monoclonal antibody to stain TNF-α-secreting leukocytes.
9.21.12 Supplement 25
Current Protocols in Cytometry
B 104
SS
103 102 R7 101 100
CD45-PECy5 red fluorescence
A
Gated on CD45+leukocytes (R7) 104 103 102 101 100
100 101 103 104 102 CD45-PECy5 red fluorescence Gated CD45+/CD14+ events (R7 and R8) 104
100 101 103 104 102 CD14-FITC green fluorescence
103 102 101 100 102 100 101 103 104 CD14-FITC green fluorescence
Gated CD45+/CD14+ events (R7 and R8)
D TNF-α-PE orange fluorescence
TNF-α-PE orange fluorescence
C
R8
104 TNFα+CD14+ 103 102 101 100 100 101 103 104 102 CD14-FITC green fluorescence
Figure 9.21.4 Identification of monocytes within the whole leukocyte cell population and detection of TNF-α+ monocytes. Panel A shows the six different cytokine population–capturing beads and the CD45+ leukocytes (R7). In panel B, monocytes are identified as cells with high CD45-PE-Cy5associated red fluorescence and high CD14-FITC-associated green fluorescence (R8). Panel C shows a representative TNF-α-PE-associated orange fluorescence versus CD14-FITC-associated green fluorescence bivariate dot plot of a non-stimulated sample in which a threshold line is establish to include all TNF-α-negative monocytes. Panel D shows an LPS-stimulated sample in which TNF-α-expressing monocytes (TNFα+CD14+) are identified.
16. At the end of the incubation period, add 1 ml PBS and wash the cells by centrifuging 5 min at 200 × g, room temperature, to remove antibody excess; carefully aspirate the supernatant and discard it. Set up instrument and acquire data 17. Perform calibration of the flow cytometer as described in Basic Protocol, steps 19 to 32. 18. Perform data acquisition as described in Basic Protocol, steps 33 to 41. Studies of Cell Function
9.21.13 Current Protocols in Cytometry
Supplement 25
Construct standard calibration curves for soluble cytokine measurements 19. Read the file containing the data from the negative control (0 pg/ml) standard tube and perform Basic Protocol, steps 42 to 47, elaborating separate log/log standard calibration curves for TNF-α, IL-1β, IL-6, IL-8, IL-10, and IL-12. Analyze data for non-stimulated sample 20. Read the file containing the data from the non-stimulated sample. 21. Draw two bivariate dot plots of FS versus SS and SS versus CD45-PE-Cy5 red fluorescence. 22. Select the bead population in the FS versus SS bivariate dot plot and further gate it in the SS versus CD45-PE-Cy5 bivariate dot plot as described in Basic Protocol, step 50. 23. Gate those CD45+ events corresponding to cells as shown in Figure 9.21.4A (R7). 24. Draw a new CD45-PE-Cy5 red fluorescence versus CD14-FITC green fluorescence bivariate dot plot in which only those events that are in R7 and correspond to CD45+ cells are displayed. Place monocyte activation threshold 25. Identify the monocyte population by drawing an electronic region (R8) around the cluster of events with high CD45-PE-Cy5 red fluorescence and high CD14-FITC green fluorescence as shown in Figure 9.21.4B and gate them. 26. Draw a TNF-α-PE orange fluorescence versus CD14-FITC green fluorescence bivariate dot plot in which only those events fulfilling the criteria to be used in the monocyte gate, as described in step 25 (R7 and R8), are included (see Fig. 9.21.4C). 27. Place a threshold line to include ≥99% of the negative cells in the TNF-α-PE orange fluorescence axis as shown in Figure 9.21.4C and save it for later retrieval. See UNIT 6.2 for more detail on the proper analysis of lymphoid cells labeled with multiple antibodies by placing arbitrary markers for positivity in negative controls.
Calculate basal cytokine amounts 28. Draw a new red fluorescence versus orange fluorescence dot plot in which only those events corresponding to the six different bead populations are displayed as shown in Figure 9.21.3F. 29. Calculate the PE MFI for each of the six different bead populations, as described in Basic Protocol, steps 44 and 45. 30. Extrapolate these MFI values to the standard calibration curves constructed previously and calculate the soluble amounts (pg/ml) present in the non-stimulated sample for each of the six cytokines analyzed (TNF-α, IL-1β, IL-6, IL-8, IL-10, and IL-12). Analyze data for stimulated sample 31. Read the file containing data from the stimulated sample. 32. Identify and gate on the monocyte population as described in step 25. Analysis of Cytokine Responses in Stimulated Whole Blood
33. Select a TNF-α-PE orange fluorescence versus CD14-FITC green fluorescence bivariate dot plot in which only the monocyte population (R7 and R8) is displayed.
9.21.14 Supplement 25
Current Protocols in Cytometry
34. Retrieve the same threshold line placed in the non-stimulated sample file (step 27) to calculate the proportion of monocytes with TNF-α-PE orange fluorescence values above this threshold (TNF-α+ cells) as illustrated in Figure 9.21.4D. 35. Within the TNF-α+ monocytes, calculate the TNF-α-PE-associated MFI values. Within TNF-α+ monocytes, the MFI of TNF-α-PE-associated orange fluorescence will vary depending on the amount of TNF-α produced per monocyte.
Calculate cytokine production 36. Select the six different bead populations as described in Basic Protocol, step 50. 37. Calculate the MFI values for each cytokine-capturing bead population as described above in step 29. 38. Extrapolate these MFI values to the standard calibration curves corresponding to each cytokine constructed previously and calculate the soluble amounts (pg/ml) present in the stimulated sample for each of the six cytokines analyzed (TNF-α, IL-1β, IL-6, IL-8, IL-10, and IL-12). To calculate the specific amount of each soluble cytokine secreted in response to LPS, the levels found for each cytokine in the non-stimulated sample (step 30) should be subtracted from the amounts found for the same cytokine in the stimulated sample. Due to the inhibitory effect of the TACE inhibitor, soluble amounts of TNF-α detected would in this assay represent only ∼5% to 10% of all TNF-α produced; the other 90% to 95% is retained on the surface membrane of the secreting cells.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Bacterial endotoxin lipopolysaccharide (LPS), 2.5 ìg/ml Prepare an aqueous stock solution of the bacterial endotoxin lipopolysaccharide (LPS; Sigma) lyophilized powder by diluting it in PBS at a final concentration of 2.5 µg/ml; store the stock solution up to 6 months at –20°C. LPS is used to induce a generic inflammatory response through monocytes and dendritic cells. For sample stimulation, a final concentration of 100 ng/ml is typically used.
Cell culture medium Supplement RPMI 1640 medium with L-glutamine to a final concentration of 2 mM and add an antibiotic mixture (10,000 UI/ml penicillin and 4000 µg/ml streptomycin; GIBCO or Invitrogen) to a final concentration of 100 UI/ml penicillin and 100 µg/ml streptomycin. Store up to 1 month at 4°C. Ionomycin, 50 ìg/ml Prepare 50 µg/ml ionomycin (Sigma) solution in absolute ethanol and store up to 6 months at −20°C. Phorbol myristate acetate (PMA), 100 ìg/ml Prepare phorbol myristate acetate (PMA; Sigma) stock solution as follows: dilute PMA in DMSO and store it in 100-µg/ml aliquots up to 6 months at −20°C. Just before use, make a 1:100 (v/v) dilution of the stock solution in PBS (APPENDIX 2A).
Studies of Cell Function
9.21.15 Current Protocols in Cytometry
Supplement 25
Recombinant IFN-γ, TNF-α, IL-2, IL-4, IL-5, IL-10 cytokine standards To prepare “standard” tubes with known concentrations (0 to 5000 pg/ml) of a mixture of lyophilized recombinant IFN-γ, TNF-α, IL-2, IL-4, IL-5, and IL-10 cytokines, reconstitute one vial of the lyophilized mixture of cytokine standards in 1 ml of assay diluent (both reagents are included in the BD-CBA kit) by gentle mixing on a roller mixer for 15 min. The stock standard concentration of each cytokine is 10,000 pg/ml. Perform a 1:2 (v/v) serial dilution of the 10,000 pg/ml standard in the following order: 1:2 (5000 pg/ml), 1:4 (2500 pg/ml), 1:8 (1250 pg/ml), 1:16 (625 pg/ml), 1:32 (312.5 pg/ml), 1:64 (156 pg/ml), 1:128 (80 pg/ml), 1:256 (40 pg/ml), and 1:512 (20 pg/ml). Alternatively, other sources of standards for these cytokines can be used and the mixture prepared for any desired concentration ranges. Store up to 6 hr at 4°C. Recombinant TNF-α, IL-1β, IL-6, IL-8, IL-10, and IL-12 cytokine standards To prepare “standard” tubes with known concentrations (0 to 5000 pg/ml) of a mixture of lyophilized recombinant TNF-α, IL-1β, IL-6, IL-8, IL-10, and IL-12 cytokines, reconstitute one vial of the lyophilized mixture of cytokine standards in 1 ml of assay diluent (both reagents are included in the BD-CBA kit) by gentle mixing on a roller mixer for 15 min. The stock standard concentration of each cytokine is 10,000 pg/ml. Perform a 1:2 (v/v) serial dilution of this 10,000 pg/ml standard in the following order: 1:2 (5000 pg/ml), 1:4 (2500 pg/ml), 1:8 (1250 pg/ml), 1:16 (625 pg/ml), 1:32 (312.5 pg/ml), 1:64 (156 pg/ml), 1:128 (80 pg/ml), 1:256 (40 pg/ml), and 1:512 (20 pg/ml). Alternatively, other sources of standards for these cytokines can be used and the mixture prepared for any desired concentration ranges. Store up to 6 hr at 4°C. TACE inhibitor: BB3103, 2 mM Dissolve BB3103 (British Biotechnology) in DMSO to make stock solution of 2 mM. Store up to 6 months at –20°C. Although several thaw and freeze cycles do not seem to alter the product, it is better to store it in 10- to 20 µl aliquots. COMMENTARY Background Information
Analysis of Cytokine Responses in Stimulated Whole Blood
The relevance of cytokine measurements in different physiological and pathological conditions is currently well established. Cytokines play a crucial role in immune responses against infection and tumor cells, as well as in vaccination and transplantation, among other situations. Through complex networks, cytokines are essential elements in priming, amplifying, directing, or inhibiting specific immune responses (Arai et al., 1990). At present, several methods are available to measure blood cell responses based on cytokine production. Most of these methods use bulk assays that detect and/or quantify specific cytokine mRNA and proteins. To measure mRNA, northern blot analysis, RT-PCR, and in situ hybridization techniques can be used. Despite their sensitivity, these approaches have several major limitations: (1) in most instances they can not link cytokine responses to single cells; (2) quantitation can be achieved only with some procedures
(e.g., real-time PCR) and does not refer to the producing cells but to total mRNA; (3) the amount of mRNA typically shows only a marginal correlation with the total amount of protein secreted. In most studies, cytokine production and secretion is assessed through quantitative immunospecific methods using either enzyme-linked immunoassays (ELISA) or, less frequently, western blot analyses. Once again, despite the high sensitivity and specificity of these methods, none enables the simultaneous quantitation of secreted cytokines and the identification of the secreting cells. Accordingly, since specific identification of those cells that produce cytokines within a sample is necessary, analysis of cytokine production at the singlecell level is mandatory. At present, several different methods can be used to detect cytokine production in single cells. Among these, the ELISPOT assay as well as different flow cytometry single cell–based techniques are the most widely used. However, in these methods,
9.21.16 Supplement 25
Current Protocols in Cytometry
the study of cytokine production at the singlecell level enables an accurate evaluation of the overall amount of the cytokines produced. Accordingly, these methods require extensive sample manipulation, including either a generic blockade of protein secretion into the extracellular medium, cell fixation and permeabilization to allow cytokine detection in the cell cytoplasm (Prussin et al., 1995), or a limited diffusion of the secreted cytokines by using an antibody-based trapping matrix on the cell surface (Manz et al., 1995). This further leads to an inhibition of autocrine and paracrine interactions, and to an alteration in the natural cell-cytokine network. The protocols described here have the advantage of allowing simultaneous assessment of the phenotype of cytokine-secreting cells and quantitation of the multiple cytokines secreted. This makes them unique in regards to overall evaluation of immune responses. Although evaluation of immunological responses by peripheral blood cells against generic stimuli are described in the Basic Protocol and Alternate Protocol, both procedures can also be used to study antigen-specific responses. In both protocols, sample manipulation is significantly reduced to a minimum, which allows cells to respond under conditions that mimic those occurring in vivo. In addition, the use of a highly-specific TACE inhibitor does not affect cytokine secretion into the extracellular medium, except for TNF-α. One limitation of this technique is that it relies on lack of a direct link between the amount of a specifically secreted cytokine and an individual cell.
Technical Considerations Selection of a universal and specific marker for identifying cytokine-producing cells In selecting a useful marker for the identification of cytokine-secreting cells, four major conditions were considered: (1) it should be present in all—or at least most—of the cytokine-secreting cells; (2) it should be expressed only on cytokine-secreting cells; (3) it should be easily and clearly detected on the surface of cytokine-secreting cells; and (4) its expression on the surface of the responding cells should be stable. TNF-α is one of the most ubiquitous and earliest released cytokines (Mascher et al., 1999). It is expressed in both inflammatory and antigen-specific T cells. Moreover, its secretion is not restricted to immune cells (such as monocytes, macrophages, lymphocytes, dendritic cells, or NK cells) but is also produced by
astrocytes, endothelial cells or smooth muscle cells, and others, which broadens the utility of the methodology described here. This already makes TNF-α an interesting candidate for use as a marker for cytokine-secreting cells. Most TNF-α molecules are produced as cytoplasmic membrane proteins that are cleaved by the TNF-α-converting enzyme (TACE) into a secreted soluble form. The use of specific TACE inhibitors (Bueno et al., 2002) has been associated with progressive accumulation of TNF-α on the surface of the cytoplasmic membrane of TNF-α-producing cells, allowing for their specific and unequivocal identification. Once highly specific TACE inhibitor is used to blockade shedding, ∼90% to 95% of all TNF-α molecules produced from the cell membrane is obtained, without interfering with secretion of other cytokines, thereby allowing the natural cytokine network to be maintained. CD69 is probably the most commonly used alternative marker for these purposes. Although no direct comparison between the two markers has been reported so far, CD69 seems to be less specific than TNF-α (not all CD69+ cells produce cytokines) and provides worse discrimination between CD69-negative and -positive cells, even when highly sensitive fluorochromes (i.e., PE) are used for its detection. Selection of an assay to quantify soluble cytokines Bead-based immunoassays have been developed with the major goal of increasing the amount of information obtained from volumelimited samples by multiplexing quantitation of several proteins in a single tube or well (Cook et al., 2001). The principle of the system is as simple as an ELISA: a capture MAb coupled to a solid-phase substrate—the bead—that works like a “well” in a conventional assay and a detector MAb to quantify the target protein. By using different fluorochromes to dye antibodies on one side and beads on the other side and exploiting the impressive capabilities of the flow cytometer for single-event analysis, this technique is able to simultaneously quantify several proteins with sensitivity, specificity, and precision similar to that described for conventional ELISA techniques (Chen et al., 1999). At present, several different formats for these bead-based immunoassays are available from different manufacturers. Typically, in these assays, one must be able to distinguish the different specific anti-cytokine immunoassays in a mixture. For that purpose, mixtures of
Studies of Cell Function
9.21.17 Current Protocols in Cytometry
Supplement 25
Analysis of Cytokine Responses in Stimulated Whole Blood
bead populations (each specific for a given cytokine) are used where distinction between bead populations can be achieved by combining different sizes and/or amounts of dyes specific for the beads with the use of a single and common distinct fluorochrome for the different anti-cytokine detectors. Requirements for selection of a specific bead mixture for the assay described include the following criteria: (1) after measurement in a flow cytometer, beads should be clearly distinguishable from PB cells; (2) mixed bead populations for the quantitation of desired cytokines should be compatible, allowing for the unequivocal identification of each multiplexed immunoassay; (3) they should allow quantitation of proteins below saturating conditions, even when each secreted protein is produced in different amounts; and (4) each bead immunoassay should have a high dynamic range that is optimal for the quantitation of the expected concentration ranges of the secreted cytokines. In order to keep the distinction between beads and cells simple, a mixture of bead populations with identical size and light-scatter properties is preferred. The possibility of clearly distinguishing between beads and PB leukocytes solely on the basis of differing light-scatter properties allows the simultaneous use of combinations of up to four different fluorochromelabeled reagents on beads with a second combination of up to four different fluorochromelabeled reagents on cells in conventional benchtop flow cytometers. Accordingly, extensive subsetting of PB leucocytes can be performed to selectively analyze TNF-α expression in a desired specific cell subpopulation. In turn, despite the fact that the selected bead mixtures in the two protocols described use only one fluorescence emission (red) to distinguish among the different bead populations and a second fluorescence emission (orange)—the most sensitive—for simultaneous detection of all cytokines measured, other bead mixtures in which up to four fluorescence emissions are used to distinguish between bead populations and quantify a specific cytokine within each bead population are also compatible with the assay. The potential use of flow cytometers with the ability to simultaneously detect and quantify five or more distinct fluorescence emissions will increase the capabilities of the assay. Bead populations coupled to anti-cytokinecapturing antibodies can also be mixed with bead populations to which antibodies showing other specificities are linked.
Critical Parameters and Troubleshooting The sample Peripheral blood samples must be collected in tubes containing heparin as the anticoagulant. Other anticoagulants, such as EDTA (a potent Ca2+ chelating agent), should be avoided because Ca2+ is required for cell activation for PMA plus ionomycin, LPS, and antigen-specific stimulation models. Once collected, samples should be processed within a period of 4 to 6 hr. Although acceptable results can be obtained from samples that have been properly stored at room temperature and in a sterile environment up to 24 hr after collection, the shorter the period between sample draw and stimulation, the better. The 1:5 dilution of the PB sample in culture medium has been optimized for the protocols described in this unit (PMA plus ionomycin and LPS). This dilution is used to obtain optimal cell activation and maximal levels for the secreted cytokines at the stimulation time points described (3 or 4 hr, respectively). However, in other types of samples (or stimuli) in which the number of target cells to be activated is very low, the dilution in culture medium can be reduced to 1:1 (v/v) in order to increase both the number of target cells present in the sample and the concentration of secreted cytokines accumulated in the culture supernatant. Stimulation of cytokine secretion The stimuli. In the assays described in this unit, generic non-specific stimuli were used. Concentrations employed for both PMA plus ionomycin and LPS have been shown to induce optimal responses in regards to cell activation and cytokine production for T cells and monocytes, respectively. Other concentrations of the same stimuli can also be used, in which case the time of stimulation may need to be adapted to the new conditions. For example, with the amount of LPS recommended in the Alternate Protocol (100 ng/ml), ∼80% of all monocytes from normal PB are TNF-α+ after a stimulation period of 4 hr; if the concentration of LPS is reduced to 10 ng/ml (a concentration also broadly used), only ∼50% to 60% of the monocytes will be TNF-α+ after 5 hr of culture. Although only generic non-specific stimuli were used in this unit, identical procedures may be applied in antigen-specific stimulation assays. However, under these conditions, the potential need for longer stimulation periods should be considered.
9.21.18 Supplement 25
Current Protocols in Cytometry
Duration of the stimulation period. According to the protocols described, several cytokines could be simultaneously detected after 3 hr (see Basic Protocol) or after 4 hr (see Alternate Protocol) of stimulation. It should be noted that optimal secretion time varies from one cytokine to another. Therefore, for each cell subtype, combination of cytokines, or even stimuli, different periods of stimulation may be required. Once long stimulation periods are used, secreted cytokines are expected to progressively accumulate in the culture supernatant. Nevertheless, this may be associated with the presence of high amounts of some soluble cytokines that may reach saturating conditions and/or exceed the upper limit of the calibration standard curves. In addition, apoptosis of activated cells may also occur under these conditions. Therefore, the authors recommend keeping stimulation periods as short as possible. Cell culture conditions. During stimulation, samples must be placed in a 37°C, 5% CO2 humidified cell incubator. A sterile environment is necessary to prevent activation of cells by some other non-specific stimuli. The incubation of the sample should be performed directly in polystyrene tubes specific for flow cytometer data acquisition in order to avoid cell loss due either to sample manipulation or to specific adhesion of cells because of potential up-regulation of adhesion molecules during the activation process. Alternatively, siliconcoated tubes can be used. Inhibitor of TACE activity. In order to obtain an efficient inhibition of TACE, highly specific inhibitors at saturating conditions should be used. Other generic protease inhibitors or mixtures of protease inhibitors are not able to efficiently prevent TNF-α from shedding nor induce its accumulation on the cell surface at levels that can be reliably detected by conventional flow cytometry. Data acquisition. Acquisition of data from non-lysed whole blood samples requires a narrow sample stream in order to avoid coincidence of red cells with beads and/or leukocytes during interrogation of particles by the laser beam. Accordingly, in these settings, acquisition should be performed at low flow rates. A major disadvantage of this is that it significantly increases total acquisition time. However, if erythrocyte-free cell samples are analyzed, acquisition can be set at a medium or even high flow rate, depending on the final concentration of cells/beads in the acquisition tube. The overall number of events (cells and beads) to be acquired varies widely and essentially depends
on the expected frequency of TNF-α-secreting cells after activation of the sample, provided that the threshold for exclusion of red cells, platelets, and debris is properly set and that a minimum of 300 events is acquired for each bead population. Accordingly, if <1% of TNFα+ cells are present within the whole leukocyte population, information on a minimum of 104 cells needs to be acquired (minimum of 102 TNF-α+ cells). The lower the number of activated TNF-α+ cells, the greater the number of total cells needed. In cases where there are very low frequencies of TNF-α+ cells, the number of beads to be interrogated is not restricting. Each bead is similar to a “well” in an ELISA assay and thus an average of 500 “wells” (beads) are measured per analyte, providing a high-precision assay. However, it should be noted that the use of a greater number of bead populations for the measurement of a higher number of soluble proteins may also require an increase in the total number of events acquired. Data analysis. Despite the large volume of information and apparent complexity of the sample files acquired, their analysis is at least as simple as any multiparameter flow cytometer study performed on a three- or four-color instrument. Bead and cell populations are perfectly distinguishable by their different lightscatter properties and therefore can be gated and analyzed as two separate files. Note that the clear distinction between beads and cells allows for the simultaneous use of different monoclonal antibodies conjugated with identical fluorochromes to detect proteins on the cell surface (e.g., anti-TNF-α-PE) and to detect captured cytokines on a bead population (e.g., anti-IFN-γ-PE). Similarly, since each population of beads with different anti-cytokine specificities is identified by its unique amount of red fluorescence emission, all MAb detectors can be labeled with the same fluorochrome, phycoerythrin, and several cytokines (e.g., IFN-γ, IL-2, IL-4, IL-5, IL-10) and quantified simultaneously.
Anticipated Results Stimulation of a PB sample with either PMA/ionomycin or LPS typically results in d iff er en t respo nse patter ns. After PMA/ionomycin stimulation, major responses associated with cytokine production are observed among lymphoid cells and especially within T lymphocytes. With the reagents in the Basic Protocol, after stimulation of normal PB samples for a period of 3 hr, between 50% and 70% of the T cells show surface TNF-α expres-
Studies of Cell Function
9.21.19 Current Protocols in Cytometry
Supplement 25
sion. Typically, both secreting and non-secreting cells are found among the major CD4+ and CD8+ T cell subsets. T cell responses are associated with secretion into the extracellular compartment of Th1-associated cytokines such as IFN-γ, IL-2, and to a lesser extent, IL-4. Under identical stimulatory conditions, IL-10 and IL5 levels are undetectable. In turn, stimulation of PB cells with LPS will induce a response mainly by inflammatory cells such as monocytes and to a smaller extent, circulating dendritic cells. According to the Alternate Protocol, monocytes can be easily distinguished by their FS/SS and CD45/CD14 staining characteristics, while identification of dendritic cells will require the use of other specific stainings (Bueno et al., 2001). Monocyte response is reflected by the expression of TNF-α on the cell surface of the majority of the stimulated monocytes. In non-stimulated normal PB samples, the percentage of TNF-α+ cells is <0.1% using the Basic and Alternate Protocols. At the soluble level, high amounts of inflammatory cytokines such as IL-1β, IL-6, and IL-8 become detectable after a stimulation period of 4 hr. The use of different markers to further stain lymphocytes and/or inflammatory cells together with the CD45 antigen allows the specific identification and enumeration of TNF-α+ cells among different subsets of T cells, monocytes, and/or dendritic cells. In turn, the exact cytokines measured at the soluble level may vary depending on the combination of cytokine-capturing beads and detector antibodies used in the assay. The levels of soluble cytokines detected after stimulation of normal PB samples with PMA/ionomycin according to the Basic Protocol fall within the following ranges: IFN-γ, 3746.9 ± 1726.7 pg/ml; IL-2, 1163.5 ± 380.6 pg/ml; and IL-4, 49.8 ± 22.1 pg/ml. For the Alternate Protocol, mean levels of IL-1β, IL-6, and IL-8 are ∼4247.3 ± 2368.5 pg/ml, ∼3389.8 ± 762.5 pg/ml, and ∼3664.4 ± 1240.8 pg/ml, respectively. In non-stimulated normal PB samples, levels of soluble IFN-γ, IL-1β, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, and IL-12 would typically be undetectable after a culture period of up to 4 hr.
Time Considerations Analysis of Cytokine Responses in Stimulated Whole Blood
Using current benchtop flow cytometers, the entire protocol is quite long, taking approximately half a day. The duration of the cellstimulation period depends on the combination of stimulated cells and the stimuli used. For T cells stimulated with PMA plus ionomycin and
for monocytes stimulated with LPS, minimum incubation periods of 3 to 4 hr are required. During the incubation period, both the bead mixture and the standard dilutions can be prepared. An additional 2 hr are required for the incubation of the stimulated samples with the mixture of cytokine-capturing beads. Incubation with monoclonal antibodies directed against cell surface molecules can be performed during the last 15 min of the incubation of the samples with the bead mixture. After these incubation periods, a 5-min washing step is required prior to data acquisition. Acquisition of the standards will take ∼1 min per tube and acquisition of the sample will typically take between 10 and 15 min per tube. Therefore, the overall time estimated for the whole process (considering a short-term activation) will vary between 7 and 8 hr.
Literature Cited Arai, K.I., Lee, F., Miyajimza, A., Miyatake, S., Arai, N., and Yokota, T. 1990. Cytokines: Coordinators of immune and inflammatory responses. Annu. Rev. Biochem. 59:783-836. Bueno, C., Almeida, J., Alguero, M.C., Sanchez, M.L., Vaquero, J.M., Laso, F.J., San Miguel, J.F., Escribano, L., and Orfao, A. 2001. Flow cytometric analysis of cytokine production by normal human peripheral blood dendritic cells and monocytes: Comparative analysis of different stimuli, secretion-blocking agents and incubation periods. Cytometry 46:33-40. Cook, E.B., Stahl, J.L., Lowe, L., Chen, R., Morgan, E., Wilson, J., Varro, R., Chan, A., Graziano, F.M., and Barney, N.P. 2001. Simultaneous measurement of six cytokines in a single sample of human tears using microparticle-based flow cytometry: Allergics vs. non-allergics. J. Immunol. Methods 254:109-118. Manz, R., Assenmacher, M., Pflüger, E., Miltenyi, S., and Radbruch, A. 1995. Analysis and sorting of live cells according to secreted molecules, relocated to a cell-surface affinity matrix. Proc. Natl. Acad. Sci. U.S.A. 92:1921-1925. Mascher, B., Schlenke, P., and Seyfarth, M. 1999. Expression and kinetics of cytokines determined by intracellular staining using flow cytometry. J. Immunol. Methods 223:115-121. Prussin, C. and Metcalfe, D.D. 1995. Detection of intracytoplasmic cytokines using flow cytometry and directly conjugated anticytokine antibodies. J. Immunol. Methods 188:117-128.
Key References Bueno, C., Rodriguez-Caballero, A., García-Montero, A., Pandiella, A., Almeida, J., and Orfao, A. 2002. A new method for detecting TNF-α-secreting cells using direct-immunofluorescence surface membrane staining. J. Immunol. Methods 264:77-87.
9.21.20 Supplement 25
Current Protocols in Cytometry
First description of the method of detection of TNFα on the membrane of the secreting cells, showing its sensitivity compared to the intra-cytoplasmatic detection. Chen, R., Lowe, L., Wilson, J.D., Crowther, E., Tzeggai, K., Bishop, J.E., and Varro, R. 1999. Simultaneous quantitation of six human cytokines in a single sample using microparticlebased flow cytometric technology. Clin. Chem. 45:1693-1694. Short report in which authors describe the beadbased immunoassay and its specificity, sensitivity, and range compared with a similar conventional ELISA assay.
Contributed by Arancha RodríguezCaballero, Andrés García-Montero, Clara Bueno, and Alberto Orfao Universidad de Salamanca Salamanca, Spain
Studies of Cell Function
9.21.21 Current Protocols in Cytometry
Supplement 25
CHAPTER 10 Data Processing and Analysis INTRODUCTION This chapter presents the principles of data processing and analysis for both flow and image cytometry. Flow and image cytometers have the capability to rapidly overwhelm a user with data. The amount of data available, and the precision and accuracy possible in its acquisition, provides researchers with enhanced capabilities in their experimental programs, but also puts great demands on their data processing methods. Database management, providing for optimized storage and access to the data, is often neglected in the operation of a laboratory. This is an especially important consideration when more than the raw data are stored (for example, data about the type of sample and its method of preparation, instrument parameters, and user and operator names). Without rapid and efficient methods of analysis, investigators would soon be reduced to relying on only subjective judgments of their data. Laboratories that operate several instruments with active research programs have learned that a thorough, well-maintained data processing operation is mandatory. Because the types of data obtained vary greatly, a large assortment of methods of analyzing data must be available. This chapter describes the various types of data that are obtained from flow and image cytometers, and presents the methods of data analysis that are currently available, with special attention to those that are available online utilizing microcomputers. The first three units deal with general aspects of data management and format. The requirements for a comprehensive data management program, which is of major importance with the rapid accumulation of data available with current computer-based instruments, are described in UNIT 10.1. To permit the sharing of data between laboratories, and to provide some uniformity in data formats from various instruments, a data file format standard (FCS 3.0), described in UNIT 10.2, has been developed. At present most data are acquired and stored in listmode format, whereby all measurements on each cell are stored separately in a list, rather than in histograms as with early instruments. The desirability of using listmode format and techniques for handling such data is described in UNIT 10.3. As the field of analytical cytology has expanded, the number of detectors on flow cytometers has increased as well, providing researchers with the opportunity to measure many variables on each cell. To process such data requires multidimensional data analysis methods; UNIT 10.4 describes methods that have been applied in multimarker immunophenotyping. Processing and analysis of image data, on the other hand, requires very different methods than those used in flow cytometry. An excellent introduction to the topic of two-dimensional image processing and analysis is presented in UNIT 10.5. The high data rates of flow cytometers make it imperative to use the best data presentation methods available. UNIT 10.6 discusses techniques designed to maximize user understanding of the data. Flow cytometers are used to collect data in an attempt to understand some aspect of the biological system being studied. As described in UNIT 10.7, mathematical models are developed to extract relevant population features from the data, to help users attain their goals. As the number of parameters measured on a single cell increases, the difficulty of analysis increases as well. Simple models and graphical methods no longer work and more sophisticated methods must be applied to the data. UNIT 10.8 describes the basic concepts of multivariate analysis and illustrates how such techniques are applied to data using both public domain and commercial software.
Data Processing and Analysis
Contributed by Phillip N. Dean
10.0.1
Current Protocols in Cytometry (2003) 10.0.1-10.0.2 Copyright © 2003 by John Wiley & Sons, Inc.
Supplement 23
Fluorescence in situ hybridization (FISH) techniques are frequently utilized in flow cytometry. One particular application is the detection of hybridization domains on intact metaphase chromosomes. UNIT 10.9 describes an automated software procedure developed to detect and locate these spots with high accuracy. describes the extension of conventional optical microscopy and image analysis techniques (described in UNIT 10.5) to the visualization and analysis of three-dimensional images, which are usually obtained through the use of confocal microscopy (see UNIT 2.8). UNIT 10.10
discusses two different methods of extracting morphological data from images, model-based and design-based. The first method uses mathematical models to derive such features as area, perimeter, shape, and brightness. These techniques necessarily require some knowledge of the objects being measured and are considered to be biased. The second method of analysis uses stereological methods and probabilistic geometry to extract data. These approaches make no assumptions as to the shape or size of the objects and therefore may be considered unbiased. Both model- and design-based methods are valid; selection of which to use depends on the circumstances. UNIT 10.11 describes both methods of analysis and their application.
UNIT 10.11
Researchers often make their data available to others for comparative analysis utilizing central data-storage devices/computers. This means that the data must be transmitted over a network, often over telephone lines. Even at 56K speeds, the large files now routinely collected require long times for data transfer. When repeated analyses are made, e.g., changing the parameters, repeated transfers of the data set are required, slowing the process even further. UNIT 10.12 presents a method whereby the central data-storage system contains a program that can generate data arrays based on requests from a remote user, and transmit only the display array, not the entire data set. Transfer times are such that many display arrays can be generated and transmitted in much less time than is needed for one transmission of the entire data set. Flow cytometers generate very large amounts of multivariate data that contain vastly more information than is extracted by current analytical methods. The relatively new field of data mining, defined as the process of information discovery, is now being applied to flow cytometry to help alleviate this problem. UNIT 10.13 describes this technique, using immunophenotyping of hematologic neoplasms as an example. Quantitative light microscopy can suffer from two physical problems with the microscope. Nonuniform illumination of the sample can result from the design of the light source, which is typically not a point source. Nonuniform detection efficiency of the light collection system, including the camera, can result from imperfect optics. UNIT 10.14 describes a simple method for correcting for these effects, using concentrated fluorophore solutions. In multicolor flow cytometry a given fluorescence detector will see to a greater or lesser extent emission from every fluorochrome present. The unwanted fluorescence is removed by a process known as compensation. This correction may be performed by hardware (UNIT 1.14), but if the compensation is not properly done the resulting data are useless. Alternatively, data can be compensated by software after collection. UNIT 1.15 discusses the advantages of software compensation and provides protocols for its accomplishment. Future updates to this chapter will describe additional methods devised to analyze data from both flow and image cytometry. Phillip N. Dean Introduction
10.0.2 Supplement 23
Current Protocols in Cytometry
Data Management This unit presents a broad view of flow cytometry data management, emphasizing ways to create and manage data so as to maximize its immediate and long-term usefulness. Often too little attention has been paid to this critical issue, resulting in the loss of valuable information. As a general rule, a management system needs to be more than a numerical listing of experimental results; to be comprehensive, it must incorporate a variety of information. There are a number of things that can be done to make the best use of whatever software a facility has available. With the current rapid developments in software design, exemplified by Web-based programming, flow cytometry data management can be expected to become easier and more flexible over time.
WHY FLOW CYTOMETRY DATA IS WORTH MANAGING Many flow cytometry facilities do not provide long-term data storage support to researchers, who are expected to copy and manage data for themselves on a long-term basis. One reason for this is the lack of comprehensive data management support in current software. Facility-wide data management may not seem necessary when only one individual knows the details of an experiment and can make sense of the recorded data. On the other hand, flow cytometry data files are often the result of costly and difficult work carried out over a period of months, as part of projects that may extend over several years. Therefore, it is important to get full use of these files and keep them from getting lost or becoming uninterpretable. When work on a project is being prepared for publication it is common to need to reanalyze early data in the light of what has been learned in later experiments. As flow cytometry measurements have become more complex, there are more and more reasons to reexamine old data. Therefore, careful data management is clearly worthwhile to ensure that the stored data contains enough information about an experiment to allow it to be analyzed and interpreted by someone else years later.
A BROAD VIEW OF DATA MANAGEMENT Requirements for Good Data Management The crucial first step in data management is Contributed by David R. Parks Current Protocols in Cytometry (1997) 10.1.1-10.1.6 Copyright © 1997 by John Wiley & Sons, Inc.
UNIT 10.1 recording the relevant information in a form that can be managed. Not all components of the recorded data may need to be managed explicitly, but unrecorded data are unmanageable. Archivist Terry Cook has discussed the formal criteria for creating data that will remain available and useful over time (Cook, 1995). Cook draws on the work of Richard Cox, David Bearman, and John McDonald to define a “set of needs for capturing, maintaining and using electronic records”: 1. “Records must be comprehensive: a record reflecting who, what, when, where, why, with whom, and so on, must be created for every business transaction.” 2. “Records must be authentic: authorizations for access to the data...must be recorded and traceable to each record and transaction.” 3. “Records must be tamper-proof: no deletion or alteration to a record should occur...If a record is changed or corrected, a second record must be created and linked to the first. Moreover, each use...of a record is also a transaction and thus must generate its own record.” Although Cook is specifically addressing the requirements for managing business data, essentially the same needs must be met by anyone wanting to ensure the validity of conclusions drawn from stored flow cytometry data.
Recording the Relevant Information The relevant information in flow cytometry data is much more than numerical lists of measurements. It is important that the data include specific information identifying the content and purpose of each sample, the machine parameters of the run, and any unusual features that could be important for future analysis. To date most commercial flow cytometry software has been oriented toward the display and analysis of data from single samples, and has provided very limited facilities for linking the full array of other possible information to cell measurement data. Staff of flow cytometry facilities should be familiar with the data enrichment options that their software does offer and should encourage facility users to take full advantage of them. As much of the relevant context information as possible should be included, linked to the numerical data. A data management system needs to be consistent in the types of information that are included. To avoid omitting impor-
Data Processing and Analysis
10.1.1
tant information, it is useful to define the types of information that routinely need to be specified, automatically including as much information from the instrument as possible; this is particularly helpful to novice users.
SPECIFIC FEATURES OF FLOW CYTOMETRY DATA MANAGEMENT There are a number of areas in which improvements may be made in the management of flow cytometry data, including ways for users and flow cytometry facility staff to help maximize the usefulness and longevity of data.
Experiment Structure Information
Data Management
Most flow cytometry experiments are quite highly structured in ways that are not conveyed by simply listing the cells and reagents contained in each sample: for example, some samples are controls; other samples may constitute a time series or a reagent titration series; or a number of samples may come from different tissues of the same mouse. A special designation should be given to samples containing fluorescence compensation standards. In systems using analog fluorescence compensation, single-dye-labeled samples are needed to set the fluorescence compensation adjustments or to check standard compensation settings. Data on these standard samples should be recorded to check the settings used in the rest of the experiment. When computed off-line compensation is used, the necessary coefficients are derived from an analysis of the single-dye samples. The standard format for flow cytometry data files, the Flow Cytometry Standard (FCS; UNIT 10.2), is single sample–oriented and is not structured to describe the relationships among different samples in an experiment or to relate one experiment to another. At present these must be recorded separately. Some efforts have been made to systematize the organization and use of experiment level information within particular software systems. For example, software under development at Stanford University (Treister et al., 1996) organizes all information about the samples in an experiment into one “workshop” document. The user can define and name groups of related samples within a workspace and apply complex analysis specifications to all the samples in a group in a single step. In a different approach to encoding experiment level information, a group at Purdue has defined a Tube Identification Parameter (TIP) and used it to discriminate different cell sam-
ples within a single composite FCS file (Robinson et al., 1991, 1992; Durack et al., 1991). The TIP is included as a parameter in each cell record in the listmode file and is incremented between cell samples. Gatings or other analyses common to the whole set of samples can be done in a single operation on all of the data in the file. By gating on the TIP, individual samples or groups of samples can be selected from the set. This method retains the structural information within the FCS file format, but the inclusion of different samples in one FCS file will often invalidate standard FCS keywords (such as the ones naming the reagents in each fluorescence channel) that are not common to all the samples in the set. These keywords represent the key distinguishing features of the samples for normal FCS compliant software, so although the system has proven useful, it also illustrates the need to incorporate higher-level information into the Flow Cytometry Standard in a systematic way. UNIT 10.2 describes the latest update of the standard format for flow cytometry data files, FCS 3.0. The next major advance in flow cytometry data standards should be to develop experiment- and project-level data specifications, but as of early 1997 there seem to be no plans to move in this direction within the FCS 3.0 project.
Sample Annotation Information It is is very important to be consistent in annotating samples to ensure that data are accessible and interpretable. Consistency is required both in the recording of relevant features and in the specification of a particular cell, reagent, or condition. Use of synonyms and different abbreviations from one experiment to the next makes it much harder to connect related data. Ideally all cell and reagent specifications should be drawn from a database. This ensures consistency in naming and provides a central place to keep information, such as the titers of different reagent lots, or data, such as the number of antibody molecules needed to produce one unit of fluorescence. With access to the latter information, a data analysis program can draw plots with axes based on antibody molecules per cell.
Instrument Conditions and Time-of-Collection Annotations In general, all of the computer-readable instrument settings used during data collection should be recorded so that they can be retrieved
10.1.2 Current Protocols in Cytometry
in the future (Parks and Bigos, 1990). It is redundant to record this fully in every FCS file, but at present there is no generally accepted way to store this information more efficiently and still ensure that a cell data sample remains linked to the instrument condition record. Instrument settings and test particle measurements are important data for verifying the cytometer measurements of cells. In addition, text annotations of observations during the measurement and, where needed, notes clarifying the data (e.g., explaining different sets of data collected to document a sort) should be recorded. The time of collection is an important part of the data. This makes it possible to check the order in which samples were run and provides a simple link to time-stamped log files. The data rate or duration of the data collection can also be useful in evaluating unexpected data results.
Numerical Measurements For logarithmic amplified immunofluorescence measurements, 9-bit data (steps of ~2% per channel) are adequate. Higher resolution may be useful for cell cycle analysis. Measurements used to gate the recorded data events should be included in the data collection. Such gating is usually used to exclude dead cells or debris, but it is a good idea to set “loose” gates and refine them in the data analysis process. Thus, the values used for gating should be recorded. The number of data events that should be collected depends on the frequencies of populations to be evaluated and on the required accuracy of those evaluations (Parks and Bigos, 1997). In general, the lower the frequency of the population of interest, the more data events are needed.
Data Compression Most software allows data storage in one- or two-dimensional plots as well as in list form. When computer storage was expensive and data were limited to two or three measurements per cell, histograms provided an adequate alternative; however, with ample computer storage currently available, there is little reason to record primary data in anything but listmode (see UNIT 10.3 for more extensive discussion of this point). It is still valuable to keep the lists as compact as possible, both to make the best use of storage space and to minimize transmission time during data transfer. The use of bit packing rather than full bytes is advantageous, particularly for 9- to 12- bit
data, as it takes only 56% (9-bit) to 75% (12-bit) as much space. Bit-packed data conforms to the FCS. Unfortunately it appears that most commercial flow cytometry software does not support it. For large data files with lower bit resolution and nonuniform distributions of data values, Huffman encoding can be more efficient than packed bits (Bigos and Moore, 1996). Tested on an eight-dimensional 50,000cell data set, Huffman encoding gave 18% more compression than simple packed-bit storage at 8-bit resolution but only 6% at 12 bits.
Data Storage, Maintenance, and Recovery At present it seems that most flow cytometry facilities rely on the underlying structure of their computer file system to manage their data. With this arrangement consistent organization of data files into experiments and projects is not ensured. To facilitate consistent data management, each instance of data collection needs to have a unique identifier assigned so that any outputs (e.g., plots or tables) derived from that data can be tagged with the identifier and traced back to the original data source. The FCS (UNIT 10.2; Data File Standards Committee of the Society for Analytical Cytometry, 1990) has been defined to provide consistent data forms and facilitate data interchange. Files written to this standard should be readable by any program that supports FCS (as almost all commercial analysis programs claim to do). At this time, however, full intercompatibility has not been realized. That will require production of standard coding/decoding packages that implement the full FCS and can be incorporated into all flow cytometry software, an effort now being undertaken by the FCS Committee. Current data collections often include eight to ten primary fluorescence and light-scatter measurements. As the ability to delineate lowfrequency cell populations increases, the average number of cells that must be measured also increases. This increase in the number of measurements and the number of cells measured has led to the production of very large data files. These days, however, data storage is the simplest aspect of data management. Large, inexpensive disks make it easy to keep a large quantity of data online, and high-volume tape storage is inexpensive and compact. Recovery of data from tape, however, is slow, and the life expectancy of data tapes is shorter than the minimal 10 years that is appropriate for flow
Data Processing and Analysis
10.1.3 Current Protocols in Cytometry
cytometry data storage. Optical compact disks are now coming into routine use and are much better than tape in these two respects. Their only limitation is that they generally are not rewritable (see UNIT 10.3), so the original data files cannot be updated with analysis specifications and results. It is possible to keep all experiment and sample information, including the analysis specifications and results, online while storing the numerical data themselves elsewhere. In this case the numerical data are separated from the online data, which must include information for recovery of the numerical data.
Records of Data Analysis History and Results
Data Management
Routine data analysis includes data gating, specification of plots to display signal distributions, and computation of statistical results. Records of these analyses should be retained with the rest of the information about the sample. Graphs and tables derived from data (e.g., fluorescence signal medians for export to a spreadsheet program) should include sufficient information to trace back to the original flow cytometer data records. Computed data transformations can be an important part of the data analysis record. Analog fluorescence compensation is discussed in Parks (1997). Off-line fluorescence compensation, resulting in the computation of new data dimensions, can be useful when stored fluorescence data are uncompensated or undercompensated. In multiple-laser systems, a dye may be excited by more than one laser, leading to unwanted signal contributions that cannot be corrected by ordinary analog fluorescence compensation. In this case, off-line compensation may be required prior to analysis of cell populations and extraction of numerical results. Analysis of data from singly stained cells of each type provides the subtraction coefficients used to correct mixed dye results. When off-line fluorescence compensation has been done, the matrix that specifies the transformation should be recorded as part of the analysis. This documents the transformation and allows comparisons between the adjustments required in different experiments. The FCS 3.0 includes space for adding analysis results to the data file (UNIT 10.2; http://nucleus.immunol.washington.edu/ISAC. html). Because there is no fixed format for analyzing data, results are recorded as unformatted text. Therefore, no standard method of retrieving specific analysis results is available. One general method for recording cell popu-
lations is to define new 1-bit list parameters that identify which cells fall into particular gated populations. One-bit lists may be an effective way to represent and store the results of an analysis, particularly when the analysis identifying a population is complex, such as that derived from a cluster analysis.
Database Use At present flow cytometric databases have several useful roles. They provide a flexible way for the user to organize information; thus, database interactions may well become the predominant way in which flow cytometry users interface with the data. For example, rather than starting a data analysis program and naming a file for it to operate on, the database could be asked for experiments from the previous month under a specific name. An analysis program could then be launched on one of these experiments. If experiment and sample information have been entered fully and consistently, a database will be able to provide useful responses to queries such as “Who has been doing staining for CD192 recently?” or “Which experiments used SJL mice?” Flow cytometry service facilities and larger laboratories should organize their current data collections with the expectation that they will soon be porting it into a database. Should this not happen, they will certainly benefit from better-organized data in any case.
Security and Access Control It is good practice for a service facility to retain a copy of all data it generates. However, when data are well documented, controlling their access becomes a critical issue; unauthorized competitors might be able to steal the results. Therefore, a center with central data management has a responsibility to prevent unauthorized use of data. Although most flow cytometry data should be publicly available after publication, at certain times users should be allowed to restrict access to their data. Another major aspect of data security is the prevention of accidental or intentional data loss. Users should be allowed to “delete” data that they believe to be incorrect or obsolete; however, the system should retain these data but add a “deleted” annotation so that it does not appear in routine data listings.
Stanford University Experience with the DESK System An example of a comprehensive data man-
10.1.4 Current Protocols in Cytometry
agement system for flow cytometry is the DESK system developed at Stanford University. DESK is a comprehensive system for designing flow cytometry experiments and for collecting, annotating, managing, and analyzing flow cytometry data (Moore, 1984, 1987; Moore and Bigos, 1990). It has been in use since the mid-1980s at Stanford and in a small number of other laboratories around the world. The data management aspects of DESK have proven to be very valuable in organizing large volumes of data and in keeping it usable and accessible over a period of years. Based on this experience, a number of conclusions and recommendations can be made: 1. The experiment design editor in DESK provides for one-time entry of cell and reagent information for an experiment and makes it easy to specify cell and reagent combinations for all samples in the experiment. This is useful for ensuring consistency in cell and reagent names among all samples in the experiment. This consistency can be maintained over a series of related experiments by copying and re-editing the original design document. Ideally, the reagent labels should be obtained from a database that would contain full information on each reagent. The experiment design editor makes it possible to code related samples into parallel positions in a grid array, but future software should provide explicit ways to group samples and specify other relationships within an experiment. 2. For each cytometer the DESK system maintains a log of the instrument configurations and adjustments and of the results of the instrument standardization run that precedes each data collection run. The log is valuable for retrospective investigation when users obtain questionable data, and it can be used to monitor trends and abrupt changes in the operating conditions of the instrument. 3. Each data collection record is automatically assigned a unique identifier, which assures that different data sets can never be confused with each other even if all the user-supplied labels are the same. 4. Automatic data identification means that the user never has to deal with actual file names in the computer system. When a user specifies analyses on a sample whose numerical data are not online, a request for the appropriate data storage tape is automatically dispatched. After a request tape is loaded, the required data are transferred to a disk, the requested analysis is performed, and the results are returned to the user’s DESKtop file automatically.
5. For each cell sample the user’s DESKtop retains a record of all analysis gating, specifications for each plot produced, and the results of numerical analyses. This makes it possible to reconstruct previous analyses exactly. Printouts of plots and numerical results include user, date, experiment, sample, cell, and reagent labels as well as gating information leading to the data subset plotted. Thus, data outputs can always be traced back to the data from which they were produced. 6. The DESK data archive lists all experiments in the system by user name, date, and title. The lists can be browsed and used to gain access to the data in archived experiments. This archive consists of lists and is not a general database, and it only includes experiment level information. Data systems should include a real database that contains both the experiment level and sample level information.
LITERATURE CITED Bigos, M. and Moore, W.A. 1996. Data compression for flow cytometry. Cytometry 8(Suppl.):125. Cook, T. 1995. It’s 10 o’clock—Do you know where your data are? Technol. Rev. 98:48-53. Data File Standards Committee of the Society for Analytical Cytology, 1990. Data file standard for flow cytometry. Cytometry 11:323-332. Durack, G., Lawler, G., Kelley, S., Ragheb, K., Roth, R.A., Ganey, P., and Robinson, J.P. 1991. Time interval gating for analysis of cell function using flow cytometry. Cytometry 12:701-706. Moore, W.A. 1984. A modern computer system for collection, management and analysis of flow cytometry data. XII International Meeting of the Society for Analytical Cytology Abstr. C49. Moore, W.A. 1987. Software for management of flow cytometry data. Cytometry 1(Suppl.):12. Moore, W.A. and Bigos, M. 1990. A distributed data processing system for flow cytometry. Cytometry 4(Suppl.):88. Parks, D.R. 1997. Flow cytometry instrumentation and measurements. In Handbook of Experimental Immunology, Vol. 2 (L.A. Herzenberg, D.M. Weir, C. Blackwell, and L.A. Herzenberg, eds.) pp. 50.1-50.11. Blackwell Science, Cambridge, Mass.. Parks, D.R. and Bigos, M. 1990. Automatic standardization, compensation adjustment and instrument diagnosis for flow cytometers. Cytometry 4(Suppl.):70. Parks, D.R. and Bigos, M. 1997. Collection, display, and analysis of flow cytometry data. In Handbook of Experimental Immunology, Vol. 2 (L.A. Herzenberg, D.M. Weir, C. Blackwell, and L.A. Herzenberg, eds.) pp. 47.1-47.12. Blackwell Science, Cambridge, Mass. Data Processing and Analysis
10.1.5 Current Protocols in Cytometry
Robinson, J.P., Durack, G., and Kelley, S. 1991. An innovation in flow cytometry data collection and analysis producing a correlated multiple sample analysis in a single file. Cytometry 12:82-90. Robinson, J.P., Ragheb K., Lawler, G., Kelley, S., and Durack, G. 1992. Rapid multivariate analysis and display of cross-reacting antibodies on human leukocytes. Cytometry 13:75-82. Treister, A.S., Bigos, M., Herzenberg, L.A., Moore, W.A., Parks, D.R., Roederer, M., and Herzenberg, L.A. 1996. Experiment-based flow cytometry data analysis software. Cytometry 8 (Suppl.):121-121.
INTERNET RESOURCES http://nucleus.immunol.washington.edu/ISAC.html Homepage of the International Society for Analytical Cytology (ISAC), with link available to a full listing of FCS 3.0.
Contributed by David R. Parks Stanford University Stanford, California
Data Management
10.1.6 Current Protocols in Cytometry
Data File Standard for Flow Cytometry, FCS 3.0 BACKGROUND In the early days of flow cytometry, both commercial and “home-built” instruments acquired and saved data in unique file formats designed to meet their builders’ own needs. Because of this mixture of unique formats, data could not easily be shared among instruments, individuals, or institutions. Also, users were limited to the analysis routines and graphics displays included with the instrument on which the data files were acquired. In 1984 Murphy and Chused addressed these problems by proposing a standard format for flow cytometry data files, Flow Cytometry Standard 1.0 (FCS 1.0; Murphy and Chused, 1984). The purpose of the data-file standard was to provide a clearly defined and uniform file format that would allow data collected by one instrument to be correctly read for analysis by other software on another computer. FCS 1.0 introduced the basic four-part file structure—consisting of HEADER, TEXT, DATA, and ANALYSIS (described in more detail below)—that is still used in the current version. In 1990 the Society for Analytical Cytology (now the International Society for Analytical Cytology, or ISAC) formed a Data Files Standards Committee to recommend a standard datafile format (Dean et al., 1990). The committee used the FCS 1.0 file structure as the basis for an updated file standard, FCS 2.0. The new standard was designed to accommodate changes in technology and in the needs of the flow cytometry community since FCS 1.0 had been proposed. Several new features, including required keyword-value pairs and support for saving multiple data sets in a single file, were introduced in the updated standard. Because the FCS was both simple and flexible, it gained wide acceptance and has now been implemented by most instrument manufacturers.
RECENT DEVELOPMENTS Once again, because of advances in biotechnology, computer technology, and data communications, the data-file standard has had to be revised. The new version, FCS 3.0, has now been developed by the Data File Standards Committee (Table 10.2.1). There were several problems with earlier FCS versions that needed to be addressed. (1) 8-byte HEADER fields limited the size of data sets to <100 megabytes Contributed by Larry Seamer Current Protocols in Cytometry (1997) 10.2.1-10.2.5 Copyright © 1997 by John Wiley & Sons, Inc.
UNIT 10.2
(see discussion of HEADER below for a detailed explanation). With the growing number of measurement parameters, the use of high-bit analog-to-digital converters (ADCs), the requirements of rare-event detection, the long kinetic files that are now common, and the inclusion of calculated parameters, list-mode data files now approach and occasionally reach the 100-megabyte limit. (2) Previous versions of the FCS were limited to the 256-character ASCII text set. Flow cytometry is international in its scope; therefore, text portions of data files should be able to accommodate a broader range of internationally used text characters. (3) Computer network data transfer has become a common means of moving data; therefore, a mechanism to verify data integrity is now needed. (4) As third-party analysis software has become widely used, it has become increasingly desirable for files to include more descriptive information regarding data acquisition and signal amplification. (5) Development of cluster-analysis software to help identify and classify cell subsets has created a need for new keyword support in the FCS (Redelman and Coder, 1994). The goal in revising the FCS was to meet these needs while causing as little disruption as possible to those who create data acquisition software (instrument manufacturers), to those who write analysis software, and especially to end-users. In contemplating format changes, retaining backwards compatibility with previous versions (defined here as the ability of previous releases of analysis software to correctly read and interpret FCS 3.0 data files) was a priority. The FCS 3.0 Standard is accessible in its entirety through the World Wide Web homepages of the ISAC (http://nucleus.immunol. washington.edu/ISAC.html) and of the John Wiley & Sons, Inc., journal Cytometry (http://journals.wiley.com/cytometry).
FILE ORGANIZATION FCS 1.0 introduced the basic data-file structure that has been maintained in all subsequent versions, including the current proposal (FCS 3.0). All FCS-compliant data files are divided into four segments: HEADER, TEXT, DATA, and ANALYSIS (the last optional). Data Processing and Analysis
10.2.1
HEADER The HEADER segment must always be located at the beginning of the file. The first 10 bytes of the HEADER (and therefore of the file) give the FCS version in ASCII text (e.g., FCS 3.0) followed by four ASCII space characters. The remainder of the HEADER segment provides byte offsets to the other segments of the file. Each offset value occupies exactly 8 bytes, right justified. The positions of the byte offsets are given in Table 10.2.2. The 8-byte limitation of the byte offset value establishes a file size limit of 99,999,999 bytes. The proposed FCS 3.0 standard circumvents file-size limits through the following mechanism: If any segment of a data set falls outside the 99,999,999 byte limit, “0”s are placed in the
HEADER segment for both the begin and end byte offsets for that segment. The byte offset values for that segment are then found in keyword-value pairs in the data-set TEXT segment. The new byte offset keywords are as follows: $BEGINDATA, $ENDDATA, $BEGINANALYSIS, $ENDANALYSIS, $BEGINSTEXT, and $ENDSTEXT. There are no byte offset keywords for the beginning and ending of the primary TEXT segment because that segment must be placed entirely within the first 99,999,999 bytes of a data set. For example, if the DATA segment starts at byte 257 and ends at byte 100,345,679, HEADER fields 2633 (begin DATA) and 34-41 (end DATA) would contain the value “0”. The actual byte offsets would be found in the $BEGINDATA and
Table 10.2.1 Data File Standards Committee of the International Society for Analytical Cytology (ISAC)
Larry Seamer, Chair Director, Flow Cytometry Facility University of New Mexico Cancer Center, Cytometry 900 Camino de Salud NE Albuquerque, New Mexico 87131 TEL (505) 277-6206 [email protected] Bruce Bagwell Verity Software House P.O. Box 247 Topsham, Maine 04086 [email protected] Luther Barden Div. of Computer Research and Technology Building 12A Room 2015 National Institutes of Health 9000 Rockville Pike Bethesda, Maryland 20892 [email protected] Marc Christofferson Becton Dickinson Immunocytometry Systems 2350 Qume Drive San Jose, California 95131-1807 TEL (408) 954-2058 [email protected]
Data File Standard for Flow Cytometry, FCS 3.0
Louise E. Magruder Division of Clinical Laboratory Devices FDA/CDRH/ODE 72 Gaither Road Rockville, Maryland 20850 [email protected]
George Malachowski Cytomation, Inc. 400 East Horsetooth Road Fort Collins, Colorado TEL (303) 226-2200 Robert F. Murphy Department of Biological Sciences and Center for Light Microscope Imaging and Biotechnology Carnegie Mellon University 4400 Fifth Avenue, Box 52 Pittsburgh, Pennsylvania 15213 TEL (412) 268-3480 [email protected] Doug Redelman Sierra Cytometry 3150 Susileen Dr. Reno, Nevada 89509 Gary C. Salzman Life Sciences Division Los Alamos National Laboratory Mail Stop M888 Los Alamos, New Mexico 87545 TEL (505) 667-5503 [email protected] James C.S. Wood Coulter Corporation Mail Code 52-A01 11800 SW 147th Avenue Miami, Florida 33196-2500 TEL (305) 380-2449 or 344-1290 FAX (305) 344-5240 [email protected]
10.2.2 Current Protocols in Cytometry
Table 10.2.2
HEADER Segment Positions of ASCII(32)-Encoded Byte Offsets
Contents
Bytes
FCS 3.0
00–05
ASCII(32) space characters
06–09
ASCII-encoded offset to first byte of TEXT segment
10–17
ASCII-encoded offset to last byte of TEXT segment
18–25
ASCII-encoded offset to first byte of DATA segment
26–33
ASCII-encoded offset to last byte of DATA segment
34–41
ASCII-encoded offset to first byte of ANALYSIS segment
42–49
ASCII-encoded offset to last byte of ANALYSIS segment ASCII-encoded offset to user-defined segments
50–57
$ENDDATA keyword values in the data-set TEXT segment. This system optimizes backwards compatibility. In the rare event that a data set is ≥100 megabytes, file-reading programs designed for previous FCS versions will read the “0” byte offset and fail to read the data, avoiding partial or erroneous data reads. However, for the vast majority of data sets that are <100 megabytes, software designed to read previous FCS files will have no trouble correctly reading FCS 3.0-compliant files.
TEXT New to FCS 3.0 is the allowance for primary and supplemental TEXT segments. The primary TEXT segment is required. It must be located completely within the first 99,999,999 bytes of a data set and must contain all required keyword-value pairs, as well as any number of optional keyword-value pairs. The byte offsets to the TEXT segment in the HEADER point to the primary TEXT segment. The first character of the primary TEXT segment contains the delimiter character. The delimiter separates keywords from the keyword value. The remainder of the TEXT segment contains a series of keyword-value pairs that describe various asp ec ts o f the d ata set. For example, “$TOT/5000/” is a keyword-value pair indicating that the total number of events in the file is 5000. “$TOT” is the keyword, “5000” is the value, and “/” is the delimiter character. The “$” character flags this keyword as a standard FCS keyword. The supplemental TEXT segment is optional and may be located anywhere in the data-set after the HEADER segment. The supplemental TEXT segment can contain only optional keyword-value pairs. The byte offsets to
58–beginning of next segment
the supplemental segment are found in the values for the keywords $BEGINSTEXT and $ENDSTEXT, which are placed in the primary TEXT segment.
DATA The DATA segment contains raw data in one of three modes (list, correlated, or uncorrelated), which is indicated in the TEXT segment by the $MODE keyword value. The data are written to the DATA segment in one of four allowed formats (binary, floating-point, double-precision floating-point, or ASCII), which is indicated by the $DATATYPE keyword value. The most common form of data storage is list-mode storage in the form of binary integers ($DATATYPE/I/ $MODE/L/). The $PnB set of keywords specifies the bit width for the storage of each parameter. The $PnR set of keywords specifies the channel number range for each parameter. For example, $P1B/16/ $P1R/1024/ specifies a 16-bit field for parameter 1 and a range for the values of parameter 1 from 0 to 1023. This implies that 10 bits of the 16-bit field are used to store the data. The remaining bits are usually unused and set to “0”; however, some file writers store non-data information in that bit space. Implementors should use a bit mask when reading these listmode parameter values to ensure that erroneous values are not read from the unused bits.
ANALYSIS ANALYSIS is an optional segment that, when present, contains the results of data processing. The ANALYSIS segment has the same keyword-value pair structure as the TEXT segment. There are no required keywords for the ANALYSIS segment. The ANALYSIS seg-
Data Processing and Analysis
10.2.3 Current Protocols in Cytometry
ment typically contains information added to a copy of the original file. For example, the results of cell cycle analysis or immunophenotype determinations often involve more complex analyses than can be performed in “real time” as the data are collected and stored. When these analyses are performed the results can be added to the file within the structure of the ANALYSIS segment.
MAJOR NEW FEATURES IN FCS 3.0
Data File Standard for Flow Cytometry, FCS 3.0
FCS 3.0 is different from FCS 2.0 in the following ways. 1. The FCS 3.0 TEXT segment may be split into primary and supplemental TEXT segments. The primary TEXT segment must contain all required keyword-value pairs and be located entirely within the first 99,999,999 bytes of a data set. The supplemental TEXT segment can contain only optional keywordvalue pairs and may be located anywhere in a data-set after the HEADER segment. The byte offsets to the primary TEXT segment are found in the HEADER segment. The byte offsets to the supplemental text segment are found in keyword-value pairs in the primary TEXT segment. 2. The HEADER has been modified to accommodate data sets longer than 99,999,999 bytes. A data-offset value that requires more than 8 bytes is now represented by placing a “0” in the HEADER for both the begin and end values of the DATA segment. The actual byteoffset values are found in the primary TEXT segment of the data set. The new keywords that describe offset values are $BEGINSTEXT and $ENDSTEXT for the byte offsets to the beginning and end of the optional supplemental TEXT segment, $BEGINDATA and $ENDDATA for the byte offsets to the beginning and end of the DATA segment, and $BEGINANALYSIS and $ENDANALYSIS for the byte offsets to the beginning and end of the optional ANALYSIS segment. This system allows the vast majority of data files to be backwards compatible with analysis software designed for previous FCS versions. However, a “0” byte offset in the HEADER will prevent these older versions of analysis software from reading very large data sets, avoiding read errors or partial data reads. 3. An optional 16-bit Cyclic Redundancy Check (CRC) word has been added to the end of each data set. This internal check word allows for data-set integrity checks (Press et al., 1992).
4. To enable third-party or off-line analysis software to read and interpret data correctly, the keyword $PnE is now required for each parameter. The $PnE keyword describes the method of amplification used for a given parameter using two floating point values f1 and f2. With logarithmic amplification, f1 specifies the number of logarithmic decades and f2 gives the linear value that would have been obtained for a signal with a log value of 0. When linear amplification is used, f1 and f2 are set to 0. For example, “/$P1E/3,0.1/” specifies that the data for parameter 1 were collected using a threedecade logarithmic amplifier and the 0 channel is equivalent to the linear value 0.1. 5. The new keyword $COMP has replaced $DFCiTOj to describe the amount of fluorescence compensation employed in the collection of the data. The keyword value of $COMP represents a compensation matrix. The matrix has n rows and n columns, where n is the number of parameters. Both positive and negative values are allowed. A positive or unsigned value indicates additive compensation, while a negative value indicates the more common case of subtractive compensation. The elements are stored in row-major order, i.e., the elements in the first row appear first. For example, with a four-parameter file, matrix element seven of sixteen total elements (which would fall in row 2, column 3) indicates the percentage of parameter 3 that has been electronically subtracted from or added to parameter 2. 6. There are a number of other new optional FCS TEXT Segment keywords. $CSMODE, $CSTOT, $CSVBITS, and $CSVnFLAG specify added parameters to identify cell subsets. $CYTSN specifies the cytometer serial number. $TIMESTEP has been added to more accurately define how a TIME parameter is measured. The following optional ANALYSIS segment keywords have been added to enable specification of cell subsets: $CSDATE, $ CSDEFFILE, $CSEXP, $CSnN, and $CSnNUM. 7. $UNICODE enables the specification of certain keywords in languages that cannot be represented in ASCII. UNICODE is an international standard that enables computer representation of most of the world’s languages (Unicode Consortium, 1991). The characters for each language are represented as 2-byte codes. The first byte represents the UNICODE “page” on which the character is found and the second byte represents the individual character. There are 65536 characters available.
10.2.4 Current Protocols in Cytometry
8. The keyword values representing a year are increased by two bytes to include a century designation (e.g., 1996 rather than 96) for the keywords that represent a date ($DATE and $CSDATE). 9. The $PnG keyword has been added, describing the linear gain applied to a signal.
SUMMARY The FCS Data File Standard has evolved to keep pace with technology and current practice in the field of flow cytometry. In 1984 few envisioned that the cost of data storage would reach 3 cents per megabyte, that data file size would approach 100 megabytes, or that data would routinely be transmitted world-wide. FCS 3.0 reflects the state of the art of flow cytometry in 1996 while retaining a high level of backwards compatibility with previous versions of the standard.
Literature Cited Dean, P.N., Bagwell, C.B., Lindmo, T., Murphy, R.F., and Salzman, G.C. 1990. Data file standard for flow cytometry. Cytometry 11:323-332.
Press, W.H., Teukolsky, S.A., Vetterling, W.T., and Flannery, B.P. 1992. Numerical Recipes in C., 2nd ed. Cambridge University Press, Cambridge, UK. Redelman, D. and Coder, D.M. 1994. Cell subset (CS) parameter to record the identities of individual cells in flow cytometric data. Cytometry 18:95-102. Unicode Consortium. 1991. The UNICODE Standard, Version 1.0, vol. 1. Addison-Wesley Publishing, Reading, MA.
INTERNET RESOURCES http://nucleus.immunol.washington.edu/ISAC.html Homepage of the International Society for Analytical Cytology (ISAC), with link available to a full listing of FCS 3.0. http://journals.wiley.com/cytometry Homepage of the John Wiley & Sons, Inc., journal Cytometry, with link available to a full listing of FCS 3.0.
Contributed by Larry Seamer University of New Mexico Albuquerque, New Mexico
Murphy, R.F. and Chused, T.M. 1984. A proposal for a flow cytometric data file standard. Cytometry 5:553-555.
Data Processing and Analysis
10.2.5 Current Protocols in Cytometry
Listmode Data Processing BACKGROUND Once signals are received and digitized by the analog-to-digital converters (ADCs) of a flow cytometer, they are recorded as data that can then be processed by various means. There are essentially three ways to record flow cytometric data: as single-parameter “histogram” data, two-parameter “bivariate histogram” data, and multiparameter “listmode” data (i.e., where the data are taken serially in a list on a cell-by-cell basis). The advantages and disadvantages of each system revolve around two central issues: the amount of information about each cell that one wishes to store and analyze (and the method of analysis to be used), and the cost in space, money, and time of doing so. The processing of listmode data from flow cytometry experiments has become increasingly sophisticated over the past 25 years, as discussed in a number of recent reviews (Dean, 1990; Watson, 1992; Bagwell, 1993). As flow cytometry listmode data are not inherently any different from similar data from other fields, listmode data processing has borrowed heavily from techniques developed in fields such as image processing.
Effective Storage of Data If only one parameter is of importance, storing data in the form of a histogram is the most effective method in terms of both information content and space. If two parameters are measured per cell, however, the issue becomes more complex. It is necessary to decide whether it is sufficient to view the values of only one of the parameters provided the other falls within a “valid” range, or whether the exact values of each parameter need to be seen on a cell-by-cell basis. In the former case, data can be stored as a one-parameter histogram “gated” (see Gating of Data) on a valid subrange of a second parameter; this format is quite commonly used, and perhaps for good reason, because it is simple and contains only the data of interest. In the latter case, the data should be stored as one or more two-dimensional arrays (“bivariate data files”), each of which can then be gated on one or more additional parameters. Unfortunately, many experiments are sufficiently complex that it is far simpler to store all of the data ungated, as a multiparameter listmode data file, rather than as multiple gated-histogram data files. This also eliminates the need to make a
Contributed by James F. Leary Current Protocols in Cytometry (1997) 10.3.1-10.3.5 Copyright © 1997 by John Wiley & Sons, Inc.
UNIT 10.3
priori decisions regarding data processing. Gated histograms and bivariate displays can be easily generated or regenerated from the listmode data file; on many computer systems this can even be done in a semiautomated fashion. Devoting computer disk space to storing extra data is much less expensive than rerunning an experiment because data were not collected due to improper gating.
Gating of Data Gating requires that the data satisfy one or more conditions: typically that they lie within some boundary, which may be defined by lower and upper limits of the parameter for gated single-parameter data or by a more general shape for data with two or more parameters. The ability to gate one parameter on another can be extended to gating one parameter on multiple parameters or gating multiple parameters on “valid” ranges of each parameter. Another advantage of gating is that some data analysis can be performed in “real time.” For experiments involving large numbers of samples for which similar types of data must be collected, this can be extremely important. Instruments processing clinical samples can easily process the results of a sample in the 1 to 2 min that it takes to run the next sample. It is technically possible to produce a report containing results for a large number of patients soon after the last sample is run. If the desired data analysis is well established and does not require user intervention, it can be entirely automated.
Bivariate Displays The power of bivariate data can be increased further by gating on a third, fourth, and even more parameters. If more than three parameters must be measured (and remain correlated) on a cell-by-cell basis, there are serious limitations in the bivariate data approach. To represent n-parameter data as all possible pairwise (r = 2) permutations (and if one considers y vs. x the same as x vs. y), the number of bivariate data sets required is: n! r !( n − r )!
∆
For example, to generate bivariate displays of all possible two-parameter combinations of Data Processing and Analysis
10.3.1
four-parameter flow cytometry listmode data, the number of bivariate displays required is: 4! 2 !( 4 − 2 )!
=
4 × 3× 2 ×1 2 × 1(2 × 1)
= 6
The number of bivariate displays required rises rapidly as n increases; with six parameters per cell, it is: 6! 2 !(6 − 2 )!
=
6 × 5 × 4 × 3× 2 ×1 2 × 1( 4 × 3 × 2 × 1 )
= 15
It should be obvious from the above example that with more than four parameters per cell, it is far simpler and, depending on the bit resolution of the stored data, may be more economical to store the data as listmode data. It is important to note that some cell subpopulations may not be visible in any of the possible bivariate displays. This is because cell subpopulations in multiparameter dataspace may be projected down on top of each other in bivariate space due to the specified angle of projection, much as cell subpopulations in bivariate dataspace may be projected down on top of each other in a histogram. Two solutions to this problem are possible: to display data in three-dimensional data sets or to use principal components (Jolliffe, 1986) to choose a more favorable projection angle.
More Advanced Data Visualization Methods Modern flow cytometry data analysis packages, both commercial and home-built, apply “color gating” to subsets of data identified by gating in one bivariate display, in order to visualize the correlation of the parameters of those data in another one of the possible bivariate displays. More advanced software allows dynamic “brush plotting” of the data not only in all the bivariate displays but also in rotating three-dimensional displays of trivariate combinations of the data. Color gating is described as part of the discussion of multidimensional data analysis in UNIT 10.4.
Storage of Large Amounts of Data
Listmode Data Processing
When considering data storage requirements, it should be kept in mind that histogram data generally require little space, and this requirement does not increase with the number of cells measured. This assumes that no individual channel overflows with more data than can be stored—typically 65,000 data points on software written to accept 16-bit integers for cell numbers in a given channel. However, with
computers using 32-bit software, even this limitation is not inherent. Similarly, and with the same proviso, bivariate data do not require more space when more cells are measured. Listmode data, however, require more data space in direct proportion to the number of parameters per cell and the number of cells measured. For a relatively small number of cells and four or more parameters measured per cell, it is more economical in terms of space to store data in listmode. For larger data sets (e.g., 100,000 cells or more), listmode data storage requirements become considerable. For very large data sets (e.g., 100,000,000 cells or more), such as are needed in the analysis of very rare cell subpopulations, space requirements make it unrealistic to store data from all cells as listmode data. Such data need to be partially classified in real time as they are being acquired, using either “live” data acquisition gates or, on more advanced research instruments, real-time data classification to ensure that only data of interest are stored for subsequent off-line analyses. With the dramatic decrease in the costs of computer memory and disk space and with the ability of modern computers to store megabytes and gigabytes, storage space is no longer as important a criterion for selecting a data format as it once was. Now, except in the extreme case of multiparameter analysis of rare cell subpopulations, the most important criterion is maintaining the degree of correlation between parameters for subsequent data analysis. The result is that researchers are moving fairly rapidly toward acquiring data sets of modest size (e.g., 10,000 to 30,000 cells at four to six parameters per cell) as listmode data that can subsequently be analyzed without prior gating limitations. Even most clinical instruments are now routinely capable of listmode data acquisition and analysis.
DATA FILE STRUCTURES There are three basic types of data files: histogram, bivariate histogram, and listmode. Each of these file types is structured to efficiently store and reprocess the corresponding data type. It is very important to try to anticipate the type of data analysis that will be performed, because saving data in a particular data file structure may preclude some alternative forms of data analysis (either existing or still to be developed). There are advantages and disadvantages to each data file structure, and these depend on the needs and resources of the experimenter.
10.3.2 Current Protocols in Cytometry
Histogram Data Files All flow cytometric data files, whether histogram, bivariate, or listmode, result from grouping, or binning, the data from each single flow cytometric parameter into a number of bins whose resolution is determined by the bit resolution of the ADC that digitized the original data. For example, a typical 10-bit-resolution ADC found in commercial instruments will group the data into 210 (1024) bins, or “channels”; a 12-bit-resolution ADC will bin the data into 212 (4096) bins; and so on. For some flow cytometric applications, such as ploidy analysis and chromosome analysis, where both precision and a wide range of signals are required, ADCs with higher bit resolutions (e.g., 12 or even 16 bits) are used. The critical thing to remember about histogram data files is that each histogram is totally uncorrelated with any other histogram generated in the experiment. For example, acquiring green fluorescence and light scatter data as individual histograms provides no way of determining the light scatter of any particular cell with a given amount of green fluorescence, or vice versa. Typically a single parameter (e.g., fluorescence) is acquired and gated on one or more other parameters (e.g., forward or side scatter). Thus, histograms, although they take very little computer or disk space, are an important way of quantitating the data from a single flow cytometric parameter. However, this data file structure fails to take advantage of the powerful capabilities of the flow cytometer to correlate multiple measurements per cell, and should be used only in those limited situations where quantitation of a single measurement per cell is desired. Because of the ease with which multiple measurements per cell may be acquired and the analytical power that can be obtained by looking at the correlations between parameters on a cell-by-cell basis, histogram data files are being used less and less. This is true even in clinical settings, where the measurement process may be well defined for a particular diagnostic test. In research laboratories most data are now acquired as listmode data.
Bivariate Data Files One way to overcome the limitations of histogram data files is to store two measurements per cell in a two-dimensional array such that the correlations between the two parameters are preserved. In the example above, for instance, the correlation between green fluorescence and light scatter could be preserved.
However, this approach produces over 65,000 times the amount of data as would the single measurement; to store the data from such an experiment, even at 8-bit resolution, requires 256 × 256 (65,536) array elements. This means that a relatively large number of data points must be taken if there are to be more than a few cells per array element. Usually the bit resolution is reduced to 6-bit, so that the array has a size of 64 × 64 (4096) array elements. Even at this resolution the number of cells measured per array element is typically very small, unless data from at least 10,000 cells are stored per distribution and the data are not spread widely over many array elements. Bivariate data files work well when only two parameters are being measured for each cell. However, if three or more parameters are measured, storage as bivariate data files does not permit correlation of any two parameters with any other parameter. If bivariate data are gated on other parameters, it is wise to store the ungated listmode data of the complete data set to preserve data that may be of interest in the future. Data from three or more measurements per cell can be visualized by looking at all possible bivariate combinations of the listmode data, as discussed in UNIT 10.4 on multidimensional data analysis. For example, three parameters may be viewed as a combination of three bivariate histograms. However, to view four parameters per cell requires six bivariate histograms; six parameters requires fifteen bivariate histograms; and so forth. Thus, it takes increasing amounts of computer space to store these large arrays, and the entire data set still cannot be seen by merely looking at all possible bivariate combinations of the data. With more than two parameters, storing the experimental data in listmode data file structures may be the best approach.
Listmode Data Files The most important feature of listmode data is that each cell retains its own identity in the data structure. The parameter measurement values are written serially parameter by parameter (on a cell-by-cell basis) to a data file. The entire experiment can be replayed while applying different logical windows or gates, including gating based on complex Boolean combinations of multiple regions that may be difficult to discern during an actual experiment. Clearly it takes more computer space to store the acquired listmode data at a relatively high (e.g., 10- or 12-bit) resolution, especially when the data are usually structured in lower-resolution (e.g., 6-bit, 64 × 64–channel, or 8-bit, 256 ×
Data Processing and Analysis
10.3.3 Current Protocols in Cytometry
256–channel) bivariate displays. Storing these data at the higher resolution may, however, prove useful at a later time if it becomes possible to apply a new analysis method not originally considered. The time-ordered sequence of the stored data also allows quality control, as “bad data” in the file can be easily excluded from an analysis (Watson, 1987). Listmode data file structure has become the format of choice. Standardized formats for listmode data Few serious attempts were made to standardize data structures between different research laboratories or commercial vendors until the International Society for Analytical Cytology Meeting at Asilomar, California in 1984. At this meeting a listmode standard (FCS 1.0) data file structure was agreed upon, which was subsequently published (Murphy and Chused, 1984). Research laboratories and commercial vendors had previously written data files in proprietary forms; specialized programs were required to translate data from one format to another, making it difficult for scientists to exchange data. Moreover, researchers without programming expertise were left at the mercy of the commercial vendors from whom they bought their flow cytometers. Although it has taken some time to get research labs and commercial vendors to agree to adhere to this standard, it is now generally followed. To more firmly establish the standard, an updated version called FCS 2.0 was published (Dean et al., 1990). The positive results of using standardized data formats are already being felt. An increasing number of data analysis software programs are being offered by researchers and small commercial vendors, and competition has led to a considerable improvement in software quality and cost. This has also allowed listmode data to be transferred from the proprietary computers of instrument manufacturers to personal computers for more convenient analysis. A further development of the standard, version FCS 3.0, has now just been released; for more detailed discussion, see UNIT 10.2.
Listmode Data Processing
Storage of listmode data Listmode data can be stored in a variety of ways: as (1) temporary storage, (2) mediumterm storage, and (3) long-term “archival” storage. The preferred method of storage depends upon the needs and requirements of the user. Flow cytometric data can be temporarily stored directly in the random access memory (RAM) of the computer acquiring the data. An
advantage of this method is the speed with which data can be written into memory, which makes the data available for immediate processing, including potential real-time data classification. Input and output (I/O) to peripheral devices such as disks and tapes can be very time-consuming, particularly for listmode data. Indeed, RAM memory can be used as a “RAM disk” at speeds much greater than those of such peripheral devices. However, because RAM memory on most computers is limited, it may be difficult to store data from large listmode data sets in active memory. Also, the volatile RAM memory is lost as soon as this memory is reallocated or the computer is turned off. Computer “hard” (nonremovable) disks are the form of initial storage most commonly used for listmode data. The listmode data can then be either immediately archived to removable disks (e.g., floppy disks, optical disks, zip disks, or CD-ROMs) or magnetic tape. Removable disks equivalent to 100 or more floppy disks in storage space are a reasonably efficient way to store large listmode data sets. Removable hard and optical disks typically provide anywhere from 20 Mb to several gigabytes (Gb) of storage space and are usually less expensive than floppy disks on a cost-per-megabyte basis. Another advantage of removable hard or optical disks is that their access times are much quicker than those of floppy disks or tapes. Nonremovable hard disks, although increasingly less expensive, pose a problem: because they are nonremovable, they must be thought of as temporary storage only. Such disks must be either periodically purged or backed up onto another archival storage medium so that more data can be acquired. Removable hard disks or optical disks tend to be a more expensive way to store data, particularly if such storage is primarily archival in nature. Typically hard disks have been backed up on magnetic tapes, which are relatively inexpensive, for long-term storage. However, access to listmode data from tapes is slow, and this can lead to significant time problems with large tapes holding gigabytes or more of archived data. Removable optical disks and CD-ROMs provide possible solutions to this problem. Removable optical disks can store several gigabytes of data. Initially these disks had very slow disk access times compared to nonremovable hard disks, and were “write only” (WORM: write only read many times). They were, however, much faster than tapes in terms of data access times, and relatively inexpensive. Unfortunately, the initial cost of the disk drive they
10.3.4 Current Protocols in Cytometry
required was high. Recent advances in the optical disk industry have lowered the price of drives and even resulted in the development of erasable optical disks, but the disks remain quite expensive and are not a particularly economical archive medium if very large amounts of data are involved. Another medium for archiving listmode data is recordable CD-ROMs (CDR). This inexpensive technology has become one of the archival tools of choice. One of its major advantages is that CDR disks can be read by any CD-ROM drive with any computer operating system. Although erasable optical disks are otherwise quite convenient, they cannot be read anywhere but on another erasable optical disk drive from the same manufacturer. One danger with storing just a single backup copy of data is that the backup may become damaged and unreadable, lost, or destroyed. Backup redundancy and off-site data storage are therefore important if data integrity is to be assured. These problems are addressed in more detail in the discussion of data management in UNIT 10.1. Importance of archiving listmode data As a general rule, listmode data should always be archived. This is true for both clinical and basic research data, although the precise requirements in these two cases may be different. There are at least three reasons to archive basic research data. First, research data tends to consume the largest amounts of data space due to extensive use of listmode data file structures. It is in the nature of research that researchers frequently need to “play” with their data extensively to discern the underlying structures and mechanisms. A second reason to save data is the expectation that new data analysis techniques not now available may be able to uncover new aspects of archived data. Lastly, although it is not yet compulsory, federal funding agencies could someday require that all raw
data be stored for a period of several years in order to monitor potential scientific fraud. As listmode data are increasingly used in clinical settings, they will fall under the same laws that require storage of patient data for several years. As regulatory agencies come to realize that the data and data processing are as important as the reagents and instruments, rules will be developed specifying how data need to be acquired, processed, and stored. In the meantime, professional societies have begun voluntary efforts in this direction that have already helped improve the quality of both research and clinical data.
LITERATURE CITED Bagwell, C.B. 1993. Theoretical aspects of flow cytometry data analysis. In Clinical Flow Cytometry: Principles and Application (K.D. Bauer, R.E. Duque, and T.V. Shankey, eds) pp. 41-61. Williams & Wilkins, Baltimore. Dean, P.N. 1990. Data processing. In Flow Cytometry and Sorting, 2nd ed. (M.R. Melamed, T. Lindmo, and M.L. Mendelsohn, eds.) pp. 415444. Wiley-Liss, New York. Dean, P.N., Bagwell, C.B., Lindmo, T., Murphy, R.F., and Salzman, G.C. 1990. Introduction to flow cytometry data file standard. Cytometry 11:321-322. Jolliffe, I.T. 1986. Principal Component Analysis. Springer-Verlag, New York. Murphy, R.F. and Chused, T.M. 1984. A proposal for a flow cytometric data file standard. Cytometry 5:553-555. Watson, J.V. 1987. Time, a quality-control parameter in flow cytometry. Cytometry 8:646-649. Watson, J.V. 1992. Flow Cytometry Data Analysis: Basic Concepts and Statistics. Cambridge University Press, Cambridge, U.K.
Contributed by James F. Leary University of Texas Medical Branch Galveston, Texas
Data Processing and Analysis
10.3.5 Current Protocols in Cytometry
Multidimensional Data Analysis in Immunophenotyping Immunofluorescence techniques can be applied to the study of very complex biological systems (Loken et al., 1992). Cell populations in blood, bone marrow, lymphoid tissue, and body fluids contain multiple cell types that differ not only in lineage but in maturational stage and activation. The complexity of these cell populations requires careful selection of reagents to detect cells of interest and to distinguish them from other cell types that may confound results. Antibodies are seldom specific for individual cell groups, because cellsurface antigens often appear on multiple cell types (Knapp et al., 1989). In addition, autofluorescence from different types of cells is variable; therefore, individual “negative” thresholds must be established for each cell type. The problem of cell population complexity has been addressed by using additional reagents, usually monoclonal antibodies, to provide independent criteria to identify cells of interest. For example, CD4+ T cells can be distinguished from CD4+ monocytes by the presence of CD3, which is characteristic of T lymphocytes but is not found on monocytes (CDC, 1992). The use of two or three monoclonal antibodies in combination with forward and right angle light scattering generates a data set that is difficult to visualize, because the data must be represented in four- or five-dimensional space. The separation between cell populations provided by the multiple characteristics is best visualized by multidimensional analysis using all parameters simultaneously to identify populations within the resulting hyperspace. Groups of cells are distinguished based on a combination of characteristics not apparent in any usual two-dimensional projection of the data. The combination of several antibodies may give complete separation of populations even when none of the antibodies by itself is specific. Thus, to maximize the information content of the data obtained, data analysis techniques and antibody selection for a particular question must be linked.
DEFINING THE PROBLEM OF MULTIDIMENSIONAL DATA ANALYSIS Single-parameter flow cytometry data can be represented as a histogram plotting frequency of events versus signal intensity. TwoContributed by Michael R. Loken Current Protocols in Cytometry (1997) 10.4.1-10.4.7 Copyright © 1997 by John Wiley & Sons, Inc.
parameter data can be displayed as a dot plot or as a two-dimensional histogram. With three or more parameters, the data cannot be directly visualized without mathematical reduction or manipulation. A significant difference exists between multiparameter and multidimensional data analysis (Terstappen and Loken, 1992). In multiparameter analysis, various cellular characteristics are analyzed independently. A cell population is assessed using multiple parameters; however, the parameters are not correlated with each other. In a multidimensional analysis, several characteristics are measured simultaneously for each cell. The correlation between characteristics is maintained and becomes a powerful tool to understand the cellular composition of a complex population. For example, a cell population may be stained using three different antibodies in three different tubes. These tubes, when analyzed in sequence, provide a multiparameter analysis of that cell population: X% of the cells are reactive with antibody 1, Y% of the cells are reactive with antibody 2, and Z% of the cells are reactive with antibody 3. In contrast, by labeling each antibody with a different color, it is possible to stain a cell population with all three antibodies in a single tube, so that the relationships between the staining of a cell with each different reagent is preserved. This permits multidimensional analysis, as each event (cell) generates multiple signals simultaneously. For multidimensional data analysis, data must be collected in listmode format, thereby maintaining the correlation of the signals for each event. In a standard three-color immunofluorescence protocol, two light-scatter parameters (forward and right-angle light scatter) are collected along with green (fluorescein isothiocyanate, or FITC), orange (phycoerythrin, or PE), and red (Caltag’s Tandem Conjugates, Pharmingen’s Cychrome Conjugates, or Becton Dickinson’s peridinim chlorophyll complex, PerCP) signals to generate listmode data with five characteristics per event (see UNIT 6.2). These five characteristics represent the coordinates for that event in a five-dimensional space. In multidimensional analysis, these coordinates are used to visualize cells with similar characteristics and to distinguish cell populations from one another.
UNIT 10.4
Data Processing and Analysis
10.4.1
Visualization of a multidimensional space is difficult because most means of human communication are limited to two-dimensional space. The most frequently used approach is to display the combinations of five parameters, taken two at a time as either dot plots or contour plots. With five separate characteristics, there are ten possible combinations of two-parameter displays: 1. Forward light scatter/right angle light scatter 2. Forward light scatter/green fluorescence 3. Forward light scatter/orange fluorescence 4. Forward light scatter/red fluorescence 5. Right-angle light scatter/green fluorescence 6. Right-angle light scatter/orange fluorescence 7. Right-angle light scatter/red fluorescence 8. Green fluorescence/orange fluorescence 9. Green fluorescence/red fluorescence 10. Orange fluorescence/red fluorescence These two-parameter dot plots can be visualized as the faces of a hypercube. Instead of a cube with three unique faces, five-dimensional space has ten faces, six of which are displayed in Figure 10.4.1. Populations or clusters of cells are identified by different colors. It is important to understand that these two-dimensional displays are simply projections of five-dimensional data. Populations that appear close or overlapping in one projection may be clearly separate in other projections. The multiple projections of the data permitted by multidimensional analysis facilitate maximum separation of event clusters. The basis of multidimensional analysis is the use of color eventing to follow groups of cells in a display and to correlate those groups when viewed from a different perspective (Terstappen et al., 1989).
COMBINING REGIONS USING BOOLEAN LOGIC
Multidimensional Data Analysis in Immunophenotyping
Groups of events are identified in multidimensional space by constructing multiple regions and viewing their overlapping relationships. Each region is assigned a color; the overlap between two or more regions appears as a new color, thereby distinguishing these events as unique. The combination of regions
follows the rules of Boolean logic, as illustrated in Figure 10.4.2A. In this example, two circles represent regions R1 and R2. The space defined by parameters 1 and 2 is divided into four parts based on the relationships between the two regions. Table 10.4.1 shows various logical combinations of R1 and R2 that can be used to identify specific regions in Figure 10.4.2A. This list is not exhaustive—other combinations are possible. These relationships can also be applied to multidimensional space (Fig. 10.4.2B). The same relationships hold in three dimensions as in the two-dimensional plot: the events colored red or blue are confined to cylinders that pass through the cube. The intersection of the two cylinders is a volume, shown yellow in this example. Additional dimensions can be added to more precisely identify the characteristics of the cells of interest. It should be kept in mind that a region drawn in two-dimensional space represents a tube that extends through multidimensional space. Populations are defined based on the intersection of such tubes in the multidimensional space, although these may not be directly viewed. The Boolean combination most often used in immunofluorescence is the intersection of two regions R1 and R2. For example, lymphocytes identified by forward and right-angle light scatter [R1] are further subdivided by an antibody reactive with lymphocyte subsets [R2]. Because some subsets are not completely identified using a single reagent, two or more antibodies must be used to describe the population of interest.
MULTIDIMENSIONAL GATING TO SEPARATE OVERLAPPING POPULATIONS Multidimensional data analysis can be used to separate populations seemingly inseparable in any two-dimensional plot. Figure 10.4.3 shows four-parameter data that were computergenerated using WinList 2.0 (Verity Software) to mimic immunofluorescence data. Two populations were created with centers at channels 16 and 28 (128 channels full scale) in each of the four parameters. The two populations had the same dispersion around the central points with respect to each of the four parameters. Six combinations of two parameters were constructed from the four-parameter data (Fig. 10.4.3A-F). Identical triangular regions (R1 to R6) were drawn in each of the six plots to include the dimmer, channel 16–centered population. A Boolean combination of these
10.4.2 Current Protocols in Cytometry
103 40
CD4PE
Forward scatter
60
20
102 101
Right-angle scatter
CD8 FITC
CD 45 C Per P
101
102 103 103
CD4 PE
101
102 103
CD8 FITC
Right-angle 102 light scatter 101
101 102 103 CD4 PE
Figure 10.4.1 Six of the ten two-parameter faces of a five-dimensional hypercube. Color identifies populations recognized in other projections. Normal peripheral blood cells were stained with CD8-FITC, CD4-PE, and CD45-PerCP. Forward and right-angle scatter along with the three antigens define the five-dimensional space. CD4+ lymphocytes are red, CD8+ lymphocytes are blue, monocytes are dark green, and neutrophils and eosinophils are green. See color figure.
A
B
80
R1
R2
R1
R2
80
40 40 80 Parameter 1
120
40 80 Parameter 1
3
120 40
Pa ra m et er
Parameter 2
120
120
Figure 10.4.2 (A) Schematic diagram depicting the relationship between two regions R1 and R2 in a single plane defined by parameters 1 and 2. Table 10.4.1 shows various logical combinations of R1 and R2 that can be used to identify specific regions herein. (B) Schematic diagram depicting similar relationships between R1 and R2 in a three-dimensional display. See color figure.
Data Processing and Analysis
10.4.3 Current Protocols in Cytometry
regions was made: Gate 1 = [R1 & R2 & R3 & R4 & R5 & R6]. Using this logic, an event must lie within all of the triangular regions to be included in the gate. If an event is outside of any of the regions, it is not included within this gate. Events that satisfied Gate 1 are colored red (Fig. 10.4.3B and C). A second gate was constructed to identify all events that lie outside any of the regions: Gate 2 = [not R1 or not R2 or not R3 or not R4 or not R5 or not R6]. These events are colored green (Fig. 10.4.3B and D). Figure 10.4.3B displays all the events, but colored either red or green depending on whether they are included in Gate 1 or Gate 2. For compari-
son, only the red population is depicted in Figure 10.4.3C, whereas the green population is shown in Figure 10.4.3D. Note that many of the green dots lie within the triangle of R4 (Fig. 10.4.3D), indicating that they must lie outside the triangle in one of the other projections. The effectiveness of multiparameter separation of these two populations using combinations of the six triangular regions is demonstrated by the analysis of each population separately, shown in Figure 10.4.4. A histogram of the combined populations (Fig. 10.4.4A) can be compared with the overlapping histograms of the red and green populations after multi-
Table 10.4.1 Identification of Regions in Figure 10.4.2 by Boolean Logic Combinations of R1 and R2
Regions identified
[R1] [R1 and not R2] [R1 and R2] [R1 or R2] [Not R1 and not R2] [Not R1 or not R2] [R1 or not R2]
Red, yellow Red Yellow Red, yellow, blue Gray Red, blue, gray Red, yellow, gray
A
80 R1 40
R2
40 80 120 Parameter 1
E
R4 40
40 80 120 Parameter 3
Multidimensional Data Analysis in Immunophenotyping
40 80 120 Parameter 1
F Parameter 3
D
80
R3
40 80 120 Parameter 1
Parameter 2
Parameter 2
120
C Parameter 4
B Parameter 3
Parameter 2
120
Combination
R5
80 120 40 Parameter 4
R6
40 80 120 Parameter 4
Figure 10.4.3 Six two-dimensional projections of four-parameter data. WinList 2.0 was used to generate two populations with identical dispersions in all four parameters, one centered at 16, the other at 28. Triangular regions (R1 to R6) were defined in each projection and gates were defined as described in the text. Gate 1 events are colored red, while Gate 2 events are colored green. Both populations are displayed in B. Only the red population is displayed in C and only the green in D. See color figure.
10.4.4 Current Protocols in Cytometry
parameter separation using the six identical regions (Fig. 10.4.4B). These can be compared with overlapping histograms generated from two separate listmode files in which the two populations were never mixed (Fig. 10.4.4C). The positions and shapes of the two histograms are identical whether the populations were obtained from two separate listmode files (Fig. 10.4.4C) or mixed and then separated using the multidimensional approach (Fig. 10.4.4B). To quantify the separation, 4% of the dimmer population was excluded from the red population (recovery = 96%) when this population was analyzed alone. Less than 2% of the brighter population was included in the gate (purity = 98%) when assessed separately. This example illustrates that even when populations overlap, it is possible to achieve almost complete separation using multidimensional analysis. The ability to separate populations depends on several factors: the variability within a population, separation between populations, and availability of multiple independent parameters by which to separate the populations. It is crucial to use independent parameters that have no correlation with each other. In this example, complete separation was obtained using four separate parameters.
USING NONRECTANGULAR REGIONS TO IDENTIFY DISCRETE POPULATIONS One of the goals of multidimensional gating is to identify populations that are discrete, thereby minimizing the variations related to
A
B
slight changes in boundary settings when drawing any specific region. A population of interest is identified by combining several different plots and colors to facilitate its assignment to a specific group. This strategy can be illustrated using the same data set used in Figure 10.4.1, in which human peripheral blood was stained with CD8-FITC, CD4-PE, and CD45-PerCP. Cells of different lineages can be identified by a combination of CD45 intensity and rightangle light scatter, as shown in Figure 10.4.5B (also see Stelzer et al., 1993). In this example monocytes and neutrophils were identified first and colored green and yellow, respectively, based upon CD45 staining intensity and rightangle scatter. The remaining population (lymphocytes) was then divided based on CD4 or CD8 expression plotted versus right-angle light scatter (Fig. 10.4.5C and D). CD8+ cells were colored red (Fig. 10.4.5C) and CD4+ cells were colored blue (Fig. 10.4.5D). It should be noted that a single reagent in combination with rightangle light scatter is often more informative in discriminating lymphoid staining than in combination with forward light scattering. A nonrectangular region for CD4+ cells was drawn to exclude monocytes, which stain dimly with CD4 (Fig. 10.4.5D). The red and blue populations then aided in defining the lymphocytes in the CD45 versus right-angle light-scatter plot (Fig. 10.4.5B). The correlation between CD4 and CD8 staining for the entire data set is shown in Figure 10.4.5E and compared with the staining for just lymphocytes in Figure 10.4.5F. This allows the
C
Number of events
500
300
100
40 80 120 Parameter 1
40 80 120 Parameter 1
40 80 120 Parameter 1
Figure 10.4.4 Computer-generated histograms of the data from Figure 10.4.3. (A) Histogram of the combined populations, which is unimodal even though two populations have been added together. (B) Overlapping histograms of the red population (in which events reside within all six triangular regions) and the green population (in which events lie outside at least one triangular region) after multiparameter separation. (C) Overlapping histograms from two separate listmode files where unmixed red and green populations had characteristics identical to those mixed together in A and B. See color figure.
Data Processing and Analysis
10.4.5 Current Protocols in Cytometry
validity of boundaries established in Figures 10.4.5C and 10.4.5D to be checked in a different projection of the data. Thus, the process of defining populations is iterative with respect to refinement of the gate settings as additional projections of the data are assessed. This technique of multidimensional gating allows individual cells (dots) to be assigned to appropriate groups with similar characteristics based on multiple parameters rather than on a single parameter. This method for assigning events to
Multidimensional Data Analysis in Immunophenotyping
groups minimizes the problems caused by changes in autofluorescence and/or improper compensation.
LITERATURE CITED CDC (Centers for Disease Control). 1992. Guidelines for the performance of CD4+ T-cell determinations in persons with human immunodeficiency virus infections. Morb. Mortal. Wkly. Rep. 41(RR-8):1-17. Knapp, W., Dorken, B., Gilks, W.R., Reiber, E.P., Schmidt, R. E., Stein, H., and von dem Borne,
Figure 10.4.5 Multidimensional analysis of peripheral blood stained with CD8, CD4, and CD45. The data file is the same as in Figure 10.4.1. In B, monocytes (mono) and myeloid cells are colored green and yellow. In C, CD8+ lymphocytes have been identified by excluding cells colored green or yellow. In D, CD4+ lymphocytes have been identified by excluding the green and yellow events. Blue and red events define the lymph region in B. All events are depicted in E, whereas only those that were in lymph (B) are displayed in F. See color figure.
10.4.6 Current Protocols in Cytometry
A.E.G.Kr. 1989. Leukocyte Typing IV, White Cell Differentiation Antigens. Oxford University Press, Oxford, U.K. Loken, M.R., Civin, C.I., Shah, V.O., Fackler, M.O., Segers-Nolten, I., and Terstappen, L.W.M.M. 1992. Characterization of erythroid, lymphoid, and monomyeloid lineages in normal human bone marrow. In Flow Cytometry in Hematology (O.D. Laerum and R. Laerum, eds.) pp. 31-42. Academic Press, London. Stelzer, G.T., Shults, K.E., and Loken, M.R. 1993. CD45 gating for routine flow cyometric analysis of human bone marrow specimens. Ann. N.Y. Acad. Sci. 677: 265-280. Terstappen, L.W.M.M., Civin, C.I., Shah, V.O., Hurwitz, C.A., and Laerum, M.R. 1989. Multidi-
mensional flow cytometry as a new approach for discrimination between normal and leukemic cells in peripheral blood and bone marrow. In Progress in Cytometry (A. Jansen, ed.) pp. 4-29. Becton Dickinson, Erembodegem, Belgium, and San Jose, Calif. Terstappen, L.W.M.M. and Loken, M.R. 1992. Leukocyte differential counting. In Flow Cytometry in Hematology (O.D. Laerum and R. Laerum, eds.) pp. 95-110. Academic Press, London.
Contributed by Michael R. Loken HematoLogics, Inc. Fred Hutchinson Cancer Research Center Seattle, Washington
Data Processing and Analysis
10.4.7 Current Protocols in Cytometry
Two-Dimensional Image Processing and Analysis INTRODUCTION The success of an image cytometry project can depend upon the processing and analysis of the specimen image. Image processing is used to correct inaccuracies in an image and make it easier to interpret, whereas image analysis is used to extract quantitative data about a specimen (Jain, 1989; Pratt, 1991; Gonzalez and Woods, 1992; Baxes, 1994; Russ, 1995; Castleman, 1996). This unit provides an overview of image processing methods; for further detail, consult Castleman (1996).
Definitions Images must be converted to numerical form before they can be processed by computer. Digitization is the process of generating a rectangular array of numbers to represent an optical image. The image is divided into small regions called picture elements, or pixels. The most commonly used subdivision scheme is a rectangular sampling grid. The image is divided into rows and columns of adjacent pixels. At each pixel location, the image brightness is sampled and quantized. The number inserted into the digital image at each pixel location is called the gray level; it reflects the brightness of the image at the corresponding point. Each pixel then has an integer location, or address (row number and column number), and an integer gray-level value. In this context, then, a digital image is a rectangular array of numbers generated by sampling an optical image in a rectangular grid pattern, and quantizing it in equal intervals of intensity. See UNIT 2.3 for further discussion of image digitization. Digital image processing generates a modified version of an image; thus, it converts one image into another. Digital image analysis, on the other hand, converts a digital image into something other than a digital image, such as a set of measurement data or a decision. For example, an image analysis program might count the cells in an image and measure their size. The term contrast refers to the amplitude of gray-level differences in an image. Noise is an additive contamination of the image signal. Gray-scale resolution is the number of gray levels per unit of light intensity. Storing a digital image in 8-bit bytes, for example, yields a 256-level gray scale. Sampling density is the Contributed by Kenneth R. Castleman Current Protocols in Cytometry (1997) 10.5.1-10.5.14 Copyright © 1997 by John Wiley & Sons, Inc.
UNIT 10.5
number of pixels per micrometer of the specimen. Its reciprocal is the pixel spacing. Magnification refers to the size relationship between the objects in the specimen and the same objects in a displayed or printed image
Uses of Digital Imaging Image correction. Historically, one of the most productive uses of image processing has been image correction (or “calibration”)—that is, the removal or reduction of the noise, distortion, and resolution loss introduced by image digitizing systems (optics, cameras, etc.). Noise can be random, periodic, or fixed-pattern. Distortion comes in two types: geometric warping and photometric nonlinearity. Resolution loss reduces the visibility of small detail in an image. Enhancement. The goal of image enhancement is to improve the quality of an image, which can increase contrast and resolution and make the image easier to interpret. Image analysis. Digital image analysis techniques can locate the objects in an image, measure them, and identify each one. They are used to quantify the size, shape, density (or brightness), texture, and color of the objects of interest and determine the number and proportion of each type.
Types of Digital Image Processing Gray-scale transformations. The response of charge-coupled device (CCD) cameras is normally quite linear with respect to light intensity. Should it occur, however, photometric nonlinearity is easily corrected by a suitable gray-scale transformation operation, whereby the gray scale is made linear with respect to some other photometric property, such as optical density. Algebraic operations. Random noise can be reduced by averaging multiple images of a stationary specimen. Because the noise pattern differs from image to image, it tends to average out, whereas the image of a stationary specimen does not. Fixed-pattern noise from shading— caused, for example, by nonuniform illumination or imperfections in CCD chips—is best handled by background subtraction, a format of algebraic operation. Geometric operations. Geometric distortion is seldom a problem with good-quality
Data Processing and Analysis
10.5.1
microscope optics and CCD cameras. With serial sections, however, each section is subject to different mechanical forces as it is transferred to the microscope slide. Resulting geometric distortion can be corrected by a geometric operation in which an image is copied to another location using a suitable position-altering rule. Normally one section image is used as a reference and the others are warped to conform to it. Filtering. Many important image-processing operations, including enhancement and noise reduction, come under the general heading of filtering. Image filtering is implemented by the mathematical process of convolution, in which each pixel of an image is replaced with a weighted average of itself and its neighbors.
IMAGE PROCESSING Gray-Scale Transformations Gray-scale transformations modify an image one point at a time, considering only the gray level at that point.
Gray-level histograms Gray-level histograms are useful digital image processing tools that are simple to compute and provide a handy summary of the gray-level content of an image. A gray-level histogram shows, for each gray level, the number of pixels in an image that have that level. Figure 10.5.1B shows a gray-level histogram calculated from the image in Figure 10.5.1A. A histogram indicates whether or not an image is properly digitized—that is, properly scaled within the available range of gray levels. A digital image should make use of almost all the available gray levels, as in Figure 10.5.1. If the brightness range of the image is greater than the digitizer can handle, gray levels will be “clipped” at 0 or 255, producing spikes at one or both ends of the histogram. A quick check of the histogram during digitizing can identify problems early on. In an image of light objects on a dark background, such as the one in Figure 10.5.1, the dark pixels in the background produce the leftmost peak in the histogram, whereas the lighter
A
pixels (x 103)
B 4 3 2 1 0 0
Two-Dimensional Image Processing and Analysis
40
80
120 160 Gray level
200
240
Figure 10.5.1 Gray-level histogram. (A) Cell image obtained by fluorescence in situ hybridization (FISH). (B) Corresponding gray-level histogram, in which the number of pixels in the image having each specific gray level (vertical) is plotted versus gray level (horizontal).
10.5.2 Current Protocols in Cytometry
pixels inside objects give rise to the broader right-hand peak. The relatively few mid-gray pixels around the edges of the objects contribute to the dip between the two peaks. Gray-level thresholds can be used to separate objects from background. For instance, all pixels having a gray level greater than some threshold value can be labeled as interior points. Choosing the gray level at the dip in the histogram as the threshold often yields reasonable boundaries for objects. Another useful fact is that the total area of the objects is equal to the area of the right-hand peak in the histogram. Implementation Gray-scale transformations allow the user to modify the way in which image data fill the available range of gray levels, which affects how the image will appear when displayed. In a gray-scale transformation, each output pixel’s gray level depends only upon the gray level of the input pixel with the same coordinates. Thus, a gray-scale transformation does not modify the spatial relationships within an image. Grayscale transformations may be viewed as pixelby-pixel copying operations, except that the gray levels are modified according to a specified function. A gray-scale transformation that converts an input image A(x,y) into an output image B(x,y) may be expressed as B(x,y) = f(A(x,y)). A gray-scale transformation is completely specified by the transformation function, f(D), which specifies the mapping of input gray level to output gray level. Gray-scale transformations are used to overcome limitations of image digitizers and improve image display quality. They may also be used to convert the units of a gray scale. If a microscope image has been digitized with graylevel values proportional to the specimen’s transmittance, a gray-scale transformation can make the gray levels linear with respect to optical density. Gray-scale transformations in which output gray level is a linear function of input gray level are called linear gray-scale transformations, and the gray-scale transformation function takes the form DB = f(DA) = aDA + b, where DB is the output gray level corresponding to an input pixel having gray level DA. If a = 1 and b = 0, this gives an identity operation that merely copies A(x,y) into B(x,y). When a > 1, the contrast is increased, whereas for a < 1, the contrast is reduced. If a = 1, the operation merely scales the gray-level values of all pixels up or down by the amount b, making the entire image darker or lighter. If a is negative, the image is complemented (re-
versed). Gray-scale transformations can be designed to make the histogram flat, make it match that of another image, or make it fit a specified form. They can also compensate for known nonlinearity of a digitizer or display.
Algebraic Operations Algebraic operations produce an output image that is the pixel-by-pixel sum, difference, product, or quotient of two or more input images. One or more of the images may be a constant, but this often reduces to a gray-scale transformation. Image averaging One useful application of image addition is noise reduction: if multiple images of a stationary scene are contaminated by an additive random noise source, averaging them reduces the noise. The stationary component of the image is unchanged by the averaging process, whereas the noise pattern, being different from one image to the next, builds up more slowly in the summation. The signal-to-noise ratio improves by the square root of the number of images averaged. Image addition may also be used to superimpose the contents of one image upon another, producing a “double exposure.” Background subtraction Image subtraction can remove a superimposed background from an image. Such a background may take the form of a slowly varying shading pattern, a periodic noise pattern, or any other additive contamination that is known at every point in the image. It is usually possible to capture an image of the background alone and then subtract that from each newly digitized image. Another way to remove shading is to fit a two-dimensional surface to the background and subtract the resulting function from the image. Subtraction can also accentuate any differences between two images of the same scene. For example, one can detect motion by subtracting sequential images of a scene with a moving component. Image subtraction is also useful for locating the edges of an image: imperfect registration between the two images causes the edges of the stationary structures to appear in the difference image.
Geometric Operations Definitions Geometric operations change the spatial relationships between objects in an image. The effect is similar to printing the image on a
Data Processing and Analysis
10.5.3 Current Protocols in Cytometry
rubber sheet, stretching the rubber sheet, and tacking it down at various points. In general, any point in the input image may move to any position in the output image; however, geometric operations are usually more constrained to preserve order. A geometric operation requires two separate algorithms. One defines the spatial transformation itself: this specifies the path of each pixel as it moves from its initial to its final position. A second algorithm is required for gray-level interpolation, because the integer x,y positions in the input image will generally move to fractional (noninteger) positions in the output image. The general definition for a geometric operation is g(x,y) = f(x′,y′) = f[a(x,y),b(x,y)], where f(x′,y′) is the input image and g(x,y) is the output image. The functions a(x,y) and b(x,y) uniquely specify the spatial transformation. g(x,y) is generated, pixel by pixel, line by line, by looking back at f(x′,y′) for the appropriate gray level. Because x′ and y′ generally are noninteger coordinates, interpolation among four neighboring pixels in the input image is required. Nearest-neighbor interpolation is the simplest to implement, but it often creates so-called “stairstep edges.” Bilinear interpolation, which is only slightly more complex, creates smoother output images. For simple operations such as rotation, translation, and scaling, the two functions a(x,y) and b(x,y) that specify warping can be written in algebraic terms. For more complex warps, it is possible to specify a set of pixels in the input image to be used as “control points,” along with the position where each of these
should end up in the output image. The computer can then use interpolation to determine where the remaining points should go. Uses of geometric operations Geometric operations can remove distortions—most often caused by optics or, with serial sectioning, physical distortion of the specimen—from a specimen image. It can bring two images of the same specimen into registration by warping one to match the other, and it can rearrange the content of an image to aid interpretation (e.g., automatic karyotyping).
Filtering Lowpass filtering (smoothing) is used to correct random noise. In a properly digitized image, the specimen dominates the low and middle ranges of the frequency spectrum, but random noise often dominates the upper end. Thus, the noise reduction filter must be designed to discriminate against only the smallest (roughly pixel-sized) structures in the image. Strong periodic noise often indicates a hardware problem that is best solved as such. However, periodic noise manifests itself very compactly as “spikes” in the Fourier frequency domain, and can therefore be isolated and eliminated by frequency-domain filtering. Some loss of resolution is inherent in optical imaging because of diffraction phenomena that result from the wave nature of light. Figure 10.5.2 shows the modulation transfer function (MTF) of a perfect (“diffraction-limited”) lens. The MTF is a function that shows by what
MTF
1.0
0 0 Frequency (f )
Two-Dimensional Image Processing and Analysis
fc
Figure 10.5.2 Modulation transfer function (MTF) of a perfect lens. The MTF indicates the extent to which a lens passes image detail of different spatial frequencies. The cutoff frequency, fc = 2NA/λ, improves with higher numerical aperture (NA) of the objective lens and with shorter wavelength illumination (λ). Here spatial frequency is referred to the specimen plane.
10.5.4 Current Protocols in Cytometry
factor image components at different frequencies are multiplied as they pass through the lens. An MTF equal to 1.0 throughout the frequency band would have no effect on the image. For a perfect lens, however, this is not the case. Midsized structures lose contrast, and detail small enough to correspond to frequencies greater than the cutoff frequency fc is not passed at all. This degradation is unavoidable, because it is inherent to the wave nature of light. Furthermore, actual microscope optics can only approach this theoretic optimum behavior. Inexpensive objectives frequently do considerably worse. A highpass filter can partially restore the loss of contrast of midsize objects. However, this process (“deconvolution”) competes with lowpass filtering for noise reduction. Thus, a properly designed filter combination must boost the midrange frequencies to compensate for the optics while attenuating the highest frequencies that are dominated by noise. Highpass filtering can enhance the detail in an image, making it easier to interpret, and can also compensate for known shortcomings in image display or printing hardware.
is rotated 180° and its origin is shifted to the coordinates (i,j) in the input image. The two arrays are then multiplied together, element by element, and the resulting products are summed to give the output value. The output may then require scaling to put it into the proper range. Although the kernel in Figure 10.5.3 is 3 × 3, any size can be used. It is the size and content of a kernel that determine what effect convolution will have on an image. The number of multiply-and-add operations required is equal to the number of pixels in the image times the number of weights in the kernel. Unless the kernel is small, convolution can become computationally expensive. Near the border of an image, pixels lack a full set of neighbors. In such a region convolution can wrap the input image around upon itself (by assuming that the first column comes immediately after the last, etc.), fill in a constant (e.g., zero) for input pixels beyond the border, or simply eliminate output rows and columns that cannot be computed by full convolution. It is important to digitize an image so that no important information falls nearer the border than half the width of the kernel.
Convolution Convolution, illustrated by Figure 10.5.3, is a process of replacing each pixel with a weighted average of itself and its neighbors. The weights are taken from an array (matrix) called the convolution kernel. To create an output pixel for a position (i,j), the kernel array
Smoothing Convolution using a kernel matrix in which all the values are positive will have a smoothing or blurring effect on the image, as in Figure 10.5.4B. Small structures will lose contrast more than larger ones. This can be effective for reducing random noise. The process will also
kernel matrix column j
column j
input image
output image
row i
row i
array of products
output pixel
xC summer
scaling constant
Figure 10.5.3 Digital convolution. The output image is made up of scaled, weighted averages of the pixels in neighborhoods of the corresponding input pixels. The kernel matrix specifies the weights.
Data Processing and Analysis
10.5.5 Current Protocols in Cytometry
blur edges, making them less sharp, so that an image may look out of focus. The larger a kernel, the more pronounced its effect: a 3 × 3 kernel with equal weights produces a relatively mild effect, whereas larger kernels produce more significant blurring. Enhancement Convolution using a kernel containing one or more positive weights at the center, with negative weights around the periphery, increases the contrast of small structures at the expense of larger ones. The effect of applying a kernel of this type, known as an edge-enhancement filter, is shown in Figure 10.5.4C. Small kernels will enhance only the smallest structures in the image; larger kernels are required to enhance mid-sized objects as well. If the weights add up to one, the gray level of large areas is preserved, and the output image looks the same except that the edges are sharper. If the weights add up to zero, the gray level of large areas is lost completely, and only edges remain (Figure 10.5.4D). Optimal filter design In addition to the guidelines mentioned above, there are well-developed mathematical
Two-Dimensional Image Processing and Analysis
procedures for designing convolution kernels that are optimal for specific purposes. For example, if an image has been blurred by poorly focused optics, camera motion, etc., and the amount of blur can be measured and characterized, the image can be refocused using a deconvolution filter. The proper values to use in a kernel matrix can be computed with the aid of the Fourier transform. Good results can be obtained if the blur is carefully characterized, but noise in the image limits the degree to which such damage can be corrected. An image that has been contaminated by additive random noise can be processed using a Wiener filter to reduce the noise. The Wiener filter is a kernel designed to be the best kernel for estimating what the image looked like before the noise was added; it minimizes the mean square difference between its output and the original. Other kernel design techniques, variations on the Wiener filter, yield filters that produce results more pleasing to the human eye. When it is desirable to locate a specific type of object in an image, the image can be processed using a matched filter: this is a kernel that is designed to best enhance those objects at the expense of other components of the scene.
A
B
C
D
Figure 10.5.4 Convolution. (A) Input image; (B) smoothed image (5 × 5); (C) enhanced image; (D) severely enhanced image.
10.5.6 Current Protocols in Cytometry
Ideally, the output image will have distinct peaks where objects of interest are located in the input image. This sort of filter is sometimes called a matched detector, because it seeks to tag the locations of specific objects rather than to enhance their appearance.
The assignment is based solely on the feature vector. A misclassification error occurs if an object is assigned to the wrong class. The probability of this happening is the misclassification error rate.
System Design IMAGE ANALYSIS Image analysis is concerned with image content—that is, finding out what is in the picture. There are two approaches to this process of pattern recognition, in common use, one statistical and the other involving artificial neural networks. Classical statistical pattern recognition involves locating the objects in an image and identifying them using statistical techniques. Artificial neural networks use an approach that is patterned after the action of neurons in the central nervous system.
Pattern Recognition As illustrated in Figure 10.5.5, pattern recognition consists of three major steps (discussed in more detail below). The first phase is image segmentation, in which each object is found and isolated from the rest of the scene. The second step, called feature extraction, is where the objects are measured. A feature is some significant, quantifiable characteristic of an object. The feature extraction process produces the feature vector. This vector, which contains an amount of information that is drastically reduced from that of the original image, represents all the knowledge upon which the subsequent classification decision must be based. It is sometimes useful to think of a multidimensional feature space in which all possible feature vectors reside; any particular object then corresponds to a point in feature space. The third step is classification. Each object is assigned to one of several preestablished groups (classes) that represent all possible types of objects expected to occur in the image.
The design of a pattern recognition system for classifying objects involves five operations. 1. Algorithm selection: choosing an algorithm that will isolate the individual objects in the image segmentation. 2. Feature selection: deciding which characteristics of the objects (size, shape, etc.) best distinguish among the various classes. 3. Classifier design: choosing the mathematical basis for the classification procedure. 4. Classifier training: pinning down the various adjustable parameters of the classifier itself (e.g., decision thresholds). 5. Classifier testing: estimating the rate of misclassification errors that can be expected when the system is put into operation. Pattern recognition is used in many cell counting and proportion estimation applications. The clinical tasks of chromosome karyotyping, white blood cell differential counting, Pap smear analysis, and FISH dot counting have been automated in commercial instruments that use pattern recognition.
Image Segmentation The image segmentation process partitions an image into disjoint (nonoverlapping) regions in which all the pixels are adjacent (touching). Thresholding Thresholding is a simple image segmentation technique that is useful for scenes containing solid objects on a contrasting background. It never fails to divide the image into disjoint regions with closed, continuous boundaries.
x1 x2 xn input image
image segmentation
object image
feature extraction
Figure 10.5.5 The three steps of pattern recognition.
"bar" classification
feature vector
object type Data Processing and Analysis
10.5.7 Current Protocols in Cytometry
These boundaries, however, do not always correspond to the edges seen by the human eye. A threshold rule for image segmentation assigns all pixels with gray levels below some threshold gray level to the background. All pixels at or above the threshold fall inside objects. The object boundaries are made up of the interior points that are next to a background point. Thresholding works well if both the objects and the background have relatively uniform, but different, gray levels. If the background gray level is not uniform or the contrast of the objects differs, a threshold value that works well in one area of the image might work poorly in another area. In this case, a threshold gray level that is a slowly varying function of position in the image can be used to ensure that each object gets an appropriate threshold gray level.
Two-Dimensional Image Processing and Analysis
Edge-based methods Image segmentation can be accomplished by searching for edge points directly using one of two approaches. In boundary tracking, an image is transformed into one that shows edges as high gray level using a gradient magnitude operator that gives each output pixel a value proportional to the derivative (slope) in the neighborhood of the corresponding input pixel. Working in the gradient image, the boundary-tracking process is initiated by identifying a pixel of locally highest gray level as the first boundary point. Next, the 3 × 3 neighborhood of that pixel is searched for the neighbor with the maximum gray level, which is designated the second boundary point. If the same maximum gray level is shared by two or more of the neighbors, the choice is arbitrary. The next boundary point is found based on the current and previous boundary points. Working in the 3 × 3 neighborhood centered on the current boundary point, the neighbor diametrically opposite the last boundary point and the neighbors on each side of it are examined. The next boundary point is the one of those three that has the highest gray level. If two or all three adjacent boundary points share the highest gray level, then the middle one is chosen. If the two nonadjacent candidate points share the highest gray level, the choice is arbitrary. In low-noise images this algorithm usually follows the maximum gradient boundary, but image noise can send the process temporarily or hopelessly off the track. Noise can be reduced by smoothing the gradient image before tracking, but it cannot guarantee that the tracking algorithm will not
get lost and run off to the border of the image. Another approach is edge detection and linking, which involves identifying a large proportion of the edge points and trying to connect them together to form boundaries. An edge-detection filter, such as the gradient operator, can be followed by a thresholding operation to label the edge points. This will produce a boundary that is several pixels thick in some places and has gaps in others. A thinning operation can shrink it to single pixel width. Then an algorithm that connects together nearby edge points can link them up to form continuous boundaries. Curve-fitting techniques can be used to draw smooth boundaries through sparsely located edge points. Region growing Region growing is an image segmentation procedure that begins by dividing an image into many tiny regions. These initial regions may be small neighborhoods or even single pixels. Next, the boundaries between adjacent regions are examined one at a time. If two neighboring regions differ significantly in gray level (and perhaps other relevant properties), the boundary is considered to be “strong” and is allowed to stand. If they do not, the “weak” boundary is dissolved and the adjacent regions merged into one. This process is continued until none of the remaining boundaries is weak enough to be dissolved. At this point the surviving regions will have grown until their boundaries coincide with the edges of the objects. Segmentation by region growing is computationally more expensive than simpler techniques, but it can prove useful for complex images.
Feature Extraction (Object Measurement) After the objects in the image have been isolated they need to be measured. The feature measurements can be used both for quantitation and for identification of the objects. It is usually most convenient to compute spatial (size) measurements first in terms of pixels and photometric (brightness) measures in terms of gray levels. In a later step, distance and area measurements can be calibrated by multiplying them by the pixel spacing or the area of a pixel, respectively. Gray level values can be multiplied by a suitable photometric calibration constant. A gray scale transformation based on the photometric calibration curve of the digitizer can be used to linearize the gray scale if necessary. For simplicity, the following discussion deals with uncalibrated measurements.
10.5.8 Current Protocols in Cytometry
Size measurement Area is a convenient measure of overall object size. For accurate size measurement, however, it is necessary to determine whether boundary pixels are completely or only partially contained inside the object. If the object is defined by a polygon with its vertices at the centers of the boundary pixels, then each boundary point is, on the average, only half inside the object. Subtracting half the number of boundary points from the number of pixels in the object will give the area. Perimeter is particularly useful for discriminating between objects with simple and complex shapes. A simply shaped object uses less perimeter to enclose its area. The perimeter measurement can be easily computed during the extraction of an object from a segmented image. The formula is simply the number of 2 times the number of diagolateral steps plus √ nal steps on the boundary. Sometimes it is necessary to smooth the boundary prior to perimeter measurement, because raggedness will elevate the measurement value.
tation, size, shape, and brightness. Many ways have been proposed to measure texture. One usually seeks a texture measure that can discriminate among different types of objects. Often considerable trial and error is required, particularly given that the human eye is blind to many types of textural differences. A common way to quantify texture amplitude is using the standard deviation of the interior gray level. The width of the autocorrelation function (Castleman, 1996) reflects texture size. Pattern textures can be characterized by measurements that quantify the size and directionality of the pattern. Sometimes objects differ from each other and the surrounding background in texture, rather than in brightness. In this case image segmentation can be done based on texture. This is done by first computing a texture image in which the gray level of each pixel is a texture feature computed in a small neighborhood local to that pixel. This image can then be segmented by conventional means.
Classification Shape measurement The most commonly used shape measurement is the shape factor, computed as the perimeter squared divided by area. The value of the shape factor is at its minimum of 4π for a circle, and becomes larger for noncircular shapes. This measurement is roughly correlated with the subjective concept of shape complexity. There are also other computationally simple ways to quantify shape (Castleman, 1996). Photometric measurement The average gray level inside an object specifies its overall brightness or density. The integrated optical density (IOD) is the sum of the gray levels of all pixels in the object, and is numerically equal to the area multiplied by the mean interior gray level. In bright-field microscopy, this measurement reflects the “mass” or “weight” of the object. Texture measurement The term texture refers to the nature of the brightness variations within an object. Texture can be random, as with the grainy appearance of noise in an image, or patterned, as with the texture of a cloth. Measuring texture is an attempt to quantify the gray level nonuniformity of an image. A texture feature is a value that quantifies some characteristic of the gray level variation within an object. A good texture feature is independent of object position, orien-
There are two forms of classification that may be used: statistical pattern recognition and classification using artificial neural networks. Statistical classification The statistical pattern recognition approach assumes that the image contains objects that belong to several predetermined types, or classes. To identify objects by classifying them into groups, one must first decide which descriptive characteristics of the objects should be measured to form the feature vector for each object. Feature selection Using the right features is important, because these values are all that are used to identify the objects. Frequently intuition guides the listing of potentially useful features. The list can then be pared down to the few best features. Good features should be: 1. Discriminating: Features should take on significantly different values for objects belonging to different classes. For example, diameter would be a good feature for separating large cells from small ones. 2. Reliable: Features should take on similar values for all objects of the same class. For example, diameter might be good for identifying red blood cells because it is relatively uniform for this class of cell.
Data Processing and Analysis
10.5.9 Current Protocols in Cytometry
3. Independent: The various features used should not be correlated with each other. For example, the diameter and area of circular cells would constitute highly correlated features, as area is proportional to the square of diameter; both essentially reflect the same property, namely, the size of the cell. Although highly correlated features can be combined (e.g., by averaging them together), they generally should not be used as separate features. 4. Few in number: The complexity of a pattern recognition system increases rapidly with the number of features used. More importantly, the amount of data required to train a classifier, and to measure its performance, increases exponentially with the number of features. With too many features, therefore, it may be impractical to acquire enough data to train the classifier adequately. Finally, adding features that are either noisy or highly correlated with existing features can actually degrade the performance of a classifier. In practice, the feature selection process usually involves testing a set of intuitively reasonable features and reducing this set to the few best. Usually none of the available features meets all of the above criteria. Classifier design Classifier design involves establishing the logical structure of the decision rule. The classifier computes, for each object encountered, values that indicate the degree to which that object resembles the objects of each class. This value is computed from the features, and the object is assigned to the class for which the value is largest. Classifier decision rules often reduce to a threshold rule that sets ranges on each feature measurement that correspond to each class. If the measurements fall within a particular range, the object is assigned to the corresponding class. One or more such class may be labeled “unknown.”
Two-Dimensional Image Processing and Analysis
Classifier training Once the basic decision rules for the classifier have been established, it is necessary to determine the particular threshold values that best separate the classes. This is generally done by training the classifier on a group of known objects called the training set. This is a collection of objects from each class that have been previously identified by some accurate method. The training set objects are measured and the ranges are determined to maximize the accuracy of the classifier when operating on the training set. When training a classifier, one can
use a simple rule such as minimizing the total number of classification errors. If some misclassifications are worse than others, one can establish a cost function that accounts for this by weighting the different errors appropriately. The ranges are then set to minimize the overall “cost” of operating the classifier. To be representative, the training set should include examples of all types of objects that might be encountered, including those rarely seen. If the training set excludes some uncommon objects, it is considered unrepresentative. If it contains items that have been erroneously classified, it is biased. If the training set is representative of the objects at large, the classifier should perform about as well on new objects as it did on the training set. Classifier testing A classifier’s performance can be estimated directly by tabulating it on a “test set” of known objects. If the test set is large enough to be representative of the objects at large, and if it is free of classification errors, the resulting performance estimate can be quite useful. It is tempting to take classifier performance on the training set as a measure of overall performance, but this estimate is usually biased optimistically. A better approach is to use a separate test set for performance evaluation. This, however, increases significantly the requirement for preclassified data. If previously classified objects are at a premium, one can use a roundrobin procedure in which the classifier is trained on all but one of the available objects, and that object is then classified. Doing this for all objects in turn yields an estimate of overall performance. Maximum-likelihood classifiers A commonly used statistical classification technique is the maximum-likelihood classifier. Suppose, for example, a classifier that can separate red blood cells (RBCs) from white blood cells (WBCs) on the basis of diameter is needed. Figure 10.5.6A shows a hypothetical probability density function (pdf) of diameter for each type of cell. These curves can be estimated from diameter histograms of training set cells. Because the area under a pdf is always unity, the histograms must be normalized to unit area. If they are unimodal and symmetrical, it is usually safe to assume they are Gaussian and simply compute the mean and standard deviation of the training set. Suppose it is also known that the specimens contain two-thirds RBCs and one-third WBCs.
10.5.10 Current Protocols in Cytometry
A
0.4 p1(x)
p(x)
0.3
p2(x) RBCs
0.2
WBCs
0.1 0
B
0.3 P1p1(x)
p(x)
0.2 P2p2(x)
0.1
0 0
T
5
10
15
x
Figure 10.5.6 Cell classifier probability density functions (pdfs). The pdfs are p1(x) and p2(x); the prior probabilities are P1 and P2.(A) Hypothetical diameter pdfs for RBCs and WBCs; (B) pdfs scaled by a prior probability that two-thirds of the specimen are RBCs. T is the maximum-likelihood decision threshold, and the shaded area is P2 times the probability that a WBC will be misclassified.
Thus, before a cell has been measured, it is known to be twice as likely to be a WBC as an RBC. It can then be stated that the prior probability of an RBC (class 1) is 2⁄3 and that of a WBC (class 2) is 1⁄3. Figure 10.5.6B shows the pdfs scaled by the prior probabilities. In this case, collecting the pdfs and prior probabilities constitutes training the classifier. Before an object has been measured, the knowledge available about it consists merely of the prior probabilities. After measurement, however, it should be possible to use the measurement and the conditional pdfs to increase the information available about class membership. After measurement, the probability that an object belongs to class i is given by Bayes’ theorem:
P(Ci | x ) =
p( x| Ci ) P(Ci ) p( x )
where p(x|Ci) is the probability density function of feature x in class i, P(Ci) is the prior probability of class i, and 2
p( x ) =
∑ p( x| C )P(C ) i
i
i=1
is the normalization factor required to make the probabilities sum to one.
Using Bayes’ theorem it is possible to combine the pdfs, the prior probabilities, and the object’s measurement together to compute the probability that an object belongs to each class. Given this information, each object can be assigned to its most likely class. In the blood cell example, an object would be assigned to class 1 (call it an RBC) if P(C1|x) ≥ P(C2|x) and to class 2 (WBC) otherwise. The maximumlikelihood classification rule dictates that the decision threshold (T) be placed at the point where the two scaled pdfs are equal—that is, where they cross, as in Figure 10.5.6b at x = T. This placement minimizes overall misclassification error. There are two possible errors this classifier can make: labeling an RBC as a WBC, and vice versa. The probabilities of these two errors can be obtained by integrating the tails of the pdfs. For example, the probability of misclassifying a WBC is equal to the area under the WBC pdf that falls to the left of the decision threshold. Normally there are more than two classes of objects, and more than one feature is measured on each object. The approach outlined here generalizes readily to these situations. Data Processing and Analysis
10.5.11 Current Protocols in Cytometry
Classification using artificial neural networks Artificial neural networks provide a different approach to object classification. Their design is inspired by the structure of biological nervous systems. A neural network is a collection of identical nodes, or processing elements (PEs), interconnected in layers. Each PE receives inputs from the PEs in the previous layer and sends its output to those in the next layer. PEs in the first layer, or input layer, receive the object measurements as inputs. PEs in the intermediate layers, or hidden layers, pass the signal on. The last layer, or output layer, has one output terminal for each class to which an object might belong. The object is assigned to the class corresponding to the output presenting the largest value. The number of layers, and of PEs in each layer, are design choices. The basic processing element of a neural network sums the product of each of its inputs times a weight factor, transforms the result according to a nonlinear transformation function, and sends the result to the nodes in the next layer. The output (O) of a PE with N inputs is simply
N O = g x i wi i = 1
∑
where xi is an input and wi is the weight for that processing element. The weights are adjusted during the training process and remain fixed in ordinary use. The nonlinear transformation function g[⋅] has a sigmoid (S) shape, and it limits the output of the PE to the range between 0 and 1. Although the outputs are always positive, weights can be either positive or negative. During a training exercise, measurements of known objects from the training set are presented to the network in random order. The interconnection weights of the PEs are adjusted slightly each time to “nudge” the output of the network in the direction of the correct answer. As training proceeds, performance improves, until the network has finally “converged” upon the proper set of weights in each PE. Relative advantages and disadvantages of statistical and neural network classifiers The advantages most often stated as favoring neural networks over statistical classifiers
are that neural networks (1) require less prior knowledge about a problem, (2) are capable of implementing more complex decision rules, and (3) are amenable to high-performance parallel processing implementations. Advocates of neural nets also point out that (4) the tremendous pattern recognition capabilities of the human brain (which uses ten billion neurons) suggest that artificial neural networks may have the potential to approach that level of performance. However, neural net implementations tested to date tend only to approach the pattern recognition performance of well-designed statistical classifiers. Disadvantages of neural network solutions relative to statistical approaches include (1) the large amount of training time they require, (2) their slower operation when implemented as a simulation on a conventional computer, and (3) the unavailability of detailed understanding of their decision-making process. There is also (4) the possibility of overtraining (allowing the training process to go on too long). If the size of the training set is not large, this can allow the network merely to “memorize” (i.e., tune itself to) that training set, rather than adjusting itself to recognize all members of the classes at large. With neural nets it is always necessary to use separate training and test sets. In general, if the objects are dissimilar, and the features that are used reflect that dissimilarity, then either a statistical or neural net classifier can perform with reasonable accuracy. In the absence of exploitable dissimilarities, however, the outlook is less encouraging.
IMAGE PROCESSING SOFTWARE There are a number of image processing and data analysis software packages available (see the November 1995 issue of IEEE Spectrum [volume 32, no. 11] for reviews). These can be used for developing and testing algorithms and for routine processing in biomedical research work. Some of the more popular ones are listed, roughly in order of increasing cost, in Table 10.5.1. In most cases demo versions are available at no charge. Because the various packages differ in emphasis and user interface philosophy, it is advisable to evaluate the software for a particular application prior to making a commitment.
Two-Dimensional Image Processing and Analysis
10.5.12 Current Protocols in Cytometry
Table 10.5.1
Image Processing Software for Flow Cytometry
Name
Source and Internet address(es)
Platform(s)
Description
NIH Image
National Institutes of Health http:/rsb.info.nih.gov/nih.image/ (free binary and source code)
Macintosh and Windows 95
Contains a library of image-processing and analysis functions, and has been extended by macros; originally developed for Macintosh, but has been ported to Windows 95; in the public domain
SCIL Image
TPD Institute of TNO, The UNIX Netherlands (originally developed by a group of Dutch universities) [email protected] (email; commercial version and full demo with 3-month license) http://www.ph.tn.tudelft.nl/software. html (free limited demo and program extensions)
TIMWIN
Technical University of Delft, The Netherlands http://www.ph.tn.tudelft.nl/software. html (free demo; binary and C source code)
Image Tool
Windows 95 and Department of Dental Diagnostics, Windows NT University of Texas Health Science Center, San Antonio, Texas http://ddsdx.uthscsa.edu/dig/itdesc. html ftp://maxrad6.uthscsa.edu (FTP; free binary and C++ source code for the executable)
LView Pro
MMedia Research Corp. http://world.std.com/mmedia
Windows 3.1 and Primarily a graphics viewer, it incorporates a Windows 95 considerable number of image processing functions that are useful for learning and algorithm development
Khoros
Khoral Research, Inc. (originally created at the University of New Mexico) http://www.khoros.unm.edu/ (free source code)
UNIX
Image-processing software development environment that uses a flowchart-type graphical user interface (GUI) called Cantata and has a large, extensible library of image-processing functions
Analytical Imaging Station
Imaging Research http://imaging.brocku.ca/
OS/2
Low-cost software package designed specifically for bioscience image analysis applications
Alice
Hayden Image Processing Group http://perceptive.com/
Designed for scientific and medical image Macintosh, processing PowerMac, Windows 95, and Windows NT
WiT
Logical Vision, Ltd. http:/web.ucs.ubc.ca./lvision/
Windows 3.1, Windows 95, Windows NT, and UNIX
Offers a collection of linear and morphologic filters, segmentation algorithms, measurement functions, and arithmetic operations for twoand three-dimensional binary, gray-scale, complex-valued, and color images
Windows 3.1 and Designed for manipulating and measuring 3.11 images, with main focus on extracting information from the images analyzed
Free image-processing and analysis program
Has a drag-and-drop dataflow GUI and a rather complete expandable library of image-processing and analysis functions; can share computational load among machines on a network continued
10.5.13 Current Protocols in Cytometry
Table 10.5.1
Name
Image Processing Software for Flow Cytometry, continued
Source and Internet address(es)
Platform(s)
Description
Image-Pro Plus
Media Cybernetics (sold through authorized dealers) http://www.mediacy.com/
Windows 3.1, Scientific image analysis software Windows 95, and Windows NT
Visilog
Noesis Vision http://www.cam.org./noesis/
UNIX, Windows 3.1, and Windows 95
Complete library of image processing and analysis functions
KBVision and Aphelion
Amerinex Applied Imaging, Inc. http://www.aai.com
Windows 95, Windows NT, and UNIX
KBVision includes a visual programming environment and a powerful library of processing and analysis functions, and can be purchased in modules to reduce cost; separate Aphelion product has extensive algorithm libraries
LITERATURE CITED Baxes, G. A. 1994. Digital Image Processing: Principles and Applications. John Wiley & Sons, New York. Castleman, K. R. 1996. Digital Image Processing. Prentice-Hall, Englewood Cliffs, N. J. Gonzalez, R. C. and Woods, R. E. 1992. Digital Image Processing. Addison-Wesley, Reading, Mass. Jain, A. K. 1989. Fundamentals of Digital Image Processing. Prentice-Hall, Englewood Cliffs, N. J. Pratt, W. K. 1991. Digital Image Processing, 2nd ed. John Wiley & Sons, New York.
Russ, J. C. 1995. The Image Processing Handbook, 2nd ed. CRC Press, Boca Raton, Fla.
KEY REFERENCE Castleman, 1996. See above. Describes in detail all operations discussed in this paper.
Contributed by Kenneth R. Castleman Perceptive Scientific Instruments, Inc. League City, Texas
Two-Dimensional Image Processing and Analysis
10.5.14 Current Protocols in Cytometry
Data Presentation
UNIT 10.6
Flow cytometry data, like any other form of data, should be presented so that it conveys information in a way that is easy to understand and interpret. There are several basic rules to follow: 1. Data should be presented in the simplest form possible. There is generally a tradeoff between simplicity and amount of information presented. Information of lesser importance may be omitted in order to simplify presentation, but at times it may be necessary to maintain complexity so that essential information is not lost. It is important to consider a potential observer’s skills and needs when deciding what information to include. Irrelevant information can serve as a distraction and add unnecessary complexity. Therefore, only pertinent data should be presented. 2. Data must be clearly identified. Poor labeling wastes time and can result in incorrect assumptions and invalid conclusions. 3. Data should be presented using standard methods whenever possible. The difficulty of understanding a new data presentation format may obscure the information that was intended to be conveyed.
NUMBERS VERSUS PICTURES It has been said that a picture is worth a thousand words. This may be true, but often in flow cytometry a single number can be worth a thousand pictures. Flow cytometric data may be presented as numbers, tables of numbers, or graphs. Choosing the optimal format requires careful consideration of the volume of data being presented and of the level of knowledge of those to whom it is being presented. In some
cases a single number resulting from the analysis of flow cytometry data may be more informative than a complex array. The typical flow cytometer collects five or more parameters simultaneously at a rate of thousands of events per second. It is neither practical nor desirable to try to examine and interpret these raw data. Instead, flow cytometry software takes these raw data values and displays them as graphs by creating distribution plots that show the number of events with the same data value over a range of possible values. The distribution of the data within these plots can then be analyzed to determine the relative proportion of events within any portion of a plot. This relative proportion is generally reported as a percentage of the total events. Historically, this has generally proved to be the number that is of the most interest.
SINGLE-PARAMETER DATA Data from only one parameter at a time is typically displayed as a simple two-dimensional distribution plot or distribution histogram (Fig. 10.6.1). Note that the y axis in this example is labeled “number of events,” and the x axis is labeled “relative intensity.” Given that a flow cytometer counts the events and measures the intensity of either scattered or fluorescent light on a relative scale, these generic labels are appropriate for any parameter. However, other labels that more specifically describe the data being displayed may be used: e.g., “cell count” for the y axis when the particles analyzed are cells, or “green fluorescence” for the x axis when displaying data from that parameter. The x axis label may even identify
Number of events
100
0 100
101
102 Relative intensity
103
104
Figure 10.6.1 Single-parameter histogram. This distribution histogram shows the relative intensity of detected light on the x axis and the number of events on the y axis. Contributed by W. Roy Overton Current Protocols in Cytometry (1997) 10.6.1-10.6.7 Copyright © 1997 by John Wiley & Sons, Inc.
Data Processing and Analysis
10.6.1 Supplement 1
the fluorescent reagent that was used: e.g., “CD45-FITC” for cells identified with CD45– fluorescein isothiocyanate. The bars that form the histogram may be unshaded or shaded, depending on the flexibility of the software and personal preference. Histograms shaded with black bars are clearest and are the best choice for achieving publication-quality diagrams. However, it is sometimes useful to compare two or more histograms by overlaying them on a single plot; in this case a combination of unshaded, lightly shaded, and darkly shaded histograms should be used. Similarly, displaying computed histogram subsets, such as for analysis of DNA content, is best done using unshaded or lightly shaded histograms (Fig. 10.6.2). Most flow cytometry software is able to determine the number and percentage of events that fall within any region of interest in a distribution histogram. A single number indicating the percentage of “positive cells”—cells that evidence a particular characteristic or marker
A
of interest—is often the only result desired from the analysis of a flow cytometry specimen, especially in the clinical laboratory. For example, this number may represent the percentage of CD4-positive cells in a blood specimen from an AIDS patient or the relative number of kappa-positive and lambda-positive B cells in a patient suspected of having a B cell leukemia. In addition to determining the number of “positive cells” in a specimen, it is often useful to quantify the relative intensity of the emitted fluorescence. Fluorescence intensity is usually reported as the average channel number of the cells of interest, but may also be reported as units of molecules of equivalent soluble fluorochromes (MESF).
DUAL-PARAMETER DATA All parameters collected from a specimen can be displayed separately as single-parameter data. However, more information about the specimen can be gained from a display that correlates the parameters. For example, in Fig-
200 aneuploid
diploid
Cell number
160
G2
120
G2
S 80
S
40
0
0
32
64
96
128 160 DNA content
192
224
256
B Cell number
1200 900 600 300 0 0
Data Presentation
50
100 150 DNA content
200
250
Figure 10.6.2 Shading of histograms. The two histogram plots show the results of DNA analysis using modeling algorithms. (A) Generated using MultiCycle (Phoenix Flow Systems). (B) Generated using the ModFit program (Verity Software).
10.6.2 Supplement 1
Current Protocols in Cytometry
ure 10.6.3 data from both red and green fluorescence have been displayed as two single-parameter histograms. It is apparent that there is a negative population and a positive population for each color; however, the correlation between the two colors is not indicated, so it is impossible to tell whether the red-positive population is green-positive, green-negative, or a combination.
The simplest way to display correlated dualparameter data is a density plot, or dot plot. In Figure 10.6.4 the data from Figure 10.6.3 have been displayed as correlated data on an x-y plot. On this plot it is possible to distinguish three major populations of cells: one positive for both markers (T helper cells), another positive for only the green CD3 marker (other T cells), and a third negative for both antibodies (B cells and
Cell count
150
0 CD3-FITC
Cell count
150
0 100
101
102 CD4-PE
103
104
Figure 10.6.3 Single-parameter histogram displays of a dual-parameter antibody-stained blood specimen. (A) Distribution of CD3–fluorescein isothiocyanate (CD3-FITC) fluorescence (green). (B) Distribution of CD4-phycoerythrin (CD4-PE) fluorescence (red).
104
CD4-PE
103
102
101
100 100
101
102 CD3-FITC
103
104
Figure 10.6.4 Correlated dual-parameter density plot of data from Figure 10.6.3. This dot plot shows distribution of fluorescence data from an antibody-stained blood specimen. CD3-FITC fluorescence (green) is plotted on the x axis and CD4-PE fluorescence (red) on the y axis. Data Processing and Analysis
10.6.3 Current Protocols in Cytometry
Supplement 1
the data distribution. Each contour level, defined by a polygon, includes within its boundaries those portions of the plot that contain a specified density of data points. For example, the outermost contour may contain all coordinates of the plot that have at least one particle; the next contour may surround those areas of the plot with at least five particles; the next may show the areas with fifteen particles; and so on. In some software, these levels may be selected by the user, but in other programs they are determined by the software. Although contour plots and dot plots both show dual-parameter data on a two-dimensional x-y plot, they each imply a third dimension: the number of particles. Dot plots indicate the number of particles by the density of the dots, whereas contour plots use the contour levels. However, these are not true three-dimensional plots.
NK cells). It would not be possible to identify all three populations using only single-parameter histograms. The density of the dots is a visual representation of the data distribution. If the dots are too dense to resolve populations, the data may be rescaled by having a single dot represent two or more particles with the same x and y values. Most flow cytometry computer programs have a sensitivity adjustment that can be set to determine the minimum number of particles required to produce a dot. Some software programs add color to the basic dot plot to identify subpopulations. A number of methods can be used to determine color groups, such as gating or cluster analysis. Another two-dimensional method of displaying correlated dual-parameter data is the contour plot (Fig. 10.6.5). This type of plot is similar to the dot plot except that polygons are used to encircle areas of the x-y plot to illustrate
7% Lin
104
Figure 10.6.5 Correlated dual-parameter contour display of data from Figure 10.6.3. CD3-FITC fluorescence (green) is plotted on the x axis and CD4-PE fluorescence (red) on the y axis.
CD4-PE
103 102 101 100 100
101
102
103
104
CD3-FITC
Cell count
Rot, tilt = 45°,30°
CD 4
Figure 10.6.6 Correlated dual-parameter isometric display of data from Figure 10.6.3. Green fluorescence is plotted on the x axis, orange-red fluorescence on the y axis, and cell count on the z axis.
-PE
C
T -FI
3 CD Data Presentation
10.6.4 Supplement 1
Current Protocols in Cytometry
In order to show dual-parameter data and the number of particles in a three-dimensional display, the data are drawn as isometric plots (Fig. 10.6.6) that display the data as a three-dimensional histogram on an x-y-z plot, where x and y are the two parameters and the z axis shows the number of particles. Because this three-dimensional plot is actually drawn in two-dimensional space, it may be necessary to draw the display from several orientations (rotate and tilt) for optimal data visualization. A contoured form of the isometric display is available in the Isocontour program (Verity Software), in which the isometric plot also has contour levels (Fig. 10.6.7).
MULTIPLE-PARAMETER DATA
Cell count
Density plots, contour plots, and isometric plots are powerful methods for displaying dualparameter data. However, sometimes it is desirable to display more than two parameters at one time.
Approaches to three-parameter data display are similar to those for dual-parameter plots. Density plots have been used in two ways. One is to create a cube or other three-dimensional polygon and show dual-parameter density plots on each side of the polygon. This approach has been used extensively in software from Becton Dickinson (Fig. 10.6.8). The other method of displaying three-parameter data in a single density plot is to create a cloud display on an x-y-z plot (Fig. 10.6.9). Contour plots of cube, polygon, or cloud-type displays have also been used to display three-parameter data. Initially it is difficult to interpret three-parameter plots, but with practice this type of display can prove useful. However, most people find it next to impossible to interpret four- or five-parameter data displayed in a single plot. In Becton Dickinson’s Paint-A-Gate software, multiple-parameter data can be displayed in either a diamond plot (four-parameter display; Fig. 10.6.10) or an N-plot, also called a ribbon
Gr
e
en
flu
ore
e sc
nc
e
20 30
40
Red
10 30
fluor
40
esce
50
nce
60
50 60
Figure 10.6.7 Isocontour display. The Isocontour program (Verity Software) combines the isometric display with the contour display on an x-y-z plot.
Data Processing and Analysis
10.6.5 Current Protocols in Cytometry
Supplement 1
plot (Fig. 10.6.11), which can display up to eight parameters. Such multiple-parameter plots require extensive experience to interpret and can be quite frustrating to novices.
SUMMARY PLOTS Several methods of displaying summarized data for multiple parameters have been developed. The first step is to reduce the distribution
Orange Fluorescence
Figure 10.6.8 Three-parameter cube dot plot. The cube display (from Becton Dickinson’s Paint-A-Gate) shows three parameters by displaying a series of dual parameter dot plots on the three visible sides of a cube.
Green Fluorescence
Red fluorescence
of data for each parameter to a single datum value. Most commonly, the reduced data values are either a percentage of the events or a number representing the distribution of events, such as the mean fluorescence channel. The next step is to plot these reduced data values for all the parameters in a single plot. This can be done by entering the data into a spreadsheet program and then displaying them as histograms. This
ed R
ce en c s re o u Fl
60
50 50 40
30 30 20
10 10 10
20
30
40
50
60
G
n ree
flu
o
c res
en
ce
Side scatter
Data Presentation
Figure 10.6.9 Three-parameter cloud display using a dot plot or density plot on an x-y-z plot. This display was produced by WinList (Verity Software).
10.6.6 Supplement 1
Current Protocols in Cytometry
process can be automated through software (for example, Prism from Coulter and Paint-A-Gate from Becton Dickinson). Alternatively, these numbers can be presented simply in tables. A list of software suppliers can be found in the SUPPLIERS APPENDIX.
Contributed by W. Roy Overton Cooper Hospital/University Medical Center Camden, New Jersey
5 CD
6-
PE
SS
Figure 10.6.10 A four-parameter diamond display (from Becton Dickinson’s Paint-A-Gate program) shows four parameters by displaying a series of dual-parameter dot plots. FS, forward scatter; SS, side scatter.
C
D3
FIT
C
Intensity
FS
FS
SS
GF Parameter
OF
RF
Figure 10.6.11 A multiparameter N-plot or ribbon display (from Becton Dickinson’s Paint-A-Gate program) shows five parameters by graphing the data on a line graph. Each line represents the data from one cell or particle. This plot shows the data from 50 blood cells stained with CD3-FITC, CD4-PE, and CD8–peridinin chlorophyll complex (CD8-PerCP). FS, forward scatter; SS, side scatter; GF = green fluorescence; OF, orange fluorescence; RF, red fluorescence. Data Processing and Analysis
10.6.7 Current Protocols in Cytometry
Supplement 1
Data Analysis Through Modeling This unit examines what models are and how they are routinely used to extract important population features. Models, however, are more than just a convenient conduit of useful information: they can be mathematical abstractions of some portion of the real world and provide a contemporary view of the interplay between a set of measurable parameters. When models are created, they seem to come to life and begin evolving towards a sharper and more meaningful picture of reality. The other intent of this unit is to capture this evolutionary process for some models used in cytometry. Models are intrinsic to the scientific process—a process that constantly evolves models toward more accurate and generalized views of reality. Model parameters are generally obtained from a set of observed data by a leastsquares approach. The veracity of model parameters is determined by careful inspection of the overall reduced χ2, parameter error analysis, and quality of the fit graphics. In cytometry, the most common application of modeling routines is for DNA analysis.
A MODELING STORY The unit begins with a fictionalized account of the discovery of an unusual artifact. The artifact represents a puzzle that forces scientists to use the powerful attributes of models to discover three hidden messages. The content of the message turns out to be another level of models and serves to demonstrate their possible importance to society. In unraveling the mysteries of the artifact, most all the important attributes of models are introduced. The rest of the unit expands these attributes, describing in more detail how models work. Topics such as linear and nonlinear least-squares are described, as well as the limitations and possible pitfalls of modeling. Most of the remainder of the unit describes many of the models that are used in the general area of DNA analysis. The unit ends with a discussion of the relatively undeveloped field of immunofluorescence modeling and leads the reader through the process of creating a model for immunofluorescence data.
Part I Imagine that a special artifact was found on the top of the Great Pyramid at Giza in the early 1300s by an English expedition. The golden
Contributed by C. Bruce Bagwell Current Protocols in Cytometry (1997) 10.7.1-10.7.20 Copyright © 1997 by John Wiley & Sons, Inc.
surface of the artifact could not be scratched or harmed in any way. It appeared to be a picture of some kind with a surface unremarkable except for unusual smooth undulations (see Fig. 10.7.1). For two centuries no progress was made in understanding where the object came from or the significance of its unusual surface. In the sixteenth century, a brilliant mathematician attempted to solve the riddle of the pyramid artifact. He arbitrarily segmented the first panel into a set of square elements, each containing the thickness measured at the square’s center. He guessed that the true picture or message was some kind of mathematical transformation on these elements, and that perhaps the circular markings surrounding each picture were important in deducing the exact nature of this transformation. After a number of unsuccessful attempts, he tried a bell-shaped function to model the broadening of the original picture or message (see Fig. 10.7.2). He presumed the degree of blurring was determined by the four perimeter spots. The mathematician wanted to find a set of unknown values that, when broadened by the bell-shaped function, yielded the observed thicknesses. He hoped these calculated values would provide some clue to the meaning of the artifact. The mathematician reasoned that if he could somehow develop a formula that quantified the overall difference between the observed and calculated thicknesses, he could eventually figure out a way to minimize the formula and obtain the most probable set of original unbroadened data. He first tried summing up the differences between the observed and predicted data but quickly realized that positive and negative differences were canceling each other regardless of magnitude. He tried taking the absolute value of the differences but was later confounded by how to minimize the sum of absolute values. The solution to the problem became evident when he squared the differences. Not only did the sum of squared differences eliminate the sign effect, but it was also very amenable to further mathematical manipulations. Now his problem was to minimize this sum of squared differences (SSQ). When he expanded the squared term, he realized the function was amenable to minimization with respect to each unknown thickness value. The un-
UNIT 10.7
Data Processing and Analysis
10.7.1 Supplement 1
known thicknesses could be held at some constant, arbitrary value, and the “best” value of a single thickness could be found that minimized the function. By differentiating the SSQ function with respect to the single varying thickness and setting it to zero, he could obtain an equation that determined what this best thickness would be, given all the other thicknesses. By performing this same operation on all the rest of the thicknesses, he ended up with a set of simultaneous equations with an equal number of unknowns. Now, all the mathematician needed to do was solve these equations. It took two months of hard work, but when he finally had calculated the set of unbroadened values, he was able to decode part of the message this artifact had so successfully held secret for two centuries (see Fig. 10.7.3). The mathematician was elated to find the meaning of the exposed equation so clear. He had suspected that there was a relationship between force, mass, and acceleration from the
Data Analysys Through Modeling
work of Galileo and Kepler but had no idea that the relationship was so elegantly simple. Given a constant force (F) and a known mass (m), the equation or model predicted the acceleration (a) of any given object. Based on this model and another that related mass and distance to gravitational force, scientists were finally able to explain and predict the motion of planets, moons, and tides. He eventually published this great work in 1686 in “Philosophiae Naturalis Principia Mathematica.” The mathematician was, of course, Sir Isaac Newton.
Part II The rest of the artifact’s encoded pictures remained an enigma for another two and onehalf centuries. During this period, scientists did not seem driven to find out more from the artifact. In fact, by the end of the 1890s, scientists had a certain complacency about describing nature with models. Most scientists felt that all there was left to do was to make more and more precise measurements because the funda-
Figure 10.7.1 Original artifact picture found in 1307 by an imaginary English expedition. Contours indicate the surface topography.
10.7.2 Supplement 1
Current Protocols in Cytometry
mental models of nature were already known (Badish, 1972). Then in the early 1900s, a physicist working as a Swiss patent examiner in Bern became fascinated with the artifact puzzle. Where did it come from? What were the rest of the messages? He pondered these questions and decided to try his hand at a possible solution. After several years of trial and error, the physicist finally decoded the second picture. As was the case for the first picture, the key to unlocking this artifact secret was finding the correct function used to blur the original picture. The physicist decided to use a function he had developed for calculating the relative velocities of observers approaching each other. After he solved the simultaneous equations, he could hardly believe the new equation that appeared (Fig. 10.7.4). Up until this time, Newtonian mechanics predicted that a constant force acting on a mass would result in constant acceleration. But this new addition to Newton’s model indicated that as the velocity of an object (v) approached the speed of light (c), the mass
of an object increased rapidly, preventing the object from ever attaining the speed of light. The equation led the physicist to develop other models, including a new gravitational force model, which replaced the simpler Newtonian models when they explained a small irregularity in the precession of Mercury’s orbit and predicted the bending of light from distant stars by the mass of the sun. The physicist published his work as the specific and general theories of relativity. The physicist was Albert Einstein.
Part III The final artifact panel was elucidated in 2015 by a molecular geneticist when he noted the fractal nature of the patterns ...
The Real Story The above story, for the most part, is untrue. There is no such artifact as far as this author knows and the works of Isaac Newton and Albert Einstein were their own and not guided by some alien influence. Also, there are a few anachronisms such as the use of least-squares
unknown picture
broadening function
digitized picture
observed picture
Figure 10.7.2 First attempt to solve the artifact’s hidden message. The top panel of the picture (see Fig. 10.7.1) was segmented into a set of square elements, each containing the thickness measured at the square’s center. The circular markings surrounding each picture provided information on the degree of this broadening. The solution was eventually found by assuming that the hidden picture was blurred by a bell-shaped function to yield the observed picture. A least-squares solution was necessary to find the hidden message.
Data Processing and Analysis
10.7.3 Current Protocols in Cytometry
Supplement 1
analysis in the 16th century. But wound into this story is a great deal of truth about the basic principles of models. Models can be mathematical abstractions of some part of the laws that make up our universe. The language of models is mathematics because models need to be able to explain and predict observed data. This unit will explain what models are and how they are used to analyze data. It will also serve as a guide to finding out more about some of the models that have been created in cytometry and elucidate how they evolved into their current form. The artifact story demonstrates the important evolutionary nature of models. A model is the best representation of the true relationships between a selected set of parameters at the time the model is created. Scientific progress is made when the model is refined to incorporate already known data that did not fit the old model or to predict new data. The transformation of Newton’s mass term into Einstein’s mass/velocity relationship was significant because it predicted the increase of mass at relativistic speeds. This prediction was later confirmed by
high-speed particle experiments. The story also demonstrates two relationships between least-squares analysis and models. The idea of creating a function that sums the square of the differences between the observed data and a model’s prediction of the data is a powerful concept. By minimizing the sum of squares, the parameters that make up the model can be estimated with great accuracy. This estimation process is very important in cytometry because cytometrists are in the business of measuring features of cell or other particle populations. This modeling process enables the accurate estimation of important parameters that may have scientific, diagnostic, or therapeutic implications. The story also points out another important aspect of models and least-squares analysis. When attempts are made to quantify specific particle attributes such as DNA fluorescence, there is an inevitable blurring of measurement due to a number of random effects—e.g., sample flow rate fluctuations, variability of particle position in the laser beam, photomultiplier tube and amplifier electronic noise, and a host of
digital solution
image solution
Data Analysys Through Modeling
Figure 10.7.3 The mathematician’s solution to the first part of the artifact’s message. The F = ma (force equals mass times acceleration) message eventually became known as the mass action law and served as a model describing the motion of all known objects until the early 1900s.
10.7.4 Supplement 1
Current Protocols in Cytometry
other effects. If the nature of this blurring effect is known, these broadening features can be added to the model, the least-squares analysis can be performed, and their effects greatly diminished. This kind of process is known as a super resolution and is routinely used in spy satellites, general image enhancement, and myriad other applications. In other words, with knowledge of how images or data are blurred, it is possible to reverse the blurring process mathematically, which is exactly what the mathematician and physicist did in the artifact story. It is sometimes convenient to split models into two parts: the kernel and the line-spread function. The kernel is a mathematical expression that describes the relationships between the model parameters. These parameters may represent some attributes of a real world system such as S phase fraction or proliferation index. The line-spread function describes the point-
by-point broadening of the kernel due to inaccuracies in the measurement process. In cytometry models, the most common error distribution for the line-spread function is the Gaussian (Abramowitz and Stegun, 1964; Dean, 1990). In the artifact example, Newton used a two-dimensional Gaussian to broaden his model kernel to obtain a best fit to the observed set of thicknesses. The important parameter of the Gaussian that controls the degree of broadening is the standard deviation. The general mathematical operation that combines the kernel and the line spread is often referred to as convolution. Reversing this process with algorithms such as least-squares analysis is frequently called deconvolution. In cytometry, models are normally constructed to quantify observed data (e.g., histograms) and to deconvolve the broadening effects in the measurements. Eliminating the broadening effects due to measurement noise
Digital solution
Image solution
Figure 10.7.4 The physicist’s solution to the second part of the artifact’s message. This new enhancement to Newton’s model indicated that as one’s velocity (v) approached the speed of light (c), the mass (m) of an object increased rapidly, prohibiting the object from ever obtaining the speed of light. The equation led the physicist to develop other models, including a new gravitational force model. This new model supplanted Newtonian models with specific and general theories of relativity.
Data Processing and Analysis
10.7.5 Current Protocols in Cytometry
Supplement 1
essentially minimizes overlap between populations. It is important to distinguish between a modeling process and a partitioning process. When data are partitioned via regions and gates, boundaries are created separating one type of particle from another. Each channel contains either one type of particle or another. In modeling, the categorization is done on probabilistic grounds. A single channel can have one, two, or more different types of particles. The most successful use of these models has been in DNA analysis, for two probable reasons: the very tight regulation cells have on the total amount of DNA present yields a very precise and interpretable unbroadened function, and a fairly firm understanding of the mechanics of cell division and published data indicate the prognostic importance of some DNA model parameters.
LEAST-SQUARES ANALYSIS The process used to obtain model parameter estimates from a set of observed data is called
least-squares analysis (Meyer, 1975). In the story, Isaac Newton reasoned that if he formed a function that was the square of differences between model-derived and observed data, he could take the derivatives of this function with respect to each model parameter, set them to zero, and then solve for the model parameters. In his case, the model parameters were the original artifact message. Although leastsquares analysis was not invented until the middle 1800s by Gauss (Gauss, 1855), Newton could potentially have done this analysis in the middle 1600s because he knew how to differentiate functions.
Linear Least-Squares Analysis Figure 10.7.5 demonstrates how the leastsquares process works for a simple example, the best fit of a straight line through a set of observed data. If the intercept parameter of the line is fixed and the slope is allowed to vary, the least-squares function forms a parabola with its nadir at a clearly defined slope value
A Sum-of-squares (× 106)
10.0
8.0 6.0 4.0 2.0 0.0 4
B
6
8 10 12 Slope (intercept fixed)
14
16
Sum-of-squares (× 105)
7.74 7.72
7.70 7.68 7.66
7.64 7.62 7.60
8 10 12 14 16 18 20 22 24 26 28 30 32 Intercept (slope fixed)
Data Analysys Through Modeling
Figure 10.7.5 Sum-of-squares of a line through a set of data points as a function of slope (A) and intercept parameters (B). The functions are parabolic with minimums defined where the slopes of the curves are zero. Least-squares analysis involves solving a set of equations to find these minimum values.
10.7.6 Supplement 1
Current Protocols in Cytometry
(see Fig. 10.7.5A). The position of the lowest point is where the slope is zero. The second panel, Figure 10.7.5B, demonstrates the same pattern with the slope fixed and the intercept allowed to vary. When the derivatives of the least-squares function are set to zero, the resulting equations determine the point with the lowest least-squares. This point is where the line model best fits the data. In some cases, these model parameter estimates, slope and intercept, have a direct relationship with attributes of the real world. Newton in this story was lucky: the leastsquares function was easily differentiable and the derivatives were amenable to forming equations that could be solved. Functions that have this property are called linear functions (Meyer, 1975), and when linear functions are used in least-squares analysis, the process is called linear least-squares analysis. Most cytometry models, however, are more complex and require a different, perhaps less elegant, approach.
value and repeats the process until it either reaches some maximum number of iterations or the sum-of-squares changes by less than some predefined amount. As with the linear least-squares method, the best fit of the model to the observed data yields parameter estimates that may represent real world attributes if, indeed, the model correctly represents the real world. False solutions One of the potential problems in nonlinear least-squares analysis is the possibility of obtaining a false solution. If the model parameter estimates are not very accurate, it is possible that the program will find a local sum-ofsquares minimum not at the global minimum. Most modern modeling programs (e.g, ModFit 5.2 and ModFitLT, Verity Software House, and MultiCycle AV DNA Analysis, Phoenix Flow Systems) have very accurate estimating routines that mitigate the possibility of false solutions, but it is always prudent to visually inspect all fits because false solutions are rarely subtle.
Nonlinear Least-Squares Analysis In nonlinear least-squares analysis (Bevington, 1969) the process of finding the lowest sum-of-squares value is more complicated. The best fit of a Gaussian function through some data points can illustrate some key aspects of nonlinear least-squares. Gaussian functions have three parameters: mean, standard deviation, and area. For a moment, let the standard deviation of the Gaussian be some arbitrary value of the data. If the mean and area are allowed to vary between reasonable limits, a set of sum-of-square values results. A plot of a sum-of-square value for each mean and area combination describes a surface (see Fig. 10.7.6). The strategy in nonlinear least-squares analysis is to find the location of the lowest value on this surface. The most popular computer algorithm that finds this value is the Marquardt compromise (Marquardt, 1963). The nonlinear least-squares method begins by estimating the parameters of the model. Once the computer has a starting location on the sum-of-squares surface, it can estimate the curve’s gradient at that position by either changing each parameter slightly and determining the slope or by differentiating the sumof-squares function and evaluating the gradient at the current estimate position. Once this gradient is known, the Marquardt method uses a combination of approaches to approximate the location of the lowest portion of the surface. The algorithm changes its position to this new
Error analysis The local surface topography around the sum-of-squares minimum value provides information on the precision of the model parameter estimates. If the surface has a very shallow bottom near the sum-of-squares minimum, the best fit parameters can be inaccurate because different sets of parameter values can yield similar sum-of-squares. However, because the character of the surface can be measured, it is possible to estimate this type of error. The discussion of DNA models will examine error analysis in more detail (see S phase fraction error analysis). Limitations of modeling Complex models can generally be broken down into discrete components. In DNA models, for example, typical model components might be for the G0/G1, S, and G2M populations. As long as model components of similar shape do not overlap, models can provide accurate estimates when analyzed by a nonlinear least-squares method. When two or more model components of similar shape overlap, a groove forms in the sum-of-squares surface where an infinite number of parameter combinations yield identical minimal sum-of-squares solutions (see Fig. 10.7.7). This modeling limitation can be encountered in DNA tetraploid models where the DNA diploid G2M model component is under the DNA tetraploid G0/G1 component.
Data Processing and Analysis
10.7.7 Current Protocols in Cytometry
Supplement 1
In DNA near-diploid populations, the diploid and aneuploid S phases can also significantly overlap. Commonly used DNA analysis packages use different strategies to avoid these problems. Choosing the correct model can be considered another limitation in modeling. If the wrong model is selected, then no amount of sophisticated processing will provide accurate answers. Many times the decision on the appropriate model is up to the subjective whims of the software user. This limitation can be significantly reduced by appropriate education of users and automated selection of models. One course, the “Annual Course in Flow Cytometry,” is held alternately at Bowdoin College in Maine and Los Alamos National Laboratory in New Mexico. (For more information about this course contact James Jett, Life Sciences Division MS M888, Los Alamos, N. Mex. 87545.)
χ2 and Reduced χ2 Not all the data should be weighted the same in the least-squares formula. The number of events in each histogram channel is not exact and varies approximately as a Poisson distribution. The standard deviation of this distribution is equal to the square root of the number of events in a channel. Thus, a channel that has 100 events has a relative variability of 10⁄100 or 10%. A channel that has 900 events has the smaller relative variability of 30⁄900 or 3.3%. This lessening of relative variability can be visualized by comparing a histogram with 1000 events to one of 10,000 events. The 10,000event histogram will appear to be smoother due to less relative variability. If the least-squares is weighted by the variance and the standard deviation is squared, the sum-of-squares statistic transforms to the χ2 statistic (Bevington, 1969). Modeling programs usually minimize the χ2 statistic in order
Are a
Figure 10.7.7 Sum-of-squares surface of two overlapping Gaussians through a set of data points. The Gaussians were forced to have the same mean and standard deviation parameters. The resultant surface groove demonstrates what happens when two model components of similar shape overlap. There are an infinite number of solutions to the minimum sum-of-squares value resulting in an unacceptably high error in parameter estimates.
Area 2 Data Analysys Through Modeling
Me an
Figure 10.7.6 Sum-of-squares surface of a Gaussian through a set of data points. The standard deviation was fixed at some arbitrary value and the mean and area parameters were allowed to vary. Nonlinear least-squares analysis uses an iterative approach to finding the minimum.
Area
10.7.8 Supplement 1
Current Protocols in Cytometry
to properly weight the data. Although the χ2 is used internally by the nonlinear least-squares algorithm, it is rarely exposed to the user. Because χ2 is a function of the relatively arbitrary number of channels modeled, it is difficult to interpret directly. The χ2 value can be normalized by the number of degrees of freedom, which is the number of channels minus the number of model parameters (Bevington, 1969), to yield the more interpretable reduced χ2. Appropriate fits of models have reduced χ2 values ranging from 0.9 to 3.0. If the reduced χ2 is >3.0, the model may be missing important data features. If it is <0.9, the model may be overfitting the data by fitting statistical noise.
Limitation of Least-Squares Analysis In the story, Newton squared the differences between his model of the artifact’s message and the observed thicknesses to construct a sum-ofsquares formula amenable to a minimization solution. In reality, it was not until 1855 that Carl Friedrich Gauss proposed the leastsquares method as a process for obtaining model parameter estimates from a set of observed data. The idea of squaring the differences to eliminate signs at first seems heavy handed, but its advantage lies in the practicality of further mathematical manipulations, especially differentiation. Unfortunately, squaring the differences to eliminate sign effects does have some undesirable consequences. A few outlier points can have a profound effect on the final regressed parameter values. The squaring transformation
augments the outlier differences to the point where they can dominate the final solution (see Fig. 10.7.8). Note how a few outlier points yield a regression line that is not representative of the main bulk of data. A class of robust algorithms that are not as sensitive to outliers as leastsquares has been developed (Press et al., 1992). A robust algorithm is relatively insensitive to small departures from the idealized assumptions. Typical cytometry models, such as DNA modeling, do not have much of a problem with outlier data, and therefore, least-squares approaches are appropriate. However, cytometry analyses, such as automated software compensation and regression of standard beads in quantitative cytometry, have either outliers or a dynamic range that mitigates the effectiveness of least-squares approaches. These types of analyses are particularly well suited to the robust estimator type of techniques.
SPECIFIC MODELS FOR CYTOMETRY The main bulk of modeling analysis in cytometry has been restricted to DNA histogram analysis (see Fig. 10.7.9). However, there are many other specialized areas where models have been successfully tried in cytometry.
Modeling the DNA Content of Cells This section describes a set of models that describe the characteristics of DNA histograms. A DNA histogram is formed by staining cells or nuclei with DNA-specific dyes, which
15000 Parameter 2
Robust straight-line fit 10000
5000 Least-squares fit 0 − 20
0
20
40 60 Parameter 1
80
100
120
Figure 10.7.8 Demonstration of the lack of robustness of least-squares analysis with outlier data. A few outlier points can have a profound effect on the final least-squares regressed parameter values. Robust straight-line fitting algorithms are much less sensitive to outlier data.
Data Processing and Analysis
10.7.9 Current Protocols in Cytometry
Supplement 1
normally report the stoichiometric binding as fluorescence intensity. The fluorescence of each event is digitized, stored, and eventually presented in histogram format, where the x-axis is proportional to fluorescence intensity or DNA content and the y-axis is proportional to number of events. DNA models attempt to describe mathematically the underlying cycling and non-cycling populations as well as the characteristics of measurement error. The evolutionary aspect of modeling is well exemplified by the history of DNA modeling. The following descriptions highlight some of the major steps in this evolutionary process. Polynomial DNA model In 1974, Dean and Jett proposed the first DNA histogram model (Dean and Jett, 1974). They decided to use a Gaussian function to represent G0/G1 and G2M populations and settled on a broadened second-degree polynomial to represent S phase (see Fig. 10.7.9C). They felt that the zero-degree (rectangle) and firstdegree (trapezoid) polynomials were too restrictive for the shape of Chinese hamster ovary (CHO) cell S phases, and that higher-order degree polynomials (>2) did not provide significantly better fits and might have some undesirable side effects at the G0/G1 and G2M end points (P. Dean, pers. comm.). The original Dean and Jett G0/G1 and S line spread was restricted to holding the coefficient of variation constant with the G2M standard deviation allowed to be an unrestricted parameter. Later versions of the model, however, relaxed the constant coefficient of variation restriction by allowing it to vary linearly between the G0/G1 and G2M peaks.
Data Analysys Through Modeling
Multi-Gaussian DNA model In 1976, a radically different DNA histogram model was proposed by Fried (Fried, 1976), who wanted a less constrained S phase than the one proposed by Dean and Jett. He approached the problem much like Newton in the artifact story and constructed his model to have a series of eight to ten Gaussians for S phase. Fried did not want to impose any artificial restraints on the shape of S phase. Because of the unconstrained nature of the Fried model, it became a popular analytical strategy for perturbed cell populations. The problems that plague this approach were readily apparent in his first publication. In many cases the Gaussians overlapped so much that the final least-squares solution in-
volved negative Gaussians. Fried fixed that problem by having his algorithm delete the offending Gaussian(s) from the model and then reanalyzing. Fried later tried using some empirically derived methods in spacing the Gaussians so that they did not overlap too much (Fried, 1979). The fundamental problem with the Fried model was that his kernel did not approximate the reality of the S phase distribution. His model broke the S phase into a series of discontinuous spikes when in reality S phase is a more continuous function. No matter how hard Fried and others (Jett, 1977) tried to fix the problems with his model, its ultimately fatal flaw was this disparity between the model kernel and reality. Multi-rectangle/trapezoid DNA model In 1979, this author (Bagwell, 1979) reasoned that the idea of creating an unconstrained S phase model was valid, but it had to be a continuous function—there could be no gaps in the S phase kernel. The simplest extension of the Fried approach was to use broadened rectangles instead of Gaussians (see Fig. 10.7.9D). Many of the problems with the Fried approach disappeared with this unconstrained but continuous model. The model was later validated by rigorous S phase sorting experiments (Sheck et al., 1980). Two years later, the model was improved by using broadened trapezoids (Bagwell, 1981a) instead of rectangles (see Fig. 10.7.9E). For perturbed populations of cells, this model remains as one of the methods of choice in deriving accurate S phase analysis from perturbed populations. Polynomial and Gaussian DNA model The Dean and Jett polynomial could theoretically be made more flexible by resorting to higher degrees (>3), but the problem with this approach was the erratic behavior of the function at the G0/G1-S and S-G2M boundaries. The 95% confidence limits of high-order polynomials flare out significantly at the end-points. In 1980, Fox added flexibility to the second-degree polynomial by adding an extra unconstrained Gaussian to the S phase model (Fox, 1980). This model has also stood the test of time and is an accepted method in modeling S phases in perturbed cell populations. Relative rate of DNA synthesis By examining the shape of the S phase distribution, it is possible to make deductions on the relative rate of DNA synthesis as cells
10.7.10 Supplement 1
Current Protocols in Cytometry
A
F 1000
2100
Number
Number
2800
1400
800 600 400
700
200
0
0
350
B
G
280
Number
Number
280 210 140
210 140
70
70
0
0
350
C
H
280
Number
Number
280 210 140
210 140
70
70
0
0
350
D
I
280
Number
Number
280 210 140
210 140
70
70
0
0
350
E
J
280
Number
Number
280 210 140
210 140 70
70 0
0
0
40
80 120 Channels
160
0
50
100 150 200 250 Channels
Figure 10.7.9 Nonlinear least-squares fits of several models to DNA diploid and DNA aneuploid histograms. (A) DNA diploid histogram. (B) DNA diploid histogram re-scaled to better visualize S-phase. (C) Broadened polynomial fit. (D) Three broadened rectangles. (E) Three broadened trapezoids. (F) DNA aneuploid histogram with significant debris and aggregates. (G) Re-scaled DNA aneuploid histogram. (H) Broadened rectangles. (I) H plus the single-cut model component fitting the debris. (J) I plus the aggregate model component fitting aggregates. Data Processing and Analysis
10.7.11 Current Protocols in Cytometry
Supplement 1
Data Analysys Through Modeling
traverse the S phase compartment. Cells traveling through S phase can be conceptualized as cars speeding along a circular track. When cars reach a straightaway, the distance between them increases and when they slow down on curves, the distance decreases. Areas of track where cars are moving quickly will have relatively fewer cars than where they are moving slowly. By measuring the frequencies of cells in either S phase compartments (Bagwell, 1979) or separate channels (Dean and Anderson, 1975), relative DNA synthesis rates can be computed. The major assumption made in this type of analysis is that the population is homogeneous with an asynchronous exponential growth rate. This type of analysis never really became popular and now has virtually disappeared from cell-cycle analysis programs. The major reason for this loss of interest is probably the relatively restrictive assumptions about the population’s growth characteristics.
to modeling flow cytometric DNA histograms. One possible explanation of the abundance of the diverse DNA histogram deconvolution strategies is the relatively flat and featureless S phase component and the ability of a number of alternate functions to fit it. Despite all these competing strategies, current popular DNA modeling software uses only Gaussians for G0/G1 and G2M phases and either broadened rectangles, trapezoids, or polynomials for S phase. In 1983, a paper written by Hedley (Hedley, 1983) changed the course of evolution of DNA histogram modeling. Hedley described a technique for obtaining reasonable DNA histograms for cells obtained from paraffin-embedded tissue. This technique was largely responsible for a host of papers demonstrating the prognostic significance of DNA aneuploidy and later S phase fraction for a number of solid tumors.
Automatic analysis Another area of rapid evolution is automatic analysis of DNA histograms. This process begins by finding all the valid peaks in a DNA histogram. In this context, a peak is defined as an observable histogram distribution that most likely corresponds to a discrete underlying population. One approach to valid peak finding is to first smooth the data with a suitable lowpass smoothing routine, then perform univariant frequency-dependent clustering (Salzman et al., 1975; Conrad, 1987) to find all the possible peaks. These peaks are then further filtered by a set of criteria to yield a final set of peaks. The pattern of filtered-peak locations is then assigned a particular ploidy pattern. A given pattern corresponds to a particular model selection, range assignments, and subsequent estimates. Almost anticlimactically, the last step is to perform a nonlinear least-square analysis of the model. Automatic analysis strategies such as the one described have the major advantage of providing a uniform approach to DNA analysis, enabling different laboratories to obtain identical results when analyzing the same histogram. However, these automatic strategies are still evolving and can occasionally give inaccurate ploidy assessments. It is imperative that these analyses be validated by careful visual inspection of the fit, the reduced χ2 value, and the S phase error analysis.
DNA aneuploid model DNA histograms of cells from paraffin-embedded solid tumors are generally more complex than those derived from tissue culture cell lines. Many histograms have two or more cycling DNA aneuploid populations as well as a fair amount of debris and aggregates (see Fig. 10.7.9F and G). Although Fried attempted to model DNA aneuploid histograms (Fried, 1976, 1977), his multi-Gaussian S phase was not appropriate for overlapping S phases because there was no way to distinguish between one aneuploid S phase Gaussian and another. In order to successfully model overlapping S phases, DNA models evolved toward simpler model components for S phase, generally either a single broadened rectangle (see Fig. 10.7.9H) or trapezoid. The simple S phase model components can handle overlapping S phases from histograms with a DNA index 1.2 to 1.3. Obtaining accurate S phase estimates from paraffin-embedded tissue was largely thwarted by the relatively high contaminating amounts of debris and aggregates. Of the two contamination problems, debris was the major problem because it was possible to reduce the aggregate contamination by electronic pulse processing (Sharpless et al., 1975).
Clinical DNA analysis During the 1980s, scores of papers were published demonstrating some new approach
Modeling debris Preparation of paraffin-embedded cells for flow cytometry inevitably results in some level of nuclear damage. This process can fracture, mince, or slice nuclei, forming a continuous distribution throughout the DNA histogram.
10.7.12 Supplement 1
Current Protocols in Cytometry
The exponential initially used for modeling debris was a poor choice because it did not have the same underlying shape as the debris from either fresh, frozen or paraffin-embedded tissue. The beginning of a solution to the problem came when Rabinovitch (1991) created a debris model from the histogram data itself. He assumed that some constant fraction of events in a particular channel was equally distributed over all lower channels. This distribution was later enhanced into two nonuniform distributions: one representing debris from paraffinembedded tissue (see Fig. 10.7.9I) and the other from fresh or frozen tissue (Bagwell et al., 1991). The concept of using observed histograms to create model components had an important evolutionary impact on DNA histogram modeling. Modeling aggregates Although aggregate contamination can be reduced by electronic pulse gating, user-defined aggregate gates are largely subjective and create variability in the subsequent S phase fraction estimates. Software approaches to aggregate compensation were initially restricted to methods that attempted to eliminate doublets from G2M peaks by analyzing singlet and triplet peaks (Bagwell, 1979; Beck, 1980; Sato, 1990). In 1991, Rabinovitch decided to try to create a histogram-dependent model for aggregates. He assumed that the probability distribution of aggregate forms was dependent on the number of ways one particle could attach to another (Rabinovitch, 1991). This author’s attempt to enhance this method involved using physical chemistry equilibrium equations to describe the aggregate probability distribution (Bagwell, 1993). Figure 10.7.9I shows a typical analysis of a DNA aneuploid histogram with debris and aggregate compensation. S phase fraction error analysis As mentioned earlier (see χ2 and Reduced 2 χ Analysis), to properly assess the validity of a model describing a set of data, it is important to examine the reduced χ2, the best fit, observed data graphics, and the model parameter error estimates. In the case of DNA histograms, the model parameter deemed most important is the S phase fraction(s). Unfortunately, S phase fraction is not a direct model parameter. Instead, it is an algebraic expression involving a number of related model parameters. The proper way of calculating the ultimate uncertainty of a derived parameter such as S phase
fraction has been another area of relatively recent evolution. A model builder might use formulae derived from propagation of error theory (Bevington, 1969) to estimate a calculated parameter such as S phase fraction; however, when the accuracy of the error estimate is examined, it is common to find inaccuracies of the error estimate >100%. The approximations made in propagation of error theory are many times just not valid. Other means of estimating these errors have been proposed. These methods are based on Monte Carlo methods that have recently gained acceptance by statisticians (Efron, 1982). Because the Monte Carlo method is iterative, the S phase error distribution can be estimated directly. Traditional histogram acceptance criteria (Shankey et al., 1993)—number of events, coefficient of variation, debris, and aggregate contamination—can all affect the width of the S phase error distribution. By examining this distribution against a set of laboratory S phase cutoffs between low, intermediate, and high S phase states, it is possible to calculate the certainty that a model-derived S phase fraction is in one or more of these states. (See Figure 10.7.10 for a typical DNA analysis report showing cell cycle modeling and S phase error analysis.) Another approach to this problem is to analyze a histogram with a number of closely related models (Rabinovitch, 1993) and establish a range of S phase fractions. This idea is based on the fact that similar models analyzing noisy data can yield divergent answers. A good example can be found by analyzing some data with a line, parabola, and cubic equation (see Fig. 10.7.11). If the data are highly correlated (Fig. 10.7.11A), there will be little difference between the best fit solutions for these functions. However, if the data have little correlation (Fig. 10.7.11B), the best fit functions can produce quite divergent fits, resulting in a range of parameter values such as the y-axis intercept. Unfortunately, there is no means of interpreting this range probabilistically. However, it does provide valuable information on the degree of model parameter uncertainty. Future DNA model evolution As can be seen, the general area of cell-cycle analysis is well trodden by model builders. Does this mean that the same complacent attitude as scientists had in the late 1800s is appropriate? Certainly not. There are probably new
Data Processing and Analysis
10.7.13 Current Protocols in Cytometry
Supplement 1
ideas, such as the histogram-dependent model components, waiting to be unearthed and explored. In the opinion of this author, one area that deserves more exploration is in many ways the most obvious: the DNA content line-spread function. What is known about this line-spread function? Dean and Jett’s original publication assumed the line-spread function was restricted to constant coefficient of variation with the
exception of the G2M standard deviation, which was allowed to be a free model parameter. Why did they make this exception? In all probability it was because they had better fits when the standard deviation was a free parameter than when restricting it to be twice the G0/G1 standard deviation. Later, it was found that the best relationship between channel number and standard deviation was a linear one defined by a slope and
Quick Service Laboratory Nome, Alaska Date acquired: 26/10/88 File: M5487f1.fcs Source: SAMPLE ID Case: PATIENT ID Analysis type: Automatic analysis Prep: Paraffin
Number
600
DIPLOID: 47.72 % Dip G0-G1: 98.11 % at 64.59 Dip G2-M: 1.89 % at 129.17 Dip S: 0.00 % G2/G1: 2.00 Dip %CV: 3.67 ANEUPLOID 1: 52.28 % An1 G0-G1: 87.89 % at 73.30 An1 G2-M: 7.34 % at 151.30 An1 S: 4.77 % G2/G1: 2.06 An1 %CV: 4.21 An1 DI: 1.13
400
Total S-Phase: 2.49 %
Extra Pop: % Debris: 17.99 % Aggregates: 15.92 % Modeled Events: 12823 RCS: 1.866
200
Diploid B.A.D.: 14.17 % Aneuploid B.A.D.: 37.77 % Total B.A.D.: 21.99 %
0
0
50
100
150
200
250
DNA Content
S-Phase Assessment:
Impressions:
Tissue Type: All Model type: Aneuploid Aneuploid S: Low Calculated p-value: p<0.01 S-phase Boundaries: 10.0 and 20.0
This sample has a DNA aneuploid component accounting for more than 50% of the cells sampled. S-phase analysis demonstrates a low number of tumor cells in a cycling state.
Total S: Low Calculated p-value: p<0.01 S-phase Boundaries: 10.0 and 20.0
ModFitLT V2.0(Win32)
Data Analysys Through Modeling
Figure 10.7.10 Typical DNA analysis and integrated report. DNA models provide accurate estimates for S phase fraction, accounting for most debris and aggregate contamination. S phase can be automatically categorized into low, intermediate, and high states with an associated p-value. Filled triangles along the x-axis indicate detected peak locations.
10.7.14 Supplement 1
Current Protocols in Cytometry
intercept (Brunsting et al., 1979). But this does not answer the question of how the G2M standard deviation is related to the G0/G1 standard deviation. It could be a factor of two, but it also could be any other number as well. The relationship may indeed be correct, but it might be too general and flexible. Assuming that a G2M cell has two complements of DNA compared to a G0/G1 cell and further assuming that the uncertainty of measuring both these complements is uncorrelated, then simple statistics predicts that there should be a factor of square root of two difference between the G0/G1 and G2M standard deviations. Taking the other extreme and assuming that the uncertainties are perfectly correlated, the relationship should be a factor of two. The true answer seems to be between these two
A
extremes (see Fig. 10.7.12). There probably is more structure within this relationship, and any discovery that refines line-spread functions will have profound effects on all standard DNA models.
Chromosome Analysis Single-parameter histogram distributions of chromosomes can be modeled with a set of Gaussians (Gray, 1974). Usually these models need some interaction with the user because chromosome homologues can appear anywhere in the histogram. Dual-parameter chromosome histograms (e.g., chromomycin versus Hoechst 33258) can be modeled with two-dimensional Gaussians also known as bivariates (Dean, 1990).
70 60
Y values
50 40 30 20 10 0 0
2
4
6 X values
8
10
0
2
4
6 X values
8
10
B
Y values
60
40
20
0
Figure 10.7.11 Demonstration of the multiple model method of assessing parameter error. Fit of line, parabola, and cubic with correlated and uncorrelated data. If the data are highly correlated (A), there will be little difference between the best fit solutions for these functions. However, if the data have little correlation (B), the best fit functions can produce quite divergent fits, resulting in a range of parameter values such as the y-axis intercept.
Data Processing and Analysis
10.7.15 Current Protocols in Cytometry
Supplement 1
Proliferation Models
Relative SD
In 1987, while investigating the staining characteristics of some long-chain hydrocarbon dyes, Paul Karl Horan found a dye that seemed to stain tissue culture cells without any significant cytotoxicity. He also noted that the dye would not leave the cells after successive washings. It was later determined that the dye, now known as PKH, inserted itself more or less permanently between the lipid layers of the cellular membranes and did not seem to significantly alter cell function (Horan and Slezak, 1989). Thus, cells stained with PKH could be tracked when delivered into a host organism. Cell-tracking dyes such as PKH have many possible research and therapeutic applications (Horan et al., 1990), but the one application germane to this discussion is the analysis of cellular proliferation. Under the appropriate staining conditions, these dyes do not seem to interfere with cell division. Because PKH molecules are distributed randomly in cellular membranes, when a cell divides, approximately half the molecules of dye migrate with each daughter cell. When the daughter cells divide, PKH dye density is again reduced by one-half. The dye dilution with each successive generation is easily seen in flow cytometric log histograms as a series of Gaussian-like distributions separated by a constant number of channels representing a factor of two shift in fluorescence intensity (see Fig. 10.7.13). If the cells are assumed to divide exactly in half, the
10 9 8 7 6 5 4 3 2 1 0
Modeling Immunofluorescence If the grounds of cell-cycle analyses are well trodden, then immunofluorescence analysis should be considered a pristine forest awaiting a new group of fearless pioneers in model building. First, many of the myths and pseudomodels that have been proposed for immunofluorescence analysis must be swept
Observed SD Constant CV Square root
0
Data Analysys Through Modeling
standard deviation associated with each generation population is constant in the log domain. A model that describes each of these generation populations can be constructed (Yamamura et al., 1995) and analyzed by a nonlinear leastsquares type of analysis. The resultant areas from each generation population can be further analyzed to yield important cell kinetic parameters such as proliferation index and nonproliferative fraction. The main advantage of using these tracking dyes rather than conventional tritiated thymidine uptake is that they can be combined with other fluorescence markers to simultaneously examine the proliferation of a heterogeneous mixture of populations. If each successive generation population is distinguishable in the histogram, the proliferation model can be extended to allow the standard deviation of each population to vary linearly with channel number. The slope of this linear relationship, called the division error index, quantifies the degree of fidelity of the population’s division during mitosis. This derived parameter may be important in studying tumor heterogeneity.
2 4 6 Aggregate peak number
8
Figure 10.7.12 Standard deviation (SD) versus aggregate peak number for chicken thymocyte nuclei aggregate populations (squares). If the staining variability of nuclei is perfectly correlated, the standard deviations should increase linearly (circles). If the staining variability is uncorrelated, it should increase as a square root function (triangles). The true relationship seems to be between these two extremes.
10.7.16 Supplement 1
Current Protocols in Cytometry
away to permit better appreciation of the potential new view of model building. Immunofluorescence analysis has traditionally been concerned with finding the most probable proportions of negative and positive staining distributions in a test histogram given the shape and position of the negative distribution (Bagwell, 1996). This problem turns out to be one of those minimal information problems where there is not quite enough information
given to yield an exact solution, but just enough to tantalize one to try. Enhanced integration Start with a test histogram that has some proportion of positive and negative staining cells for some antibody. Make the problem nontrivial by assuming there is heavy overlap between these two distributions. Further assume that the integration boundary is chosen as
Proliferation Wizard Model
Date acquired: 31-Oct-1993 Date analyzed: 22-Oct-1996 File: Proliferating.fcs
60
Number
Parent: 8.31 % at 200.00 Generation 2: 12.07 % at 180.81 Generation 3: 22.93 % at 161.62 Generation 4: 27.54 % at 142.43 Generation 5: 20.16 % at 123.24 Generation 6: 8.51 % at 104.05 Generation 7: 0.00 % at 84.86 Generation 8: 0.37 % at 65.67 Generation 9: 0.00 % at 46.48 Generation 10: 0.09 % at 27.29
40
20
Proliferation Index: 3.99 Nonproliferative Fraction: 0.08 Division Error Index: 1.20 Spacing of generations: 19.19
0 0
50
100
150
200
Background: 0.00 % Number of Cells Analyzed: 4473 Reduced Chi-Square: 0.845
Log red fluorescence
120 R1
FL2-height LOG
90
60
30
0 0
30
60
90
120
FL1-height LOG Figure 10.7.13 (A) Proliferation model analysis of cells stained with PKH tracking dye and immunofluorescence marker. The important model parameters are the proliferation index, nonproliferative fraction, and the division error index. The main advantage of the technique is that it can be combined with fluorescence markers to simultaneously examine proliferation of a heterogeneous mixture of cellular populations. Filled triangles along the x-axis indicate detected peak locations. (B) This panel demonstrates the ability to gate on other fluorescent markers such as CD3 (FL2) and CD4 (FL1).
Data Processing and Analysis
10.7.17 Current Protocols in Cytometry
Supplement 1
the boundary that separates a suitable control histogram into 95% and 5% compartments (i.e., 5% false positives). On integration of this test histogram, there will be some false positives to the right of the boundary and some false negatives to the left. Assuming that there are no false negatives, the information from the control can be used to adjust the answer for the false positive contamination. This type of enhanced integration has been around for many years and is used in many disciplines. Normalized subtraction In 1981, a seemingly unrelated method for estimating positive fraction was proposed (Bagwell, 1981b; Hoffman, 1981). In normalized subtraction the control histogram is normalized to have the same area as the test histogram within a certain match range. The normalized control histogram is then subtracted from the test histogram. The resultant residue is counted as positive staining cells.
Data Analysys Through Modeling
Dmax method In 1988, another seemingly unrelated method was proposed (Overton, 1988) involving operations with control and test cumulative distributions. A cumulative distribution is formed by successively integrating each channel back to the origin. The maximum difference between these two distributions, Dmax, was found to be an estimator for positive fraction. The myth is that all these methods were fundamentally different. In fact, they all are approximations to a single equation (Bagwell, 1996) involving just three distributions: cumulative control, test, and positive. The first two are known and the last is not. However, the positive distribution can be approximated by another equation involving three smaller distributions, two known and one not. This pattern repeats itself over and over until there are only single-channel distributions. This type of pattern is called recursion where part of the solution involves the solution, much like part of a mirror image is a mirror image when two mirrors are facing each other. Carrying out the above recursive solution with exact artificial distributions leads to an answer that is very close to being correct. However, if the distributions are slightly perturbed by shifting the control histogram, the result is a large disparity between the calculated and theoretical answers. A similar problem was encountered earlier when the Dean and Jett polynomial was increased to a higher-order
polynomial. Although the new formulation is more accurate with ideal distributions, it has little “robustness” when subjected to small irregularities as found in real-world data. The best solution to the positive fraction problem was found by making a single modest approximation to the cumulative positive distribution (Enhanced Normalized Subtraction, Bagwell, 1996). The single broadened rectangle is another example of compromising accuracy with simple, but less accurate, robust model formulations. Good models are those that work under nonideal conditions. The algorithms discussed thus far for positive fraction estimation are really not of the same caliber as those described for DNA analysis. There is no kernel or measurement linespread function. The algorithms do not attempt to describe the underlying distributions of receptors in cells or the antibody-binding process. They are more a mathematical trick than a model as defined in this chapter. However, there is a superficial similarity between these algorithms and DNA analysis models. They both involve mathematics that approximate some useful parameters; therefore, the positive fraction algorithms are probably best thought of as pseudomodels rather than true models. Currently used immunofluorescence models There have been relatively few published models for immunofluorescence analysis. Sladek and Jacobberger (1993) compared a simple model of immunofluorescence analysis to some of these pseudomodels and found their Weibull model better in providing accurate estimates for positive fraction. A Weibull function can be skewed to either the right or left which sometimes fits immunofluorescence data better than a simple Gaussian.
STEPS IN CREATING A MODEL How could a model be created for the general area of immunofluorescence? For this exercise, sophisticated tools are not necessary because the most important tool is the mind. The model should describe immunofluorescence histograms. There is no set formula for creating models. Some people will begin with careful definitions and proceed in a methodical and deliberate fashion; others will have a vague idea that slowly gains substance as it is repeatedly illuminated with mental scrutiny and data analysis. No matter what the approach is, the most important step is the first. A reasonable first step would be to attempt to leverage what is already known. It is prob-
10.7.18 Supplement 1
Current Protocols in Cytometry
ably safe to assume that there will be a linespread function similar to the one already described in DNA modeling. The line-spread function, however, will need some modification because immunofluorescence histograms are normally formed by transforming raw electronic pulses by three-or four-decade logarithmic amplifiers. Remove the line spread and look at true immunofluorescence measurements. What more information is available? If there is any significant debris or aggregates, conceivably the same theories developed for DNA analysis can be used with the appropriate log transforms. What is the interaction of the fluorescently labeled monoclonal with the target population? Is it important to account for the different number of fluorochromes on the antibody? Does this factor have some important influence on the final shape of the histogram? Probably not. If the cell has thousands of receptors, the final signal will be due to a large number of these antibodies, averaging out any individual differences. The model should assume that the antibody concentration is in saturating conditions; otherwise, an antibody concentration parameter is necessary. The distribution of autofluorescence and background staining can be incorporated by using the negative isotype control as a histogram-dependent model component. If the fraction of cells specifically binding the antibody has the same autofluorescence distribution, the control histogram can be convolved with a function representing the receptor distribution. What is a reasonable receptor distribution? It is unlikely that receptor distribution can be calculated from first principles, so the logical recourse is to search the literature. If there are no references, a reasonable guess at a suitable function will be required. Use the Weibull function (Abramowitz and Stegun, 1964) because it has the flexibility to skew right or left and has been used already in immunofluorescence modeling (Sladek and Jacobberger, 1993). The convolution formula will contain at least three parameters; two will control the shape of the Weibull function and the third will control the location of the distribution. This third parameter will be linearly related to the actual number of receptors. However, this method assumes that only one homogeneous subpopulation is interacting with the monoclonal antibody. It is entirely possible that there are two or more different reacting populations. It is possible to create a set of reacting model components to handle
each population. However, the components should not significantly overlap; otherwise, the model will have the same kind of problems Fried did with his multi-Gaussian models. With this beginning, build the model and try it with an appropriate data set where the real answers are known. First generate some test data just to make sure the system is working properly. Later, it will be necessary to use real data. If an appropriate testing cell system is not available, consider writing one of the authors of previous modeling efforts to obtain their data set. With an idea of how some biologic parameters may interact, construct a model and put it to the test. After the model is published, probably in a year or more, someone will take the ideas, refine them, and present a new formulation of the model. Later, some other investigator might improve on the model again. This process may continue until the model eventually becomes scientific dogma, much like Newton’s and then Einstein’s theories.
LITERATURE CITED Abramowitz, M. and Stegun, I. 1964. Handbook of Mathematical Functions. Dover, New York. Badish, L. 1972. The completeness of nineteenthcentury science. Isis 63:48-58. Bagwell, C.B. 1979. Theory and application of DNA histogram analysis. Ph.D. dissertation, University of Miami, Fla. Bagwell, C.B. 1981a. PARA2 Program. Coulter Cytometry, Hialeah, Fla. Bagwell, C.B. 1981b. IMMUNO program, EASY2 Software. Coulter Cytometry, Hialeah, Fla. Bagwell, C.B. 1993. Theoretical aspects of flow cytometry data analysis. In Clinical Flow Cytometry (K.D. Bauer, R.E. Duque, and T.V. Shankey, eds.) pp. 41-61. Williams & Wilkins, Baltimore. Bagwell, C.B. 1996. A journey through flow cytometric immunofluorescence analyses—finding accurate and robust algorithms that estimate positive fraction distributions. Clin. Immunol. Newsletter 16:33-37. Bagwell, C.B., Mayo, S.W., Whetstone, S.D., Hitchcox, S.A., Baker, D.R., Herbert, D.J., Weaver, D.L., Jones, M.A., and Lovett, E.J. III. 1991. DNA histogram debris theory and compensation. Cytometry 12:107-118. Beck, H.P. 1980. Evaluation of flow cytometric data of human tumors. Correction procedures for background and cell aggregations. Cell Tissue Kinet. 13:173-181. Bevington, P.R. 1969. Data Reduction and Error Analysis for the Physical Sciences. McGrawHill, New York. Data Processing and Analysis
10.7.19 Current Protocols in Cytometry
Supplement 1
Data Analysys Through Modeling
Brunsting, A., Collins, J., Kaine, F., and Bagwell, C.B. 1979. An examination of some basic assumptions in DNA histogram analysis. J. Cell Kinet. 12:123-134. Conrad, M.P. 1987. A rapid, nonparametric clustering scheme for flow cytometric data. Patt. Recog. 20:229-235. Dean, P.N. 1990. Data processing. In Flow Cytometry and Sorting, 2nd ed. (M.R. Melamed, T. Lindmo and M.L. Mendelsohn, eds.) pp. 415444. Wiley-Liss, New York. Dean, P.N. and Anderson, E. 1975. The rate of DNA synthesis during S phase by mammalian cells in vitro. In Pulse Cytophotometry (C.A.M Haenen, H.F.P. Hillen, and J.M.C. Wessels, eds.) p. 77. European Press Medikon, Ghent, Belgium. Dean, P.N., and Jett, J.H. 1974. Mathematical analysis of DNA distributions derived from flow microfluorometry. J. Cell Biol. 60:523-527. Efron, B. 1982. The jacknife, the bootstrap, and other resampling plans. In Numerical Recipes in C (W.H. Press, S.A. Teukolsky, W.T. Vetterling, and B.P. Flannery, eds.) pp. 691-692. Cambridge University Press, Cambridge. Fox, M.H. 1980. A model for the computer analysis of synchronous DNA distributions obtained by flow cytometry. Cytometry 1:71. Fried, J. 1976. Method for the quantitative evaluation of data from flow microfluorometry. Comp. Biomed. Res. 9:263-276. Fried, J. 1977. Analysis of deoxyribonucleic acid histograms from flow cytofluorometry. Estimation of the distribution of cells within S phase. J. Histochem. Cytochem. 25:942-951. Fried, J. 1979. Multi-user system for analysis of data from flow cytometry. Comput. Programs Biomed. 10:218-230. Gauss, C.F. 1855. Méthode des Moindres Carrés. Ballet-Bachelier, Paris. Gray, J.W. 1974. High-resolution chromosome analysis: Chromosome measurement and sorting by flow systems. Proc. Natl. Acad. Sci. U.S.A. 72:1231-1234. Hedley D.W. 1983. Method for analysis of cellular DNA content of paraffin-embedded pathologic material using flow cytometry. J. Histochem. Cytochem. 31:1333-1335. Hoffman, R. 1981. Simple analysis of immunofluorescence histograms. Cytometry 2:104. Horan, P.K., Melnikoff, M.J., Jensen, B.D., and Slezak, S.E. 1990. Fluorescent cell labeling for in vivo and in vitro cell tracking. Methods Cell Biol. 33:469-490.
Marquardt, D.W. 1963. An algorithm for leastsquares estimation of nonlinear parameters. J. Soc. Ind. Appl. Math. 11:431-441. Meyer, S.L. 1975. Data Analysis for Scientists and Engineers. p.388. John Wiley & Sons, New York. Overton, R.W. 1988. Modified histogram subtraction technique for analysis of flow cytometry data. Cytometry 9:619-626. Press, W.H., Teukolsky, S.A., Vetterling, W.T., and Flannery, B.P. 1992. Numerical Recipes in C. Cambridge University Press, New York. Rabinovitch, P.S. 1991. Numerical compensation for the effects of cell clumping on DNA content histograms. Cytometry Suppl. 4:27. Rabinovitch, P.S. 1993. DNA Content and Cell Cycle Analysis. In Clinical Flow Cytometry (K.D. Bauer, R.E. Duque, and T.V. Chankey, eds.) pp. 117-142. Williams & Wilkins, Baltimore. Salzman, G.C., Crowell, J.M., Goad, C.A., Hansen, K.M., Hiebert, R.D., Labauve, P.M., Martin, J.C., Ingram, M., and Mullaney, P.F. 1975. A flow-system multiangle light-scattering instrument for cell characterization. Clin. Chem. 21:1297-1304. Sato, S. 1990. Doublet correction of DNA histogram. Cytometry Suppl. 4:87. Shankey, T.V., Rabinovitch, P.S., Bagwell, B., Bauer, K.D., Duque, R.E., Hedley, D.W., Mayall, B.H., and Wheeless, L. 1993. Guidelines for implementation of clinical DNA cytometry. Cytometry 14:472-477. Sharpless, T., Tragonos, F., Darzynkewicz, Z., and Melamed, M.R. 1975. Flow cytofluorometry: Discrimination between single cells and cell aggregates by direct size measurements. Acta Cytol. 19:577-581. Sheck, L.E., Muirhead, K.A., and Horan, P.K. 1980. Evaluation of the S phase distribtuion of flow cytometric DNA histograms by autoradiography and computer algorithms. Cytometry 1:109-117. Sladek, T.L. and Jacobberger, J.W. 1993. Flow cytometric titration of retroviral expression vectors: Comparison of methods for analysis of immunofluorescence histograms derived from cells expressing low antigen levels. Cytometry 14:23-31. Yamamura, Y., Rodriguez, N., Schwartz, A., Eylar, E., Bagwell, B., and Yano, N. 1995. A new flow cytometric method for quantitative assessment of lymphocyte mitogenic potentials. Cell. Mol. Biol. 41 (Suppl. I):S121-S132.
Horan, P.K. and Slezak, S.E. 1989. Stable cell membrane labeling. Nature 340:167-168. Jett, J. 1977. Mathematical analysis of DNA histograms from asynchronous and synchronous cell populations. In Proceedings of the Third International Symposium on Pulse-Cytophotometry, Vienna, Austria (D. Lutz, ed.) pp. 93-102. European Press Medikon, Ghent, Belgium.
Contributed by C. Bruce Bagwell Verity Software House Topsham, Maine
10.7.20 Supplement 1
Current Protocols in Cytometry
Multivariate Analysis
UNIT 10.8
BACKGROUND INFORMATION This unit introduces basic concepts of multivariate analysis that are generally relevant to flow cytometry, and illustrates how to carry out some typical analyses using public domain utilities and/or commercial statistical software. Many possible goals can be addressed using multivariate methods. This unit will focus on: • determination of the degree of correlation between different variables
(parameters) • identification of the number of subpopulations present in a sample • partitioning of a sample into subpopulations, either manually or automatically. Experimenters acquiring data on more than three variables via flow cytometry face the classic challenge of multivariate statistics: how conclusions can be drawn from data that cannot be easily visualized. Both graphical and numerical methods can be used to meet this challenge. The goal of graphical methods is to represent multidimensional data in two dimensions without unacceptable loss of information. The goal of numerical methods is to draw quantitative or qualitative conclusions based on operations performed in the multidimensional space. Covariance Matrix The starting point for consideration of multivariate data is the determination of the degree of correlation between the different variables. This can be done by examining the variance-covariance matrix (also referred to simply as the covariance matrix) of a set of data. Consider a set of n observations (measurements) on p variables (parameters) forming a matrix X, where the element xij is the value of the jth variable for the ith observation. The variance-covariance matrix is a p × p symmetric matrix that describes how (on average) each variable changes with respect to each other variable (the diagonal elements are the variances of each variable). The definition of the elements of the covariance matrix (often denoted by Cov(X), Σ, or S) is n
1 s jk = ( x ji − x j )( xki − xk ) n − 1 i =1
∑
_ _ where xj and xk are the mean values of variables j and k, respectively. The determinant S is called the generalized sample variance, and is a convenient descriptor of the overall variation in a set of multivariate data. Examination of the off-diagonal elements of the covariance matrix reveals whether variation in one variable is associated with variation in another variable. For example, a positive value of s12 indicates that events with higher values of variable 1 are more likely to have higher values of variable 2. A negative value of s12 indicates that events with higher values of variable 1 are more likely to have lower values of variable 2 (and vice versa). A zero value indicates no correlation. Figure 10.8.1 shows histograms for two sets of data: one in which two variables are perfectly correlated (Fig. 10.8.1A) and one in which they are perfectly uncorrelated (Fig. 10.8.1D). The covariance matrices are shown here. 1 2
2 2
1 0 0 2
Contributed by Michael V. Boland and Robert F. Murphy Current Protocols in Cytometry (1998) 10.8.1-10.8.21 Copyright © 1998 by John Wiley & Sons, Inc.
Data Processing and Analysis
10.8.1 Supplement 3
A
B No. of events
PAR2
6 4 2 0 –2 –2
0
2
500 300 100 0
4
–2
0
PAR1
C
6
D
600
PAR1
No. of events
4
PAR2
800
400 200 0 –2
2
0
2
4
PAR1
6
6 4 2 0 –2 –4 –2
0
2
4
6
PAR2
Figure 10.8.1 Illustration of covariance and projection ambiguity. Two data sets were generated using a two-dimensional Gaussian random number generator. Both data sets had the mean values for both variables set to two, but they had different covariance matrices (see text). Contour maps are shown for the data set with (A) and without (D) correlation between the variables. The univariate histograms shown (B,C) are the same for both populations.
Note that the univariate histogram of the first variable (parameter 1 or PAR1) for the first case is the same as that for the second case (similarly for PAR2), so the two data sets cannot be distinguished based on the examination of univariate histograms only. Since the units of the elements of S are the units of the original data (and therefore can vary widely in magnitude), it can be difficult to assess whether a particular element of S is significantly different from zero. For this purpose, a measure of covariation that is somehow normalized for the variations in the variables themselves is useful. The covariance matrix can be converted into a correlation matrix (often denoted by Cor(X), ρ, or R), which provides just such a scaled measure of covariation. Its elements can be calculated using the definition rjk =
s jk s jj skk
The possible values of R range from −1 to 1. rjk2 represents the fraction of the total variation in variable j that can be accounted for solely by variation in variable k. For example, if r12 is equal to 1, the value of variable 1 for each observation can be perfectly predicted from the value of variable 2 for that observation (i.e., there exists a linear relationship between the values of variable 1 and variable 2). Thus, there is duplication of information in the data set, indicating that in this case the dimensionality of the data set can be reduced by at least 1 without losing any information. Multivariate Analysis
The nature of the experiment and the fluorescent probes used will influence the expectation that correlation will be observed between variables. In the case of a heterogeneous
10.8.2 Supplement 3
Current Protocols in Cytometry
cell population labeled with a combination of different surface markers (where both labeled and unlabeled populations are expected for each marker), no overall correlation between the amounts of the different markers may be expected (although there may be correlation in amounts of markers within a particular subpopulation). The measured amount of a surface marker may be correlated with cell size (e.g., as reflected in light scatter measurements), but this correlation would not be expected to be perfect. If initial examination of the correlation (or covariance) matrix reveals significant correlation, methods designed to reduce dimensionality may be appropriately employed. The most common starting point for dimensionality reduction, principal components analysis, is discussed in the next section. Principal Components Analysis The goal of principal components analysis is to linearly combine the original variables into a smaller set of variables that accounts for most of the variation in the data. By appropriate manipulation of the covariance matrix, a new set of variables (the principal components) can be calculated such that all of the information present in the original data is preserved. In geometric terms, the principal component transformation represents a rotation of the axes such that the observations are spread as much as possible along lines parallel to the new axes rather than along lines at an angle to the axes. As part of the transformation, the principal components are ordered such that the first principal component accounts for the largest amount of the total variation possible, the second principal component accounts for as much of the remaining variation as possible, and so on. If significant correlation exists between variables, the highest numbered principal components will have little variation. A choice may then be made to utilize only a certain number of the principal components to “capture” the original data—the elimination of some of the principal components is a form of dimensionality reduction. Figure 10.8.2 illustrates how principal components analysis can be used to generate a univariate histogram that distinguishes two populations in a bivariate data set. This point can be extrapolated to data for three or more variables: multiple populations present in multivariate data may be distinguishable in two dimensions if sufficient correlations exist. As an example, consider two data sets: one in which four populations are present at the vertices of a tetrahedron and one in which eight populations are present at the vertices of a cube. If the vertices of the tetrahedron are shared with four of those from the cube, and if the variances of all of the populations are the same, all two-dimensional projections of the data sets are indistinguishable (Fig. 10.8.3). Examination of the two-dimensional views cannot be used to determine the number of populations contained in either data set. The number can be determined by viewing the principal components (Fig. 10.8.4): a contour map of the second and third principal components reveals eight populations for the second data set (Fig. 10.8.4D) but only four for the first (Fig. 10.8.4B). Cluster Analysis The goal of cluster analysis is to divide (or partition) a multivariate data set such that subpopulations are resolved from each other and can then be separately described and quantitated. A variety of cluster analysis methods have been developed. This discussion will focus on the k-means method, which is most commonly used for large data sets such as those typically encountered in flow (and sometimes image) cytometry. In brief, k-means cluster analysis is performed by choosing seed points for an initial number of clusters (usually denoted by k, hence the name), and then iteratively assigning each observation in the data set to the cluster to which it is closest. A variety of refinements on this basic approach have been described, including methods for joining clusters to eliminate artificial “splitting” of a cluster.
Data Processing and Analysis
10.8.3 Current Protocols in Cytometry
Supplement 3
B
A 8 No. of events
800
PAR2
6 4 2 0
600 400 200
–2
0 –2 0
2 4 6 PAR1
8
–2 0 2 4 6 8 PAR1
D No. of events
2 0 –2
800
500
600 1000 400 1500
200 –4
0 –6 –4–2 0 2 4 6 PAR1
–2 0 2 4 6 8 PAR2
F
E
4 PAR2
C
0 –6 –4–2 0 2 4 6 PAR1
–4 –2
0
2
4
PAR2
Figure 10.8.2 Dimensionality reduction using principal components. Data for two subpopulations were generated using a two-dimensional Gaussian random number generator. A contour map (A) and univariate histograms (B,C) are shown. Note that although the univariate histograms indicate that two populations are present, they cannot be used to separate the populations. A contour map (D) and univariate histograms (E,F) are also shown for the two principal components. Note that the separation between the two populations is improved in the first principal component histogram, and that only one peak is seen in the histogram for the second principal component. The percent of variation accounted for by the first principal component is 90%.
Available Software An excellent selection of specialized statistical analysis packages is currently available, and current microprocessors are sufficiently fast that analyses that were once carried out using centralized computing facilities can now be done on desktop computers. There is therefore little reason why the average cytometry user cannot perform basic tasks of multivariate analysis using well-documented, thoroughly tested packages such as S-Plus, SAS, or Minitab. Each of these packages is available for one or more of the most readily available operating systems, including Microsoft Windows, MacOS, and UNIX, and commands for a given analysis package tend to be consistent across operating systems. The detailed examples below were generated on a UNIX workstation running S-Plus version 3.4 (MathSoft) and SAS version 6.11 (SAS Institute). These analyses were repeated using S-Plus version 3.3 for Microsoft Windows and SAS version 6.10 for MacOS. It is anticipated that the reader can also use the examples below to develop methods in other statistical analysis environments. If it is not known whether the computer has a particular software package installed, contact the system administrator or person otherwise in charge of the computer. Multivariate Analysis
For those without access to statistical packages, an illustration of cluster analysis using the public domain program Autoklus (Bakker Schut et al., 1993) is also provided.
10.8.4 Supplement 3
Current Protocols in Cytometry
Variable Naming Conventions Any of the S-Plus variable names used below can be changed as needed. This includes text before or after the period included in some S-Plus names. For instance, myfile. pcclust1 or myfile.princom1 can be used in place of llb08.pcclust1. However, text appearing after a $ in an S-Plus variable name cannot be changed. Such text refers to an
12 10 8 PAR2
6 4 2 0 –2 –2
0
2
4
6
8
10
12
PAR1
Figure 10.8.3 A contour plot depicting the projection of four or eight populations onto the PAR2-PAR1 plane. This projection is identical for all 2-D plots using parameters 1 through 3, assuming populations as described in the text.
Prin. comp. 2
A
B
5 0 –5 –5
0
5
–5
Prin. comp. 1
Prin. comp. 2
C
0
5
Prin. comp. 3
D
5 0 –5 –5
0
5
Prin. comp. 1
10
–5
0
5
Prin. comp. 3
Figure 10.8.4 Contour plots of the principal components of samples with four (A,B) or eight (C,D) populations, as described in the text.
Data Processing and Analysis
10.8.5 Current Protocols in Cytometry
Supplement 3
attribute of the variable, the naming of which is not up to the user. SAS data set and parameter names may also be changed, except where noted below. Note that SAS data set names can have no more than 8 characters. Example Files and Procedures Links to data files and S-plus procedures referred to below may be found on the web page http://www.stc.cmu.edu/murphylab/protocols/flow/currprotcyt.html. The examples provided in the protocols below use a common FCS file (llb08.fcs) and/or a text file generated from it (llb08.txt). Any reference to llb08 below can be replaced by the appropriate name for any data file(s). PRINCIPAL COMPONENTS ANALYSIS The starting point for this protocol is an ASCII text file containing the data to be analyzed, with one line for each observation (event) and one column for each variable (parameter). Any descriptive text at the beginning or end of the file should be deleted (e.g., a header row). If data are in listmode FCS files, see Conversion of FCS Files to ASCII Text Files. For S-Plus analysis, two alternative plotting strategies are presented in steps 7a and 7b. S-Plus 1. Use the matrix() and scan() functions to import a text file. > llb08.data <- matrix(scan(“llb08.txt”), ncol=4, byrow=T)
This particular command line will read the file llb08.txt into the variable llb08.data. When importing a different file, change the name of the data set (llb08.data), the number of parameters in the file (the number 4 in ncol=4), and the name of the input file (llb08.txt), making sure to include the complete path to the file if necessary (e.g., /users/name/filename.txt). Note that trailing tabs or spaces at the end of lines of text can cause the import of text files to fail and such tabs or spaces should therefore be removed before import. 2. Optional: Examine the covariance and/or correlation matrices. > var(llb08.data) > cor(llb08.data)
These commands will print the covariance and correlation matrices of the data set on the screen. If little correlation is observed, principal components analysis may not be useful. 3. Generate the principal components. > llb08.pc <- princomp(llb08.data)
The S-Plus princomp() function returns a data structure that contains a number of attributes, including the loadings (i.e., the contribution of each parameter to a particular principal component), and the scores (i.e., each parameter of each event multiplied by its loading). The command shown above places all of these attributes in the variable named llb08.pc. 4. Optional: Determine how much of the total variation is accounted for by each of the principal components. > llb08.pc$sdev^2/sum(llb08.pc$sdev^2) Multivariate Analysis
10.8.6 Supplement 3
Current Protocols in Cytometry
The sdev attribute contains the standard deviations of each of the principal components. By squaring these values (to obtain variances), and then dividing by the sum of the variances, the fraction of variance accounted for by each of the principal components is obtained. Principal components accounting for little of the total variance may be ignored in later processing. For the llb08 file, the fractions are 0.50, 0.30, 0.11, and 0.09. 5. Open a window for on-screen graphics. > motif() [when using S-Plus for the UNIX/X-Window environment] > win.graph() [when using S-Plus for Microsoft Windows]
Before data can be viewed in S-Plus, it is necessary to open a window to which the graphics output will be sent. See the S-Plus User’s Manual for details, or type help(motif) or help(win.graph) from the S-Plus command line. 6. Plot the principal component scores and identify clusters. > llb08.pcclust1 <- brush(llb08.pc$scores, hist=T, spin=F)
The S-Plus brush() function is useful for visualizing data of greater than two dimensions. It generates plots of all pairs of parameters and allows the user to interact with the data by selecting points of interest. First, select a brush size that is appropriate for the cluster to be isolated, and then use the brush to select that cluster of points. Finally, quit brush() and return to S-Plus. The points selected will be saved in the variable that was named in the command above (i.e., llb08.pcclust1 in this case). To view data in three dimensions, use spin=T in the brush() command. It is then possible to view and manipulate a 3-D plot in addition to the 2-D plots and histograms (if hist=T was used as well). In Figure 10.8.5, a cluster has been selected in the left side of the Comp. 3 versus Comp. 1 plot. The selected points are highlighted (note the dark regions in all plots) and included as downward bars in the histograms. For demonstration purposes, the region has been intentionally selected to include multiple populations of cells. To better illustrate the capabilities of brush(), the number of events plotted for Figure 10.8.5 was reduced from 50,176 to 1000 using the following command. > brush(llb08.pc$scores[sample(1000),], hist=T, spin=F)
In general, this is not necessary as a separate step; the command is provided in the interest of completeness. It is also important to know that when using brush(), the S-Plus command line does not accept input. To regain the ability to input commands after using brush(), press the Quit button in the graphics window. 7a. To plot the identified clusters using the original, untransformed parameter values: > brush(llb08.data[c(llb08.pcclust1),], hist=T, spin=F)
This command plots only those points that were selected when using brush() in step 6 (note the use of llb08.pcclust1 in the command). Use the 2-D plots to confirm that the chosen clusters are truly homogeneous. The ability to plot histograms of each parameter (i.e., hist=T, above) is helpful in this regard. In this case, there are actually multiple populations visible among the selected events in the plots in Figure 10.8.5. To correctly isolate a single population, one can either repeat step 6 to select another, better-isolated population, or apply brush() to the points returned from step 7a in order to isolate the events that constitute the desired population. The latter approach is implemented by this command: Data Processing and Analysis
10.8.7 Current Protocols in Cytometry
Supplement 3
Figure 10.8.5 Application of the S-Plus brush() command to the principal components of the events in the llb08 file.
> llb08.pcclust2 <- brush(llb08.data[c(llb08.pcclust1),], hist=T, spin=F)
Figure 10.8.6 shows the placement of the brush so as to select a single subpopulation from the points returned in step 7a. These points will be saved in llb08.pcclust2 and can be plotted using the following command. > brush(llb08.data[c(llb08.pcclust1[c(llb08.pcclust2)]),], hist=T, spin=F)
7b. To generate contour plots for a single cluster using all parameters: > plotall2d(llb08.data[c(llb08.pcclust1),], parnames=c(“PAR1",”PAR2","PAR3","PAR4"))
plotall2d() is a short program written in S-Plus and can be downloaded from http://www.stc.cmu.edu/murphylab/protocols/flow/currprotcyt.html. The plots generated by this command are shown in Figure 10.8.7.
Multivariate Analysis
10.8.8 Supplement 3
Current Protocols in Cytometry
Figure 10.8.6 A second use of the brush() function allows better isolation of a single population of cells.
SAS 1. The following set of commands will read a four-parameter text file into an SAS data set. Enter the commands into the Program Editor window and then select Submit from the Locals menu. data llb08 ; infile ’llb08.txt’ EXPANDTABS; input PAR1 PAR2 PAR3 PAR4 ; run ;
When importing a different file, change the name of the data set (llb08), the number and naming of the parameters (PAR1 to PAR4), and the name of the input file (llb08.txt), making sure to include the complete path to the file if necessary (e.g., /users/name/filename.txt). Look at the SAS LOG window to ascertain the success or failure of the program. 2. Generate the principal components. As before, enter the commands into the Program Editor window and then select Submit from the Locals menu. proc princomp data=llb08 out=llb08pc ; var PAR1 PAR2 PAR3 PAR4 ; run ;
This set of commands will generate the principal components for the data contained in the llb08 database, and will save the results in a temporary database, llb08pc.
Data Processing and Analysis
10.8.9 Current Protocols in Cytometry
Supplement 3
PAR3
PAR2
150 100 50 0 0
50
100
150
250 200 150 100 50 0
200
0
50
150 100 50 0 0
50
100
200
150
250 200 150 100 50 0
200
0
50
100
150
PAR2
PAR4
PAR1
PAR4
150
PAR1
PAR3
PAR4
PAR1
100
150 100
150 100 50
50
0
0 0
50
100 PAR2
150
0
50
100
150
200
PAR3
Figure 10.8.7 S-Plus contour plots provide another method of visualizing clusters. In this case, all possible pairs of parameters were used to plot a single cluster.
Multivariate Analysis
Figure 10.8.8 An SAS-generated plot of the third versus the second principal component for the llb08 data.
10.8.10 Supplement 3
Current Protocols in Cytometry
3. Plot the principal components. Make sure to submit these commands. symbol1 c=default, i=none, v=point, cv=default ; proc gplot data=llb08pc ; plot (PRIN1-PRIN4) * (PRIN1-PRIN4) ; run ;
This set of commands will generate plots using all pairs of principal components. Note that many of these plots contain no information (e.g., PRIN1 versus PRIN1) or are redundant (e.g., PRIN2 versus PRIN1 and PRIN1 versus PRIN2). Figure 10.8.8 is a plot generated by SAS of the third principal component versus the second. 4. Identify clusters in the principal component plots generated in step 3. By viewing the plots of the principal components, one can identify clusters that may or may not have been present in the untransformed data. Visual identification of these clusters is helpful for setting the initial conditions of any cluster analysis. K-MEANS CLUSTER ANALYSIS The starting point for this protocol is an ASCII text file containing the data to be analyzed, with one line for each observation (event) and one column for each variable (parameter). Any descriptive text at the beginning or end of the file should be deleted (e.g., a header row). If data are in listmode FCS files, see Conversion of FCS Files to ASCII Text Files. For S-Plus analysis, five alternative plotting strategies are presented in steps 6a through 6e. S-Plus 1. Use the matrix() and scan() functions to import a text file. > llb08.data <- matrix(scan(“llb08.txt”), ncol=4, byrow=T)
This particular command line will read the file llb08.txt into the variable llb08.data. When importing a different file, change the name of the data set (llb08.data), the number of parameters in the file (the number 4 in ncol=4), and the name of the input file (llb08.txt), making sure to include the complete path to the file if necessary (e.g., /users/name/filename.txt). Note that trailing tabs or spaces at the end of lines of text can cause the import of text files to fail and such tabs or spaces should therefore be removed before import. 2. Choose a number of clusters, k, to be found within the data. The number of clusters to generate is one of the most important choices in k-means cluster analysis, although there are no specific criteria that can be reliably applied to choosing an initial number of clusters. Remember that it is possible to reprocess the data after visualization of the initial clusters, so a correct initial choice is not necessary. 3. Generate the k clusters using the number chosen above. > llb08.09means <- kmeans(llb08.data, llb08.data[1:9,])
In this example, k = 9 was chosen. Using the first k events from the data is an acceptable method of setting the initial centers of the k clusters (i.e., llb08.data[1:9,], above). The S-Plus kmeans() function returns a data structure including a vector indicating which cluster each of the events was placed in (llb08.09means$cluster), the centers of each cluster (llb08.09means$centers), the within-cluster sum of squared error (llb08.09means$withinss), and the number of events in each cluster (llb08.09means$size).
Data Processing and Analysis
10.8.11 Current Protocols in Cytometry
Supplement 3
Value of cluster center
200
150
100
50
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Parameter Figure 10.8.9 A line plot can be useful in characterizing the means of each parameter for the clusters returned by k-means analysis.
4. Identify the clusters that constitute the largest fractions of the total population. > llb08.09means$size
Selection of the largest clusters for further analysis is useful when studying the characteristics of events that are the most frequent. It is also valid to choose smaller clusters to analyze minor populations. 5. Open a window for on-screen graphics. > motif() [when using S-Plus for the UNIX/X-Window environment] > win.graph() [when using S-Plus for Microsoft Windows]
Before data can be viewed in S-Plus, it is necessary to open a window to which the graphics output will be sent. See the S-Plus User’s Manual for details, or type help(motif) or help(win.graph) from the S-Plus command line. 6a. To plot the value of the center of each cluster versus the parameter number: > plot(rep(c(1:number_of_parameters),number_of_clusters_ to_plot), llb08.09means$centers[c(number_of_clusters_ to_plot), ],type="n") > for (i in 1:number_of_clusters_to_plot) { lines(c(1: number_of_parameters),llb08.09means$centers[i,],col=i) }
The italicized text in this command should be replaced with appropriate numbers from the data set being analyzed. For instance, to plot clusters 1, 2, 5, 6, and 9 using all four parameters from the llb08 data, use the following commands:
Multivariate Analysis
> plot(rep(c(1:4),5), llb08.09means$centers[c(1,2,5,6,9),], type="n") > for (i in 1:5) { lines(c(1:4),llb08.09means$centers[i,],col=i) }
10.8.12 Supplement 3
Current Protocols in Cytometry
PAR3
all events 250 200 150 100 50 0
0
50
100
150
200
250
PAR2
PAR3
cluster 1
cluster 5
250 200 150 100 50 0
0
50
100
150
200
250
0
50
100
150
200
250
200
250
PAR2
PAR3
cluster 6
cluster 9
250 200 150 100 50 0
0
50
100
150
200
250
0
50
100
150
PAR2
Figure 10.8.10 Visualizing clusters one at a time using only two parameters can be useful in interpreting the results of cluster analysis.
This sequence of commands will generate a plot like the one in Figure 10.8.9. This kind of analysis is useful for visualizing the character of each cluster (e.g., there is a population that is low in parameter 1, high in parameter 2, low in parameter 3, and of moderate value in parameter 4). If the colors of the lines generated in this step are not acceptable, it may be necessary to edit the color scheme used by S-Plus. If using S-Plus for UNIX/X-Window, this can be done by selecting Color Scheme under the Options menu in the S-Plus graphics window. Once there, create a new scheme and set the colors for Lines, Text, Polygons, and Background. The examples in this protocol use white for Background, and black, grey, brown, red, orange, yellow, green, blue, cyan, and magenta for Lines, Text, and Polygons. Give the color scheme a name and select Save and then Close. For similar problems while running S-Plus for Microsoft Windows, select Options from the main menu, then Colors. The option titled Windows Standard has 12 colors defined for Line, Fill, and Text and will work with the examples provided here. 6b. To choose two parameters and generate contour plots for some or all clusters: > plotkmeans(llb08.data[,c(2,3)], llb08.09means, c(“PAR2",”PAR3"), which=c(1,5,6,9))
An example of this analysis is shown in Figure 10.8.10, where the four clusters containing the greatest number of events were chosen to plot using parameter 2 and
Data Processing and Analysis
10.8.13 Current Protocols in Cytometry
Supplement 3
Figure 10.8.11 Plotting a single cluster using all parameters allows for complete characterization of the corresponding population of cells.
parameter 3. The short S-Plus program that was used to generate this figure, plotkmeans(), can be downloaded from http://www.stc.cmu.edu/murphylab/protocols/flow/currprotcyt.html. Note that clusters 5 and 9 contain more than a single population of events. This indicates that the number of initial clusters could be increased in an attempt to separate these smaller clusters from the larger ones. This kind of visualization can be repeated for other meaningful pairs of parameters (i.e., changing one or both of PAR2 and PAR3). 6c. To generate dot plots for a single cluster using all parameters: > brush(llb08.data[llb08.09means$cluster[] == 1,], hist=T)
The ability of the brush() function to visualize multidimensional data is again of use. As shown in Figure 10.8.11, the output of brush() makes it possible to verify the homogeneity of a cluster that is output from kmeans(). The cluster visualized in Figure 10.8.11, for instance, shows no evidence of containing another, smaller population. 6d. To generate contour plots for a single cluster using all parameters: > plotall2d(llb08.data[llb08.09means$cluster[] == 1,], parnames=c(“PAR1",”PAR2","PAR3","PAR4"))
Select the cluster to plot by changing the number 1 in llb08.09means$cluster[] == 1. The results of this S-Plus statement are shown in Figure 10.8.12. Although they use the same information, the contour plots represent an alternative to the dot plots implemented in brush() (see step 6c). As before, the S-Plus function plotall2d() can be obtained from http://www.stc.cmu.edu/murphylab/protocols/flow/currprotcyt.html. Multivariate Analysis
10.8.14 Supplement 3
Current Protocols in Cytometry
PAR3
PAR2
100
120 100 80 60 40 20
80 60 40 20
40
60
80
100
120
140
40
60
80
PAR1
100
120
140
PAR1
PAR3
PAR4
100 140 120 100 80
80 60 40 20
40
60
80
100
120
20
140
40
60
140 120 100 80 20
40
60
80
80
100
120
PAR2 PAR4
PAR4
PAR1
100 120
PAR2
140 120 100 80 20
40
60
80
100
PAR3
Figure 10.8.12 Contour plots of individual clusters can be useful in characterizing each population of cells.
6e. To construct dot plots of all events using all parameters and identify clusters via plot color: > dotplotall(llb08.data, llb08.09means, c(“PAR1",”PAR2","PAR3","PAR4"), c(1,2,3,4,5,6,7,8,9)
dotplotall() is another S-Plus program that can be downloaded from http://www.stc. cmu.edu/murphylab/protocols/flow/currprotcyt.html. The output from this command is a series of plots using all pairs of parameters, and showing each cluster as a single color in all of these plots. This means of visualizing the data is similar to that used in Autoklus (see below). If the color output from this command is not satisfactory, refer to the discussion under step 6a. SAS 1. The following set of commands will read a four-parameter text file into an SAS data set. Enter the commands into the Program Editor window and then select Submit from the Locals menu. data llb08 ; infile ’llb08.txt’ EXPANDTABS; input PAR1 PAR2 PAR3 PAR4 ; run ;
When importing a different file, change the name of the data set (llb08), the number and naming of the parameters (PAR1 to PAR4), and the name of the input file (llb08.txt), making sure to include the complete path to the file if necessary (e.g., /users/name/filename.txt). Look at the SAS LOG window to ascertain the success or failure of the program.
Data Processing and Analysis
10.8.15 Current Protocols in Cytometry
Supplement 3
Figure 10.8.13 An example of SAS output. The program will depict each cluster as a unique color. The color version of this figure can be seen on the CD version of Current Protocols in Cytometry.
2. Choose a number of clusters, k, to be found within the data. The number of clusters to generate is one of the most important choices in k-means cluster analysis, although there are no specific criteria that can be reliably applied to choosing an initial number of clusters. Remember that it is possible to reprocess the data after visualization of the initial clusters, so a correct initial choice is not necessary. 3. Generate the k clusters. proc fastclus data=llb08 out=llb0809m maxc=9 ; run ;
The SAS fastclus() function implements the k-means algorithm for clustering. It is the maxc option that specifies how many clusters are to be found in the data (9 in this case). 4. Plot the clusters using two parameters at a time. proc gplot ; symbol2 v=point ; plot PAR3 * PAR2 = cluster ; run ;
Each cluster will be plotted as a different color on a simple X-Y plot, as in Figure 10.8.13. Some populations might overlap in some dimensions and not in others, so it is important to view all combinations of parameters (i.e., change PAR3 and PAR2, above).
Multivariate Analysis
Autoklus 1. There is no need to generate a text version of the FCS file as Autoklus can read data directly from the FCS format.
10.8.16 Supplement 3
Current Protocols in Cytometry
Figure 10.8.14 Plots generated by Autoklus using all specified parameters. The program will depict each cluster as a unique color. The color version of this figure can be seen on the CD version of Current Protocols in Cytometry.
2. Go to the directory containing Autoklus and execute the program. C:\>autoklus
3. Press Enter after viewing the startup screen. 4. When presented with the menu of disks to use for working memory, select one with as much free space as possible. 5. Select Readdata from the main menu and then Dir *.*. This allows navigation to a directory containing FCS file(s). The double dot (..) is used to move to the parent of the current directory (i.e., move up one level in the directory structure). Once the file has been located, select it and press return. If the file is read successfully, proceed to the next step. During preparation of this protocol, it was noted that Autoklus did not read an FCS file with more than ∼30,000 events, whereas the documentation claims that the program can accommodate >64,000 events. If the program stops working at this point and the file is more then 30,000 events long, try reducing its size and then reading it again. The llb08.fcs file was shortened to 30,208 events for use in this part of the protocol (this particular shortened file, llb08s30.fcs, is also available on the web at http://www.stc.cmu.edu/murphylab/protocols/flow/currprotcyt.html). 6. After the data have been imported and plotted by Autoklus, select Utility from the main menu and then Changepar. Alter the plotting parameters to suit the data. After all parameters have been changed for the six plots on the screen, answer “n” when asked about making additional changes. If it is necessary to generate additional plots, Autoklus provides another “page” of plots that can be configured. Select Page from the Utility menu to move to the other page of plots. Select Page again to go back to the original plots. Figure 10.8.14 shows the parameters to use for each plot when using the llb08 example file.
Data Processing and Analysis
10.8.17 Current Protocols in Cytometry
Supplement 3
7. Return to the main menu by selecting Main. Now select Cluster and then Cluster again to begin the clustering process. 8. Indicate the number of events to use in generating the clusters. HOW MANY CELLS TO CLUSTER (MAX:30208) ? 30208
The program will sometimes stop working if the default number of events is chosen (i.e., by pressing Return). It is therefore best to actually type in the desired number of events. 9. Enter the number of large clusters to be found. HOW MANY LARGE CLUSTERS ? 12
This entry selects the number of clusters that should remain once Autoklus is done with the clustering and subsequent joining of clusters that are sufficiently close to one another. 10. Enter the number of seed points. HOW MANY SEED POINTS (MAX 255) ? 20
This number tells Autoklus how many cluster centers to start with. This number should be larger than the number of large clusters specified in step 9, because Autoklus will join closely adjacent clusters to reduce the number of final clusters. 11. Unless there is a reason to do otherwise, accept the default values for the next three options: the percent of values to use in preclustering, the termination condition for the clustering algorithm, and whether Autoklus should show the cluster process. HOW MANY PERCENT FOR PRECLUSTERING (DEFAULT 100) ? <ENTER> AT HOW MANY PERCENT CHANGES SHOULD I STOP (DEFAULT 5) ? <ENTER> SHOW THE CLUSTER PROCESS (N/Y) ? <ENTER>
Autoklus will now proceed to cluster the data. It will provide some indication of its progress as it works. Once it is done, there should be a set of plots like those shown in Figure 10.8.14. It is possible to identify all of the major clusters described in Table 2 of Murphy (1985) in the clusters generated by Autoklus using the steps above. 12. To better view individual clusters, select Showclus (found under Main and then Cluster). To see all clusters separately, press “y” when prompted. Autoklus will plot one cluster at a time. If “n” is pressed, Autoklus will plot all clusters and then flash them one at a time. Selecting “n” at this point allows for better comparison among clusters. When plotting one cluster at a time, it is difficult to assess the relative positions of the clusters. Autoklus will also display the fraction of events that are contained in each cluster. Autoklus provides several other plotting modes that may be useful in the particular application being used. They can be found by selecting Plot under Main and then Utility.
Multivariate Analysis
10.8.18 Supplement 3
Current Protocols in Cytometry
CONVERSION OF FCS FILES TO ASCII TEXT FILES Data collected by most flow cytometers are stored in files which adhere to the Flow Cytometry Standard (FCS) file format (Murphy and Chused, 1984; Dean et al., 1990). Most general purpose statistical packages (with the exception of SAS JMP) cannot read these FCS files. Data must therefore be extracted from FCS files and placed in simple ASCII text files. Note that most associated descriptive information (annotation) contained in an FCS file (e.g., sample source, reagents used, and instrument settings) is not preserved in the resulting text file, so the connection between the original file and the text file must be maintained via filename only. Note also that text files are usually much larger than the FCS files from which they are derived (as FCS files normally store parameter values in binary), so large amounts of temporary disk space may be needed to complete analysis of even a few samples. Data acquisition and analysis packages supplied with commercial flow cytometers may provide the ability to export data as formatted text. If this is not the case, public domain software or shareware for DOS and MacOS can be used to convert FCS files to text. It is sometimes necessary to perform minor edits on the resulting text files in order to read them using statistical packages. A text editor capable of handling large text files is required for this step (e.g., edit in DOS, vi or emacs in UNIX, or a commercial word processor for MacOS or Microsoft Windows). Conversion Using MFI in DOS MFI (written by Eric Martz) is a general purpose, public domain flow cytometric data analysis package that runs under DOS (and in DOS windows under various flavors of the Windows operating system). It is available from Eric Martz’s catalog of free flow cytometry software at http://marlin.bio.umass.edu/mcbfacs/flowcat.html#mfi and can also be found on the Purdue Cytometry CD ROM (http://www.cyto.purdue.edu/flowcyt/ cdseries.htm). Use the following procedure to generate text files from FCS files using MFI version 3.4J2. 1. Change the default directory to the one containing the files to be converted. 2. Run MFI (its directory must be in the path). 3. Tag (select) files to be converted using T and U and then press Enter. 4. Press Esc to skip dot plots. 5. Press C to modify the configuration. 6. Press 9 to modify the ASCII file creation configuration. 7. Press 2 and then Y to turn ASCII file writing on, then press any key. 8. Press Enter to return to main configuration menu. 9. Press G to toggle graphics off. 10. Press Enter to exit configuration editing. 11. Press Enter to proceed with processing. 12. Press Space when prompted to continue between screens of information and between files. 13. Press Q to exit MFI. MFI will create files with the same name as the input file but with a .ALS extension.
Data Processing and Analysis
10.8.19 Current Protocols in Cytometry
Supplement 3
Conversion Using FCS Assistant in MacOS FCS Assistant (written by Ray Hicks) is a shareware utility for converting listmode FCS files to ASCII; it can output either listmode data or ungated histograms. It is available over the Internet at http://facsmac.med.cam.ac.uk/FCSA.html and can also be found on the Purdue Cytometry CD ROM. The following applies to version 1.2.9 beta. Use of FCS Assistant is straightforward. Select Open from the File menu and choose the FCS file using the dialog box. Note that FCS Assistant requires that the $PnN keywords and the $BYTEORD keywords be present in the file. Once the file has been read, verify that the file description information is correct. Entire parameters can be deleted, if desired (select Remove Parameter from the Modify menu). To write the listmode data to a text file, select Export and then Raw Tab Text from the File menu and specify the output filename. Note that FCS Assistant puts an extra tab at the end of the first line of files it creates, causing some data import routines to believe that the data set contains one more column than it actually does. This extra tab can be deleted by editing the file with a text editor or word processing program. Note also that FCS Assistant (at least as of version 1.2.9 beta) cannot read 8-bit FCS files (files where $PnB is equal to 8). CONVERSION OF ASCII TEXT FILES TO FCS FILES After transformation or partitioning of flow cytometric data sets, it is often desirable to utilize the interactive and publication graphics capabilities found in many commercial flow cytometry packages. To do so, ASCII text files created using general purpose statistical packages must often be converted to FCS format. No public domain utilities for performing this task under MacOS exist as of this writing, but two options are available under DOS. Conversion Using A2FCS in DOS A2FCS is a public domain utility that runs under DOS (and in DOS windows under various flavors of the Windows operating system). It is part of the MFI/FCS Verification Suite available at http://marlin.bio.umass.edu/mcbfacs/flowcat.html#mfiverif and can also be found on the Purdue Cytometry CD ROM. Conversion Using TEXT2FCS in DOS TEXT2FCS is a public domain utility that runs under DOS (and in DOS windows under various flavors of the Windows operating system). It is available at http://marlin.bio. umass.edu/mcbfacs/flowcat.html#text2fcs and can also be found on the Purdue Cytometry CD ROM. IMPORTING S-PLUS DATA OBJECTS 1. Download the S-Plus data object from http://www.stc.cmu.edu/murphylab/protocols/ flow/currprotcyt.html.
Multivariate Analysis
Use a web browser (e.g., Netscape Navigator or Microsoft Internet Explorer) to find the desired object to download and select the appropriate link in the page above. A cryptic looking text file should be displayed. Find the Save As option in the browser and then save the file as text.
10.8.20 Supplement 3
Current Protocols in Cytometry
2. At the S-Plus prompt, type the following: > data.restore(“file_just_saved”)
This will create the desired S-Plus object in the working directory. LITERATURE CITED Bakker Schut, T.C., De Grooth, B.G., and Greve, J. 1993. Cluster analysis of flow cytometric list mode data on a personal computer. Cytometry 14:649-659. Dean, P.N., Bagwell, C.B., Lindmo, T., Murphy, R.F., and Salzman, G.C. 1990. Data file standard for flow cytometry. Cytometry 11:323-332. Murphy, R.F. 1985. Automated identification of subpopulations in flow cytometric list mode data using cluster analysis. Cytometry 6:302-309. Murphy, R.F. and Chused, T.M. 1984. A proposal for a flow cytometric data file standard. Cytometry 5:553-555.
Contributed by Michael V. Boland and Robert F. Murphy Carnegie Mellon University Pittsburgh, Pennsylvania
Data Processing and Analysis
10.8.21 Current Protocols in Cytometry
Supplement 3
Detection and Location of Hybridization Domains on Chromosomes by Image Cytometry
UNIT 10.9
Some fluorescence in situ hybridization (FISH) applications require the detection of hybridization domains on intact metaphase chromosomes so that the domains can be measured, located, or compared. Manual detection and analysis of these domains can be subjective and may depend on the judgment of the individual operator. This unit describes the automated computer algorithms that have been developed to detect and locate hybridization domains computationally. These algorithms reduce the tedium of manual detection and improve objectivity and repeatability. USING CUSTOM ALGORITHMS TO DETECT AND LOCALIZE HYBRIDIZATION DOMAINS IN METAPHASE CHROMOSOMES Locating hybridization domains along metaphase chromosomes involves using a series of algorithms to isolate the chromosome of interest, calculate its length, identify regions of interest, and assign a position to the domains relative to the end of the chromosome. First, a chromosome segmentation algorithm separates the chromosome of interest from its background. The method computes the average intensity along the edge of the chromosome and applies that value to the chromosome image as a threshold. Next, algorithms are used to determine the medial axis of the chromosome, a line that divides the chromosome into two equal halves. The medial axis is drawn by computing the chromosomal skeleton and extending the ends of the skeleton to intersect with the tips of the chromosome using the center of mass (COM) extension technique. A dynamic regional thresholding (DRT) method allows the reliable identification of hybridization domains, even when the domains vary substantially in size and their total fluorescence varies in intensity. DRT unites multiple thresholding and morphological techniques, permitting robust detection of DNA probe hybridization domains. Figure 10.9.1 illustrates the progression of images generated by the chromosome segmentation, hybridization domain detection, and medial axis determination methods. Finally, the locations of the hybridization domains within the chromosome are determined by calculating the fractional length location of each probe, or its FLpter—the distance of the probe from the tip of the short arm relative to the total length of the medial axis of the chromosome.
BASIC PROTOCOL
Materials DNA sample counterstained with 4′,6-diamidino-2-phenylindole (DAPI) or propidium iodide (PI) and hybridized with probes of interest (UNITS 8.1-8.3) Fluorescence microscope with filter wheel containing single-bandpass filters at the excitation source and corresponding triple- (or double-) bandpass emission filters High-resolution CCD or other digital camera Computer for controlling image acquisition and performing image analysis Software for image processing and analysis (e.g., Khoros from http://www. khoral.com, NIH image from http://rsb.info.nih.gov/NIH-image/download.html, or SCIL-Image from http://www.tno.nl/instit/tpd/product/scil/; or see references in protocol steps for instructions on generating the appropriate algorithms) Collect and prepare image 1. Mount sample on fluorescence microscope and collect multicolor digital images, one showing counterstained DNA and one or two others showing the hybridization signals. Data Processing and Analysis Contributed by Laura Mascio Current Protocols in Cytometry (1998) 10.9.1-10.9.11 Copyright © 1998 by John Wiley & Sons, Inc.
10.9.1 Supplement 3
Use single-wavelength excitation filters and multi-bandpass emission filters during image acquisition. This will allow the hybridization probe and DNA counterstain images to be collected in near-perfect registration relative to each other.
2. Cut out the region of interest in each of the associated images, generating a new image that contains a single chromosome (see Fig. 10.9.1A). Isolate and segment chromosome 3a. Prepare a mask: Apply an isodata threshold to the DNA counterstain image to evenly separate the object and background intensity populations. The resulting binary image is called a “mask.” This mask (Ridler and Calvard, 1978) will later be combined with the smoothed Kuwahara image generated in step 3b (see Fig. 10.9.2). To computationally isolate the chromosome from its surroundings in the DNA counterstain image, it is necessary to understand the intensity populations in the image. Most hybridization images acquired with a cooled CCD camera have very low, nonvarying background signals, and do not require an initial background subtraction. When this is not the case,
A
B
C
D
E
F
Location of Hybridization Domains on Chromosomes by Image Cytometry
Figure 10.9.1 Progression of steps toward FLpter determination. (A) The DNA counterstain image. (B) The FITC emission image. (C) DRT segmentation of FITC hybridization domains (spots inside white patches). (D) Texas red (TR) emission image. (E) DRT segmentation of TR hybridization domains. (F) Superposition of hybridization domain segmentations (from C and E) onto the chromosome, along with the results from medial axis determination. Figure (from Mascio et al., 1995) used by permission of the University of California, Lawrence Livermore National Laboratory, and the U.S. Department of Energy.
10.9.2 Supplement 3
Current Protocols in Cytometry
Number of pixels
initial position of marker background pixels
final position of marker (threshold value)
object pixels
0
128
256
Pixel intensity = final intensity value of the COM for each population
Figure 10.9.2 Histogram of a sample image. With the isodata threshold method, a marker starts at the mid-intensity level and then is adjusted until it converges to a position that is midway between the centers of mass of the two parts into which it divides the histogram. Figure reprinted with permission from Mascio et al. (1995).
background subtraction techniques should be used to correct for variations in the background before proceeding (see UNIT 10.5).
3b. Prepare a Kuwahara image: Apply an edge-preserving, smoothing Kuwahara filter (Kuwahara et al., 1976) to the DNA counterstain image. Use a square filter, size 1 + 8X on a side. Save an additional copy of this image, because it will be needed for two different computations (as shown in Fig. 10.9.3). X is a multiplier that will vary with the image resolution. X = 1 works well for images acquired with a magnification factor of 100 and a numerical aperture of 1.3.
4. Smooth the Kuwahara image from step 3b with two iterations of a uniform (average) square filter, size 1 + 4X (where X is the same multiplier used in step 3b). Find the intersection of this image and the segmentation mask from step 3a using a logical “and” operator. Apply an isodata threshold to the resulting image. This usually generates an isodata/contour image with a boundary that is larger than the chromosome.
5. Compute the contour of this binary image, then dilate it with 2X iterations and erode that result with 1X iterations (X is the same multiplier used in steps 3b and 4). This step restricts the results to those pixels closest to the chromosomal border (Fig. 10.9.3C).
6. Calculate the zero-crossings of the second derivative of the Kuwahara image from step 3b to produce fine “threads” that follow the intensity gradients throughout the image. This can be accomplished by computing the dynamic gist (DYG) using filter size 1 + 4X, thresholding this result at zero, and inverting. Computing the DYG is described in detail by Verbeek et al. (1988). The result of this gradient image step is illustrated in Figure 10.9.3D.
7. Compute the intersection of the isodata/contour image (from step 4) and the gradient image (from step 6) using a logical “and” operator. Compute the contour of the
Data Processing and Analysis
10.9.3 Current Protocols in Cytometry
Supplement 3
A apply edge-preserving, smoothing Kuwahara filter to original image (step 3b)
B further smooth image and compute isodata threshold (step 4)
D calculate zero crossings of second derivative (step 6)
C contour and dilate threshold result for rough border (step 5)
E compute logical "and" (step 7)
Figure 10.9.3 Segmenting the chromosome. (A) An edge-preserving, smoothing filter is applied to the original image and then two thresholds are applied independently. (B, C) The isodata method yields a rough estimate for the location of the border. (D) The second-derivative method yields threads that follow intensity gradients. (E) A union of these two produces only the threads that lie within the rough border. The result is used as a mask to extract corresponding intensity values from the original image. The average of these intensities is then used as a single-value threshold on the original DNA counterstain image. Figure reprinted with permission from Mascio et al. (1995).
resulting image. Intersect this contour with the estimate of the rough border from step 5. The remaining threads fall within the area of the object border and give a precise, though often not continuous, estimate for the edge of the object (Fig. 10.9.3E).
Location of Hybridization Domains on Chromosomes by Image Cytometry
8. Measure the average intensity value in the original image that is masked by the final intersection result of step 7. Use this average value to threshold the DNA counterstain image. This can be accomplished by computing the intersection of the contour result from step 7 and the original image (from step 2). Compute a histogram for this result, and then set the number of pixels in the zero bin of the histogram to zero. The average of the remaining values in the histogram is the grayscale average of the contour. Apply this threshold to a slightly smoothed version of the DNA counterstain image. The chromosome should now be segmented or isolated from the background.
10.9.4 Supplement 3
Current Protocols in Cytometry
Compute the medial axis 9. Using the binary image generated in step 8, compute the skeleton of the segmented chromosome. The skeleton (Verwer, 1988) is essentially a medial axis but it stops short of the extremities of the chromosome (see Fig. 10.9.4A).
10. Define the area of the chromosome that has no skeleton by drawing lines that are orthogonal to the tip of the skeleton until they intersect the chromosome boundaries (Fig. 10.9.4B). 11. Compute the geometric center for the area defined in step 10 (Fig. 10.9.4C). 12. Extend the skeleton by drawing a straight line from the skeletal endpoint to the geometric center (Fig. 10.9.4D). 13. Repeat steps 10, 11, and 12 on the resulting extension twice (Fig. 10.9.4E–I). Finally, extend the backbone through the center point until it reaches the tip of the chromosome (Fig. 10.9.4J). The tip of the chromosome is defined here as the point where the extended skeleton intersects the border of the convex hull of the chromosome. Determine medial axis length 14. Use a Freeman chain code representation of the medial axis and determine its length using the Monte Carlo derived equation: length = 0.980No + 1.406Nd − 0.091Nc, where N is the number of pixels in a given class; o denotes orthogonal direction, or
A
B
C
D
E
F
G
H
I
J
Figure 10.9.4 Extending the chromosomal skeleton (A) to form the medial axis (J). (B) Form a line orthogonal to the end of the skeleton. (C) Determine the geometric center of the area with no backbone. (D) Extend skeleton to the geometric center. (E–I) Repeat steps (B) through (D) twice. (I) Complete a line across the tip of the chromosome. Extend the skeleton to this line. Figure reprinted with permission from Mascio et al. (1995). Data Processing and Analysis
10.9.5 Current Protocols in Cytometry
Supplement 3
“even” Freeman code; d denotes diagonal direction, or “odd” Freeman code; and c denotes a chess knight’s move (consecutive orthogonal-diagonal or vice-versa pairs). For a more detailed description of the Freeman chain code representation and the derivation of the Monte Carlo equation, see Freeman (1970), Vossepoel and Smeulders (1982), and Dorst and Smeulders (1987).
A first region second region third region
B
opening x2
C opening x1
D
Location of Hybridization Domains on Chromosomes by Image Cytometry
Figure 10.9.5 Pictorial summary of the four major steps in the DRT algorithm. (A) The original image is divided into three regions of interest. (B) The algorithm computes a conservative threshold value (one that is likely to include all object pixels or blobs). This is followed by a step to remove spurious signals left by the conservative threshold. (C) The algorithm computes another threshold for each region, this time using a specific and strict threshold value that is likely to include only object pixels. This step leaves objects with a diameter greater than two pixels (pinnacles) and eliminates any remaining spurious noise pixels. (D) The algorithm combines the results of steps (B) and (C) with a logical “or” operator, which adds the images together to yield the hybridization domains of interest. Figure (from Mascio et al., 1995) used by permission of the University of California, Lawrence Livermore National Laboratory, and the U.S. Department of Energy.
10.9.6 Supplement 3
Current Protocols in Cytometry
Detect hybridization domains Define regions of interest 15. Apply a small uniform smoothing filter (3 × 3) to the probe image from step 2. To detect hybridization domains using the DRT algorithm (Fig. 10.9.5), first isolate regions of interest by locating areas that are higher in intensity than their immediate surroundings (Fig. 10.9.6; Verbeek et al., 1988).
16. Apply a large minimum (grayscale erosion) filter to the image from step 15. Use a filter size that is large enough to completely erode objects away. A one-dimensional example of this effect is sketched on the right side of Figure 10.9.6A.
17. Apply a small, approximately round, maximum (grayscale dilation) filter to the image from step 15 to ensure that the regions of interest fully encompass the hybridization domains. The “round” structuring element increases the probability that separate regions will form around closely spaced hybridization domains. A one-dimensional example of this effect is sketched on the left side of Figure 10.9.6A.
18. Subtract the image generated in step 16 from the image generated in step 17. A one-dimensional example of this effect is sketched in Figure 10.9.6B.
19. Apply a “noise threshold” (typically 10 to 50 of 256 gray levels) to the image generated in step 18. This threshold determines the size of the region and must be low enough to ensure that each domain is fully enclosed by the region. A one-dimensional example of the result of this threshold, and the subsequent masking which will be described in step 21, are sketched in Figure 10.9.6C.
20. Use a labeling algorithm to uniquely identify each individual region in the images generated in step 19. 21. Isolate each irregularly shaped region identified in step 20 by masking the original hybridization image. These regions of relatively high intensity (Fig. 10.9.5A) are now ready to be further delimited by a combination of “blob” and “pinnacle” detection (Fig. 10.9.5B–D).
Detect blobs 22. Apply a restricted isodata algorithm to each region resulting from step 21. This algorithm is a modification of the isodata algorithm. To implement this function, first generate a histogram of the image. Then, if there are more than ten pixels with an intensity value of zero in the histogram of the region, alter the histogram so that the zero-intensity bin contains ten pixels. Then proceed with the standard isodata algorithm. The restricted isodata compensates for the arbitrary threshold used in the medial axis determination (step 19). It does not matter if too much background was included in the irregularly shaped region of interest; the size of the zero-valued background can no longer influence the isodata algorithm and thus the value chosen for a noise threshold has no effect on the results. When the region contains three populations (e.g., background, object 1, and object 2), the resulting isodata threshold will fall between the background and the objects as long as the latter have similar intensities. Otherwise, the threshold will separate the objects. This can be useful if the middle population consists of noise.
23. Compute a 2×-opening (two erosion operations followed by two dilation operations) on the binary image resulting from step 22 to remove any spurious noise. The resulting image contains the blobs.
Data Processing and Analysis
10.9.7 Current Protocols in Cytometry
Supplement 3
A Intensity
original signal along one line in image
Location
small maximum filter (dilates objects)
large minimum filter (estimates background)
maximum minus minimum
B
resulting regions defined
C
original signal
max minus min
(reduces background values for dilated objects)
region 1
Location of Hybridization Domains on Chromosomes by Image Cytometry
region 2
Figure 10.9.6 Detecting regions with local contrast. The graphs shown are of intensity vs. location along a line through an object. (A) The effect of dilating and eroding the regions containing some signal. (B) The eroded result is subtracted from the dilated result so as to remove the slowly changing background signal and leave the dilated regions level with one another. (C) Choose a “noise threshold” nt (indicated by the horizontal dotted line) such that pixels with intensity over the threshold define region masks; the masks are used to “cookie cut” regions from the original signal. This threshold defines the minimum strength of an acceptable signal after background subtraction, but the exact value of this threshold is not critical. The goal is to isolate large regions that contain within them a signal of interest. The selection of the noise threshold will only nominally affect the size of the region, and the region will be analyzed more carefully during the blob and pinnacle detection steps of the DRT algorithm (Fig. 10.9.5B–C). To complete the region definition, the region masks that result from the noise threshold are used to isolate the regions of interest in the original image. Figure reprinted with permission from Mascio et al. (1995).
10.9.8 Supplement 3
Current Protocols in Cytometry
The 2×-opening removes very small, bright regions that are less than 4 pixels in diameter (Serra, 1982; Giardina and Dougherty, 1988). This manipulation may eliminate an object that includes the desired hybridization domain, an outcome that is undesirable. Figure 10.9.5 shows that the restricted isodata algorithm preserves three potential objects from two of the regions, and the 2×-opening eliminates one of those. Domain elimination can be overcome by using pinnacle detection.
Detect pinnacles 24. Compute the histogram of each individual region isolated in step 21. 25. Determine the highest intensity bin of the histogram that contains at least five pixels. Use this as an estimate for the intensity value of the brightest part of the object.
26. Multiply this value by 90% and apply a threshold at that value. Such a high threshold may lead to loss of shape and area information, but does not interfere with detection of the objects.
27. Apply a single binary opening to the binary image generated in step 26. The result image contains the pinnacles. A single binary opening will eliminate the spurious noise pixels that might remain even after the strict threshold. The single opening preserves potential hybridization domains that are 2 to 3 pixels in diameter. These domains would be eliminated by the 2×-opening used in the blob detection, as shown in Figure 10.9.4. Note also that the area of larger objects is reduced by this method as compared to the method of blob detection.
Combine blobs and pinnacles 28. Combine the results of the blob detection (from step 23) and pinnacle detection (from step 27) with a logical “or” operator to yield the detected hybridization domains (Fig. 10.9.5D). Remove spurious noise pixels. Localize hybridization domains along the medial axis 29. Compute the center of mass (COM) for each detected hybridization domain. 30. Overlay the COM of each hybridization domain with the results of the chromosome medial axis determination (from step 14). 31. Determine the shortest path between the hybridization COM and the medial axis. The result represents the location of the hybridization domain. This is the orthogonal projection of the hybridization domain onto the medial axis. The point where this projection meets the medial axis is the location of the domain along the axis.
COMMENTARY Background Information FISH has proven to be a powerful tool for assembling physical maps of the human and other genomes (Pinkel et al., 1988; Lawrence, 1990; Lichter et al., 1990; Kallioniemi et al., 1994; Sakamoto et al., 1995). The technique allows a cloned DNA sequence to be localized along the length of a metaphase chromosome with high precision. The locations of probes have been determined relative to chromosome bands, the end of the chromosome, and the locations of previously mapped probes. Local-
ization relative to the end of the chromosome short arm is appealing because the approach is well suited to automation. The automation algorithms described here have been used for FLpter analysis and have been evaluated by Sakamoto et al. (1995) and by Kallioniemi et al. (1994).
Critical Parameters The sizes of the algorithm filters suggested for chromosome segmentation, medial axis determination, and hybridization domain detec-
Data Processing and Analysis
10.9.9 Current Protocols in Cytometry
Supplement 3
tion should serve as examples; they contain information relative to one another. Filter sizes may need to be adjusted for adequate analysis of individual images, depending on the magnification of each system. It is important to note that the lengths of the chromosomes as derived by the chromosome segmentation process will be affected by the intensity of the DNA counterstain at the ends of the chromosomes. As a result, comparison of location information should be limited to images from similarly prepared samples. The combination of thresholding and morphological techniques in the DRT is highly nonlinear and depends heavily on local information relative to each object. This allows dynamic analysis and enables the detection of spots with a wide variety of characteristics, but it prevents the prediction of what fraction of an object will be included in a DRT segmentation result. Therefore, DRT segmentation results are best used to detect hybridization signals and to accurately determine domain location. The total fluorescence intensity associated with the hybridization domains should not be determined by DRT segmentation alone. Once the potential domains are located by DRT segmentation, they can be isolated and individually segmented to a judiciously predetermined fraction using a more straightforward method. The accuracy with which this center of mass can be computed depends on the intensity of the object and on the intensity and smoothness of the surrounding background. Relatively bright objects carry more weight per pixel than dim objects, so intensity has more influence than geometrical segmentation errors on the centerof-mass calculation. Similarly, large objects have more sampled data that can statistically outweigh small errors in segmentation. Small or dim objects, or objects in a noisy environment, contain less information, which can lead to larger errors in segmentation and center-ofmass calculation.
Troubleshooting The DRT algorithm efficiently detects potential hybridization domains and often highlights spots in FISH images that are not the hybridization domains of interest. Thus, it is useful to build a tool that allows the operator to choose the hybridizations of interest before computing FLpter values. Location of Hybridization Domains on Chromosomes by Image Cytometry
Anticipated Results The chromosome segmentation and medial axis determination methods described appear
to work well for well-separated chromosomes that are relatively straight and not overly condensed. The overall performance of this mapping system is based on the speed with which probes can be mapped and the precision with which FLpter values can be determined. Mapping speed for the DRT method has been reported to be about one probe per hour for dual-color mapping. Data from Mascio et al. (1995), Kallioniemi et al. (1994), and Sakamoto et al. (1995) suggest that the standard deviation for FLpter analysis using this system is ∼3 Mb. Furthermore, the measurements are more or less normally distributed. Thus, probes can be mapped with a precision (standard error of the mean) of <1 Mb, and probes separated by more than ∼2.5 Mb can be properly ordered based solely on their FLpter values (assuming that they have been mapped onto chromosomes from the same individual so that heteromorphic variability is not a factor).
Time Considerations The time for chromosome segmentation, medial axis determination, and probe detection for two different hybridization probes is ∼20 sec on a DEC station 5000/200 (33 MHz).
Literature Cited Dorst, L. and Smeulders, A.W.M. 1987. Length estimators for digitized contours. Comput. Graphics Image Process. 40:311-333. Freeman, H. 1970. Boundary encoding and processing. In Picture Processing and Psychopictorics (B.S. Lipkin and A. Rosenfeld, eds.) pp. 241266. Academic Press, New York. Giardina, C.R. and Dougherty, E.R. 1988. Morphological Methods in Image and Signal Processing. Prentice-Hall, Englewood Cliffs, N.J. Kallioniemi, A., Kallioniemi, O.-P., Mascio, L., Sudar, D., Pinkel, D., Deaven, L., and Gray, J.W. 1994. Physical mapping of chromosome 17 cosmids. Genomics 20:125-128. Kuwahara, M., Hachimura, K., Eiho, S., and Kinoshita, M. 1976. Processing of RI-angiocardiographic images. In Digital Processing of Biomedical Images (K. Preston and M. Onoe, eds.) pp. 187-203. Plenum, New York. Lawrence, J.B. 1990. A fluorescence in situ hybridization approach for gene mapping and the study of nuclear organization genome analysis. Genet. Phys. Mapping 1:1-39. Lichter, P., Tang, C., Call, K., Hermanson, G., Evans, G.A., Housman, D., and Ward, D.C. 1990. High-resolution mapping of human chromosome 11 by in situ hybridization with cosmid clones. Science 247:64-69.
10.9.10 Supplement 3
Current Protocols in Cytometry
Mascio, L.N., Verbeek, P.W., Sudar, D., Kuo, W.-L., and Gray, J.W. 1995. Semiautomated DNA probe mapping using digital imaging microscopy: I. System development. Cytometry 19:5159. Pinkel, D., Landegent, J., Collins, C., Fuscoe, J., Segraves, R., Lucas, J., and Gray, J.W. 1988. Fluorescence in situ hybridization with human chromosome-specific libraries: Detection of trisomy 21 and translocations of chromosome 4. Proc. Natl. Acad. Sci. U.S.A. 85:9138-9142. Ridler, T.W. and Calvard, S. 1978. Picture thresholding using an iterative selection method. IEEE Trans. Systems Man Cybernet. 8:630-632. Sakamoto, M., Pinkel, D., Mascio, L., Sudar, D., Peters, D., Kuo, W.-L., Yamakawa, K., Nakamura, Y., Drabkin, H., Jericevic, Z., Smith, L., and Gray, J.W. 1995. Semi-automated DNA probe mapping using digital imaging microscopy. II. System performance. Cytometry 19:6069. Serra, J. 1982. Image Analysis and Mathematical Morphology. Academic Press, London.
Verwer, B.J.H. 1988. Improved metrics in image processing applied to the Hildritch skeleton. Proceedings of the 9th International Conference on Pattern Recognition. Rome, Italy, Nov. 14-17, 1988, pp. 137-142. Computer Society Press, Washington, D.C. Vossepoel, A.M. and Smeulders, A.W.M. 1982. Vector code probabilities and metrication error in the representation of straight lines of finite length. Comput. Vision Graphics Image Process. 20:347-364.
Key Reference Mascio, L.N., Verbeek, P.W., Sudar, D., Kuo, W.-L., and Gray, J.W. 1995. See above. This paper presents and discusses the algorithm described in this unit.
Contributed by Laura Mascio Lawrence Livermore National Laboratory Livermore, California
Verbeek, P.W., Vrooman, H.A., and Van Vliet, L.J. 1988. Low-level image processing by max-min filters. Signal Process. 17:249-258.
Data Processing and Analysis
10.9.11 Current Protocols in Cytometry
Supplement 3
Three-Dimensional Image Visualization and Analysis The intent of this unit is to give the reader an understanding of the principles and applications of three-dimensional (3D) image analysis (IA) as applied in cytometry. The unit does not give a detailed description of how to undertake 3D IA for specific applications. For further information, please refer to the appropriate textbooks (e.g., Castleman, 1996); for scientific articles pertaining to specific applications, see Literature Cited. This unit is an extension of UNIT 10.5 and UNIT 2.8 which deal, respectively, with twodimensional image processing and analysis and confocal microscopy. Conventional optical microscopy acquires a two-dimensional (2D) image of the specimen. In cytometry, the image is either a direct representation of a thinly cut section of the (tissue) specimen or a projection through a thicker, three-dimensional (3D) specimen. In either case, the 2D image is only a partial representation of the specimen and, consequently, techniques have been developed to acquire 3D images of biological specimens (see UNIT 2.8 on confocal microscopy), and to visualize and computationally analyze them. A 3D image can be considered to be a series of 2D slice images recorded at successively increasing depths through the specimen. This unit briefly describes the techniques for visualization and analysis of 3D images, and also describes an alternative acquisition approach based on conventional epifluorescence microscopy and deconvolution. Following these descriptions are some examples of how 3D analysis has been used in practice in cytometry. An appendix of additional resources is at the end.
DEFINITIONS The following is a list of definitions for abbreviations used in this unit. The reader should refer to Inoué and Spring (1997) for an extensive glossary of terms used in microscopy and image analysis. 2D two dimensional. 3D three dimensional. Animation the process of displaying a series of 2D images in rapid succession. Anisotropic term used to describe unequal physical properties of an image along different axes (in this unit referring to the dimensions of a voxel). Contributed by Stephen J. Lockett Current Protocols in Cytometry (1999) 10.10.1-10.10.13 Copyright © 1999 by John Wiley & Sons, Inc.
UNIT 10.10
Binary image a digital image where each voxel has a value of either 0 or 1. Deconvolution the mathematical procedure for removing the distortion in an image cased by the PSF (see below and UNIT 2.6) of the image acquisition system; it is also commonly called “deblurring” and is implemented by application of computer programs on digital images. Depth direction the direction through a 3D image that is parallel to the optical axis of the microscope; generally the z dimension of the image is assigned this direction. Gallery a series of 2D images displayed adjacent to one another (for an example, see see Fig. 10.10.2, panel B). Gray image a digital image where each voxel has a scalar value; normally the value is an integer in the range 0 to 255. Interpolation the mathematical procedure for increasing or decreasing the sampling density of an image (see discussion of Display of 3D Images). Neighborhood the set of voxels directly adjacent to a specific voxel in an image (see discussion of Neighborhood of a Voxel). PI propidium iodide. Pixel a rectangular element within a digitized 2D image (see definition of “voxel” below). PSF point spread function; the image that would be obtained if it were possible to image an infinitely small point source of light; it is either calculated from theory or measured using a small light source (also see UNIT 2.6). Segmentation the mathematical procedure of separating a digitized image into background and one or more types of objects (see Example of Segmentation of a 3D Confocal Microscope Image) SF shape factor; a measure of the irregularity in the shape of an object. Threshold intensity levels calculated from the image that can be used to segment the image into background and one or more types of objects (see Fig. 10.10.7A and B for an example) Voxel a box-shaped volume element within a digitized 3D image (see Description of a 3D Image). Data Processing and Analysis
10.10.1 Supplement 10
z x y
A
B
Figure 10.10.1 The 3D digital image. (A) The image is composed of box-shaped elements, called “voxels,” in contrast to 2D images, which are composed of rectangular elements called “pixels.” The image shown here has 3 voxels in the x direction, 5 in the y direction, and 4 in the z direction; therefore the total number of voxels in this 3D image is 3 × 5 × 4 = 60 voxels. (B) A 2D slice image in the xz plane from the 3D image.
A
B
z x y
C
D
x
x y z
y z z
y x
Three-Dimensional Image Visualization and Analysis
x
y
z
Figure 10.10.2 Display of 2D slices cut from a 3D image. (A) Schematic representation of a 3D image as a series of 2D slices. (B) Display of a 3D image in the form of a gallery of 2D slices. Shown are 31 slices comprising a 3D image of a solid white object on a black background. (C) Schematic showing the orientation of orthogonal slices through a 3D image. Nonorthogonal slices can also be cut from the image, but they are computationally complex to calculate. (D) The three orthogonal slices that intersect at the center of the 3D image shown in gallery form in B (the xy slice is the same as the slice at row 2, column 8 in panel A). Note that the orthogonal slices reveal that the object is a dumbbell consisting of two superimposed spheres (see Lockett et al., 1998a, for more details). This is not obvious from the gallery in panel B due to the orientation of the dumbbell with respect to the slices.
10.10.2 Supplement 10
Current Protocols in Cytometry
DESCRIPTION OF A 3D IMAGE Figure 10.10.1 illustrates (schematically) a digital 3D image and defines the basic terminology. A typical 3D image encountered in cytometry would be 512 voxels in the x direction, 512 voxels in the y direction, and 50 voxels in the z direction, for a total of 13.1072 × 106 voxels. In optical microscopy, a voxel represents a box-shaped volume of the specimen, with a typical size of 0.3 µm in the x direction, 0.3 µm in the y direction, and 0.6 µm in the z direction. This is a volume of 5.4 × 10−20 m3. These values can vary considerably, but frequently a voxel is set to a larger size in the z direction than in the x and y (i.e., is anisotropic), because of the lower spatial resolution of optical microscopy in this direction. This anisotropy has important implications for image display, processing, and analysis.
DISPLAY OF 3D IMAGES Since the most powerful image analysis system is the human visual system, which excels at object recognition (although object measurement is not accurate without additional computational aids) and which in many cases is the only image analysis required, appropriate display of the image is an essential prerequisite to image analysis. While the presentation of 2D images to the (2D) human visual system via a computer screen is straightforward, the presentation of 3D images is generally not simple. Some form of image processing is required to select only a subset of the 3D data for display. Well established methods exist for presenting 3D images. The easiest method is to display each of the 2D slices that comprise the 3D image separately, beside each other in an array format (known as a “gallery”; see Fig. 10.10.2B), or to display each slice in rapid succession (known as “animation”). Such displays have the advantage of showing the intensity information at every voxel in the 3D image, but the spatial relationships between voxels in the depth direction (the direction parallel to the optical axis of the microscope and perpendicular to the plane of the slices) is partially lost, making it difficult to observe depth-associated changes in the 3D image. Reslicing the 3D image in a different direction, e.g., in the xz plane and thus displaying xz slices (see Figs. 10.10.2C and 10.10.2D), might solve the problem, but the choice of the new direction is arbitrary and thus there is no guarantee that it will help. The reader should be aware that the display of 3D images will be distorted if the anisotropy (differences in the properties of an
image in one direction versus another dimension) of the voxels is not corrected for. The most common correction method is to stretch the acquired image in the z direction using interpolation so that voxels in the expanded image are isotropic. For example, a 512 × 512 × 50 voxel image with a voxel size of 0.3 × 0.3 × 0.6 µm would be expanded by a factor of 2 in the z direction, resulting in an interpolated image of 512 × 512 × 100 isotropic voxels of size 0.3 × 0.3 × 0.3 µm. The reader should refer to the Purdue University Cytometry Laboratories Microscopy CD-ROM (see Internet Resources) for examples of 3D image rendering.
VOLUME RENDERING One of the two most popular methods for displaying 3D images is volume rendering, which is based on projecting the 3D image along a given direction into a 2D “projection” image. The method is illustrated in Figures 10.10.3 and 10.10.4. Figure 10.10.3 illustrates how the intensity at coordinate (x′, y′), IPx′,y′ in the projection image is a function of the intensities along the line (x′, y′, z1) to (x′, y′, zn) through the 3D image, where n is the number of slices in the 3D image. Mathematically, IPx′, y′ = f(ix′, y′, z1, ix′, y′, z2, .... , ix′, y′, zn), where ix′,y′,zk is the intensity at voxel x′, y′, zk in the 3D image. The function f() depends on the type of 3D image and on the information that the analyst wishes to visualize in it. Often the function simply averages the intensities along each line, because this gives equal weights to all the intensities. In cytometry, however, 3D images are generally of fluorescence labels, and thus the information of interest is predominantly the high-intensity voxels corresponding to high concentrations of label. In this situation the maximum operator is chosen for f() in order to “pick out” the highest intensity along each projection line for display in the 2D image. If there is more than one bright spot along a given line, only the brightest will be projected into the displayed image. This limitation is greatly reduced by generating a series of projection images where the 3D image is incrementally rotated along a line perpendicular to the projection direction (see Fig. 10.10.3). The series of projections is then displayed as an animation which gives a “rotating movie” impression to the viewer. Figure 10.10.4B shows examples of maximum intensity projections at different angles through a 3D image of fluorescently stained cell nuclei.
Data Processing and Analysis
10.10.3 Current Protocols in Cytometry
Supplement 10
rotation axis x
x z
y
(x′,y′)
y
(x′,y′,z1)
(x′,y′,zn )
3D image
projection image Figure 10.10.3 Generation of projection images in volume rendering. The intensity at coordinate (x′, y′) in the projection image is a function of the set of intensities along the line (x′, y′, z1) to (x′, y′, zn) in the 3D image. The same transformation is applied at all coordinates in the xy plane. Often a series of projection images is generated, by incrementally rotating the 3D image about an axis orthogonal to the projection direction.
A common application of volume rendering is stereo pair generation, where two projection images are calculated with the 3D image rotated by approximately 6° to 10° between them. When the two images are viewed simultaneously, one with the left eye and the other with the right eye, an impression of looking directly into the 3D image is obtained. Figure 10.10.4C is an example of a stereo pair image.
SURFACE RENDERING
Three-Dimensional Image Visualization and Analysis
Surface rendering is a method for displaying only a surface of a 3D object (normally the outermost surface) to the human eye. Thus, it deliberately does not attempt to display all the information contained in the 3D image, and is the method of choice for “opaque” objects whose insides do not contain any information of interest. It is also the most natural display method because it mimics the way we view the world around us. However, surface rendering must be preceded by the image analysis procedure of “segmentation” (see below) in order to extract the surfaces of interest from the 3D image. It should be noted that significant errors in the segmentation of microscope images of biological specimens are not uncommon, causing the rendering to not accurately represent the contents of the specimen. Figure 10.10.5 shows examples of surface rendering using the same 3D image as in Figure 10.10.4. Figure 10.10.5A shows the surface of the cell nuclei in Figure 10.10.4B (0°). This image is disappointing be-
cause the nuclei look flat, making it impossible to determine the orientation of the surfaces and to delineate the nuclei from each other. To remove these limitations, it is common practice to introduce shading (Fig. 10.10.5B) into the rendering. Shading is performed by shining an imaginary light on the surface and determining at every point on the surface how much light is reflected to the observer. The amount of reflection at each point is a function of the orientation of the surface at that point to the light source and observer. In addition, the surface can be made partially transparent or represented by a wire frame so that other, internal surfaces can also be displayed (Fig. 10.10.5C). A similar method for rendering surfaces is the “3D perspective view,” where the objects of interest are defined by a contour in each slice of the 3D image (see Fig. 2.8.9 in UNIT 2.8 for an example).
3D IMAGE ANALYSIS Analysis of 3D images is generally performed by computer programs which are a combination of interactive and automatic steps. Generally, the interactive steps enable the user to perform recognition tasks (e.g., identifying the individual cell nuclei in an image of tissue) which the human visual system can do accurately and efficiently, whereas the tasks which are tedious for the user (e.g., calculating the volume of a segmented nucleus) are done automatically by the program. Below are described some of the image analysis steps (operations)
10.10.4 Supplement 10
Current Protocols in Cytometry
A
B
0˚
30˚
C
60˚
90˚
D
Figure 10.10.4 Example of 3D volume rendering of cell nuclei in the epidermis of normal human skin labeled with propidium iodide. (A) Three (nonadjacent) slices from the 3D image. (B) “Maximum intensity” projections at 30° increments through the 3D image. (C) Stereo pair generated from the 3D image. To view, look at the right image with the left eye and the left image with the right eye. (D) Overlay of the pair of images in C after first detecting the edges of the nuclei using a gradient magnitude operator (see UNIT 10.5).
which are commonly applied in 3D cytometry, followed by an example illustrating how they can be compiled into an image analysis program. Much of this section is a straightforward extension of UNIT 10.5, so this unit covers only those aspects of image analysis of special relevance to 3D cytometry.
NEIGHBORHOOD OF A VOXEL An important concept to understand in 3D image processing and analysis is the “neighbor-
hood” surrounding a voxel—that is, the set of voxels directly adjacent to a specific voxel in an image. The most common neighborhood encountered in 3D imaging is the 3 × 3 × 3 neighborhood, which is illustrated in Figure 10.10.6. There are 3 types of 3 × 3 × 3 neighborhood. The first is the 26-connected, shown in Figure 10.10.6A; note that the distance from the center of voxel c to the center of each neighbor is not the same. For example, the voxel at (1,0,0) is 1 unit of distance away, the
Data Processing and Analysis
10.10.5 Current Protocols in Cytometry
Supplement 10
green
A
purple
C
B
red
Figure 10.10.5 Surface rendering of the 3D image shown in Figure 10.10.4. (A) The surfaces of the nuclei in Figure 10.10.4B without any shading. (B) Shading added by reflecting an imaginary light off the surface, which reveals the orientation of each point on the surface and thus makes it possible to delineate the individual nuclei. (C) By using wire frames to render surfaces, it is possible to see objects inside the nuclei. The white dots inside the nuclei are the centromere of chromosome 1, which has been labeled using fluorescence in situ hybridization (FISH; see Chapter 8).
–1
0
1 x
–1 z 1 0
C
–1
C
0
1
y
A
B
Figure 10.10.6 Neighborhood of a voxel. (A) The 3 × 3 × 3 neighborhood surrounding the central voxel, c. The slices have been separated in the z direction for clarity. Voxel c has 26 neighbors, 9 each in the xy planes above and below and 8 in the xy plane containing c. Therefore this neighborhood is called the 26-connected neighborhood. When the coordinates of c are set to the origin (i.e., x, y, z = 0, 0, 0), the 26 neighbors have coordinates: (−1, −1, −1), (−1, −1, 0), ...... , (−1, 0, −1), ...... , (1, 0, −1), ...... , (1, 1, 1) leaving out (0, 0, 0). (B) The 6-connected neighbors of C: (1, 0, 0), (−1, 0, 0), (0, 1, 0), (0, −1, 0), (0, 0, 1), (0, 0, −1).
10.10.6 Supplement 10
Current Protocols in Cytometry
voxel at (1,1,0) is √2 away and the voxel at (1,1,1) is √3 away. The set of voxels that are 1 unit of distance from voxel c (Fig. 10.10.6B) is the second type, called the 6-connected neighborhood. The third type is the set of 6-connected neighbors plus the voxels √2 unit away, which forms the 18-connected neighborhood. Other-sized neighborhoods may be used, such as 5 × 5 × 5, and neighborhoods need not be cubes. For example, neighborhoods often are designed to represent a cube of the actual specimen; thus if the voxel size is x = 0.3, y = 0.3, and z = 0.5 µm, then a 5 × 5 × 3 neighborhood would represent a cube of the specimen measuring 1.5 × 1.5 × 1.5 µm.
NOISE-REMOVAL METHOD FOR 3D FLUORESCENCE IMAGES 3D fluorescence images, particularly those acquired with photomultiplier tubes operating at high gain, such as in confocal microscopes, contain shot noise. The resulting effect is that random voxels obtain very high intensity values and thus cause the image to look as if it has been sprinkled with salt. One method to reduce this noise is image averaging, which must be performed during image acquisition (see UNIT 10.5). The other common method is median filtering, which can be performed after the image is acquired. This is a local filtering operation, meaning that the result of applying the filter at each voxel depends on the intensities of all the voxels in a neighborhood around the voxel being filtered but not on voxels outside the neighborhood. Generally, a 3 × 3 × 3, 26-connected neighborhood is used, but because of anisotropy, a 5 × 5 × 3 neighborhood might be used instead. The median filter is applied at each voxel in the image by listing the intensities of all the neighborhood voxels in increasing order. Then, the middle value (i.e., the 14th values in the list for a 3 × 3 × 3, 26-connected neighborhood) is the new intensity for the voxel being filtered. Since the highintensity voxels from shot noise are nearly always at the end of the list, they are replaced by intensity values more representative of the specimen in that neighborhood. Noise removal should be done before interpolation to correct for anisotropy.
DEPTH ATTENUATION AND CORRECTION In 3D fluorescence microscopy, the full thickness of the specimen is exposed to the excitation light, which causes continuous photobleaching of the fluorescence labels in the
specimen. The effect is that the later slices to be imaged are less intense than the earlier ones, with the loss of intensity approximately following the exponential decay law (see UNIT 2.8). It is necessary to correct for this attenuation to obtain accurate measurements of fluorescence intensity and certain other image-processing and analysis tasks. A simple procedure for attenuation correction is as follows. First, a sequence of 2D images (5 to 10 = N) is taken of a representative area of the specimen at the same position and depth using the same exposure for each image as for each slice in the 3D image. These conditions ensure that the sequence number of the image (k: 1 to N) is proportional to the total exposure time of the specimen. Next the first image is segmented into regions of fluorescence label and background (see below). The region of fluorescence label is used as a mask to locate the same fluorescence region in the subsequent images. For each region, the total fluorescence is calculated and the background subtracted (see below). This series of background-subtracted total intensities is fitted using the least-squares method (see, e.g., Young, 1962) to the exponential decay curve: loge(Ik) = loge(I1) – λk, where λ is the photobleaching rate and I1 and Ik the background-subtracted fluorescence intensity of the first and kth images respectively. The correction is implemented by multiplying the background-subtracted fluorescence signal in slice k by exp(λk).
EXAMPLE OF SEGMENTATION OF A 3D CONFOCAL MICROSCOPE IMAGE The 3D image used in this example is of cell nuclei labeled with the fluorescent DNA label propidium iodide (PI). A slice from the image is shown in Figure 10.10.7A. The series of steps used to segment the nuclei in this image could also be applied to a 2D image; for an example of a more advanced 3D segmentation algorithm, see Ortiz de Solorzano et al. (1999). The first step is thresholding (see UNIT 10.5) to divide the image into a number of disconnected regions of nuclei and background (Fig. 10.10.7B). Some of the regions represent nuclei, while others represent a cluster of several nuclei (Fig. 10.10.7C). Next, each cluster is repeatedly shrunk until it breaks up into two or more smaller regions, each of which represents one nucleus (Fig. 10.10.7D). Shrinking is done using a local filter called a binary erosion, which uses as input a binary image (Fig. 10.10.7B) where voxels can have only the value
Data Processing and Analysis
10.10.7 Current Protocols in Cytometry
Supplement 10
A
C
B
D
E
red green
F
yellow
red green
yellow
Figure 10.10.7 Example of segmentation of a 2D image of fluorescence-labeled cell nuclei. (A) 2D slice image from a 3D image of cell nuclei labeled with propidium iodide (PI; a red dye, rendered in dark gray here). (B) The image in panel A, after thresholding to convert it to a binary image of nuclear regions (white) and background (black). (C) A nuclear region containing a cluster of several nuclei. (D) The region in panel C after application of the binary erosion operator to shrink the region and divide it into multiple regions, each one representing a single nucleus. (E) Labeling of each region in panel D. (F) Dilation of the regions back to their original size.
1 or 0. In the filtering operation, which is normally done in a 3 × 3 × 3, 26-connected neighborhood, if any voxel in the input image has value 0, then the voxel being filtered is set to 0 in the output image. In the next step, the shrunk regions are labeled by giving every voxel in each region the same value, but each region has a different value (Fig. 10.10.7E). Then each region is enlarged at an equal rate using repeated application of the binary dilation operator (opposite of erosion) until the regions fill the volume they occupied before erosion (which requires maintaining a copy of the image from before the erosion process). Figure 10.10.7F shows the result. Three-Dimensional Image Visualization and Analysis
MEASUREMENT OF 3D OBJECTS A very large number of measurements can be made on 3D objects after they are seg-
mented. Only the simplest few are described here. The volume (V) of an object is the number of voxels in it. Shape factor (SF) is a useful parameter for measuring the regularity of the shape of an object. 1.5
SF =
(SA )
6V π
where SA is the surface area of the object. SA can be approximately calculated by counting the number of voxels in the object that are touching background voxels or voxels of other objects. SF has a minimum value of 1 for a sphere. The total fluorescence intensity of an object can be calculated by adding up all the voxel values of the object. However, this procedure has two common sources of error. First, some of the intensity from the object lies outside the
10.10.8 Supplement 10
Current Protocols in Cytometry
border of the segmented object because of the point spread function (PSF) of the image-acquisition system. The PSF is the image that would be obtained if it were possible to image an infinitely small point of light (see UNIT 2.6). Second, the voxel intensities contain background signal, such as detector noise, autofluorescence, and nonspecific labeling. The first error is corrected by dilating the object before measurement of total intensity. The second is corrected by making the best possible estimate of the background signal from the background voxels surrounding the dilated object. By simple arithmetic it can be shown that the total fluorescence intensity of the object after correction for both these errors is (T1V2 − T2V1)/(V2 − V1), where T1 is the sum of the voxel intensities of the dilated object, V1 is the volume of the dilated object, T2 is the sum of the voxel intensities of the object after additional dilation to deliberately include surrounding background signal, and V2 is the volume of the object after additional dilation.
DECONVOLUTION OF 3D IMAGES ACQUIRED USING CONVENTIONAL EPIFLUORESCENCE MICROSCOPY Although the most common method for acquiring 3D images is confocal microscopy (UNIT 2.8), the pinhole in the emission light path of the confocal microscope causes inefficient detection of the fluorescence signal. This can be a serious problem when the fluorescence labels are very sensitive to photobleaching or when live cells are damaged by the excitation light. An alternative acquisition approach is to acquire the 3D image using conventional microscopy, which has the advantage of detecting a much larger proportion of the emitted light. However, the 3D image is not directly related to the spatial distribution of the fluorescing molecules in the specimen. This is because the intensity at each voxel consists of a component proportional to the fluorescence at the corresponding location in the specimen (the in-focus signal) plus an additional component from nearby fluorescing molecules above and below the focal plane (the out-of-focus component). It is therefore necessary to subtract this out-offocus component using the image-processing method of 3D deconvolution, in order to obtain a 3D image where the voxel intensities are approximately proportional to the 3D fluorescence distribution of the specimen. Deconvolution is a mathematical procedure for removing the distortion in an image caused by the out-of-focus light. It is commonly called “de-
blurring” and is implemented by application of computer programs on digital images. Software for the method is available from several vendors (see Table 10.10.1), and the principles are described by Agard et al. (1989) and Castleman (1996). Examples showing the result of application of deconvolution to 3D images acquired by conventional microscopy are shown in the vendors’ Web sites (see Table 10.10.1 and see Internet Resources) and on the Purdue University Cytometry Laboratories Microscopy CD-ROM (see Internet Resources).
3D RECONSTRUCTION OF TISSUE FROM SERIAL THIN SECTIONS Analysis of cells within their broader tissue context is an important capability necessary to understand, e.g., the processes of tissue development and the spreading of cancerous lesions. However, such analysis is limited by the small field of view in high-spatial-resolution 3D optical microscopy (typically a volume of 200 × 200 × 100 µm), and in certain studies by the limited penetration depth of (fluorescence) labels. The field of view in the x and y dimensions can be easily expanded by tiling together an array of images recorded at different positions of the microscope’s stage. The approach for increasing the range in the z direction is first to section the tissue thinly (4 µm), then to acquire images of each section, and finally, by registering each pair of adjacent images, to reconstruct the 3D image of the tissue. In practice, the registration is nontrivial because of differential damage and distortions introduced by the sectioning. It is beyond the scope of this unit to discuss registration methods, but, generally, the most practical approaches require the analyst to visually identify common, well localized features in pairs of images of adjacent sections (Verbeek, 1995; Lockett et al., 1998b).
OTHER FORMS OF 3D IMAGING Thus far, it has been assumed that the term 3D imaging refers to the three spatial dimensions. In studies of dynamic processes in live cells, 3D imaging is often in the form of two spatial dimensions with the third dimension being time. Generally the objective in these studies is to analyze intracellular dynamic processes, for example, intracellular calcium waves (Camacho and Lechleiter, 1993) or the tracking of cells in tissue (Wessels et al., 1998). The latter example usually involves delineation of the moving cell in each 2D image followed by measurement of the cell’s location between images. The image-analysis tools for such mo-
Data Processing and Analysis
10.10.9 Current Protocols in Cytometry
Supplement 10
Table 10.10.1
Table of 3D Image Processing and Analysis Softwarea
Name
Source and Internet address
Platforms
Additional comments
SCIL-Image
Pattern Recognition Group, Department of Applied Physics Delft University of Technology The Netherlands http://www.ph.tn.tudelft.nl/software. html National Institutes of Health http://rsb.info.nih.gov/nih-image Free Universal Imaging Corporation http://www.image1.com
PowerMac, Windows, Unix
Selection of image processing and analysis algorithms, mainly for 2D. Separate library available (DIPlib) for 3D.
Macintosh and Windows 95
2D image processing and analysis
Windows 95
Unix, PC
Complete imaging systems with applications to live cell image analysis —
Unix, Mac
—
Unix
Emphasis on 3D visualization Deconvolution Software specifically for confocal microscopy
NIH Image
MetaMorph imaging system 3DVIEWNIX
ANALYZE
AVS DeltaVision ImageSpace
Voxblast VoxelView XCOSM
Scanalytics IDL
EIKONA3D
Medical Image Processing Group Department of Radiology University of Pennsylvania Philadelphia, Pa. http://www.mipg.upenn.edu Mayo Medical Ventures Rochester, MN http://www.mayo.edu/bir/home.html Advanced Visualization Systems http://www.avs.com http://www.api.com Molecular Dynamics, Sunnyvale, Calif. http://www.mdyn.com Vaytek, Inc., Fairfield, Iowa Vital Images, Fairfield, Iowa http://www.vitalimages.com/ Institute for Biomedical Computing Washington Univ., St. Louis, Mo. http://www.ibc.wustl.edu/bcl/xcosm/ xcosm.html Scanalytics, Inc., Fairfax, Virginia http://www.signalanalytics.com/ Research Systems, Inc. 2995 Wilderness Place Boulder, Colo. Alpha Tec Ltd., Thessaloniki, Greece http://www.alphatecltd.com
SGI SGI
Unix, PC, Mac SGI, Mac
3D visualization
Unix
Deconvolution
PC, Mac
Deconvolution
Unix, PC, Mac
Very complete image processing and analysis package 3D Volumetric image processing and analysis
Windows
aSee UNIT 10.5 for additional image processing and analysis software.
Three-Dimensional Image Visualization and Analysis
10.10.10 Supplement 10
Current Protocols in Cytometry
tion analysis are available in several imageprocessing programs (e.g., NIH-Image and the Metamorph imaging system; see Table 10.10.1). The Purdue University Cytometry Laboratories Microscopy CD-ROM (see Internet Resources) contains several examples of time-lapse images. The study of dynamic processes extends to four dimensions (4D), when a time series of three-spatial-dimension images is acquired and analyzed. The state of the art in 4D imaging with applications to cytometry has been reviewed by Thomas and White (1998). Spectral imaging is another form of 3D imaging, where two dimensions are spatial and the third is the fluorescence emission spectrum, representing the intensity as a function of wavelength. Spectral imaging has applications in the fields of quantitative histology and cytogenetics, and has been reviewed by Rothman et al. (1998).
EXAMPLES OF 3D IMAGE ANALYSIS IN CYTOMETRY This section provides a few examples from the scientific literature illustrating the application of 3D image analysis in cytometry. The reader should refer to the literature for details of the methods used (see Literature Cited). One of the earlier applications of 3D image analysis in cytometry was to measure DNA ploidy distributions in solid tissue. The standard flow cytometry method for this measurement (see Chapter 7) is not well suited for solid tissues, because nuclei can be difficult to disperse and small fractions of nondiploid nuclei can be missed. Furthermore, the use of thinly sliced (4-µm) tissue sections with 2D imaging leads to large errors because whole nuclei are not present for measurement. These problems have been solved by the use of thick (>20 µm) tissue, fluorescence labeling, confocal microscopy, and 3D image analysis (Rigaut et al., 1991). The 3D image-analysis steps involved are correction for depth attenuation, segmentation of the individual cell nuclei, and measurement of the total fluorescence from the DNA stain in each nucleus. The ability to segment individual cell nuclei from within intact tissue is an important prerequisite for the quantitative analysis of intranuclear organization and cell-to-cell organization of tissue. Segmentation is often difficult in solid cancer specimens where tight clusters of nuclei are difficult to divide up. This problem has now been solved to a large extent by Ortiz de Solorzano et al. (1999), who included a visual classification step in the otherwise auto-
matic segmentation program. In the classification step, the user used a 3D rendering of each initially segmented object to tell the program if each object was an individual nucleus or a cluster of nuclei. Clusters were then automatically divided. Although the program does not segment all nuclei in every specimen, the inclusion of the visual step means that it is known which nuclei in a specimen have and have not been correctly segmented up to the accuracy of visual judgment. Several studies have used 3D image analysis to measure internal molecular organization of the nucleus. An example is the spatial localization patterns of specific DNA loci. The technique involves labeling of the loci (usually the centromeric regions of chromosomes) by fluorescence in situ hybridization (FISH; Thompson et al., 1994), 3D segmentation of individual nuclei, and measurement of the Euclidean distance of FISH signals to the center of mass of the nucleus and to each other. Some of these studies have revealed nonrandom distributions of DNA loci within their nucleus (Höfers et al., 1993). Recently, more generalized approaches for measuring the randomness of the 3D spatial distributions of fluorescentlabeled spots or protein in the nucleus have been described (Noordmans et al., 1998). The present state of knowledge about the territorial organization in human cell nuclei derived from using FISH, 3D microscopy, and 3D image analysis is summarized by Cremer et al. (1996). Another application of 3D image analysis to intracellular analysis has been the measurement of the redistribution of calcium ions from the cytosol to the nucleus (Yelamarty et al., 1990). The technique involved deconvolution of 3D images, since the presence of out-of-focus fluorescence could have obscured the calcium gradients. Analysis of the cell-to-cell organization of tissue is another key application of 3D image analysis in cytometry. In an attempt to improve the classification of poorly differentiated epithelial specimens, Albert et al. (1992) first segmented the nuclei and then used graphtheoretic methods to measure the distribution of center-to-center distances between the nuclei. When differentiating dysplasia from carcinoma, 2D analysis of the distance distribution in 2D resulted in a 20% error, which reduced to zero error when analysis was done in 3D. Wartenberg et al. (1998) evaluated the depth distribution of the drug doxorubicin in multicellular cancer spheroids. Their technique required confocal image acquisition, correction
Data Processing and Analysis
10.10.11 Current Protocols in Cytometry
Supplement 10
for depth attenuation, and quantitative analysis of the radial distribution of the drug concentration in the spheroids. In other cytometric studies, the goal is to study long-range properties of tissue, for example in 3D modeling of nerve cell morphology (Carlbom et al., 1994) or following the branching pattern of ducts in breast tissue. These studies require tissue volumes too large for direct 3D imaging, so the tissue is serially sectioned, and each section is imaged using (2D) microscopy followed by registration of the adjacent 2D images in order to reconstruct the 3D image of the tissue. This method has been used by Kaufman et al. (1998) to prepare a digital atlas of mouse development, which represents a valuable teaching aid and research tool in anatomy. Kay et al. (1998) have used analogous techniques for studying the architecture of microvessels in prostate cancer.
LITERATURE CITED Agard, D.A., Hiraoka, Y., Shaw, P., and Sedat, J.W. 1989. Fluorescence microscopy in three dimensions. Methods Cell Biol. 30:353-77. Albert, R., Schindewolf, T., Baumann, I., and Harms, H. 1992. Three-dimensional image processing for morphometric analysis of epithelium sections. Cytometry 13:759-765. Camacho, P. and Lechleiter, J.D. 1993. Increased frequency of Ca2+ waves in Xenopus laevis oocytes expressing a Ca2+-ATPase. Science 260:226-229.
Lockett, S.J., Sudar, D., Thompson, C.T., Pinkel, D., and Gray, J.W. 1998a. Efficient, interactive, and three dimensional segmentation of cell nuclei in thick tissue sections. Cytometry 31:275-286. Lockett, S.J., Fernandez, C., Rodriguez, E., Wesselmann, U., Bastian, B.C., Sudar, D., Pinkel, D., and Gray, J.W. 1998b. Interactive system for registering adjacent tissue sections. Proc. SPIE 3260:154-161. Noordmans, H.J., van der Kraan, K., van Driel, R., and Smeulders, A.W.M. 1998. Randomness of spatial distributions of two proteins in the cell nucleus involved in mRNA synthesis and their relationships. Cytometry 33:297-309. Ortiz de Solorzano, C., Rodriguez, E.G., Jones, A., Sudar, D., Pinkel, D., Gray, J.W., and Lockett, S.J. 1999. Automatic nuclear segmentation for 3D thick tissue confocal microscopy. J. Microsc. 193:212-226. Rigaut, J.P., Vassy, J., Herlin, P., Duigou, F., Masson, E., Briane, D., Foucrier, J., Carvajal-Gonzalez, S., Downs, A.M., and Mandard, A.-M. 1991. Three-dimensional DNA image cytometry by confocal scanning laser microscopy in thick tissue blocks. Cytometry 12:511-524. Rothman, C., Bar-Am, I., and Malik, Z. 1998. Spectral imaging for quantitative histology and cytogenetics. Histol. Histopathol. 13:921-926. Thomas, C.F. and White, J.G. 1998. Four-dimensional imaging: The exploration of space and time. Trends Biotechnol. 16:175-182.
Carlbom, I., Terzopoulos, D., and Harris, K.M. 1994. Computer-assisted registration, segmentation, and 3D reconstruction from images of neuronal tissue sections. IEEE Trans. Med. Image. 13:351-362.
Thompson, C.T., LeBoit, P.E., Nederlof, P.M., and Gray, J.W. 1994. Thick-section fluorescence in situ hybridization on formalin-fixed, paraffinembedded archival tissue provides a histogenetic profile. Am. J. Pathol. 144:237-243.
Castleman, K.R. 1996. Digital Image Processing. Prentice Hall, Englewood Cliffs, N.J.
Verbeek, F.J. 1995. Three-Dimensional Reconstruction of Biological Objects from Serial Sections Including Deformation Correction. Ph.D. Thesis, Technical University of Delft, The Netherlands.
Cremer, C., Münkel, C., Granzow, M., Jauch, A., Dietzel, S., Eils, R., Guan, X.Y., Meltzer, P.S., Trent, J.M., Langowski, J., and Cremer, T. 1996. Nuclear architecture and the induction of chromosomal aberrations. Mutation Res. 366:97116. Höfers, C., Baumann, P., Hummer, G., Jovin, T.M., and Arndt-Jovin, D.J. 1993. The localization of chromosome domains in human interphase nuclei: Three-dimensional distance determinations of fluorescence in situ hybridization signals from confocal laser scanning microscopy. Bioimaging 1:96-106.
Three-Dimensional Image Visualization and Analysis
Kay, P.A., Robb, R.A., and Bostwick, D.G. 1998. Prostate cancer microvessels: A novel method for three-dimensional reconstruction and analysis. Prostate 37:270-277.
Wartenberg, M., Hescheler, J., Acker, H., Diedershagen, H., and Sauer, H. 1998. Doxorubicin distribution in multicellular prostate cancer spheroids evaluated by confocal laser scanning microscopy and the “optical probe technique.” Cytometry 31:137-145.
Inoué, S. and Spring, K.R. 1997. Video Microscopy. Plenum, London.
Wessels, D., Voss, E., Von Bergen, N., Burns, R., Stites, J., and Soll, D.R. 1998. A computer-assisted system for reconstructing and interpreting the dynamic three-dimensional relationships of the outer surface, nucleus and pseudopods of crawling cells. Cell Motil. Cytoskeleton 41:225246.
Kaufman, M.H., Brune, R.M., Davidson, D.R., and Baldock, R.A. 1998. Computer-generated threedimensional reconstructions of serially sectioned mouse embryos. J. Anat. 193:323-336.
Yelamarty, R.V., Miller, B.A., Scaduto, R.C., Yu, F.T.S., Tillotson, D.L., and Cheung, J.Y. 1990. Three-dimensional intracellular calcium gradients in single human burst-forming units-
10.10.12 Supplement 10
Current Protocols in Cytometry
erythroid-derived erythroblasts induced by erythropoietin. J. Clin. Invest. 85:1799-1809. Young, H.D. 1962. Statistical Treatment of Experimental Data. Waveland Press, Prospect Heights, IL.
INTERNET RESOURCES http://genex.hgu.mrc.ac.uk/ The Mouse Atlas and Gene Expression Database Project. http://biocomp.stanford.edu/3dreconstruction/ index.html
Comprehensive list of image processing and analysis software packages. http://www.cyto.purdue.edu Purdue University Cytometry Laboratories. Provider of the Microscopy CD-ROM containing examples of 2D, 3D, and time-lapse fluorescence microscope images, software, reference and educational material, and microscopy Web sites.
Contributed by Stephen J. Lockett Lawrence Berkeley National Laboratory Berkeley, California
Data Processing and Analysis
10.10.13 Current Protocols in Cytometry
Supplement 10
Image Processing and 2-D Morphometry The relative ease of processing and extracting information from digital images has led to a rapid growth in the use of image processing/analysis and morphometry (UNITS 10.5 & 10.10). As defined by Weibel (1979), morphometry is the quantitative description of a structure. Currently, there are two different approaches to extracting measurements from images. The model-based method commonly employed in image-analysis software programs uses algorithms to extract measurements (e.g., area, perimeter; UNIT 10.7). These morphological operations make certain assumptions about the nature of the objects being measured, and therefore have an inherent bias in the measurement process. Design-based image analysis uses modern stereological methods and probabilistic geometry to extract quantitative information from images. Data are gathered by systematic sampling and point counting, and no assumptions are made as to the size or shape of the structures to be measured (Gundersen and Jensen, 1987; Gittes, 1990; Jensen, 1991). Stereological techniques may therefore be considered an unbiased method of analysis. Each of the above approaches to analysis is valid, and depending upon the nature of the sample, one approach may be preferred or more easily used. The human visual system is unsurpassed in the ability to recognize subtle qualitative changes in an image. However, some inherent weaknesses in our visual system could result in wrong assumptions about the nature of a sample, unless analytical methods are used to view
A
UNIT 10.11
the image. The human visual system is very sensitive to changes in contrast. This may be demonstrated by the simultaneous contrast illusion and the Mach band effect. In Figure 10.11.1A the diamond on the left appears lighter than the diamond on the right, although they are exactly the same gray level. Another phenomenon observed in the gray-scale bands of Figure 10.11.1B is the appearance of first a slightly lighter band and then a darker band as the eye travels from light to dark bands, when, in reality, these lighter and darker bands do not exist. This figure illustrates that human visual perception of contrast is related to surrounding intensity levels and that at contrast boundaries there is actual visual sharpening of image detail. This sensitivity to contrast can lead to erroneous conclusions in the interpretation of scientific specimens. A good example is in the evaluation of a staining reaction using immunocytochemistry to locate epitopes in a tissue section. A sample with several focal, high-contrast stained regions may be interpreted as having more antigen present than a sample with a more diffuse and lightly staining reaction.
BASICS OF DIGITAL IMAGES Image Resolution Image resolution is measured in pixels per inch (ppi). An image with a resolution of 150 ppi contains 22,500 pixels in a square inch (150 pixels wide × 150 pixels high = 22,500). The higher the resolution, the more pixels in the
B
Figure 10.11.1 (A) Simultaneous contrast illusion. The gray diamond against a white background will appear darker than the gray diamond against a black background. (B) Mach band effect. There appears to be a lighter and then a darker band at the transition to a darker gray-scale band.
Contributed by John Turek Current Protocols in Cytometry (2000) 10.11.1-10.11.11 Copyright © 2000 by John Wiley & Sons, Inc.
10.11.1 Supplement 13
image. A 2 × 2–in. image with a resolution of 100 ppi would have 40,000 pixels. The same image with a resolution of 300 ppi would have 360,000 pixels in the same 2 × 2–in. area. To perform image analysis, it is necessary that the features of interest be captured so that the measurements are performed on images containing sufficient information to preserve a smooth object boundary. For example, a round object with an insufficient number of pixels will have a jagged edge rather than a smooth boundary.
Bit Resolution or Pixel Depth Bit resolution or pixel depth is a measurement of the amount of information per pixel. The pixel depth will determine how much color or gray-scale information is available for each pixel. Greater pixel depth means more available gray-scale levels or colors and more accurate image representation. Pixels in binary images have a depth of 1 (on or off), and these images are black and white. Gray-scale images typically have a pixel depth of 8, which means they have 256 (28) possible gray levels. For grayscale image analysis, the minimum requirement, in most instances, is an 8-bit image, but some images may need greater depth (i.e., 10bit (1024 gray scales) or higher) for accurate analysis. Digital color images are usually represented as a combination of shades of red, green, and blue pixels (i.e., RGB). The pixels in these images will have 8 bits each of red, green, and blue, which means that they have 224, or 16 million, possible colors. For color image analysis, a 24-bit image is the minimum requirement, although it is often advantageous to have higher pixel depths (e.g., 36- or 48-bit color) or to represent the color image as something other than RGB color.
Image Acquisition Digital images are commonly acquired via a video camera coupled with a frame grabber capable of converting the analog signal into a digital image, or via charged-coupled device (CCD) cameras, which may either produce a digital image directly or be coupled to a frame grabber (UNIT 2.3). The technology and hardware for creating digital images are constantly changing and evolving; digital cameras and flat-bed scanners are now virtually ubiquitous. While some cameras are capable of capturing 4K × 4K images, the majority of existing digital CCD cameras capture images that range be-
tween 1 million (1K × 1K) and 4 million (2K × 2K) pixels. When using a video camera coupled to a frame grabber or a CCD camera, it is usually best to utilize the maximum resolution available for capturing the image. The image pixels can then be resampled to a lower resolution in image-processing software if the highest resolution is not needed for image display, or if storage capacity is a limiting factor. Special care should be given to digital cameras interfaced with microscopy equipment. It is better to use the microscope optics to magnify the specimen rather than perform digital enlargement after image capture. Often, the image to be analyzed is not one acquired using a CCD camera, but rather a conventional photographic negative or print. In this instance, the most frequently asked question concerns the level of resolution (dots per inch, or dpi) needed to create the digital image and preserve specimen information. A theoretical answer to this question is provided by the Nyquist criterion for resolution, which states that to digitally preserve the resolution of detail (spatial frequency) in an image, it is necessary to sample the image at a rate twice as high as the highest spatial frequency of the detail to be preserved. In simple terms, spatial frequency may be considered to be rate of change from black to white in an image. In Figure 10.11.2A, the thin lines are closely spaced compared to the thicker lines of Figure 10.11.2B. The rate of change (i.e., spatial frequency) from black to white and returning to black is higher in Figure 10.11.2A than in Figure 10.11.2B. To explain what this means on a practical level, consider the case of an electron micrograph, which has some of the highest point-to-point resolution (i.e., spatial frequency) of any scientific image. If one examines an electron microscopic image of a section through an animal cell, some of the fine structure may include a trilaminar cell membrane and small nonmembranous organelles and cytoskeletal components such as microtubules and microfilaments. The electron micrograph of a mitochondrion in Figure 10.11.3A was digitized at 100-dpi resolution, which was insufficient to preserve the fine detail in the image, indicating that the sampling was less than that required by Nyquist’s criterion. Figure 10.11.3B was digitized at 300-dpi resolution, where the fine detail is clearly represented. The appropriate level of resolution to preserve detail for the human eye
Image Processing and 2-D Morphometry
10.11.2 Supplement 13
Current Protocols in Cytometry
A
B Figure 10.11.2 Example of different spatial frequencies. Panel (A) has a higher spatial frequency than panel (B). To preserve the detail in panel A would require a higher sampling rate than for panel B.
A
B
Figure 10.11.3 (A) Image digitized at 100 dpi. (B) Same image digitized at 300 dpi. Both images were made from a photographic print of an electron microscopic image of a mitochondrion. The sampling rate in A was insufficient to preserve the detail in the image and information was lost.
is easily determined empirically; for electron micrographs of thin tissue sections, 600-dpi resolution is usually adequate for most specimens.
IMAGE PROCESSING AND ENHANCEMENT Once an image is acquired at the appropriate sampling frequency (dpi) to preserve the details of interest, the next step is to process and enhance the image to facilitate the identification of the features that are to be analyzed or measured. It is necessary to use care at this step so as not to alter the basic characteristics of the
features to be analyzed (i.e., size, shape). One of the easiest and most frequently used methods of enhancing an image is the modification of contrast and brightness of the image by manipulation of the image histogram. An 8-bit gray-scale image will display 256 different brightness levels ranging from 0 (black) to 255 (white). All the pixels in an image may be displayed as a frequency distribution of the number of pixels versus pixel value, which is basically a graph or plot of the different grayscale pixel values in an image. An image that has pixel values throughout the entire range has
Data Processing and Analysis
10.11.3 Current Protocols in Cytometry
Supplement 13
A
0
B
255
0
255
Figure 10.11.4 Example of histogram sliding. (A) Original image. (B) Brightened image.
a large dynamic range, and may or may not display the appropriate contrast for the features of interest. However, it is not uncommon for a histogram to display most of the pixel values clustered to one side or distributed around a narrow range in the middle of the histogram. This is where the power of digital imaging to modify contrast exceeds the capabilities of traditional photographic optical methods.
Histogram Manipulation
Image Processing and 2-D Morphometry
There are a variety of software programs for manipulation of image brightness and contrast, and the user interface and methods will vary with the programs. However, all these programs perform several basic operations. The most common methods of modifying images that are overly dark or bright are histogram sliding and stretching, or a combination of both. In histogram sliding, a constant brightness is added to or subtracted from all the pixels in the image or just to/from pixels falling within a certain
gray-scale level (e.g., 64 to 128). The end result is to brighten or darken the whole or selected parts of the image. In Figure 10.11.4A, the image has some pixels throughout the entire dynamic range, but is still too dark. Sliding the histogram (i.e., lightening all the pixels) in Figure 10.11.4B produces an acceptable image. A somewhat similar operation is histogram stretching, in which all or a range of pixel values in the image are multiplied or divided by a constant value. The result of this operation is to have the pixels occupy a greater portion of the dynamic range between 0 and 255, and thereby increase or decrease image contrast. In Figure 10.11.5A, the entire image is dark, with the majority of the pixels having gray-scale values below 64 and none at the lighter levels. To produce an acceptable image in this instance it was necessary both to slide (brighten) some of the pixels and also to stretch the histogram so that the pixels were mapped to 256 gray levels (Fig. 10.11.5B). It is important to note
10.11.4 Supplement 13
Current Protocols in Cytometry
A
B
0
255 0
255
Figure 10.11.5 Example of histogram sliding with histogram stretching. (A) Original image.
that histogram stretching or sliding does not alter the inherent resolution of the image.
Gamma Adjustment The gamma of a histogram curve is the slope, expressed as a ratio of the logs of the output to input values. A gamma value of 1.0 equals an output/input ratio of 1:1, and signifies that no correction is applied (Fig. 10.11.6A). Some video cameras have a nonlinear light response which can lead to image-analysis errors, especially when dealing with densitometric measurements. In this instance, it is necessary to apply an inverse gamma function to the output lookup table to restore the proper values to the pixels (Fig. 10.11.6B). The degree of nonlinear response can be obtained from the camera manufacturer.
Removing Noise in an Image Images collected under low illumination conditions can have a poor signal-to-noise ratio. It is possible to reduce image noise by using image-averaging techniques during the image acquisition phase. Frame grabbers can capture and average multiple frames (e.g., 16 to 32 frames) to increase the image information and decrease noise. Likewise, a series of noisy individual images may be combined in the com-
puter to reduce noise. Noise may also be decreased by utilizing spatial filters. Averaging and gaussian filters reduce noise, but also cause some image blurring. Median filters that use a kernel such as a 3 × 3 or 5 × 5 to replace the central pixel luminance value with the median value of the neighboring pixels cause less image blurring. The gray specks in the radiograph of Figure 10.11.7A are an example of gross image defects. These were removed using a 5 × 5 median filter (Fig. 10.11.7B), which caused some softening, but not blurring, of image detail.
Feature Identification and Classification Following image enhancement, discrimination of the structures to be analyzed is necessary. Most image data may be classified as areas with closed boundaries (e.g., a cell), points (i.e., discrete solid points or objects that may be areas), and linear data. Objects need to be segmented and isolated from the background before analysis. A simple way of segmenting image features is to use thresholding techniques. Thresholding may be performed on gray-scale or color images. For gray-scale images of biological samples, one can usually segment an object such as a cell from the background for counting purposes. However, it is
Data Processing and Analysis
10.11.5 Current Protocols in Cytometry
Supplement 13
A
0
255 1
B
0
255 1.8
Figure 10.11.6 The gamma function. (A) A gamma function of 1. (B) A gamma function of 1.8, which brightens the pixels of the darker gray-scale bands.
visualization and analysis. In pseudocolor images, the various gray scales are mapped to different colors using an output lookup table. Because human vision is more sensitive to color than to gray scales, it is therefore easier to identify small gray-scale level differences. Capturing color digital images can provide the possibility for greater discrimination during the
often difficult to accurately segment portions of a gray-scale object based upon threshold alone. To the computer, biological samples, such as cells, are a series of nested blobs (i.e., cells, nuclei, organelles, or inclusions) that may have similar shapes and gray-scale intensities. Conversion of the gray-scale images to pseudocolor images often is useful for both feature
A
Image Processing and 2-D Morphometry
B
Figure 10.11.7 (A) Original image, showing noise (gray specks). (B) Image after applying a 5 × 5 median filter to remove noise.
10.11.6 Supplement 13
Current Protocols in Cytometry
intensity
white
saturation
hue 0°
Figure 10.11.8 An HSI image may be represented as a double cone. The cone axis is intensity and ranges from black to white. The radius of the cone is the degree of color saturation, and the perimeter is the color hue. See color plate. black
Figure 10.11.9 Top: RGB images. The area of interest contains some diffuse blue staining that could not be segmented by threshold (shown by arrows). Bottom: Converting the image to HSI and then thresholding only the saturation band (lower left) results in an accurate discrimination of the area of interest. See color plate.
10.11.7 Current Protocols in Cytometry
Supplement 13
Figure 10.11.10 Identification of areas of interest by the computer. Because of boundaries touching one another (arrows), an inaccurate count will result (see Fig. 10.11.11).
Image Processing and 2-D Morphometry
threshold process. Image segmentation may be based upon red, green, and blue (RGB) values in the image. However, it is often difficult to differentiate between different shades of the same color in RGB format. A more powerful method is to use hue, saturation, and intensity (HSI; Fig. 10.11.8). RGB images may be converted to HSI in various software programs. The HSI method of color discrimination is similar to the way the human brain interprets color. HSI can be represented as a double cone with hue (i.e., wavelength of light reflected from or transmitted through an object) being a color wheel at the base of the cones. Saturation is any point along the radius of the cone and represents the amount of gray relative to hue (0% = gray, 100% = fully saturated). Intensity is the vertical axis of the cones and represents relative lightness or darkness (0 = black, 100 = white). The power of HSI compared to RGB format for image segmentation is illustrated in Figure 10.11.9. In the top panel of Figure 10.11.9, the region of interest for measurement (bounded by arrows) contains diffuse blue staining. In RGB mode it is impossible to segment the region using threshold because of the spillover into other areas. In the lower panel, the image was converted to a 48-bit HSI image and thresholding performed only on the degree of color saturation in the image. This resulted in successful identification of the region of interest. Another method to both enhance image features and segment the image is to use edge
detection. Filters for edge detection, such as Laplace and Sobel, are often useful prior to object counting, but if quantitative measurements of size are needed these filters can induce a small amount of error. Edge-detection filters emphasize gradients and increase the transition between luminance values. However, these filters may still leave some objects with discontinuous boundaries and will fail to separate objects that touch each other. Algorithms may be used to connect together nearby edges, and watershed filters may be used to separate touching objects. One of the common procedures in image analysis is object counting. If a digital image is displayed in any of a variety of image-analysis software programs, one has the option of identifying a region of interest (ROI) for analysis. In most circumstances it is important that the area chosen for analysis be derived via systematic sampling, and not because the area has the features of interest. The entire image may be the ROI, or a polygonal or irregular region may be specified. The bounding box in Figure 10.11.10 is the ROI. If a threshold is set on the gray-scale value of the blobs in the image, the areas can then be identified for analysis. However, because some blobs are touching each other, the analysis program may see several blobs as a single object and report an inaccurate count. When performing object counting it is often best to exclude areas touching the region of interest, i.e., bounding box, because these may be partial
10.11.8 Supplement 13
Current Protocols in Cytometry
Figure 10.11.11 Excluding areas touching the region of interest, the computer incorrectly identifies only 14 objects owing to boundaries touching one another.
profiles. After excluding such blobs in Figure 10.11.10, indicated by arrows, the computer inaccurately identifies 14 objects (Fig. 10.11.11). To separate the blobs and obtain an accurate count, morphological operations, such as erosion and dilation, and watershed filters, are used. Mathematical morphology provides a method to process an image based upon shape features. Erosion and dilation are complementary functions. In erosion, a structuring element such as a 3 × 3 pixel array may be used as a
probe to contract or reduce foreground features. With the erode function, individual objects will contract away from surrounding features, and holes or gaps between objects will increase in size. The dilate function produces the opposite effect, where image features in the foreground will grow and expand. Simple erosion and dilation are iterative processes that require user interaction and judgment to determine visually when the objects are separated. A more rapid method is to use a watershed filter. Watershed
Figure 10.11.12 After converting the image to binary format and applying a watershed filter to separate object boundaries, 22 objects are identified.
Data Processing and Analysis
10.11.9 Current Protocols in Cytometry
Supplement 13
Figure 10.11.13 Objects can also be counted in an unbiased manner by using a counting frame. Objects that touch the solid line or fall outside the boundary of the dashed line are not counted. This method may be used to determine numerical density (objects/area).
filters utilize erosion and dilation in combination with an algorithm to identify and separate object boundaries. During a watershed operation, each successive erosion of a foreground object may be thought to represent a higher elevation, as in a topographical map. The lower elevations of each peak in the map represent the boundary, or watershed. Once the “peaks” are identified, a dilation operation is then imposed
Image Processing and 2-D Morphometry
on the image, but the algorithm prevents the boundaries of the individual peaks from touching each other. For our object-counting example (Fig. 10.11.10), the image is converted to binary (i.e., black and white) bitmap format and a watershed filter applied. One can now identify 22 objects, excluding those touching the edge of the region of interest (Fig. 10.11.12).
Figure 10.11.14 A point counting grid with a counting frame may be used to determine the area of objects. If an object or its boundary falls in the quadrant to the upper right of the counting point (+) it is tallied.
10.11.10 Supplement 13
Current Protocols in Cytometry
After segmentation and counting, it is often necessary to further classify objects for measurement. One useful parameter is shape. The shape factor of an object is defined as (4π × area)/perimeter2. A perfect circle will have a shape factor of 1. Departure from circularity (e.g., oval or irregular border) will lower the shape factor below one. The inverse relationship, perimeter2/(4π × area), has also been called compactness. Shape can also be expressed as the perimeter2/area. With this formula, a circle has a minimum value of 4π and noncircular objects have larger values. Shape is a useful criterion for distinguishing cancer cells from normal cells. Cancer cells and/or their cell nuclei are usually more irregular and less circular than normal cells. In addition to shape and size, features such as minimum and maximum diameter, perimeter, area, centroid, and orientation in an X-Y plane are useful for some samples. As mentioned already, there are two approaches to image analysis. The methods previously described apply to model-based systems that use algorithms to extract the quantitative data. Image analysis based upon modern stereological methods utilizes systematic sampling and point counting to get morphometric data, and no assumptions are made as to the size or shape of the objects. Using the unbiased techniques of modern stereology (Gundersen, 1988; Gundersen et al., 1988a,b; Gittes, 1990), a counting frame is placed over the image previously used for counting objects (Fig. 10.11.13). Objects completely within the counting frame and those that touch the upper and right dashed lines are tallied, while profiles touching the solid lines are excluded. The result is a tally of 23 objects. If the size of the counting frame is known, the number of objects may be expressed as numerical density (i.e., number of objects/frame area). If we assume arbitrarily that the frame represents 100 µm2, then the numerical density (QA) = 23/100 µm2 = 0.23 objects per µm2. Stereological methods may also be used to determine the size or area of the objects. If a counting frame with regularly spaced points is placed on the image (Fig. 10.11.14), then the mean area fraction of the objects may be determined. An object is counted if it falls within the quadrant to the upper right of one of the counting points (+) on the grid. The objects falling under the regularly spaced points are counted with no regard to their relationship to the counting frame. The area fraction (AA) is equal to the number of points hitting objects/total number of
points. For Figure 10.11.14, AA = 60/88 = 0.682. The 60 points are circled in the figure. This value may be used to determine the mean object area provided that the frame area is known. The mean object area (a) is equal to AA/QA, where QA is the relative number of objects. For Figure 10.11.14, QA = 0.23/µm2; therefore, the mean area (a) of the objects is 0.682/0.23 = 2.96 µm2.
LITERATURE CITED Gittes, F. 1990. Estimating mean particle volume and number from random sections by sampling profile boundaries. J. Microsc. 158:1-18. Gundersen, H.J.G. 1988. The nucleator. J. Microsc. 151:3-21. Gundersen, H.J.G. and Jensen, E.B. 1987. The efficiency of systematic sampling in stereology and its prediction. J. Microsc. 147:229-263. Gundersen, H.J.G., Bagger, P., Bendtsen, T.F., Evans, S.M., Korbo, L., Marcussen, N., Moller, A., Nielsen, K., Nyengaard, J.R., Pakkenberg, B., Sorensen, F.B., Vesterby, A., and West, M.J. 1988a. The new stereological tools: dissector, fractionator, and point sampled intercepts and their use in pathological research and diagnosis. Acta Pathol. Microbiol. Immunol. Scand. 96:857-881. Gundersen, H.J.G., Bendtsen, T.F., Korbo, L., Marcussen, N., Moller, A., Nielsen, K., Nyengaard, J.R., Pakkenberg, B., Sorensen, F.B., Vesterby, A., and West, M.J. 1988b. Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. Acta Pathol. Microbiol. Immunol. Scand. 96:379-394. Jensen, E.B. 1991. Recent developments in the stereological analysis of particles. Ann. Inst. Statist. Math. 43:455-468. Weibel, E.R. 1979. Stereological Methods. Practical Methods for Biological Morphometry. Academic Press, London.
KEY REFERENCES Jähne, B. 1993. Digital Image Processing: Concepts, Algorithms, and Scientific Applications, 2nd ed. Springer-Verlag, New York. General textbook on digital imaging covering more advanced concepts. Robinson, J.P. and Turek, J.J. 1999. Microscope image processing and analysis. Encyclopedia of Electronics and Engineering, Vol. 13 pp. 9-21. John Wiley & Sons, New York. General basic reference for digital imaging. Includes background on confocal and electron microscopy. Serra, J. 1983. Image Analysis and Mathematical Morphology. Academic Press, London. Foundational text for understanding morphological operations in digital imaging.
Contributed by John Turek Purdue University West Lafayette, Indiana
Data Processing and Analysis
10.11.11 Current Protocols in Cytometry
Supplement 13
Dial-In Flow Cytometry Data Analysis UNIT 10.1 presents arguments for facility-wide management of flow cytometry data and UNIT 10.3 further weighs the advantages and disadvantages of data storage in listmode. An obvious disadvantage of listmode storage is, of course, the burgeoning size of the data files, which in modern times contain more measured parameters, at greater resolution and for greater numbers of cells per file. Although current computer hardware can cope with the storage space required, there are problems with accessing such large files via any kind of network connection because of the enormous traffic that can be generated. Furthermore, for those connecting via telephone modem, the limited speed of transmission all but precludes analysis of any but the smallest data files. Whether this access is by way of a file transfer protocol (FTP) or by reading the files across the network using a conventional flow cytometry data-analysis program makes no difference to the speed or volume of the transmission; in both cases the entire file must be transmitted. This unit describes a way around this traffic problem that arises from a consideration of the methods inherent in flow cytometry data analysis and display. Most commonly, 1-dimensional (1-D) and 2-dimensional (2-D) arrays are extracted from the data and displayed as histograms, contour maps, or dot plots. This is not to say that the bulk of the data in the original file plays no part in the display; in fact it may all play a part. A 2-D display of two measured parameters may be extracted from a data file by selecting data from just those cells having prescribed characteristics of some or all of the nondisplayed parameters (the process called “gating”). However, the extracted 1-D or 2-D array can be described using very few numbers. For example, a 1-D histogram is usually described in 256 channels; a 2-D histogram, generically referred to as a “contour” display whether or not actual contour lines are drawn, can be satisfactorily described in as few as 64 × 64 channels (i.e., 4096 numbers). Thus, the network traffic problem would be greatly relieved if only this display array were transmitted. For maximum usefulness, this should be achieved without restricting the user’s immediate and interactive control over the display format and the gating of the listmode data. A solution that fulfills these criteria involves the installation at the central storage facility of
Contributed by Francis L. Battye Current Protocols in Cytometry (2002) 10.12.1-10.12.7 Copyright © 2002 by John Wiley & Sons, Inc.
a small computer program called a Web servlet, operating in concert with a Web server, which assists the analysis by extracting the display array from the data file and then organizing its transmission over the network to a remote client program that creates the display. This solution may be contrasted with another proposed by Olivares et al. (1998). In a proposal to enable public access to a central flow cytometry data repository, Olivares et al. suggested a Web site to which users may submit requests for analysis of selected groups of files. Requests would be processed by an off-line analysis program and the results viewed by a Web browser. That scheme would remove the necessity of reading files across the network but would also remove the user’s immediate and interactive control over the format of the displayed results and would prevent side-by-side or overlaid comparisons of local and remotely stored data. An implementation of the solution described here has recently been reported (Battye, 2001). The operational starting point was a general-purpose flow cytometry data-analysis program, WEASEL (http://www.wehi.edu.au/ cytometry/WEASEL.html). WEASEL generates the standard display types—1-D histograms, 2-D and 3-D scatter plots (“dot plots”), and 2-D histograms—which may be displayed as probability densities (Moore and Kautz, 1986), standard contour plots, or 3-D perspective “hills and valleys” displays. It will extract numerical data like mean or median fluorescence or coefficients of variation. The program is written in the Java computer language as a freestanding application, thus allowing it to run on workstations under Macintosh, Windows, Unix, and other operating systems. More importantly in the present context, the Java language contains application program interfaces (APIs) for communication with Web servers and these can be built into the application (http://java.sun.com/nav/whatis/). Java was also used for creation of the Web servlet, called FCSServlet, which could accept requests for display data and extract it from flow cytometry standard (FCS) data files. At the same time, WEASEL was modified to enable it to hold a dialog with FCSServlet. Communication between the WEASEL program and FCSServlet is hidden from the user, to whom the process appears identical with that of reading a locally stored file. The
UNIT 10.12
Data Processing and Analysis
10.12.1 Supplement 19
typical operations of first reading an ungated file, then drawing regions on the relevant 1-D or 2-D displays, constructing a gate as a logical expression involving those regions, and applying it to the original or subsequently read files all proceed in apparently the same way whether the files are local or remote. Frequent reference will be made to this implementation in describing the principle of servlet-assisted data analysis. New implementations incorporating this principle may be constructed by any competent computer programmer with or without specific Java language experience and using only standard or freely available programming tools as documented below.
COMMUNICATION BETWEEN THE ANALYSIS PROGRAM AND THE SERVLET The setup of the dialog between the analysis program and servlet proceeds in 3 steps. 1. The analysis program connects with the servlet and requests a list of available data files. If the request is recognized, the servlet responds with the data file list that the analysis program then displays for its operator in the form of a file dialog. The servlet also sends a client identification number that must be quoted by the analysis program in all subsequent requests. The operator selects a file from the directory list presented. 2. The analysis program requests the file header and text or keyword portion for the selected file. Header and text portions of FCS files may be as small as ≤500 bytes or >5000 bytes, e.g., Becton Dickinson ConsortVax files (Seamer et al., 1997; UNIT 10.2). The servlet interprets the header and text section, extracting parameter information (e.g., number of parameters, bit size, range) that is stored for later reference, and then transmits the text in its entirety. The analysis program receives and interprets the text, displaying a format dialog from which the operator selects parameters, gating, and display format. 3. The analysis program constructs and transmits a request containing display type, selected parameters, and gating, including specification of any polygonal or rectangular regions that have been drawn by the operator and that are required for the gate. The servlet reads the data file but transmits only a compressed display array. Dial-In Flow Cytometry Data Analysis
DATA COMPRESSION Not only is the display array likely to be much smaller than the original file, the data, by nature, also lend themselves to further compression.
Integer Arrays While histograms (1-D or 2-D) normally hold integer values, these values never require the full 4 bytes of a Java integer to define them. Most often the content of each channel will be <256 cells and such a value can be encoded as 1 byte. In particular, since 2-D histograms are often quite sparse, there are many repeated zero values. A data compression scheme that was simple to implement, involved negligible computer processing overhead in data encoding and decoding, and was especially suited to flow cytometry data was devised for this application (Battye, 2001). The scheme required assignment of a default number of bytes for expressing each integer value. Values that required more than this default number of bytes were flagged and the full value was encoded. Long runs of repeated values were encoded as the value followed by a special flag byte and the number of repeats. Thus, for integers, substantial compression was achieved without any loss of precision.
Floating-Point Arrays Probability density displays have two characteristics: a smoothing of the data to a degree dependent on the local density of cells and a choice of contour levels that encloses, between adjacent lines, fixed percentiles of the total population. Arrays of such smoothed data are best expressed as floating-point numbers. These are difficult to compress without loss. However, by assigning some minimum precision, the normal 4-byte floating-point numbers can easily be trimmed to 3 bytes. This still gives ∼5 decimal digits of precision for each data point, which is ample given that the data encode a visual display only. Long sequences of numbers that are repeated to within the assigned precision can be specially encoded as for the integers. Although more sophisticated techniques may be found that will give marginally better compression, the advantage of this method is its simplicity.
Dot Plots Dot-plot displays result from the plotting, often in chronological order, of individual dots, each reflecting data from one cell, in a 2-D panel or pseudo 3-D cube. This display form
10.12.2 Supplement 19
Current Protocols in Cytometry
may be criticized because of the obscuration of earlier dots by those subsequently plotted. However, it has the great advantage of allowing color coding of each dot to reflect some additional feature of the cell it represents (i.e., magnitude of a measured parameter, cluster membership, or enclosure within a defined region). Dot-plot resolution can be limited to 256 channels in order that the dot coordinates and color code each can be transmitted as 1 byte. Thus, the net effect of this scheme of data extraction by servlet on reducing the data transmission is the product of (1) the reduction from the file size to the required data array size, with (2) the additional reduction resulting from compression of the data.
COMPUTER SOFTWARE CONSTRUCTION Although more sophisticated programming tools are available, e.g., Code Warrior (Metrowerks), Java language code may be written using any simple text editor. That code then may be compiled and packaged on Windows PCs, using tools in the Java Development Kit available from Sun Microsystems’ Java Web site (http://java.sun.com/j2se/), or on a Macintosh using tools available from the Apple Web site (http://developer.apple.com/java/). Also available at those Web sites are the Java Runtime Environments to allow execution of the resultant program on each respective target system. For programmers with no prior Java language experience, an excellent source of help for construction of the software components is the Java tutorial on the Sun Microsystems Web site (http://java/sun.com/docs/ books/tutorial/).
Analysis Program Construction The analysis program should have the usual displays and controls already familiar to operators. There must be space in a “properties” or “preferences” dialog window to allow operator entry of a network address or universal resource locator (URL) for the remote servlet. Thereafter, the dialog between it and the remote servlet should be invisible to the operator. In functional terms, the modifications, compared to a conventional data analysis program, should include duplication of the file reading and data extraction section, as shown diagrammatically in Figure 10.12.1. In terms of the Java language, a typical transaction proceeds as follows: 1. Open a connection to the servlet URL.
2. Set the connection to allow output. 3. Create an ObjectOutputStream for that connection. 4. Write an object containing a request (for header information or for a data array) to the connection. 5. Create an ObjectInputStream for the connection. 6. Read the response object (containing the requested data). For each session, the remote servlet assigns an identification number that is stored by the analysis program and is quoted in each transaction. The file specifications for all displays are also numbered in agreement with those used by the servlet (http://java.sun.com/docs/books/ tutorial/networking/urls/index.html).
Servlet Construction The servlet is an extension of the Java HttpServlet class. To handle the special flow cytometry data requests from the analysis program, the doPost(request, response) method is replaced. The servlet side of the transaction is: 1. Create an ObjectInputStream for the request. 2. Read the request object. 3. Check the request’s identification and read the requested data from the file repository. 4. Create an ObjectOutputStream for the response. 5. Write the response object. Since the servlet may handle multiple remote sessions simultaneously, an identification number is assigned for each and must be quoted within each data request. Also, since each session may operate on multiple files, the files are also given numerical identifiers in agreement with the numbers used by each analysis program. As each file is read, a data structure containing the interpreted header information is created and given its identifier. This avoids multiple reading and interpreting of the FCS text. As each session closes, its files’ data structures can be disposed of (http://java.sun.com/ docs/books/tutorial/servlets/index.html).
Deployment The application program can be installed on any workstation for which a Java runtime environment is available. For testing the application and servlet on a Windows PC, the Java Servlet Development Kit (JSDK2.1; http://java.sun.com/products/
Data Processing and Analysis
10.12.3 Current Protocols in Cytometry
Supplement 19
A analysis program user interface menu system file selection dialog display format dialog
utility functions read directory interpret FCS text extract display array create display
directory header listmode
file repository
local or remote network connection
B
ry
analysis program utility functions read local directory or receive remote directory interpret FCS text from local file or from remote file extract display array from listmode or receive array from servlet create display
local file repository
o ct re er i d d a he de directory o tm il s header
display array remote network connection
servlet read directory interpret FCS text extract display array
central file repository
Figure 10.12.1 (A) Computer software layout for conventional flow cytometry analysis. Whether files are locally or remotely stored, construction of any display requires reading of the entire listmode data across the connection. (B) For servlet-assisted data analysis, sections of the analysis program that are involved with file reading are duplicated. Standard methods for reading locally stored files are retained but are augmented by methods for communication with a remote servlet operating at the central file repository computer system. The servlet extracts the required display arrays and transmits them over the network connection.
servlet/archive.html) can be used, with the included server providing the function of a servlet container. Procedures for final servlet deployment will depend on the Web server used. Further information is available from http://java.sun.com/products/servlet/industry. html. Users of the popular Apache Web server may o btain the Tom cat ad d-on (http://jakarta.apache.org/tomcat/). Tomcat can also be used in standalone mode, which is particularly useful for testing and debugging.
ANTICIPATED RESULTS
Dial-In Flow Cytometry Data Analysis
To test the efficacy of this scheme, two typical data files have been used for model transmissions. For comparisons, the transmission times that would be experienced by a remote operator using a hypothetical 28,000 bits/sec modem are calculated. Since transmission times are inversely proportional to data transfer rates, they may be translated readily for
faster or slower network connections. Actual transfer rates will be affected by server performance, network traffic, and telephone line quality as well as by the theoretical maximum rates of the modems at the local and remote sites. The time required for file reading and data array creation is not calculated because this is not an additional task but merely one moved from the local to the remote computer. Data compression and decompression times, on the order of tens of milliseconds at most, are negligible in comparison to transfer times. The first file tested (file 1) is from an analysis of 10,062 cells for which eight parameters, including four immunofluorescence parameters, were measured; then a ninth parameter, representing the result of an offline cluster analysis, was added later. Table 10.12.1 summarizes results for representative display types 1-D histograms (256 integers), 2-D histograms (64 × 64 integers), 2-D probability density
10.12.4 Supplement 19
Current Protocols in Cytometry
Table 10.12.1 Comparison of the Transmission Times for File 1 by File Transfer Protocol (FTP) and for Data Arrays Extracted from the File by Web Servleta
Transmit type Data values FTP 256 integers 64 × 64 integers 64 × 64 float Dot plot
Nominal bytes
Transmitted File Data Total Transmission bytes compression compression compression time (sec)
10,062 cells 181,116 256 1024 4096 16,384
181,116 243 1460
1.0 176.9 11.1
1.0 4.2 11.2
1.0 745.3 124.1
64.7 0.1 0.5
4096 30,186
9932 30,186
11.1 6.0
1.6 1.0
18.2 6.0
3.5 10.8
16,384 30,186
aNominal bytes are calculated assuming 4-byte integers and floating-point numbers. File compression is given by file size divided by nominal bytes of the transmitted array. Data compression (nominal bytes/transmitted bytes) is the further compression obtained using the techniques described. Transmission time is calculated for a hypothetical 28,000 bits/sec modem.
displays (64 × 64 float), and dot plots. Transmission times shown are just for the data component. The contribution of the header and text information to the total transmission is discussed below. Only 243 bytes of data need to be transmitted, in <0.1 sec, for the single-parameter 1-D histogram of a chosen immunofluorescence parameter. Because the transmission time for one parameter is so short, the servlet was designed to send histograms for all parameters in each transmission, thereby allowing the operator the flexibility of displaying any or all without further access to the data file. For the example of nine-parameter data tested, all nine single-parameter histograms could be encoded in 1790 bytes that could be transmitted in 0.6 sec. Data for a 64 × 64 2-D array required 0.5 sec. Transmission of a 64 × 64 floating-point probability density array using the same parameters took more than three-fold longer than the integer data, firstly because each data point required 3 bytes, and secondly because, after smoothing, there are fewer long runs of zero values than for the comparable integer array. Nevertheless, the 3.5 sec required is still acceptable. Data for a color-coded dot plot are shown for transmission of 10,062 dots, perhaps more than may be required for some purposes. Once the data array is received, it may be replotted by the local application in a number of formats. Also, the statistics markers, lines or quadrants, may be shifted at will for extraction of numerical information from the data array without further access to the data file. There are cases where the 64 × 64–channel 2-D displays do not offer sufficient resolution. One example is time-course measurements, in which elapsed time is stored as an FCS parame-
ter. The results summarized in Table 10.12.2 are from an analysis of intracellular Ca2+ as determined using the ratiometric dye indo-1 (file 2). In this case, a satisfactory result may be obtained from a 2-D integer array of 256 × 256 channels representing elapsed time versus indo-1 fluorescence ratio. Using just that transmitted data, it is possible for the analysis program to calculate and display the mean fluorescence ratio for all cells received at each time channel without plotting any contour lines. This gives a clear picture of the response of the whole sample over time. However, in that particular experiment, this did not show the full picture since not all cells were responding to the stimulus, and plotting of the contour lines was required to show responding and nonresponding populations. Because of the nature of these fairly widely dispersed data with 150,276 cells distributed over 65,536 channels, the results could be shown more clearly and with less visual “noise” by using a probability density display that required transmission of a smoothed floating-point array rather than an integer array. Transmission time for those floating-point data is rather long but still 26-fold less than for transmitting the entire file. Again the data, once received, could be replotted in other formats. Tables 10.12.1 and 10.12.2 show the extent of the reduction in the quantity of transmitted data due to extraction of data arrays (file compression) and that due to the compression techniques (data compression). The numbers shown omit the contribution to the transmission time of the text or keyword portion of the FCS data files, because much of this section may be unnecessary to the analysis. However, for the WEASEL program, the entire keyword section
Data Processing and Analysis
10.12.5 Current Protocols in Cytometry
Supplement 19
Table 10.12.2 Comparison of the Transmission Times for File 2 by File Transfer Protocol (FTP) and for Data Arrays Extracted from the File by Web Servleta
Transmit type Data values
Nominal bytes
FTP
150,276 cells 65,536
4,508,280 4,508,280
1.0
1.0
1.0
1610.1
262,144
38,308
17.2
6.8
117.9
13.7
65,536
262,144
173,239
17.2
1.5
26.1
61.9
256 × 256 integers 256 × 256 float
Transmitted File Data Total Transmission bytes compression compression compression time (sec)
aNominal bytes are calculated assuming 4-byte integers and floating-point numbers. File compression is given by file size divided by nominal
bytes of the transmitted array. Data compression (nominal bytes/transmitted bytes) is the further compression obtained using the techniques described. Transmission time is calculated for a hypothetical 28,000 bits/sec modem.
was indeed transmitted so that all keywords could be available to the operator for annotation of the displays as desired. Hence, for file 1, a ConsortVax file, 5023 bytes of FCS text were transmitted in 1.8 sec via 28,000 bits/sec modem. For file 2, a Cyclops Summit file, 1352 bytes required <0.5 sec. The time required for transmission of information from the application program to the servlet also has been ignored because it will always be quite short. In step 2 of the communication dialog described above, the application program sends only a folder and file name string. In step 3, the largest components of the transmission are region specifications, each of which will amount to only a few hundred bytes even for a quite intricately drawn region, and optionally a fluorescence compensation matrix, also not larger than a few hundred bytes. In addition to data arrays for graphical display, the servlet can also extract numerical statistics from FCS data files. Typically these will be requested by the operator in the context of a 2-D display and may include mean, median, peak position, and coefficient of variation for the horizontal and vertical parameters. They may be calculated for subsets of the cells that are enclosed within any of the defined regions or gates. In any case, the number of values requested will rarely exceed 100 or 200 floating-point (4-byte) numbers with a total transmission time of ≤0.3 sec.
CURRENT AND FUTURE IMPLICATIONS Dial-In Flow Cytometry Data Analysis
This method has been shown to be very effective at reducing the data transmission times for the particular styles of information
extraction and for the large data files that are both characteristic of flow cytometry analysis. Multiparameter listmode files always contain more information than can be conveyed by any single display, and the now common technique of showing 1-D or 2-D projections of a data space that is always of much larger dimension is very well served by this Web servlet data extraction concept. A further important point is that the transmission of data arrays leaves the client program free to choose the display format. Once data are received, dot plots may be plotted with or without color coding and a subset of the total number of dots may be selected if the trends are thereby made clearer. Plotting of 2-D-histogram integer arrays may be done as either pseudo-color or contour displays with free choice of the number and spacing of the contour lines. These and their smoothed, floating-point counterparts may also be plotted as 3-D perspective displays that may be animated. Thus, this system greatly speeds access to remotely stored data while still retaining the flexibility of data manipulations normally associated with local access. In fact, building it into an existing general-purpose analysis program enables the side-by-side display and comparison of local and remote data. Moreover, the fact that different processes are used to read remote and local files may be completely hidden from the operator. It is the practice at some institutions to implement some form of database storage, classification, and retrieval for their flow cytometry data. Of course, it is possible to view the identification and location of data as a separate issue from that of reading and extracting the stored
10.12.6 Supplement 19
Current Protocols in Cytometry
information. However, since techniques are well established for querying databases using Java-language computer programs (Hamilton et al., 1997), the possibility arises of building the database connectivity into the very same client program that produces the flow cytometry displays. In any case, there should be no impediment to the implementation of a system similar to that described here, with the same savings of network transmission time, at any site.
http://java.sun.com/j2se/
LITERATURE CITED
http://java.sun.com/docs/books/tutorial/ networking/urls/index.html
Battye, F.L. 2001. Web servlet-assisted, dial-in flow cytometry data analysis. Cytometry 43:143-149. Hamilton, G., Cattell, R., and Fisher, M. 1997. JDBC Database Access with Java: A Tutorial and Annotated Reference. Addison Wesley Longman, Reading, Mass. Moore, W.A. and Kautz, R.A. 1986. Data analysis in flow cytometry. In Handbook of Experimental Immunology, 4th edition (D.M. Weir, ed.) Vol. 1, Chpt. 30. Blackwell Scientific, Oxford.
Source for downloading software development kits and runtime environments for Java on Windows, Solaris, and Linux operating systems. http://developer.apple.com/java/ Source for downloading software development kits and runtime environments for Java on Macintosh OS X or Classic operating systems. http://java.sun.com/docs/books/tutorial/ Sun Microsystems’ Java tutorial dealing with all aspects of Java Programming.
Section of Sun Microsystems’ Java Tutorial dealing with the way applications may connect to and read from network sites defined by universal resource locators (URLs). http://java.sun.com/docs/books/tutorial/servlets/ index.html Section of Sun Microsystems’ Java Tutorial dealing with servlets.
Olivares, G., Habbersett, R., and Coder, D. 1998. The National Flow Cytometry Resource Cytometry Reference Database. Cytometry Supplement 9:57.
http://java.sun.com/products/servlet/archive.html
Seamer, L.C., Bagwell, C.B., Barden, L., Redelman, D., Salzman, G.C., Wood, J.C.S., and Murphy, R.F. 1997. Proposed new data file standard for flow cytometry, version FCS 3.0. Cytometry 28:118-122.
http://java.sun.com/products/servlet/industry.html
INTERNET RESOURCES http://www.wehi.edu.au/cytometry/WEASEL.html Description of the features and operation of the WEASEL flow cytometry data analysis program. http://java.sun.com/nav/whatis/ A guide to what the Java platform is and how it can be used.
Site for downloading the Java servlet development kit.
Documents the support for Java servlets among many popular Web servers. http://jakarta.apache.org/tomcat/ Description of the Tomcat addition to the Apache Web server that is used as a servlet container.
Contributed by Francis L. Battye The Walter and Eliza Hall Institute of Medical Research Melbourne, Victoria, Australia
Data Processing and Analysis
10.12.7 Current Protocols in Cytometry
Supplement 19
The Application of Data Mining to Flow Cytometry INTRODUCTION TO DATA MINING Data mining is defined as automating the process of information discovery (Cabena et al., 1998). This is also known as knowledge discovery in databases (KDD). While tools exist to facilitate access and query on data, more work is needed to detect useful patterns, correlations, and trends in existing data (Parsaye and Chignell, 1993). Data mining is the process of automating this information discovery. An essential element of data mining is data modeling (UNIT 10.7). Existing data need to be fitted into a representative model upon which useful information can be detected based on a variety of different algorithms. The data-mining algorithms generally group or categorize existing data into distinct sets using various attributes. A variety of algorithms have been utilized, ranging from neural net to statistics. Only a few representative algorithms will be described in this unit. Central to the increasing use of data mining is recent advancement in computational hardware. The high cost of mainframe computers for data-mining projects in the past has limited their use to a very small number of large corporate users. By contrast, most data-mining tools today can run on high-end personal computers. Data mining also requires access to large amounts of data stored online. Now data are more easily stored than ever before with large hard disk drives at very low cost.
DATA MINING AND FLOW CYTOMETRY Large amounts of data are routinely generated from flow-cytometric analysis. The database management system (DBMS) of a typical flow cytometric instrument contains all the re-
study definition
data preparation (training data)
UNIT 10.13
sults for analyzed cases, which can be used as the data source for the data-mining process. Flow cytometry has been applied to many diverse areas from nucleic acids to immunophenotypes and microbiology. It would be impractical to cover data mining in all areas of flow cytometric analysis. The scope of this unit will be limited to application of data mining to immunophenotyping of hematologic neoplasms. Nevertheless, the general concepts and techniques of data mining discussed here should be applicable to all other areas of flow cytometric analysis. A few potential areas of application of data mining to immunophenotyping by flow cytometry are suggested in this unit: 1. Development of a diagnostic module to offer differential diagnosis for a new case based on its marker results. 2. Detection of useful diagnostic criteria for certain types of neoplasm. For example, stain intensity of a certain marker (dim versus bright) may facilitate more accurate diagnosis of neoplasms with an overlapping pattern of positive markers. 3. Profiling patient prognosis. For example, data mining may detect certain specific markers that are characteristic of patients with a poorer prognosis based on prognostic data of patients with a certain neoplasm. 4. Discovery of undesirable quality control (QC) trends in marker results. For example, technical problems that cause false-negative or false-positive results can be detected by comparing marker results with expected results for a large number of cases. Note that the process of information discovery is typically based on vast amounts of historical data. Practicing flow cytometrists often
data modeling
validation (validation data)
iterations prediction (application data)
Figure 10.13.1 Five steps in data mining.
Data Processing and Analysis
Contributed by Andy N.D. Nguyen
10.13.1
Current Protocols in Cytometry (2002) 10.13.1-10.13.6 Copyright © 2002 by John Wiley & Sons, Inc.
Supplement 20
notice some interesting trends in analytical results. However, these observations are often ad hoc in nature and cannot be definitely confirmed without a systematic approach in analyzing a large number of cases (McDonald et al., 1998).
THE DATA-MINING PROCESS While details of data mining vary among different techniques (Adriaans and Zantinge, 1996), there are generally five key steps (Fig. 10.13.1): Study definition Data preparation Data modeling Validation Prediction. One of the potential areas of application mentioned previously, development of a diagnostic module to offer differential diagnosis, will be used for discussion purposes.
Study Definition This involves articulating a goal, selecting an output that characterizes that goal, and specifying the attributes that are needed to mathematically describe that output (PiatetskyShapiro, 1996). The example project chosen for this unit focuses on development of a diagnostic module for a new hematologic neoplasm case. The output is obviously the diagnosis (or a small set of differential diagnoses) with attributes that closely match the marker data of the case. The essential attributes needed to describe a diagnosis should contain a textual description of the diagnosis (for example, “follicular lymphoma”) and the result value (positive versus negative) of all markers used in flow cytometric analysis. Other attributes useful to identify a diagnosis include stain intensity of a marker (“dim” versus “bright”), histological features (e.g., “nuclear size”), and cytogenetic results. The number of attributes used in analysis is known as “dimension.” Typically, more attributes than necessary are specified in the beginning of a project. Subsequent steps in data modeling and validation should help filter out attributes that do not provide any useful input for analysis.
Data Preparation
The Application of Data Mining to Flow Cytometry
This process converts raw data into a usable set for analysis. Several important points for data preparation in a data-mining project must be considered, as described below.
Data cleaning The raw data are usually not “clean.” For instance, final diagnosis in an anatomical report may not be stored in the database in a consistent manner (e.g., “mantle cell lymphoma” versus “mantle zone lymphoma”). This is known as variance in terminology. Further data manipulation is required to convert the raw data into another data set with more consistent naming structure. Other data quality issues to be resolved in the data preparation process include redundant data, incorrect or inconsistent data, and typographical errors. Data merging In order to analyze data coming from a variety of different sources (marker results, final anatomical diagnosis, cytogenetic results, etc.), all the data have to be merged into a single database for analysis. This task seems straightforward, but, in reality, may be a very difficult one if the data are stored in different formats. Data-conversion tools are available to combine tables in different formats and from different databases into a resultant table. Three different types of data sets will eventually be used in a data-mining project: the first data set for constructing the model (training data set), the second one to validate the model (validation data set), and the third one to make a prediction using the constructed model (application data set).
Data Modeling Once a study has been defined and data are prepared, a data-mining tool reads the training data set and attempts to construct a mathematical model. While all models vary depending on the underlying techniques (see discussion of Data Mining Algorithms, below), the key concept is essentially the same. The data-mining tool scans vast amounts of data and fits them into parameters in the mathematical model.
Validation The completed model must go through a rigorous validation process using the validation data set. Output for the cases in the validation data set will be generated using the completed model and compared against the actual known outcome available in the validation data set. The performance of the model can be evaluated using various criteria specified by the user. If the model does not perform as expected, it will be modified accordingly and the whole process of data modeling and validation will be repeated until a satisfactory model is obtained.
10.13.2 Supplement 20
Current Protocols in Cytometry
The data-preparation step may also be modified to obtain a more adequate training data set, if needed, for further iterations. In the example project, the diagnostic model will be constructed with data from the training data set, which consists of a large number of retrospective cases (i.e., previously analyzed cases with all pertinent attributes available). A validation data set will then be tested on the model. This validation data set also contains retrospective data but is usually smaller in size than the training data set. If the differential diagnosis given by the model for the cases in this validation data set agrees with the known outcome, the model is deemed satisfactory and is ready for application to prospective cases.
Prediction After the model has been completed, it can be used to perform predictions on any new data set of interest. The algorithms employed in certain models also allow for providing the rationale behind the selected output. In the example, the diagnostic model will offer the differential diagnosis for a new case using the marker results and other pertinent data. Rationale for the output can be presented by comparing the data in the current case with a summary of retrospective data for the given differential diagnosis. Various numerical results have been used by different algorithms to quantify the degree of accuracy for the output: certainty factor, probability, and confidence factor.
DATA-MINING ALGORITHMS As described in the previous section on the data-mining process, at the heart of all datamining tools is the data-modeling step. What may not be common to all data-mining tools is
Input layer Xi
how the model is built. This difference is due to the various algorithms upon which data modeling is based (Fayad et al., 1996). An abridged list of algorithms includes statistics, neural net, genetic, link analysis, visualization, fuzzy logic, decision tree, association, and agent network. It is beyond the scope of this unit to go into details of how these algorithms work. Only a few important and representative algorithms will be described: Statistics Neural net Genetic algorithm Hybrid algorithm. Interested readers can obtain more information on other algorithms from the references at the end of this unit (see Literature Cited). In all the algorithms described below, the term “input” or “independent variable” refers to an attribute of a neoplasm in the example project, such as marker result, nuclear size, or cytogenetic result. The term “output” or “dependent variable” refers to the differential diagnosis given by the model.
Statistics Statistics has long been used to create data models for information discovery. Some mathematicians have argued that all algorithms used in data mining are really statistical methods in different forms. While there is some element of exaggeration in this argument, statistical methods no doubt have a strong influence on almost all data-mining algorithms. Two particular statistical methods will be described here: regression modeling and discriminant analysis.
Hidden layer
Outer layer Yi
CD5 CD10 Neoplasm 1 CD19 Neoplasm 2 CD20 CD23
Figure 10.13.2 A neural net.
Data Processing and Analysis
10.13.3 Current Protocols in Cytometry
Supplement 20
Regression modeling The process of regression is an aggregate method of predicting the difference between the predicted and the actual data sets using the concept of regression towards the mean. A regression equation is a mathematical expression that estimates a dependent variable using a set of independent variables and a set of constants. For example, Y = A1 X1 + A2 X2 + .... + Ak Xk is a regression equation where Y is the dependent variable, X is the set of independent variables, A is the set of constants, and k is the number of independent variables. Because of its linear format, this type of equation is referred to as a linear regression equation. A variety of techniques have been used to generate linear regression equations, including the popular least-squares technique. For situations where no linear functions can fit the data, one needs to use nonlinear regression equations. Coefficients of nonlinear regression equations can be determined using a variety of different statistical methods. Discriminant analysis This algorithm is widely used for clustering analysis (UNIT 10.8) to find groupings of data that have similar trends and patterns. Its objective is to identify a set of discriminant coefficients (or weights) that describes the linear classification function (LCF): LCF = w1 V1 + w2 V2 + ..... + wk Vk where w is the set of discriminant coefficients, V is the set of independent variables, and k is the number of variables. The discriminant coefficients are chosen to maximally separate groups of data. Threshold values for LCF are used to classify a data object into different groups. Discriminant analysis, like other clustering methods, has no specified dependent variables; hence, it is also known as unsupervised learning or segmentation. Once clusters have been identified, further investigation will be needed to determine the significant difference between them.
Neural Net
The Application of Data Mining to Flow Cytometry
In 1943, McCulloch and Pitts initiated the concept of a neural net, which mimics a neuron in the human brain. The original neural net, Perceptron, consists of only two layers and uses linear threshold units. Perceptron was introduced in 1959 but fell into disfavor due to performance limitations. With its simple design, Perceptron was unable to solve certain simple prediction problems. Neural nets came into favor again when they overcame their
shortcomings with nonlinear threshold units and multiple-layer design. Figure 10.13.2 illustrates a typical neural net with three layers. Each of the processing units in the second layer, called “hidden nodes,” takes many inputs from the first layer and contributes to generating outputs in the third layer. Each output is calculated using a nonlinear function of the weighted sum of the inputs. The weight assigned to each input is obtained during a training process called back-propagation, in which the output generated by the neural net is compared to the targeted output, and the deviation between them is minimized by adjusting the weights. The number of inputs, layers, and outputs, and the weighing algorithms, determine the complexity of a neural net. A variety of techniques are available to determine the optimal number of hidden nodes and to adjust the weights in the training process. The hidden middle layer in Figure 10.13.2 may be designed as multiple hidden layers in a neural net. The set of outcomes Yi is dependent on the input: Yi = f(WiXi) where Xi is the set of inputs and Wi is the set of weights associated with the inputs. The function f(WiXi) is typically a nonlinear equation in a modern neural net. Applied to the example project, Xi includes the set of markers used in analysis (CD5, CD10, and so on), and Yi is the set of diagnoses.
Genetic Algorithm This is essentially a combinatorial optimization method based on biological evolution. The basic idea behind genetic algorithm is that over time, evolution will select the fittest organism in a given environment. This algorithm typically optimizes the data model to best fit the available data. This is still a very active research area, and a variety of techniques have been utilized in genetic algorithm. In general, genetic algorithm in data mining attempts to divide a data set into different groups through an iteration process similar to that seen in biological evolution. The algorithm typically starts out with a random grouping of data into several different sets, or clusters. Each data cluster can be thought of as an organism. A “fitness function” is defined to assess how well each data cluster performs in comparison to other clusters. As each data cluster is evaluated, it is ranked accordingly among all the data clusters. Genetic algorithm utilizes “fitness operators” to mimic the functions of reproduction and mutation in nature. These operators will selectively retain highly ranked data clusters for
10.13.4 Supplement 20
Current Protocols in Cytometry
crossing-over with other equally fit clusters. The repeated process of crossing-over between the best-fit clusters in turn may generate even better data clusters. The poor-fit clusters can be eliminated from the data pool. Random change in data clusters may yield better data clusters, in a manner that is similar to advantageous mutations seen in nature. Most of the random changes, however, result in poorer data clusters that will eventually be eliminated from the data pool.
Hybrid Algorithms There are many modeling techniques that use combinations of different algorithms. Datamining tools that rely on such an approach are referred to as hybrid systems. For example, genetic algorithm has often been used in conjunction with neural net to model data. In such a hybrid algorithm, genetic algorithm is useful to determine the optimal weights associated with inputs in the output function of a neural net.
RESOURCES The following is a short list of data-mining vendors in the market. This list is by no means complete, but represents a few dominant companies with their commercially available tools. International Business Machines Corporation 1133 Westchester Avenue White Plains, New York 10604 (800) 426-4968 http://www.ibm.com Their product, Intelligent Miner, is a suite of data-mining tools running on many platforms (AIX, OS/390, OS/400, Solaris, Windows). Algorithms include statistics and visualization. NeuralWare Inc. 230 East Main Street, suite 200 Carnegie, Pennsylvania 15106-2700 (412) 278-6280 http://www.NeuralWare.com Their data-mining products, NeuralWorks Predict and NeuralWorks Professional II/Plus, are suites of tools running on most Windows, UNIX, and Linux platforms. All these tools utilize neural net as the key algorithm in data mining. Some of them are hybrid systems using neural net together with other algorithms such as genetic algorithm, fuzzy logic, statistics, and rule-based expert systems. Oracle Corporation 500 Oracle Parkway
Redwood Shores, California 94065 (650) 506-7000 http://www.oracle.com Their product, Darwin, utilizes parallel computing to rapidly deliver results. Darwin was originally developed by Thinking Machines Corp., which was acquired by Oracle in 1999. This software runs on most platforms, including high-end mainframes and personal computers. Various algorithms available in this software include decision tree, regression, genetic, neural net, and visualization. Oracle Corp. is considered one of the world’s largest suppliers of database systems. SAS Institute Inc. 100 SAS Campus Drive Cary, North Carolina 27513-2414 (919) 677-8000 http://www.sas.com SAS has two products for data-mining projects: Enterprise Miner, a complete data-mining solution based on neural net; and JMP, a statistical software that links statistics with graphics to explore data for information discovery. These two products run on almost all computer platforms. SAS is considered a market leader in statistical softwares. Silicon Graphics 1600 Amphitheatre Parkway Mountain View, California 94043 (650) 960-1980 http://www.sgi.com Their data-mining product, Silicon MineSet, is a suite of five tools for visual exploration of data. Visualization is enhanced by animated 3-D landscapes. These tools run on SGI UNIX and Windows NT servers. They employ a variety of different data-mining algorithms: decision tree, probability, and association.
LITERATURE CITED Adriaans, P., and Zantinge, D. 1996. Data Mining. 1st ed. Addison-Wesley, Harlow, England. Cabena, P., Hadjinian, P., Stadler, R., Verhees, J., and Zanasi, A. 1998. Discovering Data Mining: from Concept to Implementation. 1st ed. Prentice Hall, Inc., Upper Saddle River, N.J. Fayad, U., Piatetsky-Shapiro, G., and Smyth, P. 1996. From data mining to knowledge discovery: An overview. In Advances in Knowledge Discovery and Data Mining (U. Fayad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy, eds.) pp. 1-34. AAAI Press/MIT Press, Menlo Park, Calif. McDonald, J.M., Brossette, S., and Mose, S.A. 1998. Pathology information systems: Data mining leads to knowledge discovery. Arch. Pathol. Lab Med. 122:409-411.
Data Processing and Analysis
10.13.5 Current Protocols in Cytometry
Supplement 20
Parsaye, K., and Chignell, M. 1993. Intelligent Database Tools and Applications. 1st ed. John Wiley & Sons, New York. Piatetsky-Shapiro, G. 1996. An overview of issues in developing industrial data mining and knowledge discovery applications. Proceedings of the Second International Conference on Knowledge Discovery and Data Mining,p 89. AAAI Press/MIT Press, Menlo Park, Calif.
Contributed by Andy N.D. Nguyen University of Texas Houston, Texas
The Application of Data Mining to Flow Cytometry
10.13.6 Supplement 20
Current Protocols in Cytometry
Calibration and Shading Correction for Fluorescence Microscopes Cell fluorescence, just like any light reaching a digital camera, is quantified in units of the gray scale. Such quantitation permits reliable comparison between samples only if they are all imaged within a short time interval and on the same instrument; however, any change in microscope alignment or camera settings, let alone moving the sample to a different microscope, is likely to affect the measured gray level.
UNIT 10.14
BASIC PROTOCOL
There are two basic types of variability associated with quantitation in fluorescence microscopy. The first has to do with the overall intensity of illumination and the efficiency of light collection and detection. A fluorescent object will appear darker or brighter depending on the light source, optics throughput, and camera gain. Correction for these factors requires calibration with a standard of reproducible quantum yield. The other type of variability reflects heterogeneity of illumination and/or detection efficiency across a single field. This effect is known as shading. In wide-field microscopes, a Hg or Xe arc burner is typically used in the fluorescence path. The brightness of the arc is nonuniform, which results in a nonuniform distribution of light intensity in the focal plane of the objective. Although illumination can be controlled by centering and focusing the lamp, a perfectly even field can seldom be achieved. When uniformity of illumination is critical, a fiberoptic scrambler can be used. Detection efficiency can also vary depending on the position of the object within a field. Usually there is little control over this factor; however, dust on the camera window can be one trivial cause of persistent shadows in the images. The most versatile method of shading correction in fluorescence microscopy utilizes a continuous and spatially uniform (“flat”) fluorescent field. After background subtraction, the image of a specimen is divided by the image of a flat field, which corrects for shading caused by both illumination and detection factors. Naturally, if the fluorescent field also has a constant quantum yield, both calibration and shading correction can be accomplished at the same time. A sample suitable for calibration and shading correction can be prepared from highly concentrated solutions of fluorophores. Several microliters of 10% (w/v) sodium fluorescein or 30% (w/v) rhodamine 6G are put under a coverslip and imaged under the same conditions as the specimen of interest. Concentrated fluorophores result in an even fluorescent field of reproducible brightness, resistant to photobleaching. A simple mathematical operation on images converts the specimen intensity into standard units of fluorescence. Table 10.14.1 presents the terminology that will be used throughout the unit. In the numerical examples, it will be assumed that the image intensity is digitized to 8 bits. Image names are given in italics, and angular brackets are used to denote the average gray level of an image. Subscripts i,j after an image name indicate the gray level of a pixel with coordinates i,j. The subscripts are used to emphasize the fact that mathematical operations on images, such as subtraction or division, are performed on a pixel-by-pixel basis. CAUTION: Sodium fluorescein, though not toxic, can easily stain the skin. Use gloves when handling this chemical. Even more care is needed when working with solutions of rhodamine 6G since it is a suspected carcinogen and methanol is used as the solvent. Image Acquisition Generally, four images must be collected: Raw, Background, Shade, and Blank (Table 10.14.1). Contributed by Michael A. Model Current Protocols in Cytometry (2002) 10.14.1-10.14.7 Copyright © 2002 by John Wiley & Sons, Inc.
Data Processing and Analysis
10.14.1 Supplement 21
Table 10.14.1 Definitions of Terminology
Term
Definition
Background
Image of the background fluorescence. It can be either directly acquired or obtained by transformation of Raw Image of a nonfluorescent sample (i.e., water) Image of the specimen Unedited image of the specimen Image of the fluorescein or rhodamine standard solution
Blank Corrected Raw Shade
Raw. Optimize the conditions (e.g., filters, exposure) for imaging of fluorescent specimens. The signal from all the specimens should be sufficiently strong without saturating the pixels. Background. If Raw has enough area free of cells or tissue, it can be used to obtain information about the background (see Image Processing). Otherwise, take a separate image (Background) of a different and mostly empty area in the specimen. Shade. Use the fluorescein standard solution (see recipe) if the specimen is stained with a green fluorophore (e.g., fluorescein) or fresh rhodamine standard solution (see recipe), if the specimen is stained with a red fluorophore (e.g., Texas Red). Optical filters must be the same as for the specimen; however, the exposure can be the same or different if the intensities of the specimen and the standard are approximately equal (or can be made such by using a neutral density filter). Using the same exposure time for all images may offer some advantages (see Troubleshooting). As with Raw, the gray level of Shade should be sufficiently far from saturation, since, in the future, the illumination may become brighter. Take several images of standard slides during the imaging session (i.e., Shade1, Shade2, ..., ShadeN). To make standard fluorescent slides, clean glass slides and coverslips thoroughly and keep them protected from dust. Put several microliters of the fluorescein or rhodamine standard solution on a slide and cover with a coverslip. Use just enough volume to make the solution spread over the entire area under the coverslip. Even though some areas under the coverslip may appear darker than others due to uneven depth of the liquid, the fluorescent signal should be unaffected by this. When using the rhodamine solution, cover quickly before evaporation occurs. Seal coverslip with nail polish and use within several hours. Check that all slides come out with reproducible fluorescence (in the author’s hands, reproducibility is 97% to 98%). A standard slide appears rather uniform. For reproducible focusing, close the fluorescence field aperture and bring it into sharp focus, then open again. The shadow of the field aperture can be observed only on the thin layer of fluorescent solution nearest the objective because the light does not penetrate into deeper layers. Thus, when the shadow is sharp, the objective is focused precisely on the coverglass-solution interface. Imaging of a standard slide can also be used to optimize lamp alignment. Blank. Put water or sodium bicarbonate under a coverslip. Use the same exposure as was used for Shade. Calibration and Shading Correction for Fluorescence Microscopes
10.14.2 Supplement 21
Current Protocols in Cytometry
Image Processing General information on image processing can be found in numerous reviews and monographs, such as Oberholzer et al. (1996), Cardullo and Alm (1998), and Russ (1999). Shade. Dust on coverslips may appear as bright speckles. These can be removed from the image by applying the median filter (see Fig. 10.11.7 for an example). After correction of defects, replace all individual Shade images with the averaged image ShadeAv: ShadeAvi,j = (Shade1i,j + Shade2i,j + ... ShadeNi,j)/N. Subscripts i,j indicate that this operation is performed on a pixel-by-pixel basis. Some programs have a command that does the averaging of multiple images in one step. Blank. The image of a nonfluorescent sample should be almost uniform. Calculate the average intensity of the field, , using an image-statistics tool. Background. Use the cursor or any other appropriate analysis method to measure the average gray level of the background. When fluorescent objects occupy only small discrete areas, the entire image of the background can be restored by transforming the original image, as described by Likar et al. (2000), Russ (1999), or UNIT 2.11. If the medium is strongly fluorescent, such that the heterogeneity of the background cannot be disregarded, use the entire Background image in the equation for Correctedi,j below rather than its average intensity. Around three-dimensional fluorescent objects there is often a halo resulting from out-of-focus planes. For finding from an image containing some fluorescent material, choose only the areas where the intensity levels off. Corrected. Perform shading correction and normalization of the image according to the following formula:
Corrected i , j =
Raw i, j − 〈 Background〉 ShadeAvi , j − 〈 Blank 〉
×C
where C is an arbitrary constant and and are the mean gray levels of images Background and Blank, respectively. If the intensities of Raw and ShadeAv are comparable, C can be set to approximately one-half maximal possible gray level. Corrected will have integer gray values computed by rounding the values of the expression on the right side of the equation above. If the value of C chosen is too small, the range of gray values will be limited. As an example, suppose that: ri , j =
Rawi , j − 〈 Background〉 ShadeAvi, j − 〈 Blank 〉
where r is a ratio which varies from 0.18 to 2.5. For C = 10, the gray values in Corrected will range from 2 to 25. Only 24 gray levels out of the available 256 will be used to display the image. If C = 100, the image will be represented by gray levels between 18 and 250—i.e., by almost the entire 8-bit gray scale; however, for C = 200, the parts of the image where r >1.3 will appear saturated, and the intensity information will be lost.
Data Processing and Analysis
10.14.3 Current Protocols in Cytometry
Supplement 21
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Fluorescein standard solution Dissolve 100 mg sodium fluorescein in 1 ml of 0.1 M NaHCO3 in a microcentrifuge tube. Vortex until no visible undissolved particles remain. Filter through a 0.2-µm filter. Store 4 months in the dark at room temperature or 4°C. More concentrated fluorescein (up to 75% w/v) can be used if desired. More concentrated solutions tend to yield a marginally better reproducibility when using objectives with large numerical apertures; however, they may be less suitable for low power objectives because in this concentration range the brightness decreases with increasing concentration.
Rhodamine standard solution Dissolve 300 mg rhodamine 6G in 1 ml methanol and vortex vigorously. Because methanol is volatile, filtration cannot be used; therefore, microcentrifuge the solution at maximum speed for 1 min, room temperature. Quickly remove the supernatant and store in a capped microcentrifuge tube protected from light. Prepare fresh before use. The author has not tested the stability of rhodamine solution. Even though it may be stable chemically, evaporation of solvent can result from multiple uses of the same aliquot or storing many small aliquots. Until more data are available on the stability of rhodamine 6G in solution, its prolonged storage should be avoided.
COMMENTARY Background Information
Calibration and Shading Correction for Fluorescence Microscopes
Calibration of fluorescence microscopes has been achieved in the past using a variety of standards—e.g., beads, slides made of fluorescent materials, chambers or capillaries filled with diluted fluorescent solutions (Sisken, 1989; Rost, 1991; Galbraith et al., 1991; Turney et al. 1996; Wolf, 1998; Ghauharali and Brakenhoff, 2000). The method using concentrated solutions of fluorophores was first described by Model and Healy (2000) and in more detail by Model and Burkhardt (2001). A combination of several useful features makes this technique practical: 1. Due to extremely high optical density, only a very thin layer of solution immediately adjacent to the coverslip is illuminated. That makes the total depth of the liquid between a coverslip and a slide unimportant, resulting in high reproducibility. 2. The shallow depth of the illuminated layer allows rapid diffusional exchange of molecules. As a result, the intensity of the signal remains constant upon prolonged illumination because bleached molecules are rapidly replaced with fresh ones. 3. Fluorescein at high concentrations exhibits quenching (this also applies to rhodamine but to a lesser extent). The brightness of a 10% (w/v) solution is well suited for observation
under a microscope and is comparable to the brightness of many stained biological objects. 4. The chemicals are inexpensive and readily available, so one does not have to depend on the availability of a unique commercial product. 5. The continuous nature of the standard makes possible a straightforward correction for brightness and shading at the same time. 6. Shading correction using concentrated fluorophores is more precise than with deep translucent samples.
Critical Parameters Standardization using concentrated solutions of fluorophores achieves successful correction for variable illumination, optics throughput, and sensitivity of the CCD camera. It is important, however, to match the objectives and optical filters. It is also essential to focus on the specimens in a consistent manner. Objectives The numerical aperture of the objective lens determines the depth of the imaged layer (Taylor and Salmon, 1989). Measurements made with objectives with different numerical apertures may not be comparable since different parts of the specimen will be recorded.
10.14.4 Supplement 21
Current Protocols in Cytometry
Optical Filters A concentrated standard may have spectral properties different from those of the specimen. Because of this, the optical filters (i.e., excitation, emission, and beamsplitting) must always be consistent. Small deviations from the spectral characteristics of the original filters may be tolerated but large deviations are likely to result in the failure of standardization. If one wishes to switch to different filters, it is important to verify that Corrected images obtained with both filter sets have equal intensities.
Focus If the coverslip over the specimen is not horizontal, some areas of the field may not appear in focus. The efficiency of light collection depends on focusing, and blurry regions should be rejected before quantitation.
Troubleshooting Lamp instability The light source should be stable enough to permit reliable quantitation. If large fluctuations are observed, replace the illumination lamp. A new lamp may have to burn for a few
A
B
C
Figure 10.14.1 Example of shading correction. Images were acquired with an Axiovert TV 100 microscope (Carl Zeiss) through a Plan Neofluar 20/0.5 air objective. The microscope was equipped with a 100-W Hg lamp and a Micromax 1300Y CCD camera (Roper Scientific). (A) Image of a 10% fluorescein slide (Shade). There is more than a two-fold difference in the intensity between the brightest area below the center of the field and the dark upper corners. (B) Original image (Raw) of Jurkat cells stained with anti-phosphotyrosine Alexa 488 (for details see Model and Burkhardt, 2001). (C) Image Corrected. Note in particular the significant brightening of cells in the upper left corner. The background level was determined from measuring the empty areas in Raw; it has a uniform intensity an order of magnitude less than the intensity of a typical cell.
Data Processing and Analysis
10.14.5 Current Protocols in Cytometry
Supplement 21
hours to achieve a stable regime. If lamp replacement does not help, consider a more stable power supply. Dry slides If a standard slide appears too dim, it may have dried. Make a fresh one immediately before use.
Calibration and Shading Correction for Fluorescence Microscopes
Unusually bright images It may happen that a particular specimen turns out to be brighter than was originally anticipated. Raw may have gray levels within the scale, but values after normalization may become greater than can be displayed. In this case, the normalization constant C for that image can be reduced. A subsequent adjustment will be needed to allow comparison with other specimens. For instance, let C = 100 in the example considered above. Suppose a specimen is encountered where the ratio r = 5 instead of the expected 0.18 to 2.5. Just for that image, C can be reduced to 40, so that the normalized intensity becomes 200. After the intensity has been measured, it is multiplied by 100/40 = 2.5. Even though the imaging software may not be able to handle a gray level of 500 (2.5 × 200), this image can still be included in the specimen statistics. If C is not reduced (i.e., a wider range is not chosen) for a bright Raw, Corrected will be saturated. In 8-bit images, assumed in these calculations, the maximal brightness level that can be resolved corresponds to gray level 255. All areas that are brighter will be reduced to 255 and quantitative analysis of such an image will produce values that are too low; therefore, the brightness of Corrected must first be scaled down to avoid saturation. The scaling factor is chosen to ensure that all pixels of interest have gray levels less than 255 so that the image analysis software can correctly measure the image and generate the average intensity. Then, in order to compare this image with others where C was different, these differences must be taken into account. Once the problem becomes dealing with single numbers (average intensity of an image) there is obviously no limit of 255. A slightly different situation can arise if Raw (taken at exposure time TRaw1) becomes saturated. Then one can reduce the exposure time to TRaw2 so that the gray levels fall below 255 As long as pixels are not saturated, there is a linear proportionality between the gray level I and the exposure time T: I = B + a × T, where B is the camera offset and a is a constant. Thus,
if intensity I2 at exposure TRaw2 is known, it is possible to calculate what the intensity would have been at exposure TRaw1:
I1 = B + ( I 2 − B ) ×
TRaw1 TRaw2
This holds true both for Raw and . For their difference we have
Raw1i , j − 〈 Background1〉 = ( Raw2i, j − 〈 Background2〉 ) ×
TRaw1 TRaw2
which leads to a simple relationship between normalized images taken at different exposures:
Corrected1i , j = Corrected 2i , j ×
TRaw1 TRaw2
As before, the fact that the gray levels of Corrected1 will exceed 255 does not preclude quantitation and inclusion of the image in the statistics. Likewise, if Shade becomes overexposed, the exposure for both standard and blank is reduced from TShade1 to TShade2. Since the difference between Shade and is present in the denominator of the equation for Correctedi,j (see Image Processing), the conversion is done according to:
Corrected1i , j = Corrected 2i , j ×
TShade1 TShade 2
Finally, if both the specimen and the standard are imaged at different exposures TRaw2 and TShade2, then:
Corrected1i , j = Corrected 2i , j ×
TRaw1 TShade 2 × TRaw2 TShade1
It is evident from the last equation that if the specimen and the standard are always imaged at the same exposure (TRaw1 = TShade1, TRaw2 = TShade2) then:
TRaw1 TShade 2 × =1 TRaw2 TShade1
10.14.6 Supplement 21
Current Protocols in Cytometry
and no correction is needed. The concept that, whenever possible, the same acquisition time should be used for specimens and for standards, refers expressly to this fact (see Image Acquisition, Shade).
Anticipated Results
Model, M.A. and Healy K.E. 2000. Quantification of the surface density of a fluorescent label with the optical microscope. J. Biomed. Mater. Res. 50:90-96. Oberholzer, M., Östreicher, M., Christen, H., and Brühlmann, M. 1996. Methods in quantitative image analysis. Histochem. Cell Biol. 105:333355.
Shade should be a featureless field with gradually varying intensity. Corrected looks similar to Raw. The absolute values of their intensity levels are different, though, and the shading effect should be minimized in Corrected. An example of Shade, Raw, and Corrected is given in Figure 10.14.1.
Rost, F.W.D. 1991. Quantitative Fluorescence Microscopy. Cambridge University Press, Cambridge.
Time Considerations
Taylor, D.L. and Salmon, E.D. 1989. Basic fluorescence microscopy. Methods Cell Biol. 29:207-37
Preparation of the standard solution may take 10 to 20 min, and preparation of one slide takes ∼1 min.
Literature Cited Cardullo, R.A. and Alm, E.J. 1998. Introduction to image processing. Methods Cell Biol. 56:91115. Galbraith, W., Wagner, M.C.E., Chao, J., Abaza, M., Ernst, L.A., Nederlof, M.A., Hartsock, R.J., Taylor, D.L., and Waggoner, A.S. 1991. Imaging cytometry by multiparameter fluorescence. Cytometry 12:579-596. Ghauharali, R.I. and Brakenhoff, G.J. 2000. Fluorescence photobleaching-based image standardization for fluorescence microscopy. J. Microsc. 198:88-100. Likar, B., Maintz, J.B.A., Viergever, M.A., and Pernus, F. 2000. Retrospective shading correction based on entropy minimization. J. Microsc. 197:285-295. Model, M.A. and Burkhardt, J.K. 2001. A standard for calibration and shading correction of a fluorescence microscope. Cytometry 44:309-316.
Russ, J.C. 1999. The Image Processing Handbook, Third Edition. CRC Press/IEEE Press, Boca Raton, Fla. Sisken, J.E. 1989. Fluorescent standards. Methods Cell Biol. 30:113-126.
Turney, S.G., Culican, S.M., and Lichtman, J.W. 1996. A quantitative fluorescence-imaging technique for studying acetylcholine receptor turnover at neuromuscular junctions in living animals. J. Neurosci. Methods 64:199-208. Wolf, D.E. 1998. Quantitative video microscopy. Methods Cell Biol. 56:117-134.
Key Reference Model, M.A. and Burkhardt, J.K. 2001. See above. Detailed description of calibration method using concentrated fluorophore solutions.
Acknowledgment The author wishes to thank Dr. Janis K. Burkhardt for helpful comments.
Contributed by Michael A. Model Cleveland Clinic Foundation Cleveland, Ohio
Data Processing and Analysis
10.14.7 Current Protocols in Cytometry
Supplement 21
A Software Method for Color Compensation
UNIT 10.15
Loken (1977) was the first to describe an electronic method (called spectral compensation) for removing the undesired overlap of one fluorochrome signal from another, which occurs when two or more fluorochromes are measured simultaneously. The general practice has been to take particles or cells separately stained with each fluorochrome and adjust the instrument compensation. Although very successful for two and three fluorochromes, this method becomes increasingly inadequate when more than three fluorochromes are mixed together and can lead to artifactual data and misinterpretation. Roederer describes most of these problems in UNIT 1.14. This unit addresses the unique problems associated with the tandem dyes that have become so popular. The tandem dyes are a complex consisting of a donor molecule and an acceptor molecule. The donor is excited and transfers its energy to the emitter molecule by fluorescence resonant energy transfer (FRET; UNIT 1.12). The most common donor molecules are the phycobiliproteins; the most common emitters are Texas Red and the long-wavelength cyanine dyes. There are three major problems with the tandem dyes. The most troublesome and significant problem is inter- and intra-manufacturer variation in required compensation, caused primarily by overconjugation of the antibody followed by poor purification procedures. While some manufacturers’ products exhibit less variation, every antibody requires some change in compensation, ranging from <1% to >10%, as discussed by Stewart and Stewart (1999, 2001a,b). Consequently, each lot of tandem-conjugated antibody must be uniquely compensated. The second problem is the binding of PE-cyanine tandems to monocytic-lineage cells. Van Vugt et al. (1996) suggested that the binding was to the FcRI receptor (CD64) and their data support this notion. However, the authors find that this tandem can also bind to cells that are CD64 negative, suggesting that the binding is more complicated. Some manufacturers now supply tandems that do not exhibit this monocyte-binding problem. Finally, the PE-cyanine tandems are ambient-light sensitive. This causes the cyanine component to decompose, thereby yielding pure PE. When stained samples are exposed to light prior to data acquisition, the amount of PE fluorescence increases so that the compensation required slowly changes and artifactual data are obtained. Protection of samples from light eliminates this problem. For establishing the correct compensation, Roederer (UNIT 1.14) makes the correct point that the autofluorescence of unstained cells must be homogeneous and the same as the stained cells used for setting compensation. This might be misinterpreted to mean that different compensation settings will be needed depending on the autofluorescence of the cells, e.g., lymphocytes versus monocytes. When setting compensation values, it is necessary that the same homogeneous population be used for establishing all settings. Lymphocytes are often chosen because they are the most homogeneous and least autofluorescent, but once determined, the parameters of compensation can be applied to any population. However, the position of the marker for evaluating positive or negative cells may not be the same for other cells, as their autofluorescence will vary depending on the excitation and emission wavelengths. If any instrument setting is changed, new compensation must be performed. The appropriate compensation for more than three colors cannot be reliably obtained using cells stained with a single antibody (Stewart and Stewart, 1999). As Roederer explains (UNIT 1.14), the correct way to set compensation is to align the medians for autofluorescence and stained cells. This will necessarily prevent the use of quadrant markers; when the median of a very bright marker is set equal to the median of the unstained cells, the variance of the stained population is generally much larger than that Contributed by Carleton C. Stewart and Sigrid J. Stewart Current Protocols in Cytometry (2003) 10.15.1-10.15.12 Copyright © 2003 by John Wiley & Sons, Inc.
Data Processing and Analysis
10.15.1 Supplement 23
of the autofluorescent population and its fluorescence will always extend above the quadrant marker set by the unstained cells. Either a marker must be set at the correct angle or compensation must be incorrectly set so that the “tops” of the populations line up. The latter practice always causes overcompensation. For practical reasons, however, most people follow this pathway. If one is concerned only with the percentage of double-positive cells, this is a practical way to solve the problem of statistics. While it is true that the overlap of one fluorochrome emission into another is a fixed percentage for a given system, the PMT high voltages and the compensation circuits are interactive. The percentage of compensation on the instrument may not accurately reflect the actual percentage of overlap. Thus, setting compensation at a fixed percentage and adjusting the PMT voltage will be just as effective in changing compensation as keeping the PMT voltage constant and setting the percentage compensation. Adjusting compensation using the PMT voltage, however, is not recommended because it will change the position of the unstained cells (autofluorescence). Isotype controls should not be used for setting the position of negative cells. It is very important to maintain saturation stoichiometry in all staining reactions of cells. If an antibody conjugated with a particular fluorochrome is so bright that it is off scale when the unstained cells are properly adjusted, never dilute the antibody to bring it on scale, because one will lose conditions of saturation (for a more complete discussion, see Stewart and Mayers, 2000). Instead, an antibody conjugated with a fluorochrome of lower quantum efficiency should be used or the conjugated antibody should be diluted with unconjugated antibody so the fluorescence is reduced while stoichiometry is maintained. The tandem dyes that are constructed using PE or PerCP as the donor molecule will be excited by the 488-nm laser beam. Those constructed using APC as the donor molecule will be excited by the 633-nm beam. This latter beam will also excite the emitter molecules: Texas Red, PerCP (to some extent), the red cyanine dyes, and the long-wavelength Alexa dyes. Therefore, compensation must be performed between the PE detector and the acceptor molecule detector for 488-nm excitation and between the PE detector and the detector used for the 635-nm excitable dyes. Emission spectra for tandem dyes can be quite wide because the energy transfer is not always optimal. Usually, bivariate plots are used to evaluate fluorescence because every combination of fluorescence parameters needs to be displayed for multiparameter flow cytometry. An improperly compensated cluster may project itself onto an otherwise irrelevant bivariate plot, or interaction between compensation networks can result in correct compensation for one pair of fluorochromes and incorrect compensation for a second pair. These effects may lead the operator into compensating the wrong parameter combination. Four commonly encountered problems associated with incorrect compensation are illustrated in Figures 10.15.1 to 10.15.4. The effect of incorrect compensation may not be seen with some antibody combinations because it will depend on the relative intensities of their fluorochromes and the epitope density on the cells. For subsequent discussion, the following terms will be used for convenience and simplicity. Note that flow cytometers may differ in the placement and collection range of detectors. FL1= green fluorescence detector (525 nm); FITC FL2 = orange fluorescence detector (575 nm); PE FL3 = red fluorescence detector (630 nm); PE-Texas Red FL4 = deeper red fluorescence detector (675 nm, with 633-nm excitation); PerCP PE-Cy5, PE-Cy7 A Software Method for Color Compensation
10.15.2 Supplement 23
Current Protocols in Cytometry
D
E
F
G
FL4
C
FL2
B
FL2
A
H
CD8-PE-Cy5
I
FL4
CD8-PE-Cy5
Figure 10.15.1 Projection of uncompensated data. For clarity, only three bivariate histograms are shown and parameters without any antibody of that color are designated by the parameter name. Otherwise the antibody name and fluorochrome are used.
B
CD4-PE
A
FL4
FL3
C
D
FL3
FL4
Figure 10.15.2 Overcompensation. For clarity, only two bivariate plots are shown and parameters without any antibody of that color are designated by the parameter name. Otherwise, the antibody name and fluorochrome are used.
Data Processing and Analysis
10.15.3 Current Protocols in Cytometry
Supplement 23
B
A
C
b
n
d
b
CD45-APC
n
d
CD8-PC
Figure 10.15.3 Correcting overcompensation. Overcompensation can also result in the apparent loss of expression of an epitope. This is exacerbated when very bright staining is encountered.
uncompensated
undercompensated
overcompensated
CD3-APC
correct incorrect correct
CD8-PE-Cy5
CD8-PE-Cy5
incorrect
CD8-PE-Cy5
Figure 10.15.4 Equal fluorescence detected by two detectors.
A Software Method for Color Compensation
10.15.4 Supplement 23
Current Protocols in Cytometry
In the top row of Figure 10.15.1 (A, B, C), compensation is required for FL4-%FL3 (C). Note that cells also appear uncompensated in both A and B. When FL4-%FL3 is correctly compensated (F), the projection onto FL2 is eliminated (E), but some FL2-%FL3 compensation is still required (D). The correctly compensated sample is shown in the last row (G, H, I). In Figure 10.15.2, FL3-%FL2 is overcompensated (A), as shown by the events along the y-axis. Now the cells stained with CD4-PE appear undercompensated in B, and one might be tempted to compensate FL4-%FL2. If, however, FL3-%FL2 is correctly compensated (C), FL4-%FL2 (D) is too. This problem is encountered when cells exhibit high epitope density and are stained with a very bright fluorochrome. Figure 10.15.3A shows an uncompensated view of cells stained with CD45, some of which are co-expressing CD8. Populations of cells negative (n), dim (d), and bright (b) for CD8 can be resolved. The vertical line represents the median fluorescence intensity of CD8 brightly stained cells. If FL3-%FL4 is overcompensated (B), as might occur using single-stained cells, the CD8-positive cells all appear to be dim. A marker detected by FL3 that stains cells with low intensity might completely disappear. When correctly compensated (C), the intensity of brightly stained cells is identical to that of uncompensated ones. One can achieve this verification only if cells are stained simultaneously with both markers. Another problem occurs when a fluorochrome emission is found on an ∼45° angle between two detectors, as illustrated in Figure 10.15.4. This happens when the antibody is overconjugated and cells exhibit the expression shown in A. In this situation, compensation cannot be properly performed because both detectors are measuring essentially the same fluorescence intensity. PE-Texas Red reagents and PE-Cy5 reagents in the fourcolor mode can be particularly troublesome. In Figure 10.15.4B, the cells expressing dim fluorescence are correctly compensated, but the bright cells are overcompensated so some of them appear negative. When the compensation is reduced so that the bright cells are correct (C), the dimly stained cells are incorrectly compensated. There are two possible solutions to this problem. In one solution, the detector high voltage is decreased so that its signal for the undesired fluorochrome is lower. The problem with this solution is that the signal for the desired fluorochrome is also reduced. The best solution is not to use this particular antibody/fluorochrome conjugation. The problem can be eliminated by selecting the proper fluorochrome conjugation for the antibody. The strategy is to select FITCor PerCP-conjugated antibodies for highly expressed epitopes and PE- or PE-Cy5-conjugated antibodies for lower expressed epitopes. APC-conjugated antibodies can be used for either. Whether software or hardware compensation is performed, every detector will see some degree of fluorescence from every fluorochrome, and these will interact in a complex way (UNIT 1.14). The more fluorochromes used, the less forgiving incorrect compensation becomes, and if compensation is incorrect, the resulting data analysis will be wrong and misinterpreted. However, obtaining correct results is easy. When data are collected uncompensated, software compensation provides the user with the flexibility of setting correct compensation every time for every sample. Instruments using incorrect hardware compensation settings provide data that are of little value. Software methods do have their own problems. The linearization assumptions made by the software algorithms may be more or less in error, depending on how logarithmic the amplifiers actually are. Binning effects become more of a problem with increasing numbers of compensated parameters. Data Processing and Analysis
10.15.5 Current Protocols in Cytometry
Supplement 23
The protocols to follow illustrate the process of software compensation utilizing matrix algebra that provides for elements of all possible PMT detection combinations. This makes it very easy to establish the correct compensation, every time, on data that have been acquired uncompensated. Although the details here are limited to four colors, the principles described can be applied to any number of desired colors. While the authors use WinList (Verity Software House) to illustrate software compensation, any program (e.g., Flo Jo; Tree-Star) capable of providing this feature will utilize the same basic process. ACQUISITION OF DATA FROM COMPENSATION STANDARDS This protocol uses human blood lymphocytes as target cells for establishing the compensation matrix because they are homogeneous and exhibit low autofluorescence. Blood from larger mammals and rodent spleen cells are equally good sources. For staining cells with antibodies, refer to UNIT 6.2. The following protocol will require the following materials: a lymphocyte suspension (UNIT 5.1), 12 × 75–mm tubes suitable for the flow cytometer, and conjugated antibodies: CD45-FITC, CD4-PE, CD45-APC, CD3-FITC, CD4-PE, CD8-PE-Cy5, plus all desired tandem conjugates. 1. Set up compensation standards by staining lymphocytes (e.g., spleen cells from rodents, human blood) in separate tubes as follows: Tube 1: CD45-FITC Tube 2: CD4-PE Tube 3: CD45-APC Tube 4: CD3-FITC, CD4-PE, CD8-PE-Cy5, and CD45-APC (this combination will be used for verification of compensation) Tubes 5 through n: each desired tandem to generate their unique compensation matrices. 2. Create an acquisition display (Fig. 10.15.5) with FS versus SS, FL1 versus FL2, FL3 versus FL2, and FL3 versus FL4. 3. Set all instrument compensations to zero and acquire a file of unstained cells (Fig. 10.15.5, top row). Establish a region R1 around the lymphocytes and gate all bivariate fluorescence histograms on R1. 4. Adjust PMT voltages so that the linearized median channel fluorescence for all fluorochromes is the relative intensity of 3.0 ± 0.2. Do not do this for Beckman-Coulter instruments because virtually all cells will be in channel 1. 5. Acquire a data file from each tube, as shown in Figure 10.15.6, rows 1 to 4. Acquire data for the four-color combination (tube 4). Acquire data for all samples stained with tandem dyes. Beckman-Coulter instruments acquire integrated pulse area, whereas, Becton Dickinson and Cytomation use pulse peak height. Negative events appear considerably different in the respective bivariate plots. For a Coulter instrument, most events are along the axis because of the integration threshold setting. In contrast, peak detection does not require an integration threshold, so pulse height data can be set at any value. For additional information, see Shapiro (1995).
A Software Method for Color Compensation
10.15.6 Supplement 23
Current Protocols in Cytometry
FL4
FL2
FL2
R1 FL3
FL4
FL3
FL2
FL2
FL1
R1 FL3
FL3
FL4
CD4-PE
SS
CD4-PE
CD45-FITC
R1 FL3
FL4
FL3
FL2
FL2
FL1
R1 CD2-PE-Cy5 CD45-APC
CD2-PE-Cy5
FL2
FL2
FL1
R1 FL1
FL3
FL3
Figure 10.15.5 Setting up the instrument. Human leukocytes were used and a region R1 drawn around lymphocytes, shown in the left view, was used as a gate. Parameters without any antibody of that color are designated by the parameter name. Otherwise, the antibody name and fluorochrome are used. The top row shows unstained cells used for adjusting the high-voltage settings. Row 2 is uncompensated CD45-FITC in FL1, row 3 is uncompensated CD4-PE in FL2, row 4 is uncompensated CD2-PE-Cy5 in FL3, and row 5 is uncompensated CD45-APC in FL4. The software will automatically compensate each parameter. Refer to the software help menu if additional information is required.
CORRECTION OF COMPENSATION FOR ALL NONTANDEM FLUOROCHROMES This protocol creates a generic compensation matrix for all the fluorochromes except the tandems. This matrix will subsequently be modified by compensating each tandem. As each tandem is different, each must have its own unique matrix. In addition, a region file of the correctly compensated data is created to verify instrument performance. This protocol is based on the program WinList (Verity Software), but other software can be used. 1. Start WinList and create histograms for all fluorescence parameters as shown in Figure 10.15.6.
Data Processing and Analysis
10.15.7 Current Protocols in Cytometry
Supplement 23
CD45-FITC
FL4
FL4
FL1
FL4
FL4
R12 FL1
CD45-APC
FL1
R10
CD2-PE-Cy5
FL1
CD45-APC
FL4
FL3
R9
FL3
R11
FL1
FL4
FL3 CD2-PE-Cy5
FL1
FL4 FL2
FL2
CD2-PE-Cy5
FL3
FL3
CD45-FITC
FL4 FL2
FL2
R8
R3
R7
FL3
FL1 FL2
R6
FL4 CD4-PE
CD4-PE
CD4-PE FL2
FL1
FL1
FL3
FL3
CD45-FITC R5
FL2
FL2
FL2
R2
R4
FL3
R13
CD45-APC
Figure 10.15.6 Software compensation. The software has automatically compensated the uncompensated data shown in Figure 10.15.5. Parameters without any antibody of that color are designated by the parameter name. Otherwise, the antibody name and fluorochrome are used. The first row shows compensated CD45-FITC in FL1, the second row shows CD4-PE in FL2, the third row shows CD2-PE-Cy5 in FL3, and the fourth row shows CD45-APC in FL4. Regions R2 to R13 are drawn around each cluster. Daily verification of instrument performance requires that cells stained with a single antibody appear in these regions; the process fails if they do not.
Turn on fluorescence parameters
Apply compensation to data
OFFSET events along axis
A Software Method for Color Compensation
Select to automatically compensate
Figure 10.15.7 Compensation toolbox.
10.15.8 Supplement 23
Current Protocols in Cytometry
2. Open the file containing CD45-FITC data. Open the compensation dialog box, shown in Figure 10.15.7, and click on each fluorescence parameter in the list on the left. From the list on the right highlight and click on the FITC parameter (FL1). This will automatically compensate this fluorochrome, as shown in Figure 10.15.6. 3. Repeat step 2 for the files containing CD4-PE, one of the desired tandems (CD2-PECy5 is illustrated), and CD45-APC. 4. Save the compensation matrix as single.cmp using the Save button in the dialog box. WinList provides for a log offset, also shown in the bottom right of the dialog box, that assigns a value to any event along the axis, so that all events are normally distributed among channels 1 to 10. This is a cosmetic function for improving the appearance of the histogram. 5. Draw a region around each cluster as shown in Figure 10.15.6 and save the file as single.reg. VERIFICATION OF COMPENSATION
CD45-APC
CD4-PE
CD4-PE
CD4-PE
The data have now been compensated using cells stained with a single-color fluorochrome. Compensation must be verified before proceeding with the unique compensation for the tandems, because it is not possible to obtain correct compensation using cells
A
CD3-FITC
CD8-PE-Cy5
CD45-APC
CD8-PE-Cy5
CD3-FITC
CD8-PE-Cy5
CD45-APC
CD4-PE
CD4-PE
CD4-PE
R2
R3
CD45-APC
B
R4 R5
CD8-PE-Cy5
Figure 10.15.8 Verifying compensation. The compensation matrix single.cmp has been applied and All Trace Lines is selected in the dialog box. In the top row, FL3-%FL2 and FL3-%FL4 may require adjustment, shown as heavy trace lines. The trace lines for the other views, shown as light lines, do not require adjustment and the trace lines A and B are correctly set for CD8-PE-Cy5. These lines are reset here to zero so that basic.cmp can be applied to other tandem fluorochromes for their unique adjustment. A trace line parallel to an axis means there is no software compensation for that parameter combination.
Data Processing and Analysis
10.15.9 Current Protocols in Cytometry
Supplement 23
stained with single colors, as shown in Figures 10.15.2 and 10.15.3, even if they are mixed together in the same tube. The reason is that cells go through the laser beam one at a time, so that the compensation error associated with expression of more than one color does not occur. The basis for this error has been previously described by Stewart and Stewart (1999). In order to verify compensation, cells must be stained simultaneously with a PE-conjugated antibody, a mutually exclusive PE-Texas Red- or PE-Cy5-conjugated antibody, and a co-expressed APC-conjugated antibody. A good combination for this is CD3-FITC, CD4-PE, CD8-PE-Cy5, and CD45-APC (see Acquisition of Data from Compensation Standards, step 1, tube 4). The protocol makes use of trace lines, which are a continuum of the values produced by the compensation matrix so that the compensation being applied can be seen. Clicking on the line and dragging it changes the line positions, and new values are then computed. Thus, compensation can be set manually by the user or automatically computed. 1. Open the file containing the four-color combination shown in Figure 10.15.7. 2. Using the Open button in the dialog box, open the compensation matrix single.cmp. 3. In the dialog box shown in Figure 10.15.7, remove compensation by unchecking the box in the lower left corner. Also select All Trace Lines from the right list (currently highlighted in Figure 10.15.7). If necessary, click and drag to adjust only the bold trace lines shown in Figure 10.15.8, top row. Do not adjust the light trace lines. (They will all appear to be bold on the monitor.) 4. Apply compensation by checking the box in the lower left corner. A “+” and “–” will appear in small boxes (not shown) on the uppermost or the rightmost corners of each bivariate histogram. Apply or remove small increments of compensation by clicking on the “+”or “–” box, respectively, until the cell cluster medians are equal to that of their comparison cluster. 5. Draw regions R2 through R5 to circumscribe each cluster as shown in the lower view of Figure 10.15.8 and save as verify.reg using the region dialog box. 6. Since each tandem must be set individually, set trace line A (FL2-%FL3) and trace line B (FL4-%FL3) to zero by moving the trace lines parallel to the their respective axes and save this matrix as basic.cmp. This will be the basic compensation matrix for obtaining the unique matrices for each antibody combination made with different PE-Texas Red or PE-Cy5 tandems. Trace lines A and B will be the ones that will require adjustment.
A Software Method for Color Compensation
The bottom row shows the correctly compensated data for this combination only. The regions R2 through R5 have been drawn so that newly stained cells can be evaluated to insure continuity of instrument performance on a daily basis. These cells should always be found in these boxes. As shown in Figures 10.15.6 and 10.15.8, this process requires that the cell clusters formed by the cells stained with either single antibodies or the combination prepared for the verification of instrument compensation fall into the defined regions saved in the files single.reg and verify.reg. If they do not, one must determine if there is an instrument problem or a staining problem. For example, new lots of antibody will almost always require adjustment of the compensation matrix and region files. For a FACSCalibur, it is absolutely essential that the instrument timing adjustment be also correctly made on a daily basis and that the red diode laser be in perfect operating condition.
10.15.10 Supplement 23
Current Protocols in Cytometry
CD33-PECy5
FL4
CD33-PECy5
FL2
B
FL4
FL2
A
CD33-PECy5
CD33-PECy5
Figure 10.15.9 Unique compensation matrix for each tandem. The listmode file of monocytes (gated by moving R1 from the lymphocyte to the moncyte cluster shown in Figure 10.15.5) stained with CD33-PE-Cy5 is shown with the basic.cmp matrix applied. The trace lines are adjusted to provide the correct compensation. Parameters without any antibody of that color are designated by the parameter name. Otherwise the antibody name and fluorochrome are used.
GENERATE COMPENSATION MATRICES FOR EACH ANTIBODY COMBINATION The compensation in basic.cmp will be correct for FL1-%FL2, FL2-%FL1, FL1%FL4, FL4-%FL1, FL2-%FL4, and FL4-%FL2. All combinations of FL3 are dependent upon what antibody-conjugated tandem is used, so a unique compensation matrix must be generated for each combination of antibodies. This is accomplished by correcting the basic.cmp matrix just generated for each tandem. 1. Open the listmode file containing the tandem to be compensated, as shown in Figure 10.15.9 for CD33-PE-Cy5. 2. Open the compensation matrix basic.cmp. Remove compensation by unchecking the box in the lower left corner and select trace lines. 3. Adjust only the trace lines (A) FL2-%FL3 and (B) FL4-%FL3 (bold in Fig. 10.15.9) in the same way as shown in Figure 10.15.8. Name and save the compensation matrix for this combination, e.g., comboname.cmp. 4. Repeat this process for each tandem.Whenever a tandem antibody lot is changed, it is imperative that the compensation matrix be compared to the one produced by the old lot. The likelihood that they will be the same is rare. This means that over time, several different compensation matrices for the same antibody combination will be generated. To avoid logistical problems, the correct matrices must be linked to their listmode files. One method to do this is to archive all listmode files along with their compensation matrices on a daily basis. Data Processing and Analysis
10.15.11 Current Protocols in Cytometry
Supplement 23
DATA FROM SPECIMENS Once the compensation matrices have been saved, they can be applied to any data acquired uncompensated from samples generated using each antibody combination. Open the data file containing data to be analyzed. Open the appropriate compensation matrices, using the open button in the compensation dialog box (Fig. 10.15.7); that is all there is to it. USING A MACRO It is possible to automate the entire data-analysis process by using a macro. This makes data analysis of multiple samples very fast. For these steps, refer to WinList help or the tutorial, if necessary. 1. Create a setup macro by producing the desired views such as those shown in Figure 10.15.6. Save as setup.mac using the Save Protocol command in the File dialog menu. 2. Place listmode data files in a batch using the File dialog menu. 3. Starting with the first file, record a macro. “Get next batch, Open file comboname.cmp. Perform any other desired steps, such as applying a special region file or gate logic, or adding histograms to a multigraph. 4. Save the macro as comboname.mac. Each macro can be batched and synchronized with the batched listmode files and analyzed automatically. The correct compensation matrix will be applied each time.
Literature Cited Loken, M.R., Parks, D.R., and Herzenberg, L.A. 1977. Two-color immunofluorescence using a fluorescenceactivated cell sorter. J. Histochem. Cytochem. 25:899-907. Shapiro, H.M. 1995. Practical Flow Cytometry. 3rd ed. pp.160-162.Wiley-Liss, New York. Stewart, C.C. and Mayers, G.L. 2000. Kinetics of antibody binding to cells. In Immunophenotyping, (C.C. Stewart and J. Nicholson, eds.) pp. 1-21. J.Wiley & Sons, Inc., New York. Stewart, C.C. and Stewart, S.J. 1999. Four color compensation. Comm. Clin. Cytometry 38:161-175. Stewart, C.C. and Stewart, S.J. 2001a. Cell preparation for the identification of leukocytes. In Methods in Cell Biology, vol. 64 (Z. Darzynkiewicz, H. Crissman, and J.P. Robinson, eds.) pp. 218-270. Academic Press, New York. Stewart, C.C. and Stewart, S.J. 2001b. Multiparameter data acquisition and analysis of leukocytes by flow cytometry. In Methods in Cell Biology, vol. 64 (Z. Darzynkiewicz, H. Crissman, and J.P. Robinson, eds.) pp 289-312. Academic Press, New York. van Vugt, M.J., van den Herik-Oudijk, I.E., and van de Winkle, J.G. 1996. Binding of PE-CY5 conjugates to the human high-affinity receptor for IgG (CD64). Blood 88:2358-2361.
Contributed by Carleton C. Stewart and Sigrid J. Stewart Roswell Park Cancer Institute Buffalo, New York
This work was supported by a NIH core grant and NYS DOH. The authors thank David Sheedy and Earl Timm Jr. for their editing and clarifications.
A Software Method for Color Compensation
10.15.12 Supplement 23
Current Protocols in Cytometry
CHAPTER 11 Microbiological Applications INTRODUCTION or more than a decade, flow cytometry has been repeatedly proposed as a valuable new addition to the microbiologist’s toolbox, but it has taken a far longer time to be adopted than was initially expected. The purpose of this chapter is to present potentially valuable applications and to introduce the basic tenets underlying the use of flow cytometry in microbiology. The goals governing the selection of material have been two-fold: to assist new users by presenting accurate, detailed, and complete procedures, and to ensure that the protocols are as user-friendly and easy to follow as possible, so that flow cytometrists can present them to their colleagues outside the field as attractive alternatives to existing microbiological methods.
F
outlines the capabilities of microbial flow cytometry technology and provides an overview of current and future applications. A related unit in the chapter on flow cytometry instrumentation, UNIT 1.11, discusses the technical issues involved in designing flow cytometry instrumentation to measure microbes, illustrating how issues in microbial technology relate to broader issues in flow technology. UNIT 11.1
The next three units serve as a broad introduction to the field. UNIT 11.2, on flow cytometry and environmental microbiology, provides an excellent discussion of the intrinsic problems encountered in examining small particles. It also gives a very useful introduction to molecular identification techniques, using tools that are likely to be of increasing importance as the fields of flow cytometry and microbiology merge. In evaluating microbial samples by flow cytometry, it is necessary to deal with the question of organism viability—especially since the viability criteria applied to microbial systems are quite different from those for the mammalian systems discussed elsewhere in this manual. UNIT 11.3 presents a thorough discussion on the estimation of microbial viability and explains why some of the true and trusted methods are inadequate. The unit also includes several sample protocols, as a basis for users, although the assays must be adjusted for the particular application and microorganism of interest and then carefully validated before being applied in any research, clinical, or service form. Another very useful application of flow cytometry is in sorting mixed populations of organisms to establish pure cultures of a particular species or other definable subset of cells. UNIT 11.4 provides detailed instructions for sorting organisms using a cytometric cell sorter, using the separation of GFP-expressing from GFP− bacteria as an example; information on culture, sorting, reculture, and verification of sort purity and GFP expression is included. One of the goals of Current Protocols in Cytometry is to develop a set of clinically relevant applications of flow cytometry. UNIT 11.5 deals with identification of the spirochete Borrelia burgdorferi, the causative agent of Lyme disease, and detecting the presence of serum antibodies against it. B. burgdorferi has traditionally been detected using a plate culture assay; the alternative method presented here is a clever application of flow cytometry in which acridine orange uptake is used as a means to detect killed bacteria, which reflect the presence of serum antibody. Microbiological Applications Contributed by J. Paul Robinson Current Protocols in Cytometry (2004) 11.0.1-11.0.3 C 2004 by John Wiley & Sons, Inc. Copyright
11.0.1 Supplement 29
UNIT 11.7 presents an assay for Mycobacterium tuberculosum testing by flow cytometry. This unique and innovative implementation demonstrates a rapid, safer approach to the identification of antibiotic resistance in tubercle bacilli. UNIT 11.8 provides a general overview of antibiotic susceptibility testing and is an excellent starting point for those interested in applying flow cytometry in this area.
introduces a practical approach to utilizing flow cytometry for the identification of foodborne pathogens. The unit deals specifically with Escherichia coli, although the method could be adapted to numerous organisms. Such approaches have the potential to significantly speed up the detection and characterization of food contamination.
UNIT 11.6
UNIT 11.9 represents a significant advance in the application of cytometry to microbes. This
unit develops the principle of biomass determination using flow cytometry and provides all the tools necessary to complete biomass calculations. Standard curves are calculated using the relationship between forward scatter and cellular dry mass for bacteria of a particular shape. The unit provides an algorithm and computer program to enable rapid application of this technique. deals with the cytometry of yeasts. This unit provides all the knowledge necessary for understanding yeast analysis by flow cytometry, including staining, cell cycle, and data analysis.
UNIT 11.10
broadens the coverage of microbial cytometry by discussing methods for enumeration of phytoplankton, bacteria, and viruses. It provides detailed methodology as well as numerous examples of sample data plots to give experimentalists a sound idea of what one should see in terms of data analysis. While the authors focus on marine samples, the principles hold for other sample types as well.
UNIT 11.11
UNIT 11.12 is a very comprehensive unit dealing with the application of cell-cycle analysis to marine organisms. Significant differences exist between phytoplankton DNA and mammalian or plant-cell DNA, both of which have already been discussed in detail in other units. In addition, this unit introduces the use of fluorescent oligonucleotide probes targeted to 16S rRNA that permit discrimination of specific taxa in the heterogeneous communities of phytoplankton.
Staining protocols generally designed for the flow cytometric analysis of the cell cycle in mammalian cells are frequently not satisfactory for quantification of the various cell-cycle phases in yeasts. UNIT 11.13 specifically discusses the cell cycle of yeasts and provides a detailed accounting of the methods for dealing with yeasts and the interpretation of the data. The authors carefully explain how significant reduction in CVs can be obtained with a staining procedure using the sensitive nucleic acid stain SYBR Green 1, thereby improving the accuracy of DNA content measurement and estimates of populations in cell-cycle compartments. presents protocols on the flow cytometric assessment of drug susceptibility in Leishmania infantum promastigotes. This rather safer and easier flow cytometric method, based on 5-(and-6)-carboxyfluorescein succinimidyl ester, a relatively inexpensive system which works very well in flow cytometry, allows one to move away from the more traditional 3 H-thymidine assays for determining cell division. This very comprehensive unit details the preparation and staining of the Leishmania promastigotes using PI and SYBR-14 as well as assessment of promastigote cellular protein content using FITC. With its useful protocols for evaluating drug resistance, this unit expands the application of flow cytometry techniques to important microbial-related areas such as parasite resistance, parasite-drug interactions, and cellular toxicity. UNIT 11.14
Introduction
11.0.2 Supplement 29
Current Protocols in Cytometry
Flow cytometry is particularly suited for analysis of aquatic bacteria because it is able to perform rapid multiparametric analysis on individual cells at cell numbers representative of natural environments. UNIT 11.15 presents a nucleic-acid double-staining protocol (NADS) for flow cytometry that can distinguish the fractions of viable, damaged, or membrane-compromised cells within the marine or freshwater free-bacterial community, while distinguishing them from debris. The NADS protocol is based on the simultaneous utilization of two nucleic acid stains, SYBR Green and PI. The efficiency of the double staining is magnified by the FRET from SYBR Green to PI when both are bound to the nucleic acids. Oxidative stress plays a key role in a number of physiological, pathological, and toxicological issues, and the analysis of reactive oxygen species has become an important application for functional flow cytometry. The use of such assays in live bacteria is limited to a certain extent by the presence of the bacterial cell wall, which impedes penetration of vital dyes. UNIT 11.16 presents protocols for studies of oxidative stress in E. coli WP2 strains, which possess an altered cell-wall lipopolysaccharide that results in increased cell membrane permeability. Analysis of WP2 strains offers a convenient alternative for assays of bacterial function. An increasing number of methods originally designed for mammalian cell analysis are being adapted to microbiology. One area of considerable interest is disease prevention through the screening of food, water, and raw materials prior to human consumption. UNIT 11.17 focuses on the application of cytometry to the detection and enumeration of pathogens encountered in foods and pharmaceuticals, with recommendations on dealing with real-world samples. The first protocol demonstrates system setup, validation, and optimization for pathogen detection. Other protocols present several methods of labeling microorganisms: direct and indirect labeling with fluorochromes, and fluorescence in situ hybridization using peptide nucleic-acid probes. A support protocol outlines the steps to establish correlation between flow cytometry data and traditional microbial plate counts. This long and detailed unit establishes a solid set of standards and analysis routines for safe handling of pathogens in the cytometry environment. J. Paul Robinson
Microbiological Applications
11.0.3 Current Protocols in Cytometry
Supplement 29
Overview of Flow Cytometry and Microbiology In recent years flow cytometry has become a relatively common, everyday technique in immunology and hematology laboratories. In almost any situation where it is necessary to phenotype a cell population, identify an antigen of interest, or determine the cell cycle status of a population, the time and effort required to develop a flow cytometric approach will be richly justified. Flow cytometry is widely accepted as a mature technology, and the applications as necessary and even desirable. For a long time, however, the application of flow cytometry to microbiology—although in theory inviting—has in practice remained an idea whose time has never quite come.
HISTORICAL DEVELOPMENT The application of flow cytometry to the analysis of microbial systems has been a slow and difficult process. Early attempts in the 1970s showed promise (Hercher et al., 1979; Hutter and Eipel, 1979), with initial possibilities that appeared almost too good to be true— for example, rapid identification, rapid determination of antibiotic resistance, rapid enumeration, and an ability to provide quantitative information in a field that otherwise essentially lacked high technology. The 1980s saw the implementation of several innovations in system design along with an enhanced understanding of the nature of small-particle light scatter (Salzman et al., 1982), both of which facilitated the detection of small particles (Steen, 1986). It was freely predicted that flow cytometry would rapidly develop into a clinically applicable technique (Ingram et al., 1982; Boye et al., 1983). Although flow cytometry gained almost immediate acceptance in the hospital pathology and immunology environment, microbiology
UNIT 11.1
laboratories were essentially oblivious to the technology, for a number of reasons. One immediate stumbling block was cell size. As discussed by Shapiro (2003), the difference in size and volume between microbial and mammalian cells is enormous (see Table 11.1.1). Because of the lack of interest from microbiologists and the difficulty of the engineering concepts, flow cytometry instruments were not designed to measure microorganisms, but rather cells in the range of 5 to 15 µm. In practice, the measurement of smaller particles, while possible, often required modifications to the instrument or a greater understanding of and interest in the technological aspects of cytometry than was generally possessed by those with any expertise in microbiology. A second problem was the notion that the fluorescent dyes used in flow cytometry were better understood in mammalian systems and relatively poorly understood in microbial systems. A great deal of effort has been put into developing fluorescent probes appropriate to the biochemical characteristics of mammalian cells. Unfortunately, this focus on mammalian systems resulted in a general failure to link microbiologists with those interested in fluorescence measurement, and hence the philosophy negatively affected the widespread acceptance of the technology. A third factor was the cost of instrumentation. Microbiology has never been considered a high-technology field, and few microbiology laboratories would normally consider making the enormous outlays on technology that are considered reasonable in a pathology laboratory. The cost of flow cytometry instrumentation was generally felt to be prohibitive; so were the costs per operation, when compared to the few cents per test for a bacterial identification using traditional techniques. For all
Table 11.1.1 Relative Size Ratios for Bacteria, Yeast, and Eukaryotes
Measurement
Bacteria
Yeast
Eukaryote
Diameter
0.5-5
3-5
10-30
Surface area
3-12
30-75
300-3000
Volume
0.3-3
20-125
500-1500
1
10
300-3000
Dry cell mass
Microbiological Applications Contributed by J. Paul Robinson Current Protocols in Cytometry (2004) 11.1.1-11.1.4 C 2004 by John Wiley & Sons, Inc. Copyright
11.1.1 Supplement 30
these reasons, flow cytometry technology has heretofore interested only research microbiologists, and has not had any substantial impact on the field of microbiology as a whole (which includes environmental, public health, medical, food science, industrial, and military applications).
APPLICATIONS Environmental Microbiology Marine and environmental microbiologists have been among the first to recognize the potential of flow cytometry (Amann et al., 1990; Allman et al., 1993; Cunningham, 1993; Edwards et al., 1993; Tarran and Burkill, 1993; Troussellier et al., 1993). Environmental microbiology brings with it certain problems for which flow cytometry can to some extent provide solutions. Perhaps the most significant of these is the issue of culturability of organisms. There is a traditional microbiological viewpoint that an organism must be culturable in order to be designated “viable.” However, an enormous number of organisms remain unstudied, unclassified, even undiscovered, because their specific culture conditions are not known. Currently a fierce debate rages within the microbiology community concerning the existence of organisms in a state termed viable but not culturable (VBNC). The controversy highlights the fact that many of the organisms that exist cannot be cultured, and therefore cannot be identified (Bogosian, 1998). The use of flow cytometry as a microbiological tool casts new light on the controversy: with this technique, it is not necessary to culture an organism to determine its viability, at least in terms of metabolic state. Using vital fluorescent dyes, it is possible to identify populations of organisms that, although they may still be unculturable, are at least definitely “not dead.” Work on this complex issue has opened a fertile area for creating new detection methods by linking up-to-date molecular techniques with rapid analytical technologies such as flow cytometry. Molecular tools can be used to create new microbial probes that can easily be converted to fluorescent conjugates suitable for flow cytometry.
Bioterrorism and Detection of Biological Warfare Agents Overview of Flow Cytometry and Microbiology
With the development of advanced genetic engineering technologies, it is possible to develop microorganisms that can produce as much as 100 times more pathogen or toxin per cell than that which is produced by naturally
occurring strains. Such enhanced pathogens can be significant weapons in the hands of bioterrorists. Bioterrorism agents are classified in five groups: bacteria, viruses, rickettsiae (which have characteristics common to both bacteria and viruses, but like the latter will grow only within other cells), chlamydia (which are obligate intracellular parasites incapable of generating their own energy sources), fungi (and importantly fungal spores), and finally toxins. Each of these agents presents a significantly different problem in the area of detection. While many current biowarfare detection kits depend on antibodies reacting with the antigenic surface coatings of pathogenic bacteria or viruses, immunologically based detection mechanisms have advantages and disadvantages.
Detection systems related to flow cytometry Flow cytometry can provide rapid, accurate, and quantitative information about airborne and waterborne pathogens and perhaps even toxins. One area where these characteristics may be of considerable utility is the detection of biological warfare agents. Culture systems that require several hours to identify microorganisms are of little value to front-line soldiers or civilian populations faced with possible biological assault. Flow cytometry has the advantage of being able to differentiate between nonbiological and biological particles, and, perhaps even more importantly, determine whether or not any organisms that are found are alive. Although molecular tools are often considered to be superior because tests can usually be done in batches, and because of the perception that they are more accurate, it is not clear that this is always the case. For example, consider the situation of a possible biological weapon. A sample is collected that may contain biological agents. Using molecular techniques, an organism or a spore is determined to be present, but there is no way to tell if it is alive or viable. In contrast, viability can be determined for many organisms in a relatively short time using flow cytometry. In addition, flow cytometry can differentiate very quickly between biological and nonbiological samples— a task that is considerably more difficult using molecular tools, since a negative answer may be somewhat less convincing and less conclusive. Unfortunately, after a major drive in the late 1990s to develop new instrumentation and fluorescent indicators for microbiology, it appears this effort has again fallen short. It is conceivable that in rapid-detection scenarios, flow
11.1.2 Supplement 30
Current Protocols in Cytometry
cytometry will fail to gain acceptance until a future crisis emphasizes once again how powerful this technology is for identifying small particles in suspension. Regardless, the recent short burst of activity has raised the level of interest and knowledge concerning the application of flow cytometry to microbial systems. One new opportunity is offered by multispectral technologies with the capability of using advanced mathematically based classification systems. Although these technologies are only just emerging, it is predictable that such developments will impact both the speed and accuracy of flow cytometry–based technologies. A number of technologies were recently discussed as “next-generation” detection possibilities, but it is clear that at present detection solutions contain more proposal than reality (Spencer and Lightfooty, 2001).
Food Microbiology Perhaps one of the most useful potential applications of flow cytometry is in food microbiology. Foods are easily reduced to liquid form, the natural sample state for flow cytometry. The difficulty lies in the need to remove the perhaps 99.9999% of the particles present that are normal, and irrelevant to the measurement, before one can observe the 1 in 106 that is a living organism, and potentially requiring of further study. Thus, enrichment of the microbial population is often still necessary. It is also necessary to be able to identify the organism of interest—for example, E. coli O157:H7 in juices or foods—among the many other microbes that may be present; identifying a particular organisms or strain requires specific monoclonal antibodies. Routine use of flow cytometry technology may not be feasible given the lack of easy-to-use flow-based protocols, but in epidemic situations its use should be considered a real possibility. Of course, the development of protocols for sorting bacteria, such as those in UNIT 11.4, will help to change this state of affairs. Another, related application might be the detection of Cryptosporidium or Giardia in water supplies. Recent outbreaks of Cryptosporidium in normally safe drinking systems (in the state of Wisconsin, USA, and the city of Sydney, Australia) have brought it to the attention of microbiologists that there is some confusion and lack of understanding even among experts. In Sydney, for example, Cryptosporidium was detected repeatedly over a period of several months in the normally safe water supply. No clear foci of infection have been identified, and few if any
cases of cryptosporidiosis have been clinically identified. Even with expert flow cytometry available, no persuasive evidence has been provided that the organisms can be reproducibly identified. The problems are many: the lack of good antibodies for identifying Cryptosporidium, the very small numbers of organisms, and the difficulty routinely experienced in identifying these particular organisms absolutely. The impact of flow cytometry—admittedly a valuable research tool—in this sort of situation remains unclear.
FUTURE DEVELOPMENT New molecular techniques may hold promise for flow microbiology. Flow identification and sorting using fluorescence in situ hybridization (FISH) techniques offer great potential. Amann (1995) has demonstrated a clearly effective application of 16S rRNA probes in microbial ecology. More recently Wallner et al. (1995) were able to use this approach to separate labeled subpopulations which were subsequently used as templates for PCR amplification of the 16S rRNA gene. Application of such combinations of flow and molecular techniques to highly mixed populations such as those in sludge or soil provides unique solutions only achievable by a combination of flow and molecular techniques. Among the real issues that will affect the use of flow cytometry technology in microbiology, the first and foremost is bound to be cost. The perception exists in clinical microbiology at least, that flow cytometry and similar technologies will never be able to perform as cost effectively as traditional culture methods. “Pennies a test” is an expression frequently quoted in clinical laboratories; this is certainly difficult to achieve when the expenditure of $100,000 or more for a flow cytometer must be factored in. The second real problem is the lack of knowledge and understanding about the capabilities of flow cytometry on the part of clinical microbiologists who have been told over and over for the past 10 years that the time for flow cytometry in microbiology is imminent. It is also true that microbial diversity and the unique ability of organisms to alter their antigenic expression make it difficult to use antibodies as effectively as in human immunophenotyping. It is clear, however, that new alternatives using bead technologies and cheaper, easier-to-use instruments will make flow cytometry more attractive and perhaps even less expensive than the current techniques it can replace. Bead technologies promise to bring major changes to flow cytometry and
Microbiological Applications
11.1.3 Current Protocols in Cytometry
Supplement 30
microbiology. The ability to create these small “identification” laboratories within test tubes may be a powerful addition to the tools flow cytometry can offer the microbiologist. Ever since Current Protocols in Cytometry was in the planning stages, coverage of flow cytometry and microbiology has been under continued discussion. The editors now believe that bringing specific methods and ideas to the bench scientist in accurate, reproducible, and detailed format can foster a real and measurable growth in application of flow cytometry to microbiology. The availability of high-quality methods should encourage experiments by many people who previously might have felt the techniques were too complex and the instrumentation too difficult to manage. The provision of both commentary and protocol units will, we hope, be a self-fulfilling prophecy in the long-awaited progression of flow cytometry into the field of microbiology. Those who have for years been developing expertise at the interface between the two fields should now be encouraged to share this hardearned knowledge and provide all the assistance they can to colleagues who wish to implement the sometimes difficult assay systems in flow cytometry. By combining all the above developments, together with a greater desire of microbiologists to encompass new technologies, it is possible that flow cytometry and microbiology will finally complement each other.
Literature Cited Allman, R., Manchee, R., and Lloyd, D. 1993. Flow cytometric analysis of heterogeneous bacterial populations. In Flow Cytometry in Microbiology (D. Lloyd, ed.) pp. 27-47. Springer-Verlag, New York. Amann, R. 1995. Fluorescently labeled, rRNAtargeted oligonucleotide probes in the study of microbial ecology. Mol. Ecol. 4:543-554. Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R., and Stahl, D.A. 1990. Combination of 16S rRNA–targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56:1919-1925. Bogosian, G. 1998. Viable but nonculturable, or dead? ASM News 64:547.
Boye, E., Steen, H.B., and Skarstad, K. 1983. Flow cytometry of bacteria: A promising tool in experimental and clinical microbiology. J. Gen. Microbiol. 129:973-980. Cunningham, A. 1993. Analysis of microalgae and cyanobacteria by flow cytometry. In Flow Cytometry in Microbiology (D. Lloyd, ed.) pp. 131-142. Springer-Verlag, New York. Edwards, C., Diaper, J.P., Porter, J., and Pickup, R. 1993. Applications of flow cytometry in bacterial ecology. In Flow Cytometry in Microbiology (D. Lloyd, ed.) pp. 121-129. Springer-Verlag, New York. Hercher, M., Mueller, W., and Shapiro, H.M. 1979. Detection and discrimination of individual viruses by flow cytometry. J. Histochem. Cytochem. 27:350-352. Hutter, K.-J. and Eipel, H.E. 1979. Microbial determinations by flow cytometry. J. Gen. Microbiol. 113:369-375. Ingram, M., Cleary, T.J., Price, B.J., Price, R.L., and Castro, A. 1982. Rapid detection of Legionella pneumophila by flow cytometry. Cytometry 3:134-137. Salzman, G.C., Griffith, J.K., and Gregg, C.T. 1982. Rapid identification of microorganisms by circular-intensity differential scattering. Appl. Environ. Microbiol. 44:1081-1085. Shapiro, H.M. 2003. Practical Flow Cytometry, 4th ed. Wiley-Liss, New York. Spencer, R.C. and Lightfooty, N.F. 2001. Preparedness and response to bioterrorism. J. Infect. 43:104-110. Steen, H.B. 1986. Simultaneous separate detection of low angle and large angle light scattering in an arc lamp-based flow cytometer. Cytometry. 7:445-449. Tarran, G.A. and P.H., Burkill 1993. Flow cytometry at sea. In Flow Cytometry in Microbiology (D. Lloyd, ed.) pp. 143-158. Springer-Verlag, New York. Troussellier, M., Courties, C., and Vaquer, A. 1993. Recent applications of flow cytometry in aquatic microbial ecology. Biol. Cell 78:111-121. Wallner, G., Erhart, R., and Amann, R. 1995. Flow cytometric analysis of activated sludge with rRNA-targeted probes. Appl. Environ. Microbiol. 61:1859-1866.
Contributed by J. Paul Robinson Purdue University West Lafayette, Indiana
Overview of Flow Cytometry and Microbiology
11.1.4 Supplement 30
Current Protocols in Cytometry
Flow Cytometry and Environmental Microbiology Environmental microbiology is a methodslimited science. Microorganisms are of fundamental importance in energy flows through ecosystems, responsible for many rate-limiting steps in biogeochemical cycles. Although microbiological populations comprise huge numbers of individual cells, only limited knowledge is available on a few criteria regarding the majority of these cells. New methods provide hints about the complexities of microbiological cells and their interactions in ecosystems, as well as possible oversimplifications and errors in some previous assumptions. Many studies that address the structure and functioning of food webs are based upon in situ measurements of the turnover of organic and inorganic carbon, nitrogen, phosphorus, and sulfur. The information gained has been invaluable, and has allowed modeling of global food webs, but has not provided sufficient details on the interactions of individual cells or species within an environment. Many of the methods used by environmental microbiologists have been inherited from medical microbiologists. Understandable interest in infection prompted investigations into pathogenic organisms, and crudely reproducing the body environment in a petri dish and incubator allows culture of many infectious bacteria. However, when applied to environmental samples, these methods have allowed culture of only a small fraction (typically 1% to 10%) of the cells visible on microscopic examination. Thus considerable effort has been devoted to developing direct methods of analyzing cells from such samples, avoiding the need of a culture step. These new techniques, particularly those featuring some aspect of molecular biology and fluorescence dye technology, have opened up many new and exciting avenues of basic microbiological research, making a comprehensive review beyond the scope of this contribution. Finding specific cells against a background that may consist of millions of morphologically similar nontarget cells as well as contaminating particles is both time-consuming and difficult. Traditionally, target cell numbers are increased by a culture step of some sort, allowing subsequent detection. However, given the selectivity introduced by culture from most situations, this approach has obvious limitations. When Contributed by Jonathan Porter Current Protocols in Cytometry (2004) 11.2.1-11.2.13 Copyright © 2004 by John Wiley & Sons, Inc.
UNIT 11.2
applying direct methods, it is often necessary to extract and/or purify target cells before attempting the analysis step. For ecological studies, it is easy to criticize these procedures as introducing bias and leading to results unrepresentative of reality. However, most extraction steps report cell recoveries >30% of the total number, and many reports suggest much more efficient recoveries, demonstrating that the bias introduced during cell extraction is not as great as that introduced by a culture step. Flow cytometry (FCM) and cell sorting cannot solve all the methodological problems of environmental microbiology, and their application is limited in certain situations. However, the ability to generate multiparameter data on millions of individual cells while allowing maintenance of cell viability, and subsequent recovery of subpopulations of interest, is unique in microbiological studies. Even standard instruments are sensitive enough to detect bacterial spores using light scatter alone, and indigenous free viruses (and viral-like particles) in marine waters have been detected after staining with fluorescent nucleic acid dyes (Marie et al., 1999; UNITS 11.11 & 11.12). An enormous amount of information can be gained from each cell, especially given the huge and ever-expanding range of fluorescent marker dyes that can be applied to the analyses. Finally, the option of cell sorting allows physical separation of specific cells of interest into culture media, or into defined buffers, enriched and purified to allow successful application of most culture or molecular biological techniques. FCM is not a stand-alone method, but should be thought of as an analysis technique that leads to biologically relevant data and sample processing.
APPLYING FLOW CYTOMETRY AND CELL SORTING TO ENVIRONMENTAL MICROBIOLOGY Sample preparation and instrument setup will determine the success of any FCM analysis. Flow detection of microorganisms requires more stringent preparation procedures than detection of larger cells, and it is worthwhile to ensure that background signals are as low as possible. All buffers and laboratory reagents are filtered at least three times before use, and it is
Microbiological Applications
11.2.1 Supplement 27
Flow Cytometry and Environmental Microbiology
preferable to rinse glassware and the like with particle-free solutions. When working with cell extractions from soil, sediment, or sludge, it is difficult enough to recognize bacteria against the inherent background of bacterial-sized particles without adding unnecessary noise. Thus the first step in applying FCM to environmental samples is to establish a cell extraction procedure that leaves representative cells in a suspension free from any particles large enough to block the nozzle of the instrument. Laboratory cultures are generally suitable after dilution, without the need for a cleanup step, although the culture medium may be filtered before use. Sample preparation may range from none at all (e.g., when analyzing lake water bacteria) to extensive blending, centrifugation, and enrichment steps (e.g., when processing soil samples). Extensive sample preparation may introduce bias into the composition of the final cell suspension. Generally aquatic samples require only a vigorous shaking and settling step, followed by dilution or crude filtering through nylon mesh before analysis, resulting in minimal change. FCM analysis of bacteria from more challenging environments has been performed (Page and Burns, 1991; Lange et al., 1997) but the degree of success has generally depended upon the ease and effectiveness of cell extraction. Instrument setup is also important, with regular cleaning and sterilization required. Calibration and alignment for bacterial studies are critical. Monodisperse beads of 0.5-µm diameter are useful for instruments using a high-numerical-aperture objective lens for light collection. However, for jet-in-air systems and for highly particulate samples, it may be preferable to use larger beads so as to allow cells to follow slightly different paths through the sensing region. Subsequent data handling and analysis will depend upon the instrument specifications, although most machines now save files in an FCM standard format (UNIT 10.2). Several software packages (including freeware examples) exist to help in FCM data analysis (see also Chapter 10). Once a cell suspension that will not block or clog the cytometer tubing has been obtained, it is necessary to label cells in order to distinguish them from noncellular particles. Generally this will require a fluorescent label, although some morphologically distinct cells can be differentiated from background using inherent light scatter characteristics (Fig. 11.2.1). Prokaryotes such as Achromatium oxaliferum (Head et al., 1996), although never cultured, may be
extracted into a suspension sufficiently clean for most applications. These large cells, with many cytoplasmic inclusions, are distinguishable on the basis of light scatter alone (Fig. 11.2.1A), but use of fluorescent probes still increases confidence in the accuracy of the FCM detection. Preliminary work targeting Ochrobium cells (another genus of larger, morphologically distinct bacteria) from anoxic, hypereutrophic lake water and sediments used cell sorting to attempt cell purification (Gray, N.D., Head, I.M., Pickup, R.W., and Porter, J., unpub. observ.). As predicted, the cells appeared at the high end of the light-scatter distribution, but they were not the largest bacterial cells present in the samples. Cell sorting using gates set on forward scatter yielded samples enriched in Thiopedia cells, whereas samples sorted using gates with a decreased scatter were enriched in Ochrobium cells (Fig. 11.2.1B). However, exceptions of this sort tend to occur only with larger cells; for the majority of bacteria, a fluorescence discriminator label is required. Some bacterial populations can be distinguished on the basis of autofluorescence of specific pigments, but many other procedures require an added label. Choice of fluorescent label is dependent upon the experimental aims. Samples may be probed for a total bacterial cell count, a viable or active cell count, a specific cell count, or an indication of cell macromolecular content (e.g., DNA, RNA, or total protein). Nucleic acid labeling allows analysis of the cell cycle in natural populations. Sample labeling protocols are obviously dictated by the fluorescent dye being used. In some cases, dye binding is strongly influenced by salt concentration, which has caused problems when labeling nucleic acid in marine bacteria. Protocols often require washing and resuspension steps to remove unbound dye. If enumeration of cells is important in the experimental aims, it may often be better to adjust sample conditions using concentrated buffers, and/or choose dyes and protocols that do not require washing steps, in order to avoid cell damage and loss. Each centrifugation step causes the loss of some bacteria, particularly those with a high cell surface hydrophobicity (e.g., Aeromonas salmonicida; Deere et al., 1996). An immense (and ever-increasing) number of fluorescent probes now exist for use in biological research. Many were developed for the study of eukaryotic cell biology, but bacterial applications are rapidly increasing. The majority of these applications have used dyes for cell
11.2.2 Supplement 27
Current Protocols in Cytometry
A 1000
Achromatium
Side-scatter intensity
800
600
400
200
0 0
B
200
400
600
800
1000
Forward-scatter intensity 300
250
Relative number
enriched in Ochrobium 200
150
enriched in Thiopedia 100
50
0
0
200
400
600
800
1000
Forward-scatter intensity
Figure 11.2.1 Detection of indigenous bacteria from freshwater environment using light scatter alone. (A) Achromatium oxaliferum; (B) Ochrobium spp. and Thiopedia spp. Despite their size, all three cell types required confirmatory analysis, either by fluorescent in situ hybridization with specific oligonucleotide probes against Achromatium (Head et al., 1996) or by cell sorting and confirmatory microscopic examination (Ochrobium and Thiopedia; Gray, N.D., Head, I.M., Pickup, R.W., and Porter, J., unpub. observ.).
Microbiological Applications
11.2.3 Current Protocols in Cytometry
Supplement 27
enumeration and viability assessment. Specific detection is achieved through use of antiserum labeled with standard fluorochromes such as fluorescein or phycoerythrin or through fluorescent in situ hybridization (FISH) using ribosomal RNA (rRNA)–directed oligonucleotides. These oligonucleotides are also often labeled with standard dyes, although modern, brighter alternatives have been investigated. The choice of fluorochrome is influenced by the aims of the work (e.g., measuring other dyes or pigments) and by the instrument light source. As cytometers become more amenable to work aboard ship and in the field, the demand for dyes that will allow analysis of photosynthetic pigments, cell cycle analysis, nucleic acid content, and membrane integrity from single-lightsource instruments will grow. This sort of development has been important, for example, in the study of phytoplankton (Veldhuis et al., 1997). There has also been a move toward dyes that excite and emit further into the red, to overcome problems with autofluorescence from environmental samples.
ENVIRONMENTAL MONITORING OF MICROORGANISMS USING FLOW CYTOMETRY AND CELL SORTING
Flow Cytometry and Environmental Microbiology
The pioneering work demonstrating the potential of FCM in environmental bacteriology has been summarized by Yentsch and Horan (1989). Use of FCM in environmental microbiology has steadily increased since then (for reviews, see Lloyd, 1993; Davey and Kell, 1996; Porter et al., 1996, 1997a) and has already been of considerable utility in certain situations—e.g., in the discovery of the significant marine genus Prochlorococcus (Chisholm et al., 1988). Observations using photosynthetic pigments and DNA analysis have shown the limitations of traditional methods for analyzing photosynthetic marine bacterial plankton biomass (Campbell et al., 1994). Some workers have also used light scatter as an estimator for carbon biomass, and red fluorescence as an estimator for chlorophyll biomass, to differentiate the numbers and contributions of Prochlorococcus, Synechococcus, and eukaryotic algae to oceanic ultraphytoplankton populations and activities (Li, 1995). Although ideally suited to water analysis, where cells come ready-suspended, FCM has been performed on cells sampled and extracted from most types of environment (Table 11.2.1), including soil (Page and Burns, 1991; Porter et al., 1997b), activated
sludge (Wallner et al., 1995, 1997), and air (Lange et al., 1997; Henningson et al., 1998).
SPECIFIC DETECTION OF MICROORGANISMS IN ENVIRONMENTAL SAMPLES USING FLOW CYTOMETRY Autotrophic microorganisms can generally be differentiated by their signature pigments; Prochlorococcus, Synechococcus, and algal species are routinely analyzed in marine samples in this fashion (Blanchot et al., 1997; Campbell et al., 1997; Detmer and Bathmann, 1997). Such signatures have proved useful when estimating grazing rates, as loss of prey cells from suspension can be measured in conjunction with accumulation of prey inside the predator. This approach has been utilized for studies on bivalve grazing on algae (Bougrier et al., 1997) as well as studies on grazing within microbial food webs; in the latter case, fluorescent beads may also be used as a prey alternative (Avery et al., 1995; Kenter et al., 1996). Immunological methods are well established in cytometry, and have much to offer in terms of specific labels for detection and enumeration, as well as purification procedures. Problems with production, specificity, changes in epitope expression with changing growth conditions, and nonspecific labeling of background material are fairly common; however, extremely sensitive and specific detection is possible, including the labeling of intracellular antigens (e.g., bacterial enzymes; Currin et al., 1990). Improvements in detection and in signal/noise ratios have been made by combining monoclonal antibody labeling with propidium iodide staining, an approach used for the detection of Legionella in cooling waters (Tyndall et al., 1985) and of Flavobacterium in soil (Page and Burns, 1991). Despite these improvements, FCM has proved to be of only limited value in testing soil samples (Page and Burns, 1991; Porter et al., 1997b), although samples from other highly particulate environments such as feces and activated sludge have been successfully analyzed (Volsch et al., 1990; van der Waaij et al., 1994, 1996). However, more routine immunofluorescent FCM applications to environmental bacteriology have used sewage or water samples. Vesey et al. (1994a) established immunofluorescence detection and cell sorting as a routine tool for the monitoring of Cryptosporidium in water samples. Sorted cysts can also be encouraged to excyst, to determine viability; changes in light scatter that result from excystation can be detected by FCM
11.2.4 Supplement 27
Current Protocols in Cytometry
Table 11.2.1 Examples of the Application of Flow Cytometry to the Study of Microorganisms from Environmental Samples
Sample type
Experimental aims
References
Soil
Specific detection using FISHa Specific detection using immunofluorescence Viability assessment Enumeration and biomass estimation Viability assessment Specific detection using immunofluorescence Specific detection using FISH Specific detection using FISH and cell sorting Specific detection using immunofluorescence and cell sorting Enumeration and viability assessment Specific detection using immunofluorescence and cell sorting Specific detection using immunofluorescence Enumeration and cell cycle analysis
Thomas et al., 1997 Page and Burns, 1991
Sand Compost Feces Activated sludge
Sewage effluent
Fresh water
Cooling water Sea water
Air
Porter et al., 1997b DeLeo and Bavaye, 1996 Diaper and Edwards, 1994 van der Waaij et al., 1994, 1996; Vesey et al., 1997c Wallner et al., 1995 Wallner et al., 1997 Porter et al., 1993, 1995d
Lebaron et al., 1998b; Porter et al., 1995a; Yamaguchi and Nasu, 1997 Vesey et al., 1994a
Tyndall et al., 1985 Marie et al., 1996, 1997; Monger and Landry, 1993; Montford and Baleux, 1992; Robertson and Button, 1989; Trousellier et al., 1995 Lebaron et al., 1998b
Enumeration and viability assessment Enumeration and specific detection Henningson et al., 1998; Lange et al., 1997
aFluorescent in situ hybridization.
(Vesey et al., 1997c). Porter et al. (1993) used polyclonal antibodies to detect Escherichia coli in lake water and sewage. Particles scoring as positive events were sorted and plated onto selective medium for confirmation. Singlefluorochrome indirect-labeling protocols of this type, however, are susceptible to nonspecific binding of antibodies (Porter et al., 1995d). Straightforward methods for evaluating nonspecific binding of antiserum in environmental samples have been proposed (Vesey et al., 1997a,b). FISH methods label rRNA sequences inside intact cells. This offers a unique opportunity in environmental bacteriology, enabling measurements to be made of the number of individual cells of a target species in a given sample. It also allows confirmation of the presence and activity of cells (i.e., from directly retrieved
rRNA sequences) without the need to culture them. Many studies have been performed (including the analyses) on microscope slides. A review by Amann et al. (1995) provides a comprehensive background to FISH, and discusses sample preparation, probe design, and hybridization conditions (see also Chapter 8). Several FCM analyses of mixed populations of cultured cells have been performed. Wallner et al. (1993) optimized FISH methods for bacterial analysis by FCM and subsequently used these to directly probe the microflora of activated sludge (Wallner et al., 1995). This approach was later extended to the use of cell sorting to separate labeled subpopulations (Wallner et al., 1997). Cells obtained in this way proved to be suitable templates for PCR amplification of the 16S rRNA gene. Data from studies such as these, and from microscopic observations, have dem-
Microbiological Applications
11.2.5 Current Protocols in Cytometry
Supplement 27
Flow Cytometry and Environmental Microbiology
onstrated that the fluorescence signal obtained is proportional to the ribosome content of the cell; hence, in nutrient-poor environments, low ribosome content may render cells difficult to detect. The method has likewise been shown to be useful in analyzing soil bacteria (in conjunction with other fluorescent discriminators; Thomas et al., 1997), airborne bacteria (Lange et al., 1997), marine nanoflagellates (Rice et al., 1997), and algae (Simon et al., 1995, 1997). The FISH approach has been extended by probing for precursor polycistronic rRNA molecules, thus measuring current levels of ribosome synthesis rather than assessing total ribosome numbers. This is thought to more accurately reflect the current state of cell activity (Oerther et al., 2000). Due to the diversity of sequences of bacterial rRNA intragenic regions, this approach may be limited to specific species or groups, and for this reason it is likely that high-level probes will continue to have a major role in FISH probing of prokaryote communities. Indeed, probes directed against higher-order classifications have uncovered some striking findings, such as data demonstrating the broad distribution and importance of Cytophaga-like bacteria in marine waters, initiated by Glockner et al. (1999). Such bacteria have since been investigated further and have been found as significant proportions of procaryote communities in soil and wastewater treatment plants. One method for overcoming the detection limits of FISH is to amplify target nucleic acid sequences inside whole cells. This approach has been developed in histopathology, where high sensitivity is needed to detect viral infections. The approach may be used to label cells containing particular genes of interest. This procedure has been successfully performed on cultured bacterial cells to detect a plasmid-encoded gene of ∼700 bp (Porter et al., 1995c) and the 16S rRNA gene (∼1600 bp; Hodson et al., 1995). The latter study also demonstrated reverse transcription and amplification from mRNA inside whole cells, an approach confirmed by later work (Chen et al., 1997). In situ amplification of nucleic acids should allow both the detection of general areas of the ribosomal molecule (∼760 bp; Tani et al., 1998) and the amplification of longer fragments using specific probes (e.g., ∼950 bp; Tani et al., 1998). In situ amplification has also been used to detect single-gene copies (Hodson et al., 1995), and has been applied to cells within environmental water samples (Tani et al., 1998, 2002). For this work, samples were examined microscopically
and required image analysis to help differentiate cells from nonbacterial particulates. In an FCM analysis, this differentiation could presumably have been achieved by gating.
VIABILITY ASSESSMENT OF MICROORGANISMS USING FLOW CYTOMETRY Fluorescent probes exist for a range of metabolic functions, which aim to reflect cell viability without the need for culture (UNIT 9.2). Bacteriologists have not as yet developed an assay that unambiguously demonstrates viability (Nebe-von Caron and Bradley, 1995; Davey and Kell, 1996; UNIT 11.3). Similar assays are used to assess viability in yeasts (Breeuwer et al., 1995; Abe, 1998), although efflux pumps can complicate matters; these assays can be used to probe either cellular or mitochondrial functioning (Dinsdale et al., 1995). Such applications have great potential in the brewing industry. Fluorescent viability probes used either separately or in combination reflect metabolic activity rather than reproductive potential. However, data obtained using FCM have increased the confidence that can be placed in direct viability assessment methods. As some of the fluorescent viability probes share common excitation and emission wavelengths, they cannot be used for simultaneous sample labeling. The speed of FCM has enabled researchers to process multiple subsamples of cells within an acceptable time period (Porter et al., 1995b; Deere et al., 1996). In these studies, cell viability was assessed based on five characteristics: growth on solid agar medium, response to nutrient addition (nalidixic acid assay, or direct viable count), presence of a membrane potential (using rhodamine 123 and oxonol dyes), membrane integrity, and intracellular membrane activity (using fluorogenic ester dyes). Total counts were also made, using nucleic acid staining and immunofluorescence. The studies show that greater variation in counts was obtained from FCM counting than from microscopy, presumably through nonuniformities in day-to-day instrument setup and operation, emphasizing the need for quality control. However, the FCM/fluorescent probe viability estimates were more similar to each other and to the direct viable (nalidixic acid) count procedure than to the count of culturable cells. The fact that assays of separate cellular functions yielded similar results provides increased confidence that the techniques are effective indicators of cell activity. Further multiparameter
11.2.6 Supplement 27
Current Protocols in Cytometry
work on bacteria in prolonged stationary phase supports this conclusion, showing that the proportion of cells in a subpopulation deemed active approximately corresponded to those in subpopulations deemed to have the higher levels of rRNA and total cell protein (Turner et al., 2000). Use of these fluorescent viability probes is applicable to the study of active cells, or cells whose stress response includes maintenance of metabolic activity. Dyes such as the fluorogenic ester dyes have been used in combination with propidium iodide for simultaneous labeling of live versus dead cells (Yamaguchi and Nasu, 1997). Membrane permeability to nucleic acid stains has also been used as an indicator of viability. The basis for the assay is that live cells maintain membrane integrity and function and exclude the dye, whereas dead or damaged cells are unable to do this, and allow the dye to enter the cell and label nucleic acid (UNIT 11.3). Studies in the gram-positive bacterium Micrococcus luteus have shown that the stress response appears to be one of metabolic shutdown (i.e., dormancy). Dormant M. luteus cells did not form colonies on solid medium, accumulated a nucleic acid dye, and failed to accumulate the membrane potential probe rhodamine 123 (Kaprelyants and Kell, 1993; Votyakova et al., 1994). After appropriate treatment, however, dormant cells could be resuscitated (Kaprelyants and Kell, 1993; Votyakova et al., 1994). When breaking dormancy, the cells showed an initially decreased permeability to the nucleic acid dye followed by accumulation of rhodamine 123, and ultimately became culturable. Cell sorting has been used to confirm that the culturable cells are those that accumulated the rhodamine dye (Kaprelyants et al., 1996); only the use of FCM allowed the intricate measurements required to elucidate this complex phenomenon. Other workers have successfully used membrane permeability to a nucleic acid label as an indicator of viability (Jernaes and Steen, 1994; Mason et al., 1995; Langsrud and Sundheim, 1996; Jacobsen et al., 1997; Joux et al., 1997), with some studies advocating their use in bacterial viability assessment (Williams et al., 1998). However, for natural populations, this approach is more problematic, and can be affected by cell status, sample salinity, and sample storage (Lebaron et al., 1998a,b). Fluorescent viability probe measurements may be further complicated if efflux pumps lead to a false recording of viability. If these, or other, direct methods gain acceptance, they will be more
rapid and more accurate than culture techniques for monitoring viability of bacteria from a variety of sources. Studies on organisms in lake water samples have correlated FCM estimates of metabolic activity (esterase activity) with microscopic estimates (respiring counts; Porter et al., 1995a). However, other studies have suggested that counts of esterase-active bacteria are greater than counts of respiring bacteria (Yamaguchi and Nasu, 1997). Thus correlation between methods in the laboratory may not extrapolate to natural populations. Both these latter studies noted a strong correlation between FCM estimates of metabolic activity and the degree of pollution of the water samples. As noted above, similar methods have proved useful in yeast (Breeuwer et al., 1995; Abe, 1998). Viability estimates may prove useful in a clinical setting, where it may become possible to test for antibiotic activity against an infectious bacterium (Mason et al., 1994; Roth et al., 1997). Such an assay may be suited, for example, to urinary tract infections.
LINKING IDENTITY, FUNCTION, AND ACTIVITY IN SINGLE MICROBIAL CELLS The importance which microbial ecologists attach to specific cell identification and enumeration is reflected in the amount of effort that has been directed into methods development and application. Similar effort has been directed toward determining in situ cell activity, to relate to cell function. The next step of integrating these two approaches is under way, to provide an indication of which cells are both present and active (i.e., responsible for a defined function). It is apparent from the preceding sections that identification is often achieved through ribosomal RNA gene sequence, especially when dealing with uncultured microorganisms. Function can likewise be determined by examining expression of a functional gene (e.g., ammonia or methane monoxygenases). In situ probing for functional genes is still problematic for many microorganisms (although the approach of in situ PCR may hold some promise). Approaches that have been tested include stable isotope incorporation of nucleotides or catabolic enzyme substrates, with subsequent separation of newly synthesized heavy isotope–labeled nucleic acid (Radajewski et al., 2000; Manefield et al., 2002) and combining radiolabeled substrate uptake (assessed by microautoradiography) with FISH (Gray et al., 2000). Neither of these approaches is amenable
Microbiological Applications
11.2.7 Current Protocols in Cytometry
Supplement 27
Flow Cytometry and Environmental Microbiology
to FCM analysis. However, the incorporation of alternative modified nucleotides (e.g., BrdU) has been tested in natural bacterioplankton assemblages with success (Pernthaler et al., 2002), and should be readily adaptable to FCM detection. This latter study combined BrdU incorporation for activity with FISH for the fluorescence identification step. An identification step with only limited resolution (Gram stain reaction using hexidium iodide) has been combined with esterase activity (to identify inactive, active, and very active populations) for bacteria from activated sludge (Forster et al., 2002). This work demonstrated that although less than 5% of the total bacteria labeled as Gram positive, a substantial fraction of the cells designated as very active were Gram positive, and thus of functional significance. Further efforts to link sequence and function using FCM have utilized cell sorting, and it is heartening to see the increasing use of this powerful tool in microbial ecology. Studies on indigenous bacteria from several environments have used a fluorescent label to assess function, followed by cell sorting to further study labeled cells. The apparent DNA content of many aquatic bacteria may be classed as high or low, and Lebaron and co-workers have investigated the correlation between apparent DNA content and activity (Lebaron et al., 2002). These workers coupled cell sorting with radiolabeled leucine incorporation, suggesting that those bacteria with a high apparent DNA content may be responsible for the bulk of activity (as determined by this method). Whiteley et al. (2002) coupled activity assessment with cell sorting to remove actively respiring cells from soil preparations. Following lysis of the sorted cells, amplified ribosomal RNA genes were used to generate diversity profiles. Sequencing of the amplified DNA would have allowed identification of the targeted bacteria. This approach has been successfully demonstrated for aquatic bacteria (Bernard et al., 2000) and for activated sludge samples (Forster, Snape, Lappin-Scott, and Porter, unpub. observ.). It is thus clear how the versatility of FCM and cell sorting could provide a significant tool in microbial ecology. However, a note of caution must be added. Given that a universally accepted activity indicator has not been developed as yet, every functional label may be open to criticism, and the use of the respiratory label 5-cyano-2,3-di4-tolyl tetrazolium chloride (CTC) has been heavily criticized, due to its impairment of cell function (Servais et al., 2001). Other workers have also suggested that the process of cell
sorting may adversely affect some bacterial cells (Resina-Pelfot et al., 2001).
DISCUSSION AND FUTURE PROSPECTS FCM has already proved useful for (environmental) microbiology, and new advances in dye technology, instrument design, and data processing are leading to constant improvements in the technique. Although of limited application in particulate environments, FCM is ideally suited to the analysis of aquatic microorganisms. The instrumentation can be partly automated for routine use, and the options for data analysis are more sophisticated than ever. Neural networks can be trained to recognize subpopulations of microorganisms from raw data and are being developed for examination of naturally fluorescent marine and freshwater planktonic species (Wilkins et al., 1996). Gauci et al. (1996) demonstrated the use of spectral fingerprinting to determine the spectral properties of particles at hundreds of wavelengths simultaneously. Particles can then be assigned to previously characterized optical fingerprints. Pulsed laser sources may also have several applications in bacteriology, allowing time-resolved fluorescence analysis to discriminate between particles according to differences in their fluorescent lifetimes. Other developments include the production of cytometers specifically designed to detect microorganisms, including a battery-operated, portable FCM with a diode laser and fixed optical alignment (eliminating machine setup and focusing) from Aber Instruments. The more sturdy construction of this instrument allows it to be operated in the field (Davey and Kell, 1996). Cytometers have also been developed for use at sea. The Europa (Jonker et al., 1995) was followed by the CytoBuoy (Dubelaar and Gerritzen, 2000), which is now commercially available. Novel dyes are also constantly being produced, and should facilitate multiple labeling of samples, thus increasing the information that can be obtained from each sample. One such area is the use of fixable viability dyes (developed by Molecular Probes). The rationale underlying this approach is that the dye is taken up due to the presence of a proton-motive force, and once inside a cell, becomes aldehyde-fixable. Such fixation would then preserve samples for a reasonable time period, and permeabilize cells for oligonucleotide probing. Labeling of cells in the field, followed by the use of other procedures such as FISH, would allow
11.2.8 Supplement 27
Current Protocols in Cytometry
simultaneous detection of viable and specific cells. Fixable viability dyes for bacterial studies are not as yet commercially available. Other proposals include the use of lectins as aids to discriminate cells from particulates (Payne et al., 1992; Vesey et al., 1994b; Porter et al., 1998) and the use of proteins such as green fluorescent protein (GFP) to report on gene expression (UNIT 9.12). This has been done cytometrically (Tombolini et al., 1997), and several constructs involving GFP have been proposed to help microbiologists study realtime gene expression in situ (Andersen et al., 1998; Christensen et al., 1998). GFP has also been successfully expressed in fungi (Fernandez-Abalos et al., 1998). Promoterless GFP constructs have also been used in promoter trapping to investigate genes that are expressed under environmentally relevant conditions (; Hansen et al., 200; Dunn et al., 2003). Once workers are able to identify which genes to study, further information on the contribution of those genes to an organism’s ecology will follow. Reporters may also be used for gene expression studies, with the presence of a reporter enzyme being detected using substrates of the enzyme that yield a fluorescent product (UNIT 9.5). In situ PCR may also prove useful for studies on gene expression, since the detection of mRNA using FISH has the disadvantage of requiring long probes with multiple labels on each probe in order to generate sufficient signal. Use of long probes requires longer incubation times, and multiple labels can increase nonspecific labeling. The more accurately workers are able to identify cells of interest, the greater the chances of successful cell sorting and recovery of clean suspensions highly enriched in target cells, suitable for culture or molecular biological procedures. Advances such as these should help to establish fluorescence techniques and FCM as necessary and routine tools in environmental microbiology.
LITERATURE CITED Abe, F. 1998. Hydrostatic pressure enhances vital staining with carboxyfluorescein or carboxydichlorofluorescein in Saccharomyces cerevisiae: Efficient detection of labeled yeasts by flow cytometry. Appl. Environ. Microbiol. 64:11391142. Amann, R.I., Ludwig, W., and Schleifer, K.-H. 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59:143-169.
Andersen, J.B., Sternberg, C., Poulsen, L.K., Bjorn, S.P., Givskov, M., and Molin, S. 1998. New unstable variants of green fluorescent protein for studies of transient gene expression in bacteria. Appl. Environ. Microbiol. 64:2240-2246. Avery, S.V., Harwood, J.L., and Lloyd, D. 1995. Quantification and characterization of phagocytosis in the soil amoeba Acanthamoeba castellanii by flow cytometry. Appl. Environ. Microbiol. 61:1124-1132. Bernard, L., Schäfer, H., Joux, F., Courties, C. Muyzer, G., and Lebaron, P. 2000. Genetic diversity of total, active and culturable marine bacteria in coastal seawater. Aquat. Microb. Ecol. 23:111. Blanchot, J., Andre, J.M., Navarette, C., and Neveux, J. 1997. Picophytoplankton dynamics in the equatorial Pacific: Diel cycling from flow cytometer observations. C.R. Acad. Sci. Ser. III Sci. Vie 320:925-931. Bougrier, S., Hawkins, A.J.S., and Heral, M. 1997. Preingestive selection of different microalgal mixtures in Crassostrea gigas and Mytilus edulis analysed by flow cytometry. Aquaculture 150:123-134. Breeuwer, P., Drocourt, J.-L., Bunschoten, N., Zweitering, M.H., Rombouts, F.M., and Abee, T. 1995. Characterization of uptake and hydrolysis of fluorescein diacetate and carboxyfluorescein diacetate by intracellular esterases in Saccharomyces cerevisiae, which result in accumulation of fluorescent product. Appl. Environ. Microbiol. 61:1614-1619. Campbell, L., Nolla, H.A., and Vaulot, D. 1994. The importance of Prochlorococcus to community structure in the central North Pacific ocean. Limnol. Oceanogr. 39:954-961. Campbell, L., Liu, H.B., Nolla, H.A., and Vaulot, D. 1997. Annual variability of phytoplankton and bacteria in the subtropical North Pacific Ocean at Station ALOHA during the 1991-1994 ENSO event. Deep-Sea Res. Part I Oceanogr. Res. Pap. 44:167-192. Chen, F., Gonzalez, J.M., Dustman, W.A., Moran, M.A., and Hodson, R.E. 1997. In situ reverse transcription, an approach to characterize genetic diversity and activities of prokaryotes. Appl. Environ. Microbiol. 63:4907-4913. Chisholm, S.W., Olson, R.J., Zettler, E.R., Goeriche, R., Waterbury, J.B., and Welschmeyer, N.A. 1988. A novel free-living prochlorophyte abundant in the oceanic euphotic zone. Nature 334:340-343. Christensen, B.B., Sternberg, C., Andersen, J.B., Erberl, L., Moller, S., Givskov, M., and Molin, S. 1998. Establishment of new genetic traits in a microbial biofilm community. Appl. Environ. Microbiol. 64:2247-2255. Currin, C.A., Paerl, H.W., Suba, G.K., and Alberte, R.S. 1990. Immunofluorescence detection and characterization of N2-fixing microrganisms from aquatic environments. Limnol. Oceanogr. 35:59-71.
Microbiological Applications
11.2.9 Current Protocols in Cytometry
Supplement 27
Davey, H.M. and Kell, D.B. 1996. Flow cytometry and cell sorting of heterogeneous microbial populations: The importance of single-cell analyses. Microbiol. Rev. 60:641-696. Deere, D., Porter, J., Pickup, R., and Edwards, C. 1996. Direct analysis of starved Aeromonas salmonicida. J. Fish Dis. 19:459-467. DeLeo, P.C. and Baveye, P. 1996. Enumeration and biomass estimation of bacteria in aquifer microcosm studies by flow cytometry. Appl. Environ. Microbiol. 62:4580-4586. Detmer, A.E. and Bathmann, U.V. 1997. Distribution patterns of autotrophic pico- and nanoplankton and their relative contribution to algal biomass during spring in the Atlantic sector of the Southern Ocean. Deep Sea Res. Part II Top. Stud. Oceanogr. 44:299-320. Diaper, J.P. and Edwards, C. 1994. Flow cytometric detection of viable bacteria from compost. FEMS Microbiol. Ecol. 14:213-220. Dinsdale, M.G., Lloyd, D., and Jarvis, B. 1995. Yeast vitality during cider fermentation: Two approaches to the measurement of membrane potential. J. Inst. Brew. 101:453-458. Dubelaar, G.B.J. and Gerritzen, P.L. 2000. CytoBuoy: A step forward towards using flow cytometry in operational oceanography. Sci. Mar. 64:255-265. Dunn, A.K., Klimowicz, A.K., and Handelsman, J. 2003. Use of a promoter trap to identify Bacillus cereus genes regulated by tomato seed exudate and a rhizosphere resident, Pseudomonas aureofaciens. Appl. Environ. Microbiol. 69:11971205. Fernandez-Abalos, J.M., Fox, H., Pitt, C., Wells, B., and Doonan, J.H. 1998. Plant-adapted green fluorescent protein is a versatile vital reporter for gene expression, protein localization and mitosis in the filamentous fungus, Aspergillus nidulans. Mol. Microbiol. 27:121-130. Forster, S., Snape, J.R., Lappin-Scott, H.M., and Porter, J. 2002. Simultaneous fluorescent Gram staining and activity assessment of activated sludge bacteria. Appl. Environ. Microbiol. 68:4772-4779. Gauci, M.R., Vesey, G., Narai, J., Veal, D., Williams, K.L., and Piper, J.A. 1996. Observation of single cell fluorescence in laser flow cytometry. Cytometry. 25:388-393. Glockner, F.O., Fuchs, B.M., and Amann, R. 1999. Bacterioplankton compositions of lakes and oceans: A first comparison based on fluorescence in situ hybridization. Appl. Environ. Microbiol. 65:3721-3726.
Flow Cytometry and Environmental Microbiology
Gray, N.D., Howarth, R., Pickup, R.W., Jones, J.G., and Head, I.M. 2000. Use of combined microautoradiography and fluoresence in situ hybridization to determine carbon metabolism in mixed natural communities of uncultured bacteria from the genus Achromatium. Appl. Environ. Microbiol. 66:4518-4522.
Hansen, L.H., Ferrari, B., Sørensen, A.H., Veal, D., an d Sørensen, S.J. 2001. Detection of oxytetracycline production by Streptomyces rimosus in soil microcosms by combining wholecell biosensors and flow cytometry. Appl. Environ. Microbiol. 67: 239-244. Head, I.M., Gray, N.D., Clarke, K.J., Pickup, R.W., and Jones, J.G. 1996. The phylogenetic position and ultrastructure of the uncultured bacterium Achromatium oxaliferum. Microbiology 142: 2341-2354. Henningson, E.W., Krocova, Z., Sandstrom, G., and Forsman, M. 1998. Flow cytometric assessment of the survival ratio of Francisella tularensis in aerobiological samples. FEMS Microbiol. Ecol. 25:241-249. Hodson, R.E., Dustman, W.A., Garg, R.P., and Moran, M.A. 1995. In situ PCR for visualization of microscale distribution of specific genes and gene products in prokaryotic communities. Appl. Environ. Microbiol. 61:4074-4082. Jacobsen, C.N., Rasmussen, J., and Jakobsen, M. 1997. Viability staining and flow cytometric detection of Listeria monocytogenes. J. Microbiol. Methods 28:35-43. Jernaes, M.W. and Steen, H.B. 1994. Staining of Escherichia coli for flow cytometry: Influx and efflux of ethidium bromide. Cytometry 17:302309. Jonker, R.R., Meulemans, J.T., Dubelaar, G.B.J., Wilkins, M.F., and Ringelberg, J. 1995. Flow cytometry—a powerful tool in analysis of biomass distributions in phytoplankton. Water Sci. Technol. 32:177-182. Joux, F., Lebaron, P., and Troussellier, M. 1997. Succession of cellular states in a Salmonella typhimurium population during starvation in artificial seawater microcosms. FEMS Microbiol. Ecol. 22:65-76. Kaprelyants, A.S. and Kell, D.B. 1993. Dormancy in stationary-phase cultures of Micrococcus luteus: Flow cytometric analysis of starvation and resuscitation. Appl. Environ. Microbiol. 59:3187-3196. Kaprelyants, A.S., Mukamolova, G.V., Davey, H.M., and Kell, D.B. 1996. Quantitative analysis of the physiological heterogeneity within starved cultures of Micrococcus luteus by flow cytometry and cell sorting. Appl. Environ. Microbiol. 62:1311-1316. Kenter, U., Zimmermann, U., and Muller, H. 1996. Grazing rates of the freshwater ciliate Balanion planctonicum determined by flow cytometry. J. Plankton Res. 18:1047-1053. Lange, J.L., Thorne, P.S., and Lynch, N. 1997. Application of flow cytometry and fluorescent in situ hybridization for assessment of exposures to airborne bacteria. Appl. Environ. Microbiol. 63:1557-1563. Langsrud, S. and Sundheim, G. 1996. Flow cytometry for rapid assessment of viability after exposure to a quaternary ammonium compound. J. Appl. Bacteriol. 81:411-418.
11.2.10 Supplement 27
Current Protocols in Cytometry
Lebaron, P., Catala, P., and Parthuisot, N. 1998a. Effectiveness of SYTOX green stain for bacterial viability assessment. Appl. Environ. Microbiol. 64:2697-2700 Lebaron, P., Parthuisot, N., and Catala, P. 1998b. Comparison of blue nucleic acid dyes for flow cytometric enumeration of bacteria in aquatic systems. Appl. Environ. Microbiol. 64:17251730. Lebaron, P., Servais, P., Baudoux, A.-C., Bourrain, M., Courties, C., and Parthuisot, N. 2002. Variations of bacterial-specific activity with cell size and nucleic acid content assessed by flow cytometry. Aquat. Microb. Ecol. 28:131-140. Li, W.K.W. 1995. Composition of ultraphytoplankton in the Central North Atlantic. Mar. Ecol. Prog. Ser. 122:1-8. Lloyd, D. 1993. Flow Cytometry in Microbiology. Springer-Verlag, London. Manefield, M., Whiteley, A.S., Griffiths, R.I., and Bailey, M.J. 2002. RNA stable isotope probing, a novel means of linking microbial community function to phylogeny. Appl. Environ. Microbiol. 68:5367-5373. Marie, D., Vaulot, D., and Partensky, F. 1996. Application of the novel nucleic acid dyes YOYO-1, YO-PRO-1, and Picogreen for flow cytometric analysis of marine prokaryotes. Appl. Environ. Microbiol. 62:1649-1655 Marie, D., Partensky, F., Jacquet, S., and Vaulot, D. 1997. Enumeration and cell cycle analysis of natural populations of marine picoplankton by flow cytometry using the nucleic acid stain SYBR Green I. Appl. Environ. Microbiol. 63:186-193. Marie, D., Brussaard, C.P.D., Thyrhaug, R., Bratbak, G., and Vaulot, D. 1999. Enumeration of marine viruses in culture and natural samples by flow cytometry. Appl. Environ. Microbiol. 65: 45-52. Mason, D.J., Allman, R. Stark, J.M., and Lloyd, D. 1994. Rapid estimation of bacterial antibiotic susceptibility with flow cytometry. J. Microsc. 176:8-16. Mason, D.J., Lopez-Amoros, R., Allman, R., Stark, J.M., and Lloyd, D. 1995. The ability of membrane potential dyes and calcafluor white to distinguish between viable and non-viable bacteria. J. Appl. Bacteriol. 78:309-315. Monger, B.C. and Landry, M.R. 1993. Flow cytometric analysis of marine bacteria with Hoechst 33342. Appl. Environ. Microbiol. 59:905-911. Montford, P. and Baleux, B. 1992. Comparison of flow cytometry and epifluorescence microscopy for counting bacteria in aquatic ecosystems. Cytometry 13:188-192. Nebe-von Caron, G. and Badley, R.A. 1995. Viability assessment of bacteria in mixed populations using flow cytometry. J. Microsc. 179:55-66.
Oerther, D.B. , Pernthaler, J., Schramm, A., Amann, R. and Raskin, L. 2000. Monitoring precursor 16S rRNAs of Acinetobacter spp. in activated sludge wastewater treatment systems. Appl. Environ. Microbiol. 66:2154-2165. Page, S. and Burns, R.G. 1991. Flow cytometry as a means of enumerating bacteria introduced into soil. Soil Biol. Biochem. 23:1025-1028. Payne, M.J., Campbell, S., Patchett, R.A., and Kroll, R.G. 1992. The use of immobilized lectins in the separation of Staphylococcus aureus, Escherichia coli, Listeria and Salmonella spp. from pure cultures and foods. J. Appl. Bacteriol. 73:41-52. Pernthaler, A., Pernthaler, J., Schattenhofer, M., and Amann, R. 2002. Identification of DNA-synthesizing bacterial cells in coastal North Sea plankton. Appl. Environ. Microbiol. 68:5728-5736. Porter, J., Edwards, C., Morgan, J.A.W., and Pickup, R.W. 1993. Rapid, automated separation of specific bacteria from lakewater and sewage using flow cytometry and cell sorting. Appl. Environ. Microbiol. 59:3327-3333. Porter, J., Diaper, J., Edwards, C., and Pickup, R. 1995a. Direct measurements of natural planktonic bacterial community viability by flow cytometry. Appl. Environ. Microbiol. 61:27832786. Porter, J., Edwards, C., and Pickup, R.W. 1995b. Rapid assessment of physiological status in Escherichia coli using fluorescent probes. J. Appl. Bacteriol. 79:399-408. Porter, J., Pickup, R., and Edwards, C. 1995c. Flow cytometric detection of specific genes in genetically modified bacteria using in situ polymerase chain reaction. FEMS Microbiol. Lett. 134:5156. Porter, J., Pickup, R.W., Robinson, J., and Edwards, C. 1995d. Recovery of a bacterial sub-population from sewage using immunofluorescent flow cytometry and cell sorting. FEMS Microbiol. Lett. 133:195-199. Porter, J., Deere, D., Pickup, R., and Edwards, C. 1996. Fluorescent probes and flow cytometry: New insights into environmental bacteriology. Cytometry 23:91-96. Porter, J., Deere, D., Hardman, M., Edwards, C., and Pickup, R. 1997a. Go with the flow—use of flow cytometry in environmental microbiology. FEMS Microbiol. Ecol. 24:93-101 Porter, J., Pickup, R.W., and Edwards, C. 1997b. Evaluation of flow cytometric methods for the detection and viability assessment of bacteria in soil. Soil Biol. Biochem. 29:91-100. Porter, J., Robinson, J., Pickup, R., and Edwards, C. 1998. An evaluation of lectin-mediated magnetic bead cell sorting for the targeted separation of enteric bacteria. J. Appl. Microbiol. 84:722-732. Radajewski, S., Ineson, P., Parekh, N.R., and Murrell, J.C. 2000. Stable-isotope probing as a tool in microbial ecology. Nature 403:646-649. Microbiological Applications
11.2.11 Current Protocols in Cytometry
Supplement 27
Resina-Pelfort, O., Comas-Riu, J., and Vives-Rego, J. 2001. Effects of deflected droplet electrostatic cell sorting on the viability and exoproteolytic activity of bacterial cultures and marine bacterioplankton. System. Appl. Microbiol. 24:31-36.
Turner, K., Porter, J., Pickup, R.W., and Edwards, C. 2000. Changes in viability and macromolecular content of long-term batch cultures of Salmonella typhimurium measured by flow cytometry. J. Appl. Microbiol. 89:90-99.
Rice, J., Sleigh, M.A., Burkill, P.H., Tarran, G.A., O’Connor, C.D., and Zubkov, M.V. 1997. Flow cytometric analysis of characteristics of hybridisation of species-specific fluorescent oligonucleotide probes to rRNA of marine nanoflagellates. Appl. Environ. Microbiol. 63:938-944.
Tyndall, R.L., Hand, R.E. Jr., Mann, R.C., Evans, C., and Jernigan, R. 1985. Application of flow cytometry to detection and characterization of Legionella spp. Appl. Environ. Microbiol. 49:852857.
Robertson, B.R. and Button, D.K. 1989. Characterizing aquatic bacteria according to population, cell size and apparent DNA content by flow cytometry. Cytometry 10:70-76. Roth, B.L., Poot, M., Yue, S.T., and Millard, P.J. 1997. Bacterial viability and antibiotic susceptibility testing with SYTOX green nucleic acid stain. Appl. Environ. Microbiol. 63:2421-2431. Servais, P, Agogue, H., Courties, C., Joux, F., and Lebaron, P. 2001. Are the actively respiring cells (CTC+) those responsible for bacterial production in aquatic environments? FEMS Microbiol. Ecol. 35:171-179. Simon, N., LeBot, N., Marie, D., Partensky, F., and Vaulot, D. 1995. Fluorescent in situ hybridisation with ribosomal RNA targeted oligonucleotide probes to identify small phytoplankton by flow cytometry. Appl. Environ. Microbiol. 61:2506-2513. Simon, N., Brenner, J., Edwardsen, B., and Medlin, L.K. 1997. The identification of Chysochromulina and Prymnesium (Haptophyta, Pymnesiophycae) using fluorescent or chemiluminescent oligonucleotide probes: A means for improving studies on toxic algae. Eur. J. Phycol. 32:393401. Tani, K., Kurokawa, K., and Nasu, M. 1998. Development of a direct in situ PCR method for detection of specific bacteria in natural environments. Appl. Environ. Microbiol. 64:1536-1540. Tani, K., Muneta, M., Nakamura, K., Shibuya, K., and Nasu, M. 2002. Monitoring of Ralstonia eutropha KT1 in groundwater in an experimental bioaugmentation field by in situ PCR. Appl. Environ. Microbiol. 68:412-416. Tombolini, R., Unge, A., Davey, M.E., de Bruijn, F.J., and Jansson, J.K. 1997. Flow cytometric and microscopic analysis of GFP-tagged Pseudomonas fluorescens bacteria. FEMS Microbiol. Ecol. 22:17-28. Thomas, J.C., Desrosiers, M., St.-Pierre, Y., Lirette, P., Bisaillon, J.G., Beaudet, R., and Villemur, R. 1997. Quantitative flow cytometric detection of specific microorganisms in soil samples using rRNA targeted fluorescent probes and ethidium bromide. Cytometry 27:224-232.
Flow Cytometry and Environmental Microbiology
Troussellier, M., Courties, C., and Zettelmaier, S. 1995. Flow cytometric analysis of coastal lagoon bacterioplankton and picophytoplankton: Fixation and storage effects. Estuarine Coastal Shelf Sci. 40:621-633.
Van der Waaij, L.A., Mesander, G., Limburg, P.C., and van der Waaij, D. 1994. Direct flow cytometry of anaerobic bacteria in human feces. Cytometry 16:270-279. Van der Waaij, L.A., Limberg, P.C., Mesander, G., and van der Waaij, D. 1996. In vivo IgA coating of anaerobic bacteria in human faeces. Gut 38:348-354. Veldhuis, M.J.W., Cucci, T., and Sieracki, M.E. 1997. Cellular DNA content of marine phytoplankton using two new fluorochromes: Taxonomic and ecological implications. J. Phycol. 33:527-541. Vesey, G., Hutton, P., Champion, A., Ashbolt, N., Williams, K.L., Warton, A., and Veal, D. 1994a. Application of flow cytometric methods for the routine detection of Cryptosporidium an d Giardia in water. Cytometry 16:1-6. Vesey, G., Narai, J., Ashbolt, N., Williams, K., and Veal, D. 1994b. Detection of specific microorganisms in environmental samples using flow cytometry. Methods Cell Biol. 42:489-522. Vesey, G., Deere, D., Gauci, M.R., Williams, K.L., and Veal, D.A. 1997a. Evaluation of fluorochromes and excitation sources for immunofluorescence in water samples. Cytometry 29:147154. Vesey, G., Deere, D., Weir, C.J., Ashbolt, N., Williams, K.L., and Veal, D.A. 1997b. A simple method for evaluating Cryptosporidium specific antibodies used in monitoring environmental water samples. Lett. Appl. Microbiol. 25:316320. Vesey, G., Griffiths, K.R., Gauci, M.R., Deere, D., Williams, K.L., and Veal, D.A. 1997c. Simple and rapid measurement of Cryptosporidium excystation using flow cytometry. Int. J. Parasitol. 27:1353-1359. Volsch, A., Nader, W.F., Geiss, H.K., Nebe, G., and Birr, C. 1990. Detection and analysis of two serotypes of ammonia-oxidizing bacteria in sewage plants by flow cytometry. Appl. Environ. Microbiol. 56:2430-2435. Votyakova, T.V., Kaprelyants, K.S., and Kell, D.B. 1994. Influence of viable cells on the resuscitation of dormant cells in Micrococcus luteus cultures held in an extended stationary phase: The population effect. Appl. Environ. Microbiol. 60:3284-3291.
11.2.12 Supplement 27
Current Protocols in Cytometry
Wallner, G., Amann, R., and Beisker, W. 1993. Optimizing fluorescent in situ hybridization with rRNA-targeted oligonucleotide probes for flow cytometric identification of microorganisms. Cytometry 14:136-143. Wallner, G., Erhart, R., and Amann, R. 1995. Flow cytometric analysis of activated sludge with rRNA-targeted probes. Appl. Environ. Microbiol. 61:1859-1866. Wallner, G., Fuchs, B., Spring, S., Beisker, W., and Amann, R. 1997. Flow sorting of microorganisms for molecular analysis. Appl. Environ. Microbiol. 63:4223-4231. Whiteley, A.S., Griffiths, R.I., and Bailey, M.J. 2003. Analysis of the microbial functional diversity within water-stressed soil communities by flow cytometric analysis and CTC+ cell sorting. J. Microbiol. Meth. 54:257-267. Wilkins, M.F., Boddy, L., Morris, C.W., and Jonker, R. 1996. A comparison of some neural and nonneural methods for identification of phytoplankton from flow cytometry data. Comput. Appl. Biosci. 12:9-18.
Williams, S.C., Hong, Y., Danavall, D.C.A., Howard-Jones, M.H., Gibson, D., Frischer, M.E., and Verity, P.G. 1998. Distinguishing between living and non-living bacteria: Evaluation of the vital stain propidium iodide and its combined use with molecular probes in aquatic samples. J. Microbiol. Methods 32:225-236. Yamaguchi, N. and Nasu, M. 1997. Flow cytometric analysis of bacterial respiratory and enzymatic activity in the natural aquatic environment. J. Appl. Microbiol. 83:43-52. Yentsch, C.M. and Horan, P.K. (eds.). 1989. Cytometry in Aquatic Sciences. Cytometry (special issue) 10:497-672.
Contributed by Jonathan Porter University of Exeter Exeter, United Kingdom
Dr. Porter wishes to acknowledge the support of the Natural Environment Research Council, Swindon, United Kingdom.
Microbiological Applications
11.2.13 Current Protocols in Cytometry
Supplement 27
Estimation of Microbial Viability Using Flow Cytometry One of the most basic questions that a microbiologist might ask of a microorganism is whether it is alive or not, and in microbiology, it is often necessary to determine the number of living (viable) cells in a sample or culture of interest. However, perhaps surprisingly, this is a question that is not always easily answered (Kaprelyants et al., 1993, 1999; Kell et al., 1998), even for macroorganisms (Watson, 1987). The gold standard for determining the number of viable microbial cells in a sample is usually achieved by plating a 0.1- to 1-ml sample of cells (diluted as required) onto an agar plate (Hattori, 1988; Postgate, 1969) and scoring as viable (a posteriori) those cells that were able to form visible colonies. The culture viability is then the ratio of these cells to the total cell count in the original sample, which is determined microscopically. However, there are several problems associated with this technique, not the least of which is the length of time required to obtain the results. For some slowly growing organisms (e.g., mycobacteria), it may take several weeks to determine how many cells were viable in the original sample. Even when the sample contains fast-growing organisms and the plates are incubated under optimal growth conditions, a minimum of overnight growth is usually required before the resulting colonies can be counted. In clinical situations and for economic reasons, such a delay is often unacceptable. Thus, many so-called rapid methods have been developed to allow a speedier assessment of the viable microbial load in a sample (e.g., Adams and Hope, 1989; Fung, 1994; Harris and Kell, 1985; Jones, 1987). These alternative rapid viability measurements include a variety of stain-based methods. The so-called vital stains that have been used in attempts to estimate microbial viability fall into three broad categories. (1) Some dyes, such as propidium iodide (PI), are excluded by the intact membranes of viable cells. Therefore, the presence of the dye within the cell indicates disruption of the cell membrane and may be expected to be correlated with cell death. (2) Other dyes, such as rhodamine 123, are actively accumulated by viable cells; thus,
UNIT 11.3
the number of brightly stained cells reflects the viability of the sample. However, in some cases, more active cells can actually pump such dyes out (Jernaes and Steen, 1994). Additionally, some dyes are less tightly bound by energized membranes, so that the more active cells are less brightly stained. (3) In the case of dyes such as fluorescein diacetate (FDA), a membrane-permeant nonfluorescent precursor is converted to a membrane-impermeant fluorescent molecule by the activity of intracellular enzymes, and thus is an indicator of metabolically active cells. Each of these dyebased approaches is discussed in more detail below. Although usually considered to be the goldstandard measure of viability, a plate count actually only indicates how many of the cells can replicate under the conditions provided for growth. In the case of environmental samples, the laboratory media, the temperature, and other factors may differ substantially from those in the original sample (Roszak and Colwell, 1987); thus, the proportion of cells that can divide and form colonies may be much lower than the number of cells that would score as viable using the dye-based rapid methods (Amann et al., 1995). Nevertheless, the plate count method has remained the gold standard, in part due to the fact that traditional microscopic analyses of stained cells are time consuming and can lead to operator fatigue; thus, conclusions are normally drawn from the analysis of at best a few hundred cells. Furthermore, microscopic examination is largely a qualitative technique, wherein a judgment of alive or dead is all that is possible, and the interpretation of the extent of a cell’s staining may vary among operators. Flow cytometry offers an alternative method of determining the amount of fluorescent dye taken up by each cell in a population (Davey and Kell, 1996; Kell et al., 1991; Lloyd, 1993; Shapiro, 1995). Since quantitative measurements can be made very rapidly on a large number of individual cells, an accurate picture of the distribution of dye uptake by many thousands of cells is possible within a few minutes. This unit begins with a discussion of the various advantages and disadvantages of Microbiological Applications
Contributed by Hazel M. Davey, Douglas B. Kell, Dieter H. Weichart, and Arseny S. Kaprelyants Current Protocols in Cytometry (2004) 11.3.1-11.3.21 C 2004 by John Wiley & Sons, Inc. Copyright
11.3.1 Supplement 29
classical (proliferative) versus cytochemical (dye-based) viability assays. It discusses the three classes of cytochemical methods in greater detail, and provides instructions for three simple protocols. Finally, it discusses the use of cell sorting in conjunction with tests for microbial viability.
THE PROBLEM OF DETERMINING VIABILITY In classical terms, a microbial cell is generally considered viable if it possesses all the components and mechanisms necessary for sustained proliferation (Greenwood and Peutherer, 1992). According to Postgate (1976): “At present one must accept that the death of a microbe can only be discovered retrospectively: a population is exposed to a recovery medium, incubated, and those individuals which do not divide to form progeny are taken to be dead. . . . There exist at present no shortcuts which would permit assessment of the moment of death: vital staining, optical effects, leakage of indicator substances and so on are not of general validity. . . . The term ‘viability’ applies to populations, not individuals (except in an all-or-none sense: an individual is either viable or nonviable).” Viability is evidently best determined by the classical method of assessing cellular proliferation directly, and scoring only those cells that have visibly multiplied. However, the underpinning assumption of rapid microbiology is that one can estimate something that might correlate with culturability by assessing the presence and functionality of individual vital factors and processes. The cytochemical approach gives an insight into the physiology of individual cells by providing data on parameters such as membrane energization and enzyme activity. Both approaches have advantages and disadvantages, as discussed in this section. Regrettably, studies in which the same cells are compared using each approach are rare indeed.
The Growth-Based Approach
Estimation of Microbial Viability Using Flow Cytometry
The classical approach requires an a priori knowledge of the suitable growth media and conditions for the organism or organisms present in the sample, as well as the use of a suitable method of growth detection for the organisms. In practice, because of limited time, materials, and prior knowledge, the most convenient method is all too often chosen. Highnutrient, complex media such as Luria broth and trypticase soy broth (TSB) are often used
for these procedures to ensure growth. Growth detection is usually measured either as colonyforming units (cfu) on a solid agar plate or as turbidity in liquid media. There are several problems associated with this kind of approach. Standard growth conditions. Many microorganisms have growth requirements that are very different from the standard conditions applied. For instance, standard conditions usually involve the aerobic incubation of a sample at a higher temperature than that at which it was collected (e.g., ≥30◦ C may be used even in the case of environmental samples). In fact, several medically important bacteria (such as Mycobacterium leprae) and the vast majority of bacteria in the environment have not yet been cultured axenically by any method devised to date (Amann et al., 1995). In many cases where success has eventually been obtained, organisms defied efforts to culture them until some critical component was added to the medium. A well-known example is Legionella (Meyer, 1983). Changes in physiological state. Microorganisms with known growth requirements may reside in a physiological state in which the (otherwise appropriate) standard culture conditions do not support growth, or do so only for a small fraction of the population, or only after long lag phases. Physiological states that can be difficult or impossible to detect include injury (stress), starvation (stationary phase), and dormancy (latency or cryptobiosis; Kell et al., 1998). Additionally, in some cases at least, normally copiotrophic bacteria can be recovered only after incubation in comparatively oligotrophic conditions (MacDonell and Hood, 1982; Mukamolova et al., 1998b). Growth determination method. In some cases, growth of viable cells can remain undetected due to the constraints of the growth determination method employed. Organisms displaying slow growth rates or long lag phases may not be capable of producing enough biomass to form visible colonies or detectable turbidity during the period of incubation allowed. In some cases cessation of growth may occur after a limited number of divisions (Kell et al., 1998), or the organism may be unable to form colonies on solid media. These factors, alone or in combination, may lead to falsenegative results. Growth factor requirements. It is of course entirely plausible and even likely that dormant and uncultured microorganisms actually need autocrine or paracrine growth factors for
11.3.2 Supplement 29
Current Protocols in Cytometry
their cultivation in vitro (Kaprelyants and Kell, 1996; Kaprelyants et al., 1999; Mukamolova et al., 1998a). Thus the main drawback of classical, growth-based viability assays is the possibility of false-negative results; false positives can be excluded by correct sterile technique. However, if the experimental design equates viability with culturability, only growth-based viability assays make any true sense, and thus may be considered to be both necessary and sufficient criteria for cellular viability.
The Cytochemical Approach There are occasions in which it is the metabolic activity of the cells that is of concern, whether they are capable of multiplication or not. Clearly a cell whose DNA has been damaged at the origin of replication could not multiply, but the rest of its activities would probably be unaffected. If, for example, these activities included the production of a toxin, then a method that detects metabolic activity would be of more interest than one that requires proliferation to score for cellular presence and activity. Cytochemical assays can have several advantages over proliferation-based assays. They are generally less time-consuming, in some cases delivering instantaneous results. They facilitate (at least potentially) a method of measuring something that might be correlated with other measures of viability (such as culturability) in organisms for which suitable growth conditions have not been established. For organisms that display extremely slow growth rates, long lag phases, or low growth yields, proliferation-based methods are often impossible or impractical, and thus the cytochemical approach offers an attractive alternative. In some cases (e.g., flow cytometry), the cytochemical approach allows simultaneous analysis of taxonomic traits by using specific antibodies or ribosomal RNA probes (Amann et al., 1995; Wallner et al., 1993). Thus, multidimensional snapshots of mixed populations can be generated, providing information on the species composition and physiological status of cell populations. However, these rapid assays also often have their own drawbacks that can make them difficult to interpret. So far, no viability assay has been developed that selectively and reliably detects viable cells without, under any circumstances, giving a signal with dead cells (Kaprelyants et al., 1993), where dead cells are operationally defined as cells unable to form a
colony on a plate under any condition tested (Barer et al., 1998; Kell et al., 1998). This is due to the fact that assays are normally based on single parameters such as membrane energization (often referred to as membrane potential, despite the absence of any direct evidence for it in bacteria; Kell, 1988, 1992), enzyme activity, or uptake of a substrate. Some of these criteria might be considered necessary to define viability in most cases, but none of them is sufficient to exclude nonviable cells. For example, a cell could display some enzyme activity but may have lost its ability to divide by lethal lesions in the chromosome. Thus, these assays can give rise to false-positive results, making it of the utmost importance to ascertain the reliability of any viability assay by negative control experiments, preferably involving samples of cells killed by a range of treatments (e.g., heat, ethanol, chlorine). Commercial viability assays can also produce false-negative results, even if the suppliers of assay kits include seemingly convincing data supporting their reliability (albeit under rather restricted conditions). In general, the test populations employed to demonstrate detection of viable microbes are either growing cells or cells subjected to rather short periods of stress (e.g., heat or cold). In natural environments, starvation and/or stress may be long-term, and the activity of cells may be reduced to extremely low levels (especially in the case of dormant cells), such that positive results might be below the limits of detection of the assay. Similarly, injured cells may have damaged membranes and score as nonviable in these kits, whereas repair of the damage during cultivation on a rich medium would allow subsequent growth (i.e., viability). The apparent paradox is avoided by the use of operational definitions (Barer et al., 1998; Kell et al., 1998) in which viability is not in fact an innate property of a cell, but is scored as a result of experimental measurements. Using these definitions, however, it should not be surprising if different experiments lead to different results. Taken together, the cytochemical viability assays might be less time-consuming and more convenient in many cases, but they do not necessarily provide reliable data because the principles on which they are based are not sufficient (nor sometimes even necessary) criteria for viability. Thus, before use, each method for assessing viability must be validated for each organism and for each type of sample in order to avoid false-positive or false-negative results.
Microbiological Applications
11.3.3 Current Protocols in Cytometry
Supplement 29
Table 11.3.1 Fluorescent Stains for Flow Cytometric Measurement of Viability
Excitation sourcea
Stain
Mode of action
BacLight kit (Molecular Probes) bis-(1,3-Dibutylbarbituric acid) trimethine oxonol: DiBAC4 (3)
Exclusion of PI and Argon (488 nm) staining with SYTO 9 Uptake by dead cells Argon (488 nm)
Calcofluor white Uptake by dead cells Carboxynaphthofluorescein Enzymic activity diacetate ChemChrome B/Y/V6
Proprietary information
5-Cyano-2,3Respiratory activity ditolyltetrazolium chloride (CTC) Ethidium bromide Exclusion Fluorescein diacetate (FDA) Enzymic activity
Joux et al. (1997), Langsrud and Sundheim, (1996), Swarts et al. (1998), Virta et al. (1998) Beck and Huber (1997), Ben Amor et al. (2002), Deere et al. (1995), Jepras et al. (1995), L´opez-Amor´os et al. (1997), Mason et al. (1994, 1997), Suller et al. (1997) He-Cd (325 nm) Berglund et al. (1987), Mason et al. (1995a) He-Ne (633 nm) Bergersen et al. (1995), Davey and Kell (1996) or laser diode (635 nm) Argon (488 nm) Clarke and Pinder (1998), Deere et al. (1998), Diaper and Edwards (1994b), Parthuisot et al. (2000) Argon (488 nm) Joux et al. (1997), Kaprelyants and Kell (1993b) Argon (488 nm) Argon (488 nm)
Fluorescein-di-β-Dgalactopyranoside (FDG) FUN-1 kit (Molecular Probes) Propidium iodide (PI)
Enzymic activity
Argon (488 nm)
Metabolic activity
Argon (488 nm)
Rhodamine 123
Uptake by live cells
SYTOX Green
Exclusion
TO-PRO-3
Exclusion
Exclusion
References
Aeschbacher et al. (1986) Aeschbacher et al. (1986), Berglund et al. (1987), Diaper and Edwards (1994b), Norden et al. (1995) Plovins et al. (1994)
Balajee and Marr (2002), Millard et al. (1997), Prudencio et al. (1998), Wenisch et al. (1997) Argon (488 nm) Auger et al. (1993), Berglund et al. (1987), or He-Ne (544 Deere et al. (1998), Gant et al. (1993), Niven nm) and Mulholland (1998) Argon (488 nm) Auger et al. (1993), Comas and Vives-Rego (1998), Davey et al. (1993), Diaper et al. (1992), Kaprelyants and Kell (1992), Porro et al. (1994) Argon (488 nm) Langsrud and Sundheim (1996), Roth et al. (1997) He-Ne (633 nm) Davey and Kell (1999) or laser diode (635 nm)
a He-Cd, helium/cadmium; He-Ne, helium/neon.
FLUORESCENT STAINS FOR MICROBIAL VIABILITY DETERMINATION BY FLOW CYTOMETRY
Estimation of Microbial Viability Using Flow Cytometry
Despite the problems associated with fluorescent staining protocols for viability measurements, useful information can be obtained providing one carefully selects and tests an appropriate protocol for the problem under investigation. To illustrate this, this section discusses the three general classes of molecules
whose uptake and/or fluorescence may be expected to reflect the viability of the cells (see Table 11.3.1), and presents sample procedures for the determination of microbial viability. This is followed by a case study (in the next section) of the use of flow cytometric viability testing in the gram-positive bacterium Micrococcus luteus. It must be stressed that the protocols given below are provided as a guide only. The large variety of microorganisms that one may wish
11.3.4 Supplement 29
Current Protocols in Cytometry
Figure 11.3.1 Distribution of the fluorescence of M. luteus cells that have been starved for 5 months, stained with propidium iodide, and assessed by flow cytometry. Octanol was added to the cells at a final concentration of 0.5%.
to study and the different modes of death that may befall them preclude the development of a universal flow cytometric viability test. The most useful advice that can be given to an experimenter is to try a range of stains with sensibly designed control samples (i.e., exposed to the same types of stress that will be present in the experimental samples), and to choose the method that gives the most reliable results. Under ideal conditions, complete separation of viable and dead cells will be achieved (see example in Fig. 11.3.1), and the analysis gates can easily be set to encompass live and/or dead cells. Under less favorable conditions, overlap may occur between the control samples on a single-parameter histogram. It may be possible to separate the two populations by using a dual-parameter dot plot of, for example, fluorescence versus forward light scatter. If this is not found to be the case, it may be necessary to choose a different staining protocol or to accept a “zone of uncertainty” in the results. As a general rule, the use of 0.22-µm (or smaller) filtered water is recommended for all aqueous solutions. Fluorescent reagents should generally be protected from light to avoid photobleaching, and buffers should be protected from light to inhibit microbial growth. Specific details for preparing and storing dyes are given in the examples below. The most important factor in experimental design is the preparation of suitable control samples.
Both positive and negative controls are suggested in the protocols below, but the choice of control samples will depend on the exact purposes of the experiment.
Dye Exclusion The exclusion of dye by an intact membrane is probably the most straightforward viability test to understand and perform. Fluorescent stains normally excluded by living cells are used to assess viability on the grounds that dead cells have leaky membranes that are permeable to the stains. Nucleic acid stains such as propidium iodide (PI) or ethidium bromide are indeed generally excluded by intact plasma membranes and their uptake is often used to indicate cell death (Aeschbacher et al., 1986; B¨ohmer, 1985; Green et al., 1994; Grogan and Collins, 1990; Jones, 1987; Lapinsky et al., 1991; L´opez-Amor´os et al., 1995; Schmid et al., 1992). PI is often the dye of choice for viability determinations in animal cells, whether the assay is done using flow cytometry or fluorescence microscopy (e.g., Garner et al., 1997; Maxwell and Johnson, 1997; Ronot et al., 1996). There is, however, an inherent danger in blindly transferring protocols developed for one cell type to another, particularly when one cell type is eukaryotic and the other is prokaryotic. In the case of ethidium bromide, for instance, efficient efflux pumps capable of removing the dye from Escherichia
Microbiological Applications
11.3.5 Current Protocols in Cytometry
Supplement 29
coli have been demonstrated by Jernaes and Steen (1994), and many other such pumps are known (Lewis, 1994). Thus, the applicability of excluded dyes for microbial viability determinations needs to be carefully considered for each type of organism.
Example 1: Measuring Viability by Dye Exclusion Using TO-PRO-3 TO-PRO-3 is a nucleic acid stain that can be used for viability testing by dye exclusion. This series of dyes has an advantage over many other exclusion dyes in that its fluorescence is enhanced some 1000-fold on binding to nucleic acids (Rye et al., 1992, 1993a,b). Undiluted solutions of TO-PRO-3 (as obtained from the manufacturer) should be stored frozen and protected from light. The diluted solution is stable for several weeks when stored in the same manner. This procedure calls for a flow cytometer with a 633-nm helium/neon (HeNe) laser or 635-nm laser diode as excitation source, and a detector set to receive emission above 650 nm. Staining. Place 999 µl of a cell sample or control, at 105 to 107 /ml, into a tube suitable for use with the flow cytometer. Dilute 1 mM TO-PRO-3 (Molecular Probes) to 0.1 mM in a buffer suitable for use with the sample of interest. Add 1 µl per sample (final 0.1 µM) and mix gently (e.g., with a pipet). Stain each sample and control immediately before use. Calibration. Stain a dead control (e.g., ethanol-fixed or heat-treated cells) and analyze on the flow cytometer, adjusting the PMT voltage as necessary to ensure that the sample fluorescence (at >650 nm) is to the right-hand side of the display. Position an analysis gate to encompass the dead cells. Stain the live control (freshly harvested cells, ≥95% alive) and analyze on the flow cytometer. Position an analysis gate to encompass the live cells. Adjust the gates so that optimum separation of live and dead cells is achieved. If good separation is not achieved, vary the final stain concentration to between 0.05 and 1 µM, alter the PMT voltages, and/or alter the staining time according to the nature of the cell sample. Analysis. Stain a sample and analyze using the settings determined during calibration. Record the percentages of viable and dead cells. Estimation of Microbial Viability Using Flow Cytometry
Dye Uptake It is well documented that the mitochondria of eukaryotic cells have the ability to accumulate lipophilic cations such as rhodamine 123
concentratively (Chen, 1988; Chen et al., 1982; Grogan and Collins, 1990; Johnson et al., 1980, 1981), in an uncoupler-sensitive fashion. The staining of mitochondria with rhodamine 123 has been used in conjunction with flow cytometry to study their activity (Darzynkiewicz et al., 1981; Iwagaki et al., 1990; Lizard et al., 1990). Viable bacteria also accumulate rhodamine 123, while nonviable ones do not (Diaper et al., 1992). Under certain conditions, the extent to which individual bacteria take up rhodamine 123 quantitatively reflects the extent of their viability—i.e., whether they are immediately culturable, nonculturable, or dormant (Kaprelyants and Kell, 1992). On average, larger cells may be expected to accumulate more molecules of rhodamine 123 than do smaller cells, but since flow cytometry allows collection of both fluorescence (rhodamine 123 uptake) and forward light scattering (cell size) from each cell, the data can be plotted as a dual-parameter histogram, enabling one to take size differences between cells into account when interpreting the data. For an example of results using an uptake dye, see Estimations of Viability of Micrococcus luteus. In contrast to some of the other viability stains such as acridine orange (Back and Kroll, 1991), the uptake of rhodamine 123 is useful not only because it does not require the use of fixatives to permeabilize the cell, but also because the concentrative uptake is dependent on an intact and energized cytoplasmic membrane. This has the great advantage that living cells can be stained and that further physiological studies may be conducted following staining, if required (Davey et al., 1993). There are, however, experimental problems with the use of lipophilic cations for microbial viability determinations. For instance, they may be pumped out of viable cells by microbial efflux pumps, causing both viable and nonviable cells to appear to be nonfluorescent. In addition, although the stain is readily concentrated by gram-positive bacteria such as Micrococcus luteus, the permeability of the stain in gram-negative organisms is low unless the cells are pretreated with EDTA (Kaprelyants and Kell, 1992). However, such pretreatment is practically impossible to standardize, and thus the extent of lipophilic cation accumulation may vary from experiment to experiment. In addition, in a protocol for viability determination, it is generally desirable that the number of preprocessing steps be kept to a minimum in order to avoid the possibility of affecting the viability of the sample.
11.3.6 Supplement 29
Current Protocols in Cytometry
An alternative approach is the use of lipophilic anions, which, in contrast to cations, bind preferentially to nonviable cells. The lipophilic anion bis-(1,3-dibutylbarbituric acid) trimethine oxonol, or DiBAC4 (3), has been shown to enter eukaryotic membranes only if the membranes are deenergized (Wilson and Chused, 1985). This stain has been used for the rapid assessment of microbial responses to antibiotics (Jepras et al., 1997; Mason et al., 1994, 1995b; Suller et al., 1997), allowing the analysis of heterogeneity within a microbial population in terms of susceptibility to an antibiotic.
Example 2: Measuring Viability using a Combination of Dye Exclusion and Dye Uptake Propidium iodide (PI) is a nucleic acid stain that is normally excluded by the intact membrane of viable cells. Uptake of bis(1,3-dibutylbarbituric acid)trimethine oxonol [DiBAC4 (3)] is dependent on deenergization rather than permeabilization of the cell membrane. Thus, these two dyes can be combined in a single protocol that provides a method for following de-energization as part of a “two-stage death” of microbial cells. Both dyes can be obtained in powder form. Stock solutions prepared as described below appear to be stable for many months when frozen and protected from the light.
It is possible to carry out this procedure on a flow cytometer with a single excitation source at 488 nm. Alternatively the PI can be excited with a second laser at ∼544 nm (e.g., a green He-Ne laser). In this case, the fluorescence of DiBAC4 (3) must be separated from the green scatter signal via temporally/spatially separated illumination of the stream of cells. Emissions of the DiBAC4 (3) and PI are detected at ∼525 nm and >600 nm respectively. Staining. PI stock is prepared as an aqueous solution at a concentration of 3.33 mg/ml. DiBAC4 (3) stock is prepared at a concentration of 1 mg/ml in acetone. A combined staining solution is prepared by adding 100 µl PI stock and 240 µl DiBAC4 (3) to 24.66 ml filtered water. This working solution is combined with an equal volume of cell suspension at a concentration of ∼106 cells/ml. Samples should be incubated 30 min in the dark at room temperature prior to analysis on the flow cytometer. Calibration. Set up a dual-parameter scatter plot of log red fluorescence (PI) versus log green fluorescence [DiBAC4 (3)]. If necessary, gate data plotted on this graph via forward and/or side scatter to exclude debris. Stain and analyze live and dead controls (as described in Example 1 above). Adjust PMT settings to give results similar to those shown in Figure 11.3.2. If this cannot be achieved, it may be necessary to vary the dye concentrations and/or the
Figure 11.3.2 Flow cytometric analysis (using a Partec PAS III) of Saccharomyces cerevisiae stained with PI and DiBAC4 (3) as described in Example 2. Events in the bottom-left quadrant represent live cells that have taken up neither stain. Events in the top-right quadrant represent cells that have taken up both stains and are (presumably) dead. Events in the bottom-right quadrant represent cells that have de-energized but intact membranes.
Microbiological Applications
11.3.7 Current Protocols in Cytometry
Supplement 29
staining time according to the nature of the cell sample. Analysis. Stain a sample and analyze using the settings determined during calibration. Record the percentage or absolute numbers of cells in each quadrant.
Metabolic Activity In certain circumstances the activity of the cells may be of more interest than their membrane integrity. For this purpose, a third class of viability stains is used in mammalian cell biology, often as a positive marker in a dualstaining protocol with ethidium bromide or propidium iodide (Aeschbacher et al., 1986). These stains are themselves nonfluorescent and membrane permeant, but are metabolically altered inside the cell to become fluorescent and, under ideal conditions, impermeant. One example is fluorescein diacetate (FDA), which is cleaved by intracellular esterases to produce fluorescein. Dead cells do not stain because they lack enzyme activity and/or the fluorescein diffuses freely through their damaged membranes. Flow cytometric analyses of mammalian cells with this class of dyes are well established (e.g., Aeschbacher et al., 1986; Frey, 1997). Diaper and Edwards (1994a,b) used flow cytometry to detect a variety of viable bacteria after staining with FDA and its derivatives, or with ChemChrome B (Chemunex). Importantly, none of the dyes tested was found to be universal for the detection of viable bacteria. However, ChemChrome B was found to stain the widest number of gram-positive and gram-negative species, whereas the FDA derivatives preferentially stained gram-positive bacteria. Breeuwer et al. (1995) showed that FDA and carboxyfluorescein diacetate (CFDA) penetrated yeast rapidly and that esterase activity was probably most limiting; an energy-dependent efflux of carboxyfluorescein from viable cells was also observed (Breeuwer et al., 1994; Ueckert et al., 1995). It is probable that fluorescein can be pumped out of or leak rapidly from viable microorganisms, thus giving the appearance of a lack of metabolic activity in cells that are nonetheless viable.
Example 3: Measuring Viability by Assessing Metabolic Activity with CFDA Estimation of Microbial Viability Using Flow Cytometry
Carboxyfluorescein diacetate (CFDA) is cleaved by intracellular enzymes to produce fluorescent carboxyfluorescein. In using this approach, the experimenter must be aware that
dormant cells (Kaprelyants et al., 1993) or cells with low metabolic activity will probably score as nonviable. The CFDA solution in DMSO should be stored frozen and protected from light, and is stable for several months. This procedure calls for a flow cytometer with a 488-nm argon laser or other suitable excitation source, and a detector set to collect emission at ∼525 nm. Staining. For cell samples, use 106 cells/ml suspended gram-positive bacteria in 50 mM KH2 PO4 , pH 7.4, or gram-negative bacteria in TE buffer (APPENDIX 2A). For dead controls, use ethanol-fixed or heat-treated cells; for live controls, use freshly harvested cells, ≥95% alive. Place 990 µl of cell samples and controls into tubes suitable for use with the flow cytometer. Add 10 µl of 1 mM CFDA in dimethyl sulfoxide (Molecular Probes; final 10 µM) to each and mix gently (e.g., with a pipet). Incubate for 30 min at the normal growth temperature of the microorganism. Calibration. Analyze the dead control on the flow cytometer, adjusting the PMT voltage as necessary to ensure that the sample fluorescence (at ∼525 nm) is to the left-hand side of the display. Position an analysis gate to encompass the dead cells. Analyze the live control and position an analysis gate to encompass the live cells. Note that with CFDA live cells are brightly fluorescent and dead cells are nonfluorescent or weakly fluorescent. Adjust the gates so that optimum separation of live and dead cells is achieved. If good separation is not achieved, adjust the stain concentration, alter the PMT voltages, and/or alter staining time according to the nature of the cell sample. Analysis. Analyze cell samples using the settings determined during calibration. Record the percentages of viable and dead cells.
Commercial Kits A variety of kits have been produced specifically for the measurement of viability of specific types of organisms. For example, the LIVE/DEAD BacLight Bacterial Viability Kit (Molecular Probes) gives a two-color viability assessment of both gram-negative and grampositive bacteria, where live cells are labeled green with SYTO 9 and dead cells are labeled red with PI. However, as freely admitted by the company, the kit equates the presence of intact plasma membranes with viability. Thus, “bacteria rendered nonviable by exposure to agents that do not necessarily compromise the integrity of the plasma membrane, such as formaldehyde, usually appear viable by this
11.3.8 Supplement 29
Current Protocols in Cytometry
criterion” (Haugland, 1996). Despite this limitation the kit is becoming widely used in microbiology (Braux et al., 1997; Buchmeier and Libby, 1997; Decamp et al., 1997; Duffy and Sheridan, 1998; Jacobsen et al., 1997; Joux et al., 1997; Korber et al., 1997; Langsrud and Sundheim, 1996; Rigsbee et al., 1997; Swarts et al., 1998; Taghi-Kilani et al., 1996; Terzieva et al., 1996; Virta et al., 1998; Weir et al., 1996). The growing use of such kits reflects, at least in part, their ease of use. In the case of the BacLight kit, the reagents are simultaneously added to the bacterial suspension, which is then incubated for a few minutes. The sample is then analyzed without washing, so “live” and “dead” bacteria can be distinguished and quantitated rapidly. There is a great danger that because of the name the uninitiated may use these tests blindly (despite the manufacturer’s warnings) without checking the reliability of the dyes with the organisms and conditions used in their experiments.
Example 4: Measuring Viability with the BacLight Kit A convenient approach to viability determination may be obtained using the Molecular Probes BacLight kits. A set of eleven 2-ml calibration samples ranging from 100% live to 100% dead cells should be prepared by mixing live and dead cells in different ratios. Full details of the sample preparation and reagent addition differ depending on the specific kit purchased (see http://www.probes.com/pis/ mp07007a.pdf). The protocol provided below is based on this information, in case Internet access is not available. BacLight kit reagents should be protected from light. The storage instructions provided with the kit should be followed exactly; in some cases, freezing or desiccation may be required. This procedure requires a flow cytometer with a 488-nm argon laser or other suitable excitation source, and detectors set to collect emission below 590 and above 610 nm. Staining. Follow the instructions provided with the kit purchased. Incubate 15 min at room temperature in the dark. Calibration. Analyze the calibration samples as described in the kit instructions. Produce a calibration curve of percentage live bacteria measured versus the actual percentage of live bacteria present in the calibration sample. Analysis. Analyze the experimental samples in the same way as the calibration samples and determine the “true” percentage of viable cells from the calibration curve.
Designing a Cytochemical Protocol When selecting a stain for a particular application there are several factors that need consideration. Some—such as the extinction coefficient, quantum yield, and photostability— are of general applicability to flow cytometric fluorescence measurements and are discussed in detail elsewhere (Davey and Kell, 1996; Shapiro, 1995). The wavelengths available for excitation must also be considered; a list of viability stains that are compatible with common flow cytometric light sources is shown in Table 11.3.1. One factor that is of particular relevance in the measurement of viability is the toxicity of the stain. Protocols used to assess viability should clearly not perturb viability. This becomes essential when one wishes to perform further physiological studies on the cells (for example, by exploiting flow cytometric cell sorting to isolate subpopulations with different fluorescence properties). In this case the toxicity of the stain (and indeed of any other chemicals used) must be assessed at the concentrations used in the protocol to ensure that they do not have any unwanted effects. Even where further physiological study is not required, it is generally desirable to use cells that have not been fixed to avoid any possible perturbation of what one is trying to measure. However, levels of cellular autofluorescence are generally higher for unfixed cells than for, say, ethanol-fixed cells. For the majority of microorganisms (chlorophyllcontaining organisms are the most notable exception), cellular autofluorescence tends to decrease substantially towards the red end of the optical spectrum, driving the development of red-excited fluorophores (Fabian et al., 1992; Patonay and Antoine, 1991; Shealy et al., 1995a,b). Such dyes can be exploited in flow cytometry using a 633-nm He-Ne laser or a 635-nm laser diode. Laser diodes can be used to construct smaller, cheaper, and more robust flow cytometers such as the Microcyte (see Internet Resources for further information). The Microcyte was developed by Gjelsnes and Tangen (1994) primarily for the analysis of microorganisms. Using this instrument it is possible to obtain both a total count and a viable count in absolute terms very rapidly (Davey and Kell, 1996, 1999). Flow cytometric analyses of samples of M. luteus stained with TO-PRO-3 are shown in Figure 11.3.3. When the appropriate stain and excitation source have been selected, it is important to perform a series of experiments to determine
Microbiological Applications
11.3.9 Current Protocols in Cytometry
Supplement 29
Figure 11.3.3 Flow cytometric analysis of M. luteus. Cells were grown overnight in Nutrient Broth E in a shaking water bath at 30◦ C. Samples were removed from the culture and stained with 0.1 µM TO-PRO-3 (Molecular Probes). The samples were then analyzed on a Microcyte flow cytometer. Under these conditions unstained samples did not give any detectable fluorescence. Three samples include cells freshly harvested from the culture, cells freeze-thaw treated prior to staining, and cells permeabilized by fixation in 70% ethanol. The frozen sample was kept at −20◦ C for 30 min, rapidly defrosted by plunging into a 50◦ C water bath, and then refrozen and rethawed in the same manner. For fixed cells, the fixative was removed by centrifugation prior to staining. In all cases the total cell count (based on forward light scatter; data not shown) has been normalized to 1 million cells.
the optimum concentration of the stain and the optimum length of time between addition of stain and subsequent analysis. The optimum concentration will inevitably be a compromise between a high one (for maximum signal) and a low one (for specificity). It may be necessary to measure and adjust the cell concentration to ensure that stain uptake is not limiting. In this case the use of a flow cytometer that allows determination of absolute cell numbers is an ideal approach. In some situations it may be desirable to exploit the multiparametric nature of flow cytometry to use two different viability stains that rely on measuring different cellular parameters (e.g., Yurkow and McKenzie, 1993). Alternatively, one may wish to combine the viability assay with measurements of other cellular properties. In either case, careful selection of all of the stains involved is required to ensure that there is minimal overlap in the emission spectra. Estimation of Microbial Viability Using Flow Cytometry
ESTIMATIONS OF VIABILITY OF MICROCOCCUS LUTEUS In nature and under conditions of stress, bacterial cultures display significant hetero-
geneity in terms of the percentage of viable (culturable) cells, and with respect to cellular metabolic activities (Kaprelyants et al., 1993). An important task is therefore to find reliable and rapid methods for estimating the number of cells with different characteristics in the whole bacterial population. To this end, the application of flow cytometric methods seems very promising, as it allows the properties of individual cells in a population to be distinguished. Because fluorescence and light scattering are measured on a quasi-continuous scale, it is possible to fully quantify the heterogeneity of a sample rather than making a simple classification into one of two classes (live or dead). The following experiments with Micrococcus luteus illustrate the application of flow cytometry for discrimination between cells in different physiological states. A variety of flow cytometric approaches have been investigated for the determination of viability in the gram-positive, non–spore forming bacterium Micrococcus luteus (NCIMB 13267; Kell et al., 1995). The bacteria are grown aerobically at 30◦ C in shake flasks in a lactate minimal medium containing L-lactate (e.g., Kaprelyants and Kell, 1992, for full
11.3.10 Supplement 29
Current Protocols in Cytometry
details). In order to obtain samples of low initial viability (as judged by plate counts), the cells may then be subjected to a starvation stress by allowing them to reach stationary phase before holding them at 30◦ C aerobically for up to 1 month, followed by a further incubation of up to 3 months at room temperature without agitation. As a result of these procedures, cell populations can be obtained in which <0.01% of the cells grow on agar plates (solidified Nutrient Broth E) that would normally support growth. However, the total cell count, estimated microscopically using a counting chamber, remains close to 100% of the initial (prestarvation) value (Kaprelyants and Kell, 1993a). In fact, as illustrated below, starved populations consist of different subpopulations that can be visualized flow cytometrically.
Estimation of Numbers of Active Versus Inactive Cells The proportion of active cells in a population of M. luteus can be estimated using the membrane energization–sensitive probe rhodamine 123. Figure 11.3.4A shows the typical fluorescence distribution of nonstarved M. luteus cells stained with rhodamine 123 and analyzed by flow cytometry. Regions 1, 2, and 3 were demarcated following the analysis of a freshly harvested sample of viable cells stained with rhodamine 123. In the absence of uncoupler, the cells exhibited a level of fluorescence between channels 80 and 136 (Fig. 11.3.4A); the use of carbonyl cyanide mchlorophenylhydrazone (CCCP) showed that this fluorescence was fully uncoupler sensitive (Fig. 11.3.4B). A good correlation is observed between the percentage of viable M. luteus cells in the population and the percentage of cells with CCCP-sensitive accumulation of rhodamine 123, judged flow cytometrically (Kaprelyants and Kell, 1992). Similar results were obtained by the flow cytometric study of M. luteus cells
stained with 5-cyano-2,3-ditolyltetrazolium chloride (CTC), the reduced form of which is a fluorescent formazan that allows one to monitor the respiratory activity of individual cells (Kaprelyants and Kell, 1993b; Rodriguez et al., 1992).
Estimation of Dormant and Dead Cells It has been shown that 10% to 50% of M. luteus cells in 3-month-old populations can be resuscitated to normal, colony-forming bacteria under conditions that exclude any significant regrowth of initially viable cells (Kaprelyants and Kell, 1992, 1993a, 1996; Kaprelyants et al., 1993, 1994, 1996, 1999; Mukamolova et al., 1998a,b; Votyakova et al., 1994, 1998). This indicates the persistence of a significant percentage of cells in the dormant state, a hypothesis that was confirmed using the most probable number (MPN) method by the resuscitation of cells from samples that, statistically, contained no initially viable cells (Kaprelyants et al., 1994). It was found that when the medium also contained spent growth medium from a culture in late log phase, a substantial increase (1000- to 100,000-fold) in the number of viable bacteria was observed compared with those estimated with the agar plate method (Table 11.3.2). These experiments were the first that served conclusively to exclude regrowth as a contributor to the observed resuscitation—an enormous problem that is rarely tackled satisfactorily in this context (Kell et al., 1998), and one that is also highly significant for the isolation of slowgrowing strains from natural ecosystems (Button et al., 1993; Schut et al., 1993). It was also concluded that viable cells of M. luteus can secrete a pheromone-like substance that is apparently necessary (though not on its own sufficient in all cases) for the resuscitation of starved, dormant cells of the same organism (Kaprelyants et al., 1994). This substance, resuscitation-promoting factor (RPF), is a small secreted protein with a molecular
Table 11.3.2 Resuscitation of Dormant M. luteus Cells in Liquid Medium
Culture
Time of starvation
Total count (cells/ml)
Viable count by cfu (cells/ml)a
Viable count by MPN (cells/ml)a,b
1
2 months
5.3 × 109
5 × 106
3.5 × 109
2
4.5 months
1010
1.3 × 106
9.2 × 109
3
6 months
1.2 × 1010
3.6 × 104
9.2 × 109
4
9 months
6.2 × 109
5.2 × 105
5.4 × 109
a Abbreviations: cfu, colony-forming units; MPN, most probable number. b Performed in the presence of RPF (for details see Mukamolova et al., 1998a).
Microbiological Applications
11.3.11 Current Protocols in Cytometry
Supplement 29
Estimation of Microbial Viability Using Flow Cytometry
Figure 11.3.4 Distribution of the fluorescence of M. luteus following staining with 0.3 µM rhodamine 123. Flow cytometry was performed using a Skatron Argus 100 instrument that was set up as described in the manufacturer’s manual. The PMT voltage for the fluorescence channel was 700 V, and the full scale of the abscissa represents 3.5 decades in fluorescence intensity. (A) Cells were grown in lactate minimal medium until late logarithmic phase, harvested, washed, and resuspended in lactate minimal medium lacking lactate (Kaprelyants et al., 1996) prior to 20-fold dilution and staining. (B) Cells were prepared as in (A), but 15 µM CCCP was added prior to analysis. (C) Cells were grown in lactate minimal medium and then starved for 5 months before dilution and staining.
11.3.12 Supplement 29
Current Protocols in Cytometry
weight of ∼17 to 18 kDa (Kaprelyants et al., 1999; Mukamolova et al., 1998a). As shown in Figure 11.3.4, the proportion of dormant cells can also be determined using flow cytometry. Figure 11.3.4C shows a typical distribution of the fluorescence of M. luteus cells starved for 5 months, stained with rhodamine 123, and analyzed by flow cytometry. A bimodal fluorescence distribution is evident. Region 1 (channels 0 to 80) represents cells that bind rhodamine 123 nonspecifically. For comparison, under nonstarvation conditions (Fig. 11.3.4B), 98% of fresh, late–logarithmic phase M. luteus cells stained with the same concentration of rhodamine 123 followed by treatment with a suitable concentration of the uncoupler CCCP exhibited fluorescence in this region. Although starved cells also occurred in region 2 (channels 80 to 136), their sensitivity to CCCP was very low, with only 2% to 5% of the cells in region 2 exhibiting a decrease in fluorescence after CCCP treatment. This phenomenon was not due to any inability of the uncoupler to act per se, since octanol treatment also failed to decrease the extent of staining of such cells (not shown). It has been suggested that cells accumulated in regions 1 and 2 represent bacteria in different physiological states (Kaprelyants et al., 1996). To determine whether this was indeed the case, a cell-sorting approach was used (see Use of Cell Sorting in Viability Studies, below).
Estimation of Injured Cells The presence of injured bacteria in starved populations or in populations subjected to stress (e.g., freezing or drying) is very likely. Commonly, such cells have an impaired membrane permeability barrier, which has been tested by the following flow cytometric approaches.
Membrane-impermeant probes In the case of bacteria, a damaged or leaky membrane (see Dye Exclusion, above) may not be a sufficient criterion for defining a cell as nonviable, but it can nevertheless be used as an indication of stress-induced injury. The permeability barrier of cells starved for 5 months was monitored by staining with PI (MacDonell and Hood, 1982; Mukamolova et al., 1998b). It was shown that PI does not penetrate the cytoplasmic membrane of intact M. luteus, while the administration of 0.5% (v/v) octanol to the cell suspension resulted in 100% of the cells being stained with PI (Fig. 11.3.1). Observation of different starved cultures of M. luteus
revealed that resuscitation was not successful in some cultures where the percentage of PI-positive cells was close to 100%, even in the presence of RPF. This indicates a correlation between the state of the permeability barrier and the ability of starved cells to recover, and thus may allow the use of PI staining for discrimination between dormant and dead cells in some populations.
NADH-induced respiration The ability to monitor the respiratory activity of individual cells allows the design of experiments for the quantitative determination of injured cells in a population following a stress such as freezing. It is well known that some bacteria in stressed populations become injured, as reflected for example in their elevated sensitivity to surface-active agents (Ray and Speck, 1973). This effect has been used for enumerating injured bacteria by plating them on selective media containing detergents (Ray and Speck, 1973). This approach, however, can only reflect injuries connected with damage to the outer portion of the cell envelope of gram-negative bacteria (Ignatov et al., 1981; Ray and Speck, 1973), whereas it is damage to the cytoplasmic membrane that is more important in determining the viability of bacteria after freezing (Ignatov et al., 1981). It has been shown that an increase in the permeability of the cytoplasmic membrane to NADH after freezing (which in contrast to normal cells resulted in the stimulation of endogenous respiration by NADH) was well correlated with a decrease in the viability of E. coli (Ignatov et al., 1981). The flow cytometric behavior of frozen/thawed M. luteus cells after the first 5 min of incubation in the presence of CTC revealed that ∼25% of the population gave significant fluorescence in the presence of NADH, but only 1% did so in its absence. After 17 min of incubation with CTC, the absolute percentage of cells fluorescing above channel number 20 had increased in both cases (i.e., with or without NADH), but the difference had not. Further incubation of the cells resulted in a decreased difference in the distribution pattern for the two types of sample. The kinetics of CTC reduction in the two samples are summarized in Figure 11.3.5. The reduction of CTC within the first few minutes of incubation in the presence of NADH indicates the existence of cells with an injured permeability barrier but with an intact respiratory chain (Ignatov et al., 1981, 1982).
Microbiological Applications
11.3.13 Current Protocols in Cytometry
Supplement 29
Figure 11.3.5 Flow cytometric fluorescence behavior of a frozen/thawed sample of M. luteus. Cells were diluted 10-fold in 50 mM phosphate buffer and incubated with 4 mM CTC for the times indicated, in either the presence or absence of 1 mM NADH. The ordinate represents the difference between samples in the percentage of cells whose fluorescence was in a channel number greater than 20.
These cells very rapidly reduce CTC to formazan, to a concentration comparable to that in intact cells (as judged by the channel number of the fluorescence), whereas some endogenous substrates left in the cells after freezing and thawing permit a slower reduction of CTC in the samples without NADH. Thus, at least 25% to 30% of the cells in a frozen population of M. luteus are injured, although the final viability of this sample (as judged by plating on a rich medium which permitted repair processes to take place) was 90% to 95% (Ignatov et al., 1982; Ray and Speck, 1972). Increased permeability of M. luteus cells resulting from exposure to the freeze/thaw stress can also be seen in Figure 11.3.3.
USE OF CELL SORTING IN VIABILITY STUDIES
Estimation of Microbial Viability Using Flow Cytometry
While flow cytometric analysis allows the investigator to perform a rapid and quantitative version of experiments that could otherwise be performed by fluorescent microscopy, flow cytometric cell sorting allows the process to be taken one very important step further. With flow cytometric analysis one can simply say that the distribution of dye uptake is cor-
related with a plate count of the same sample. However, providing that the staining protocol does not affect the viability of the cells (which may be determined by plate counts of stained and unstained samples), one can exploit the sorting capability of appropriate instruments to separate cells from the purported viable and nonviable fractions of the histogram. This allows determination of the culturability of exactly those cells whose cytological properties have already been determined directly. To this end, cultures whose fluorescence was of the type displayed in Figure 11.3.4C were sorted into two populations: (1) cells whose rhodamine 123 staining was sensitive or partially sensitive to CCCP (regions 2 + 3), and (2) cells whose rhodamine 123 staining was not sensitive to CCCP (region 1). After sorting, cells were plated on nutrient agar and examined in an MPN assay for viable count determinations, while the total count of sorted samples was also examined. These experiments revealed that the resuscitation of cells as judged by the MPN assay was successful for cells in regions 2 + 3, but not for cells in region 1. This constitutes direct evidence that dormant cells are concentrated in regions 2 + 3 (Kaprelyants et al., 1996).
11.3.14 Supplement 29
Current Protocols in Cytometry
Figure 11.3.6 (A) Flow cytometric cell sorting using the Autoclone module of the Coulter Epics Elite can be used to place individual cells onto ∼60 to 65 discrete locations on a standard 90-mm agar plate. (B) An agar plate with M. luteus colonies following sorting.
Some cell sorters, such as the Coulter Epics Elite, have a motorized stage (the Autoclone module) for the collection of single sorted cells. While this is primarily designed for the collection of cells into the wells of microtiter plates, it is also possible to modify the stage and collection protocol to allow microbial cells to be collected directly onto agar plates (Fig. 11.3.6). Thus, an event from a tightly defined region on a histogram can be correlated directly to the growth (or absence of growth) of a colony on an agar plate. This approach has been pioneered by Nebe-von Caron and colleagues, and has great potential for many microbiological applications (e.g., Nebe-von Caron and Anderson, 1996; Nebe-von Caron
and Badley, 1996; http://www.cyto.purdue. edu/flowcyt/research/micrflow/gerhard/ genmic10.htm). Although the sorting approach offers many advantages, it is important to also be aware of its potential pitfalls. Microbial cells often grow in clumps or form aggregates during sample preparation. If one cell in a clump has a leaky membrane, it will take up an exclusion dye such as propidium iodide, and the clump will thus be fluorescent and score as dead in the viability assay. However, if this clump also contains one or more live cells, a colony will result when the clump is sorted onto an agar plate. This problem can be overcome to some extent by using the forward scatter signal an
Microbiological Applications
11.3.15 Current Protocols in Cytometry
Supplement 29
Table 11.3.3 Advantages and Disadvantages of Viability Determination Methods
Multiplication assay
Advantages
Disadvantages
Provides sufficient proof that cell was alive
Slow
Generally straightforward to interpret
Requires prior knowledge of growth requirements Underestimates viable cell numbers
Cytological assay
Rapid
Can be difficult to interpret
Can be used without knowledge of growth requirements
Viability is not measured directly
Total count can be determined simultaneously
False positives and false negatives may occur
indicator of size. However, since the size of microbial cells, even within a single species, can vary greatly with growth conditions, a more robust approach may be to use two or more viability stains to give a broader picture of the physiological status of the cells. An additional problem may arise with damaged cells. The process of flow cytometric analysis followed by sorting onto an agar plate may in its own right be considered an additional stress that may convert an injured cell into a nonviable one. Such stresses can be quantified to some extent by plating injured cells before and after sample preparation. The effect, if any, of the sheath fluid on injured cells should also be determined.
CONCLUSIONS
Estimation of Microbial Viability Using Flow Cytometry
The rapid cytological estimation of true microbial viability is extremely difficult (if not impossible in principle), not least because of the problems in defining viability in microbial cells. Despite the difficulties mentioned above, the view to which the authors of this unit subscribe (Kell et al., 1988) is that only culturability can provide a gold standard for positive viability. Although the flow cytometric approach has much to offer for the determination of microbial viability, it must be emphasized that no single stain nor even cocktail is likely to be a universal indicator of viability, especially if we require that its interpretation reflect our ability to induce the cells to divide (see Table 11.3.3). A cell that is killed by exposure to environmental extremes such as heat, pH, and so forth is likely to be very different from a cell that is killed by exposure to an antibiotic or other chemical, and different again from a cell that dies (loses culturability) owing to a lack
of nutrients in its environment. Indeed, there is increasing evidence that a reversible loss of culturability may be both widespread and of adaptive significance (Mukamolova et al., 2003). Thus, the flow cytometric properties of a cell and the distribution of dye uptake within a population will depend on how the cells die, and more generally on their entire physiological state and its history. The exploitation of the sorting capability of flow cytometers permits the design of experiments that carefully evaluate the applicability of so-called viability stains, and the authors strongly recommend that others adopt this approach. In many situations, it is desirable to know both the percentage viability and the viable count (number of viable cells in a unit volume). For these purposes, instruments that allow determination of absolute cell concentration are of particular value. In conclusion, although there are as yet no perfect stains, careful protocol development currently allows valuable information to be obtained regarding specific problems. In the case of organisms that have not been exposed to excessive stress (e.g., laboratory cultures under normal conditions and in some cases clinical samples), substantial progress is being made towards the rapid and routine flow cytometric assessment of microbial viability or vitality.
Literature Cited Adams, M.R. and Hope, C.F.A. (eds.). 1989. Rapid Methods in Food Microbiology. Elsevier/NorthHolland, Amsterdam. Aeschbacher, M., Reinhardt, C.A., and Zbinden, G. 1986. A rapid cell membrane permeability test using fluorescent dyes and flow cytometry. Cell Biol. Toxicol. 2:247-255.
11.3.16 Supplement 29
Current Protocols in Cytometry
Amann, R.I., Ludwig, W., and Schleifer, K.H. 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59:143-169.
Buchmeier, N.A. and Libby, S.J. 1997. Dynamics of growth and death within a Salmonella typhimurium population during infection of macrophages. Can. J. Microbiol. 43:29-34.
Auger, J., Leonce, S., Jouannet, P., and Ronot, X. 1993. Flow cytometric sorting of living, highly motile human spermatozoa based on evaluation of their mitochondrial activity. J. Histochem. Cytochem. 41:1247-1251.
Button, D.K., Schut, F., Quang, P., Martin, R., and Robertson, B.R. 1993. Viability and isolation of marine bacteria by dilution culture—theory, procedures, and initial results. Appl. Environ. Microbiol. 59:881-891.
Back, J.P. and Kroll, R.G. 1991. The differential fluorescence of bacteria stained with acridine orange and the effects of heat. J. Appl. Bacteriol. 71:51-58.
Chen, L.B. 1988. Mitochondrial membrane potential in living cells. Ann. Rev. Cell Biol. 4:155181.
Balajee, S.A. and Marr, K.A. 2002. Conidial viability assay for rapid susceptibility testing of Aspergillus species. J. Clin. Microbiol. 40:27412745. Barer, M.R., Kaprelyants, A.S., Weichart, D.H., Harwood, C.R., and Kell, D.B. 1998. Microbial stress and culturability: Conceptual and operational domains. Microbiology. 144:20092010. Beck, P. and Huber, R. 1997. Detection of cell viability in cultures of hyperthermophiles. FEMS Microbiol. Lett. 148:11-14. Ben Amor, K., Breeuwer, P., Verbaarschot, P., Rombouts, F.M., Akkermans, A.D.L., De Vos, W.M., and Abee, T. 2002. Multiparametric flow cytometry and cell sorting for the assessment of viable, injured, and dead bifidobacterium cells during bile salt stress. Appl. Environ. Microbiol. 68:5209-5216. Bergersen, O., Ronning, O., Helleman, A.L., Vekterud, K., and Gjelsnes, O. 1995. Isolation and labelling of micro-organisms for use in flow cytometry. 1995. NORDFOOD conference, Turku (Åbo), Finland. Berglund, D.L., Taffs, R.E., and Robertson, N.P. 1987. A rapid analytical technique for flow cytometric analysis of cell viability using calcofluor white M2R. Cytometry. 8:421-426. B¨ohmer, R.M. 1985. Flow cytometric detection of a two-step cell death induced by hyperthermia. Cytometry. 6:215-218. Braux, A.S., Minet, J., TamanaiShacoori, Z., Riou, G., and Cormier, M. 1997. Direct enumeration of injured Escherichia coli cells harvested onto membrane filters. J. Microbiol. Methods. 31:1-8. Breeuwer, P., Drocourt, J.L., Rombouts, F.M., and Abee, T. 1994. Energy-dependent, carriermediated extrusion of carboxyfluorescein from Saccharomyces cerevisiae allows rapid assessment of cell viability by flow cytometry. Appl. Environ. Microbiol. 60:1467-1472. Breeuwer, P., Drocourt, J.L., Bunschoten, N., Zwietering, M.H., Rombouts, F.M., and Abee, T. 1995. Characterization of uptake and hydrolysis of fluorescein diacetate and carboxyfluorescein diacetate by intracellular esterases in Saccharomyces cerevisiae, which result in accumulation of fluorescent product. Appl. Environ. Microbiol. 61:1614-1619.
Chen, L.B., Summerhayes, I.C., Johnson, L.V., Walsh, M.L., Bernal, S.D., and Lampidis, T.J. 1982. Probing mitochondria in living cells with rhodamine 123. Cold Spring Harbor Symp. Quant. Biol. 46:141-155. Clarke, R.G. and Pinder, A.C. 1998. Improved detection of bacteria by flow cytometry using a combination of antibody and viability markers. J. Appl. Microbiol. 84:577-584. Comas, J. and Vives-Rego, J. 1998. Enumeration, viability and heterogeneity in Staphylococcus aureus cultures by flow cytometry. J. Microbiol. Methods. 32:45-53. Darzynkiewicz, Z., Staiano-Coico, L., and Melamed, M.R. 1981. Increased mitochondrial uptake of rhodamine 123 during lymphocyte stimulation. Proc. Natl. Acad. Sci. U.S.A. 78:2383-2387. Davey, H.M. and Kell, D.B. 1996. Flow cytometry and cell sorting of heterogeneous microbial populations—the importance of single-cell analyses. Microbiol. Rev. 60:641-696. Davey, H.M. and Kell, D.B. 1999. A portable flow cytometer for the detection and identification of microorganisms. In NATO Advanced Research Workshop on Rapid Methods for Monitoring the Environment for Biological Hazards. (P.J. Stopa, ed.). In press. Plenum, Warsaw, Poland. Davey, H.M., Kaprelyants, A.S., and Kell, D.B. 1993. Flow cytometric analysis, using rhodamine 123, of Micrococcus luteus at low growth rate in chemostat culture. In Flow Cytometry in Microbiology. (D. Lloyd, ed.) pp. 8393. Springer-Verlag, London. Decamp, O., Rajendran, N., Nakano, H., and Nair, G.B. 1997. Estimation of the viability of Vibrio cholerae 0139 by assessing cell membrane integrity. Microbios. 92:83-89. Deere, D., Porter, J., Edwards, C., and Pickup, R. 1995. Evaluation of the suitability of bis(1,3-dibutylbarbituric acid) trimethine oxonol, (diBA-C4 (3)− ), for the flow cytometric assessment of bacterial viability. FEMS Microbiol. Lett. 130:165-169. Deere, D., Shen, J., Vesey, G., Bell, P., Bissinger, P., and Veal, D. 1998. Flow cytometry and cell sorting for yeast viability assessment and cell selection. Yeast. 14:147-160. Diaper, J.P. and Edwards, C. 1994a. Flow cytometric detection of viable bacteria in compost. FEMS Microbiol. Ecol. 14:213-220.
Microbiological Applications
11.3.17 Current Protocols in Cytometry
Supplement 29
Diaper, J.P. and Edwards, C. 1994b. The use of fluorogenic esters to detect viable bacteria by flow cytometry. J. Appl. Bacteriol. 77:221-228. Diaper, J.P., Tither, K., and Edwards, C. 1992. Rapid assessment of bacterial viability by flow cytometry. Appl. Microbiol. Biotechnol. 38:268272. Duffy, G. and Sheridan, J.J. 1998. Viability staining in a direct count rapid method for the determination of total viable counts on processed meats. J. Microbiol. Methods. 31:167-174. Fabian, J., Nakazumi, H., and Matsuoka, M. 1992. Near-infrared absorbing dyes. Chem. Rev. 92:1197-1226. Frey, T. 1997. Correlated flow cytometric analysis of terminal events in apoptosis reveals the absence of some changes in some model systems. Cytometry. 28:253-263.
Jacobsen, C.N., Rasmussen, J., and Jakobsen, M. 1997. Viability staining and flow cytometric detection of Listeria monocytogenes. J. Microbiol. Methods. 28:35-43. Jepras, R.I., Carter, J., Pearson, S.C., Paul, F.E., and Wilkinson, M.J. 1995. Development of a robust flow cytometric assay for determining numbers of viable bacteria. Appl. Environ. Microbiol. 61:2696-2701. Jepras, R.I., Paul, F.E., Pearson, S.C., and Wilkinson, M.J. 1997. Rapid assessment of antibiotic effects on Escherichia coli by bis- (1,3dibutylbarbituric acid) trimethine oxonol and flow cytometry. Antimicrob. Agents Chemother. 41:2001-2005.
Fung, D.Y.C. 1994. Rapid methods and automation in food microbiology—a review. Food Rev. Int. 10:357-375.
Jernaes, M.W. and Steen, H.B. 1994. Staining of Escherichia coli for flow cytometry: Influx and efflux of ethidium bromide. Cytometry. 17:302309.
Gant, V.A., Warnes, G., Phillips, I., and Savidge, G.F. 1993. The application of flow cytometry to the study of bacterial responses to antibiotics. J. Med. Microbiol. 39:147-154.
Johnson, L.V., Walsh, M.L., and Chen, L.B. 1980. Localization of mitochondria in living cells with rhodamine 123. Proc. Natl. Acad. Sci. U.S.A. 77:990-994.
Garner, D., Thomas, C., and Allen, C. 1997. Effect of semen dilution on bovine sperm viability as determined by dual-DNA staining and flow cytometry. J. Androl. 18:324-331.
Johnson, L.V., Walsh, M.L., Bockus, B.J., and Chen, L.B. 1981. Monitoring of relative mitochondrial membrane potential in living cells by fluorescence microscopy. J. Cell Biol. 88:526535.
Gjelsnes, O. and Tangen, R. 1994. Liquid flow cytometer, Norway/patent W094/29695. Green, L., Peterson, B., Steimel, L., Haeber, P., and Current, W. 1994. Rapid determination of antifungal activity by flow cytometry. J. Clin. Microbiol. 32:1088-1091. Greenwood, D.R.S. and Peutherer, J. (eds.). 1992. Medical Microbiology. Churchill Livingston, London. Grogan, W.M. and Collins, J.M. 1990. Guide to Flow Cytometry Methods. Marcel Dekker, New York. Harris, C.M. and Kell, D.B. 1985. The estimation of microbial biomass. Biosensors. 1:17-84. Hattori, T. 1988. The Viable Count: Quantitative and Environmental Aspects. Springer-Verlag, Berlin. Haugland, R.P. 1996. Handbook of Fluorescent Probes and Research Chemicals, 6th ed. Molecular Probes, Inc., Eugene, Oreg. Ignatov, S.G., Krasilnikov, V.A., Pereligin, V.V., Kaprelyants, A.S., and Ostrovsky, D.N. 1981. Study of structural-functional changes in membranes of E. coli after low-temperature freezing. Biokhimiya. 46:1996-2003.
Estimation of Microbial Viability Using Flow Cytometry
tion in Molt 16 cells. Lymphokine Res. 9:365369.
Ignatov, S.G., Andreeva, O.V., Evdokimova, O.A., Kaprelyants, A.S., and Ostrovsky, D.N. 1982. Study of the repair of membrane injury after low-temperature freezing of E. coli. Biokhimiya. 47:1621-1628. Iwagaki, H., Fuchimoto, S., Miyake, M., and Oirta, K. 1990. Increased mitochondrial uptake of rhodamine 123 during interferon-gamma stimula-
Jones, R.P. 1987. Measures of yeast death and deactivation and their meaning: Parts 1 and 2. Process Biochem. 22:118-128. Joux, F., Lebaron, P., and Troussellier, M. 1997. Succession of cellular states in a Salmonella typhimurium population during starvation in artificial seawater microcosms. FEMS Microbiol. Ecol. 22:65-76. Kaprelyants, A.S. and Kell, D.B. 1992. Rapid assessment of bacterial viability and vitality using rhodamine 123 and flow cytometry. J. Appl. Bacteriol. 72:410-422. Kaprelyants, A.S. and Kell, D.B. 1993a. Dormancy in stationary-phase cultures of Micrococcus luteus: Flow cytometric analysis of starvation and resuscitation. Appl. Environ. Microbiol. 59:3187-3196. Kaprelyants, A.S. and Kell, D.B. 1993b. The use of 5-cyano-2,3-ditolyl tetrazolium chloride and flow cytometry for the visualisation of respiratory activity in individual cells of Micrococcus luteus. J. Microbiol. Methods. 17:115-122. Kaprelyants, A.S. and Kell, D.B. 1996. Do bacteria need to communicate with each other for growth? Trends Microbiol. 4:237-242. Kaprelyants, A.S., Gottschal, J.C., and Kell, D.B. 1993. Dormancy in non-sporulating bacteria. FEMS Microbiol. Rev. 104:271-286. Kaprelyants, A.S., Mukamolova, G.V., and Kell, D.B. 1994. Estimation of dormant Micrococcus luteus cells by penicillin lysis and by resuscitation in cell-free spent medium at high dilution. FEMS Microbiol. Lett. 115:347-352.
11.3.18 Supplement 29
Current Protocols in Cytometry
Kaprelyants, A.S., Mukamolova, G.V., Davey, H.M., and Kell, D.B. 1996. Quantitative analysis of the physiological heterogeneity within starved cultures of Micrococcus luteus using flow cytometry and cell sorting. Appl. Environ. Microbiol. 62:1311-1316. Kaprelyants, A.S., Mukamolova, G.V., Kormer, S.S., Weichart, D.H., Young, M., and Kell, D.B. 1999. Intercellular signalling and the multiplication of prokaryotes: Bacterial cytokines. Symp. Soc. Gen. Microbiol. In press. Kell, D.B. 1988. Protonmotive energy-transducing systems: Some physical principles and experimental approaches. In Bacterial Energy Transduction. (C.J. Anthony, ed.). Academic Press, London. Kell, D.B. 1992. The protonmotive force as an intermediate in electron transport-linked phosphorylation: Problems and prospects. Curr. Top. Cell. Regul. 33:279-289. Kell, D.B., Ryder, H.M., Kaprelyants, A.S., and Westerhoff, H.V. 1991. Quantifying heterogeneity: Flow cytometry of bacterial cultures. Antonie van Leeuwenhoek. 60:145-158. Kell, D.B., Davey, H.M., Mukamolova, G.V., Votyakova, T.V., and Kaprelyants, A.S. 1995. A summary of recent work on dormancy in non-sporulating bacteria: Its significance for marine microbiology and biotechnology. J. Marine Biotechnol. 3:24-25. Kell, D.B., Kaprelyants, A.S., Weichart, D.H., Harwood, C.L., and Barer, M.R. 1998. Viability and activity in readily culturable bacteria: A review and discussion of the practical issues. Antonie van Leeuwenhoek. 73:169-187. Korber, D.R., Choi, A., Wolfaardt, G.M., Ingham, S.C., and Caldwell, D.E. 1997. Substratum topography influences susceptibility of Salmonella enteritidis biofilms to trisodium phosphate. Appl. Environ. Microbiol. 63:33523358. Langsrud, S. and Sundheim, G. 1996. Flow cytometry for rapid assessment of viability after exposure to a quaternary ammonium compound. J. Appl. Bacteriol. 81:411-418. Lapinsky, S.E., Glencross, D., Car, N.G., Kallenbach, J.M., and Zwi, S. 1991. Quantification and assessment of viability of Pneumonocystis carinii organisms by flow cytometry. J. Clin. Microbiol. 29:911-915. Lewis, K. 1994. Multidrug resistance pumps in bacteria: Variations on a theme. Trends Biochem. Sci. 19:119-123. Lizard, G., Chardonnet, Y., Chignol, M.C., and Thivolet, J. 1990. Evaluation of mitochondrial content and activity with nonyl-acridine orange and rhodamine 123: Flow cytometric analysis and comparison with quantitative morphometry. Cytotechnology. 3:179-188. Lloyd, D. 1993. Flow cytometry in microbiology. Springer-Verlag, London. L´opez-Amor´os, R., Comas, J., and Vives-Rego, J. 1995. Flow cytometric assessment of Escherichia coli and Salmonella typhimurium
starvation-survival in seawater using rhodamine 123, propidium iodide, and oxonol. Appl. Environ. Microbiol. 61:2521-2526. L´opez-Amor´os, R., Castel, S., Comas, Riu J., and Vives-Rego, J. 1997. Assessment of E. coli and Salmonella viability and starvation by confocal laser microscopy and flow cytometry using rhodamine 123, DiBAC4 (3), propidium iodide, and CTC. Cytometry. 29:298-305. MacDonell, M. and Hood, M. 1982. Isolation and characterization of ultramicrobacteria from a gulf coast estuary. Appl. Environ. Microbiol. 43:566-571. Mason, D.J., Allman, R., Stark, J.M., and Lloyd, D. 1994. Rapid estimation of bacterial antibiotic susceptibility with flow cytometry. J. Microsc. 176:8-16. Mason, D.J., L´opez-Amor´os, R., Allman, R., Stark, J.M., and Lloyd, D. 1995a. The ability of membrane potential dyes and calcofluor white to distinguish between viable and non-viable bacteria. J. Appl. Bacteriol. 78:309-315. Mason, D.J., Power, E.G.M., Talsania, H., Phillips, I., and Gant, V.A. 1995b. Antibacterial action of ciprofloxacin. Antimicrob. Agents Chemother. 39:2752-2758. Mason, D.J., Dybowski, R., Larrick, J.W., and Gant, V.A. 1997. Antimicrobial action of rabbit leukocyte CAP18(106-137). Antimicrob. Agents Chemother. 41:624-629. Maxwell, W. and Johnson, L. 1997. Chlortetracycline analysis of boar spermatozoa after incubation, flow cytometric sorting, cooling, or cryopreservation. Mol. Reprod. Dev. 46:408-418. Meyer, R.D. 1983. Legionella infections—a review of 5 years of research. Rev. Infect. Dis. 5:258278. Millard, P., Roth, B., Thi, H., Yue, S., and Haugland, R. 1997. Development of the FUN-1 family of fluorescent probes for vacuole labeling and viability testing of yeasts. Appl. Environ. Microbiol. 63:2897-2905. Mukamolova, G.V., Kaprelyants, A.S., Kell, D.B., and Young, M. 2003. Adoption of the transiently non-culturable state: A bacterial survival strategy? Adv. Micr. Physiol. 47:65-129. Mukamolova, G.V., Kaprelyants, A.S., Young, D.I., Young, M., and Kell, D.B. 1998a. A bacterial cytokine. Proc. Natl. Acad. Sci. U.S.A. 95:89168921. Mukamolova, G.V., Yanopolskaya, N.D., Kell, D.B., and Kaprelyants, A.S. 1998b. . On resuscitation from the dormant state of Micrococcus luteus. Antonie van Leeuwenhoek. 73:237-243. Nebe-von Caron, G. and Anderson, W.A. 1996. Germination of spores characterized by fluorescent probes and flow cytometry. Cytometry. 23:115. Nebe-von Caron, G. and Badley, R.A. 1996. Bacterial characterization by flow cytometry. In Flow Cytometry Applications in Cell Culture (M. AlRubeai and A.N. Emery, eds.) pp. 257-290. Marcel Dekker, New York.
Microbiological Applications
11.3.19 Current Protocols in Cytometry
Supplement 29
Niven, G.W. and Mulholland, F. 1998. Cell membrane integrity and lysis in Lactococcus lactis: The detection of a population of permeable cells in post-logarithmic phase cultures. J. Appl. Microbiol. 84:90-96.
Roth, B., Poot, M., Yue, S., and Millard, P. 1997. Bacterial viability and antibiotic susceptibility testing with SYTOX green nucleic acid stain. Appl. Environ. Microbiol. 63:24212431.
Norden, M.A., Kurzynski, T.A., Bownds, S.E., Callister, S.M., and Schell, R.F. 1995. Rapid susceptibility testing of Mycobacterium tuberculosis (H37RA) by flow cytometry. J. Clin. Microbiol. 33:1231-1237.
Rye, H.S., Yue, S., Wemmer, D.E., Queseda, M.A., Haugland, R.P., Mathies, R.A., and Glazer, A.N. 1992. Stable fluorescent complexes of doublestranded DNA with bis-intercalating asymmetric cyanine dyes: Properties and applications. Nucl. Acids Res. 20:2803-2812.
Parthuisot, N., Catala, P., Lemarchand, K., Baudart, J., and Lebaron, P. 2000. Evaluation of ChemChrome V6 for bacterial viability assessment in waters. J. Appl. Microbiol. 89:370-380. Patonay, G. and Antoine, M.D. 1991. Near-infrared fluorogenic labels—new approach to an old problem. Anal. Chem. 63:A321-A326. Plovins, A., Alvarez, A.M., Ibanez, M., Molina, M., and Nombela, C. 1994. Use of fluorescein-diβ-D-galactopyranoside (FDG) and C12 -FDG as substrates for β-galactosidase detection by flow cytometry in animal, bacterial, and yeast cells. Appl. Environ. Microbiol. 60:4638-4641. Porro, D., Smeraldi, C., Martegani, E., Ranzi, B.M., and Alberghina, L. 1994. Flow cytometric determination of the respiratory activity in growing Saccharomyces cerevisiae populations. Biotechnol. Prog. 10:193-197. Postgate, J.R. 1969. Viable counts and viability. Methods Microbiol. 1:611-628. Postgate, J.R. 1976. Death in microbes and macrobes. In The Survival of Vegetative Microbes. (T.R.G. Gray andJ.R. Postgate, eds.) pp. 1-19. Cambridge University Press, Cambridge. Prudencio, C., Sansonetty, F., and CorteReal, M. 1998. Flow cytometric assessment of cell structural and functional changes induced by acetic acid in the yeasts Zygosaccharomyces bailii and Saccharomyces cerevisiae. Cytometry. 31:307313. Ray, B. and Speck, M.L. 1972. Repair of injury induced by freezing E. coli as influenced by the recovery medium. Appl. Microbiol. 24:258263. Ray, B. and Speck, M.L. 1973. Freeze-injury in bacteria. CRC Crit. Rev. Clin. Lab. Sci. 4:161213. Rigsbee, W., Simpson, L.M., and Oliver, J.D. 1997. Detection of the viable but nonculturable state in Escherichia coli O157:H7. J. Food Saf. 16:255262. Rodriguez, G.G., Phipps, D., Ishiguro, K., and Ridgway, H.F. 1992. Use of a fluorescent redox probe for direct visualization of actively respiring bacteria. Appl. Environ. Microbiol. 58:1801-1808. Ronot, X., Paillasson, S., and Muirhead, K.A. 1996. Assessment of cell viability in mammalian cell lines. In Flow Cytometry Applications in Cell Culture. (M. Al-Rubeai andA.N. Emery, eds.) pp. 177-192. Marcel Dekker, New York. Estimation of Microbial Viability Using Flow Cytometry
Roszak, D.B. and Colwell, R.R. 1987. Survival strategies of bacteria in the natural environment. Microbiol. Rev. 51:365-379.
Rye, H.S., Dabora, J.M., Queseda, M.A., Mathies, R.A., and Glazer, A.N. 1993a. Fluorometric assay using dimeric dyes for double- and singlestranded DNA and RNA with picogram sensitivity. Anal. Biochem. 208:144-150. Rye, H.S., Yue, S., Quesada, M.A., Haugland, R.P., Mathies, R.A., and Glazer, A.N. 1993b. Picogram detection of stable dye-DNA intercalation complexes with two-color laser-excited confocal fluorescence gel scanner. Methods Enzymol. 217:414-431. ¨ Schmid, I., Krall, W.J., Uittenbogaart, C.H., Braun, J., and Giorgi, J.V. 1992. Dead cell discrimination with 7-amino-actinomycin D in combination with dual color immunofluorescence in single laser flow cytometry. Cytometry. 13:204208. Schut, F., Devries, E.J., Gottschal, J.C., Robertson, B.R., Harder, W., Prins, R.A., and Button, D.K. 1993. Isolation of typical marine bacteria by dilution culture—growth, maintenance, and characteristics of isolates under laboratory conditions. Appl. Environ. Microbiol. 59:21502160. Shapiro, H.M. 1995. Practical Flow Cytometry, 3rd ed. Alan R. Liss, New York. Shealy, D.B., Lipowska, M., Lipowski, J., Narayanan, N., Sutter, S., Strekowski, L., and Patonay, G. 1995a. Synthesis, chromatographic separation, and characterization of nearinfrared-labeled DNA oligomers for use in DNA sequencing. Anal. Chem. 67:247-251. Shealy, D.B., Lohrmann, R., Ruth, J.R., Narayanan, N., Sutter, S.L., Casay, G.A., Evans, L., and Patonay, G. 1995b. Spectral characterization and evaluation of modified near-infrared laser dyes for DNA sequencing. Appl. Spectrosc. 49:18151820. Suller, M.T.E., Stark, J.M., and Lloyd, D. 1997. A flow cytometric study of antibiotic-induced damage and evaluation as a rapid antibiotic susceptibility test for methicillin-resistant Staphylococcus aureus. J. Antimicrob. Chemother. 40:7783. Swarts, A.J., Hastings, J.W., Roberts, R.F., and vonHoly, A. 1998. Flow cytometry demonstrates bacteriocin-induced injury to Listeria monocytogenes. Curr. Microbiol. 36:266-270. Taghi-Kilani, R., Gyurek, L.L., Millard, P.J., Finch, G.R., and Belosevic, M. 1996. Nucleic-acid stains as indicators of Giardia muris viability following cyst inactivation. Int. J. Parasitol. 26:637-646.
11.3.20 Supplement 29
Current Protocols in Cytometry
Terzieva, S., Donnelly, J., Ulevicius, V., Grinshpun, S.A., Willeke, K., Stelma, G.N., and Brenner, K.P. 1996. Comparison of methods for detection and enumeration of airborne microorganisms collected by liquid impingement. Appl. Environ. Microbiol. 62:2264-2272. Ueckert, J., Breeuwer, P., Abee, T., Stephens, P., von Caron, G.N., and ter Steeg, P.F. 1995. Flow cytometry applications in physiological study and detection of foodborne microorganisms. Int. J. Food Microbiol. 28:317-326. Virta, M., Lineri, S., Kankaapaa, P., Karp, M., Peltonen, K., Nuutila, J., and Lilius, E.M. 1998. Determination of complement-mediated killing of bacteria by viability staining and bioluminescence. Appl. Environ. Microbiol. 64:515-519. Votyakova, T.V., Kaprelyants, A.S., and Kell, D.B. 1994. Influence of viable cells on the resuscitation of dormant cells in Micrococcus luteus cultures held in extended stationary phase: The population effect. Appl. Environ. Microbiol. 60:3284-3291. Votyakova, T.V., Mukamolova, G.V., Shtein Margolina, V.A., Popov, V.I., Davey, H.M., Kell, D.B., and Kaprelyants, A.S. 1998. Research on the heterogeneity of a Micrococcus luteus culture during an extended stationary phase: Subpopulation separation and characterization. Microbiology. 67:71-77. Wallner, H., Amann, R., and Beisker, W. 1993. Optimizing fluorescent in situ hubridization with rRNA-targeted oligonucleotide probes for flow cytometric identification of microorganisms. Cytometry. 14:136-143.
Davey and Kell, 1996. See above. Extensive review of the application of flow cytometry to problems of (mainly) microbiological interest. Includes 1073 literature references. Porter, J., Deere, D., Hardman, M., Edwards, C., and Pickup, R. 1997. Go with the flow—use of flow cytometry in environmental microbiology. FEMS Microbiol. Ecol. 24:93-101. A useful review of the applications of flow cytometry in environmental microbiology. Shapiro, 1995. See above. The Book. A detailed and entertaining overview. All you ever wanted to know about flow cytometry but were afraid to ask. Troussellier, M., Courties, C., and Vaquer, A. 1993. Recent applications of flow cytometry in aquatic microbial ecology. Biol. Cell 78:111-121. Another short but useful review of applications of flow cytometry in environmental microbiology, concentrating on aquatic systems.
Internet Resources http://pcfcij.dbs.aber.ac.uk/ The Aberystwyth flow cytometry site has information on microbial flow cytometry, including viability determinations. http://www.cyto.purdue.edu/flowcyt/research/ micrflow/index.htm Microbial flow cytometry section with contributions from several authors on viability measurements. http://www.probes.com/handbook/sections/ 1503.html
Watson, L. 1987. The Biology of Death (previously published as The Romeo Error). Sceptre Books, London.
Source for many viability stains, including viability kits.
Weir, S.C., Lee, H., and Trevors, J.T. 1996. Effect of selected disinfectants on the persistence and movement of a genetically-engineered Pseudomonas sp. in soil. Syst. Appl. Microbiol. 19:421-427.
Includes useful application notes on microbial cytometry.
Wenisch, C., Linnau, K., Parschalk, B., ZedtwitzLiebenstein, K., and Georgopoulos, A. 1997. Rapid susceptibility testing of fungi by flow cytometry using vital staining. J. Clin. Microbiol. 35:5-10. Wilson, H.A. and Chused, T.M. 1985. Lymphocyte membrane potential and Ca2+ sensitive potassium channels described by oxonol dye fluorescence measurements. J. Cell. Physiol. 125:7281. Yurkow, E.J. and McKenzie, M.A. 1993. Characterization of hypoxia-dependent peroxide production in cultures of Saccharomyces cerevisiae using flow cytomtery: A model for ischemic tissue destruction. Cytometry. 14:287-293.
http://www.bdbiosciences.com/ immunocytometry systems/application notes/
http://www.partec.de/applications/notes.html Application notes on the analysis and enumeration of microorganisms.
Contributed by Hazel M. Davey University of Wales Aberystwyth, United Kingdom Douglas B. Kell University of Manchester Institute of Science and Technology Manchester, United Kingdom Dieter H. Weichart Max Planck Institut f¯ur Molekulare Genetik Berlin, Germany
Key References Amann et al., 1995. See above. Important review of microbial diversity and phytogenetics using nucleic acid probe methods that do not require microbial cultivation.
Arseny S. Kaprelyants Russian Academy of Sciences Moscow, Russia
Microbiological Applications
11.3.21 Current Protocols in Cytometry
Supplement 29
Sorting of Bacteria
UNIT 11.4
During recent years the number of documented applications for analyzing and separating a variety of bacterial types by flow sorting has increased dramatically, following along with an increase in the sensitivity levels measured by the currently available sorting technology. A variety of small bacterial types (<1 µm) can be detected based upon light-scatter measurements and/or fluorescence for subsequent sorting with either jet-inair or SortSense (Beckman Coulter) flow cell tip configurations using the appropriate water- or air-cooled lasers. Because this is a relatively new area of cytometric applications, the limits do not lie in what has been accomplished, but rather in what has yet to be tried with the existing instrumentation. This unit is designed to provide researchers with procedures to set up a flow sorter for analysis and sorting of bacteria. The initial requirement for sorting bacteria is the preparation of a cell sorter to eliminate bacterial and fungal contamination introduced by the sheath fluid and to reduce noise from small particles in the fluid. Procedures are provided to verify the sterility of the instrument and to assist in the optical and sort alignment for small particle detection. While the sorting instrument discussed in the protocols below is an ELITE ESP (Beckman Coulter), the procedures outlined should work on other existing sorting configurations, assuming the operator is proficient at standard sorting and alignment procedures. Small (0.5-µm) beads are used to demonstrate optical instrument alignment using peak side scatter as the trigger signal for both jet-in-air and SortSense tips. Basic Protocol 1 covers instrument setup, while Basic Protocol 2 outlines a procedure for sorting bacteria based on their expression of green fluorescent protein (GFP) to separate transformed from nontransformed bacteria after GFP cloning. Other applications of the sorting methodology include sorting of bacteria for culturability determination, molecular and immunological detection and isolation of specific microorganisms from a heterogeneous population, and selection of genetically altered or mutated bacteria. INSTRUMENT PREPARATION FOR BACTERIA SORTING Prior to the initiation of a bacterial sorting project it is necessary to establish and verify the sterility and alignment of the cytometer. The use of prefiltered, autoclaved sheath fluid and the addition of a 0.22-µm filter just preceding the flow-cell body provide relatively aseptic and particle-free sheath fluid. The side-scatter signal is moved to the second photomultiplier tube (PMT) to allow peak signal processing, and small (0.5-µm) beads are used to verify instrument sensitivity and optimize the alignment process.
BASIC PROTOCOL 1
Materials 70% (v/v) ethanol Phosphate-buffered saline (PBS; APPENDIX 2A) filtered to remove particles >0.22 µm and autoclaved for 20 min 0.5% (w/v) sodium hypochlorite LB liquid medium (see recipe) Cell sorter with 488-nm excitation (15 mW) Luer lock fittings: male and female luer-to-tubing adapters Sterivex GP Filter (Millipore) 10-µm alignment beads 2-µm YG beads (Polysciences) diluted 1:3000 in filtered PBS 0.5-µm YG beads (Polysciences) diluted 1:5000 in filtered PBS 12 × 75–mm tubes, sterile 17 × 120–mm (15-ml) conical tubes, sterile (Sort Collection Tubes) Contributed by Kristi R. Harkins Current Protocols in Cytometry (1999) 11.4.1-11.4.12 Copyright © 1999 by John Wiley & Sons, Inc.
Microbiological Applications
11.4.1 Supplement 7
Sterilize instrument 1. Fill sheath tank with 70% ethanol, empty the in-line sheath filter of PBS, and pressurize the cytometer. 2. Allow ethanol to fill the sheath lines while bleeding the air from the in-line sheath filter, and establish a sheath stream. Run the ethanol through the cytometer for 1 hr. Proceed with normal shutdown procedures and leave 70% ethanol in the cytometer sheath lines for 24 hr. 3. Remove the ethanol from the sheath tank and in-line sheath filter, and replace with filtered, autoclaved PBS. Pressurize the system and allow the PBS to fill the sheath lines while bleeding the air from the in-line sheath filter. Establish a sheath stream and allow the system to run fluid for 15 min. 4. Place the system in vacuum mode, cut the sheath fluid line near the flow cell, and insert the male and female luer lock fittings. Add 0.5% sodium hypochlorite to the line closest to the flow cell and allow the solution to pass through the sheath line into the flow cell to be removed through the vacuum line. Next, attach the Sterivex GP filter where the sheath line was cut so that the male side of the filter is nearest the flow cell body and represents the outlet side of the filter. Establish a sheath fluid stream. 5. Run sheath fluid through the lines for 1 hr. 6. Sterilize the sample line by running 70% ethanol through the line for 5 min. Back flush sheath fluid through the line to remove the ethanol solution. Sterilize the sample uptake probe with 70% ethanol and allow the probe to air dry prior to putting a sample on the instrument. For safety reasons, be sure to turn off the voltage to the sorting plates during this step.
Assess instrument sterility 7. Test sterility of the sheath fluid by using the sort test option on the sorter: in the sheath mode, sort 1 to 100,000 test drops increasing in 10-fold increments into 2 ml LB medium (for each sort) in 15-ml conical test tubes. 8. Test sterility of the sample line by back flushing a few drops of sheath fluid through the sample line into 2 ml LB medium in 15-ml conical test tubes. 9. Test the air around the sorter by leaving 2 ml LB medium open in the sort compartment for a period of time similar to that required for the sort process. 10. Incubate all three samples 18 hr at 37°C with shaking. The medium should remain clear. In situations where contamination appears in the LB medium, the contamination probably resides within the Sterivex filter or after the filter. Replace the filter, sample tubing, and sheath fluid line between the filter and the flow cell body. Soak the flow cell body in a 0.5% sodium hypochlorite for 15 min, then rinse thoroughly in sterile water prior to reassembly onto the sorter. Repeat steps 7 to 10. See Critical Parameters and Troubleshooting section (below) for discussion of precautions to be taken to prevent contamination.
Acquire bead data on cell sorter for alignment testing and sort calibration 11. Set up the cytometer to trigger on peak 90° side scatter by moving the side-scatter filters back to the second PMT. Sorting of Bacteria
11.4.2 Supplement 7
Current Protocols in Cytometry
12. Use the beam-focusing lens that provides the smallest beam spot (15 × 60 mm for SortSense or 40 × 80 mm confocal for jet-in-air), and follow normal optical alignment procedures using standard 10-µm alignment beads with peak 90° light scatter as the trigger signal and monitoring the peak forward scatter signal.
255 jet-in-air
mean channel = 648
Sortsense
mean channel = 617
191
127
Count
63
0
1070
802
535
267
0 0
200
400
600 Peak side scatter
800
1000
Figure 11.4.1 Analysis of the peak side-scatter trigger signal from the 0.5-µm beads used for optical alignment purposes. Top panel: beads analyzed using a jet-in-air tip (20 mW of 488-nm excitation from an air-cooled argon laser) with a mean channel of 648. Bottom panel: the same beads analyzed with SortSense tip technology (15 mW of 488-nm excitation from an air-cooled argon laser) with a mean channel of 617. The peak side-scatter signal was the trigger signal with a discriminator setting of 100. In both cases there is good separation between the peak side-scatter mean and the noise level (100 to 150).
Microbiological Applications
11.4.3 Current Protocols in Cytometry
Supplement 7
13. Run a sample of 2-µm YG beads and further optimize the optical alignment while monitoring the peak forward- and side-scatter signals. YG beads have proven to give the best results; other brands that have been tried tend to kill bacteria or cause small-particle contamination.
14. Run a sample of filtered PBS and adjust the discriminator to bring the data rate to 0. 15. Run a sample of 0.5-µm YG beads and establish a gate around the bead population on the log forward- versus log side-scatter histogram. Save the instrument protocol (gains, high voltage, laser power, discriminator setting). Use this as the standard alignment protocol for future bacterial sorting experiments (see Fig. 11.4.1 for an example of peak side scatter for these beads). At this point it will be necessary to remove any neutral-density filter in front of the forward light-scatter sensor. This will improve detection of small-particle forward-angle light scatter.
16. Adjust the crystal frequency and percent drive settings to those normally optimal for the instrument and the tip (jet-in-air or SortSense) in use, and establish a sort test stream. Only minor adjustment of the crystal drive amplitude (%) should be required to establish a stable and uniform sort test stream.
17. Proceed with a sort matrix to set up the drop delay, using the 2-µm beads for easy microscopic visualization. BASIC PROTOCOL 2
BACTERIAL SORTING FOR ENRICHMENT OR CLONING This procedure sorts green fluorescent protein (GFP)−positive bacteria from GFP-negative bacteria as a means of recovering successful transformants after GFP cloning. Settings are provided for bulk sorting of the GFP-positive bacteria with jet-in-air and SortSense tips using a 488-nm air cooled laser. GFP-expressing bacteria are reanalyzed so that purity can be assessed. The use of sorting to establish bacterial clones is also discussed. Materials Log-phase bacterial culture 0.85% (w/v) NaCl (saline solution) 70% (v/v) ethanol LB liquid medium or 100 × 15–mm LB plates (see recipes) Cell sorter with jet-in-air or SortSense tip (Beckman Coulter), cloning device (e.g., Autoclone; Beckman Coulter), and air-cooled argon laser (488 nm) Prepare sample 1. Dilute a log-phase culture 1/500 in 0.85% saline solution to achieve an optimal cell density for analysis (106/ml). Cell density can be verified by using a Petroff-Hausser counting chamber (Fisher). If necessary, aggregates can be dispersed by passing the bacteria through a 27-G needle or using a sonic dismembrator with a 3-mm-diameter tip (e.g., Fisher) at 23 kHz.
2. Sterilize the sample line by running 70% ethanol through it for 3 min, then back flush sheath fluid through the line to remove the ethanol. Sterilize the sample uptake probe with 70% ethanol and allow the probe to air dry prior to putting a sample on the instrument.
Sorting of Bacteria
This sterilization procedure should be carried out after analyzing with beads and between samples. For safety reasons, be sure to turn off the voltage to the sorting plates during this step.
11.4.4 Supplement 7
Current Protocols in Cytometry
3. Using the PMT settings from the instrument protocol saved from analysis of 0.5-µm beads (see Basic Protocol, step 15), run a filtered saline sample to determine the background noise level. Adjust the peak side-scatter discriminator to minimize this signal (i.e., reduce data rate to 0). 4. Place the diluted bacteria in a sterile sample tube and establish a data rate within the normal limits of the instrument (usually 500 to 4000 particles per second). Keep the sample differential pressure at a level similar to the one used for bead analysis and alignment. If the data rate is too high at this differential pressure, dilute the sample accordingly and reanalyze. If the data rate is too low, repeat the sample dilution at a lower dilution ratio. Be sure to include a side-scatter obscuration bar and horizontal obscuration bar as the forward-scatter mask when using the jet-in-air tip.
565 jet-in-air
mean channel = 21
423
282
Count
141
0
1269 Sortsense
mean channel = 218
951
634
317
0 0.1
1
10 Log side scatter
100
1000
Figure 11.4.2 Analysis of the side scatter (log scale) of Escherichia coli. Top panel: bacterial sample analyzed using a jet-in-air tip (20 mW of 488-nm excitation from an air-cooled laser with a 40 × 80–mm lens) with a mean channel of 21. Bottom panel: the same sample analyzed using a SortSense tip (15 mW of 488-nm excitation from an air-cooled laser with 15 × 60–mm lens). Instrument settings were the same as those described in Figure 11.4.1. Note the significantly higher mean channel value (218) with the SortSense tip, indicating a higher level of sensitivity. This figure also illustrates how different the scatter patterns of bacteria can be relative to the scatter patterns of beads, which show similar mean channels in Figure 11.4.1.
Microbiological Applications
11.4.5 Current Protocols in Cytometry
Supplement 7
325
pre-sort Sortsense tip
243
97
pre-sort jet-in-air tip
72 48
162
23%
23% 24
0 150
0 100
Count
81
112
post-sort Sortsense tip
75
94%
75
post-sort jet-in-air tip
50
85%
25
37 0 0.1
1
10
100
0 1000 0.1 GFP fluorescence
1
10
100
1000
Figure 11.4.3 Analysis and separation of Escherichia coli expressing the red-shifted variant of GFP under a constitutive promoter. The top panels represent the analysis of the bacteria mixture before sorting and the bottom panels represent reanalysis of the GFP-positive fraction after sorting. SortSense sorting provided very good separation between bacteria that were negative and positive for GFP expression using 15 mW of 488-nm excitation from an air-cooled argon laser and a 15 × 60–mm beam-focusing lens (left panels). Jet-in-air sorting provided reasonable separation between the bacteria that were negative and positive for GFP expression using 20 mW of 488-nm excitation from an air-cooled argon laser and a 40 × 80–mm beam-focusing lens (right panels).
5. Run a bacterial sample on the cytometer and accumulate a log side-scatter versus log forward-scatter distribution (see Fig. 11.4.2 for an example of log side scatter). Draw a gate around the bacterial population on the scatter display and collect a log GFP fluorescence signal (see Fig. 11.4.3, upper panels). 6. Sterilize the surfaces within the sorting compartment with 70% ethanol. If the sorter is designed with a closed sort compartment area, this should be closed after sort collection tubes are in place for sorting.
Perform bulk sorting 7. Using the sort setup that provides the highest purity, sort 30,000 bacteria using all three signals as sort criteria to separate the GFP-positive bacteria. Assess purity by reanalyzing a portion of the sorted sample (see Fig. 11.4.3, lower panels). 8. Between samples, back flush the sample tubing and run 70% ethanol through the sample lines for 3 min to clean the lines of bacteria from the previous sample. For each new experiment on the sorter, after this cleaning step, set up a control sort to verify the sterility of the instrument and provide a negative control for any post-sort assays. Using sort test drops, sort a number of empty drops equivalent to the number of bacteria sorted (e.g., 30,000) into a test tube containing the appropriate bacterial culture medium or buffer.
Sorting of Bacteria
11.4.6 Supplement 7
Current Protocols in Cytometry
Sort cells to establish individual clones 9. Program the cloning device on the instrument to sort one cell per well into liquid LB medium (200 µl/well) or buffer in a 96-well plate, or onto a petri dish containing solid LB medium. If using an Autoclone (Beckman Coulter), in order to sort cells onto solid medium it is necessary to modify the 96-well plate holder of the instrument. This is done by bending outward the center of the metal rails that normally hold a 96-well plate to allow the petri dish to sit level in the plate holder. A piece of double-sided tape on the bottom of the petri dish will keep the plate from slipping during the cloning process. Approximately 60 individual colonies can be delivered to a petri dish by this cloning. A single-cell three-drop sort is recommended to minimize dehydration when sorting onto solid medium.
10. Include a negative control plate to verify the sterility of the sheath fluid and sample lines: run a filtered saline sample and use the 96-well cloning program in conjunction with sort test to sort empty drops across a 96-well plate (200 µl LB medium/well) or the area of a petri dish of solid LB medium. REAGENTS AND SOLUTIONS Use deionized water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
LB (Luria-Bertani) liquid medium 10 g Bacto tryptone (Difco) 5 g Bacto yeast extract (Difco) 10 g NaCl H2O to 1 liter Mix until dissolved, and adjust pH to 7.0 with 5 N NaOH Sterilize by autoclaving for 20 min on liquid cycle Store in dark up to 2 months at room temperature LB plates 10 g Bacto tryptone (Difco) 5 g Bacto yeast extract (Difco) 10 g NaCl 15 g Bacto agar (Difco) H2O to 1 liter Mix, and adjust pH to 7.0 with 5 N NaOH Sterilize by autoclaving for 20 min on liquid cycle Cool to 50°C, swirl mixture, and pour ∼30 ml per dish into 100 × 15–mm petri dishes. Store the solidified plates inverted up to 4 weeks at 4°C. COMMENTARY Background Information Davey and Kell (1996) provide an excellent review article on flow cytometry of bacteria, including a section on cell sorting. Briefly, published applications include sorting of bacteria for culturability determination, molecular and immunological detection and isolation of specific microorganisms from a heterogeneous population, and the selection of genetically altered or mutated bacteria.
Sorting applications Methods employed to kill bacterial contaminants in food products must be tested to determine their effectiveness in preventing further bacterial growth. Flow sorting provides a rapid means of verifying functional measurements of viability to determine the culturability of the microorganism. The ELITE ESP sorter (Beckman Coulter) has been used to confirm the analysis of viability and culturability of stressed bacteria using a variety of dyes. Nebevon Caron et al. (1998) used propidium iodide,
Microbiological Applications
11.4.7 Current Protocols in Cytometry
Supplement 7
Sorting of Bacteria
bisoxonol, and ethidium bromide to distinguish intact cells, deenergized cells, and depolarized cells of laboratory-cultured Salmonella typhimurium. Reproductive viability was demonstrated in all three populations by colony growth on specialized solid medium. Votyakova et al. (1998) showed that it was possible to resuscitate starved Micrococcus luteus cells after sorting based upon rhodamine 123 fluorescence (see UNIT 11.3). The cultivation methods generally used in studying microbiological diversity may bias the results through selective recovery of bacterial types. Flow sorting provides a tool for selecting major and minor populations based upon defined cytometric parameters, alleviating this bias. Using water-cooled laser excitation on a FACStar Plus flow sorter (Becton Dickinson), Wallner et al. (1997) sorted and enriched bacteria up to 280-fold from activated sludge, lake water, and lake sediment based on light scattering, DNA content, and specificity of fluorescein-labeled rRNA probes. Polymerase chain reaction (PCR) amplification, used post-sorting to amplify 16S ribosomal DNA of bacteria sorted based upon fluorescence in situ hybridization of rRNA-specific probes, revealed unknown species of bacteria in activated sludge. Flow sorting of bacteria based on immunofluorescence labeling has been useful in the identification of particular bacterial species from soil and water samples. The effectiveness of this method depends on the specificity of the antibodies. Multicolored adsorbed monoclonal antibodies directly conjugated to fluorochromes are recommended to provide a higher degree of specificity in tagging bacteria from environmental samples. This has the advantage that the bacteria remain viable for further culture after sorting. However, because the antibodies used must be generated to antigens that are able to be purified, this technique is limited to the detection of culturable bacteria species (Porter et al., 1996; UNIT 11.2). The use of flow sorting to isolate genetically altered or mutated bacteria is an approach that holds great potential. While chemical selection can be used to isolate transformed bacteria, flow sorting provides a means to select for high-level production of a detectable product. Fluorescent substrates are available that allow the measurement of β-galactosidase activity (UNIT 9.5), and these have been used in conjunction with flow sorting to select viable bacteria (Nir et al., 1990). More recently, green fluorescent protein (GFP) and its variants have been
engineered into bacteria and shown to be expressed at a detectable limit (see also UNIT 9.12). Tombolini et al. (1997) engineered Pseudomonas fluorescens to express the gene for a red-shifted GFP through a constitutive promoter. Fluorescence intensity was readily detectable and stable even when the bacteria were cultured under starvation conditions. Data shown above (Fig. 11.4.3) demonstrate the results of sorting based on GFP intensity and the potential for selecting bacteria that have a high level of GFP expression, possibly resulting from a higher gene copy number. This in turn will allow researchers to use flow cytometry to monitor environmental effects on genetically altered bacteria (biosensors) released into the environment. When chemical selection is not possible, flow sorting can be used to select for mutant or transformed bacteria that have specific protein expression characteristics. A high-level gramicidin S−hyperproducing mutant of Bacillus brevis was isolated by flow sorting based on fluorescein isothiocyanate fluorescence intensity (Azuma et al., 1992). Specific fluorescent tags (lectins or antibodies) can be used to sort out genetically engineered bacteria, as in the isolation of Escherichia coli producing singlechain antibodies against digoxin (Francisco et al., 1993). Gel microdrop encapsulation of bacteria provides another approach to measuring bacterial properties by cytometry. The gel microdrops (GMDs), ∼25 µm in diameter, have been used to measure the susceptibility to various antibiotics of small mycobacteria (Ryan et al., 1995). A single bacterium was encapsulated in a GMD and cultured in medium containing various antibiotics. Growth was measured by the increase in propidium iodide fluorescence after fixation. GMD technology has been used in conjunction with cell sorting to isolate mammalian cells based upon the detection of secretory products captured and labeled within the GMD. This technology could be expanded to include the isolation of transformed bacteria secreting a specific byproduct or the isolation of antibiotic-resistant bacteria for further sequence analysis. Alternative methods Magnetic selection has proven useful in the enrichment of E. coli from fecal, soil, slurry, grass, and water samples (Safarik et al., 1995; Porter et al., 1997). This method can be used with samples that are unsuitable for flow sorting, such as soil samples, and gives higher
11.4.8 Supplement 7
Current Protocols in Cytometry
throughput than methods using traditional sorters. Purities obtained by flow sorting are equivalent to those from magnetic separation in cultured-organism samples but tend to be higher than for magnetic separation in environmental samples. As with any sorting process, the purity is dependent upon the specificity of the lectin or antibody attached to the iron particle or fluorochrome. Magnetic selection has been used in tandem with flow sorting to enrich for a subset of mammalian cells prior to sorting when it is important to have large number of a very pure sorted population. This approach may be equally useful in bulk sorting of bacteria. Optical trapping provides another alternative to flow sorting. A tightly focused infrared laser beam can be used to capture a bacterium based upon either microscopically observed morphology or fluorescence characteristics (Mitchell et al., 1993; Beck and Huber, 1997). This method has the disadvantage of a low sorting efficiency (0.2% to 2%), and the instruments generally need to be home made. The method does have the advantage of permitting a controlled environment, which cannot be attained in flow sorting; it can therefore be applied to cells that must be maintained under anaerobic conditions, or in media that are not amenable to sorting.
Critical Parameters and Troubleshooting Because of the diversity of bacterial species and of the possible applications of cell sorting, it is necessary to consider the requirements of both the organism and the application when designing a sorting experiment. Topics to consider include the optimal buffer and osmotic potential of the sheath fluid, the size of the bacteria and parameters to be used for detection, whether the sorting is for purity or yield, the desired final outcome (bulk sorting or cloning, culture or secondary assay such as PCR), and the biosafety aspects of sample handling (see also UNIT 3.3). Instrument sterility Prefiltering and autoclaving the sheath fluid insures its sterility and minimizes background particle-scatter signal noise. The additional inline sheath filter near the flow cell body insures the sterility of the fluid entering the flow cell body. The Sterivex GP filter recommended in Basic Protocol 1 has been documented to filter up to 3 liters and withstand pressures of 45 psi with a relatively high fluid flow rate so that no
significant drop in sheath pressure across the filter is observed. This means standard crystal frequency, drive percentage, and drop delay settings can be used. As the filter fills with particles, the sheath fluid pressure at the tip will be reduced and this will lead to droplet breakoff instabilities, indicating that it is time to replace the filter. The room housing the sorter should be kept clean and dust free, with reduced air flow during the sorting process, to minimize the chances of a contaminated sort resulting from dirty room air. Use of proper control samples will allow the operator to determine if contamination is coming from the sample line, sheath line, or room air (see Basic Protocol 1, step 10). Instrument sensitivity The best signal-to-noise ratios of the peak forward- and side-scatter signals occur when sorting with sense-in-quartz technology and using a 15 × 60–mm beam-focusing lens. This setup yields a relatively slow pulse width (∼8 µsec) and a more intense laser beam spot than the standard narrow elliptical beam. Under these conditions the particle is in this intense beam for a longer period of time, leading to a larger scatter signal. It is important to use a narrow forward-scatter sensor beam obscuration bar (jet-in-air tip) or small beam obscuration spot (SortSense tip) to allow greater accumulation of low-angle light scatter. Noise from small particles in environmental samples can reduce the ability to discriminate small bacteria based on scatter. If this is the case, it will be necessary to use a fluorescent tag to detect the bacteria of interest and separate them from other particles (see also UNIT 11.2). Purity It is very important to consider the potential difficulty in achieving a pure population after sorting environmental samples, because cells often form heterogenous aggregates that contribute to the impurities after sorting. Bacteria can be dissociated prior to sorting by pushing the sample through a small-bore needle. Potential problems include needle clogging and possible cellular damage from shearing. A sonic probe has been successfully employed for dispersion of aggregates (Nebe-von Caron et al., 1996), though again cellular damage may occur. In either case it is important to use traditional microbiological culture methods to determine if there is any detrimental effect on viability (see also UNIT 11.3).
Microbiological Applications
11.4.9 Current Protocols in Cytometry
Supplement 7
Additionally, environmental samples also contain extraneous materials, which increase the number of background events measured on the cytometer. Unfortunately, most environmental bacteria are small (<1 µm), so it is often necessary to increase the discriminator to reduce the data rate to a reasonable level. This may also prevent the detection of smaller bacteria that do not trigger a measurable signal, and can contribute to impurities in the sorting process. Some applications where purity is very important may require a second sorting step to further enrich for the population of interest, or may benefit from a preenrichment step using magnetic separation if a cell-specific ligand is available. Ideally, if secondary procedures such as PCR could be performed on smaller numbers of cells, the contamination problems of bulk sorting would not pose a significant problem (Wallner et al., 1997). As with any sorting procedure, it is important to monitor the data rate and remain within the signal-processing limits of the instrument. Commercial sorting technology includes signal monitoring systems that will determine when there is insufficient time between peak pulses. This makes it especially important to use a peak signal with a good signal-to-noise ratio, as with peak side scatter. Pulse pileup (PPU) is monitored on the ELITE ESP to abort the sort when a second cell enters the beam before the first cell exits the beam. With the jet-in-air tip the cell moves at a higher velocity, leading to shorter pulse widths and less chance of pulse pile up. It should be kept in mind that this higher velocity with the jet-in-air tip also means that the cell spends less time in the beam and will reduce the signal-to-noise ratio (see Fig. 11.4.2).
Sorting of Bacteria
Post-sorting processes It is important to select a sheath fluid composition that will provide sufficient ions to hold a droplet charge during the sorting process and yet not affect the osmotic potential of the cells. If one is performing a post-sorting process such as PCR or a biochemical assay, it is important to be aware of the effect the sheath fluid components might have on that assay. For example, the effect of EDTA (in the sheath fluid) in chelating magnesium would hinder the PCR process. Manufactured sheath fluids must be devoid of bactericides or fungal static agents that might affect the growth process of bacteria post-sorting. In the case of viability studies, it may be necessary to utilize specialized medium types to initiate growth as a means of verifying
the identification of injured versus dead bacteria. If one is sorting to determine the culturability of these bacteria, it should be remembered that dehydration and ionic change occur when sorting onto a dry surface and will further stress the bacteria and make the initiation of growth difficult. Using a three-drop sort during cloning reduces the effect of dehydration and ionic change on the bacterium. Safety In all cases, consider the biosafety level rating of the microorganism in question. A resource of information on microorganisms and their recommended biosafety rating can be found in the United States Health and Human Services publication Biosafety in Microbiological and Biomedical Laboratories, 3rd edition (U.S. DHHS, 1993). In the author’s laboratory anything that is considered a human pathogen or has a recommended biosafety level greater than 1 is not a candidate for sorting. Schmidt et al. (1997; see also UNIT 3.3) provide a set of guidelines for sorting unfixed cells. Bacteria expressing GFP can be fixed in 1% formaldehyde with no detrimental effect on the fluorescence when viable sorting is not important, for example when PCR will be the final process in the experiment. Wallner et al. (1998) have also shown that PCR of rDNA can be performed on formaldehyde-fixed bacteria. However, RNA-based PCR does not work well in formaldehyde-fixed animal cells.
Anticipated Results The success of each potential application will depend on whether the sorter can distinguish the bacteria from other particles in the mixture. Figure 11.4.1 depicts a very good signal-to-noise separation achieved when analyzing 0.5-µm beads with either a jet-in-air or SortSense flow cell tip in conjunction with relatively low laser power (15 to 20 mW). Although an operator can align the sorter so that these particles are detectable, it is important to remember that bacteria of a similar size will probably scatter less laser light than beads do, and may therefore be rather more difficult to detect. Figure 11.4.2 shows a reduction in the peak side-scatter intensity of bacteria using the jet-in-air versus the SortSense tip, supporting the idea that the SortSense tip will provide a better signal-to-noise ratio in measuring small particles. When discerning very small bacteria poses a problem, it is best to label the microorganism of interest with a fluorescent dye to allow detection and separation relative to other
11.4.10 Supplement 7
Current Protocols in Cytometry
small particles. This may be accomplished using an antibody or lectin dye conjugate, by fluorescence in situ hybridization of DNA or RNA, or (if applicable) based upon functional differences that can be distinguished through the use of a fluorescent dye. The final outcome of the sort will also depend upon the experimental design. Sterile sorts can routinely be achieved using the setup outlined in Basic Protocol 1. The actual growth of the organism post-sorting will depend upon the original viability state of the bacteria prior to the sort and the type of medium used to resuscitate or initiate growth. A good positive control for these variables is to use traditional microbiological techniques on a nonsorted sample to verify that growth is possible. Nonsterile sorting for post-sort processes such as PCR will be affected by the number of bacteria sorted, the purity of the sorted bacteria, and the details of the post-sort assay. Figure 11.4.3 shows a pre- and post-sort analysis of GFP bacteria. Purity in the post-sort sample ranges from 85% with the jet-in-air tip to 94% with the SortSense tip. More than likely there are aggregates of GFP-positive and -negative bacteria, which could explain the partial contamination. The purity can be improved either by performing a second sort of the material from the first sort or by cloning.
Time Considerations Initial sterilization of the sheath lines of the sorter requires 1 day. Installation of the in-line 0.22-µm filter and sterilization of the flow cell body require an additional 2 hr. Verification that the instrument is free of contaminants requires an overnight (16-hr) incubation. Once the system has been verified to be clean, it is not necessary to repeat this process again unless the sorter develops a contamination problem. The in-line filter is replaced each month, and the flow cell body and sample line are sterilized each time prior to a sort. Instrument optical alignment and setup for sorting will take another hour. Sorting for enrichment or purity will take a varying amount of time depending on the number of bacteria desired. Cloning can be completed in <30 min.
Literature Cited Azuma, T., Harrison, G.I., and Demain, A.L. 1992. Isolation of gramicidin S hyperproducing strain of Bacillus brevis by use of a fluorescence activated cell sorting system. Appl. Microbiol. Biotechnol. 38:173-178.
Beck, P. and Huber, R. 1997. Detection of cell viability in cultures of hyperthermophiles. FEMS Microbiol. Lett. 147:11-14. Davey, H.M. and Kell, D.B. 1996. Flow cytometry and cell sorting of heterogeneous microbial populations: The importance of single-cell analyses. Microbiol. Rev. 60:641-696. Francisco, J., Campbell, R., Iverson, B., and Georgiou, G. 1993. Production and fluorescence activated cell sorting of Escherichia coli expressing a functional antibody fragment on the external surface. Proc. Natl. Acad. Sci. U.S.A. 90:10444-10448. Mitchell, J.G., Weller, R., Beconi, M., Sell, J., and Holland, J. 1993. A practical optical trap for manipulating and isolating bacteria from complex microbial communities. Microb. Ecol. 25:113-119. Nebe-von Caron, G., Stephens, P., and Badley, R.A. 1996. Bacterial detection and differentiation by flow cytometry and fluorescent probes. In The Purdue Cytometry CD-ROM Series, Vol. 2 (J.P. Robinson, ed.). Purdue University, West Lafayette, Ind. Nebe-von Caron, G., Stephens, P., and Badley, R.A. 1998. Assessment of bacterial viability status by flow cytometry and single cell sorting. J. Appl. Microbiol. 84:988-998. Nir, R., Yasraeli, Y., Lamed, R., and Sahar, E. 1990. Flow sorting of viable bacteria and yeasts according to β-galactosidase activity. Appl. Environ. Microbiol. 56:3861-3866. Porter, J., Deere, D., Pickup, R., and Edwards, C. 1996. Fluorescent probes and flow cytometry: New insights into environmental bacteriology. Cytometry 23:91-96. Porter, J., Mobbs, K., Hart, C.A., Saunders, J.R., Pickup, R.W., and Edwards, C. 1997. Detection distribution and probable fate of Escherichia coli 0157 from asymptomatic cattle on a dairy farm. J. Appl. Microbiol. 33:297-306. Ryan, C., Nguyen, B.T., and Sullivan, S.J. 1995. Rapid assay for mycobacterial growth and antibiotic susceptibility using gel microdrop encapsulation. J. Clin. Microbiol. 33:1720-1726. Safarik, I., Safarikova, M., and Forsythe, S.J. 1995. The application of magnetic separations in applied microbiology. J. Appl. Bacteriol. 78:575585. Schmid, I., Nicholson, J.K.A., Giorgi, J.V., Janossy, G., Kunkl, A., Lopez, P.A., Perfetto, S., Seamer, L.C., and Dean, P.N. 1997. Biosafety guidelines for sorting of unfixed cells. Cytometry 28:99117. Tombolini, R., Unge, A., Davey, M.E., De Bruijn, F.J., and Jansson, J.K. 1997. Flow cytometric and microscopic analysis of GFP-tagged Pseudomonas fluorescens bacteria. FEMS Microbiol. Ecol. 22:17-28. United States Department of Health and Human Services (U.S. DHHS). 1993. National Institute of Health: Biosafety in Microbiological and Biomedical Laboratories, 3rd ed. Publication
Microbiological Applications
11.4.11 Current Protocols in Cytometry
Supplement 7
No. (CDC) 93-8395, 993. U.S. Government Printing Office, Washington, D.C. Votyakova, T.V., Mukamolova, G.V., Shtein-Margolina, V.A., Popov, V.I., Davey, H.M., Kell, D.B., and Kaprelyants, A.S. 1998. Research on the heterogeneity of Micrococcus luteus culture during an extended stationary phase: Subpopulation separation and characterization. Microbiology 67:85-92. Wallner, G., Fuchs, B., Spring, S., Beisker, W., and Amann, R. 1997. Flow sorting of microorganisms for molecular analysis. Appl. Environ. Microbiol. 63:4223-4231.
Internet Resources www.cyto.purdue.edu/flowcyt/research/micrflow/ Contains the material on flow cytometry and microbiology included in the Purdue Cytometry CD-ROM Series Vol. 2, 1996.
Contributed by Kristi R. Harkins Iowa State University Ames, Iowa
Key Reference Davey and Kell, 1996. See above. An excellent overview of the problems and advantages associated with cytometric analysis of microorganisms; a section on cell sorting describes applications where bacterial sorting is of proven or probable benefit to researchers, and provides references.
Sorting of Bacteria
11.4.12 Supplement 7
Current Protocols in Cytometry
Detection of Borreliacidal Antibodies by Flow Cytometry
UNIT 11.5
Borreliacidal antibodies are lethal to the spirochete Borrelia burgdorferi, the causative agent of Lyme disease. Detection of borreliacidal antibodies is useful for serodiagnosing Lyme disease and monitoring immune status after vaccination. Their detection, however, is dependent on the use of live organisms. Visual assessment of cell viability or monitoring of pH-dependent color changes in growth medium can be used to identify killing by borreliacidal antibodies, but detection by flow cytometry significantly increases sensitivity by allowing evaluation of small numbers of organisms. In addition, data from multiple assays can be rapidly acquired and analyzed objectively. CAUTION: Live organisms are potentially pathogenic. Wear appropriate protective clothing and follow universal precautions. NOTE: All solutions and equipment coming in contact with live cells must be sterile, and proper aseptic technique should be used accordingly. FLOW CYTOMETRIC DETECTION OF ANTI–B. BURGDORFERI ANTIBODIES Live B. burgdorferi and complement are combined with serum. After incubation for 16 to 24 hr, B. burgdorferi are stained with acridine orange, which intercalates into doublestranded nucleic acids and produces green fluorescence (maximum emission 530 nm) when excited by incident laser light (488 nm). If the serum contains borreliacidal antibodies, acridine orange enters the spirochete and concentrates in blebbed surface areas characteristic of cell death. The spirochetes are accurately evaluated by flow cytometry and live (viable) and dead, blebbed (nonviable) B. burgdorferi are discriminated by the intensity of fluorescence.
BASIC PROTOCOL
Materials Borrelia burgdorferi stock culture aliquots (see Support Protocol 3) in BSK medium Barbour-Stoenner-Kelly (BSK) medium (see Support Protocols 1 and 2) Mueller-Hinton agar plates containing B. subtilis spores (see Support Protocol 4) Serum to be tested for anti–B. burgdorferi borreliacidal antibodies Normal antibiotic-free control serum from the same species as serum sample to be tested Normal control serum containing 5 µg/ml doxycycline (see recipe) Amberlite XAD-16 nonionic polymeric resin (Sigma) Phosphate buffered saline (PBS; APPENDIX 2A), pH 7.2, filter-sterilized Penicillinase (optional; Sigma) Guinea pig complement (see Support Protocol 5), tested for activity (see Support Protocol 6) Acridine orange working solution (see recipe) 50-ml screw-cap centrifuge tubes (e.g., Fisher) Filter paper 0.2-µm microcentrifuge spin-filter tubes (1.5 ml; Costar) 56°C water bath Petroff-Hausser counting chamber (Fisher) Dark-field microscope 12 × 75–mm polypropylene or polystyrene tubes as required for flow cytometer Flow cytometer with a 488-nm argon laser and 530/30 band-pass filters Contributed by Steven M. Callister, Dean A. Jobe, and Ronald F. Schell Current Protocols in Cytometry (2003) 11.5.1-11.5.12 Copyright © 2003 by John Wiley & Sons, Inc.
Microbiological Applications
11.5.1 Supplement 26
Cultivate B. burgdorferi 1. Thaw a frozen aliquot containing ∼1 × 108 B. burgdorferi per ml and inoculate 200 µl (2 × 107 organisms) into 6 ml fresh BSK medium. 2. Mix by inversion and incubate 72 hr at 32°C to 35°C. Best results occur when B. burgdorferi organisms are in logarithmic growth phase. Using organisms in stationary or lag phase results in less than optimal detection of borreliacidal antibodies.
Detect antimicrobial agents The presence of antimicrobial agents in serum interferes with detection of borreliacidal antibodies and is the primary cause of false positive results. Suspect samples should be screened for antimicrobial agents prior to testing for borreliacidal antibodies. 3. To screen serum, cut wells ∼5 mm in diameter into a Mueller-Hinton agar plate containing B. subtilis spores. 4. Pipet 50 µl undiluted test serum, normal antibiotic-free control serum, and normal control serum containing 5 µg/ml doxycycline into individual wells. 5. Incubate plates, with wells facing up, 4 to 5 hr at 37°C. 6. Following incubation, examine wells for zones of growth inhibition. Sera contained in wells with significant zones of growth inhibition should be considered positive for the presence of antimicrobial agents.
Remove antimicrobial agents Antimicrobial agent must be removed prior to testing for borreliacidal activity. Removing antimicrobial agents does not significantly affect the titer of borreliacidal antibodies. 7. Wash XAD-16 resin by placing ∼25 g resin in a filter paper–lined funnel and passing 100 ml filter-sterilized PBS over the resin. Blot the washed resin on fresh filter paper to remove excess PBS and transfer to a screw-capped container for storage at room temperature. Place 1 g washed XAD-16 resin into a 50-ml centrifuge tube. 8. Combine 100 µl test serum that has been determined to contain antimicrobial agent with 400 µl filter-sterilized PBS (i.e., a 1-part to 4-part dilution) and mix well. 9. Add diluted serum to XAD-16 resin and incubate 20 min at room temperature with occasional gentle mixing. IMPORTANT NOTE: XAD-16 resin does not remove penicillins. Penicillin or related antibiotics can be removed by adding 1 U penicillinase to serum and incubating 10 min at room temperature.
10. Remove adsorbed serum with a 40- to 200-µl adjustable pipet. Remaining resin beads will be removed during filtration.
Detection of Borreliacidal Antibodies by Flow Cytometry
11.5.2 Supplement 26
Current Protocols in Cytometry
Detect borreliacidal antibodies 11. In the upper chamber of a 0.2-µm, 1.5-ml microcentrifuge spin-filter tube, combine 100 µl serum to be tested with 400 µl BSK medium (i.e., a 1-part to 4-part dilution). In a separate spin-filter tube, combine normal serum from the same species with BSK medium at the same ratio. Filter-sterilize by microcentrifuging 2 min at 3500 rpm. Samples treated for removal of antimicrobial agents can be added directly to the spin-filter tube without further dilution, since dilution has already been performed (see step 8). In this case, dilute the normal serum control with filter-sterilized PBS instead of BSK. Higher rates of centrifugation tend to increase the amount of particulate debris in the sample. If serum samples do not filter, raw serum should be prefiltered through a 0.8-ìm filter prior to dilution in BSK.
12. Aseptically transfer 100 µl of each of the diluted, filter-sterilized sera into separate 1.5-ml microcentrifuge tubes. The remaining 400 ìl can be used to set up duplicate or triplicate samples, or can be frozen for use at a later date.
13. Inactivate complement by incubating serum samples 10 min in a 56°C water bath. 14. Using a Petroff-Hausser counting chamber and a dark-field microscope, determine the number of B. burgdorferi organisms per milliliter in the BSK culture from step 2. A Petroff-Hausser counting chamber operates on the same principle as a standard hemacytometer (see APPENDIX 3A). It may be necessary to dilute the B. burgdorferi culture prior to counting. An appropriate dilution will yield ∼100 spirochetes distributed across 10 countable squares.
15. Prepare a suspension of B. burgdorferi organisms containing 1 × 106 spirochetes per ml by diluting the appropriate volume of the culture prepared in step 2 into fresh BSK medium. 16. Aseptically add 100 µl B. burgdorferi suspension (1 × 105 organisms) to the diluted sera in the microcentrifuge tubes (from step 12). 17. Add the appropriate volume of guinea pig complement to the B. burgdorferi organisms and diluted sera in the microcentrifuge tubes. Complement must be tested for activity prior to use (See Support Protocol 6). The appropriate volume is dependent on the quality and hemolytic activity of specific lots of complement. In addition, complement is heat-labile and must be used immediately after thawing; the reliability of borreliacidal antibody detection is affected if inactivated complement is used. Excess thawed complement should be discarded.
18. Gently vortex suspensions to mix thoroughly and incubate 16 to 24 hr at 32°C to 35°C. Acquire flow cytometric data 19. Adjust the flow cytometer settings to detect B. burgdorferi organisms (Table 11.5.1). The excitation and emission wavelengths are 488 and 530 nm, respectively.
20. Transfer 100 µl of each sample (from step 18) into a separate 12 × 75–mm test tube containing 400 µl PBS, pH 7.2, and 50 µl acridine orange working solution.
Microbiological Applications
11.5.3 Current Protocols in Cytometry
Supplement 26
Table 11.5.1 FACScan Flow Cytometer Settings for Detection of Borreliacidal Antibodya,b
Parameter
Voltage
Threshold
Forward scatter Side scatter AO fluorescence
E00 350 651
240 052 052
aUnused fluorescence detectors can be turned off. Abbreviation: AO, acridine
orange. bSettings are for the FACScan only. Settings for other flow cytometers will be
different.
21. Run the sample containing normal serum and B. burgdorferi on the flow cytometer. Acquire forward scatter, side scatter, and fluorescence data for a minimum of 1500 events. a. Use a forward light scatter versus 90° side scatter dot plot to examine the distribution of B. burgdorferi (Fig. 11.5.1A). b. Use a 90° side scatter versus 530-nm fluorescence dot plot to distinguish the B. burgdorferi organisms from debris (Fig. 11.5.1B). c. Draw a gate R1 around the B. burgdorferi organisms to exclude debris (Fig. 11.5.1C). Gating based on fluorescence rather than on light-scatter parameters will yield more accurate data.
Analyze listmode data files 22. Using gated events from the normal serum control (Fig. 11.5.2A), set a histogram marker (M1; Fig.11.5.2B). The histogram marker setting is arbitrary, but should not exceed 8% of the total gated population.
23. Analyze the unknown serum to determine the percentage of B. burgdorferi organisms shifted into the region defined by the histogram marker. B. burgdorferi organisms that have been killed by borreliacidal antibodies will increase in fluorescence intensity due to blebbing and increased uptake of acridine orange (Fig. 11.5.2C). This causes a significant increase in mean channel fluorescence and a corresponding shift into the region defined by the histogram marker (Fig.11.5.2D). Since borreliacidal antibodies are lytic, a significant decrease in the numbers of gated events will also occur. SUPPORT PROTOCOL 1
Detection of Borreliacidal Antibodies by Flow Cytometry
PREPARATION OF BARBOUR-STOENNER-KELLY (BSK) MEDIUM Barbour-Stoenner-Kelly (BSK) broth medium is used for in vitro cultivation of B. burgdorferi and is the primary substrate used in the borreliacidal antibody test. Quality control of BSK is necessary to ensure optimal growth of the spirochete and performance of the borreliacidal assay (see Critical Parameters). Materials HEPES (Sigma) Neopeptone (Difco) Sodium citrate (Sigma) Glucose (Sigma) Sodium bicarbonate (Sigma)
11.5.4 Supplement 26
Current Protocols in Cytometry
90° light scatter
A 104
C
B
103
R1
102 101 100 100
101 102 103 104 Forward light scatter
Acridine orange fluorescence
Figure 11.5.1 Flow cytometric detection and gating of B. burgdorferi organisms (excitation, 488 nm; emission, 530 nm). (A) B. burgdorferi and debris. (B) Differentiation of B. burgdorferi from debris using acridine orange. (C) Gating of B. burgdorferi organisms.
TC yeastolate (Difco) Pyruvic acid (Sigma) N-acetyl glucosamine (Sigma) Bovine serum albumin (Sigma) Gelatin (microbiological grade; Difco) 5 N NaOH 10× CMRL 1066 medium with L-glutamine and without sodium bicarbonate (ICN Biomedicals) Rabbit serum (Life Technologies), heat-inactivated 45 min at 56°C. 56°C water bath Positive-pressure pump Millipore filter manifold Prefilter (124 mm) 0.2-, 0.45-, and 0.8-µm filters (142-mm diameter) 0.2-µm bell filters Sterile 100-ml containers Dark-field microscope Prepare BSK medium 1. Combine the following in a 2-liter flask. 900 ml Milli-Q double-filtered or deionized distilled water 6.0 g HEPES 5.0 g neopeptone 0.7 g sodium citrate 5.0 g glucose 2.2 g sodium bicarbonate 2.5 g TC yeastolate 0.8 g pyruvic acid 0.4 g N-acetyl glucosamine 50 g bovine serum albumin (fraction V) 2. Mix components slowly using the slowest stir plate setting. Mixing should require 2 to 4 hr. Vigorous stirring may result in breakdown products toxic to B. burgdorferi.
Microbiological Applications
11.5.5 Current Protocols in Cytometry
Supplement 26
A104
B 100 shift = 8%
80
103
MCF = 248
R1
102
Counts
90° light scatter
events = 1689
101
60 40 20 M1
100
0
C104
D 100 shift = 53% 80
103
MCF = 491
R1
102
101
Counts
90° light scatter
events = 566 60 40 20
M1 100 100
101
102
0 103 104 100 101 Acridine orange fluorescence
102
103
104
Figure 11.5.2 Flow cytometric appearance of B. burgdorferi organisms incubated with complement and normal serum (A,B) or serum containing borreliacidal antibodies (C,D). “Shift” refers to the percentage of the B. burgdorferi population that falls within the region defined by marker 1 (M1). Excitation, 488 nm; emission, 530 nm.
3. While BSK components are mixing, prepare gelatin solution as follows. In a 500-ml flask, combine 200 ml Milli-Q double-filtered or deionized distilled water and 14 g gelatin. Heat at medium setting with stirring to dissolve. Autoclave 15 min at 121°C, then place gelatin solution in a 56°C water bath. 4. When BSK components are dissolved completely, adjust to pH 7.5 with 5 N NaOH. 5. Combine the 200 ml gelatin solution, 100 ml of 10× CMRL 1066 medium, and 64 ml heat-inactivated (45 min at 56°C) rabbit serum to BSK medium. Mix thoroughly. Sterilize BSK medium 6. Using the positive-pressure pump, pump BSK medium through the Millipore filter manifold loaded with 124-mm prefilter and 142-mm diameter 0.2-, 0.45-, and 0.8-µm filters stacked from smallest (bottom) to largest (top) pore size. BSK medium cannot be sterilized by autoclaving. Filtration is necessary to remove large particulates prior to filter sterilization.
7. Filter-sterilize the BSK medium into a sterile container using the positive-pressure pump and a sterile 0.2-µm bell filter.
Detection of Borreliacidal Antibodies by Flow Cytometry
Confirm sterility 8. Aseptically remove a 1-ml aliquot of sterile BSK, transfer to a sterile 1.5-ml microcentrifuge tube, and incubate overnight at 35°C. Store remaining filter-sterilized BSK at 4°C.
11.5.6 Supplement 26
Current Protocols in Cytometry
9. Following incubation, examine using dark-field microscopy to confirm sterility. If the medium is contaminated, warm BSK to room temperature and repeat filter sterilization. The shelf life of BSK is ∼6 months at 4°C.
10. Transfer BSK into sterile storage containers. When filling storage containers, keep headspace to a minimum to reduce oxidation of BSK medium.
QUALITY CONTROL OF BARBOUR-STOENNER-KELLY (BSK) MEDIUM Before using BSK medium in the borreliacidal antibody test, it is important to make sure that BSK supports the growth of small numbers of B. burgdorferi and that spirochetes grow without clumping. A highly variable component of BSK medium is bovine serum albumin (BSA). The quality control protocol outlined below must be followed to obtain optimal results.
SUPPORT PROTOCOL 2
Materials BSK medium, sterile (see Support Protocol 1) B. burgdorferi culture in logarithmic growth phase (see Support Protocol 3) 13 × 100–mm sterile culture tubes Dark-field microscope 1. Arrange ten 13 × 100–mm culture tubes in a test-tube rack. Pipet 6 ml sterile BSK into one tube (labeled tube 1) and 5.4 ml into each of the remaining tubes (labeled tubes 2 to 10). 2. Using a Petroff-Hausser counting chamber and a dark-field microscope, determine the number of B. burgdorferi organisms per ml of stock culture. 3. Prepare a dilution of the stock culture in the BSK contained in tube 1 such that the final concentration in tube 1 is 1 × 106 organisms per ml. 4. Make serial ten-fold dilutions as follows. Transfer 0.6 ml from tube 1 to tube 2. Mix well. Transfer 0.6 ml from tube 2 to tube 3. Mix well. Continue in this fashion up to tube 10. 5. Incubate 3 to 4 weeks at 32°C to 35°C. Examine weekly by dark-field microscopy for the presence of B. burgdorferi. Acceptable BSK medium will yield at least 1 × 108 healthy, unclumped B. burgdorferi per milliliter from an initial inoculum of ≤10 organisms. Less than optimal results are almost always the result of inferior BSA.
PREPARATION OF B. BURGDORFERI STOCK CULTURE The isolate of B. burgdorferi used to detect borreliacidal antibodies is critical. Borreliacidal antibodies will not be detected unless the spirochete expresses the homologous protein on the surface (Callister et al., 1996, 2002; Rousselle et al., 1998).
SUPPORT PROTOCOL 3
Materials Borrelia burgdorferi Barbour-Stoenner-Kelly (BSK) medium (see Support Protocols 1 and 2) 13 × 100–mm sterile culture tubes 1. Inoculate 200 to 500 µl B. burgdorferi organisms into 6 ml fresh BSK medium per 13 × 100–mm sterile culture tube.
Microbiological Applications
11.5.7 Current Protocols in Cytometry
Supplement 26
2. Incubate culture at 32°C to 35°C until B. burgdorferi concentration reaches ∼1 × 108 organisms per milliliter (logarithmic growth phase). 3. Place 200- to 300-µl aliquots of B. burgdorferi culture into separate 1.5-ml microcentrifuge tubes and freeze at −70°C until used. Addition of cryopreservative is unnecessary. Frozen aliquots will remain viable for several years. SUPPORT PROTOCOL 4
PREPARATION OF MUELLER-HINTON AGAR CONTAINING BACILLUS SUBTILIS SPORES Materials Mueller-Hinton agar medium (Difco) Bacillus subtilis (ATCC #6633) spore suspension (Difco) 500-ml flask Sterile 100 × 15–mm petri plates 1. In a 500-ml flask, combine 9.5 g dehydrated Mueller-Hinton agar medium with 250 ml Milli-Q deionized distilled water. 2. Heat at medium setting with stirring to dissolve. Autoclave 15 min at 121°C, then cool agar medium to 56°C in a water bath. 3. Add 1 ml B. subtilis ATCC #6633 spore suspension. Swirl gently to mix. 4. Pour 25 to 30 ml spore-seeded agar medium into sterile 100 × 15–mm petri plates and allow to solidify. 5. Store plates at 4°C up to 3 months.
SUPPORT PROTOCOL 5
REHYDRATION OF LYOPHILIZED COMPLEMENT Complement from a variety of species is readily available and generally supplied lyophilized. Guinea pig complement yields consistent results, but complement from other species may also be used. Complement must be quality controlled (see Support Protocol 6) and care must be taken to maintain maximum activity. Materials Lyophilized guinea pig complement (Sigma, Life Technologies, Rockland, or Cedarlane Laboratories) 0.2-µm and 0.8-µm syringe-tip filters or filter units Sterile microcentrifuge tubes, prechilled 1. Remove vial(s) of lyophilized complement from storage and immediately place in a container of crushed ice. Lyophilized complement should be stored at −20° to −70° C until used.
2. Refer to the supplier’s instructions to determine the amount of distilled water necessary to rehydrate the vial(s) of complement, and allow the water to become ice cold in the same container of crushed ice before adding it to vials. The amount of distilled water necessary for rehydrating will be dependent on the number and size of vials. Detection of Borreliacidal Antibodies by Flow Cytometry
11.5.8 Supplement 26
Current Protocols in Cytometry
3. Using a syringe, add the appropriate volume of ice-cold distilled water to the complement and mix by hand or vortex mixer. Do not allow complement to warm to room temperature. If reconstitution is difficult, keep cold by frequent incubations on ice.
4. Following reconstitution, filter-sterilize the complement by passage through a 0.2-µm filter into a chilled sterile container. It may be helpful to use a prefilter (0.8-ìm or larger) if large particulates are present.
5. Dispense 200- to 500-µl aliquots of sterile complement into chilled sterile microcentrifuge tubes and store at −70°C until used. QUALITY CONTROL OF COMPLEMENT The activity of complement is defined by the number of complement hemolytic units (CH50) necessary to lyse 50% of sensitized sheep red blood cells contained within a standardized suspension. The CH50 value can vary significantly, making it important to determine the optimal amount necessary for detecting borreliacidal antibodies.
SUPPORT PROTOCOL 6
Materials Normal serum and positive control serum (obtained in the laboratory) Barbour-Stoenner-Kelly (BSK) medium (see Support Protocols 1 and 2) B. burgdorferi in logarithmic phase (see Support Protocol 3) Reconstituted complement to be tested (see Support Protocol 5) 0.2-µm filters 56°C water bath Petroff-Hausser counting chamber (Fisher) Dark-field microscope 1. Prepare a 1.5-ml volume of normal serum diluted at a ratio of 1 part serum to 4 parts BSK medium. In a separate tube, prepare the same volume of positive control serum diluted to the same ratio. 2. Filter-sterilize the diluted samples by passage through a 0.2-µm syringe filter. 3. Pipet 100-µl aliquots of diluted normal serum, diluted positive control serum, and fresh BSK medium into separate sets of seven sterile microcentrifuge tubes (total of 21 tubes). 4. Heat-inactivate all samples 10 min at 56°C. 5. Using a Petroff-Hausser counting chamber and a dark-field microscope, determine the concentration of B. burgdorferi organisms per milliliter in a logarithmic phase culture and prepare a suspension of B. burgdorferi containing 1 × 106 organisms per ml by diluting into fresh BSK medium (see Basic Protocol). 6. Add 100 µl B. burgdorferi suspension (1 × 105 organisms) to each of the microcentrifuge tubes. 7. Add 0 µl, 5 µl, 10 µl, 15 µl, 20 µl, 25 µl, and 50 µl complement successively to the microcentrifuge tubes in each of three groups (containing diluted normal serum, positive control serum, and BSK alone). 8. Gently vortex to mix all reagents together and incubate 16 to 24 hr at 32°C to 35°C. Microbiological Applications
11.5.9 Current Protocols in Cytometry
Supplement 26
9. Acquire and analyze data (see Basic Protocol). The optimal amount is the maximum volume necessary to detect borreliacidal antibodies in a positive serum sample without causing nonspecific complement-mediated killing in normal controls. This volume should be used to detect borreliacidal antibodies (see Basic Protocol).
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Acridine orange 2 mg/ml stock solution: Dissolve 0.02 g acridine orange powder in 10 ml filter-sterilized Milli-Q double-filtered or deionized distilled water. Store protected from light up to several months at 4°C. 1 ìg/ml working dilution: Add 25 µl of the 2 mg/ml stock solution to 50 ml of filter-sterilized PBS, pH 7.2 (APPENDIX 2A). Store protected from light for up to three weeks at 4°C. CAUTION: Acridine orange is a possible mutagen. Use care when handling.
Doxycycline, 5 ìg/ml in normal serum 1 mg/ml stock solution: Solubilize appropriate amount of doxycycline powder in 10 ml Milli-Q water according to the following equation:
Weight (mg) =
volume (ml) × concentration (µg/ml) assay potency (µg/mg)
Mix well and filter-sterilize using 0.2-µm syringe-tip filter. Dispense 0.5-ml aliquots into sterile 1.5-ml microcentrifuge tubes and store at −70°C. 5 ìg/ml working concentration: Add 10 µl of 1 mg/ml doxycycline stock solution to 2 ml normal serum. Mix well and dispense 60-µl aliquots into sterile 0.5-ml tubes. Store at −20°C or lower temperature. COMMENTARY Background Information
Detection of Borreliacidal Antibodies by Flow Cytometry
During the early 1980s, the spirochetal bacterium Borrelia burgdorferi was implicated as the causative agent of an epidemic of oligoarthritis previously recognized in Lyme, Connecticut (Steere et al., 1983). This illness, subsequently named Lyme disease, has become the most common tick-associated illness in Europe and the United States. Lyme disease is a multisystem disorder that usually begins with a skin lesion called erythema migrans and with constitutional symptoms. If the disease is left untreated or treated inappropriately, dissemination of the organism can lead to more severe sequelae, including nervous system disorders or arthritis (Steere, 1989).
Vaccinations with B. burgdorferi or several individual B. burgdorferi proteins induce borreliacidal antibodies that provide protection against infection by inducing a complement cascade that kills the spirochetes without the necessity of scavenging by phagocytic cells (Padilla et al., 1996). Borreliacidal antibodies are also induced by infection with B. burgdorferi (Callister et al., 1993, 1996; Jobe et al., 2003), but the spirochetes are not eliminated. Specific mechanism(s) responsible for this enigma remain largely unknown. Nevertheless, detection of borreliacidal antibodies is useful for serodiagnosing Lyme disease (Callister et al., 2002) and monitoring immune status after vaccination (Golde et al., 1997).
11.5.10 Supplement 26
Current Protocols in Cytometry
Borreliacidal antibodies are detected by combining live B. burgdorferi organisms with complement and determining whether the spirochetes are killed in vitro. Live organisms are necessary to discriminate borreliacidal antibodies from antibodies that bind B. burgdorferi but are incapable of killing. The accuracy of detection is improved greatly by using flow cytometry (Callister et al., 1994, 1996). Flow cytometric analyses are objective and large numbers of spirochetes can be evaluated rapidly.
Critical Parameters The success of the flow cytometric procedure is dependent on high-quality BSK, functional complement, and a proper concentration of healthy, nonclumping B. burgdorferi organisms. Strict adherence to the outlined procedures and stringent quality control measures for each component must be followed. Even minor deviations may have a deleterious effect. In addition, antimicrobial agents in the serum to be tested may also kill B. burgdorferi and cause a false-positive reaction (Jobe et al., 1999). Quality control of BSK medium Most ingredients in BSK will not vary from batch to batch or among manufacturers. It is necessary, however, to analyze bovine serum albumin from several manufacturers to identify one with acceptable quality (Callister et al., 1990). BSK should produce and sustain cultures of >108 motile, uniformly refractile, nonclumped B. burgdorferi per ml from an inoculum of ≤10 spirochetes.
conditions. Each new lot number of complement must be evaluated separately to determine optimal volumes for detecting borreliacidal antibodies. Enumeration of B. burgdorferi cultures To ensure accurate counting, B. burgdorferi should be loaded into the Petroff-Hausser counting chamber at a concentration of ∼100 organisms distributed across 10 countable squares. Antimicrobial agents Antimicrobial agents may cause a falsepositive reaction. Serum that may contain antimicrobial agents should be screened and treated to remove antimicrobials prior to testing (Jobe et al., 1999).
Anticipated Results B. burgdorferi must be differentiated from serum debris (see Fig.11.5.1) and live-gated using the normal serum control to detect borreliacidal antibodies accurately. When borreliacidal antibodies are not present, B. burgdorferi are tightly clustered (Fig. 11.5.2A) and the mean channel fluorescence (MCF) does not increase significantly. When borreliacidal antibodies are present, dot plot profiles become diffuse (Fig. 11.5.2C) and the MCF increases significantly compared to the normal serum control (Fig. 11.5.2D). There is also a significant decrease in the numbers of gated events, presumably because B. burgdorferi are destroyed by the borreliacidal antibodies.
Time Considerations B. burgdorferi cultures B. burgdorferi spirochetes must be used during logarithmic growth phases for accurate detection. Using organisms during lag or stationery phases significantly reduces the ability to detect borreliacidal antibodies. Accurate detection of borreliacidal antibodies also depends heavily on the isolate of B. burgdorferi used in the assay. For example, borreliacidal antibodies specific for outer surface protein C (Osp C) are not detected easily unless the B. burgdorferi spirochetes do not express OspA or OspB (Rousselle et al., 1998). Quality control of complement Complement is heat-labile and will lose significant activity if stored at room temperature or above. When rehydrating lyophilized complement, use ice-cold distilled water and perform all filtration and pipetting under ice-cold
Preparing 10 to 15 assays takes ∼30 min. B. burgdorferi should be incubated 16 to 24 hr with serum and complement. Preparation for flow cytometric detection takes 10 min. Flow cytometer optimization takes 10 min. Each sample is evaluated in 1 to 2 min, and analysis takes 15 min.
Literature Cited Callister, S.M., Case, K.C., Agger, W.A., Schell, R.F., Johnson, R.C., and Ellingson, J.L.E. 1990. Effects of bovine serum albumin on the ability of Barbour-Stoenner-Kelly medium to detect Borrelia burgdorferi. J. Clin. Microbiol. 28:363365. Callister, S.M., Schell, R.F., Case, K.L., Lovrich, S.D., and Day, S.P. 1993. Characterization of the borreliacidal antibody response to Borrelia burgdorferi in humans: A serodiagnostic test. J. Infect. Dis. 167:158-164. Microbiological Applications
11.5.11 Current Protocols in Cytometry
Supplement 26
Callister, S.M., Schell, R.F., Lim, L.C.L., Jobe, D.A., Case, K.L., Bryant, G.L., and Molling, P.E. 1994. Detection of borreliacidal antibodies by flow cytometry: An accurate, highly specific serodiagnostic test for Lyme disease. Arch. Intern. Med. 154:1625-1632. Callister, S.M., Jobe, D.A., Schell, R.F., Pavia, C.S., and Lovrich, S.D. 1996. Sensitivity and specificity of the borreliacidal-antibody test during early Lyme disease: a “gold standard”? Clin. Diagn. Lab. Immunol. 3:399-402. Callister, S.M., Jobe, D.A., Schell, R.F., Kowalski, T.J., Lovrich, S.D., and Marks, J.A. 2002. Ability of the borreliacidal antibody test to confirm Lyme disease in clinical practice. Clin. Diagn. Lab. Immunol. 9:908-912. Golde, W.T., Piesman, J., Dolan, M.C., Kramer, M., Hauser, P., Lobet, Y., Capiau, C., Desmons, P., Voet, P., Dearwester, D., and Frantz, J.C. 1997. Reactivity with a specific epitope of outer surface protein A predicts protection from infection with the Lyme disease spirochete, Borrelia burgdorferi. Infect. Immun. 65:882-889. Jobe, D.A, Lovrich, S.D., Schell, R.F., and Callister, S.M. 2003. C-terminal region of outer surface protein C binds borreliacidal antibodies in sera from patients with Lyme disease. Clin. Diagn. Lab. Immunol. 10:573-578. Jobe, D.A., Rawal, N., Schell, R.F., and Callister, S.M. 1999. Detection of borreliacidal antibodies in Lyme borreliosis patient sera containing antimicrobial agents. Clin. Diagn. Lab. Immunol. 6:930-933.
Padilla, M.L., Callister, S.M., Schell, R.F., Bryant, G.L., Jobe, D.A., Lovrich, S.D., DuChateau, B.K., and Jensen, J.R. 1996. Characterization of the protective borreliacidal antibody response in humans and hamsters after vaccination with a Borrelia burgdorferi outer surface protein A vaccine. J. Infect. Dis. 174:739-746. Rousselle, J.C., Callister, S.M., Schell, R.F., Lovrich, S.D., Jobe, D.A., Marks, J.A., and Wieneke, C.A. 1998. Borreliacidal antibody production against outer surface protein C of Borrelia burgdorferi. J. Infect. Dis 178:733-741. Steere, A.C. 1989. Medical progress: Lyme disease. N. Engl. J. Med. 321:586-596. Steere, A.C., Grodzicki, R.L., Kornblatt, A.N., Craft, J.E., Barbour, A.G., Burgdorfer, W., Schmid, G.P., Johnson, E., and Malawista, S.E. 1983. The spirochetal etiology of Lyme disease. N. Engl. J. Med. 308:733-740.
Contributed by Steven M. Callister and Dean A. Jobe Gundersen Lutheran Medical Center and Microbiology Research Laboratory and Gundersen Lutheran Medical Foundation La Crosse, Wisconsin Ronald F. Schell Wisconsin State Laboratory of Hygiene and Department of Medical Microbiology and Immunology Madison, Wisconsin
Detection of Borreliacidal Antibodies by Flow Cytometry
11.5.12 Supplement 26
Current Protocols in Cytometry
Flow Cytometric Detection of Pathogenic E. coli in Food
UNIT 11.6
The major problems encountered in detection of pathogens such as E. coli O157:H7 in foods are (1) the generally low numbers of pathogens in the presence of a large background of microbial flora, and (2) the presence of interfering substances (e.g., particulates) derived from the food itself. The simplest strategy for detection of bacterial pathogens in foods by flow cytometry involves preparation of an extract of the food product followed by filtration to remove large particulates. This is followed by a resuscitation/enrichment incubation in the presence of bacterial culture medium that allows for recovery of stressed or injured organisms, and for an increase in the number of pathogens present. While this procedure is fairly straightforward, testing of food samples with negative results means little in the absence of some type of standardization. For this reason, the method described here combines a procedure for sample testing with a procedure for making a calibration curve by introducing (spiking) log dilutions of the target organism into replicate samples of beef enrichment culture as a control pursuant to testing of actual unknown samples. Once the accuracy of the method is established, it should not be necessary to repeat the calibration procedure (which involves doing microbial plate counts on selective media) every time unknown samples are tested. Enumeration of the organisms may not even be necessary. Indeed, in many applications, as when screening large numbers of samples, a simple positive or negative result may be all that is required. In either case, however, at least one positive control (spiked) sample should be included in the same milieu (e.g., enrichment culture) as the unknowns to ensure that the fluorescent reagents function in that particular setting. This unit deals specifically with detection of one pathogen-food combination, E. coli O157:H7 in ground beef, using a commercially available fluorescent antibody that reacts with the intact bacteria. The general approach and methods might well be applicable to other pathogens and foods, provided that appropriate reagents are available. CAUTION: E. coli O157:H7 is a biosafety level 2 pathogen and should be handled using the appropriate precautions, such as the use of protective gloves and/or frequent washing of hands. CALIBRATION AND STANDARDIZATION OF FLOW CYTOMETRY WITH CONVENTIONAL CULTURE METHODS TO DETERMINE LOWER LIMIT OF DETECTION AND TO DETECT PATHOGENIC E. COLI IN A FOOD MATRIX
BASIC PROTOCOL
Prior to testing of unknown food samples for the presence of E. coli O157:H7, it is important to determine the lowest concentration that can be detected under the conditions in a particular laboratory setting. If an estimation of actual numbers of organisms present is desired, then a comparison must be made between flow cytometry counts and the counts derived by conventional culture methods. The most effective approach to these standardization issues is to add dilutions of a selectable strain of the organism to enrichment cultures and to compare the linearity of plate counts of the pathogen to those derived by flow cytometry.
Microbiological Applications Contributed by Richard B. Raybourne Current Protocols in Cytometry (1999) 11.6.1-11.6.9 Copyright © 1999 by John Wiley & Sons, Inc.
11.6.1 Supplement 8
Materials Ground beef sample E. coli O157:H7, streptomycin resistant (ATCC #35150; see Support Protocol 3) Enrichment culture (EC) broth (Difco) with 20 µg/ml novobiocin (Difco) Luria-Bertani broth (Difco) Dulbecco’s PBS (Life Technologies) MacConkey sorbitol agar (Difco) containing dihydrostreptomycin or other appropriate antibiotic (Life Technologies) 10 mg/ml trypsin in PBS 0.5% (v/v) Triton X-100 FITC-labeled rabbit anti–E. coli O157:H7 (Kirkegaard & Perry) ∼106 bead/ml suspension of 10-µm polystyrene fluorescent beads (Beckman Coulter) Stomacher Model 400 laboratory blender (Seward) 12 × 75–mm polypropylene tubes (Falcon) 50°C water bath 10-ml plastic syringes with 5-µm syringe-tip filters (nylon Swinnex-type; Fisher) Flow cytometer with 480 to 490 nm excitation beam and log forward light scatter amplification, optimized for analysis of bacteria (see Support Protocol 1) Prepare enrichment cultures 1. Process 25 g ground beef in 225 ml EC broth/20 µg/ml novobiocin in Stomacher blender for 2 min at high speed. 2. Harvest slurry that passes through the mesh lining of the blender bag. Incubate this enrichment culture 18 to 24 hr at 35°C in four sterile 100-ml Erlenmeyer flasks. If the food in question is a liquid such as an unpasteurized juice, processing by blender is not necessary. These can be diluted directly into EC broth or centrifuged (15 min at 2500 × g) and the pellet resuspended in EC broth.
3. During the incubation in step 2, culture E. coli O157:H7 to exponential growth phase in LB broth at 37°C. 4. Using 12 × 75–mm polypropylene tubes for all dilutions, make six serial 10-fold dilutions of the E. coli O157:H7 from step 3 in PBS, and add 0.5 ml to each of six replicate 5-ml aliquots of the enrichment culture set up in step 2. Prepare one negative control aliquot containing no added E. coli O157:H7. This provides a positive control as well as a dilution curve for comparison bacterial counts by plating and flow cytometry, and should give a final concentration range from ∼102 to 107 E. coli O157:H7 /ml. Unknown or suspected contaminated samples should be handled in the same manner as negative controls. The distinction between “negative” and “unknown” samples is arbitrary since in any small retail sampling, it would be extremely unlikely, but possible, to find contaminated ground beef. If satisfactory correlation between flow cytometry counts and plate counts has already been established, or is not required, one or two of the mid-range dilutions can be used as positive controls for flow cytometry, and plating (step 5) need not be done.
Flow Cytometric Detection of Pathogenic E. coli in Food
5. Divide each of the enrichment culture/E. coli O157:H7 dilutions made in step 4, as well as the negative control (no added E. coli O157:H7), into two equal aliquots. Uniformly streak 100 µl/plate from one of the aliquots onto each of three replicate MacConkey sorbitol agar plates containing 500 µg/ml dihydrostreptomycin. Incubate the plates 18 to 24 hr at 37°C.
11.6.2 Supplement 8
Current Protocols in Cytometry
Any unknown or suspect samples would not be plated, since they would not be expected to contain streptomycin-resistant organisms. The other set of aliquots including unknown or suspect contaminated samples is processed for flow cytometry (step 6).
Process samples for flow cytometry 6. To 2 ml of the second aliquot from step 5, add 0.5 ml of 10 mg/ml trypsin in PBS and 2 ml of 0.5% Triton X-100. Incubate 10 min at 50°C. 7. Using a 10-ml plastic syringe, filter the sample through a 5-µm filter and collect the filtrate. 8. Centrifuge a 1-ml aliquot of the filtrate 3 min at 5000 × g, room temperature. Resuspend in 1 ml PBS. Repeat centrifugation and resuspension. 9. Dilute the washed pellet in 1 ml PBS and dilute to a concentration suitable for flow cytometry (1–2 × 106 cells/ml). Typically this dilution is 1:5 to 1:20. This is usually determined by trial and error.
Forward scatter (FS)
A
1000
2 100
10
1
1
0.1 0.1
1
10
100
1000
Side scatter (SS)
B
131
Cell count
98
65
32 A 0 0.1
1
10
100
1000
Green fluorescence (FITC) Figure 11.6.1 (A) Two-parameter histogram of a beef enrichment culture. Gate 1 is drawn around the bacterial population, Gate 2 is drawn around the 10-µm calibration beads. (B) FITC-fluorescence histogram derived from Gate 1. Region A is E. coli O157:H7.
Microbiological Applications
11.6.3 Current Protocols in Cytometry
Supplement 8
10. Add FITC-labeled anti–E. coli O157:H7 antibody to 1 ml of the diluted sample to give a final antibody dilution of 1:1000 to 1:2000. Incubate 20 min at 5°C. Dilutions can be made by adding 10 to 20 ìl of a 1:10 stock solution—if fresh stock is made for each experiment—or 1 to 2 ìl of the manufacturer’s concentrate to the 1-ml sample. In addition to the negative enrichment samples in step 4, a control containing only PBS and antibody should be included for the purpose of establishing the instrument background (background control).
11. Add 100 µl of a 106 bead/ml suspension of 10-µm fluorescent beads to each sample. Count the beads in each sample using a hemacytometer and fluorescence microscope. The beads are added as a reference standard for comparing flow cytometric counts of E. coli O157:H7 to plate counts or counts from some other parallel method. A simple positive or negative result and the lower limit of detection can be obtained without adding beads.
Acquire flow cytometry data 12. Using one of the mid-range-dilution E. coli O157:H7 samples, establish a flow rate of 200 to 400 particles per second. Acquire a log side light scatter (log SS; x axis) versus log forward scatter (log FS; y axis) histogram. The flow cytometer should be optimized for analysis of bacteria (see Support Protocol 1).
13. Establish a gating region around the bacterial population, which will appear just above the log FS threshold, and another gate around the beads, which should be well separated from the bacteria at the upper right corner of the histogram (see Fig. 11.6.1). 14. Use the gating region around the bacteria population to produce a one-parameter log green fluorescence (LGFL) histogram. This histogram should contain well defined positive and negative populations (see Fig. 11.6.1) which are used to derive percent positives and the number of E. coli O157:H7.
Number of cells detected (× 106/ml)
100
PLATE
10
FC
1
0.1
0.01
0.001 10 Flow Cytometric Detection of Pathogenic E. coli in Food
50
100 500 5,000 Cell dilution (reciprocal)
50,000
none
Figure 11.6.2 Comparison of the number of E. coli O157:H7 detected by plate counting (solid boxes) and flow cytometry (open boxes). The reciprocal dilution of E. coli O157:H7 is given on the x axis.
11.6.4 Supplement 8
Current Protocols in Cytometry
15. Use the gating region around the beads to determine the number of events for comparison with the number of E. coli O157:H7 in the same sample and the hemacytometer counts. 16. Collect data on at least 20,000 events in the bacteria region for all of the samples. Because no bacterial population will appear in the background control sample, data collection should be based on the approximate average run time of the enrichment samples.
17. Determine the lower limit of detection using the point at which the number of positive region events in the E. coli O157:H7–containing samples becomes indistinguishable from the negative control and the background control. 18. Calculate an estimate of the number of E. coli O157:H7 cells per ml for each sample as a function of the ratio of the number of events in the bead region versus the concentration determined by hemacytometer counts using the following equation. E. coli 0157:H7 cells/ ml =
number of beads/ ml × events in E. coli 0157:H7 region events in bead region
19. Following the 18- to 24-hr incubation, perform plate counts on the plates that were set up in step 5 using a microbiological colony counter. There should be good agreement between the replicates at each dilution of E. coli O157:H7. At higher concentrations of bacteria, the colonies may be too numerous (>500/plate) to count the entire plate. In this event, one quadrant of each plate can be counted using the colony counter grid.
20. Plot the data from flow cytometry and from plate counts to determine the relative and absolute accuracy of E. coli O157:H7 determination (see Fig. 11.6.2). DIRECT ENUMERATION OF E. COLI O157:H7 USING METERED SAMPLE FLOW
ALTERNATE PROTOCOL
Because most flow cytometers use differential pressure to introduce the sample, the precise volume and flow rate of the sample analyzed is not known. This necessitates using an internal calibration standard (e.g., fluorescent beads) for quantitative purposes. Other, more specialized flow cytometers use a microliter syringe driven by a precision motor to deliver a measured sample volume at a predetermined rate (Steen, 1990). This allows for direct determination of the number of events in any operator-defined region without the use of calibration beads. Additional Materials (also see Basic Protocol) Flow cytometer with metered sample delivery; e.g., Bryte HS (Bio-Rad) 1. Prepare enrichment cultures of E. coli O157:H7 (see Basic Protocol, steps 1 to 11). 2. Set sample flow rate at 2 to 10 ml per min. Allow counts/µl to stabilize for ∼30 sec. 3. Collect log green fluorescence data gated from log forward light scatter (log FS). 4. Directly obtain the number of fluorescence-positive organisms/ml by creating a region around the positive population. 5. Plot the data from flow cytometry and from plate counts to determine the relative and absolute accuracy of E. coli O157:H7 determination.
Microbiological Applications
11.6.5 Current Protocols in Cytometry
Supplement 8
SUPPORT PROTOCOL 1
OPTIMIZATION OF THE FLOW CYTOMETER FOR ANALYSIS OF BACTERIA With bacteria, most flow cytometers designed for analysis of eukaryotic cells will be operating at near the limits of their forward light scatter detection sensitivity. Hence, it is necessary to eliminate as much as possible any sources of light-scatter noise and to determine how the instrument settings can best be configured to achieve optimal signalto-noise ratios. Prior to attempting detection of E. coli O157:H7 in a microbiologically complex background, such as nonselective enrichment cultures from foods, it will be useful to make mixtures of pure culture bacteria and E. coli O157:H7 for use in optimizing the flow cytometer. Additional Materials (also see Basic Protocol) Standard laboratory strain (i.e., non-O157:H7) of E. coli, e.g., K12 (ATCC #10798) Sheath fluid: distilled water or PBS prefiltered through a 0.1- to 0.2-µm-pore-size filter Prepare control cultures 1. Grow overnight cultures of E. coli K12 and O157:H7 in Luria-Bertani broth. 2. Dilute E. coli K12 1:100 and 1:1000 in PBS to give ∼106 and 105 cells/ml; dilute E. coli O157:H7 1:100 in PBS. 3. Add 100 to 200 µl diluted E. coli O157:H7 to 1 ml of the 1:100 dilution of E. coli K12. Stain with anti–E. coli O157:H7 antibody (see Basic Protocol, step 10). Reserve the rest of the 1:100 and 1:1000 dilutions of K12 as unstained samples. 4. Set up flow cytometer using prefiltered distilled water or PBS for sheath fluid. Clean optical elements such as the flow cell and beam-shaping lenses to remove dust particles and salt deposits. 5. Use log forward light scatter (log FS) as the triggering signal with a blank (filtered PBS or distilled water) or no sample running. Increase forward light scatter sensitivity until noise begins to appear above threshold. Increase threshold or decrease sensitivity until the data rate is less than 10 events/sec. 6. Run the 1:100 and 1:1000 dilutions of K12 while the K12/O157:H7 mixture is staining. The rate of data acquisition should reflect the dilution factor. The purpose of this step is to ensure that the light-scatter signals are associated with bacteria, rather than with particulate debris or electronic noise.
7. Acquire data in a log SS versus log FS histogram. The bacteria should appear just above the forward light scatter threshold.
8. Run the K12/O157:H7 mixture. Acquire log green fluorescence data from the anti–E. coli O157:H7 sample using a gating region around the bacterial population on the log SS versus log FS histogram. 9. Adjust log green fluorescence sensitivity to achieve maximum separation of fluorescence-positive and fluorescence-negative bacteria.
Flow Cytometric Detection of Pathogenic E. coli in Food
11.6.6 Supplement 8
Current Protocols in Cytometry
SORTING OF FLUORESCENT E. COLI O157:H7 At times, it may be desirable or necessary to isolate the fluorescently stained bacteria for culture confirmation or other subsequent studies (also see UNIT 11.4). This can be done on flow cytometers with sorting capability. The basic sorting setup procedure is the same as for other cells, except that the increased light scatter sensitivity required for bacteria may result in light scatter noise from an LED used to “freeze” stream droplets during setup. To avoid this, the LED should be turned off after the setup, during the bacteria sort. After the sort, the correct positioning of the droplets should be verified. To confirm the accuracy and purity of the sort, different numbers of organisms (e.g., 100 and 500) are sorted and plated on selective and nonselective agar.
SUPPORT PROTOCOL 2
Additional Materials (also see Basic Protocol) Selective and nonselective agar: LB agar (Difco) with and without 500 µg/ml dihydrostreptomycin, respectively Flow cytometer capable of cell sorting Microplate deposition device: e.g., Autoclone (Beckman Coulter) Sterile 96-well plates (optional) 1. Prepare and stain cells (see Basic Protocol). 2. Create a sorting region around the FITC-positive cells. 3. Sort 100 and 500 bacteria into eight replicate tubes containing 0.5 ml sterile PBS or, if the sorter is equipped with a microplate deposition device, into eight replicate wells of a sterile 96-well plate, each containing 100 µl sterile PBS. 4. Plate the entire volume of four replicates from each set on antibiotic selective agar and four on nonselective agar. Incubate for 24 to 48 hr at 37°C, until countable colonies appear. Colony counts for the 500 sorted plates should be roughly five times that for 100 sorted plates, although less than 500 colonies may be present, since not all sorted bacteria will be culturable. Good agreement between the selective and nonselective agar counts indicates a high level of purity for the sorted E. coli O157:H7.
SELECTION OF STREPTOMYCIN-RESISTANT E. COLI O157:H7 Use of an antibiotic-resistant strain makes it possible to easily identify E. coli O157:H7 colonies in a microbiologically complex background such as nonselective enrichment cultures. Streptomycin resistance is usually easily selectable.
SUPPORT PROTOCOL 3
Additional Materials (also see Basic Protocol) Luria-Bertani (LB) broth (Difco) with 0, 50, and 500 µg/ml dihydrostreptomycin (Life Technologies) Luria-Bertani (LB) agar plates (Difco) with 50 and 500 µg/ml dihydrostreptomycin Glycerol, sterile 1. Culture E. coli O157:H7 in LB broth for 18 to 24 hr at 37°C. 2. Streak LB agar plates containing 50 µg/ml dihydrostreptomycin with 50 µl of the culture from step 1. Incubate plates 18 to 24 hr at 37°C. 3. Pick one discrete colony from the plate in step 2, and inoculate 5 ml LB broth containing 50 µg/ml dihydrostreptomycin. Incubate for 18 to 24 hr at 37°C. 4. Repeat steps 2 and 3 using media containing 500 µg/ml dihydrostreptomycin. 5. Freeze bacterial stocks by mixing 1 part broth culture and 1 part sterile glycerol, and placing in a −80°C freezer. Store at −80°C.
Microbiological Applications
11.6.7 Current Protocols in Cytometry
Supplement 8
COMMENTARY Background Information Escherichia coli O157:H7 contamination of foods, especially ground beef and unpasteurized apple juice and cider, has been responsible for several large outbreaks of food poisoning with significant health and economic consequences (CAST Task Force, 1994). Because of this impact, there has been an emphasis on development of rapid DNA-based or immunological methods for detection of this and other pathogens, to replace conventional culture methods. As with flow cytometry, most of these methods require some type of enrichment procedure to enhance the specific signal relative to background in the food matrix (Fung, 1994). Flow cytometry, however, offers some unique benefits in addition to rapid detection, such as multiparameter analysis, high throughput, and potential for automation. The latter two points are of particular importance, because pathogen contamination is often sporadic, so that testing of a large number of samples is required. Applications of flow cytometry to food microbiology have primarily used fluorescent antibody probes specific for the pathogen of interest (Raybourne, 1997; McClelland and Pinder, 1994a). A limiting factor in the adaptation of this approach has been the relative scarcity of such probes with defined specificity for intact pathogens. In some studies, two-parameter fluorescence involving the combined use of specific antibodies and a DNA-binding dye such as propidium iodide or ethidium bromide has been used to improve the sensitivity of detection (Donnelly and Baigent, 1986; McClelland and Pinder, 1994b). Other potential approaches that may prove useful in pathogen detection and analysis are the use of fluorescent oligonucleotide probes directed against 16S ribosomal RNA (Amann et al., 1990) and the use of dyes that can distinguish viable, nonviable, and metabolically active bacteria (UNIT 11.3). The latter point may be of particular relevance to food microbiology, where the presence of pathogenic organisms that have been injured, but not killed, by processing steps is a major concern.
Critical Parameters
Flow Cytometric Detection of Pathogenic E. coli in Food
As with other rapid methods of microbial pathogen detection, the use of an enrichment phase is necessary. Good microbiological technique is required for setting up these cultures. The type of enrichment used in the Basic Pro-
tocol is nonselective, which allows for growth of organisms that may be injured, and relies on the specificity of the antibody to distinguish the target pathogens from background. The lower limit of detection with this system is ∼104 organisms/ml. The precise length of the enrichment incubation time to reach this level is a function of the particular strain of the organism, the initial level of contamination, and the characteristics of the food matrix. In studies with apple juice, it was found that with a 10-hr enrichment time, a single E. coli O157:H7 cell per ml in the original sample could be detected by flow cytometry (Tortorello et al., 1998). The enrichment procedure used in this unit is specific for E. coli O157:H7. If it were adapted to other organisms of food-borne importance, optimal conditions for those organisms could be obtained from a food microbiology manual, such as the FDA Bacteriological Analytical Manual (FDA, 1995).
Troubleshooting Certain aspects of the instrument operation are significantly different from what is required for eukaryotic cell analysis, especially when using one of the standard commercially available instruments found in most laboratories. In working with small particles such as bacteria, it is essential to be able to distinguish between bacteria, light-scatter noise, electronic noise, and contaminating particulates in the sheath or in the sample diluents. It is important, therefore, that all reagents be filtered, and that instrument background be assessed using diluent blanks and sheath fluid (no sample) blanks.
Anticipated Results Generally good agreement can be expected between the flow cytometric method and other methods of measuring E. coli O157:H7 in food, including conventional plate counts and direct fluorescent microscopic counts. The lower limits of direct detection with flow cytometry, ∼104 cells/ml under these conditions, is somewhat higher than for plate counting or microscopic methods, but the time per sample is much shorter. Application of this protocol to other pathogens or foods would be dependent on the availability of antibodies that react with a surface antigen on the organism in question.
Time Considerations
Extract preparation requires ∼10 min/sample. Enrichment cultures require ∼18 hr. Sam-
11.6.8 Supplement 8
Current Protocols in Cytometry
ple filtration and staining require ∼45 min. Data acquisition requires 1 to 2 min.
Literature Cited
McClelland, R.G. and Pinder, A.C. 1994b. Detection of low levels of specific Salmonella species by fluorescent antibodies and flow cytometry. J. Appl. Bacteriol. 77:440-447.
Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R., and Stahl, D.A. 1990. Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56:1919-1925.
Raybourne, R.B. 1997. Flow cytometry in food microbiology: Detection of Escherichia coli O157:H7. In Food Microbiological Analysis New Technologies. (M.L. Tortorello and S.M. Gendel, eds.) pp. 57-68. Marcel Dekker, New York.
CAST Task Force. 1994. Foodborne Pathogens: Risks and Consequences. Council for Agricultural Science and Technology. Ames, Iowa.
Steen, H.B. 1990. Light scattering measurement in an arc lamp-based flow cytometer. Cytometry 11:223-230.
Donnelly, C.W. and Baigent, G.J. 1986. Method for flow cytometric detection of Listeria monocytogenes in milk. Appl. Environ. Microbiol. 52:689695.
Tortorello, M.L., Reineke, K.F., Stewart, D.S., and Raybourne, R.B. 1998. Comparison of methods for determining the presence of Escherichia coli O157:H7 in apple juice. J. Food Prot. 61:14251430.
Food and Drug Administration (FDA). 1995. Bacteriological Analytical Manual, 8th ed. AOAC International, Gaithersburg, Md. Fung, D.Y.C. 1994. Rapid methods and automation in food microbiology: A review. Food Rev. Int. 10:357-375. McClelland, R.G. and Pinder, A.C. 1994a. Detection of Salmonella typhimurium in dairy products with flow cytometry and monoclonal antibodies. Appl. Environ. Microbiol. 60:4255-4262.
Contributed by Richard B. Raybourne U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition Laurel, Maryland
Microbiological Applications
11.6.9 Current Protocols in Cytometry
Supplement 8
Mycobacterium tuberculosis Susceptibility Testing by Flow Cytometry
UNIT 11.7
Mycobacterium tuberculosis is a facultative intracellular pathogen that can infect every organ system in the body, especially the lungs. Infection is acquired by inhalation of droplet nuclei containing the tubercle bacillus that are dispersed into air when individuals with open-cavity tuberculosis cough, sneeze, talk, or sing. Approximately one-quarter of the world’s population is infected with M. tuberculosis, with 15 to 20 million people capable of transmitting the infectious agent. Each year there are 100 million new cases of M. tuberculosis in the world and three million deaths. In the United States, the incidence of tuberculosis steadily decreased from 1900 until 1985. From 1985 to 1992, however, the number of tuberculosis cases increased by 20%. Furthermore, several major outbreaks of multi-drug-resistant tuberculosis occurred. Subsequently, the public health departments of most states implemented control measures that reduced the number of cases of tuberculosis by 26% (CDC, 1998). To ensure the continuation of this decline and control the disease, the Centers for Disease Control and Prevention asserted that rapid and accurate susceptibility testing for M. tuberculosis was essential (CDC, 1996). Most common bacterial pathogens replicate within 30 min, so that susceptibility results are available within 24 hr of testing. Susceptibility testing of M. tuberculosis, however, is plagued by the replication rate of the tubercle bacillus, a slow-growing aerobic organism with a generation time of 15 to 24 hr. Consequently, a long incubation period is required to obtain results. Classically, susceptibility testing for M. tuberculosis is performed by growing the tubercle bacillus on medium for 2 or 3 weeks in the presence or absence of antimycobacterial agents. The number of organisms growing on the drug-containing medium as a percentage of the number of tubercle bacilli growing on the drug-free medium is then determined and reported. This is called the proportion method. Subsequently, development of the rapid radiometric proportion method, BACTEC-460, drastically reduced the time for obtaining susceptibility results to 3 to 14 days (NCCLS, 1995; Siddiqi et al., 1985). ASSESSING SUSCEPTIBILITY OF M. TUBERCULOSIS BY FLOW CYTOMETRY
BASIC PROTOCOL
The use of flow cytometry greatly improves the quality of susceptibility testing and advances public health measures to prevent tuberculosis. Susceptibility testing of M. tuberculosis (Norden et al., 1995; Kirk et al., 1998; Moore et al., 1999; Vena et al., 2000) can be accomplished within 24 hr after the mycobacteria are incubated with antimycobacterial agents. The method is founded on the ability of the tubercle bacillus to hydrolyze fluorescein diacetate (FDA) or 5-chloromethyl fluorescein diacetate to free fluorescein by nonspecific cellular esterases. Fluorescein accumulates in metabolically active mycobacterial cells and can then be easily detected by flow cytometry. By contrast, mycobacteria that are killed or inhibited by the action of antimycobacterial agents hydrolyze significantly less FDA and therefore demonstrate less intensity of fluorescence. NOTE: In the protocol that follows, knowledge of basic methods for bacterial culture and availability of the requisite reagents and equipment are assumed. CAUTION: Samples, especially the drug-free controls, may contain viable infectious M. tuberculosis. Procedure must be performed in a Biosafety Level 3 facility. Suitable personal protective equipment is required.
Microbiological Applications
Contributed by Ronald F. Schell, Dean Thomas Nordelli, David J. DeCoster, Scott M. Kirk, and Steven M. Callister
11.7.1
Current Protocols in Cytometry (2004) 11.7.1-11.7.8 Copyright © 2004 by John Wiley & Sons, Inc.
Supplement 27
Materials Anti-tuberculosis agents (ethambutol, isoniazid, and rifampin; see recipe) M. tuberculosis isolates grown on Lowenstein-Jensen agar, 7H10 agar, or other suitable solid medium McFarland 1.0 standards 7H9 broth (see Support Protocol; may also be purchased commercially) 500 ng/ml fluorescein diacetate (FDA; Molecular Probes) in phosphate-buffered saline, pH 7.4 (APPENDIX 2A) or 50 µg/ml 5-chloromethylfluorescein diacetate in dimethyl sulfoxide (DMSO; Sigma) 16 × 125–mm screw-cap tubes 3- or 4-mm glass beads (Fisher) 0.2-µm filters 50.0-ml polypropylene screw-cap tubes 2.0-ml screw-cap microtubes (Sarstedt) CAUTION: The following procedures must be carried out in a biosafety cabinet contained in a P-3 biosafety facility. In addition, cap, gown, gloves, and safety glasses must be worn. Prepare M. tuberculosis isolates for susceptibility testing 1. Scrape freshly grown colonies (<4 weeks old) from the surface of the solid medium; avoid including agar with the colonies. Transfer the bacterial mass to a sterile 16 × 125–mm screw-cap tube containing five glass beads and 5 ml 7H9 broth. Carefully emulsify the mass using a vortex mixer until all large particles are broken down. Let culture stand for 30 min without disturbance to allow large particles to settle to the bottom of the tube. Freshly grown broth cultures (7H9 broth) of M. tuberculosis may also be used.
2. Make a visual assessment of the suspension turbidity in comparison to that of a McFarland no. 1 standard. The suspensions must have a turbidity equivalent to the standard in order for the susceptibility tests to be performed: if the turbidity is greater than that of the standard, dilute the suspension with 7H9 broth; if less, add more bacterial mass. Broth cultures in 7H9 broth may require an additional 24 to 48 hr of incubation before the turbidity is equivalent to that of the standard.
Inoculate isolates 3. Transfer 0.9 ml of actively growing (log phase) M. tuberculosis isolate to each of five 2.0-ml screw-cap microtubes. 4. Inoculate one tube each with 0.1 ml of 50.0, 10.0, and 2.0 µg/ml isoniazid working dilutions; one tube with 0.1 ml of 50.0 µg/ml ethambutol; and one tube with 0.1 ml of 10.0 µg/ml rifampin. 5. Inoculate a fresh 2.0-ml tube (drug-free control) with 0.9 ml of actively growing M. tuberculosis and 0.1 ml 7H9 broth. 6. Incubate all tubes 24 hr in a 37°C, 5% CO2 incubator. 7. After incubation, set up duplicate assays by transferring two 0.2-ml aliquots of each assay suspension (treated and control) into separate sterile 2.0-ml screw-cap microtubes each containing 0.2 ml freshly prepared FDA or 0.02 ml 5-chloromethylfluorescein diacetate. Flow Cytometric M. tuberculosis Susceptibility Testing
Keep the remainder of each suspension for flow cytometry (step 10).
8. Incubate samples 30 min at 37°C, 5% CO2.
11.7.2 Supplement 27
Current Protocols in Cytometry
Perform flow cytometry 9. Run a sample of uninoculated 7H9 medium through the flow cytometer to determine the level of background particles. Exclude electronic and other background noise using forward-angle light scatter by drawing proper gates (Fig. 11.7.1A), and run sample at a moderate flow rate (2 to 5 µl per sec) for 20 sec. 7H9 medium, medium inoculated with unstained M. tuberculosis cells, and medium containing FDA-stained M. tuberculosis cells are analyzed before running the samples exposed to anti-tuberculosis agents.
10. Optional: Load unstained M. tuberculosis cells into the flow cytometer. Events outside the gate are M. tuberculosis cells (see Fig. 11.7.1B). Running the unstained cells allows the user to determine if the inoculum of M. tuberculosis was correct (i.e., had the proper turbidity). A similar scattergram should be seen each time the assays are run.
Forward-angle light scatter
A
B
Side-angle light scatter
Figure 11.7.1 Side-angle light scatter versus forward-angle light scatter of (A) 7H9 broth and (B) 7H9 broth with unstained M. tuberculosis cells.
Microbiological Applications
11.7.3 Current Protocols in Cytometry
Supplement 27
11. Load stained untreated M. tuberculosis cells (control) to determine the intensity of fluorescence. Collect green fluorescence (520 nm) using a 515- to 565-nm bandpass filter and acquiring 5000 events per sample. Generally, actively growing M. tuberculosis cells hydrolyze FDA to yield a mean channel fluorescence of ≥900 on a scale of 2048.
12. Run the remaining samples of M. tuberculosis cells treated with the antimycobacterial agents (unstained and stained). A flow rate of 2.0 to 5.0 ìl/sec, to obtain 5000 fluorescent events, is recommended. Results are obtained within 30 sec of aspiration of sample by the flow cytometer.
13. Autoclave all samples 40 min at 121°C and 15 psi before discarding. Determine flow cytometric susceptibility index 14. Divide the mean channel fluorescence values obtained from the population of FDA-stained M. tuberculosis cells in the presence or absence of antimycobacterial agents by the number of channels per log decade. The number of channels varies among flow cytometers.
15. Take the antilog of these values to obtain the relative linear fluorescence value for each sample analyzed. 16. Finally, divide the relative linear fluorescence value for each drug-containing sample by the relative fluorescence value of the drug-free control to obtain the index. An isolate with an index value of ≤0.75 is considered susceptible. The susceptibility index eliminates variability among isolates of M. tuberculosis in their ability to hydrolyze FDA in the presence or absence of antimycobacterial agents. In addition, all isolates tested can be compared on a “standardized” linear scale. SUPPORT PROTOCOL
PREPARATION OF STERILE 7H9 BROTH Materials Tween 80 (Sigma) 2.5% ammonium sulfate 2.5% glutamic acid 10% sodium citrate dihydrate 0.1% pyridoxine hydrochloride 0.2% biotin Disodium phosphate, anhydrous Monopotassium phosphate, anhydrous 10% ferric ammonium citrate 1% magnesium sulfate heptahydrate 0.1% calcium chloride dihydrate 0.1% zinc sulfate heptahydrate 0.1% copper sulfate pentahydrate 50% glucose (see recipe) 1000 µg/ml catalase (see recipe) 5% BSA fraction V (see recipe)
Flow Cytometric M. tuberculosis Susceptibility Testing
11.7.4 Supplement 27
Current Protocols in Cytometry
1. Place 900 ml distilled water in a 1500-ml flask, add the following, and dissolve by stirring: 0.5 g liquid Tween 80 20 ml 2.5% ammonium sulfate (0.5 g) 20 ml 2.5% glutamic acid (0.5 g) 1 ml 10% sodium citrate dihydrate (0.1 g) 1 ml 0.1% pyridoxine hydrochloride (1 mg) 250 µl 0.2% biotin (0.5 mg) 2.5 g anhydrous disodium phosphate 1.0 g anhydrous monopotassium phosphate 400 µl 10% ferric ammonium citrate (40 mg) 5 ml 1% magnesium sulfate heptahydrate (50 mg) 500 µl 0.1% calcium chloride dihydrate (0.5 mg) 1 ml 0.1% zinc sulfate heptahydrate (1 mg) 1 ml 0.1% copper sulfate pentahydrate (1 mg). 2. Adjust pH to 6.6 with 10% HCl, place broth in autoclave, and heat 20 min at 121°C and 15 psi. 3. Allow broth to cool to 50°C, and add the following: 4 ml 50% glucose 2 ml 1000 µg/ml catalase 100 ml 5% BSA fraction V. 4. Aseptically dispense 10-ml aliquots of broth into sterile 20-mm-diameter 50-ml screw-cap tubes. Store at 4°C. The medium should remain efficacious for 3 or 4 months. 7H9 broth is also commercially available.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Anti-tuberculosis agents Use the following formula to prepare stock solutions of the following three antimycobacterial agents, based on the assay potency (µg/mg) listed on each vial. All three agents can be obtained from commercial sources such as Sigma, the U.S. Pharmacopoeia Convention, and individual manufacturers. Store according to instructions included by the distributor or manufacturer. ZHLJKW PJ RI DJHQW QHHGHG
=
× DJHQW FRQFHQWUDWLRQ µJ PO QHHGHG DVVD\ SRWHQF\ RI DJHQW µJ PJ
YROXPH PO QHHGHG
Ethambutol: Dissolve in deionized distilled water to a concentration of 10,000 µg/ml. Store in 1-ml aliquots up to 3 months at −70°C. Discard unused aliquots. Prepare 50.0 µg/ml working dilution fresh before use. Inoculate susceptibility tubes with 0.1 ml of 50.0 ìg/ml ethambutol working dilutions. Isoniazid: Dissolve in deionized distilled water to a concentration of 10,000 µg/ml. Store in 1-ml aliquots up to 3 months at −70°C. Discard unused aliquots. Prepare 50.0, 10.0, and 2.0 µg/ml working dilutions fresh before use. Inoculate susceptibility tubes with 0.1 ml each of the three isoniazid working dilution.
Microbiological Applications
11.7.5 Current Protocols in Cytometry
Supplement 27
Rifampin: Dissolve in DMSO to a concentration of 10,000 µg/ml. Store in 1-ml aliquots up to 3 months at −70°C. Discard unused aliquots. Prepare 10.0 µg/ml working dilution fresh before use. Inoculate susceptibility tubes with 0.1 ml of 10.0 ìg/ml rifampin working dilution. Bovine serum albumin (BSA), 5% 5.0 g bovine serum albumin, fraction V 95 ml 0.85% NaCl Adjust pH to 6.8 with 50% NaOH and place in water bath at 56°C for 30 min. Sterilize by filtration through a 0.2-µm filter. Store at 4°C until added to medium (maximum shelf time 1 week). Catalase, 1000 ìg/ml 0.02 ml catalase 20 ml 0.85% NaCl Sterilize by filtration through 0.2-µm filter Store at 4°C until added to medium (maximum shelf time 2 weeks) Glucose, 50% (w/v) 50 g glucose 60 ml H2O 1 ml 10% citric acid Sterilize by autoclaving 15 min at 121°C Store at 4°C until added to medium (maximum shelf time 5 months) COMMENTARY Background Information
Flow Cytometric M. tuberculosis Susceptibility Testing
The most frequently used susceptibility tests for M. tuberculosis require 4 to 21 days to obtain results after initiation of testing procedures. Susceptibility testing of M. tuberculosis and other mycobacteria can be accomplished more rapidly by using a flow cytometer (Norden et al., 1995; Bownds et al., 1996; Kirk et al., 1998; Moore et al., 1999). Results of tests are available 24 hr after the tubercle bacilli are exposed to antituberculosis agents. In addition, multiplication of mycobacteria is not required to obtain susceptibility results. Occasionally, a borderline susceptibility-test index result (0.75 to 0.90) is obtained. If the flow cytometric portion of the test is repeated 48 hr after the M. tuberculosis organisms were exposed to antimycobacterial agents, either no significant changes will be detected (indicating that the organism is resistant) or the susceptibility index will have decreased to 0.74 or generally further (e.g., to 0.60). This latter result would strongly suggest that the organism is susceptible to the antituberculosis agent. Thirty-five clinical isolates obtained from the Centers for Disease Control and Prevention have been tested by both the flow cytometric and proportion methods (Kirk et al., 1998). The proportion method is considered the “gold
standard” laboratory test for M. tuberculosis susceptibility testing (NCCLS, 1995). Agreement between the methods was high: 95% when testing for susceptibility or resistance to isoniazid, 92% for ethambutol, and 83% for rifampin, respectively. Consequently, determination of the susceptibility or resistance of mycobacteria can be accomplished rapidly and with acceptable accuracy by flow cytometry.
Critical Parameters and Troubleshooting A plausible explanation for minor discrepancies is that the majority of the population of M. tuberculosis was not in the exponential growth phase when tested by flow cytometry. Hydrolysis of FDA is affected by the metabolic state or activity of the M. tuberculosis cells. This may be especially true when an inoculum is scraped from solid medium for susceptibility testing (the most common case). Generally, the central portion of the colony has the lowest viability. Therefore, efforts should be made to use M. tuberculosis cells grown in broth for ≤14 days to perform susceptibility testing. Even M. tuberculosis cells grown in broth can vary in their viability. Maximum viability can be obtained by inoculating 10 ml of 7H9 broth with M. tuberculosis cells contained in 70-ml tissue
11.7.6 Supplement 27
Current Protocols in Cytometry
culture flasks with canted 0.2-µm vented blueplug seal caps (Falcon 3109). Cultures should be gently shaken daily to disperse the tubercle bacilli and to maintain exposure to atmospheric air and 5% CO2. Generally, 5 to 9 days of incubation is required to obtain sufficient M. tuberculosis organisms (1 × 107/ml) to perform susceptibility testing. Biosafety is a serious concern. Viable mycobacterial cells with or without exposure to antituberculosis agents are being processed by the flow cytometer, which can generate droplet nuclei of ≤5 µm containing a single cell of M. tuberculosis. It takes only one tubercle bacillus to establish infection in humans. Therefore, safety is primary. This procedure can be safely performed only by public health laboratories or large reference laboratories having a Biosafety Level 3 tuberculosis laboratory and experience with Biosafety Level 3 precautions. To address this issue, the authors have developed a procedure that kills the mycobacterial cells after they have been stained and yet will not compromise the differential effect of FDA to distinguish between viable and nonviable cells (Moore et al., 1999). The high cost of the flow cytometer has also been a critical issue for universal implementation and acceptance of this rapid susceptibility test. However, considering the high costs of supplies for performing the radiometric proportion tests, including the cost of the instrument, the flow cytometer is less expensive, especially with a refurbished instrument. The reagents used for flow cytometry are also relatively inexpensive. Cost are restricted to the purchase of 7H9 broth, microtubes, FDA, and the antituberculosis agents. Technician times for performing the radiometric proportion and flow cytometric methods are similar. In conclusion, flow cytometry with FDA staining of M. tuberculosis cells is a simple and accurate method for obtaining susceptibility test results in 24 hr, and should greatly assist public health personnel in the control of tuberculosis.
Anticipated Results Isolates of M. tuberculosis susceptible to ethambutol, isoniazid, or rifampin will yield a susceptibility index of ≤0.75. Resistant isolates will yield a susceptibility index of ≥0.76; generally, resistant isolates have a susceptibility index of ≥0.95.
Time Considerations Results of the flow cytometric susceptibility tests are available after 24 hr of initiation (setup) of the testing procedure. Generally, 98% of the susceptibility tests can be processed by the flow cytometer 24 hr after incubating the antimycobacterial agent with an isolate of M. tuberculosis. The remaining tests (2%) may require an additional 24 hr of incubation before results are obtained by flow cytometry. Other methods of susceptibility testing require 4 to 21 days of incubation before susceptibility results are available.
Literature Cited Bownds, S.E., Kurzynski, T.A., Norden, M.A., Dufek, J.L., and Schell, R.F. 1996. Rapid susceptibility testing for non-tuberculosis mycobacteria using flow cytometry. J. Clin. Microbiol. 34:1386-1390. Centers for Disease Control and Prevention (CDC). 1998. Tuberculosis morbidity—United States, 1997. Morbid. Mortal. Weekly Rep. 47:253-257. CDC. 1996. Tuberculosis morbidity—United States, 1995. Morbid. Mortal. Weekly Rep. 45:365-370. Kirk, S.M., Schell, R.F., Moore, A.V., Callister, S.M., and Mazurek, G.H. 1998. Flow cytometric testing of susceptibilities of Mycobacterium tuberculosis isolates to ethambutol, isoniazid and rifampin in 24 hours. J. Clin. Microbiol. 36:1568-1573. Moore, A.V., Kirk, S.M., Callister, S.M., Mazurek, G.H., and Schell, R.F. 1999. Safe determination of susceptibility of Mycobacterium tuberculosis to antimicrobial agents by flow cytometry. J. Clin. Microbial. 37:479-483. National Committee for Clinical Laboratory Standards (NCCLS). 1995. Antimycobacterial Susceptibility Testing for Mycobacterium tuberculosis. Proposed Standard M24-T. NCCLS, Villanova, Pa. Norden, M.A., Kurzynski, T.A., Bownds, S.E., Callister, S.M., and Schell, R.F. 1995. Rapid susceptibility testing of Mycobacterium tuberculosis (H37Ra) by flow cytometry. J. Clin. Microbiol. 33:1231-1237. Siddiqi, S.H., Hawkins, J.E., and Laszio, A. 1985. Interlaboratory drug susceptibility testing of Mycobacterium tuberculosis by a radiometric procedure and two conventional methods. J. Clin. Microbiol. 22:919-923. Vena, R.M., Munson, E.L., DeCoster, D.J., Feh, D.B., Callister, S.M., and Schell, R.F. 2000. Flow cytometric testing of Mycobacterium avium to amikacin, ciproflaxin, clarithromycin and rifabutin in 24 hours. Clin. Microbiol. Infect. 6:365-375.
Microbiological Applications
11.7.7 Current Protocols in Cytometry
Supplement 27
Contributed by Ronald F. Schell, Dean T. Nardelli, and David J. DeCoster University of Wisconsin Madison, Wisconsin Scott M. Kirk Amersham Biosciences Sunnyvale, California Steven M. Callister Gundersen Lutheran Medical Center LaCrosse, Wisconsin
Flow Cytometric M. tuberculosis Susceptibility Testing
11.7.8 Supplement 27
Current Protocols in Cytometry
Antibiotic Susceptibility Testing by Flow Cytometry
UNIT 11.8
Bacterial susceptibility to antibiotics can be detected by monitoring changes in morphological and physiological characteristics of cells through the application of specific fluorescent probes and flow cytometry. This technique confers important advantages over more conventional methods. Flow cytometry permits rapid analysis of a given microbial population and can provide information relating to the heterogeneity of individual bacterial responses to antibiotics. Antibiotic-induced changes in bacterial membrane potential, cytoplasmic membrane integrity, or DNA content are the physiological characteristics most commonly probed. Bis(1,3-dibutylbarbituric acid) trimethine oxonol, or DiBAC4(3), has proven to be a robust membrane potential–sensitive dye. DiBAC4(3) is an anionic lipophilic ion that undergoes membrane potential–dependent distribution between the cytoplasm and external medium. This type of dye has an increased binding affinity for depolarized membranes. Propidium iodide (PI) is a popular dye for determining membrane integrity. In contrast to DiBAC4(3), PI is a small cationic molecule that is excluded from cells with intact cytoplasmic membranes. If membrane integrity is compromised, then PI will penetrate cells to bind tightly to nucleic acids. SYBR Green I was originally produced as an ultrasensitive gel stain for double-stranded DNA, but has been successfully adopted for determining DNA content by flow cytometry. Unlike PI, SYBR Green I is cell permeable. The Basic Protocol and Alternate Protocol 1 detail the individual use of DiBAC4(3) or PI, respectively, for testing bacterial susceptibility to various classes of antibiotic. Alternate Protocol 2 outlines the application of SYBR Green I for detecting sensitivity to DNA and protein synthesis inhibitors. All the protocols have been developed using either 0.2-µm-filtered nutrient broth or Iso-Sensitest broth as the bacterial growth medium. ANTIBIOTIC SUSCEPTIBILITY TESTING WITH DiBAC4(3) The use of membrane potential dyes as indicators of bacterial antibiotic susceptibility is based on the following hypothesis: an antibiotic, regardless of target site, will disrupt metabolic activity of susceptible cells, resulting in a change of membrane potential. DiBAC4(3) has been shown to be effective at detecting sensitivity of gram-positive and gram-negative organisms to a range of antimicrobial agents. This dye is excitable at 488 nm with a peak emission at 516 nm and is therefore suitable for use on the majority of commercially produced flow cytometers. No calibration is necessary, although an appropriate positive control is needed to confirm that the dye is responding to changes in membrane potential. This control is provided by treating bacterial cells with an ionophore such as gramicidin S, which results in the depolarization of the bacterial membrane.
BASIC PROTOCOL
Materials Bacterial cultures grown to log phase in 0.2-µm-filtered nutrient broth or Iso-Sensitest broth (Oxoid) 200 µg/ml gramicidin S solution (see recipe) 100 µg/ml DiBAC4(3) solution (see recipe) Sample tubes as required by the flow cytometer Flow cytometer with 488 nm excitation and band-pass filter centered at or around 515 nm Microbiological Applications Contributed by David J. Mason, Fiona C. Mortimer, and Vanya A. Gant Current Protocols in Cytometry (1999) 11.8.1-11.8.9 Copyright © 1999 by John Wiley & Sons, Inc.
11.8.1 Supplement 8
Prepare positive control for DiBAC4(3) uptake 1. Incubate a broth culture of the test bacterial strain to log phase. 2. Aseptically divide the culture into two subcultures. 3. Add 200 µg/ml gramicidin S solution to one of the subcultures at a final concentration of 20 µg/ml and continue to incubate both cultures for a further 60 min. 4. Remove a 200-µl aliquot from each culture. 5. Add 100 µg/ml DiBAC4(3) solution directly to both aliquots at a final concentration of 10 µg/ml. 6. Incubate samples 3 to 5 min at room temperature. 7. Set up the flow cytometer for excitation at 488 nm with a band-pass filter centered at or around 515 nm. Use forward-angle light scatter and side scatter to monitor changes in cell size or morphology. Collect dye emission using the green fluorescence detector gated by light-scatter parameters. 8. Analyze gramicidin-treated organisms to identify a region of interest on the green fluorescence histogram corresponding to DiBAC4(3)-stained cells. Organisms from the untreated culture will be nonfluorescent and should produce data outside this region.
Perform DiBAC4(3) antibiotic susceptibility test 9. Incubate a broth culture of the test bacterial strain to log phase. 10. Aseptically divide the culture into the necessary number of subcultures: one for each different antibiotic or antibiotic concentration to be used and one for a negative control. 11. Add antibiotics to subcultures at the desired concentrations and continue to incubate. 12. Remove 200-µl aliquots from each culture at 30-min intervals. Continue to remove samples up to 90 min. Following this incubation period, antibiotic-treated cultures which exhibit fluorescence profiles similar to that of the negative control may be considered nonsusceptible.
13. Add 100 µg/ml DiBAC4(3) solution directly to samples to give a final concentration of 10 µg/ml. 14. Incubate samples 3 to 5 min at room temperature and analyze by flow cytometry. Fluorescence data from organisms exhibiting antibiotic-induced uptake of DiBAC4(3) should fall within the region of interest determined in step 8. ALTERNATE PROTOCOL 1
Antibiotic Susceptibility Testing by Flow Cytometry
ANTIBIOTIC SUSCEPTIBILITY TESTING WITH PI Antibiotic-induced changes in bacterial membrane potential are not always detected. Propidium iodide provides an alternative approach to DiBAC4(3) for detecting susceptibility to bactericidal agents and can be used with gram-positive and gram-negative bacteria. When cell death occurs, membrane integrity is lost and the dye renders the cell fluorescent. This stain is also excitable at 488 nm and emits at 605 nm. DiBAC4(3) and PI can be used together in a sample at the final concentrations described, provided that electronic compensation for spectral overlap of the dyes is set in the instrument software. Additional Materials (also see Basic Protocol) 80% (v/v) ethanol, ice cold 0.2-µm-filtered broth 100 µg/ml PI solution (see recipe) Band-pass filter centered at or around 605 nm
11.8.2 Supplement 8
Current Protocols in Cytometry
Prepare positive control for PI uptake 1. Incubate a broth culture of the test bacterial strain to log phase. 2. Remove 1 ml of culture and centrifuge at 14,000 × g, room temperature, for 30 sec. Discard supernatant and resuspend pellet in 100 µl sterile distilled water. 3. Add 900 µl ice-cold 80% ethanol and incubate 60 min at 4°C. 4. Centrifuge and discard supernatant as before. 5. Wash pellet by resuspending in 1 ml of 0.2-µm-filtered broth, centrifuging and discarding supernatant as before, then resuspending in another 1 ml filtered broth. 6. Remove a 200-µl aliquot from both the ethanol-fixed suspension and the original culture, and add 100 µg/ml PI solution to both aliquots at a final concentration of 10 µg/ml. 7. Incubate at room temperature for ≥5 min. 8. Set up the flow cytometer for excitation at 488 nm with a band-pass filter centered at or around 605 nm. Use forward-angle light scatter and side scatter to monitor changes in cell size or morphology. Collect dye emission using the red fluorescence detector with logarithmic amplification and gated by light-scatter parameters. 9. Analyze fixed organisms to identify a region of interest on the red fluorescence histogram corresponding to PI-stained cells. Organisms from the unfixed sample will be nonfluorescent and should produce data outside this region.
Perform PI antibiotic susceptibility test 10. Set up control and antibiotic-treated samples as described for DiBAC4(3) procedure (see Basic Protocol, steps 9 to 12). 11. Add 100 µg/ml PI solution directly to the samples to give a final concentration of 10 µg/ml. 12. Incubate at room temperature for 5 min and analyze by flow cytometry. Fluorescence data from organisms exhibiting antibiotic-induced uptake of PI should fall within the region of interest determined in step 9.
ANTIBIOTIC SUSCEPTIBILITY TESTING WITH SYBR GREEN I Agents that inhibit protein or nucleic acid synthesis, such as chloramphenicol or rifampicin, may fail to alter membrane potential significantly or to compromise membrane integrity. Susceptibility to these antibiotics may be detected by monitoring the nucleic acid content of treated organisms. SYBR Green I can be used to detect differences in DNA content between cells of control and antibiotic-treated cultures. The dye is excitable at 488 nm with an emission at 521 nm and can be added directly to cells suspended in growth medium.
ALTERNATE PROTOCOL 2
Additional Materials (also see Basic Protocol) 100 U/ml SYBR Green I solution (see recipe) Prepare positive control for SYBR Green I uptake 1. Incubate a broth culture of the test bacterial strain to log phase. 2. Remove a 200-µl aliquot from the culture.
Microbiological Applications
11.8.3 Current Protocols in Cytometry
Supplement 8
3. Add 100 U/ml SYBR Green I solution directly to the sample to give a final concentration of 10 U/ml. Incubate sample 3 to 5 min at room temperature. 4. Set up flow cytometer for excitation at 488 nm as described for DiBAC4(3) procedure (see Basic Protocol, step 7). Use linear amplification for green fluorescence acquisition. 5. Analyze sample to verify successful staining and set instrument parameters. Observation of the sample using epifluorescence microscopy may also help to confirm bacterial staining.
Perform SYBR Green I antibiotic susceptibility test 6. Set up control and antibiotic-treated samples as described for DiBAC4(3) procedure (see Basic Protocol, steps 9 to 12). 7. Add 100 U/ml SYBR Green I solution directly to the samples to give a final concentration of 10 U/ml. 8. Incubate 3 min at room temperature and analyze on the flow cytometry. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
DiBAC4(3) (bis[1,3-dibutylbarbituric acid] trimethine oxonol) solution, 100 ìg/ml Dissolve DiBAC4(3) (mol. wt. 517; Molecular Probes) in absolute ethanol to give a stock solution of 1 mg/ml. Store in aliquots up to 12 months protected from light at −20°C. Before use, dilute aliquots with 70% (v/v) ethanol to obtain a working solution of 100 µg/ml. Discard working solution after use. Gramicidin S solution, 200 ìg/ml Dissolve gramicidin S (mol. wt. 1214.4; Sigma) in 100% ethanol to 1 mg/ml. Store in aliquots up to 6 months at 4°C. Before use, dilute aliquots with 70% (v/v) ethanol to obtain a working solution of 200 µg/ml. Discard working solution after use. Propidium iodide (PI) solution, 100 ìg/ml Dissolve PI (mol. wt. 668; Sigma) in sterile deionized water to give a stock solution of 1 mg/ml. Divide into aliquots and store up to 12 months protected from light at −20°C. Before use, dilute with sterile deionized water to obtain a working solution of 100 µg/ml. Store working solution up to 3 months protected from light at −20°C. SYBR Green I solution, 100 U/ml Dilute SYBR Green I concentrate stock (10,000 U/ml in DMSO; Molecular Probes) with sterile deionized water to give a 100 U/ml working solution. Store stock up to 12 months and working solution up to 3 months, both protected from light at −20°C. COMMENTARY Background Information
Antibiotic Susceptibility Testing by Flow Cytometry
General discussion Bacterial resistance to antibiotics is an escalating problem (Russell and Day, 1996). Selection of the correct antibiotic treatment regime has become increasingly important. With standard laboratory techniques, it can take up
to 4 days to determine the antibiotic sensitivity of a clinical isolate. Moreover, established methods such as the disc diffusion susceptibility test can provide information only on the bacterial population as a whole. The potential benefits offered by flow cytometry to bacterial susceptibility testing have been explored by a number of workers. Initial
11.8.4 Supplement 8
Current Protocols in Cytometry
by some classes of antibiotics (e.g., β-lactams or quinolones) can be detected by variations in light scatter characteristics (Gant et al., 1993; Mason et al., 1994, 1995). Although a useful indicator of bacterial susceptibility, morphology change is relevant only to a limited number of situations. Recently, the application of membrane potential–sensitive dyes to antimicrobial susceptibility testing has become a popular approach. Ordonez and Wehman (1993) successfully used dipentyloxacarbocyanine iodide to detect sensitivity of Staphylococcus aureus to penicillin. Oxacarbocyanine dyes are cationic lipophilic ions that fluoresce in viable bacterial cells but lose intensity if the magnitude of the cytoplasmic membrane potential is reduced. Use of these dyes with gram-negative organisms requires pretreatment with EDTA to permeabilize the outer membrane. Therefore, any result may depend on the susceptibility of organisms to the action of either EDTA, the antibiotic, or both. In addition, uptake of these dyes may affect the viability of cells regardless of antibiotic action. For all protocols described in this unit, sample preparation has been kept as simple and nonperturbing as possible. Fluorescent indicators are not included in the growth medium, but are added to a sample prior to analysis in order
Relate cell number (arbitrary units)
methods concentrated on the use of nucleic acid–binding dyes. A popular technique used by Boye et al. (1983) employed a DNA-specific dye combination of mithramycin and ethidium bromide, which they used to demonstrate bacterial susceptibility to agents such as chloramphenicol, erythromycin, doxycycline, streptomycin, and benzylpenicillin. Unlike the SYBR Green I protocol presented here, this method requires that cells be permeabilized, which was traditionally achieved by fixation and washing procedures. More recently, permeabilization of cells to mithramycin and ethidium bromide has been achieved by cold shock, eliminating the need for cell washing and considerably reducing sample preparation time (Walberg et al., 1997). Using propidium iodide as a probe of membrane integrity, Gant et al. (1993) demonstrated bacterial sensitivity to gentamicin, mecillinam, cefuroxime, and ampicillin. There are alternative probes of membrane integrity, such as SYTOX Green (Molecular Probes), which is purportedly more sensitive than PI and has proved effective at detecting antibiotic susceptibility (Roth et al., 1997). The dye is excitable at 488 nm and emits a green fluorescence. It is, however, more expensive than PI. Changes in cell morphology of gram-negative bacteria caused
0
50
100
150
200
250
DiBAC4 (3) fluorescence (channel numbers) Figure 11.8.1 A single-parameter histogram showing the different fluorescent responses of DiBAC4(3) to gramicidin-treated (broken line) and untreated (solid line) Staphylococcus aureus NCTC 6571.
Microbiological Applications
11.8.5 Current Protocols in Cytometry
Supplement 8
Relative cell number (arbitrary units)
100
101
102 PI fluorescence intensity
103
104
Figure 11.8.2 A single-parameter histogram showing the different fluorescent responses of PI to ethanol-fixed (broken line) and untreated (solid line) Staphylococcus aureus NCTC 6571.
to avoid possible dye-induced effects on bacterial cells or on antibiotic mechanism. An inherent advantage of dyes such as DiBAC4(3) and PI is that they are excluded from viable bacterial cells. This further minimizes the risk of the dyes themselves causing harm to the organisms. Choice of protocol The choice of protocol is determined by the biological system (see Critical Parameters) and antibiotic. Membrane-potential probes can offer high sensitivity to small changes in membrane potential and a rapid response.
Antibiotic Susceptibility Testing by Flow Cytometry
B DiBAC4 (3) fluorescence
Forward scatter
Forward scatter
A
DiBAC4(3) has been shown to work with a number of antibiotics such as methicillin, gentamicin, and ciprofloxacin (Mason et al., 1994; Jepras et al., 1997; Suller et al., 1997). Unfortunately, membrane energization is not the only determinant that influences the response of this type of probe. Other elements that should be considered include the effects of solvents, pH, and self-quenching. As an impermeant nucleic acid dye, propidium iodide is limited to detecting susceptibility to bactericidal antibiotics, for which it has proved successful. Monitoring of DNA content offers an alternative if antimicro-
DiBAC4 (3) fluorescence
Figure 11.8.3 Dual-parameter histograms of forward-angle light scatter versus DiBAC4(3) fluorescence from gentamicin-treated E. coli NCTC 10418 at 0 min (A) and 90 min (B). Gentamicin was used at 0.1 mg/liter. Data to the right of the heavy line represent cells rendered fluorescent by DiBAC4(3). The untreated control (not shown) gave a pattern similar to that seen in panel A.
11.8.6 Supplement 8
Current Protocols in Cytometry
bial action fails to significantly disrupt membrane potential or cytoplasmic membrane integrity. This method, however, can provide only indirect evidence of cell inhibition or cell death.
Critical Parameters All the protocols described here involve handling live bacterial cells; therefore, appropriate precautions for containment and disposal of waste should be taken.
Antibiotic action continues within samples after removal from the culture, so they should be stained and analyzed as quickly as possible. Moreover, if samples are left to stand for too long before analysis (>10 min), any cell perturbation subsequently detected may not be attributable solely to the antibiotic. This illustrates the importance of a control culture for monitoring possible cell perturbation induced by factors other than the presence of an antibiotic.
Relative cell number (arbitrary units)
A
B
0
50
100
150
200
250
SYBR Green I fluorescence (channel numbers)
Figure 11.8.4 Single-parameter histograms of SYBR Green I fluorescence from E. coli KL16. Untreated (A) and rifampicin-treated (B) after 90 min exposure to the antibiotic at 150 µg/ml.
Microbiological Applications
11.8.7 Current Protocols in Cytometry
Supplement 8
When applying DiBAC4(3) or PI to susceptibility tests with a new bacterial strain, the response of the dyes should be checked by preparing positive controls as described. Although the success of both dyes with strains of Escherichia coli and S. aureus has been documented, it is conceivable that responses may vary between species and even between strains of a single species. Flow cytometric testing of fastidious organisms such as Streptococcus pneumoniae, Haemophilus spp., Neisseria spp., and Campylobacter spp. is more difficult. These organisms are comparatively slow growing and require supplemented media and different gas mixtures. It is important to confirm that the chosen dye will behave identically in different medium conditions.
Troubleshooting One of the most important considerations when using either DiBAC4(3) or PI is establishing a positive control for dye uptake. If gramicidin S treatment yields little response from DiBAC4(3), incubation of bacterial cells with valinomycin (25 µg/ml final concentration) in the presence of 200 mM KCl provides an alternative method of membrane depolarization. However, valinomycin is extremely toxic and should be handled using appropriate precautions. Ethanol fixation is generally reliable as a method of membrane permeabilization. However, if this technique fails, heat treatment of bacterial cells for 30 min at 70°C should achieve the desired effect. Some bacterial species, such as Pseudomonas spp., may prove to be less permeable than others to SYBR Green I. If this is suspected, then the concentration of SYBR Green I can be increased 10-fold. Should a chosen dye fail to detect expected cell perturbation following antibiotic treatment, an alternative fluorochrome probing for a different physiological component could be used.
Anticipated Results
Antibiotic Susceptibility Testing by Flow Cytometry
Figures 11.8.1 and 11.8.2 show typical responses from DiBAC4(3) and PI following treatment of S. aureus cells with gramicidin S or fixation with alcohol, respectively. These fluorescence profiles may vary with bacterial species and external medium (see Critical Parameters). When antibiotic-exposed bacteria are incubated with DiBAC4(3) or PI, susceptibility is indicated by an increase in cell-associated fluorescence to a level similar to that seen in the respective positive controls (Fig. 11.8.3).
Conversely, fluorescence profiles obtained from antibiotic-susceptible cells stained with SYBR Green I will differ with antibiotic mechanism. Figure 11.8.4 shows fluorescence data obtained from E. coli KL16 treated with rifampicin (an inhibitor of RNA polymerase) for 150 min at the minimum inhibitory concentration and then stained with SYBR Green I. The fluorescence peaks from the treated sample represent cells with an integral number of chromosomes, indicating that the rifampicin has inhibited new rounds of DNA synthesis while allowing any ongoing chromosome replication to continue. In contrast, Walberg et al. (1996) used a DNA-specific dye combination of mithramycin and ethidium bromide to demonstrate increased fluorescence from E. coli undergoing filamentation following treatment with ampicillin (a β-lactam antibiotic).
Time Considerations Positive controls for DiBAC4(3) and PI should be prepared in tandem with the antibiotic-treated cultures. The initial growth of cultures to log phase takes ≥90 min. During this period, flow cytometric equipment can be set up and dye solutions can be prepared. Incubation with antibiotic(s) will require an additional 90 min.
Literature Cited Boye, E., Steen, H.B., and Skarstad, K. 1983. Flow cytometry of bacteria: A promising tool in experimental and clinical microbiology. J. Gen. Microbiol. 129:973-980. Gant, V.A., Warnes, G., Phillips, I., and Savidge, G.F. 1993. The application of flow cytometry to the study of bacterial responses to antibiotics. J. Med. Microbiol. 39:147-154. Jepras, R.I., Paul, F.E., Pearson, S.C., and Wilkinson, M.J. 1997. Rapid assessment of antibiotic effects on Escherichia coli by bis(1,3-dibutylbarbituric acid) trimethine oxonol and flow cytometry. Antimicrob. Agents Chemother. 41:2001-2005. Mason, D.J., Allman, R.J., Stark, M., and Lloyd, D. 1994. Rapid estimation of bacterial antibiotic susceptibility with flow cytometry. J. Microsc. 176:8-16. Mason, D.J., Power, E.G.M., Talsania, H., Phillips, I., and Gant, V.A. 1995. Antibacterial action of ciprofloxacin. Antimicrob. Agents Chemother. 39:2752:2758. Ordonez, J.V., and Wehman, N.M. 1993. Rapid flow cytometric antibiotic susceptibility assay for Staphylococcus aureus. Cytometry 14:811-818. Roth, B.L., Poot, M., Yue, S.T., and Millard, P.J. 1997. Bacterial viability and antibiotic susceptibility testing with SYTOX green nucleic acid stain. Appl. Environ. Microbiol. 63:2421-2431.
11.8.8 Supplement 8
Current Protocols in Cytometry
Russell, A.D. and Day, M.J. 1996. Antibiotic and biocide resistance in bacteria. Microbios 85:4565. Suller, M.T.E., Stark, J.M., and Lloyd, D. 1997. A flow cytometric study of antibiotic-induced damage and evaluation as a rapid antibiotic susceptibility test for methicillin-resistant Staphylococcus aureus. J. Antimicrob. Chemother. 40:7783. Walberg, M., Gaustad, P., and Steen, H.B. 1996. Rapid flow cytometric assessment of mecillinam and ampicillin bacterial susceptibility. J. Antimicrob. Chemother. 37:1063-1075.
Lloyd, D. (ed.) 1993. Flow Cytometry in Microbiology. Springer-Verlag, London. A useful compilation of work covering many aspects of the application of flow cytometry to microbiology, including antibiotic susceptibility testing. Pore, R.S. 1994. Antibiotic susceptibility testing by flow cytometry. J. Antimicrob. Chemother. 34:613-627. A thorough and critical review of the various techniques applied to flow cytometric susceptibility testing.
Walberg, M., Gaustad, P., and Steen, H.B. 1997. Rapid assessment of ceftazidime, ciprofloxacin, and gentamicin susceptibility in exponentiallygrowing E. coli cells by means of flow cytometry. Cytometry 27:169-178.
Contributed by David J. Mason University of Warwick Coventry, United Kingdom
Key References
Fiona C. Mortimer Formerly of King’s College London London, United Kingdom
Davey, H.M., and Kell, D.B. 1996. Flow cytometry and cell sorting of heterogeneous microbial populations: The importance of single cell analyses. Microbiol. Rev. 60:641-696. An informative review for anyone applying flow cytometry to microbiology.
Vanya A. Gant University College Hospital and Trust London, United Kingdom
Microbiological Applications
11.8.9 Current Protocols in Cytometry
Supplement 8
Determination of Bacterial Biomass from Flow Cytometric Measurements of Forward Light Scatter Intensity Biomass measurements in microbiology are an integral part of determining growth rates and cell yields, and of describing the kinetics that relate rates of growth and nutrient accumulation to substrate concentration. The biomass of bacteria analyzed by flow cytometry is determined from the relationship between forward light scatter intensity and cellular dry mass according to Rayleigh-Gans theory. The forward light scatter intensity of bacteria is a function of cellular dry mass. Production of a standard curve involves consideration of the influences of size, axial ratio, and orientation of cells in the flow stream; aldehyde preservatives; and excitation wavelength on the shape of the curve. It also involves selection of an appropriate standard for calibration. Samples are analyzed according to the intensity of forward light scatter and fluorescence from a DNA-bound probe. Orthogonal light scatter intensity of an internal standard at known concentration is used for ratiometric determination of population density.
UNIT 11.9
BASIC PROTOCOL
This protocol measures the forward light scatter (FS) intensity of formaldehyde-preserved, DAPI-stained bacteria resolved by flow cytometry, and determines cellular dry mass from a standard curve computed with an algorithm based on Rayleigh-Gans theory and calibrated with an appropriate bacterial standard of known cell mass. Rod-shaped bacteria are assumed to be aligned with the flow stream, perpendicular to the laser beam. Total biomass is calculated from mean cell mass and population density, the latter determined ratiometrically from the event frequency of bacteria and internal standard microspheres. For calibration of a standard curve, the relationship between FS intensity and cellular dry mass for bacteria of a particular shape can be characterized by executing the algorithm in Table 11.9.1, plotting the theoretical values and formulating the resulting curve (Fig. 11.9.2A), and calibrating the curve (Fig. 11.9.2B) with a suitable standard to relate measured values for FS to theoretical values. The algorithm requires input for the axial ratio (length/width) of the organisms analyzed, the excitation wavelength used, and the inclusive angle of FS detection. Materials Cell suspension at ≤107 cells/ml Formaldehyde (formalin, supplied as 37% formaldehyde), filtered (0.2-µm pore) just before use DAPI/Triton X-100 staining solution (see recipe) 0.96-µm-diameter fluorescent microspheres (Polysciences) Internal standard microsphere mixture: 0.6- and 0.9-µm fluorescent microspheres (Fluoresbrite Plain Microspheres; Polysciences) in filtered water, with 0.9-µm spheres at a known concentration of 108/ml as determined by Coulter counter (APPENDIX 3A) 1⁄ dilution of internal standard microsphere mixture (0.9-µm spheres at 107/ml) 10 Basal medium (carbon-source free) with osmolarity similar to that of the bacterial sample, filtered (0.2-µm pore) Sheath fluid (water or basal medium), filtered (<0.1-µm pore; Millidisk cartridges, Millipore) Radiolabeled cells for calibration standard (Robertson et al., 1998) Microbiological Applications Contributed by B.R. Robertson and D.K. Button Current Protocols in Cytometry (1999) 11.9.1-11.9.10 Copyright © 1999 by John Wiley & Sons, Inc.
11.9.1 Supplement 9
BASIC software program (Table 11.9.1) Curve-fitting software (e.g., SigmaPlot; Jandel Scientific) Syringe filters, 0.2- and 1.0-µm pore Flow cytometry tubes 10°C incubator Flow cytometer with: UV excitation Flat-sided quartz flow cell Photomultiplier detector for forward light scatter (FS) with the beam blocker positioned to eliminate laser beam while maximizing FS collection at <20°, 3.5-decade dynamic range provided by a calibrated logarithmic amplifier or by a linear circuit 429-nm long-pass dichroic filter to separate UV light scatter from DAPI fluorescence 310- to 370-nm band-pass filters to isolate UV light scatter in both forward and orthogonal directions 450- to 490-nm band-pass filter for isolating DAPI fluorescence Data analysis software that gives values for FS intensity proportional to amplitude of the input signal Prepare instrument and sample 1. Mix a suspension of ≤107 cells/ml with 0.5% (w/v) formaldehyde and incubate at least overnight, refrigerated and in the dark. This step arrests cell growth and allows complete binding of the preservative to cell constituents. It is important that samples have consistent and complete binding before analysis, as formaldehyde contributes mass to treated organisms. Cells can be in growth medium or washed and resuspended in basal medium before fixation. Natural aquatic samples are fixed directly because populations are usually ≤106/ml. Cells can be counted using a hemacytometer or Coulter counter (APPENDIX 3A) or by epifluorescence microscopy.
2. Turn on the flow cytometer prior to analysis, allowing sufficient time (∼1 hr) for the laser output to stabilize. 3. Meanwhile, add 20 µl DAPI/Triton X-100 staining solution to an appropriate flow cytometry tube containing a 1-ml test sample of ≤1 × 106 bacteria/ml from step 1. Incubate 1 hr at 10°C in the dark. Samples should be at ≤106/ml for analysis to avoid coincidence problems. If dilution is necessary, use filtered basal medium.
4. With amplifiers in linear mode and processors set for pulse integration, align the instrument with 0.96-µm-diameter fluorescent microspheres to optimize signal for forward scatter (FS), DAPI fluorescence, and orthogonal light scatter (side scatter or SS). Other fluorescent microspheres of diameter ≤1 ìm can also be used. These particular spheres have been used because they have tight fluorescence and scatter.
5. Set amplifiers for logarithmic processing. 6. Add 10 µl undiluted internal standard microsphere mixture to the test sample so that the concentration of 0.9-µm spheres is 1 × 106/ml. Vortex the sample.
Bacterial Biomass from Forward Light Scatter Intensity
The concentration of internal standard in the test sample is ten times the concentration normally used in bacterial samples (see step 10) to make signal recognition easier at the low flow rates used for bacterial analyses.
7. Use the test sample to set photomultiplier gains and to equilibrate the sample line with stain. Acquire data for FS versus DAPI fluorescence intensity, triggering on the
11.9.2 Supplement 9
Current Protocols in Cytometry
fluorescence signal, and set gains so that the bacterial signal is on scale (Fig. 11.9.1A). Gate on the cluster of 0.9-µm spheres (used for population determination) to adjust the gain for SS so that it is low enough to detect only the spheres and not bacteria (Fig. 11.9.1B). Aquatic bacteria such as Marinobacter arcticus (Fig. 11.9.1) and Cycloclasticus oligotrophus (Robertson et al., 1998) give FS intensities less than the intensity of 0.6-ìm microspheres. The intensity of E. coli will be closer to that of the 0.9-ìm microspheres (Robertson et al., 1998).
Table 11.9.1 Algorithm for Determining the Relationship Between Forward Scatter Intensity and Cell Massa
Command
Function
Calculation of π and conversion of degrees to radians AR = 3 Axial ratiob LOWER = 0.25 Lower angle of light scattering UPPER = 179.75 Upper angle of light scattering DEL = 0.25 Angular interval LIMIT = 20 Angular limit of light collection by flow cytometryb LAMBDA = 0.36 Wavelength of argon laser light in µmb IR = 1.333 Index of refraction of the medium DEF OPEN ’TEXTQKPT’:CLS Text format:clear screen OPEN ’O’,1,"LS.DA Open data file FOR J = 1 TO 60 Range of sizes V = 0.0001*10^(J/10) Range of volumes A = (3*V/(4*PI*AR))^1/3 Calculation of radius from volume and axial ratio IC = 0:IT = 0 Set cumulative values to zero FOR THETA = LOWER TO UPPER STEP DEL Scan light scattering at all angles X = 4*PI*A*SIN (THETA*C/2)IR/LAMBDA Calculate the variable for spherical particles FOR DELTA =−90 TO 90 STEP 10 Integral over an annulus of the detector BETA = 90-THETA*SIN(DELTA*C) Calculate the variable for desired Y = X*SQR((SIN(BETA*C))^2 + (AR*COS(BETA*C))^2 orientation P = (3*(SIN(Y) − Y*COS(Y))/Y^3)^2 Calculate the ‘P-function’ for the ellipsoid I = P*(1 + (COS(THETA*C))^2) Calculate the intensity of light scattered to *SIN(THETA*C)*C*DEL a cone IT = IT + I Integrate total light scattered IF (THETA>0.5 AND THETA
aModified from Table 2 in Koch et al. (1996) and printed with permission from Elsevier Science Publishing. bInput specific to application.
Microbiological Applications
11.9.3 Current Protocols in Cytometry
Supplement 9
1
Relative frequency
A
3 2 1
N 8993
S 967
100
B 288
10
FS intensity
0.03 0.01
0.6 µm
1 1
10 100 1000 DAPI fluorescence intensity
Relative frequency
Relative frequency
B
1000
0.9 µm
0.003
Dry mass (pg/cell)
0.1
2
N 1560
1
10
100 1000 SS intensity
Dry mass (pg/cell)
C
Relative frequency
0.003
B #1 261 #2 462
1
Bacterial Biomass from Forward Light Scatter Intensity
0.01
10
0.03 N 7851 838
0.1
1
100 FS intensity
2
1000
Figure 11.9.1 Biomass of a typical population of marine isolate Marinobacter arcticus. (A) Bivariate histogram of forward scatter (FS) and DAPI fluorescence intensity of bacteria and internal standard spheres. Regions 1 and 2 represent subpopulations containing cells with one and two chromosome copies, respectively. Region 3 bounds the whole bacterial population. Left and upper panels give FS and fluorescence intensity profiles, respectively, for Region 3. S and B are the FS intensity of the 0.6-µm spheres and bacteria, respectively. N is the number of events acquired for any particular population (7851 for region 1, 838 for region 2, and 8993 for region 3). (B) Orthogonal scatter (SS) distribution of internal standard 0.9-µm spheres, where N = 1560 is equivalent to the set concentration of 1 × 105/ml. (C) Distributions of FS intensity for the two subpopulations in (A) are converted to profiles of dry mass according to Equation 11.9.1, with K from Figure 11.9.2B and If determined as in the Basic Protocol, steps 19 and 20.
11.9.4 Supplement 9
Current Protocols in Cytometry
8. Prepare the remaining bacterial samples (from step 1) at ≤1 × 106 cells/ml in filtered basal medium, and stain for 1 hr at 10°C in the dark, using appropriate time intervals (e.g., 10-min intervals for 5-min data acquisition). Filtration of natural aquatic samples (1.0-ìm pore) before staining removes excess debris and large organisms that can interfere with analysis.
9. Prepare a cell-free control for evaluating bacterial contamination of the medium, stain solution, and internal standard. 10. Just before analysis of each sample, add 10 µl diluted (1/10) internal standard microsphere mixture per 1-ml sample so that the concentration of 0.9-µm spheres is 1 × 105/ml. Acquire sample data 11. Collect sufficient sample data for statistical relevance and save in listmode files. For a sample of >10,000 bacteria, acquire data for 5 min at 35 particles/sec. Low flow rates are preferred to maintain constant illumination over the small sample stream when the diameter of the focused laser beam is narrow for optimizing bacterial resolution (Robertson and Button, 1989) and to minimize artifacts caused by intercellular stain.
12. Replay listmode files to obtain a bivariate histogram of FS intensity versus DAPI fluorescence intensity (Fig. 11.9.1A). Set region boundaries for the bacterial population(s) of interest and for the cluster of 0.6-µm spheres to obtain data for each group. Gate on the cluster of 0.9-µm spheres to produce a histogram of SS for determining event frequency (Fig. 11.9.1B). Determine calibration factor 13. Prepare a radiolabeled cell sample at 106 cells/ml as described above (steps 1, 3, and 10). Detailed methodology for producing a calibration standard is beyond the scope of this protocol. The calibration standard should be similar in shape and composition to bacteria in the samples, and should be treated (preserved, diluted, and analyzed) in the same manner.
14. Measure representative bacteria by microscopy to determine the axial ratio (length/width) needed to formulate the standard curve. Calibrating for axial ratio is usually sufficient. However, if the composition of the dry matter varies a lot, or if a value for wet mass has to be determined, it is best to calibrate for each species (see discussion in Robertson et al., 1998).
15. Collect flow data as described for the sample (see steps 11 and 12). Collect radioactivity data using a liquid scintillation counter, so that the dry mass per cell can be calculated for the standard. The specific procedure for measuring radioactivity will vary depending on the sample being analyzed (see step 18). Generally, a 1-ml sample should be measured and should contain sufficient radioactivity (e.g., 5,000 to 100,000 counts/ml).
16. Run the BASIC program in Table 11.9.1 with input for axial ratio, FS collection angle, and excitation wavelength (footnote b).
Microbiological Applications
11.9.5 Current Protocols in Cytometry
Supplement 9
B 15
0.6
10
0.6
5
0.0 0.0
0.3
0
Dry mass (pg/cell)
Particle volume (V, µm3)
A
0.4 C. oligotrophus AR = 3 λ = 360 nm K = 1.618 x 10–5
0.2 0.0
0
10 20 30 FS intensity (I )
40
0.0
0.1 0.2 0.3 FS intensity (I )
Figure 11.9.2 Standard curve for dry cell mass from forward scatter intensity. (A) Theoretical dependency of FS intensity (I) on particle volume (V). Individual values (open circles) computed with the BASIC program in Table 11.9.1 with input for an axial ratio (AR) of 3, a collection angle of 0.5° to 20°, and an excitation wavelength of 360 nm. The fitted curve is described by Equation 11.9.1, where Xdry ≈ V, KIf = I, a = 1.62 × 10−4, b = 0.0144, c = 0.480, and d = 0.274. (B) Calibration of the curve in (A) with formaldehyde-fixed oligobacterium Cycloclasticus oligotrophus (diamond). Xsdry = 0.022 pg/cell and If = 58.7, giving K = 1.618 ×10−5 from Equation 11.9.1 (Robertson et al., 1998).
17. Import data for volume (V) and FS intensity (I) from BASIC into software with curve-fitting capability. Substitute Xdry (also from BASIC) for V, and plot and formulate the theoretical curve (Fig. 11.9.2A) according to Xdry = e
a ln( KIf )3 + b ln( KIf )2 + c ln( KIf ) + d
Equation 11.9.1
Where K is the proportionality constant or calibration factor used to convert FS intensity measured by flow cytometry (If) to intensity expected from theory (KIf = I), and a, b, c, and d are constants determined by the curve-fitting program (Robertson et al., 1998). Dry mass (Xdry) is substituted for V, because Xsdry varies with V for organisms of similar composition, because cell density can vary among species, and because light scatter intensity is a function of dry mass (Koch et al., 1996). Xsdry is used in this step as a general term for dry mass and is calculated by the algorithm.
18. Determine the dry mass per cell of calibration standard (Xsdry) from the measured radioactivity. Xsdry has been determined for the radioactivity of the marine isolate Cycloclasticus oligotrophus maximally labeled with 14C (dpm 14C/cell) after extended growth on radiolabeled acetate with high, known specific activity (dpm 14/g C) used as the sole carbon source, and from the contribution of carbon to dry weight obtained by CHN analysis (g dry wt/g C), as shown in Equation 11.9.2 (Robertson et al., 1998). s Xdry =
Bacterial Biomass from Forward Light Scatter Intensity
dpm dpm
14 14
C / cell × g dry wt /g C C/g C
Equation 11.9.2
11.9.6 Supplement 9
Current Protocols in Cytometry
19. Substitute the mean Xsdry of the radiolabeled cells (step 18) and the mean If per cell into Equation 11.9.1 to determine K. Analyze sample data and calculate biomass 20. Using sample data from steps 11 and 12, find the mean FS intensity of the bacterial population of interest (B) and of the internal standard 0.6-µm spheres (S). Normalize the data (B/S) for comparison among samples and to account for day-to-day variation. For the whole bacterial population in Figure 11.9.1A, B = 288 and S = 967.
21. Determine If for the sample for a specified value of S. Calculations for cell mass are simplified if K is computed for a specific value of intensity of the internal standard spheres. In work with small aquatic bacteria, the authors specify an intensity of 950 for the spheres. For the whole bacterial population in Figure 11.9.1A, If = (B/S) × 950 = (288/967) × 950 = 283.
22. Calculate Xsdry for bacteria in any particular sample using Equation 11.9.1, with K as determined in step 19 and If as calculated in step 21. For the whole bacterial population in Figure 11.9.1A, If × K = 283 × 1.618 × 10−5 = 4.58 × 10−3, giving a mean value of 50.0 fg/cell for Xsdry.
23. Determine the bacterial population density (N) from the ratio of the number of events recorded for the bacterial population(s) of interest in the bivariate histogram (Fig. 11.9.1A, left panel) to the number of events for the 0.9-µm standard spheres in the gated histogram of SS intensity (Fig. 11.9.1C). For the total bacterial population in Figure 11.9.1A/B, N = (8993/1560) × 1 × 105/ml = 5.76 × 105 bacteria/ml.
24. Calculate bacterial biomass from N × Xsdry. For the total bacterial population in Figure 11.9.1A, biomass = (5.76 × 105 bacteria/ml) × 50 fg/cell = 28.8 ìg/liter.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
DAPI/Triton X-100 staining solution Prepare an aqueous stock solution of 0.5 mg/ml 4′,6-diamidino-2-phenylindole (DAPI; Sigma), and store up to 6 months at −20 °C in the dark. Prepare a 5% (v/v) aqueous solution of Triton X-100 and store indefinitely at −20°C. For staining solution, mix 50 µl DAPI stock solution in 1 ml Triton X-100 stock solution and pass through a 0.2-µm filter. Prepare immediately before use. COMMENTARY Background Information Values for biomass critical to many microbiological studies have traditionally been obtained for bacteria in bulk suspension by measurements of optical density based on the relationship between dry mass and FS intensity according to Rayleigh-Gans theory (Koch, 1961; Koch and Ehrenfeld, 1968). Both the theoretical basis for the use of FS intensity measurements to determine the mass of cells
analyzed by flow cytometry (Koch et al., 1996) and an application to dilute mixed cultures have been been described (Robertson et al., 1998). A method to resolve subpopulations in a mixture of bacteria stained with DAPI (Button and Robertson, 1993) in combination with the method presented here for the determination of biomass has been used to characterize bacterial populations in Lake Zürich (Button et al., 1996) and to help characterize the marine oligobac-
Microbiological Applications
11.9.7 Current Protocols in Cytometry
Supplement 9
A Particle volume (V, µm3)
1.4 1.2
AR = 1
1.0 AR = 3
0.8 0.6
0.4 0.2 0.0
B Particle volume (V, µm3)
1.4 1.2
360 nm
1.0 0.8 0.6
488 nm
0.4 0.2 0.0
C Particle volume (V, µm3)
1.4 1.2 5o
1.0 0.8
10o
0.6 0.4
20o
0.2 0.0 0.0
0.2
0.4
0.6
0.8
FS intensity (I )
Figure 11.9.3 Parameters affecting the shape of the standard curve. The theoretical relationship between FS intensity and particle size as determined using the algorithm in Table 11.9.1 with various input parameters. (A) Cell shape, as indicated by the comparison of curves obtained for cells with axial ratios (AR) of 1 and 3 (cocci and rods). (B) Excitation wavelength, shown for UV (360 nm) and 488 nm. (C) FS detection angle, as computed with a lower limit of 0.5° and upper limits as shown. Intensities for the 5° and 10° upper limits are normalized to those for 20° at 0.001 µm3 for easier comparison of curve shape.
terium C. oligotrophus, isolated from Resurrection Bay, Alaska, by extinction culture (Button et al., 1998).
Critical Parameters and Troubleshooting
Bacterial Biomass from Forward Light Scatter Intensity
This method’s lower limit is set mainly by the sensitivity of the instrument, while the upper limit is dependent upon the size, shape, and refractive index of the bacteria. For the Ortho Cytofluorograf instrument configured for maximal sensitivity (Robertson and Button,
1989), the range for bacteria with an axial ratio of three and a refractive index relative to water (m) of 1.03 is 0.005 to 1.2 pg dry mass/cell (Robertson et al., 1998). The Rayleigh-Gans approximation is applicable only to microorganisms with m < 1.05 (Heller et al., 1959; Koch, 1961); it is not appropriate for standard latex particles (m = 1.19; Robertson et al., 1998). It is important that the calibration standard be as similar in cell shape and sample treatment as possible to the unknown samples, as signifi-
11.9.8 Supplement 9
Current Protocols in Cytometry
Table 11.9.2
Biomass in a Sample of M. arcticus from Figure 11.9.1
Regiona
Cell mass (fg [dry]/cell)
Population (105 cells/ml)
Biomass (µg/liter)
1 2 3
47.7 65.1 50.0
5.03 0.54 5.76
24.0 3.5 28.8
aAs indicated in Figure 11.9.1A, regions 1 and 2 are subpopulations containing cells with
one or two genome copies, respectively, and region 3 includes the entire population.
cant error can be introduced if a standard curve is formulated for rod-shaped bacteria with specified axial ratio but cocci are analyzed (Fig. 11.9.3A), if media differ in osmolarity (which would affect relative refractive index), or if there is nonuniform treatment with (or absorption of) aldehyde preservatives, which add to the cell mass (Robertson et al., 1998). The importance of using correct values for excitation wavelength and forward light scatter detection angle in the algorithm employed to produce a standard curve is shown in Figures 11.9.3B and 11.9.3C by the differences in the shapes of the theoretical curves obtained for two common excitation wavelengths and various collection angles. It appears that very narrow ranges of collection should be accurately defined in the algorithm to eliminate significant error, as shown by the large difference between the shape of the curves with an upper limit of 5° and 10°. A standard curve can be verified by independent measurements (Robertson et al., 1998). There should be agreement between dry mass determined by flow cytometry and dry mass computed from cell volume and buoyant density according to relationships previously reported (Robertson et al., 1998).
Anticipated Results Typical results are shown in Table 11.9.2, which gives the biomass of various subpopulations of M. arcticus as analyzed by flow cytometry (Fig. 11.9.1). With appropriate software (e.g., SigmaPlot), FS distributions can be easily converted to dry mass profiles (Fig. 11.9.1A, left panel; Fig. 11.9.1C) with the calculations demonstrated above. With this method, biomass as low as 5 pg in a 1-ml sample has been determined for a resolved bacterial subpopulation of 1000 cells/ml in a mixture of three species (Robertson et al., 1998).
Time Considerations Most of the effort is expended in obtaining a standard curve, which involves formulating
and fitting the applicable theoretical curve and preparing a suitable calibration standard. This may require several days. If formaldehyde-preserved cells are used, treatment should be done a day in advance of analysis by flow cytometry to ensure complete fixation (mass accumulation; Robertson et al., 1998). Staining with DAPI at 10°C takes an hour and can commence midway during the hour allowed for laser stabilization. Instrument alignment and equilibration of sample lines with DAPI take ∼20 min, and data acquisition takes 5 min for samples with ∼106 cells/ml. Sample staining at 7- to 10-min intervals can be convenient. Data analysis should be <15 min/sample, providing that the calculations for obtaining dry mass from the standard curve have already been defined.
Literature Cited Button, D.K. and Robertson, B.R. 1993. Use of high-resolution flow cytometry to determine the activity and distribution of aquatic bacteria. In Handbook of Methods in Aquatic Microbial Ecology (P.F. Kemp, B.F. Sherr, E.B. Scherr, and J.J. Cole, eds.) pp. 163-173. Lewis Publishers, Ann Arbor, Mich. Button, D.K., Robertson, B.R., and Jüttner, F. 1996. Microflora of a subalpine lake: Bacterial populations, size, and DNA distributions, and their dependence on phosphate. FEMS Microbiol. Ecol. 21:87-101. Button, D.K., Robertson, B.R., Lepp, P.W., and Schmidt, T.M. 1998. A small, dilute, high-affinity, novel bacterium isolated by extinction culture and having kinetic constants compatible with growth at ambient concentrations of dissolved nutrients in seawater. Appl. Environ. Microbiol. 64:4467-4476. Heller, W., Nakagaki, M., and Wallach, M.L. 1959. Theoretical investigations on the light scattering of colloidal spheres. V. Forward scattering. J. Chem. Phys. 30:444-450. Koch, A.L. 1961. Some calculations on the turbidity of mitochondria and bacteria. Biochim. Biophys. Acta. 51:429-441. Koch, A.L. and E. Ehrenfeld. 1968. The size and shape of bacteria by light scattering measurements. Biochim. Biophys. Acta. 165:262-273.
Microbiological Applications
11.9.9 Current Protocols in Cytometry
Supplement 9
Koch, A.L., Robertson, B.R., and Button, D.K. 1996. Deduction of the cell volume and mass from forward scatter intensity of bacteria analyzed by flow cytometry. J. Microbiol. Methods 27:49-61. Robertson, B.R. and Button, D.K. 1989. Characterizing aquatic bacteria according to population, cell size and apparent DNA content by flow cytometry. Cytometry 10:70-76. Robertson, B.R., Button, D.K., and Koch, A.L. 1998. Determination of the biomasses of small bacteria at low concentrations in a mixture of species with forward light scatter measurements by flow cytometry. Appl. Environ. Microbiol. 64:3900-3909.
Contributed by B.R. Robertson and D.K. Button University of Alaska Fairbanks Fairbanks, Alaska
Bacterial Biomass from Forward Light Scatter Intensity
11.9.10 Supplement 9
Current Protocols in Cytometry
Flow Cytometry of Yeasts
UNIT 11.10
Flow cytometry has been used extensively to monitor many different properties of yeasts; the information thereby obtained enables determination of the population distribution with respect to that property. This unit provides various protocols for the study of yeasts. Basic Protocol 1, monitoring the cell cycle, is concerned with the events of the cell division cycle (e.g., growth and budding in Saccharomyces cerevisiae, growth and fission in Schizosaccharomyces pombe, the DNA replication cycle). Basic Protocol 2 deals with viability assessment. Yeast cell viability is most easily assessed by plasma membrane permeability or transmembrane electrochemical potential. Respiratory activity is evaluated in Basic Protocol 3, and in vivo β-galactoside activity in Basic Protocol 4. The power of those methods resides in the observation that, despite the increasing usefulness of digital imaging techniques, flow cytometry remains the method of choice for the resolution of population heterogeneities. NOTE: Expertise is assumed for flow cytometry techniques and for the culture and harvesting of yeast, the latter covered generally in Rose and Harrison (1969). MONITORING THE CELL DIVISION CYCLE OF YEASTS A precise duplication of nuclear DNA and a mechanism for the segregation of chromosomes to each of the daughter cells is essential for cell survival. The replication of nuclear DNA is restricted to a limited time interval (the S phase) within the total cell cycle time. Thus, organisms in G1 have exactly half the nuclear DNA of those in G2, the stage after replication. Determination of the frequency distribution of cellular DNA content of individual organisms within a population provides the most reliable measure of their cell cycle traverse.
BASIC PROTOCOL 1
Materials Washed yeast suspension 70% (v/v) ethanol Tris⋅Mg2+ buffer (see recipe) 0.5% (w/v) pepsin solution (see recipe) 10 µg/ml RNase A solution (see recipe) 50 µg/ml PI solution (see recipe) 1. Harvest washed yeast cells by centrifuging 2 min at 1000 × g, room temperature. 2. Resuspend in 70% ethanol at 1–5 × 106 organisms/ml and fix for ≥1 hr, but no longer than 24 hr. 3. Centrifuge again for 2 min at 1000 × g. 4. Wash sedimented organisms with 5 vol Tris⋅Mg2+ buffer and resediment. 5. Resuspend in 1 vol fresh 0.5% pepsin solution and incubate for 1 hr. 6. Centrifuge, wash with Tris⋅Mg2+ buffer, and resuspend in 1 vol of 10 µg/ml RNase A solution. Incubate at 30°C for 3 hr. 7. Collect cells by centrifugation, resuspend in 1 vol of 50 µg/ml PI solution, and incubate at room temperature for 75 min. 8. Analyze on a flow cytometer using excitation at 488 nm and collecting emission at wavelengths >620 nm. G1 cells (first peak) and G2 cells (second peak) are separated by a population of S-phase organisms. Sample data are shown in Figure 11.10.1. Microbiological Applications Contributed by David Lloyd Current Protocols in Cytometry (1999) 11.10.1-11.10.8 Copyright © 1999 by John Wiley & Sons, Inc.
11.10.1 Supplement 9
Number of cells
1000
500
0 0
10
20
30
40
50
60
70
80
Relative fluorescence
Figure 11.10.1 DNA histogram for Saccharomyces cerevisiae (n = 70,000) after propidium iodide staining, indicating 41,000 G1 cells (first peak) and 22,000 G2 cells (second peak). The third small peak consists of aggregated cells. The S-phase region lies between the two major peaks (Hutter and Eipel, 1978).
BASIC PROTOCOL 2
DETERMINING THE VIABILITY OF YEAST Loss of the plasma membrane electrochemical potential provides an excellent indication of cell death in a population of yeasts (Dinsdale et al., 1995, 1999; Seward et al., 1996; Willetts et al., 1997). Flow cytometry provides the most satisfactory method for enumeration of dead cells after brief exposure of the organisms to the anionic voltage-sensitive oxonol dye bis-(1,3-dibutylbarbituric acid) trimethine oxonol, often referred to as DiBAC4(3). Viable cells exclude the fluorophore so that only dead cells become fluorescent. Materials Yeast suspension Tris⋅Mg2+ buffer (see recipe) 1 µg/ml DiBAC4(3) solution (see recipe) 1. Dilute 1 ml yeast suspension in 20 mM Tris⋅Mg2+ buffer to give 105 organisms/ml. 2. Add 1 µg/ml DiBAC4(3) solution to give a final concentration of 0.1 µg/ml. 3. Incubate 5 min at room temperature. 4. Analyze using 488 nm excitation and measuring emission at 510 nm. Non-viable cells are fluorescent. Sample results are shown in Figure 11.10.2.
Flow Cytometry of Yeasts
11.10.2 Supplement 9
Current Protocols in Cytometry
A
10 20 FALS 30 40 50 60
10
fluorescence 20 30 40 50
B 60
10 20 FALS 30 40 50 60
C
10 20 FALS 30 40 50 60
10
fluorescence
20
30
40
50
fluorescence 20 30 40 50
60
D
10 20 FALS 30 40 50 60
10
fluorescence 20 30 40 50
60
E
10 20 FALS 30 40 50 60
10
10
fluorescence 20 30 40 50
60
F 60
10 20 FALS 30 40 50 60
10
fluorescence 20 30 40 50
60
Figure 11.10.2 Flow cytometry of Saccharomyces cerevisiae after DiBAC4(3)-based viability assessment of cells grown in the presence or absence of 10% ethanol. Organisms from control cultures (without added ethanol) are shown after (A) 24 hr or (B) 6 days. Results from heat-killed yeasts are shown in (C). Yeasts from cultures grown with 10% (v/v) added ethanol are shown after (D) 24 hr, (E) 3 days, and (F) 6 days. Although vitality decreases in the cultures containing ethanol, viability (i.e., ability to exclude the fluorophore) is hardly altered. FALS, forward-angle light scatter.
Microbiological Applications
11.10.3 Current Protocols in Cytometry
Supplement 9
BASIC PROTOCOL 3
EVALUATING THE MITOCHONDRIAL (RESPIRATORY) FUNCTION OF YEAST The respiratory activity of yeast can be evaluated indirectly by flow cytometric measurement of rhodamine 123 uptake (Porro et al., 1994; Lloyd et al., 1996). In anaerobic fermentations (e.g., in brewing or cider-making), the capacity for uptake of this cationic dye is lost early in the process. Materials Washed yeast suspension at 1–5 × 106 organisms/ml 10 mM glucose 1 µg/ml rhodamine 123 solution (see recipe) 1. Harvest 5 ml washed yeast suspension by centrifuging 2 min at 1000 × g, room temperature. 2. Add 1 ml of 10 mM glucose and 5 ml of 1 µg/ml rhodamine 123 solution, and mix thoroughly. 3. Allow 5 min for dye uptake at room temperature. 4. Analyze using 488 or 546 nm excitation (Ar-ion or Hg-arc, respectively) and collecting fluorescence emission at 580 nm. Mitochondrial respiratory activity generates a transmembrane electrochemical potential; this drives dye accumulation in the organelles.
BASIC PROTOCOL 4
ASSAYING â-GALACTOSIDASE ACTIVITY IN VIVO A flow cytometric assay for β-galactosidase in S. cerevisiae uses the fluorogenic substrate resorufin β-D-galactopyranoside (Wittrup and Bailey, 1988). After the organisms have been permeabilized with Triton X-100, a steady state is established between product formation and leakage. Bovine serum albumin in the extracellular fluid is used to prevent reuptake by other cells. The reaction is measured at low temperature. Materials 2% (v/v) Triton X-100 Substrate solution: 2 mg/ml resorufin β-D-galactopyranoside (Molecular Probes) in dimethyl sulfoxide PBS (APPENDIX 2A), prechilled to 0°C 2% (w/v) bovine serum albumin (BSA; Sigma) in PBS (store and use at 0°C) Yeast suspension (106 cells/ml) washed twice in PBS after centrifugation from growth medium and kept at 4°C 1. Place the following (in order) into a 1.5-ml microcentrifuge tube kept on ice: 40 µl 2% Triton X-100 40 µl substrate solution 0.55 ml ice-cold PBS 0.4 ml ice-cold 2% BSA 50 µl washed yeast suspension. 2. Shake well and place in an ethanol/ice bath in the sample chamber of a flow cytometer. 3. Immediately initiate flow and measure distributions after 5 min using 568 nm excitation (Kr-ion laser) and collecting emission at >590 nm.
Flow Cytometry of Yeasts
β-Galactosidase-positive organisms show fluorescence, and fluorescence intensity is proportional to enzyme activity.
11.10.4 Supplement 9
Current Protocols in Cytometry
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. Keep fluorophores as frozen solutions at −18°C in the dark. For solvents, use absolute ethanol or dimethyl sulfoxide. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
DiBAC4(3) solution, 1 ìg/ml Stock solution: Dissolve 1 mg DiBAC4(3) (Sigma) in 1 ml absolute ethanol and store up to 1 year at −18°C in the dark. Working solution: Dilute 1 µl to 1 ml in water (final 1 µg/ml) before use. Pepsin solution, 0.5% (w/v) 5.5 ml 1 N HCl 94.5 ml H2O 0.5 g pepsin Store in 1-ml aliquots for up to 1 year at −18°C Propidium iodide (PI) solution, 50 ìg/ml Stock solution: Dissolve 1 mg PI in 1 ml water and store in 50-µl aliquots for up to 1 month at −18°C in the dark. Working solution: Dilute 50 µl to 1 ml in water (final 50 µg/ml) before use. Rhodamine 123 solution, 1 ìg/ml Stock solution: Dissolve 1 mg rhodamine 123 (Sigma) in 1 ml water and store up to 1 month at −18°C in the dark. Working solution: Dilute 1 µl of to 1 ml in water (final 1 µg/ml) before use. RNase A solution, 10 ìg/ml Stock solution: Dissolve 5 mg bovine pancreas RNase A (DNase-free, Sigma) in 5 ml Tris⋅Mg2+ buffer (see recipe). Store in 0.5-ml aliquots for up to 1 week at −18°C. Working solution: Dilute 10 µl to 1 ml in Tris⋅Mg2+ buffer (final 10 µg/ml) before use. Tris⋅Mg2+ buffer 1.21 g/liter Tris base (mol. wt. 121; final 10 mM) 1 g/liter MgCl2 (5 mM) Adjust pH to 7.0 using 1 M HCl Store up to 1 month at 4°C COMMENTARY Background Information DNA-reacting fluorophores Basic Protocol 1 is suitable for use in studies of the effects of inhibitors on the DNA replication cycle of yeast (Eilam and Chernichovsky, 1988), cell cycle mutants of Schiz. pombe (Costello et al., 1986), and signaling pathway mutants (Hayashi et al., 1998). Propidium iodide (PI) is a nonspecific fluorochrome, reacting with all double-stranded nucleic acids to give complexes excitable at 488 nm and emitting above 600 nm. It has been used to measure total cell RNA, for example in Schiz. pombe (Agar and Bailey, 1982), where the RNA/DNA ratio is 100/1, and where most of the RNA is
double-stranded ribosomal RNA. After RNase treatment, PI is the most commonly employed fluorophore for cell-cycle analysis (nuclear DNA replication, see protocol above). Chromomycin A3 (Sigma) binds preferentially to GC base pairs and when excited at 457 nm emits fluorescence above 590 nm. Both Hoechst 33258 (Calbiochem) and 4′,6-diamidino-2-phenylindole (DAPI; Sigma) have a high affinity for AT base pairs. They react preferentially with mitochondrial DNA, and both fluorophores require UV excitation (329 to 425 nm; Sazer and Sherwood, 1990). DAPI has been used extensively to stain mitochondrial DNA for fluorescence microscopy (Williamson and Fennell, 1979), although in living cells
Microbiological Applications
11.10.5 Current Protocols in Cytometry
Supplement 9
there may also be uptake of this dye into mitochondria by virtue of their transmembrane electrochemical potential (Pringle et al., 1989). Flow cytometry of arrested cell-division-cycle mutants of Schiz. pombe shows that care must be exercised in assigning DNA-reacting fluorophores to mitochondrial DNA (Carlson et al., 1997). Exposure of cells to Triton X-100 and hypotonic conditions after cell wall digestion was shown to remove cytoplasmic material and enable nuclear DNA content to be evaluated using either mithramycin plus ethidium bromide or Hoechst 33258. The former fluorophore mixture, used extensively for bacteria (Skarstad et al., 1985), gave the more reproducible results. Mithramycin (excited at around 440 nm using a line of the Hg arc) has a preference for GC base pairs in DNA, and effectively transfers excitation energy only to adjacent DNA-associated ethidium bromide. Similar results were also obtained with S. cerevisiae. The fluorescein diacetate/propidium iodide method also provides a technique for estimating viability of yeast (Willetts et al., 1997); other methods use ChemChrome Y (Chemunex; Brailsford and Gatley, 1993; Carter et al., 1993; Bruetschy et al., 1994; Deere et al., 1998) and the FungoLight kit (Molecular Probes; Lloyd and Hayes, 1995; Deere et al., 1998). In the last cited study, it was shown that propidium iodide exclusion and ChemChrome Y uptake gave similar results, but that for rehydrated dried yeast, DiBAC4(3) stained a higher proportion of organisms (107%) than did propidium iodide. These data compared well with those obtained by the traditional colony-counting method, but the great advantage of the flow cytometric assessment was that it took 20 min, rather than 36 hr. FungoLight, although useful for light microscopy, was not suitable for flow cytometric analyses. Cell sorting showed that viability was not affected by the dyes themselves or by the sorting procedure.
Flow Cytometry of Yeasts
Membrane potential–sensitive dyes Membrane potential–sensitive dyes, such as DiBAC4(3), can be used to assess the energy status of yeast mitochondria in vivo. Rhodamine 123, tetramethyl rhodamine ethyl ester (TMRE), 3,3′-dipropylthiacarbocyanine iodide [DiSC3(3)], and 3,3′-dipropyloxacarbocyanine iodide [DiOC3(3)] have also been used (Denksteinova et al., 1996; Plásek and Sigler, 1996). The production of fluorescent formazan from 5-cyano-2,3-ditolyl tetrazolium chloride,
a method used for bacteria (López-Amorós et al., 1995), requires prior cell permeabilization as yeasts are not permeable to this redox indicator. Other assays Intracellular flavins and NAD(P)H redox state. A dual excitation of flavins and NAD(P)H is informative about the redox state of organisms and cells. For rat liver cells, Theorell (1983) described the simultaneous flow cytometric measurement of blue and green fluorescence using excitations at 351 to 363 nm (argon UV) and at 488 nm (argon), respectively. This method can also be applied to yeast. Cellular protein. The analysis of protein distributions provides an estimate of several cell cycle parameters for steady-state and perturbed cultures (Alberghina and Porro, 1993). An excellent summary of the use of flow cytometry in investigations of the growth of yeast is given by Davey and Kell (1996). Sterol content of yeast. Fluorescein isothiocyanate–labeled nystatin A1, a macrolide reacting specifically with 3β-hydroxy sterols (80% ergosterol), has been used to measure the sterol distribution within the yeast population (Müller et al., 1992). Hydrogen peroxide production. The probe 2′,7′-dichlorofluorescin diacetate is concentrated within cells as it is hydrolyzed by esterases. Reaction with intracellular peroxides gives a product that emits intense green fluorescence when excited at 488 nm (Yurkow and McKenzie, 1993). Simultaneous viability analysis, using the propidium iodide method, indicated that aeration after a period of anoxia gives peroxide generation leading to cell damage and death. Antibiotic sensitivity testing. A flow cytometric method for detection of the sensitivity of Candida species to amphotericin has been developed (Ordónez and Wehman, 1995). The cationic membrane potential–sensitive dye 3,3′- dip en tylo xacarbocyanine iodide [DiOC5(3)] is accumulated by cells growing normally. Incubation in the presence of antibiotic for 30 min reduced dye uptake of sensitive strains in a dose-dependent manner, whereas that of resistant strains was unaffected. Propidium iodide, ChemChrome Y, and DiBAC4(3) also show promise in the flow cytometric assessment of drug sensitivity (Carter et al., 1993). Selection of novel yeast hybrids. Flow cytometry has applications in the development of improved industrial yeast strains (e.g., the rapid
11.10.6 Supplement 9
Current Protocols in Cytometry
isolation of rare mating hybrids; Bell et al., 1998). Yeast as a spoilage organism. ChemChrome Y uptake has been used as a rapid and sensitive marker to detect yeasts as contaminants of soft drinks (Pettipher, 1991; Brailsford and Gatley, 1993).
Critical Parameters For live-cell assays, cells should be treated with respect (i.e., minimally perturbing conditions employed for centrifugation). All procedures should be carried out as quickly as possible and without change of temperature.
Anticipated Results The most valuable characteristic of flow cytometric data is the ability to provide information on the distribution of activities or constituent amounts in individuals of the cellular population, thereby highlighting subpopulations (e.g., of nonviable cells or of cell cycle stages where enzyme activities are very high or very low).
Time Considerations Analysis of the DNA replication cycle (Basic Protocol 1) uses fixed organisms and can be accomplished after performing other measurements, provided that storage of fixed cells (at 4°C) is for no more than 24 hr. The usefulness of the other methods using live organisms (Basic Protocols 2 to 4) lies in the speed of application: analyses require only a few minutes.
Literature Cited Agar, D.W. and Bailey, J.E. 1982. Cell cycle operation during batch growth of fission yeast populations. Cytometry 3:123-128. Alberghina, L. and Porro, D. 1993. Quantitative flow cytometry: Analysis of protein distributions in budding yeast. A mini review. Yeast 9:815823. Bell, P.J.L., Deere, D., Shen, J., Chapman, B., Bissinger, P.H., Attfield, P.V., and Veal, D.A. 1998. A flow cytometric method for rapid selection of novel yeast hybrids. Appl. Environ. Microbiol. 64:1669-1672. Brailsford, M. and Gatley, S. 1993. Rapid analysis of microorganisms using flow cytometry. In Flow Cytometry in Microbiology (D. Lloyd, ed.) pp. 171-180. Springer-Verlag, London. Bruetschy, A., Laurent, M., and Jacquet, R. 1994. Use of flow cytometry in oenology to analyse yeasts. Lett. Appl. Microbiol. 18:343-345. Carlson, C.R., Grallert, B., Bernander, R., Stokke, T., and Boye, E. 1997. Measurement of DNA content in fission yeast. Yeast 13:1329-1335. Carter, E.A., Paul, F.E., and Hunter, P.A. 1993. Cytometric evaluation of antifungal agents. In
Flow Cytometry in Microbiology (D. Lloyd, ed.) pp. 111-120. Springer-Verlag, London. Costello, G., Rodgers, L., and Beach, D. 1986. Fission yeast enters the stationary phase G0 state from either mitotic G1 or G2. Curr. Genet. 11:119-125. Davey, H.M. and Kell, D.B. 1996. Flow cytometry and cell sorting of heterogenous microbial populations: The importance of single-cell analyses. Microbiol. Rev. 60:641-696. Deere, D., Shen, J., Vesey, G., Bell, P., Bissinger, P., and Veal, D. 1998. Flow cytometry and cell sorting for yeast viability assessment and cell selection. Yeast 14:147-160. Denksteinova, B., Sigler, K., and Plásek, J. 1996. Three fluorescent probes for the flow-cytometric assessment of membrane potential in Saccharomyces cerevisiae. Folia Microbiol. 41:237242. Dinsdale, M.G., Lloyd, D., and Jarvis, B. 1995. Yeast vitality during cider fermentation: Two approaches to the measurement of membrane potential. J. Inst. Brew. 101:453-458. Dinsdale, M.G., Lloyd, D., McIntyre, P., and Jarvis, B. 1999. Yeast vitality during cider fermentation: Assessment by energy metabolism. Yeast 15:285-293. Eilam, Y. and Chernichovsky, D. 1988. Low concentrations of trifluoperazine arrest the cell division cycle of Saccharomyces cerevisiae at two specific stages. J. Gen. Microbiol. 134:1063-1069. Hayashi, M., Ohkuni, K., and Yamashita, I. 1998. Control of division arrest and entry into mitosis by extracellular alkalisation in Saccharomyces cerevisiae. Yeast 14:905-913. Hutter, K.-J. and Eipel, H.E. 1978. Flow cytometric determinations of cellular substances in algae, bacteria, molds and yeasts. Antonie van Leeuwenhoek 44:269-278. Lloyd, D. and Hayes, A.J. 1995. Vigour, vitality and viability of microorganisms. FEMS Microbiol. Lett. 133:1-7. Lloyd, D., Moran, C.A., Suller, M.T.E., Hayes, A.J., and Dinsdale, M.G. 1996. Flow cytometric monitoring of rhodamine 123 and a cyanine dye uptake by yeast during cider fermentation. J. Inst. Brew. 102:251-259. López-Amorós, R., Mason, D.J., and Lloyd, D. 1995. Use of two oxonols and a fluorescent tetrazolium dye to monitor starvation of Escherichia coli in sea water by flow cytometry. J. Microbiol. Methods 22:165-176. Müller, S., Lösche, A., and Bley, T. 1992. Flow-cytometric investigation of sterol content and proliferation activity of yeast. Acta Biotechnol. 12:365-375. Ordónez, J.V. and Wehman, N.M. 1995. Amphotericin suseptibility of Candida species assessed by rapid flow cytometric membrane potential assay. Cytometry 22:154-157. Pettipher, G.L. 1991. Preliminary evaluation of flow cytometry for the detection of yeasts in soft drinks. Lett. Appl. Microbiol. 12:109-112. Plásek, J. and Sigler, K. 1996. Slow fluorescent indicators of membrane potential: A survey of different approaches to probe response analysis. J. Photochem. Photobiol. B. 33:101-124.
Microbiological Applications
11.10.7 Current Protocols in Cytometry
Supplement 9
Porro, D., Sneraldi, C., Martegani, E., Ranzi, B.M., and Alberghina, L. 1994. Flow cytometry for monitoring yeast growth. Biotechnol. Prog. 10:193-199. Pringle, J.R., Preston, R.A., Adams, A.E.M., Stearns, T., Drubin, D.G., Haarer, B.K., and Jones, E.W. 1989. Fluorescence microscopy methods for yeast. Methods Cell Biol. 31:357435. Rose, A.H. and Harrison, J.S. (eds.). 1969. The Yeasts, Vol.1: Biology of the Yeasts, Academic Press, New York. Sazer, S. and Sherwood, S.W. 1990. Mitochondrial growth and DNA synthesis occur in the absence of nuclear DNA replication in fission yeast. J. Cell Sci. 97:509-516. Seward, R., Willetts, J.C., Dinsdale, M.G., and Lloyd, D. 1996. The effects of ethanol, hexan-lol and 2 phenylethanol on cider yeast growth, viability and energy status; synergistic inhibition. J. Inst. Brew. 102:439-443. Skarstad, K., Steen, N.B., and Boye, E. 1985. Escherichia coli DNA distributions measured by flow cytometry and compared with computer simulations. J. Bacteriol. 154:656-662.
Theorell, B. 1983. Flow-cytometric monitoring of intracellular flavins simultaneously with NAD(P)H levels. Cytometry 4:61-65. Willetts, J.C., Seward, R., Dinsdale, M.G., Suller, M.T.E., Hill, B., and Lloyd, D. 1997. Vitality of cider yeast grown micro-aerobically with added ethanol, butan-l-ol or iso-butanol. J. Inst. Brew. 103:79-84. Williamson, D.H. and Fennell, D.J. 1979. Mitochondrial DNA. In Methods in Cell Biology, Vol. 12 (D.M. Prescott, ed.) pp. 335-351. Academic Press, New York. Wittrup, K.D. and Bailey, J.D. 1988. A single-cell assay of β-galactosidase activity in Saccharomyces cerevisiae. Cytometry 9:394-404. Yurkow, E.J. and McKenzie, M.A. 1993. Characterization of hypoxia-dependent peroxide production in cultures of Saccharomyces cerevisiae using flow cytometry: A model for ischaemic tissue destruction. Cytometry 14:287-293.
Contributed by David Lloyd University of Wales Cardiff, United Kingdom
Flow Cytometry of Yeasts
11.10.8 Supplement 9
Current Protocols in Cytometry
Enumeration of Phytoplankton, Bacteria, and Viruses in Marine Samples
UNIT 11.11
The first applications of flow cytometry to the analysis of phytoplankton, in the mid1980s, revolutionized the study of the smallest organisms in this community—those small enough to pass through 2-µm-pore-size filters, called picophytoplankton. These tiny organisms are generally unicellular and are naturally quite concentrated in seawater, so that they can be analyzed without prior concentration or sonication. Picophytoplankton are present in all aquatic environments, although their relative contribution to the photosynthetic biomass is greatest in the central regions of oceans (90% of the total surface), which are nutrient depleted and relatively poor in chlorophyll (0.2 mg/m3). Data obtained by flow cytometry have helped confirm that picophytoplankton constitute the bulk of the photosynthetic biomass in subtropical waters. This unit presents a method for enumerating phytoplanktonic cells on the base of their natural parameters (see Basic Protocol 1). This protocol can be performed either on board ship or in the laboratory, and does not require any pretreatment of samples. If samples cannot be tested when freshly obtained, they can be preserved with formaldehyde or glutaraldehyde (see Support Protocol 1) and assayed later. Highly sensitive nucleic acid–specific stains such as TOTO-1, YOYO-1, and the SYBR Green family (all available from Molecular Probes) have also made it possible to detect and enumerate heterotrophic bacteria and, very recently, viruses in marine samples. Two further protocols detail the enumeration of bacteria (see Basic Protocol 2) and viruses (see Basic Protocol 3) in culture and in natural seawater samples. Both require fixation (see Support Protocol 3) and the use of nucleic acid–specific stains. Also included is a procedure for calibrating cytometer flow rates (see Support Protocol 2), replacing the standard approach using fluorescent microsphere standards, which is less suitable when working with seawater samples. FLOW CYTOMETRIC ENUMERATION OF PICOPHYTOPLANKTON BASED ON SCATTER AND AUTOFLUORESCENCE
BASIC PROTOCOL 1
The different populations present in a natural sample are discriminated on the basis of their scatter signals and the fluorescence of their natural phytoplanktonic pigments (see Fig. 11.11.1), which can vary throughout the water column (see Fig. 11.11.2). Flow cytometry is particularly suited to the analysis of picophytoplankton, which are difficult to study with traditional methods. Generally three major groups of these organisms, two cyanobacteria and a range of picoeukaryotes (algae), can be distinguished; see Anticipated Results for details. Marine samples may be obtained, for example, from the Provasoli-Guillard National Center for Culture of Marine Phytoplankton (CCMP), McKown Point, West Boothbay Harbor, Maine 04975, USA; http://CCMP.bigelow.org. Samples can be used fresh within 12 hr of being obtained (they should be stored at 4°C, but need not be fixed) or can be fixed and frozen (see Support Protocol 1), then thawed before being analyzed. Materials Natural marine samples or cultures, either fresh or frozen (see Support Protocol 1 for freezing procedure) 0.95-µm fluorescent microspheres (Polysciences) diluted to ∼105 beads/ml (as assessed by epifluorescence microscopy) in distilled water Seawater Contributed by Dominique Marie, Frédéric Partensky, Daniel Vaulot, and Corina Brussaard Current Protocols in Cytometry (1999) 11.11.1-11.11.15 Copyright © 1999 by John Wiley & Sons, Inc.
Microbiological Applications
11.11.1 Supplement 10
0.2-µm-pore-size cartridge filter units Flow cytometer equipped with a 488-nm argon laser (e.g., FACSort, Becton Dickinson) Additional reagents and solutions for flow cytometer calibration (see Support Protocol 2) 1. If sample has been frozen, thaw at 37°C. Transfer 1 ml of sample to a suitable flow cytometer tube. If the cell suspensions are too concentrated (as may be the case with culture samples, for example), they can be diluted in seawater previously filtered through a 0.2-ìm-pore-size filter.
A
B 104 Euk
Euk
103
Red fluorescence
Red fluorescence
104
Syn 102 Proc 101
103 Syn
beads
102 Proc 101
beads 100
100 100
101 102 103 Forward scatter
104
C
100
101
104
beads
Euk 103
Orange fluorescence
Red fluorescence
104
D 104
Syn 102
101 Proc
103
Syn
102
101
Euk
Proc
beads
100
100 100
Enumeration of Phytoplankton, Bacteria, and Viruses in Marine Samples
102 103 Side scatter
101 102 103 Orange fluorescence
104
100
101
102 103 Side scatter
104
Figure 11.11.1 Cytograms of scatter and fluorescence obtained for a marine sample collected in the Pacific Ocean at a depth of 65 m (OLIPAC cruise, Cast 94, 5° S to 150° W). Prochlorococcus (Proc), Synechococcus (Syn), and picoeukaryotes (Euk) are discriminated on the basis of the fluorescence of their natural pigments, chlorophyll (red) or phycoerythrin (orange). 0.95-µm beads were added as internal reference.
11.11.2 Supplement 10
Current Protocols in Cytometry
A
0 Syn Proc
Euk
Depth
50
100
150 101
B
102
103 Number of cells per ml
104
105
0
Euk Proc
Depth
50
100 Syn
150 10 1 Chlorophyll fluorescence
0.1
C
100
0 Proc Euk 50
Depth
Syn
100
150 0.01
0.1
1
Side scatter
Figure 11.11.2 Vertical profiles obtained for samples collected in the Pacific Ocean and analyzed fresh on board the N.O. l’Atalante during the OLIPAC cruise (Cast 94, 5° S to 150° W). Phytoplanktonic cell abundance (A), chlorophyll fluorescence (B), and side scatter (C) per cell normalized to 0.95-µm beads versus depth. Proc, Prochlorococcus; Syn, Synechococcus; Euk, picoeukaryotes.
Microbiological Applications
11.11.3 Current Protocols in Cytometry
Supplement 10
2. Add 10 µl of an ∼105 beads/ml suspension of 0.95-µm fluorescent microspheres (as an internal reference). 3. Filter 5 to 10 liters of seawater through 0.2-µm-pore-size cartridge filter for use as sheath fluid. Because cell scatter (especially forward scatter) is dependent on the nature of the sheath fluid, the use of filtered seawater as sheath fluid is recommended. If the fluidics system of the flow cytometer is equipped with an in-line filter, this should be removed, because it is likely to become contaminated quickly and will thereafter release particles.
4. Calibrate the flow rate of the cytometer (see Support Protocol 2). 5. Set the discriminator to red fluorescence and set all parameters on logarithmic amplification. For a surface sample from a moderately oligotrophic area, typical settings on a FACSort flow cytometer are forward scatter (FS) = E01, side scatter (SS) = 450, green fluorescence (FL1) = 650, orange fluorescence (FL2) = 650, and red fluorescence (FL3) = 650.
6. Insert the sample, allow ∼15 sec for the flow rate to stabilize, and then begin data acquisition. Data for natural samples are typically collected in listmode files for 2 to 4 min with a flow rate of 50 to 100 ìl/min.
7. Record the time of analysis to determine precisely the cell concentrations of each population. 8. Compute the absolute cell concentration for each population in a given sample as follows: Cpop = T × Npop/R × (Vtotal/Vsample) where Cpop = concentration of population in cells/µl, Npop = number of cells acquired, T = acquisition time (min), R = sample flow rate (µl/min) as determined for the sample series, Vtotal = volume (µl) of sample plus additions (fixatives, beads, etc.), and Vsample = volume of sample (µl). 9. Report parameters relative to the beads added to the samples: Xrel = Xpop/Xbeads where Xpop is the average value of a cell parameter (scatter or fluorescence) for a given population and Xbeads the same parameter for the beads. Before calculation of the ratio, Xpop and Xbeads must be expressed as linear values (not numbers of channels) after conversion from the logarithmic recording scale. BASIC PROTOCOL 2
Enumeration of Phytoplankton, Bacteria, and Viruses in Marine Samples
FLOW CYTOMETRIC ENUMERATION OF BACTERIOPLANKTON BY DNA STAINING AND FLUORESCENT DETECTION Because of its accuracy, its speed, and the lack of interference from dissolved organic matter, flow cytometry has been increasingly used to analyze heterotrophic bacteria (Shapiro, 1988; Robertson et al., 1998). In contrast to the photosynthetic prokaryotes Prochlorococcus and Synechococcus, bacteria do not contain any pigments and cannot be counted based on autofluorescence. Staining of cell DNA has been used as a means to discriminate and enumerate bacteria in natural seawater samples by epifluorescence microscopy (EFM; Hobbie et al., 1977) or flow cytometry (Button and Robertson, 1989; Monger and Landry, 1993; Li et al., 1995; Marie et al., 1997). The combination of DNA and chlorophyll fluorescence allows discrimination of autotrophic from heterotrophic
11.11.4 Supplement 10
Current Protocols in Cytometry
picoplankton (Monger and Landry, 1993; Campbell et al., 1994). For details of the bacterial populations generally observed using flow cytometry, see Anticipated Results. Older UV-excited dyes, such as DAPI or Hoechst 33342, that require expensive flow cytometric equipment, are currently being superseded by a wide and continually expanding range of nucleic acid–specific dyes synthesized and manufactured by Molecular Probes. These novel dyes are excited at 488 nm, which means they are usable on small, low-cost flow cytometers equipped with air-cooled single-line argon lasers. The affinity of the cyanine dyes TOTO-1 and YOYO-1, and their monomeric equivalents YO-PRO-1 and TO-PRO-1 (all available from Molecular Probes), decreases significantly with increasing ionic strength, so they are inappropriate for direct analysis of seawater samples (Marie et al., 1996). Other dyes such as SYBR Greens I and II, SYTOX Green, and the SYTO family (all available from Molecular Probes) are less dependent on medium composition and can be used for enumerating bacteria in marine environments (Marie et al., 1997; Lebaron et al., 1998). Because SYBR Green I (SYBR-I) has a very high fluorescence yield, the authors strongly recommend the use of this dye to enumerate bacteria in marine samples. Samples must be fixed before bacterial enumeration can be performed, since fixation allows the nucleic acid–specific stain to penetrate into the cell. A 10,000-fold dilution of the commercial SYBR-I stock solution is used. NOTE: All stock solutions except the dye must be prefiltered through a 0.2-µm- (or smaller) pore-size filter to avoid contamination. Materials Natural marine samples or cultures, either fresh or frozen (see Support Protocol 1 for freezing procedure) 10% paraformaldehyde (see Support Protocol 3) and/or 25% glutaraldehyde (Sigma) DNA-specific stain such as SYBR Green I, YOYO-1, TOTO-1, or TO-PRO-1 (Molecular Probes) 0.95-µm fluorescent microspheres (Polysciences) diluted to ∼105 beads/ml (as assessed by epifluorescence microscopy) in distilled water Seawater 0.2-µm-pore-size cartridge filter units Flow cytometer equipped with a 488-nm argon laser (e.g., FACSort, Becton Dickinson) Additional reagents and solutions for flow cytometer calibration (see Support Protocol 2) 1a. If samples are live: Add 1% paraformaldehyde or 0.1% glutaraldehyde (final concentrations) and let stand 20 min. Paraformaldehyde and glutaraldehyde give equivalent results.
1b. If samples have been preserved and frozen: Thaw samples at 37°C. If the cell suspensions are too concentrated (as may be the case with culture samples, for example), they can be diluted in seawater previously filtered through a 0.2-ìm-pore-sizefilter.
2. Add SYBR-I at a final concentration of 1 part in 10,000 and incubate 15 min at room temperature in the dark. Microbiological Applications
11.11.5 Current Protocols in Cytometry
Supplement 10
Figure 11.11.3 Cytograms of side scatter and fluorescence obtained for a natural sample collected in the Pacific Ocean at a depth of 65 m (OLIPAC cruise, Cast 94, 5° S to 150° W) and stained with SYBR-I. Three different bacterial populations (B-I, B-II, and B-III) can be discriminated from Prochlorococcus by the combination of the different parameters recorded. 0.95-ìm fluorescent beads were added as internal reference.
3. To 1 ml of sample, add 10 µl of an ∼105 bead/ml suspension of 0.95-µm fluorescent microspheres (as an internal reference). 4. Filter 5 to 10 liters of seawater through 0.2-µm-pore-size cartridge filter for use as sheath fluid. Distilled water can be used as sheath fluid, but for natural seawater samples, 0.2-ìm-poresize-filtered seawater is preferable, since cell scatter (especially forward scatter) is dependent on the nature of the sheath fluid. If the fluidics system of the flow cytometer is equipped with an in-line filter, this should be removed, because it is likely to become contaminated quickly and will thereafter release particles.
5. Set the discriminator to green fluorescence. 6. Calibrate the flow rate of the cytometer (see Support Protocol 2). 7. Set all parameters on logarithmic amplification. It is recommended that no more than 80,000 events be acquired in listmode, in order to avoid very large files. Typical settings on a FACSort flow cytometer are FS = E01, SS = 450, FL1 = 650, FL2 = 650, and FL3 = 650.
8. Run the sample, adjusting the flow rate and cell concentration to avoid coincidence. Enumeration of Phytoplankton, Bacteria, and Viruses in Marine Samples
Typically, the authors run samples for 1 to 2 min at a flow rate of 30 to 50 ìl/min and keep the number of events below 1000 per sec (by diluting samples that are too concentrated). Some samples, particularly those obtained in coastal areas, contain copious quantities of small particles and debris that will increase the level of background noise. This can induce
11.11.6 Supplement 10
Current Protocols in Cytometry
coincidence or lead to the generation of large listmode files. In such cases, the threshold can be increased to reduce the number of events seen by the flow instrument, and/or a “bitmap” window (non-regular region) can be defined that includes the population of bacteria so that only the events belonging to this area will be recorded (see Fig. 11.11.3).
FLOW CYTOMETRIC ENUMERATION OF VIROPLANKTON BY DNA STAINING AND FLUORESCENT DETECTION
BASIC PROTOCOL 3
The existence of bacteriophages in marine environments has been known for many years (Kriss and Rukina, 1947; Spencer, 1955, 1960), but they were not really investigated until fairly recently (Bergh et al., 1989; Bratbak et al., 1990; Proctor et al., 1990). Viroplankton clearly constitute the most abundant population of biological particles in the ocean and their ecological role has only recently been investigated. These studies initially required techniques such as transmission electron microscopy (TEM) that are time consuming and allow only limited numbers of samples to be analyzed. During the past decade, investigations using epifluorescence microscopy (EFM) in conjunction with nucleic acid–specific dyes such as DAPI (Hara et al., 1991) or with cyanine dyes (Hennes and Suttle, 1995; Weinbauer and Suttle, 1997) have considerably improved knowledge of marine viruses. Very recently, flow cytometry has been successfully applied to the analysis of viruses in solution, using the nucleic acid–specific dye SYBR Green I (Marie et al., 1999). This has permitted the analysis of viruses with reduced DNA content, down to 40 Kbp (Brussaard et al., unpub. observ.). Other dyes, such as SYTOX, PicoGreen, OliGreen, SYBR Green II, SYBR Gold, or RiboGreen (all from Molecular Probes), can be used with the same efficiency as SYBR-I (Brussaard et al., unpub. observ.). For details of the viroplankton populations generally observed using flow cytometry, see Anticipated Results. NOTE: All stock solutions except the dye must be prefiltered through a 0.2-µm- (or smaller) pore-size filter to avoid contamination. Materials Natural marine samples or cultures, either fresh or frozen (see Support Protocol 1 for freezing procedure) 10% paraformaldehyde (see Support Protocol 3) or 25% glutaraldehyde (Sigma) TE buffer, pH 7.2 (APPENDIX 2A) DNA-specific stain(s) such as SYBR Green I or II, OliGreen, or RiboGreen (Molecular Probes) 0.95-µm fluorescent microspheres (Polysciences) diluted to ∼105 beads/ml (as assessed by epifluorescence microscopy) in distilled water Distilled water 0.2-µm-pore-size filtration units for plastic syringe Flow cytometer equipped with a 488-nm argon laser (e.g., FACSort, Becton Dickinson) Additional reagents and solutions for flow cytometer calibration (see Support Protocol 2) Prepare sample 1a. For fresh samples: Add 1% paraformaldehyde or 0.1% to 0.5% glutaraldehyde (final concentrations) and let stand 20 min. No significant differences have been found between results for virus enumeration performed on samples fixed with paraformaldehyde, glutaraldehyde, or a mixture of both.
1b. For fixed and frozen samples: Thaw samples at 37°C. Microbiological Applications
11.11.7 Current Protocols in Cytometry
Supplement 10
2. Dilute samples in TE buffer, pH 7.2, to three different concentrations: typically 10-, 50-, and 100-fold for natural seawater samples and 100-, 1,000-, and 10,000-fold for cultured samples. Preparation of three different dilutions is necessary because the concentration of viruses is not known beforehand. Analysis must be performed with a suspension of ∼2 × 105 to 2 × 106 viruses/ml (final concentration). To avoid generating large files, samples can be run for 1 or 2 min at a rate ranging from 20 to 50 ìl/s. Different buffers have been tested for diluting virus samples. Tris-based buffers give the best result.
3. Add SYBR-I at a final concentration of 5 parts in 100,000 and incubate 15 min at room temperature in the dark. 4. To 1 ml of sample, add 10 µl of an ∼105 beads/ml suspension of 0.95-µm fluorescent microspheres (as an internal reference). For virus samples that are freshly fixed (i.e., have not been frozen), or for hard-to-stain material, it is necessary to heat the samples 10 min at 80°C in the presence of detergent (e.g., Triton X-100 at 0.1% final) to improve dye uptake.
Acquire data 5. Using distilled water as sheath fluid (even for marine samples), begin the cytometric procedure by calibrating the flow rate (see Support Protocol 2). Since samples are diluted in TE, use of seawater is not necessary.
6. Turn the discriminator to green fluorescence (FL1). Typical settings on a FACSort flow cytometer are FS = E03, SS = 600, FL1 = 600, FL2 = 650, and FL3 = 650.
7. Before starting data acquisition, wait for the sample flow rate to stabilize (this can take up to 20 sec). 8. Run the sample at a rate allowing <1000 events/sec (to avoid coincidence; see Basic Protocol 2, step 8, for discussion of this problem). SUPPORT PROTOCOL 1
PRESERVATION AND STORAGE OF PICOPHYTOPLANKTON If samples cannot be run immediately, they may be kept up to 12 hr at 4°C without significant change in abundance or optical parameters. If they cannot be analyzed within that time interval, they must be fixed for 15 to 20 min with formaldehyde or glutaraldehyde, then deep frozen in liquid nitrogen and stored at −80°C until analysis. Frozen samples can be kept for at least 1 year. NOTE: All stock solutions except the dye must be prefiltered through a 0.2-µm- (or smaller) pore-size filter to avoid contamination. Materials 10% paraformaldehyde (see Support Protocol 3) and/or 25% glutaraldehyde (Sigma) 1. Add paraformaldehyde or glutaraldehyde, or both, to freshly obtained water samples at final concentrations of 1% and 0.1%, respectively. 2. Wait 15 min.
Enumeration of Phytoplankton, Bacteria, and Viruses in Marine Samples
3. Freeze the samples in liquid nitrogen. 4. Store at −20°C for a few weeks or at −80°C for longer periods.
11.11.8 Supplement 10
Current Protocols in Cytometry
CALIBRATION OF THE CYTOMETER FLOW RATE Accurate calibration of the sample flow rate is essential for obtaining reliable cell counts. As most flow cytometers do not allow delivery of defined volumes of samples, fluorescent microspheres with a known concentration are often used to determine the flow rate. The authors do not use this method, because (1) microspheres are electrostatic, and seawater makes them stick on plastic tube walls, changing their initial concentration, and (2) the usual method of determining microsphere concentration, epifluorescence microscopy, generates large counting errors. The authors present below a method that is usable for flow cytometers such as the FACScan, FACSort, or FACScalibur (Becton Dickinson), but can be extended to most existing flow instruments.
SUPPORT PROTOCOL 2
NOTE: All stock solutions except the dye must be prefiltered through a 0.2-µm- (or smaller) pore-size filter to avoid contamination. 1. Select a rate (Low, Medium, or High). 2. Fill a tube with the same liquid as the one containing samples (i.e., seawater for marine samples). 3. Measure the volume of sample (or weigh precisely the tube containing the sample). 4. Remove the outer sleeve of the injection system. The sheath fluid will drop down the sample needle.
5. Wait until a droplet just falls. Before the next one forms, put on the sample tube and close the sample arm in the running position. Simultaneously, start the chronometer running. 6. Run the sample for at least 10 min. 7. Remove the sample tube and simultaneously stop the chronometer. 8. Measure (or weigh) the remaining volume. 9. Calculate the rate (R), expressed in microliters per minute, by one of the following two methods. Volume measurement: R = (Vi − Vf)/T where Vi = initial volume (µl), Vf = final volume (µl), and T = time (min). Weight measurement: R = (Wi − Wf)/(T × d) where Wi = initial weight (mg), Wf = final weight (mg), T = time (min), and d = density of the liquid used for calibration (distilled water = 1.00, seawater = 1.03). The weight measurement provides better precision.
PREPARATION OF BUFFERED 10% PARAFORMALDEHYDE STOCK SOLUTION
SUPPORT PROTOCOL 3
To preserve marine samples, the authors generally use either 1% paraformaldehyde, 0.1% glutaraldehyde, or a mixture of 1% paraformaldehyde and 0.05% glutaraldehyde (final concentrations), with a preference for paraformaldehyde. The solution is buffered so that it will not significantly modify the pH of seawater samples. The following protocol describes the preparation of a 10% paraformaldehyde solution.
Microbiological Applications
11.11.9 Current Protocols in Cytometry
Supplement 10
CAUTION: Paraformaldehyde is an irritant. Wear protective clothing and work in a fume hood. Materials Paraformaldehyde (e.g., Sigma) Distilled water 1 M NaOH Phosphate-buffered saline (PBS; APPENDIX 2A), pH 7.5 1 M HCl 0.2-µm-pore-size filtration unit for plastic syringe 1. Weigh 10 g paraformaldehyde under a fume hood. 2. Add 85 ml boiling distilled water. 3. Stir vigorously at 70°C for at least 2 hr under a fume hood, until the formaldehyde dissolves and saturates the water. 4. Add small amounts of 1 M NaOH until the solution becomes clear. 5. Add 10 ml of PBS, pH 7.5. 6. Adjust the pH to 7.5 with 1 M HCl, then dilute to 100 ml with distilled water. 7. Filter first through filter paper, then through a 0.2-µm-pore-size syringe filter. 8. Divide into aliquots in 15-ml tubes and store at −20°C. Unfrozen aliquots can be stored up to 1 week at 4°C. Paraformaldehyde is the polymerized form of formaldehyde which, in contrast to formaldehyde, lacks cross-linking properties. When liquid, the solution is unstable over time.
COMMENTARY Background Information
Enumeration of Phytoplankton, Bacteria, and Viruses in Marine Samples
General considerations Photoacclimation is widely observed in oceanic samples. Phytoplankton acclimate to changes of photon-flux densities by changing pigment content (Partensky et al., 1996). This leads to variations in the chlorophyll fluorescence per cell along a depth profile (see Fig. 11.11.2) that are extreme for Prochlorococcus (up to 50-fold) and less pronounced for picoeukaryotes (5- to 10-fold). The intensity of other cellular parameters, such as scatter and orange fluorescence, also varies throughout the water column. Cell size and chlorophyll fluorescence will also vary with the level of available nutrients, typically decreasing as nutrients become limiting (see Fig. 11.11.2). Thus the photomultiplier (PMT) voltages must be adjusted depending on the size of organisms of interest and on the depth sampled, so that the relative position of the organisms of interest remains approximately the same.
Acquisition and data analysis Parameters are collected on logarithmic scales in order to obtain the multidecade dynamic range necessary to analyze the majority of the organisms present, which in natural samples can range widely in size and fluorescence properties. Data are always collected as listmode files; typically 20,000 to 40,000 events are collected for enumeration of phototrophs and up to 80,000 for bacteria or viruses. Listmode files are then analyzed by using the free software CytoWin (Vaulot, 1989), available at http://www.sb-roscoff/Phyto/cyto.html, which is very efficient at rapidly processing a large number of files. The different populations are discriminated based on the combination of their scatter signals and the fluorescence of their natural pigments or of the nucleic acid–stain complex. To allow comparison between different samples, the cell parameters for each sample are normalized to those for 0.95-µm microspheres added as an internal reference, by dividing the mean value of each parameter by the mean value for the beads.
11.11.10 Supplement 10
Current Protocols in Cytometry
Flow cytometer The choice of flow cytometer is critical for the analysis of picoplanktonic cells. The chief criterion is sensitivity. Prochlorococcus cells are very dim in the upper layer of oligotrophic waters, and can very easily be missed. Natural viroplankton display a wide range of sizes and are often difficult to separate clearly from background noise. The instrument must also be compact enough to be used on board ship during oceanographic cruises. The FACSort and FACScalibur flow cytometers from Becton Dickinson fulfill these criteria and are suitable for the analysis of small picoplanktonic cells. Nevertheless, some populations still may not completely resolved—for instance, Prochlorococcus in surface waters of extreme oligotrophic areas. Custom modifications have been proposed to deal with this inadequacy (Dusenberry and Frankel, 1994). Another problem is that the volume of sample typically analyzed by flow cytometers (<1 ml) is too small for accurate enumeration of larger and less concentrated phytoplanktonic populations such as dinoflagellates or diatoms. Custom modifications and even new instruments have been developed for this purpose (Dubelaar et al., 1989; Cavender-Bares et al., 1998). Sample preservation Measurements of phytoplankton abundance are best obtained on fresh unfixed seawater samples, which may be kept at 4°C for up to 12 hr. If samples cannot be analyzed within this time interval, it is necessary to preserve them for delayed analysis. Fixation will always results in a loss of cells (Vaulot et al., 1989); the choice of fixative is critical and should be optimized for the species of interest. The fixation procedure must also be simple enough to be usable on board ship. Physical treatments such as centrifugation and classical or tangential filtrations must be avoided because they induce variable losses of cells. Because phytoplanktonic cells are discriminated on the basis of scatter and pigment fluorescence, the fixation procedure must preserve these properties—which rules out such classical methods as formalin and Lugol fixations that modify cell shape or significantly affect fluorescence. Similarly, alcohol fixation will extract lipophilic pigments and lead to a loss of autofluorescence. Natural seawater samples are best preserved with paraformaldehyde (0.5% to 1% final concentration), glutaraldehyde (0.1% to 1%), or a mixture of the two. If careful preparation of paraformaldehyde is not possible, the
use of a commercial 25% glutaraldehyde solution is preferable. Fixed samples must be stored at −80°C, as they degrade within a few months at −20°C. For a mixture of Prochlorococcus, Synechococcus, and a picoeukaryote, the authors observed no significant loss after 1 month of storage at −20°C, but 50% of Prochlorococcus and up to 80% of eukaryote cells were lost after 6 months. Identification of phytoplankton population Before attempting to analyze natural samples, it is critical to analyze laboratory cultures of each cell type likely to be observed (see Anticipated Results). Failure to do so could result in misinterpretation of natural samples. Analysis of depth profile For analysis of samples taken at different depths along a vertical profile, it is best to start with a sample collected at a depth corresponding to the chlorophyll concentration maximum as measured remotely. Use the same setup to analyze samples obtained from deeper levels, where cells of interest become too scarce and it is difficult to adjust PMTs precisely. Then continue by going up the water column, progressively increasing the voltage of the PMT as needed to detect the cells of interest. If only surface samples are available, set the threshold at the minimum value and increase the red PMT voltage until some noise appears (∼50 events per sec). Then fix the PMT voltage and run the sample. Bacterial staining For bacterial enumeration, if samples are in suspension in a nonsaline solution, or can be diluted enough to minimize the effects of seawater or ionic strength, the authors recommend final concentrations of 1 µg/ml for DAPI or Hoechst, 30 nM for TOTO, YOYO, TO-PRO, or YO-PRO, 1 part in 1000 for PicoGreen, and 1 part in 10,000 for SYBR Green I or SYTOX Green. Viroplankton analysis Viruses are too small to be discriminated solely on the basis of their side- or forwardscatter properties on flow cytometers such as the Becton Dickinson FACS series. Nucleic acid–specific staining is therefore necessary. Because flow cytometry was not designed for the analysis of such small particles, care must be taken in order to obtain reliable data. If samples are too diluted, there will be loss in the emission signal of the nucleic acid–dye com-
Microbiological Applications
11.11.11 Current Protocols in Cytometry
Supplement 10
B
A
103
104
M. pusilla bacteria viruses
beads
102
101
beads bacteria
103
Green fluorescence
Green fluorescence
104
102 V-II
101
100
V-I
100 100
101
102 103 Side scatter
104
C
100
101
102 103 Side scatter
104
D V-II Counts
Counts
viruses
V-I bacteria
bacteria
100
101 102 103 Green fluorescence
104
100
101 102 103 Green fluorescence
104
Figure 11.11.4 Side scatter versus green fluorescence and monoparametric DNA distribution obtained for a non-axenic culture of Micromonas pusilla infected by the virus MpV-01 (A and C) and for a natural sample (B and D) collected in the Raunefjorden, Western Norway, both stained with SYBR-I. 0.95-µm fluorescent microspheres were added as internal reference.
Enumeration of Phytoplankton, Bacteria, and Viruses in Marine Samples
plex; if they are insufficiently diluted, coincidence will occur. Analysis of different dilutions of natural seawater samples has shown that coincidence occurs for viruses above 800 events per second. However, on flow cytometers such as the FACSort, for suspensions of beads, bacteria, or small algae coincidence normally occurs above 2000 events per sec. For concentrated suspensions, above 800 objects per second, more virus doublets are observed, which result in an increase of the fluorescence signal due to viruses passing simultaneously through the laser beam. Because all the V-II and 20% of the V-I virus populations can pass through 0.2-µm-pore-size filters, 0.2-µm-pore-size-filtered seawater cannot be used to dilute the samples. Reasonable alternatives are 0.05-µm-pore-size-filtered
seawater and buffers such as TE; the use of TE improves the emission signal of stained viruses, making this the best option.
Troubleshooting Detection of Prochlorococcus in surface waters In highly oligotrophic waters, such as those of the subtropical Pacific Ocean, it is not always possible with unstained samples to detect the whole Prochlorococcus population in surface water samples because its chlorophyll fluorescence is too weak. In such cases, after staining with SYBR-I, Prochlorococcus cells will be included in the heterotrophic bacteria population, from which they cannot be discriminated
11.11.12 Supplement 10
Current Protocols in Cytometry
(see Fig. 11.11.3A); this must be taken into consideration in interpreting results. Preservation of the samples Preliminary tests must be performed to choose the best fixative for the cells of interest. Seawater is naturally buffered at about pH 8. Glutaraldehyde is acidic, and when dissolved at final concentrations of 0.1%, 0.25%, 0.5%, and 1% in seawater will produce pH values of 7.84, 7.42, 6.85, and 6.35, respectively. pH <7 is particularly damaging to fragile phytoplanktonic cells such as Prochlorococcus or small eukaryotes. Virus staining Viruses contain small amounts of nucleic acids. The critical point for virus staining resides in the equilibrium between dye concentration and virus abundance. If viruses are too concentrated, a decrease in fluorescence will result. For virus numbers that do not saturate the machine’s acquisition capacity, an increase in dye concentration will also result in loss of signal. Moreover, virus abundance determined by flow cytometry on fixed but unfrozen samples is 3- to 10-fold lower than that found for frozen samples. This suggests that live viruses may have a structure that prevents access of SYBR-I to nucleic acids. Detergent or heat treatment up to 95°C may be needed to denature the virus capsid and allow the stain to penetrate. For unknown virus material, the concentration of dye as well as the effect of heating must be assessed. In general, the authors observed that using half the concentration of SYBR-I used for bacterial staining (to 5 parts in 100,000) and heating between 70° to 80°C are suitable for the majority of viruses analyzed.
Anticipated Results Picophytoplankton Picophytoplankton are categorized into three major groups (see Fig. 11.11.1). Prochlorococcus, a cyanobacterium 0.6 µm in size, is a photosynthetic organism that contains divinyl derivatives of chlorophylls a and b. Its discovery in 1988 was one of the most significant results obtained so far from the application of flow cytometry to aquatic sciences. This organism is undoubtedly the most abundant genus of phytoplankton, reaching >105 cells/ml, and its vertical distribution extends from the surface to depths of 150 m or more within the intertropical belt (see Fig. 11.11.2; Chisholm et al., 1988,
1992; Partensky et al., 1999). Prochlorococcus is responsible for about half the biomass and the primary production in warm offshore waters. Synechococcus, also a cyanobacterium, is 1 µm in size and was discovered in 1979 (Waterbury et al., 1979). It is characterized by the dual fluorescence of its pigments: orange from phycoerythrin and red from chlorophyll. Synechococcus is found at low concentrations in oligotrophic waters (Campbell and Vaulot, 1993) but can be very dense (∼105 cells/ml) in mesotrophic and coastal areas (Olson et al., 1988; Partensky et al., 1996). Picoeukaryotes belong to a variety of algal classes, in particular Prasinophyceae, Pelagophyceae, and Bolidophyceae. Field data indicate a typical abundance of 103 cells/ml in open ocean waters (Li and Wood, 1988; Campbell and Vaulot, 1993) and up to ten times higher concentrations in coastal waters (Courties et al., 1994). Bacterioplankton Using TOTO-1 or TO-PRO-1, some authors have distinguished two different populations of bacteria (B-I and B-II) in natural seawater samples that differ both in scatter and in DNA content (Li et al., 1995). With SYBR-I two or three different populations can be distinguished (see Fig. 11.11.3). The B-I group seems to dominate in oligotrophic areas, while the B-II group increases in less oligotrophic conditions (Li et al., 1995; Marie et al., 1997). The third group is commonly found in coastal areas. Viroplankton Using SYBR-I, the authors have been able to distinguish at least two different populations of viruses (V-I and V-II) in natural seawater samples (see Fig. 11.11.4B and 11.11.4D). Viroplankton belonging to the V-I population have a higher fluorescence similar to that of cultured viruses specific to eukaryotic algae such as Micromonas pusilla (see Fig. 11.11.4A and C).
Time Considerations Phytoplanktonic cells do not require any treatment prior to enumeration, so samples of such cells can be analyzed immediately. When analysis must be delayed or when enumerating bacteria and viruses, where fixation is required, samples must be fixed for 15 min and then frozen. Staining is achieved by incubating samples for 15 min at room temperature in the dark before flow cytometric analysis.
Microbiological Applications
11.11.13 Current Protocols in Cytometry
Supplement 10
Literature Cited Bergh, Ø., Børsheim, K.Y., Bratbak, G., and Heldal, M. 1989. High abundance of viruses found in aquatic environments. Nature 340:467-468. Bratbak, G., Heldal, M., Norland, S., and Thingstad, T.F. 1990. Viruses as partners in spring bloom microbial trophodynamics. Appl. Environ. Microbiol. 56:1400-1405. Button, D.K. and Robertson, B.R. 1989. Kinetics of bacterial processes in natural aquatic systems based on biomass as determined by high-resolution flow cytometry. Cytometry 10:558-563. Campbell, L. and Vaulot, D. 1993. Photosynthetic picoplankton community structure in the subtropical North Pacific Ocean near Hawaii (station ALOHA). Deep-Sea Res. 40:2043-2060. Campbell, L., Nolla, H.A., and Vaulot, D. 1994. The importance of Prochlorococcus to community structure in the central North Pacific Ocean. Limnol. Oceanogr. 39:954-961.
Lebaron, P., Catala, P., and Parthuisot, N. 1998. Effectiveness of SYTOX Green stain for bacterial viability assessment. Appl. Environ. Microbiol. 98:2697-2700. Li, W.K.W. and Wood, A.M. 1988. Vertical distribution of North Atlantic ultraphytoplankton: Analysis by flow cytometry and epifluorescence microscopy. Deep Sea Res. 35:1615-1638. Li, W.K.W., Jellett, J.F., and Dickie, P.M. 1995. DNA distribution in planktonic bacteria stained with TOTO or TO-PRO. Limnol. Oceanogr. 40:1485-1495. Marie, D., Vaulot, D., and Partensky, F. 1996. Application of the novel nucleic acid dyes YOYO-1, YO-PRO-1 and PicoGreen analysis of marine prokaryotes. Appl. Environ,. Microbiol. 6 2 : 1649-1655.
Cavender-Bares, K.K., Frankel, S.L., and Chisholm, S.W. 1998. A dual sheath flow cytometer for shipboard analyses of phytoplankton communities from the oligotrophic oceans. Limnol. Oceanogr. 43:1383-1388.
Marie, D., Partensky, F., Jacquet, S., and Vaulot, D. 1997. Enumeration and cell cycle analysis of natural populations of marine picoplankton by flow cytometry using the nucleic acid stain SYBR Green-I. Appl. Environ. Microbiol. 93:186-193.
Chisholm, S.W., Olson, R.J., Zettler, E.R., Waterbury, J., Goericke, R., and Welschmeyer, N. 1988. A novel free-living prochlorophyte occurs at high cell concentrations in the oceanic euphotic zone. Nature 334:340-343.
Marie, D., Brussaard, C.P.D., Thyrhaug, R., Bratbak, G., and Vaulot, D. 1999. Enumeration of marine viruses in culture and natural samples by flow cytometry. Appl. Environ. Microbiol. 65:45-52.
Chisholm, S.W., Frankel, S.L., Goericke, R., Olson, R.J., Palenik, B., Waterbury, J.B., WestJohnsrud, L., and Zettler, E.R. 1992. Prochlorococcus marinus nov. gen. nov. sp.: An oxyphototrophic marine prokaryote containing divinyl chlorophyll a and b. Arch. Microbiol. 157:297300.
Monger, B.C. and Landry, M.R. 1993. Flow cytometric analysis of marine bacteria with Hoechst 33342. Appl. Environ. Microbiol. 59:905-911.
Courties, C., Vaquer, A., Trousselier, M., Lautier, J., Chrétiennot-Dinet, M-J., Neveux, J., Machado, C., and Claustre, H. 1994. Smallest eukaryotic organism. Nature 370:255. Dubelaar, G.B.J., Groenewegen, A.C., Stokdijk, W., Van Den Engh, G.J., and Visser, J.W.M. 1989. Optical plankton analyser: A flow cytometer for plankton analysis, II: Specifications. Cytometry 10:529-539. Dusenberry, J.A. and Frankel, S.L. 1994. Increasing the sensitivity of a FACScan flow cytometer to study oceanic picoplankton. Limnol. Oceanogr. 39:206-210. Hara, S., Terauchi, K., and Koike, I. 1991. Abundance of viruses in marine waters: Assessment by epifluorescence and transmission electron microscopy. Appl. Environ. Microbiol. 57:27312734.
Enumeration of Phytoplankton, Bacteria, and Viruses in Marine Samples
Kriss, A.E. and Rukina, E.A. 1947. Bacteriophages in the sea. Dokl. Akad. Nauk SSSR 57:833-836.
Hennes, K.P. and Suttle, C.A. 1995. Direct counts of viruses in natural waters and laboratory cultures by epifluorescence microscopy. Limnol. Oceanogr. 40:1050-1055. Hobbie, J.E., Daley, R.J., and Jasper, S. 1977. Use of Nuclepore filters for counting bacteria by fluorescence microscopy. Appl. Environ. Microbiol. 33:1225-1228.
Olson, R.J., Chisholm, S.W., Zettler, E.R., and Armbrust, E.V. 1988. Analysis of Synechococcus pigment types in the sea using single and dual beam flow cytometry. Deep Sea Res. 35:425440. Partensky, F., Blanchot, J., Lantoine, F., Neveux, J., and Marie, D. 1996. Vertical structure of picophytoplankton at different trophic sites of the tropical northeastern Atlantic Ocean. Deep Sea Res 43:1191-1213. Partensky, F., Hess, W.R., and Vaulot, D. 1999. Prochlorococcus, a marine photosynthetic prokaryote of global significance. Microbiol. Mol. Biol. Rev. 63:106-127. Proctor, L.M. and Fuhrman, J.A. 1990. Viral mortality of marine bacteria and cyanobacteria. Nature 343:60-62. Robertson, B.R., Button, D.K., and Kloch, A.L. 1998. Determination of the biomasses of small bacteria at low concentrations in a mixture of species with forward light scatter measurements by flow cytometry. Appl. Environ. Microbiol. 64:3900-3909. Shapiro, L.P. and Haugen, E.M. 1988. Seasonal distribution and temperature tolerance of Synechococcus in Boothbay Harbor, Maine. Estuarine Coastal Shelf Sci. 26:517-525. Spencer, R. 1955. A marine bacteriophage. Nature 175:690.
11.11.14 Supplement 10
Current Protocols in Cytometry
Spencer, R. 1960. Indigenous marine bacteriophages. J. Bacteriol. 79:614. Vaulot, D. 1989. CYTOPC: Processing software for flow cytometric data. Signal Noise 2:8.
http://carl.im.uib.no/sup Provides information on marine viruses. http://www.flowcytometry.org
Vaulot, D., Courties, C., and Partensky, F. 1989. A simple method to preserve oceanic phytoplankton for flow cytometric analyses. Cytometry 10:629-635.
Contains a wide range of resources for marine applications of flow cytometry.
Waterbury, J.B., Watson, S.W., Guillard, R.R.L., and Brand, L.E. 1979. Widespread occurrence of a unicellular, marine planktonic, cyanobacterium. Nature 277:293-294.
Catalogs and maintains algal strains to be used for protocol development.
Weinbauer, M.E. and Suttle, C.A. 1997. Comparison of epifluorescence and transmission electron microscopy for counting viruses in natural marine waters. Aquat. Microb. Ecol. 13:225-232.
Contributed by Dominique Marie, Frédéric Partensky, and Daniel Vaulot Station Biologique Roscoff, France
Internet Resources
Corina Brussaard University of Bergen Bergen, Norway
http://www.sb-roscoff.fr/Phyto/cyto.html Lists marine applications of flow cytometry and provides a downloadable copy of the Cyto Win software.
http://CCMP.bigelow.org
The authors wish to acknowledge the support of the European Community MAST III programs (MAS3-CT96-015- MEDEA, MAS3-CT97-0128- PROMOLEC, MAS3-CT965033[DG12-ASAL]), the JGOFS-France PROSOPE program, and the Research Council of Norway (project number 121425/420). The FACSort flow cytometer was funded in part by CNRS-INSU and the Région Bretagne.
Microbiological Applications
11.11.15 Current Protocols in Cytometry
Supplement 10
DNA/RNA Analysis of Phytoplankton by Flow Cytometry
UNIT 11.12
In the last two decades, flow cytometry has been widely used in oceanography to estimate the abundance of the different populations composing the picophytoplankton and to obtain general information based on their cell size and pigment contents. Nucleic acid stains are used by aquatic biologists to enumerate heterotrophic bacteria, and more recently, viruses, by flow cytometry. For this type of application, the stoichiometry of staining is not critical since the dye is used as a simple coloring tool. The following basic protocols describe methods that allow a more precise characterization of the whole genome of very small organisms. These methods have several major applications, including the precise determination of the genome size and ploidy level (see Basic Protocol 1) and cell cycle analysis (see Basic Protocol 2) of phytoplanktonic cells. One further step in the characterization of phytoplanktonic populations is the use of fluorescent oligonucleotide probes targeted to 18S rRNA that permits discrimination of specific taxa in the heterogeneous natural communities of picophytoplankton (Basic Protocol 3). DETERMINING GENOME SIZE AND PLOIDY LEVEL OF PHYTOPLANKTON During the last two decades, numerous studies have focused on the measurement of genome size and ploidy level of land plants by flow cytometry (Galbraith et al., 1983; Brown et al., 1991). These analyses, which were performed directly on nuclei isolated from fresh plant tissues, have been adapted for marine purposes. In general, estimation of genome size is best performed on isolated nuclei. Nevertheless, when a complex membrane composition prevents the release of nuclei from algae or when isolated nuclei are not stable over time, these analyses can be performed on fixed cells.
BASIC PROTOCOL 1
Little information is available on the genome size and the base composition of marine algae (Veldhuis et al., 1997). In addition to being an important item of information by Table 11.12.1 Dyes Used for Determination of Genome Size and Cell Cycle Analysis of Planktonic Cells
Dye
MW
Absorption maximum (λA; nm)
Fluorescence Working maximum conc. (λF; nm)
Final conc.
Ethidium bromide Propidium iodide Mithramycin Chromomycin A DAPI Hoechst 33342 TOTO-1 YOYO-1 TO-PRO-1 YO-PRO-1 PicoGreen SYBR Green I
394 668 1085 1183 350 652 1303 1271 645 629 NAb NAb
518 535 433 420 357 365 514 491/450a 515 491/450a 480 494
605 617 570 560 451 502 533 509/550aa 531 509/550a 520 521
5 µg/ml 5 µg/ml 30 µg/ml 30 µg/ml 1 µg/ml 1 µg/ml 30 nM 30 nM 30 nM 30 nM 1/1000 1/10,000
0.1 mg/ml 0.1 mg/ml 1 mg/ml 1 mg/ml 0.1 mg/ml 0.1 mg/ml 1 µM 1 µM 1 µM 1 µM 1/10 1/100
aNumber before slash is according to manufacturer; number after slash is according to Hirons et al. (1994). bNA, molecular weight not available from manufacturer.
Microbiological Applications
Contributed by Dominique Marie, Nathalie Simon, Laure Guillou, Frédéric Partensky, and Daniel Vaulot
11.12.1
Current Protocols in Cytometry (2000) 11.12.1-11.12.14 Copyright © 2000 by John Wiley & Sons, Inc.
Supplement 11
itself, genome size can be used to discriminate species that have similar morphological features, a common occurrence in both nano- and picoplankton (Partensky and Vaulot, 1989; Boucher et al., 1991; Simon et al., 1994; Vaulot et al., 1994; Veldhuis et al., 1997). Determining ploidy level in cultured populations is also critical for assessing the sexual cycle of eukaryotic microalgae (Vaulot et al., 1986; Le Gall et al., 1993). Nucleic acid stains are useful for assessing the genome size of algal cells, but no universal staining procedure exists because of the large number of nucleic acid dyes and the wide taxonomic diversity of algae. The intercalary dyes, such as propidium iodide or ethidium bromide, bind with both RNA and DNA and allow precise measurement of nucleic acid content. However, they interfere with the emission spectrum of chlorophyll, and thus cannot be used directly on whole cells. The G-C-specific dyes mithramycin and chromomycin A are excited by violet wavelengths (453 to 457 nm) and are often helpful for marine species. The UV-excited dyes, such as DAPI or Hoechst 33342, are currently used for the study of A-T-rich species. In addition to genome size they can be used along with the G-C-specific stains for the determination of the percent G-C (Le Gall et al., 1993; Simon et al., 1994). A new generation of dyes made available by Molecular Probes bind both DNA and RNA and present the advantages of high quantum yield and of excitation at 488 nm, a wavelength available on small flow cytometers equipped with an air-cooled laser. Some of them, such as TOTO-1, YOYO-1, TO-PRO-1, YO-PRO-1, or PicoGreen, are sensitive to the ionic strength and composition of seawater. To deal with this problem, samples can be diluted in low hypotonic buffer in order to minimize the interaction of culture medium or seawater (Marie et al., 1996). If samples cannot be diluted, the most sensitive dye usable for the cell cycle analysis of photosynthetic microorganisms is SYBR Green I, which has a much stronger affinity for double-stranded DNA than for RNA (Haugland, 1996; Marie et al., 1997), and is not sensitive to ionic strength. Materials Cell suspension (1–5 × 105 cells/ml) Nucleus isolation buffer (NIB; see recipe; add 5 µl of 1 M sodium bisulfite per ml buffer immediately before use and filter through 0.2-µm pore-size filter) Internal reference (i.e., suspension of nuclei or cells from species for which genome size is known, at ∼1 × 105 cells/ml) 1% (w/v) RNase A (type IA, Sigma) in distilled water (heat 10 min at 90°C to degrade contaminating DNase) Nucleic acid–specific stain (see Table 11.12.1 for concentrations) 0.2-µm pore-size filters for plastic syringes 10 µm nylon mesh Flow cytometer with laser emitting UV, violet, and blue lines (also see Critical Parameters) 1. Add 10 to 50 µl cell suspension to 1 ml nucleus isolation buffer (NIB). The nuclei will be released by hypotonic shock and will remain stable in NIB. If isolated nuclei cannot be obtained, fix the culture with 1% paraformaldehyde.
2. Add 10 µl internal reference and mix the sample by vortexing.
DNA/RNA Analysis of Phytoplankton by Flow Cytometry
The measurement of DNA content for an uncharacterized marine species is obtained by comparing the mean fluorescence of this species with that of a species for which the genome size is known (Fig. 11.12.1). This reference must be added before staining. If one wants to compare two strains to determine whether they have the same genome size, it is critical to work with a mixture of the strains as well as with each strain separately.
11.12.2 Supplement 11
Current Protocols in Cytometry
B
A
2C
Frequency (AU)
Proc
E. coli Proc
E. coli 1C
1C 2C 1C
4C
2C 1C 0
100 200 DNA (linear scale)
100
101
102 103 DNA (log scale)
104
Figure 11.12.1 Estimation of the genome size of the photosynthetic prokaryote Prochlorococcus (Proc), using a culture of E. coli pretreated with rifampicin as internal reference. Data are presented on both linear (A) and logarithmic (B) scales.
Red fluorescence
A
B 104
104
103
103
10 2
10 2
Proc
101 100 100
ProcG1
ProcG2
101
Bact 101 10 2 103 DNA (logarithmic)
Bact 100 104 100
101
10 2 103 DNA (linear)
104
Figure 11.12.2 Green versus red fluorescence obtained for a sample collected at 65 m in the Pacific Ocean (5°S by 150°W) during the OLIPAC cruise, and stained with SYBR Green I. The green fluorescence signal was recorded on both logarithmic (A) and linear (B) scales.
3. If fixed samples are used, pretreat samples with RNase A: Add 10 µl of 1% RNase A for every 1 ml of sample and incubate for 30 min at 37°C. 4. Filter through 10 µm nylon mesh to eliminate cellular debris or cell clumping. 5. Add the nucleic acid-specific stain (Table 11.2.1). Incubate 15 to 30 min on ice, in the dark. 6. Set up the flow cytometer using distilled water as the sheath fluid. Set the discriminator for the fluorescence corresponding to the dye used. Measurement of DNA content is performed with both linear and logarithmic amplification (Fig. 11.12.2).
7. Run the sample. Collect both linear and log fluorescence.
Microbiological Applications
11.12.3 Current Protocols in Cytometry
Supplement 11
BASIC PROTOCOL 2
CELL CYCLE ANALYSIS OF PHYTOPLANKTON Flow cytometry has been used extensively in the past to determine the cell cycle of different phytoplanktonic species in culture, and more recently in the field. In addition to providing a basic understanding of the relationships between cell cycling and environmental factors such as nutrient levels or light (Vaulot et al., 1994), the determination of the percentage of cells within the different phases of the cell cycle also provides a very elegant way to assess the division rate of phytoplankton in oceanic waters (Carpenter and Chang, 1988; Vaulot et al., 1995). Cell cycle analysis of marine species is performed on fixed cells. Materials 0.05% (w/v) glutaraldehyde/1% (w/v) paraformaldehyde Cell suspension (1–5 × 105 cells/ml) 1% (w/v) RNase A (type IA, Sigma) in distilled water (heat 10 min at 90°C to degrade contaminating DNase) 1 M potassium citrate SYBR Green I working solution (dilute commercial stock solution from Molecular Probes 1:100 in distilled water) 1× 105 beads/ml suspension of 0.95-µm fluorescent microspheres (Polysciences) Flow cytometer equipped with a laser emitting at 488 nm (also see Critical Parameters) 1. Fix the sample by adding 100 µl of 0.05% glutaraldehyde/1% paraformaldehyde to 900 µl of cell suspension. Incubate for up to 15 min. If samples cannot be run immediately after fixation, they must be deep frozen in liquid nitrogen and stored at −80°C for delayed analysis. Frozen samples should be thawed at 37°C.
2. Add 10 µl of 1% RNase A for every 1 ml of sample. Incubate 30 min at 37°C.
%G1: 62.2 %S: 10.7 %G2: 21.1
Frequency (AU)
G1-like
CV G1 : 8.9
G2-like S-like
0
DNA/RNA Analysis of Phytoplankton by Flow Cytometry
DNA (linear scale)
255
Figure 11.12.3 Cell cycle analysis using MultiCycle, on a natural sample of Prochlorococcus collected in the Pacific Ocean (Cast 94, 65 m). The coefficient of variation of the G1-like peak is an indicator of the quality of the DNA staining.
11.12.4 Supplement 11
Current Protocols in Cytometry
3. Add 10 µl SYBR Green I working solution (final concentration corresponds to a 10,000-fold dilution of the commercial solution) and 30 µl 1 M potassium citrate. Incubate 15 min at room temperature in the dark. 4. Add 10 µl of a 1 × 105 beads/ml suspension of 0.95-µm fluorescent microspheres per 1 ml of sample. 5. Set up the flow cytometer using either distilled water or 0.2-µm-filtered sea water as sheath fluid. 6. Connect a T-connector to the output of the PMT collecting green fluorescence. Reconnect the analog-to-digital converters (ADCs) for green and orange to the connector. Cell cycle analysis of phytoplankton must be performed with both logarithmic and linear amplifications (Fig. 11.12.2). The logarithmic scale is used to discriminate the populations of interest. The linear scale is required to perform cell cycle analysis (Fig. 11.12.3). On flow cytometers such as the FACS series, it is not possible to record both signals. To overcome this difficulty, a T-connector is set on the output of the green (FL1) photomultiplier (PMT); then the output of the orange PMT is disconnected. The analog-to-digital converters (ADC) of the green and orange are then reconnected to the T-connector. The green ADC is used to collect the logarithmic signal, and the orange ADC to collect the linear signal of the fluorescence of the DNA–SYBR Green I complex.
7. Put the discriminator on the red fluorescence. Typical settings on the authors’ FACSort flow cytometer are FSC = E01, SSC = 450, FL1 = 660, FL2 (linear amplification X10), FL3 = 650 for Procholorococcus cell cycle analysis.
8. Run the sample for 4 min at low rate (<50 µl/min). IDENTIFYING PHYTOPLANKTON USING WHOLE-CELL HYBRIDIZATION WITH TAXON-SPECIFIC FLUORESCENT OLIGONUCLEOTIDE PROBES
BASIC PROTOCOL 3
The vast amount of sequence data obtained during the last 10 years for the small subunit of ribosomal RNA has led to the development of signature sequences as a phylogenetic determinative tool in microbial ecology. With this method, target species are detected and their phylogenetic position is determined by the binding of a fluorescently labeled oligonucleotide probe to a homologous RNA sequence in the cells. The use of flow cytometry to determine the hybridization signal allows both quantification of the intensity of the hybridization signal and rapid screening of many isolates. Materials Phytoplanktonic cells of interest Culture medium for marine phytoplankton (sea water with appropriate nutrients; see http://ccmp.bigelow.org/CI/CI_01.html) 10% (w/v) paraformaldehyde stock (store at −20°C) 70:30 (v/v) ethanol/PBS, −20°C (see APPENDIX 2A for PBS) Hybridization buffer (see recipe) of optimal formamide concentration 50 ng/µl fluorescein isothiocyanate (FITC)–labeled probe to be tested (see Critical Parameters and UNIT 8.3) Washing buffer (see recipe) with the same stringency as the hybridization buffer Phosphate-buffered saline (PBS; APPENDIX 2A), pH 9.0 1 × 105 beads/ml suspension of 0.95-µm fluorescent microspheres (Polysciences) Hybridization oven, or any oven or incubator whose temperature can be precisely tuned Flow cytometer equipped with 488-nm argon laser (also see Critical Parameters)
Microbiological Applications
11.12.5 Current Protocols in Cytometry
Supplement 11
Fix and permeabilize cells 1. Centrifuge the phytoplanktonic cells, selecting appropriate time and g-force for the species used. Resuspend the cell pellet in a small volume (1 ml) of fresh culture medium for marine phytoplankton. Final concentration should be at least 107 cells/ml. Centrifugation may lead to cell loss; therefore appropriate centrifugation conditions have to be tested initially for each species.
2. Add 10% paraformaldehyde stock (fresh or freshly thawed) to a final concentration of 1% and incubate 1 hr on ice. 3. Vortex cells and centrifuge at 4°C using the appropriate conditions for the cells of interest. 4. Resuspend the cell pellet in cold (–20°C) ethanol:PBS. Care should be taken from this stage on when centrifuging and resuspending cells. To prevent cell loss, pellets should not be disturbed.
5. Vortex vigorously and centrifuge at 4°C. Vortexing is important in order to prevent cell aggregation.
Hybridize and wash cells 6. Resuspend the cells in hybridization buffer preheated at 46°C (or other optimal temperature; see Critical Parameters). Vortex vigorously and split the cell suspension into 20-µl aliquots. See Critical Parameters for discussion of how to optimize the stringency (formamide concentration) of the hybridization buffer.
7. For each aliquot, add 1 µl of the labeled probe to be tested (working stock at 50 ng/µl). For each experiment, two controls should be prepared. One (autofluorescence) consists of a cell suspension with no probe. The other (negative control) consists of cells hybridized with a non-target probe (non-specific staining). For example, Chlamydomonas concordia hybridized with the probe CHLO02 is 4 times more fluorescent than with the PRYM02 probe. Conversely, the intensity of green fluorescence of Prymnesium patelliferum is 4 times higher when hybridized with the PRYM02 probe than with the CHLO02 probe (Fig. 11.12.4).
8. Incubate 3 hr at 46°C (or optimal temperature). 9. Resuspend cells in 50 µl preheated (46°C or optimal temperature) washing buffer. 10. Incubate 15 min at 46°C (or optimal temperature). 11. Stop hybridization by adding 500 µl ice-cold PBS, pH 9.0. Keep samples on ice in the dark until analysis. Analyze cells by flow cytometry 12. Add 10 µl of a 1 × 105 beads/ml suspension of 0.95-µm fluorescent microspheres per 1 ml of sample. 13. Set up flow cytometer. Use forward light scatter and side scatter to look at size and refractive index distribution of the cells as well as red fluorescence to look at the chlorophyll fluorescence intensity after cell treatment. Collect all parameters, including green fluorescence, using logarithmic amplification. DNA/RNA Analysis of Phytoplankton by Flow Cytometry
Discrimination of the cells from particles will be obtained by setting the discriminator on the red fluorescence photomultiplier. After treatment, chlorophyll-containing cells will retain some red fluorescence, even though chlorophyll is degraded.
11.12.6 Supplement 11
Current Protocols in Cytometry
Green fluorescence
A
Green fluorescence
no probe
103
C
104
104
PRYM02
103
102
C. concordia
CHLO02 103
C. concordia
102
102
101
101
C. concordia
101 100
D
B
104
100
100 100
101
102
103
100
104
101
102
103
104
no probe
103
102
103
104
104
104
PRYM02
CHLO02
103
Beads
101
F
E
104
100
103
Beads
Beads P. patelliferum
102
P. patelliferum
102
P. patelliferum
101
101
101 100 100
102
101
102
103
100 100
104
101
102
103
104
100 100
101
102
103
104
Red fluorescence
Figure 11.12.4 Cytograms showing the distribution of Chlamydomonas concordia (Chlorophyta; A, B, and C) and Prymnesium patelliferum (Prymnesiophyceae; D, E, and F) hybridized with no probe (A and D), the probes PRYM02 (specific for the class Prymnesiophyceae; B and F), and CHLO02 (specific for the division Chlorophyta; C and E). Note the presence of cell aggregates for Chlamydomonas concordia. This may be avoided by intense vortexing in the hybridization buffer prior to incubation.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Hybridization buffer 0.9 M NaCl 20 mM Tris⋅Cl, pH 7.2 0.01% (w/v) sodium dodecyl sulfate (SDS) 10% to 50% deionized formamide Store up to 1 year at −20°C See Critical Parameters for optimization of formamide concentration (stringency).
Nucleus isolation buffer (NIB) 30 mM MgCl2 20 mM sodium citrate 120 mM sorbitol 55 mM HEPES 5 mM EDTA 0.1% (v/v) Triton X-100 Adjust pH to 7.5
continued
Microbiological Applications
11.12.7 Current Protocols in Cytometry
Supplement 11
Store at −20°C Add 5 ìl 1 M sodium bisulfite per ml buffer just before the experiment (sodium bisulfite is not stable over time)
Washing buffer Prepare 20 mM Tris⋅Cl, pH 7.2 (APPENDIX 2A) containing 0.01% (w/v) SDS and a concentration of NaCl which determines the stringency. Prepare washing buffer with the same stringency as the hybridization buffer (see recipe above). For hybridization buffers with 0%, 10%, 20%, 30%, 40%, and 50% formamide, respectively, prepare washing buffers with 0.9, 0.45, 0.225, 0.112, 0.056, and 0.028 M NaCl. Prepare the wash buffer fresh before use.
COMMENTARY Background Information DNA quantification methodology has benefited a great deal from the combined advent of DNA stain technology and flow cytometry. This combination of methods is now routinely applied in such diverse fields as medicine (e.g., screening of cancer cells) and plant biology (e.g., determination of the ploidy of seeds). In phytoplankton research, the determination of DNA content and GC percentage may be useful too as characterization tools (see Anticipated Results). Another potential application of flow cytometry is in the study of phytoplankton dynamics. By measuring the distribution of cells in the different phases of the cell cycle at regular intervals during a light-dark cycle, it is possible to compute the growth rate of these populations (Carpenter and Chang, 1988). This method applies to natural populations without the need to incubate cells, a method which is prone to biases known as the “bottle effect.” Although DNA quantification offers much potential, the use of molecular tools appears by far to be the most promising approach in microbial ecology. The specificity of molecular probes can be modulated at will to target any taxon level from kingdom to species, offering the possibility of accurate identification of populations in complex samples (for review, see Amann et al,. 1995).
Critical Parameters
DNA/RNA Analysis of Phytoplankton by Flow Cytometry
Data analysis Data are acquired as listmode files and are analyzed with the CytoWin software (Vaulot, 1989; available at http://www.sb-roscoff.fr/ Phyto/cyto.html) to obtain the monoparametric distribution of the population of interest. Cell cycle analyses are then performed on the linear green signal using MultiCycle (P. Rabinovitch)
that will fit the linear DNA distribution (Fig. 11.12.3) and allow one to determine the percentage of cells in the different phases of the cell cycle: G1, S, and G2 (prokaryotes) or G2M (eukaryotes). For in situ hybridization, the ratio of the mean fluorescence of cells hybridized with a specific probe to the mean fluorescence of cells hybridized with a nonspecific probe is computed to evaluate probe specificity and sensitivity. Genome analysis DNA content is best assessed on fresh material. The nuclei are released by hypotonic shock, and can remain stable for at least one hour in nucleus isolation buffer (NIB). Depending on the species of interest, the composition of the NIB may need to be optimized. For example, addition of extra citrate gives isolated nuclei with more condensed chromatin. Increasing concentrations of detergent may help to remove membranes or cytoplasmic material attached to nuclei, which induce background fluorescence. Examination of nuclei by epifluorescence microscopy will help the operator to determine the best conditions. With some eukaryotic algae, isolated nuclei cannot be obtained by this method because of the complex composition of membranes (e.g., for diatoms) or because nuclei are not stable and degrade very quickly. In such cases, the DNA content can be estimated on whole cells after treatment with alcohol (ethanol or methanol) to remove fluorescent pigments, followed by incubation with RNase. However, the presence of contaminating DNA (mitochondrial and chloroplastic) often leads to inaccuracy of genome measurement. For prokaryotes, which have no nuclei, DNA content can be assessed after fixation. The use of paraformaldehyde is recommended.
11.12.8 Supplement 11
Current Protocols in Cytometry
Choice of the internal reference The DNA content is measured by comparing the mean DNA fluorescence of the species to that of a standard. Many parameters can influence staining of isolated nuclei, such as the temperature of incubation, the stage of the culture, the concentration of chemicals like detergent or citrate, or the presence of seawater components (e.g., when samples are not concentrated enough to be diluted). The staining of the internal reference will be affected by these parameters in a way similar to that of the sample. This precludes the use of inert material such as fluorescent microspheres as the standard. The choice of the internal reference will also depend on the species to be measured. If the species and the standard have genome sizes that are too different, this leads to inaccuracy in the determination of the DNA content. However, genome size is known precisely only for a small number of microbial strains that have been entir ely sequenced (e.g., Synechocystis PCC6803 or E. coli). If one needs a precise estimation of the DNA content of cells of interest, this constitutes a major drawback. Chicken red blood cells (CRBC) have been used to assess the genome size of Phaeocystis (Vaulot et al., 1994). However, CRBC have a very high DNA content (2.33 pg DNA/cell) compared to the smallest phytoplanktonic species. For example, Ostreococcus tauri has a genome size of 6.62 fg (10.2 Mbp per cell; Courties et al., 1998). Another problem with CRBC is that they degrade quickly if the samples are not diluted enough to minimize the effect of seawater. When the genome size difference between the sample and the reference is large, one should use the data acquisition on logarithmic scale with the following formula to calculate the DNA content: D = 10
( F − FR ) × N / C
× DR
where D = DNA content of the unknown species; DR = DNA content of the reference species; F = mean channel of the unknown species; FR = mean channel of the reference species; N = number of decades of the logarithmic amplifier; and C = number of channels used for the acquisition (256 or 1024). This formula requires precise determination of the number of decades of the amplifier used for collecting the fluorescence (Durand, 1999). It is critical not to rely on the value given by the manufacturer (2 to 4 decades), which is never accurate.
As an example, the authors have estimated the genome size of Prochlorococcus at 1.9 Mbp by comparing its DNA distribution, after staining with SYBR-I or YOYO-1, with that of E. coli (1 C = 4.66 Mbp) treated with rifampicin (Fig. 11.12.1). This antibiotic inhibits RNA synthesis and blocks the initiation of new replication sites on the DNA, but allows DNA replication to terminate and cell division to occur. As a result, most cells are blocked with a 1C and 2C genome size. The value measured f or Prochlorococcus by flow cytometry matches very well with that established by DNA renaturation kinetics (1.89 Mbp; Rippka et al., submitted). This constitutes a good illustration of the potential use of flow cytometry to evaluate the genome size of small organisms. Coefficient of variation The quality of the cell cycle analysis is strongly dependent upon the coefficient of variation (CV) of the G1 peak of the linear DNA distribution. A high CV (>10%) makes it very difficult to estimate the fraction of cells in the different phases of the cell cycle, especially the S-phase. Broad G1-like peaks result from inadequate fixation procedures or from interaction between the stain and some components of the seawater. This can also be due to a concentration of stain that is either too low or too high, leading, respectively, to nonstoichiometric or nonspecific binding. Moreover, because many planktonic species have a small genome size, the dye concentration must be lowered compared to the quantities used with mammalian or plant cells. The usual practice of filtering stock solutions through 0.2-µm-pore-size filters may induce retention of dye such as SYBR Green I on the filter, thereby changing the concentration of the stock solution. In situ hybridization Limits of the technique: Current focuses in aquatic microbial ecology include the study of the diversity of planktonic organisms (bacteria, phytoplankton, and heterotrophic protists) as well as the structure and dynamic of the communities and populations. Fluorescence in situ hybridization with rRNA-targeted oligonucleotide probes, combined with flow cytometry, is a promising new approach for the characterization of microbial communities. Optimized for the identification of bacteria (Wallner et al., 1993), this approach has been used for the analysis of bacteria in activated sludge (Wallner et al., 1997). It has also been used to sort and further analyze target species
Microbiological Applications
11.12.9 Current Protocols in Cytometry
Supplement 11
DNA/RNA Analysis of Phytoplankton by Flow Cytometry
using molecular methods such as gene amplification and sequencing (Wallner et al., 1997). Probes for eukaryotic phytoplankton have also been designed, and conditions of hybridization have been optimized for flow cytometry (Simon et al., 1995, 1997; Lange, 1996). Basic Protocol 3 may be used for the rapid screening of phytoplankton isolates. However, to date, no application of this method to natural marine phytoplankton communities has proven successful, mainly because the intensity of the hybridization signal is too low to discriminate one phytoplanktonic species from the rest of the community. This is probably due to the fact that cells in the marine environment have a reduced rRNA content, possibly because of suboptimal growth. To overcome this problem, new and brighter fluorochromes would be needed (Schönhuber et al., 1997). Flow cytometer: Better discriminations between target and nontarget cells have been obtained using small flow cytometers, such as the FACSort equipped with air-cooled argon lasers delivering 15 mW, than with larger instruments, such as the Epics 545 equipped with a watercooled laser delivering 1.2 W. The reason for this difference in signal/noise ratio may be linked to the optical configurations of these instruments. Design and ordering of probes: The design of taxa-specific probes requires an extensive knowledge of the phylogeny of the target groups and is subject to a few rules which are reviewed in detail by Stahl and Amann (1991) and Amann (1995). To date, the SSU rDNA gene is the best candidate because (1) it includes signatures for species as well as phyla, (2) its database is very large, and (3) the targets in the cells are relatively abundant (one per ribosome). A few oligonucleotide probes targeted to phytoplankton taxa are now available, among which some were tested using wholecell hybridization combined with flow cytometry (Table 11.12.2). DNA probes may be purchased already labeled (e.g., at the 5′ end with FITC) and purified from commercial companies. Alternatively, labeling and purification can be custom-made following the protocol described by Amann (1995). Choice of species to be used for tests: Probes that are theoretically specific for a targeted group should be empirically tested on cells. At least one target species and one non-target species that is closely related to the target group should be chosen. The rRNA molecule of a
non-target species should present at least one mismatch with the probe; however, more mismatches are better. It is necessary to grow and harvest cells under optimum conditions. To optimize the signal, cells should be harvested in mid-exponential phase. Cells may be stored at −80°C either as a paraformaldehyde-fixed suspension or in the hybridization buffer. Preparation and optimization of hybridization buffers: Hybridization conditions depend upon the dissociation temperature (Td) of the oligonucleotide and must be empirically optimized for each new probe. The temperatures of hybridization and washing buffers are routinely adjusted in order to achieve a good discrimination between target and non-target sequences. For whole-cell hybridization, formamide concentration or ionic strength can be adjusted while temperature is kept constant. The melting point of hybrids is lowered by ∼0.7°C for every 1% increase of formamide concentration in the buffer. With this strategy, optimal conditions may be determined within a day using a single oven and batches of target and non-target cells. Preparation and storage of hybridization and washing buffers: Buffers with varying concentrations of formamide (i.e., 10%, 20%, 30%, 40%, and 50% for hybridization buffers with different stringency) and of NaCl (i.e., washing buffers) may be prepared in advance, divided into aliquots in 1-ml microcentrifuge tubes, and stored at −20°C until needed. Choice of species to be tested: Choose one target species and one non-target species that is most closely related to the target group—i.e., whose 18S rRNA possesses the minimum possible number of mismatches compared to the probe. The rationale is that non-target species with more mismatches will hybridize even less. Tests with different hybridization stringencies: Proceed with the hybridization protocol using the full range of hybridization and corresponding washing stringencies and analyze the cells by flow cytometry. Plot the intensities of green fluorescence of both species against the percentage formamide used for the hybridization buffers. For subsequent use of the probe, choose the percentage formamide that best discriminates species. Ideally, using the optimal buffers, the intensity of fluorescence of the nontarget cells should not exceed the intensity of green autofluorescence of the cells, and target cells should show a relatively bright signal in the green. In practice, a light nonspecific labeling is often observed (Fig. 11.12.4).
11.12.10 Supplement 11
Current Protocols in Cytometry
Table 11.12.2 Oligonucleotide Probes Designed for the Detection and Identification of Phytoplankton Taxa by Whole-Cell Hybridizationa
Specificity Upper group level Eukaryotes
Chlorophyta Chlorophyta “Non-chlorophyta” Class level Pelagophyceae Prymnesiophyceae Genus level Phaeocystis spp. (Prymnesiophyceae) Species level Chrysochromulina polylepis (Prymnesiophyceae) Paraphysomonas vestita (nonphotosynthetic) (Chrysophyceae)
Probes
Oligonucleotide sequence (5′ to 3′)
References
EUK1209R EUK502 EUK309R EUKB CHLO01 CHLO02 NCHLO01
GGGCATCACAGACCTG ACCAGACTTGCCCTCC TCAGGCTCCCTCTCCGG TGATCCTTCTGCAGGTTCACCTAC GCTCCACGCCTGGTGGTG CTTCGAGCCCCCAACTTT GCTGGACTCCTGGTGGTG
Lim et al., 1993 Lim et al., 1993 Lim et al., 1993 Lim et al., 1993 Simon et al., 1995 Simon et al., in press Simon et al., 1995
PELA01 PRYM01 PRYM02
ACGTCCTTGTTCGACGCT ACATCCCTGGCAAATGCT GGAATACGAGTGCCCCTGAC
Simon et al., in press Lange et al., 1996 Simon et al., in press
PHAEO01
CGGTCGAGGTGGACTCGT
Lange et al., 1996
CPOLY01
GACTATAGTTTCCCATAAGGT
Simon et al., 1997
PV1 PV2 PV3 PV4 PV5
TAAAACCCATCCTATTATATC TTCCGTATGCCAGTCAGA AGTATAAATATCACAGTCCGA ATATAATCTTTTCGATGATGA CCCATCCTATTATATCAGAAA
Rice et al., 1997a, b Rice et al., 1997a, b Rice et al., 1997a, b Rice et al., 1997a, b Rice et al., 1997a, b
GGACTTCCGCCGATCCCTAGT
Simon et al., 1997
AGTCGGGTCTTCCTGCATGT
Simon et al., 1997
GCAACAATCAATCCCAATC
Simon et al., in press
Other (lineages, group of species) CLADE1 Chrysochromulina polylepis/Prymnesium parvum/P. patelliferum/P. calathiferum (Prymnesiophyceae) Chrysochromulina species included CLADE2 in a clade (Clade 2) (Prymnesiophyceae) PELA02 Aureococcus, Pelagomonas, Pelagococcus, CCMP 1395 (belonging to the Pelagophyceae)
aAll probes are specific for signature sequences on the SSU rRNA of their target organisms.
Troubleshooting Genome analysis Unstable fluorescence of nuclei (increase or decrease of the mean fluorescence of the isolated nuclei distribution) may be observed. Both increase and decrease of the mean fluorescence result from degradation of nuclei or from the effect of some chemicals that are released after hypotonic shock. An epifluorescence microscope is useful to determine if nuclei are degraded, and in such a case, the composition of the buffer must be optimized to
stabilize the chromatin. To deal with the effect of chemicals, the presence of an internal reference is required as mentioned previously. In situ hybridization Potential problems include the following. 1. No cells visible. The first possibility here is that the cells were lost during processing. The protocol requires several steps of spinning and resuspending cells. When removing supernatants, be sure not to disturb the pellet. The second possibility is that the signal of the cells Microbiological Applications
11.12.11 Current Protocols in Cytometry
Supplement 11
is lost within the noise. In order to avoid problems, always filter the buffers to be used. 2. Low signal intensity. Make sure that the cells are in mid-exponential phase before harvesting. The signal intensity is strongly dependent on the ribosome content, which drops by an order of magnitude in stationary phase.
Anticipated Results Genome size and ploidy level Very little information is currently available concerning the genome size and GC percentage of phytoplankton species, although these parameters are very useful for characterizing these cells when more classical parameters such as morphology are not discriminative enough. For instance, by determining the genome size of 16 cultured strains of Phaeocystis spp. Lagerheim by flow cytometry, Vaulot et al. (1994) have defined 6 different clusters. Similarly, using different DNA-specific dyes, Simon et al. (1994) have been able to discriminate several eukaryotic strains by flow cytometric measurement of their DNA content and GC percentage (Simon et al., 1994). Cell cycle The analysis of the cell cycle of marine species was first designed to study the dynamics of dinoflagellate cells (Chang and Carpenter, 1991). Another nice application of this technique was the determination of the growth rate of Prochlorococcus at different depths of the equatorial and tropical Pacific (Vaulot et al., 1995; Liu et al., 1997). More recently, this technique was used to study the effects of the variations in cloud cover on Synechococcus populations in the surface waters of a Mediterranean bay (Jacquet et al., 1998). In situ hybridization The combination of whole-cell hybridization with flow cytometry was first tested on cultured bacteria (Amann et al., 1990; Wallner et al., 1993) and later applied to bacteria in natural samples (Pernthaler et al., 1997; Wallner et al., 1997; Shönhuber et al., 1999). The same technique has been applied to detect phytoplankton phylogenetic groups (from the division to the species level), and especially the smallest species in culture (the photosynthetic nano- and picoeukaryotes; Lange et al., 1996; Simon et al., 1995, 1997, in press).
Time Considerations From the harvesting of the live cells to the analysis with flow cytometry, 5 hr are needed (including 3.25 hr of incubation), provided that buffers are ready to use.
Literature Cited Amann, R.I. 1995. In situ identification of microorganisms by whole cell hybridization with rRNAtargeted nucleic acid probes. In Molecular Microbial Ecology Manual (A.D.L. Akkermann, J.D. van Elsas, and F.J. de Bruijn, eds.) pp. 1-15. Kluwer Academic Publishers, Dordrecht, The Netherlands. Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R., and Stahl, D.A. 1990. Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56:1919-1925. Amann, R.I., Ludwig, W., and Schleifer, K.H. 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59:143-169. Boucher, N., Vaulot, D., and Partensky, F. 1991. Flow cytometric determination of phytoplankton DNA in cultures and oceanic populations. Mar. Ecol. Prog. Ser. 71:75-84. Brown, S.C., Bergounioux, C., Tallet, S., and Marie D. 1991. Flow cytometry of nuclei for ploidy and cell cycle analysis In Practical Guide to Plant Cellular and Molecular Techniques (I. Negrutiu and G. Gharti-Chhertri, eds.) pp. 326-345. Birkhaüser, Basel. Carpenter, E.J. and Chang, J. 1988. Species-specific phytoplankton growth rates via diel DNA synthesis cycles: Concept of the method. Mar. Ecol. Prog. Ser. 43:105-111. Chang, J. and Carpenter, E.J. 1991. Species-specific phytoplankton growth rates via diel DNA synthesis cycles. V. Application to natural populations in Long Island Sound. Mar. Ecol. Prog. Ser. 78:115-122. Courties, C., Perasso, R., Chrétiennot-Dinet, M-J., Gouy, M., Guillou, L., and Trousselier, M. 1998. Phylogenetic analysis and genome size of Ostreococcus tauri (Chlorophyta, Prasinophyceae). J. Phycol. 34:844-849. Durand, R.E. 1999. Calibration of flow cytometer detector system. In Flow Cytometry (Z. Darzynkiewicz, J.P. Robinson, and H.A. Crissman, eds.) pp. 647-654. Academic Press, San Diego. Galbraith, D.W., Harkins, K.R., Maddox, J.M., Ayres, N.M., Sharma, D.P., and Firoozabady, E. 1983. Rapid flow cytometric analysis of the cell cycle in plant tissues. Science 220:1049-1051. Haugland, R.P. 1996. Handbook of Fluorescent Probes and Research Chemicals. 6th Edition. Molecular Probes, Inc., Eugene, Oreg.
DNA/RNA Analysis of Phytoplankton by Flow Cytometry
11.12.12 Supplement 11
Current Protocols in Cytometry
Hirons, G.T., Fawcett, J.J., and Crissman, H.A. 1994. TOTO and YOYO: New very bright fluorochromes for DNA content analysis by flow cytometry. Cytometry 15:129-140. Jacquet, S., Lennon, J.F., Marie, D., and Vaulot, D. 1998. Picoplankton population dynamics in coastal waters of the northwestern Mediterranean Sea. Limnol. Oceanogr. 43:1916-1931. Lange, M., Guillou, L., Vaulot, D., Simon, N., Amann, R.I., Ludwig, W., and Medlin, L.K. 1996. Identification of the class Prymnesiophyceae and the genus Phaeocystis with ribosomal RNA-targeted nucleic acid probes detected by flow cytometry. J. Phycol. 32:858-868. Le Gall, Y., Brown, S., Marie, D., Mejjad, M., and Kloareg, B. 1993. Quantification of nuclear DNA and G-C content in marine macroalgae by flow cytometry of isolated nuclei. Protoplasma 173:123-132. Lim, E.L., Amaral, L.A., Caron, D.A., and DeLong, E.F. 1993. Application of rRNA-based probes for observing marine nanoplanktonic protists. Appl. Environ. Microbiol. 59:1647-1655. Liu, H.B., Nolla, H.A., and Campbell, L. 1997. Prochlorococcus growth rate and contribution to primary production in the equatorial and subtropical North Pacific Ocean. Aquat. Microb. Ecol. 12:39-47. Marie, D., Vaulot, D., and Partensky, F. 1996. Application of the novel nucleic acid dyes YOYO-1, YO-PRO-1 and PicoGreen for flow cytometric analysis of marine prokaryotes. Appl. Environ. Microbiol. 62:1649-1655. Marie, D., Partensky, F., Jacquet, S., and Vaulot, D. 1997. Enumeration and cell cycle analysis of natural populations of marine picoplankton by flow cytometry using the nucleic acid stain SYBR Green-I. Appl. Environ. Microbiol. 93:186-193. Partensky, F. and Vaulot, D. 1989. Cell size differentiation in the bloom-forming dinoflagellate Gymnodinium cf. nagasakiense. J. Phycol. 25:741-750. Pernthaler, J., Alfreider, A., Posch, T., Andreatta, S., and Psenner, R. 1997. In situ classification and image cytometry of pelagic bacteria from a high mountain lake (Gossenköllesee, Austria) Appl. Environ. Microbiol. 63:4778-4783. Rice, J., O’Connor, C.D., Sleigh, M.A., Burkill, P.H., Giles, I.G., and Zubkov, M.V. 1997a. Fluorescent oligonucleotide rDNA probes that specifically bind to a common nanoflagellate, Paraphysomonas vestita. Microbiology 143:17171727. Rice, J., O’Connor, C.D., Sleigh, M.A., Burkill, P.H, Tarran G.A., O’Connor, C.D., and Zubkov, M.V. 1997b. Flow cytometric analysis of characteristics of hybridization of species-specific fluorescent oligonucleotide probes to rRNA of marine nanoflagellates. Appl. Environ. Microbiol. 63:938-944.
Rippka, R., Coursin, T., Lichtlé, C., Partensky, F., Houmard, J., and Herdman, M. Prochlorococcus sp. PCC 9511, an axenic, marine, chlorophyll b-containing oxyphotobacterium. Int. J. Syst. Bacteriol. Submitted for publication. Schönhuber, W., Fuchs, B., Juretschko, S., and Amann, R. 1997. Improved sensitivity of wholecell hybridization by the combination of horseradish peroxidase-labeled oligonucleotides and tyramide signal amplification. Appl. Environ. Microbiol. 63:3268-3273. Schönhuber, W., Zarda, B., Eix, S., Rippka, R., Herdman, M., Ludwig, W., and Amann, R. 1999. In situ identification of cyanobacteria with horseradish peroxidase-labeled, rRNA-targeted oligonucleotide probes. Appl. Environ. Microbiol. 65:1259-1267. Simon, N., Campbell, L., Örnolfsdottir, E., Groben, R., Guillou, L., Lange, M., and Medlin, L.K. Oligonucleotide probes for the identification of three algal groups by dot blot and fluorescent whole-cell hybridization. J. Eukaryot. Microbiol. In press. Simon, N., Barlow, R.G., Marie, D., Partensky, F., and Vaulot, D. 1994. Flow cytometry analysis of oceanic photosynthetic picoeucaryotes. J. Phycol. 30:922-935. Simon, N., Lebot, N., Marie, D., Partensky, F., and Vaulot, D. 1995. Fluorescent in situ hybridization with rRNA-targeted oligonucleotide probes to identify small phytoplankton by flow cytometry. Appl. Environ. Microbiol. 61:25062513. Simon, N., Brenner, J., Edvardsen, B., and Medlin, L.K. 1997. The identification of Chrysochromulina a nd Prymnesium species (Haptophyta, Prymnesiophyceae) using fluorescent or chemiluminescent oligonucleotide probes: A means for improving studies on toxic algae. Eur. J. Phycol. 32:393-401. Stahl, D.A. and Amann, R.I. 1991. Development and application of nucleic acid probes in bacterial systematics In Nucleic Acid Techniques in Bacterial Systematics (E. Stackebrandt and M. Goodfellow, eds.) pp. 205-248. John Wiley & Sons Ltd, Chichester, England. Vaulot, D. 1989. CYTOPC: Processing software for flow cytometric data. Signal and Noise 2:8. Vaulot, D., Olson, R.J., and Chisholm, S.W. 1986. Light and dark control of the cell cycle in two phytoplankton species. Exp. Cell. Res. 167:3852. Vaulot, D., Birrien, J-L., Marie, D., Casotti, R., Veldhuis, M.J.W., Kraay, G.W., and ChrétiennotDinet, M-J. 1994. Morphology, ploidy, pigment composition and genome size of cultured strains of Phaeocystis (Prymnesiophyceae). J. Phycol. 30:1022-1035. Vaulot, D., Marie, D., Olson, R.J., and Chisholm, S.W. 1995. Growth of Prochlorococcus, a photosynthetic prokaryote, in the equatorial Pacific Ocean. Science 268:1480-1482. Microbiological Applications
11.12.13 Current Protocols in Cytometry
Supplement 11
Veldhuis, M.J.W., Cucci, T.L., and Sieracki, M.E. 1997. Cellular DNA content of marine phytoplankton using two new fluorochromes: Taxonomic and ecological implications. J. Phycol. 33:527-541. Wallner, G., Amann, R., and Beisker, W. 1993. Optimizing fluorescent in situ hybridization with rRNA-targeted oligonucleotide probes for flow cytometric identification of microorganisms. Cytometry 14:136-143.
Wallner, G., Fuchs, B., Spring, S., Beisker, W., and Amann, R. 1997. Flow sorting of microorganisms for molecular analysis. Appl. Environ. Microbiol. 63:4223-4231.
Contributed by Dominique Marie, Nathalie Simon, Laure Guillou, Frédéric Partensky, and Daniel Vaulot INSU et Université Pierre et Marie Curie Roscoff, France
The authors wish to acknowledge support by grants from the European Community MAST III programs (MAS3-CT96-015-MEDEA, MAS3-CT97-0128PROMOLEC), and from the JGOFS-France PROSOPE program. The FACSort flow cytometer was funded in part by CNRS-INSU and the Région Bretagne.
DNA/RNA Analysis of Phytoplankton by Flow Cytometry
11.12.14 Supplement 11
Current Protocols in Cytometry
Cell Cycle Analysis of Yeasts
UNIT 11.13
CELL CYCLE ANALYSIS OF YEASTS USING SYBR GREEN I Flow cytometry measurement of fractions of cycling cells which are at a certain point in the life of many higher eukaryotic cell populations is relatively well standardized. More recently, flow cytometry has also been applied to the study of the yeast cell cycle. However, the flow cytometry–DNA staining protocols presently available for cell cycle analysis in yeasts do not provide satisfactory resolution and quantification of the different phases of the cycle, particularly the S phase.
BASIC PROTOCOL
The accuracy of DNA content measurements and estimation of the percentage of cells in the different compartments of the cell cycle are limited by the coefficient of variation (CV) of DNA measurements. Most DNA staining protocols for yeasts achieve only low-resolution DNA measurements (i.e., CVs of ∼9% to 10% or higher). This unit provides a staining protocol using the sensitive nucleic acid dye SYBR Green I, which generates high-resolution DNA histograms and therefore significantly improves the analyses of the yeast cell cycle. With this protocol, high-resolution DNA distributions, i.e., measures with half-peak CVs of ∼3% to 4%, were obtained for two species: Saccharomyces cerevisiae (haploid strain W303 1A) and Zygosaccharomyces bailii (ISA 1307). Eventually, identical results should be obtained for most yeast species, with minor adaptations. Development of this DNA staining protocol was based on the following criteria: (1) use of SYBR Green I, a fluorophore which binds to dsDNA with very high selectivity and which has a much higher fluorescence quantum yield upon DNA binding than do most commonly used fluorophores; (2) ethanol fixation of cells, which neither disturbs the stoichiometry of DNA/SYBR Green I complexes nor interferes with the quality of the staining, and which allows samples prepared at widely different times to be assayed all at once; (3) elimination of dsRNA with an RNase treatment; (4) decrease or elimination of protein precipitates in several cell compartments with proteinase K treatment; (5) increase of cell permeability by Triton X-100 treatment of cells; (6) elimination of most cell clumps with a light/brief sonication; and (7) use of microspheres as internal references. The properties of SYBR Green I in combination with an optimized prestaining processing of the yeast sample allow DNA analysis with higher accuracy than that previously obtained with other stains. DNA distributions with G0/G1 and G2/M peaks, with half-peak CVs of ∼3% to 4%, and with a clear-cut identification of the S phase are consistently and easily obtained. Microspheres added to the yeast cell suspensions after the staining step give consistent fluorescence ratios relative to cells. Therefore, these indexes can be used to control, on the one hand, the reproducibility of DNA measures, and on the other hand, eventual changes in the total cell DNA content and/or the cell cycle kinetics. Materials Yeast cell suspension 70% (v/v) ethanol 50 mM sodium citrate buffer, pH 7.5 (see recipe) 1 mg/ml RNase A solution (see recipe) 20 mg/ml proteinase K in water (Boehringer Mannheim; store in aliquots at 4°C) SYBR Green I working solution (see recipe) Triton X-100 continued Microbiological Applications Contributed by Margarida Fortuna, Maria João Sousa, Manuela Côrte-Real, Cecília Leão, Alexandre Salvador, and Filipe Sansonetty
11.13.1
Current Protocols in Cytometry (2000) 11.13.1-11.13.9 Copyright © 2000 by John Wiley & Sons, Inc.
Supplement 13
Two sets of fluorescent microspheres, with nominal sizes and green fluorescence, giving scatter and fluorescence signals noncoincident with those of the cells of interest and with CVs ≤2.0% 50°C water bath Ultrasonic processor with a probe to sonicate small volumes of liquid (e.g., Jencons model GE50) Flow cytometer with 488-nm excitation and a band-pass filter centered at 525 nm Additional reagents and equipment for counting cells using a hemacytometer (APPENDIX 3A)
Cells 102
MB
101
MA
100
Log forward scatter
103
104
A
100
101
256
B
102 103 Log side scatter
104
G0G1
Events
Fl I = 2.5 MA
G2 /M
Fl I = 4.5 MB
0
S
0 1023 Relative green fluorescence (area)
Cell Cycle Analysis of Yeasts
Figure 11.13.1 (A) Scattergram (contour plot) for Saccharomyces cerevisiae (haploid strain W303 1A) after SYBR Green I staining and mixing with two different sets of microspheres, MA and MB. (B) Green fluorescence histogram for the same mixture of Saccharomyces cerevisiae W303.1A, after SYBR Green I staining, with microspheres MA (nominal size 1.98 µm) and MB (nominal size 3.15 µm). Total number of events measured = 30,000. Fl I = fluorescence index (mean fluorescence of microspheres/mean fluorescence of G0/G1 cells). Green fluorescence of microspheres was not affected by SYBR Green I.
11.13.2 Supplement 13
Current Protocols in Cytometry
Fix yeast cell suspensions 1. Harvest cells by centrifugation, selecting the appropriate temperature, time, and g-force for the cell species/strain studied (for example, see UNIT 11.10). 2. Wash twice in ice-cold deionized, distilled water, centrifuging again each time as in step 1. Remove supernatant and resuspend in deionized distilled water. 3. Count cells using a hemacytometer (APPENDIX 3A) and adjust cell concentration to 1 × 107 yeast cells/ml in deionized, distilled water. 4. Divide cell suspension into an appropriate number of 1-ml aliquots in 1.5-ml microcentrifuge tubes and centrifuge again as in step 1. Remove supernatant. 5. Fix cells by resuspending in 1 ml 70% ethanol overnight, 4°C. Cell number must be rigorously maintained in each tube. Fixed cells can be stored in the ethanol solution at 4°C for several weeks without affecting the quality of DNA staining.
Stain with SYBR Green I 6. Centrifuge, wash with 50 mM sodium citrate buffer, pH 7.5, and centrifuge the cells again at room temperature, as in step 1. 7. Remove the supernatant and resuspend cells in 750 µl of 50 mM sodium citrate buffer, pH 7.5. 8. Add 250 µl of 1 mg/ml RNase A solution (i.e., 250 µg enzyme per 1 × 107 yeast cells). Incubate 1 hr at 50°C. Samples treated with RNase A show a slight improvement in the CVs of the G0/G1 peak.
9. Add 50 µl of 20 mg/ml proteinase K solution (1000 µg enzyme per 1 × 107 yeast cells). Incubate 1 hr at 50°C. Treatment with proteinase K proves to be very important because it efficiently increases cell permeability and DNA accessibility to the dye, and decreases nonspecific binding.
10. Transfer cell suspension to 12 × 75-mm tubes. Add 20 µl SYBR Green I working solution. Stain overnight at 6° to 8°C, protected from light. Final concentration of SYBR Green I used is equivalent to a 500-fold dilution of the commercial solution. This concentration may need adjustment if the DNA content of the species of interest differs significantly from that of the species used to optimize this protocol. Overnight staining guarantees maximal formation of DNA/SYBR Green I complexes.
11. Add Triton X-100 at a final concentration of 0.25% (v/v) and vortex the samples. 12. Place sample tubes on ice. Using a small ultrasonic probe immersed in the cell suspension, sonicate each sample with 3 consecutive ultrasound pulses at 30 W for 1 to 2 sec, with an interval of 1 to 2 sec between each pulse. A light, brief sonication of cell suspensions contributes to the elimination of most cell clumps, avoiding the occurrence of artificial DNA polyploid peaks. Optimal sonication can easily be controlled by checking the quality of DNA histograms.
13. Optional: Add green fluorescent microspheres (~15 to 20% of total events) as internal references. Microspheres should be selected by their nominal size and fluorescence intensity, in order to avoid an overlap with the cell signals. Green microspheres added to the cell suspension may be successfully used as internal references to: (1) check eventual DNA content alterations; (2) help in the setup of the flow cytometer; and (3) define a controlled cell cycle “window,” essential for the analysis of potential cell cycle disturbances. Sample results are shown in Figure 11.13.1.
14. Analyze cells on a flow cytometer using excitation at 488 nm and collecting fluorescence emission at 525 nm. Monitor quality of cell suspensions (i.e., presence of cell debris and clumps) using a scattergram (Fig. 11.13.2).
Microbiological Applications
11.13.3 Current Protocols in Cytometry
Supplement 13
104 103 102
Log forward scatter
101 100
100
101
102 Log side scatter
103
104
Figure 11.13.2 Typical scattergram for Saccharomyces cerevisiae (haploid strain W303 1A) after SYBR Green I staining. Total number of cells measured = 30,000. This histogram is used to check the quality of the cell suspension.
15. Acquire green fluorescence signals using linear amplification and a two-parameter histogram, i.e., area versus height of green fluorescence signal (Fig. 11.13.3). 16. Analyze the one-parameter DNA histogram (area) with software suitable for cell cycle analysis (Fig. 11.13.4). REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
RNase A solution, 1 mg/ml Dissolve 1 mg bovine pancreas RNase A (DNase-free; Sigma) in 1 ml Tris-EDTA buffer (see recipe). Store in convenient aliquots at 4°C. Sodium citrate buffer, 50 mM (pH 7.5) Dissolve 14.7 g sodium citrate in <1 liter distilled water. Adjust pH to 7.5, then bring volume to 1 liter. Filter solution using a 0.22-µm filter. Store up to 1 month at 4°C. SYBR Green I working solution Dilute the commercial stock solution of SYBR Green I (Molecular Probes) 1:10 in Tris-EDTA buffer, pH 8.0 (see recipe), and store protected from light up to several weeks at −20°C. Quickly thaw immediately prior to use. The presence of EDTA in the buffer and adjustment of the pH to 8.0 is important to maintain SYBR Green I stability/sensitivity. This solution can be stored in aliquots to avoid refreezing.
Tris-EDTA buffer, pH 8.0 Dissolve 3.7 g Tris base and 0.121 g disodium EDTA in <1 liter distilled water. Adjust pH to 8.0 at 4°C, using 1 N HCl, bring up to 1 liter with water. Filter solution using a 0.22-µm filter and store up to 1 month at 4°C. Cell Cycle Analysis of Yeasts
11.13.4 Supplement 13
Current Protocols in Cytometry
1023
A
Relative DNA (peak)
G2 /M
I
II
S
G0G1
0
G0G1
0
1023 Relative DNA (area)
G2 /M
Events
B
G0G1 S G0G1
Relative DNA (peak) Relative DNA (area)
Figure 11.13.3 (A) DNA two-parameter histogram (green fluorescence signal area versus green fluorescence signal height) for Zygosaccharomyces bailii (ISA 1307) (n = 30,000) after SYBR Green I staining. This distribution shows an apparent high synchrony of cells in G2/M according to the measurement of total SYBR Green I fluorescence (area). Two cell subpopulations, I and II, are visible in this region. In identical distributions obtained for Saccharomyces cerevisiae (haploid strain W303 1A), confocal laser microscopy of these two sorted subpopulations showed that subpopulation I was composed of budding cells with mother nuclei not yet in division (G2 nuclei) and of budding cells with nuclei in all phases of karyokinesis (uncompleted nuclear mitosis: M nuclei). Subpopulation II was shown to be composed almost exclusively of pairs of undivided mother and daughter cells with finished karyokinesis (complete nuclei: G0/G1). This subpopulation represented ∼15% of the total population in the histogram shown. Insert: overlay of a confocal fluorescence image of a pair of cells with nuclei stained with SYBR Green I and the respective transmitted image. (B) 3D representation of the same distribution.
Microbiological Applications
11.13.5 Current Protocols in Cytometry
Supplement 13
415
A
Events
G0/G1
G2/M
FPCV of G0/G1 = 4.6% HPCV of G0/G1 = 3.4%
0
S
0
1023
238
Relative DNA (area)
B
Events
G0/G1
G2/M
FPCV of G0/G1 = 11.3% HPCV of G0/G1 = 8.4%
0
S
0
1023 Relative DNA (area)
Figure 11.13.4 (A) DNA single-parameter histogram for Saccharomyces cerevisiae (haploid strain W303 1A) after SYBR Green I staining. Total number of cells measured = 30,000. This histogram illustrates the possibility of obtaining DNA distributions with narrow G0/G1 and G2/M peaks and and a clearly visible S phase. (B) DNA single-parameter histogram for the same species/strain after propidium iodide (PI) staining, using exactly the same prestaining protocol as for SYBR Green I staining. Total number of cells measured = 30,000. With PI it was not possible to obtain CVs <8%. FPCV, full peak coefficient of variation. HPCV, half peak coefficient of variation.
Cell Cycle Analysis of Yeasts
11.13.6 Supplement 13
Current Protocols in Cytometry
COMMENTARY Background Information Staining of yeast DNA Among the fluorescent probes available for DNA measurements and cell cycle analysis, propidium iodide (PI), Hoechst 33342, Ho echst 3 32 58 , and 4′,6-diamidino-2phenylindole (DAPI) have been the dyes most frequently used in various types of cells (Haugland, 1999), including yeasts (Dien et al., 1994; Carlson et al., 1997; Haase and Lew, 1997). PI is an intercalating dye, which binds dsDNA with a stoichiometry of one molecule per 4 to 5 bp, and which shows no specificity for base sequences. It also binds to dsRNA, making RNase treatments necessary to assure the DNA specificity of the measurement when the cells of interest express significant amounts of dsRNA. Hoechst 33342, Hoechst 33258, and DAPI bind to the minor groove of dsDNA in AT-rich regions. Although DAPI can also bind dsRNA intercalating in AU-rich sequences, its selectivity for dsDNA is reported to be higher than that of PI (Haugland, 1999). Up until the present, cell cycle analysis in yeasts using these fluorophores did not allow an accurate quantification of the cell fractions in the different phases of the cell cycle. In this context, SYBR Green I emerges as a promising fluorophore due to its very high selectivity for dsDNA and its higher fluorescence quantum yield upon DNA binding. Applications of SYBR Green I Until now, SYBR Green I has been used only in several molecular biology techniques for staining nucleic acids. Recent studies have reported the utility of SYBR Green I for the enumeration and cell cycle analysis of marine picoplankton and for the quantification of viruses in marine and freshwater samples (Marie et al., 1997, 1999; UNITS 11.11 & 11.12). The mitochondrial and chloroplast genomes in the green alga Chlamydomonas reinhardtii have been visualized by confocal microscopy after staining with SYBR Green I (Nishimura et al., 1998). This fluorophore has also been used to assess membrane integrity in Candida albicans (Liao et al., 1999) and for antibiotic susceptibility testing (UNIT 11.8). Measuring very low cell DNA contents Yeast cells have a much smaller total DNA content than higher eukaryotic cells (>300× less). It is estimated that a Saccharomyces cere-
visiae cell has ∼22 femtograms of DNA per haploid nucleus (Brock, 1973). These very low amounts of DNA per cell correspond to very low mean amounts of DNA stain per cell. This, in turn, means that DNA-stained cells produce very weak to weak signals, which are difficult to detect and are not well resolved, even by the sensitive photomultipliers found on current flow instruments. A possible solution is the use of new fluorophores, like SYBR Green I, which display high sensitivity for detecting DNA. They combine several important characteristics such as: (1) SYBR Green I has an exceptional affinity for DNA, at least an order of magnitude greater than that of an “old” fluorophore, e.g., ethidium bromide; and (2) DNA/SYBR Green I complexes have significantly greater fluorescence quantum yield than do complexes of DNA with other stains. The high selectivity of SYBR Green I for dsDNA contributes to a low nonspecific binding to other cell components. The properties of this new DNA stain, together with an optimal sample processing and setup/calibration of the flow cytometer, make possible the measurement of very low total cell DNA contents.
Critical Parameters and Troubleshooting Optimal SYBR Green I concentration Both undersaturation and oversaturation of cells with SYBR Green I increase the CV of total cell DNA measurement. The ideal is to obtain a maximum number of DNA/SYBR Green I complexes with minimum nonspecific binding and quenching. It is therefore advisable to titrate the optimal ratio of cell number to amount of SYBR Green I, in order to obtain high-resolution DNA histograms. Optimal pH for SYBR Green I staining Staining with SYBR Green I reagent is pH sensitive. For optimal sensitivity, it is important to verify that the pH of the staining solution at the temperature used for staining (6° to 8°C) is between 7.5 and 8.0. Protease treatment Treatment of cell suspensions with proteinase K has proved to be very important for obtaining DNA measurements with low CVs. This treatment efficiently increases cell permeability and DNA accessibility and decreases nonspecific binding.
Microbiological Applications
11.13.7 Current Protocols in Cytometry
Supplement 13
Ultrasonication of the cell suspension A brief ultrasonication of the cell suspension before flow cytometric analysis significantly reduces the incidence of cell clumps, which are responsible for the appearance of artificial DNA polyploid peaks in the DNA histograms. Although these peaks can also be removed from the histogram by gating, the presence of a high percentage of aggregates in the cell suspensions may cause a relative overestimation of G2/M peaks and an underestimation of G0/G1. Mitochondrial DNA The relative proportion of mitochondrial DNA to nuclear DNA is much higher in yeasts than in mammalian cells. Mitochondrial DNA may constitute 5% to 20% of the total yeast cell DNA (Dujon, 1981). It is obvious that differences in the number of mitochondria per cell may interfere with measures of nuclear DNA, thereby increasing the variation of total DNA from cell to cell. In other words, one can obtain “lower resolution” or “less accurate” DNA measurements, due not to methodological (i.e., quality control) problems, but to true biological variation in the total DNA per cell. Therefore, experimental conditions that may interfere with or induce changes in the total number of mitochondria per yeast cell and/or mitochondrial DNA synthesis per yeast cell may give rise to DNA distributions with a pattern that exactly reflects those cell changes. Cell clumping and/or pairs of undivided mother and daughter cells In order to monitor the possible presence of aggregates in cell suspensions, two-parameter histograms relating green fluorescence signal area to signal height (peak) must be included in the cytometric acquisition protocol. With some cell types, this histogram can be used to discriminate clumps of two G0 or G1 cells from an individual G2 or M cell, because these two different types of particles, in principle, produce signals with different heights. In most measures, a pair of G0 or G1 cells produces a half-signal height compared to a single G2 or M cell. A high frequency of false G2 or M cells, if not corrected, can result in an overestimation of the G2/M fraction and an underestimation of the G0/G1. A gate of the cell clumps may permit the correction/recalculation of those fractions. In ultrasonicated yeast cell suspensions, typical signals of cell doublets are not apparent. On the other hand, namely in yeast populations with a
high synchrony of dividing cells in G2 or M phases, a subpopulation with a DNA content equivalent to G2 or M, but producing a lower height signal than the true G2 or M, is clearly visible (see Fig. 11.13.3). When sorted and observed by confocal laser microscopy, this subpopulation has been shown to be composed almost exclusively of pairs of mother and daughter cells that are undivided, but which have finished karyokinesis (i.e., possess complete nuclei). This nonsynchronization between karyokinesis and cytokinesis must be taken into account when analyzing DNA histograms to quantitate yeast cell cycle fractions correctly allocated to the respective phases. Analysis of DNA histograms Data obtained by use of specific software to analyze DNA histograms are highly dependent upon user interaction and interpretation of the results. To guarantee reproducible results, it is advisable to define a consistent analysis protocol and to apply it to control and test cell suspensions. To control the quality of the cell cycle analysis, more recent software (MPLUS, Phoenix Flow Systems; MODFITLT, Verity Software) is able to automatically calculate the confidence intervals of the measurements and to compare the results obtained with different models of distribution fitting.
Anticipated Results The budding yeast Saccharomyces cerevisiae and the fission yeast Schizosaccharomyces pombe are reference microorganisms used for the study of many processes in eukaryotic cell biology. In particular, they are attractive model systems for cell cycle research. It is expected that the improvement in the accuracy of yeast cell cycle measurements achieved with this protocol will be useful for studies whose objectives are the detection of changes in total yeast cell DNA content and/or changes in cell cycle kinetics.
Time Considerations This protocol includes an overnight incubation with SYBR Green I. Nevertheless, the time of incubation with this stain may be reduced to as little as 30 or even 15 min. However, in the authors’ experience the overnight staining normally produces DNA distributions with slightly better CVs. Samples of cell suspensions of Saccharomyces cerevisiae (haploid strain W303 1A) and Zygosaccharomyces
Cell Cycle Analysis of Yeasts
11.13.8 Supplement 13
Current Protocols in Cytometry
bailii (ISA 1307) treated with RNase A showed only a slight improvement in the CV of the G0/G1 peak, compared with nontreated samples. This may reflect the fact that in some species/strains, the expression of dsRNA does not interfere with DNA analysis. One could consider omitting this step to simplify the procedure and reduce the processing time by 1 hr.
Literature Cited Brock, T.D. 1973. Biología de los microorganismos. Ediciones Omega, S.A., Barcelona, Spain. Carlson, C.R., Grallert, B., Bernander, R., Stokke, T., and Boye, E. 1997. Measurement of nuclear DNA content in fission yeast by flow cytometry. Yeast 13:1329-1335. Dien, B.S., Peterson, M.S., and Srienc, F. 1994. Cell-cycle analysis of Saccharomyces cerevisiae. Methods Cell Biol. 42:457-475. Dujon, B. 1981. The Molecular Biology of the Yeast Saccharomyces (J.N. Strattem, E.W. Jones, and J.R. Broach, eds.) pp. 505-625. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Haase, S.B. and Lew, D.J. 1997. Flow cytometric analysis of DNA content in budding yeast. Methods Enzymol. 283:322-332. Haugland, R.P. (ed.) 1999. Handbook of Fluorescent Probes and Research Chemicals, 7th ed. Molecular Probes, Eugene, Ore.
Liao, R.S., Rennie, R.P., and Talbot, J.A. 1999. Assessment of the effect of amphotericin B on the vitality of Candida albicans. Antimicrob. Agents Chemother. 43:1034-1041. Marie, D., Partensky, F., Jacquet, D., and Vaulot, D. 1997. Enumeration and cell cycle analysis of natural populations of marine picoplankton by flow cytometry using the nucleic acid stain SYBR Green I. Appl. Environ. Microbiol. 63:186-193. Marie, D., Brussaard, C.P.D., Thyrhaug, R., Bratbak, G., and Vaulot, D. 1999. Enumeration of marine viruses in culture and natural samples by flow cytometry. Appl. Environ. Microbiol. 65:45-52. Nishimura, Y., Higashiyama, T., Suzuki, L., Misumi, O., and Kuroiwa, T. 1998. The biparental transmission of the mitochondrial genome in Chlamydomonas reinhardtii visualized in living cells. Eur. J. Cell Biol. 77:124-133.
Contributed by Margarida Fortuna, Maria João Sousa, Manuela Côrte-Real, and Cecília Leão Universidade do Minho Braga, Portugal Alexandre Salvador and Filipe Sansonetty IPATIMUP Porto, Portugal
This study was supported by a research grant (PAMAF no. 6088). We are very grateful to Alberto Alvarez-Barrientos (Centro de Citometria de Flujo y Microscopia Confocal, Facultad de Farmacia, Universidad Complutense de Madrid, Spain), who kindly did the cell sorting experiments to validate the interpretation of the results obtained with the two-parameter histograms relating green fluorescence signal area to signal height (peak).
Microbiological Applications
11.13.9 Current Protocols in Cytometry
Supplement 13
Flow Cytometric Assessment of Drug Susceptibility in Leishmania infantum Promastigotes
UNIT 11.14
The flow cytometry approaches described in this unit enable one to detect, differentiate, and quantify cellular changes in Leishmania parasites as a result of treatment with allopurinol or other drugs. In addition, the flow cytometry approaches also demonstrate differences in allopurinol susceptibility between two promastigote forms, thereby expanding the spectrum of flow cytometry applications in the field of parasitology and in studies of parasite-drug interactions as well as cellular toxicity. This unit presents protocols for assessment of Leishmania promastigote proliferation (see Basic Protocol 1), viability (see Basic Protocol 2), and cellular protein content (see Basic Protocol 3). A Support Protocol describes the production of drug-resistant promastigotes. In these protocols, the green fluorescence of CFSE, SYBR-14, and FITC and the red fluorescence of PI are excited at 488 nm (FACScalibur, Becton Dickinson). The fluorescence intensities of stained and unstained wt-p299 and allo-p229-promastigotes, treated and untreated with allopurinol, can be determined and compared. At least 10,000 cells should be acquired (log FS and log SS) and analyzed per run, and each staining experiment should be repeated at least three times. Data analysis should be performed on fluorescence intensities that exclude cell autofluorescence and cell debris. Software for cytometric analysis like CELLQuest or WinMDI can be used to determine the cell fluorescence. ASSESSING LEISHMANIA PROMASTIGOTE PROLIFERATION BY STAINING WITH CFSE
BASIC PROTOCOL 1
Kamau et al. (2000) describe a new technique for the investigation of Leishmania promastigote proliferation using 5-(and-6)-carboxyfluorescein diacetate succinimidyl ester (CFSE), which permits one to resolve and track populations of cells that have undergone different numbers of cell divisions. Until now, incorporation of [3H]-thymidine has been the most common method for determining cell division. The effects of antileishmanial drugs such as allopurinol on the proliferation of these promastigotes at different time intervals can be assessed with CFSE (Fig. 11.14.1A). This new flow cytometric application to Leishmania opens up potential studies in antileishmanial drug pharmacokinetic and toxicology studies. Materials Promastigote forms of L. infantum (MCAN/ES/89/IPZ229/I/89, zymodeme MON 1: wt-p229) isolated from a patient (see recipe for cultivation) Phosphate-buffered saline (PBS; APPENDIX 2A) 2.8 mg/ml 5-(and-6)-carboxyfluorescein diacetate succinimidyl ester (CFSE; Molecular Probes) in DMSO (store up to 1 year at −20°C) Medium for Leishmania promastigote culture (see recipe) 10 mg/ml allopurinol stock in 0.1 N NaOH (store up to 2 to 4 weeks at 4°C) 15-ml conical centrifuge tubes (e.g., Corning) 25-cm2 tissue culture flasks (Corning) 27°C incubator 5-ml round-bottom polystyrene centrifuge tubes (e.g., Falcon) Flow cytometer (e.g., FACScalibur; Becton Dickinson) Additional reagents and equipment for counting cells in a hemacytometer (APPENDIX 3A) Contributed by Rafael Nuñez, Sarah Kamau, and Felix Grimm Current Protocols in Cytometry (2001) 11.14.1-11.14.9 Copyright © 2001 by John Wiley & Sons, Inc.
Microbiological Applications
11.14.1 Supplement 15
B
PI
100
101
102
103
104
A
101
102 Sybr-14
103
104
101
102 Sybr-14
103
104
PI
100
101
102
103
104
100
Control
Day 2
Day 0
Day 3
Day 1
Day 4
100
Figure 11.14.1 (A) Promastigote proliferation determined by CFSE staining of wt-p229. Promastigotes untreated (top graph), or treated with 400 µg/ml allopurinol (bottom graph). (B) Promastigotes untreated (top graph), or treated with 400 µg/ml allopurinol (bottom graph).
1. Using a 15-ml conical centrifuge tube, centrifuge a suspension containing 10 × 107 promastigotes in logarithmic growth phase at 1200 × g, 4°C, and remove the supernatant. Resuspend the pellet in 10 ml PBS, 4°C, and centrifuge again at 4°C. Repeat wash twice more. 2. Resuspend cells in 2 ml PBS. Count in a hemacytometer (APPENDIX 3A) and adjust final concentration to 6 × 107 cells/ml. 3. Add CFSE (from 2.8 mg/ml stock) to a final concentration of 2.8 µg/ml. 4. Incubate cells 10 min at 37°C with mixing 3 to 4 times. Keep the cells protected from light. 5. Add several volumes of ice-cold medium for Leishmania culture, supplemented with 10% heat-inactivated FBS. This will stop CFSE binding. Drug Susceptibility in Leishmania Promastigotes
6. Centrifuge 10 min at 1200 × g, 4°C. 7. Remove medium and resuspend cell pellet with fresh medium for Leishmania culture.
11.14.2 Supplement 15
Current Protocols in Cytometry
8. Adjust cell density to 5 × 106 cells/ml and continue culturing in 25-cm2 tissue culture flasks at 27°C. 9. Right after staining and adjustment of the cell concentration, add drug (e.g., allopurinol) at varying concentrations (0 through 800 µg/ml from a 10 mg/ml stock solution for allopurinol). This is time 0.
10. Determine the CFSE fluorescence by flow cytometry immediately after staining, and after 24, 48, 72, and 96 hr, in 5-ml round-bottom polystyrene tubes. For FACScalibur the suggested acquisition parameters (log mode) are as follows: FS, voltage E00; SS, voltage 286; green fluorescence (530 nm), voltage 550; yellow/orange fluorescence (585 nm), voltage 584; red fluorescence (650 nm), voltage 584. Compensation is as follows: green fluorescence minus 27.2% yellow/orange fluorescence; yellow/orange fluorescence minus 65.3% green fluorescence; yellow/orange fluorescence minus 5.4% red fluorescence; and red fluorescence minus 5.6% red fluorescence. Also see UNIT 1.14. A 525±15 nm band-pass filter is required for green fluorescence, a 575±15 nm filter is required for yellow/orange fluorescence, and a 650 nm long-pass filter is required for red fluorescence.
ASSESSING LEISHMANIA PROMASTIGOTE VIABILITY BY STAINING WITH PI AND SYBR-14
BASIC PROTOCOL 2
The combination of SYBR-14 and PI has been used extensively in sperm viability studies (Garner et al., 1994). However, this protocol represents the first procedure for dual staining to determine the viability of parasites. SYBR-14 stains the nucleus and mitochondria DNA of live promastigotes brilliant green, while PI stains the nuclei of dead promastigotes red. Some promastigotes are dual-stained, resulting in a yellowish nucleus. Flow cytometry is effective in quantifying the resultant fluorescent populations: (1) SYBR-14-stained (alive), (2) PI-stained (dead), and (3) dual-stained promastigotes (dying). Both dyes label DNA, thus avoiding the ambiguity of stains that target separate cellular organelles (Garner et al., 1994). The significant increase in dead (PI-stained) and dying (PI- and SYBR-14-stained) cells in the wt-p229 promastigotes after exposure to allopurinol indicates a clear allopurinol susceptibility of these promastigotes. In contrast, the proportions of dead and dying cells in the allo-p229 promastigotes is not significantly influenced by drug exposure (Fig. 11.14.1B). Materials Promastigote forms of L. infantum (MCAN/ES/89/IPZ229/I/89, zymodeme MON 1: wt-p229) isolated from a patient (see recipe for cultivation) Medium for Leishmania promastigote culture (see recipe) 10 mg/ml allopurinol stock in 0.1 N NaOH (store up to 4 weeks at 4°C) Phosphate-buffered saline (PBS; APPENDIX 2A) 1 mg/ml propidium iodide (PI) stock in water (store up to 1 year at −20°C) 1 mg/ml SYBR-14 (Molecular Probes) in DMSO (store up to 1 year at −20°C protected from light) 25-cm2 tissue-culture flasks (Corning) 15-ml conical centrifuge tubes (e.g., Corning) 27°C incubator 5-ml round-bottom polystyrene centrifuge tubes (e.g., Falcon) Flow cytometer (e.g., FACScalibur; Becton Dickinson) Additional reagents and equipment for counting cells in a hemacytometer (APPENDIX 3A)
Microbiological Applications
11.14.3 Current Protocols in Cytometry
Supplement 15
1. Set up promastigote cultures by inoculating fresh medium with log-phase promastigotes. Grow cells to a density of 5–10 × 106 cells/ml (∼3 days after inoculation) in 25-cm2 flasks. 2. Add drug (e.g., allopurinol) at variable concentrations (0 through 800 µg/ml from a 10 mg/ml stock solution for allopurinol). 3. Incubate cultures an additional 48 hr at 27°C. 4. Count cells using a hemacytometer (APPENDIX 3A). Using a 15-ml conical centrifuge tube, centrifuge an aliquot of cells containing 4 × 106 promastigotes 10 min at 1200 × g, 4°C, and remove the supernatant. 5. Optional: Resuspend pellet in 2 ml PBS, centrifuge 10 min at 1200 × g, 4°C, and remove supernatant. Repeat wash two additional times. This step is desirable if there is tissue overgrowth.
6. Resuspend cell pellet containing 4 × 106 promastigotes in 2 ml PBS. 7. Add PI (from 1 mg/ml stock) to a final concentration of 10 µg/ml and SYBR-14 (from 1 µg/ml stock) to a final concentration of 0.1 µg/ml. Protect the promastigotes from direct light. 8. Incubate 30 min at 37°C. Gently agitate the tubes 3 to 4 times over the course of the incubation period. 9. Optional: Wash cells with PBS as in step 5. This step is desirable if there is need to remove excess stain due to fresh preparation of the stocks.
10. Determine the fluorescence by flow cytometry in 5-ml round-bottom polystyrene tubes, immediately after staining. Avoid delays in the measurement. For FACScalibur the suggested acquisition parameters (log mode) are as follows: FS, voltage E00 or E01; SS, voltage 350 or 443; green fluorescence (530 nm), voltage 582; yellow/orange fluorescence (585 nm), voltage 550; red fluorescence (650 nm), voltage 489. Compensation is as follows: green fluorescence minus 4.9% yellow/orange fluorescence; yellow/orange fluorescence minus 18.4% green fluorescence. Also see UNIT 1.14. A 525±15 nm band-pass filter is required for green fluorescence, a 575±15 nm filter is required for yellow/orange fluorescence, and a 650 nm long-pass filter is required for red fluorescence. BASIC PROTOCOL 3
ASSESSING CELLULAR PROTEIN CONTENT OF PROMASTIGOTES BY STAINING WITH FITC The total cellular protein content of fixed promastigotes can also be determined by staining with fluorescein isothiocyanate (FITC), an acidic dye that binds covalently to the positively charged groups of proteins.
Drug Susceptibility in Leishmania Promastigotes
Materials Promastigote forms of L. infantum (MCAN/ES/89/IPZ229/I/89, zymodeme MON 1: wt-p229) isolated from a patient (see recipe for cultivation) Medium for Leishmania promastigote culture (see recipe) 10 mg/ml allopurinol in 0.1 N NaOH (store up to 2 to 4 weeks at 4°C) PBS-DNase: 4 µg/ml DNase (Type I; Sigma) in PBS (APPENDIX 2A); store up to 1 year at −20°C 70% methanol
11.14.4 Supplement 15
Current Protocols in Cytometry
1 mg/ml fluorescein isothiocyanate (FITC; Molecular Probes) in PBS (prepare fresh) 25-cm2 tissue-culture flasks (Corning) 27°C incubator 15-ml conical centrifuge tubes (e.g., Corning) 5-ml round-bottom polystyrene centrifuge tubes (e.g., Falcon) Flow cytometer (e.g., FACScalibur; Becton Dickinson) Additional reagents and equipment for counting cells in a hemacytometer (APPENDIX 3A) 1. Set up promastigote cultures by inoculating fresh medium with log-phase promastigotes. Grow cells to a cell density of 5–10 × 106 cells/ml (∼3 days after inoculation) in 25-cm2 flasks. 2. Add drug (e.g., allopurinol) at variable concentrations (0 through 800 µg/ml from a 10 mg/ml stock solution for allopurinol). 3. Incubate cultures an additional 48 hr at 27°C. 4. Count cells using a hemacytometer (APPENDIX 3A). Using a 15-ml conical centrifuge tube, centrifuge an aliquot of cells containing 4 × 106 promastigotes 10 min at 1200 × g, 4°C, and remove the supernatant. Resuspend pellet in 2 ml PBS, centrifuge again 10 min at 1200 × g, and remove supernatant. Repeat wash two additional times. 5. Resuspend cell pellet (4 × 106 promastigotes) in 2 ml PBS-DNase. Incubate 10 min at 37°C. 6. Fix cells 6 hr in 70% methanol at 4°C. 7. Wash and resuspend the fixed cells in PBS (4 × 106 cells/ml). Add FITC from 1 mg/ml stock to a final concentration of 0.1 µg/ml and incubate 30 min at 37°C. 8. Wash cells as in step 4. 9. Determine the fluorescence by flow cytometry in 5-ml round-bottom polystyrene tubes immediately after staining. For FACScalibur the suggested acquisition parameters (log mode) are as follows: FS, voltage E00 or E01; SS, voltage 350 or 443; green fluorescence (530 nm), voltage 582; yellow/orange fluorescence (585 nm), voltage 550; red fluorescence (650 nm), voltage 489. Compensation is as follows: green fluorescence minus 4.9% yellow/orange fluorescence; yellow/orange fluorescence minus 18.4% green fluorescence. Also see UNIT 1.14. A 525±15 nm band-pass filter is required for green fluorescence, a 575±15 nm filter is required for yellow/orange fluorescence, and a 650 nm long-pass filter is required for red fluorescence.
GENERATION OF ALLOPURINOL-RESISTANT PROMASTIGOTES (allo-p229)
SUPPORT PROTOCOL
Promastigotes of the same isolate promastigote forms (“wild-type” promastigotes) of L. infantum (MCAN/ES/89/IPZ229/I/89, zymodeme MON 1: wt-p229) are cultured in the medium used in the above protocols, in the presence of increasing concentrations of allopurinol, until a final concentration of 800 µg/ml is reached. This process takes 12 months. The allo-p229 strain is a drug-resistant cell which permits comparison of the diverse techniques. Microbiological Applications
11.14.5 Current Protocols in Cytometry
Supplement 15
Materials Promastigote forms of L. infantum (MCAN/ES/89/IPZ229/I/89, zymodeme MON 1: wt-p229) isolated from a patient (see recipe for cultivation) Medium for Leishmania promastigote culture (see recipe) 10 mg/ml allopurinol stock in 0.1 N NaOH (store up to 2 to 4 weeks at 4°C) 1. Change medium every week for three weeks by inoculating log-phase promastigotes into fresh medium containing 100 µg/ml (final) allopurinol (added from 10 mg/ml stock). 2. Change medium every week for three weeks by inoculating log-phase promastigotes into fresh medium containing 200 µg/ml (final) allopurinol. 3. Change medium every week for two weeks by inoculating log-phase promastigotes into fresh medium containing 400 µg/ml (final) allopurinol. 4. Change medium every week for 10 months by inoculating log-phase promastigotes into fresh medium containing 800 µg/ml (final) allopurinol. REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Medium for Leishmania culture Prepare the liquid medium as described in Table 11.14.1. Adjust pH to 7.4 using 10 N NaOH. Adjust volume to 5 liters and sterilize by filtration through a 0.22-µm filter. Store up to 4 months at 4°C. Working medium: Just before use, add the following: 10% heat-inactivated FBS (APPENDIX 2A) 1.0% (v/v) sterile 10 mg/ml gentamycin 1.0% (v/v) sterile 2 mg/ml hemin Parasite cultivation Pr omastigote forms (“wild-type” promastigotes) of L. infantum (MCAN/ES/89/IPZ229/I/89, zymodeme MON 1: wt-p229) are maintained at 27°C in 25-cm2 tissue culture flasks (Corning) containing 5 ml liquid medium, pH 7.4 (Table 11.14.1), supplemented with 10% heat-inactivated fetal bovine serum. COMMENTARY Background Information
Drug Susceptibility in Leishmania Promastigotes
Leishmaniasis is a major tropical and subtropical parasitic disease. The prevalence is estimated at 12 million people worldwide with 200-350 million people at risk. In the Mediterranean region, leishmaniasis caused by Leishmania infantum has emerged as one of the important opportunistic infections of human immunodeficiency virus (HIV)–positive individuals (Montalban et al., 1990). Moreover, the prevalence of canine leishmaniasis in this region may be as high as 42%. Dogs and wild canids are important reservoirs and are mainly responsible for the persistence of the disease in this region (Baneth et al., 1998).
Sodium stibogluconate, N-methyl-D-glucamine antimoniate, amphotericin B, pentamidine, and ketoconazole are all drugs used in treatment of leishmaniasis. Some of these drugs cause severe adverse side effects, and failures of treatment are common. Allopurinol, a purine analog, has been used for treatment of leishmaniasis, alone or combined with the previously mentioned drugs. In Leishmania, allopurinol inhibits purine biosynthesis and hence inhibits protein synthesis (Frayha et al., 1997). Low cost, ease of administration (oral), and lack of toxicity make allopurinol, which is today mainly used in canine leishmaniasis, a
11.14.6 Supplement 15
Current Protocols in Cytometry
Table 11.14.1
Formula for Promastigote Culture Mediuma
Component
Amount
Supplier
Catalog No.
β-Alanine DL-Alanine L-Arginine⋅HCl L-Asparagine L-Aspartic acid L-Cysteine⋅HCl⋅H2O L-Cystine L-Glutamic acid L-Glutamine Glycine L-Histidine DL-Isoleucine L-Leucine L-Lysine⋅HCl DL-Methionine L-Phenylalanine DL-Serine Taurine DL-Threonine L-Tryptophan L-Tyrosine DL-Valine Adenosine Guanosine Biotin stock (0.2 mg/ml) MEM NEAA (100×) MEM AA (50×) MEM vitamin solution (100×) Medium 199 S-MEM (Eagle) Sodium pyruvate L-Malic acid Fumaric acid, disodium salt Folic acid Succinic acid α-Ketoglutaric acid Citric acid (anhydrous) PABA stock (0.2 mg/ml) NaH2PO4-2H2O Sodium bicarbonate CaCl2-2H2O KCl MgCl2⋅6H2O MgSO4⋅7H2O Phenol red L-Proline
5g 3225 mg 1350 mg 530 mg 275 mg 225 mg 75 mg 625 mg 4850 mg 300 mg 300 mg 225 mg 225 mg 375 mg 675 mg 700 mg 650 mg 1075 mg 1125 mg 250 mg 750 mg 525 mg 25 mg 25 mg 5 ml 25 ml 20 ml 5 ml
Sigma Sigma Sigma Life Technologies Life Technologies Life Technologies Serva Sigma Fluka Merck Fluka Sigma Life Technologies Life Technologies Sigma Life Technologies Sigma Sigma Sigma Sigma Sigma Sigma Sigma Sigma Sigma Sigma Sigma Life Technologies
A9920 A3409 A3909 11013-018 11016-029 11035-060 17880 G5638 49420 4201 53320 I6268 11077-021 066-01083 H M2768 21095-013 S5386 T7146 T1520 T0271 T1020 V6379 A4036 G6264 B4639 M7145 M7020 11120-037
5g 17.5 g 250 mg 1675 mg 195 mg
Life Technologies Life Technologies Life Technologies Sigma Sigma
31100-084 072-01400 P 11840-030 M7397 F1506
10 mg 345 mg 925 mg 1470 mg 2.5 ml 1725 mg 5.5 g 375 mg 7.45 g 7.6 g 9.25 g 50 mg 18.8 g
Life Technologies Sigma Sigma Sigma Sigma Fluka Merck Fluka Fluka Fluka Sigma Sigma Life Technologies
13370-010 S9637 K1128 C4540 A9878 71500 6329 21100 60130 63065 M1880 P5530 066-01096 E continued
Microbiological Applications
11.14.7 Current Protocols in Cytometry
Supplement 15
Table 11.14.1
Formula for Promastigote Culture Mediuma, continued
Component
Amount
Supplier
Catalog No.
D(−)Fructose
1000 mg 1750 mg 1000 mg 50 g
Sigma Sigma Sigma Life Technologies
F3510 G7021 S9031 H0763
D(+)Glucose
Sucrose HEPES
aFor additional discussion, see Brun and Schönenberger (1979), Cunningham (1977), and Grimm et al. (1991).
Drug Susceptibility in Leishmania Promastigotes
particularly appealing candidate (Quellette and Papadopoulou, 1993). The effect of allopurinol on wild-type promastigotes (wt-p229), and on promastigotes of the same isolate which had been cultivated in vitro in the presence of up to 800 µg/ml allopurinol for a period of one year (allo-p229), was monitored by diverse flow cytometric approaches. The use of CFSE determined the effect of allopurinol on the proliferation of these promastigotes at different time intervals (Fig. 11.14.1A). This new flow cytometric application to Leishmania opens up potential studies in antileishmanial drug pharmacokinetic and toxicology studies (Kamau et al., 2000). The combination of SYBR-14 and PI has been used extensively in sperm viability studies (Garner et al., 1994). However, this is the first time that this dual staining is used to determine the viability of parasites. Flow cytometry can be very effective in quantifying the Leishmania fluorescent populations (Fig. 11.14.1B): (1) SYBR-14-stained (alive), (2) PI-stained (dead), and (3) dual-stained promastigotes (dying). This staining method has the advantage of being rapid (30 min) and and also that the cells do not require extra processing prior to the staining. Thus, Kamau et al. (2000) demonstrated that SYBR-14 used in combination with PI was effective for simultaneously visualizing both the living and dead populations of Leishmania promastigotes before and after treatment with allopurinol. Moreover, the flow cytometry approaches allow one to demonstrate differences in allopurinol susceptibility between two promastigote forms, expanding the spectrum of flow cytometry applications into studies of parasite resistance. The assessment of viability and cellular changes by flow cytometry proved to be a promising way of evaluating the susceptibility and resistance of Leishmania promastigotes to allopurinol. The successful application of flow cytometry to determine cellular changes in
Leishmania cells further opens up future perspectives in determination of effects of antileishmanial compounds. Finally, the CFSE assay and the viability assay with SYBR-14 and PI are new tools in the flow cytometric measurement of cell toxicity in parasitology.
Critical Parameters and Troubleshooting The viability of the parasites, the freshness of the reagents, and strict adherence to protocols are critical elements in obtaining satisfactory results. It is mandatory to keep all the fluorescent reagents protected from light. In addition, during and after staining keep the samples (tubes) covered with aluminum foil until the very last moment before cytometry measurement.
Anticipated Results CFSE staining permits one to resolve and track population of cells which have undergone different number of cell divisions as shown in Figure 11.14.1A. Staining with SYBR-14 and PI is effective for simultaneously visualizing both the living and dead populations of Leishmania promastigotes before and after treatment with allopurinol. A significant increase in dead (PI-stained) and dying (PI- and SYBR-14-stained) wt-p229 promastigotes after exposure to allopurinol indicates a clear allopurinol susceptibility of these promastigotes. In contrast, if the proportion of dead and dying promastigotes is not significant, the cells have not been influenced by the drug exposure (Fig. 11.14.1B). Protein content of the fixed cells can also be determined by measuring FITC fluorescence intensity of the control and allopurinol-treated promastigotes. Allopurinol treatment leads to a clear decrease in the mean fluorescence intensity only in allopurinol-sensitive promastigotes, whereas no distinct effect should be detected in resistant promastigotes. FITC staining complements the two other flow cytometric assays.
11.14.8 Supplement 15
Current Protocols in Cytometry
Time Considerations The staining procedures are quite simple and do not require extensive incubations. Most of the incubations take less than 30 min. However, washing procedures should be enforced in order to keep very low background during the staining. The staining method with SYBR-14 and PI has the advantage of being rapid (30 min), and also that the cells do not require extra processing prior to the staining.
Literature Cited Baneth, G., Dank, G., Keren-Kornblatt, E., Sekeles, E., Adini, I., Eisenberger, C.L., Schnur, L.F., King, R., and Jaffe, C.L. 1998. Emergence of visceral leishmaniasis in central Israel. Am. J. Trop. Med. Hyg. 59:722-725. Brun, R. and Schönenberger, M. 1979. Cultivation and in vitro cloning of procyclic culture forms of Trypanosoma brucei in a semi-defined medium. Acta Tropica 36:289-292. Cunningham, I. 1977. New culture medium for maintenance of tsetse tissues and growth of trypanosomatids. J. Protozool. 24:325-329. Frayha, G.J., Smyth, J.D., Gobert, J.G., and Savel, J. 1997. The mechanism of action of antiprotozoal and anthelmintic drugs in man. Gen. Pharmacol. 28:273-299. Garner, D.L., Johnson, L.A., Yue, S.T., Roth, B.L., and Haugland, R.P. 1994. Dual DNA staining assessment of bovine sperm viability using SYBR-14 and propidium iodide. J. Androl. 15:620-629.
Grimm, F., Brun, R., and Jenni, L. 1991. Promastigote infectivity in Leishmania infantum. Parasitol. Res. 77:185-191. Kamau, S., Hurtado, M., Müller-Doblies, U., Grimm, F., and Nunez, R. 2000. Flow cytometric assessment of allopurinol susceptibility in Leishmania infantum promastigotes. Cytometry 40:353-360. Montalban, C., Calleja, J.L., Erice, A., Laguna, F., Clotet, B., Podzamczer, D., Cobo, J., Mallolas, J., Yebra, M., and Gallego, M. 1990. The Co-operative Group for Study of Leishmaniasis in AIDS: Visceral leishmaniasis in patients infected with human immunodeficiency virus. J. Infect. 21:261-270. Quellette, M. and Papadopoulou, B. 1993. Mechanisms of drug resistance in Leishmania. Parasitol. Today 9:150-153.
Key Reference Kamau et al., 2000. See above. The study of Kamau et al. demonstrated (1) that CFSE determined the effect of the antileishmania drug allopurinol on the proliferation of promastigotes at different time intervals, (2) that SYBR-14 used in combination with PI is effective for simultaneously visualizing both the living and dead populations of Leishmania promastigotes before and after treatment with allopurinol, and (3) that the total cellular protein content of fixed promastigotes can also be determined by staining with FITC.
Contributed by Rafael Nuñez, Sarah Kamau, and Felix Grimm University of Zürich Zürich, Switzerland
Microbiological Applications
11.14.9 Current Protocols in Cytometry
Supplement 15
Resolution of Viable and Membrane-Compromised Free Bacteria in Aquatic Environments by Flow Cytometry Cellular viability is a key parameter in the study of aquatic microbial assemblages and of their relationships with the natural environment because it allows one to correlate total activities measured by a global method with the only fraction of active cells responsible for these activities (Gasol et al., 1999). Membrane integrity is one of the main criteria generally accepted to characterize a living cell and to distinguish it from a dead or damaged one. Indeed, membranes isolate the cell from the exterior medium, regulate ion or molecule exchanges between them, and are indispensable to the establishment of the proton electrochemical potential difference driving ATP synthesis at the bioenergeticmembrane level during oxidative phosphorylation (Mitchell, 1961; Nicholls, 1982).
UNIT 11.15
BASIC PROTOCOL
In aquatic environments, free heterotrophic bacteria play an extremely important role because of their high biomass, wide panel of metabolisms, and ubiquity, as well as the toxicity of certain species (Porter, 1996). Bacteria are responsible for most of the mineralization of the organic matter in bodies of water, and the determination of their abundance and biomass are basic parameters in related investigations (Pomeroy, 1984; Kirchman et al., 1991; UNIT 11.9). Although determination of the total abundance of bacteria is routinely performed (Hobbie et al., 1977; Kogure et al., 1979; Porter et al., 1995), quantification of the viable fraction is not easily achieved. This unit presents a nucleic-acid double-staining protocol (NADS) for flow cytometry that can distinguish the fractions of viable, damaged, or membrane-compromised cells within the marine or freshwater free-bacterial community, while distinguishing them from debris. Flow cytometry is particularly suited for analysis of aquatic bacteria because it is able to perform rapid multiparametric analysis on individual cells at cell numbers representative of natural environments (Trousselier et al., 1993). Since heterotrophic bacteria do not contain natural photosynthetic pigments and are not autofluorescent, staining has to be performed to discriminate and enumerate these bacteria by flow cytometry (UNIT 11.11). The NADS protocol is based on the simultaneous utilization of two nucleic acid stains, SYBR Green and PI. SYBR Green (I or II) is a membrane-permeant dye (Haugland, 1998) whereas PI is membrane impermeant (Jones and Senft, 1985; Sgorbati et al., 1996; López-Amorós et al., 1997; Williams et al., 1998). Both are readily excited by the blue line from an argon laser or an arc lamp, the excitation sources generally found in benchtop analytical flow cytometers. The efficiency of the double staining is magnified by the FRET from SYBR Green to PI when both are bound to the nucleic acids (Barbesti et al., 2000), in which the green fluorescence of SYBR Green is quenched by the PI (for more details on FRET, refer to UNIT 1.12). According to the FRET phenomenon, full quenching of SYBR Green fluorescence by PI will identify cells with a compromised membrane fully permeable to PI and will result in cells with red-only fluorescence. Partial quenching will indicate cells with a slightly damaged membrane that allows a small amount of PI to penetrate and will result in cells with both red and green fluorescence. Lack of quenching will characterize cells with an intact membrane stained only by SYBR Green and identified as viable by their green-only fluorescence. Figure 11.15.1 displays an example of this cluster resolution for an E. coli culture sample. This approach, originally described by Barbesti et al. (2000) for cultured bacteria, has been adapted to bacteria from fresh and marine waters (Grégori et al., 2001). For marine samples, SYBR Green II is preferred to SYBR Green I because it shows a higher fluorescence intensity (Lebaron et al., 1998). Contributed by Gérald Grégori, Michel Denis, Sergio Sgorbati, and Sandra Citterio Current Protocols in Cytometry (2003) 11.15.1-11.15.7 Copyright © 2003 by John Wiley & Sons, Inc.
Microbiological Applications
11.15.1 Supplement 23
1000
Green (SYBR Green) fluorescence (au)
A 100
B
vertical limit viable bacteria
10 dead bacteria 1
D noise
0.1 0.1
noise 1
10
100
1000
0.1
horizontal limit 1
10
100
1000
Red (PI) fluorescence (au)
Figure 11.15.1 Determination of the A, B, and D quadrants used to resolve viable, membranedamaged, and membrane-compromised bacteria (dead cells), respectively, on a green- versus red-fluorescence histogram. On the left, positioning of the vertical limit between red-fluorescing dead bacteria and the background noise of an ozone-treated sample. On the right, positioning of the horizontal limit between green-fluorescing viable bacteria and background noise in an untreated sample containing mainly viable bacteria.
Samples do not require any pretreatment and this protocol can be performed either on board ship or in the laboratory. The different clusters of bacteria—i.e., intact membranes (viable), damaged membranes, and compromised membranes—present in natural marine samples are discriminated on the basis of the green and red fluorescence of the cells after staining. Analogous fluorescence quenching could be achieved by using Hoechst dyes instead of SYBR Green (Nebe-von Caron and Badley, 1995; Lebaron et al., 1998), but those fluorochromes require ultraviolet excitation, which in general necessitates the use of more expensive flow cytometers. Materials Natural fresh- or seawater samples SYBR Green I (fresh water) or II (seawater) working solution (see recipe) Cells killed by paraformaldehyde fixation, heat, or ozone treatment Freshly harvested cells: freshly harvested natural sample with >95% viability or sample from a bacterial culture in exponential growth phase 1 mg/ml propidium iodide (PI; Molecular Probes): store ≤1 month at 4°C in the dark Fresh water or seawater: filter through a 0.2-µm filter 10% bleach: dilute standard bleach (5% sodium hypochlorite) 1:10 in H2O 70% ethanol
Resolution of Viable and Membrane-Compromised Aquatic Bacteria
100-µm filter Flow cytometer: 488-nm argon laser or arc lamp Filters for collection of 525 ± 15-nm (green) fluorescence and >620-nm (red) fluorescence Sheath fluid: distilled water passed through a 0.2-µm filter 12 × 75–mm tubes, sterile
11.15.2 Supplement 23
Current Protocols in Cytometry
NOTE: All water, including fresh water or seawater used to dilute samples, and that used to make solutions, should be passed through a 0.2-µm filter prior to use. Stain cells 1. Filter natural fresh- or seawater samples through a 100-µm filter to avoid clogging the flow cytometer. 2a. For natural freshwater samples: Add 20 µl SYBR Green I working solution to 1.980 ml freshwater sample in a sterile 12 × 75–mm tube and vortex. In the same manner also stain three controls: 1.980 ml water (dye background levels) Cells killed by paraformaldehyde fixation, heat, or ozone treatment (negative control) Freshly harvested cells (positive control). 2b. For natural seawater samples: Add 20 µl SYBR Green II working solution to 1.980 ml seawater sample in a sterile 12 × 75–mm tube and vortex. In the same manner also stain three controls: 1.980 ml seawater (dye background levels) Cells killed by paraformaldehyde fixation, heat, or ozone treatment (negative control) Freshly harvested cells (positive control). The original 10,000× concentration of dye has been reduced to 1× for both types of sample.
3. Add 20 µl of 1 mg/ml PI (10 µg/ml final) and vortex. 4. Incubate 30 min at room temperature in the dark. Calibrate flow cytometer 5. Run the water-only sample on a flow cytometer in order to localize the background noise due to the dyes and adjust the detection threshold to it. 6. Resolve viable, membrane-damaged, and membrane-compromised bacteria by defining three quadrants from the background noise limits as follows: a. Run the negative (dead) control with all parameters set on logarithmic scale. b. Adjust the PMT voltages as necessary to ensure that red-fluorescent cells (>620 nm) are to the right-hand side of the display, and that their green fluorescence (515 to 530 nm) is as weak as possible. c. Position the vertical axis of the quadrant at the limit between noise and dead cells (Fig. 11.15.1). Typical settings on the Cytoron Absolute Count instrument (Ortho Diagnostic Systems) are FS = 60 (/255), SS = 60 (/255), green PMT = 80 (/255), orange PMT = 100 (/255), and red PMT = 80 (/255).
7. Run the positive (live) control. If necessary, adjust the PMT voltages to ensure that the green-fluorescent cells (525 nm) are to the upper-left side of the display and that their red fluorescence signal is as weak as possible. Position the horizontal axis of the quadrant at the limit between noise and viable cells (Fig. 11.15.1).
Microbiological Applications
11.15.3 Current Protocols in Cytometry
Supplement 23
Green (SYBR Green) fluorescence (au)
A 1000 intact membrane
B damaged membranes
100
10 compromised membranes
1
noise noise C 1
D 10
100
1000
Red (PI) fluorescence (au)
Figure 11.15.2 Application of the NADS protocol to a sample containing two different E. coli cultures. Flow cytometric analysis yields this typical cytogram characterizing cells by their fluorescence. Quadrant A includes green-fluorescent cells stained by SYBR Green only and identified as viable because membrane integrity prevents uptake of PI. Quadrant B contains green- plus red-fluorescent cells stained by both SYBR Green and PI and identified as membrane-damaged cells because they enable various amounts of PI to penetrate the cell and bind to nucleic acid, inducing a corresponding increase of red fluorescence and decrease of green fluorescence, depending on the extent of FRET from SYBR Green to PI. Quadrant D includes red-fluorescent cells, which are identified as dead because the total quenching of SYBR Green fluorescence by PI reveals the presence of highly compromised membranes. Background noise is confined to quadrant C. Note that with the NADS protocol it is possible to distinguish the two E. coli cultures by their relative nucleic-acid content.
Analyze and resolve viable, membrane-damaged, and membrane-compromised bacteria 8. Analyze bacterial samples on the flow cytometer using the settings determined by the calibration steps. 9. Record the percentages, numbers, or abundances of viable (green only; upper-left quadrant), membrane-damaged (green and red; upper-right quadrant) and membranecompromised (red only; bottom-right quadrant) bacteria (Fig. 11.15.1). If the sample is too concentrated, dilute with fresh water or seawater as appropriate. 10. Clean cytometer by running successive volumes of 10% bleach, 70% ethanol, and water passed through a 0.2-µm filter until <1 events/sec are detectable. REAGENTS AND SOLUTIONS Use deionized, distilled water passed through a 0.2-µm filter in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Resolution of Viable and Membrane-Compromised Aquatic Bacteria
SYBR Green I or II working solution In a 1.5-ml microcentrifuge tube, dilute 10 µl SYBR Green I or 100 µl SYBR Green II (Molecular Probes) in 990 or 900 µl water, respectively. Store ≤1 month at 4°C in the dark.
11.15.4 Supplement 23
Current Protocols in Cytometry
COMMENTARY Background Information To better understand the functioning of the microbial ecosystem, it is necessary to investigate the different activities of the microorganisms at the cellular level. Usually, bulk activity measurements refer to total counts and imply the involvement of all cells. This is not correct for natural samples, because total counts may include ghost, dead, or damaged cells (Zweifel and Hagström, 1995; Gasol et al., 1999). It is therefore crucial to quantify the amounts of both viable and dead cells. The nucleic-acid double-staining (NADS) protocol developed by Barbesti and co-workers (Barbesti et al., 2000) for cultured bacteria in fresh water, and adapted to natural samples from fresh and marine environments (Grégori et al., 2001), was developed with this goal in mind. This protocol specifically addresses membrane integrity, which is a widely accepted criterion for viability (Nebe-von Caron and Badley, 1995; Joux and Lebaron, 2000). Indeed the loss of membrane integrity results, among other things, in the collapse of cellular energetics and active transports, quickly leading to cell death (Nebevon Caron and Badley, 1995). The principle of the NADS protocol described in this unit is based on the excitation energy transfer from an excited donor (SYBR Green) to an acceptor molecule (PI). This fluorescence resonance energy transfer (FRET) occurs when the distance between nucleic acid–bound molecules (PI and SYBR Green) is <70 Å because the emission spectrum of SYBR Green (maximum emission at 520 nm) covers the absorption spectrum of PI (maximum absorption at 535 nm). Thus, on the one hand, red-only fluorescing bacteria and cells emitting both green and red fluorescence represent the fractions of dead and damaged membrane bacteria respectively. On the other hand, green-only fluorescing bacteria represent the fraction of viable cells with preserved membrane integrity (Fig. 11.15.2).
Critical Parameters Flow cytometer calibration The maximal data rate at which the flow cytometer performs accurately is ∼1,000 events/sec for the Cytoron Absolute Count system (Ortho Diagnostic Systems), and ∼2,000/sec for the FACScalibur (Becton-Dickinson) and EPICS XL (Beckman Coulter) systems. It is always safe to keep the data rate below 1,000 particles/sec (Gasol and Del Gior-
gio, 2000). Analyze the sample, adjusting the flow rate and/or the cell concentration (by dilution) to avoid coincidence and saturation of the flow cytometer. Typically, the authors run samples for 60 sec at a flow rate of 60 µl/min and maintain a data rate below 1,000/sec by diluting samples if necessary. Threshold limit Set the discriminator to green and red fluorescences if possible (e.g., Cytoron Absolute Count flow cytometer) or on the FS or SS signals. Adjust the threshold to the background noise to improve the signal/noise ratio in order to better resolve the different clusters of bacteria and avoid instrument saturation. Preserving sample Determination of bacterial viability with the NADS protocol is performed only on fresh natural samples. The samples must be analyzed as soon as possible after collection. Because the analyses generally have to be done in the laboratory, the authors suggest keeping samples in the dark and cold (4°C) during transport in order to preserve their freshness.
Troubleshooting Optimal dye concentrations In natural samples containing insufficient concentrations of bacteria (i.e., <0.5 × 106 cells/ml), the background fluorescence of SYBR Green dyes and PI can be high enough to affect signal quality. To solve this problem, it is necessary to adapt the concentrations of the fluorescent probes to that of bacteria by decreasing the added amount of dye working solutions. For instance, a final PI concentration of 1 µg/ml and a final dilution of 1/100,000 (v/v) for the SYBR Green I commercial solution are optimal for freshwater samples with <0.5 × 106 bacteria/ml. For more concentrated samples (up to 20 × 106 bacteria/ml), 10 µg PI/ml and a final dilution of 1/10,000 (v/v) of SYBR Green I or 1/1000 (v/v) of SYBR Green II remain the best for bacterial cultures. Instrument sterility The 0.2-µm-filtered distilled water used as sheath fluid ensures the sterility of the flow cytometer and minimizes background particlescatter noise. Nevertheless, it is also important to clean the flow cytometer when the analyses are finished in order to remove residual bacteria
Microbiological Applications
11.15.5 Current Protocols in Cytometry
Supplement 23
or dye molecules from the fluidic system. The authors accomplish this by running the following solutions through the flow cytometer: (1) a 10% bleach solution, (2) a 70% ethanol solution, and (3) 0.2-µm-filtered distilled water. The flow cytometer is considered cleaned when the detection rate is less than 1 event/sec.
Anticipated Results The NADS protocol is an assay based on analytical flow cytometry that allows one to distinguish viable from membrane-damaged and membrane-compromised bacteria and to sort out noise and debris. Viable cells include both active (e.g., measurable metabolic activity) and inactive states (e.g., lack of enzyme activity in nutrient-depleted environments; Nebe-von Caron and Badley, 1995). This protocol should not be considered a strict live/dead assay, but rather as a proxy whose efficiency depends on the characteristics (cell type, physiologic state, and so on) of the cells occurring in natural samples where several populations in different physiologic and cytostructural terms can be simultaneously encountered. This large heterogeneity of bacteria in aquatic samples, where most of the species are still unknown, is the main limitation to the NADS assay as well as to other protocols using fluorochromes. Indeed, it must be kept in mind that some as-yetunknown species may not be consistent with observations made thus far regarding bacterial membrane permeability with respect to the fluorescent probes in use. Addressing membrane integrity, the NADS protocol provides resolution of bacteria into viable (active or inactive), membrane-damaged, and membrane-compromised groups, but other complementary approaches are needed to independently demonstrate the presence of specific activity or metabolism, as well as the capacity of a cell to undergo division (Lloyd and Hayes; 1995; McFeters et al., 1995; Porter et al., 1996; Grégori et al., 2001).
Time Considerations
Resolution of Viable and Membrane-Compromised Aquatic Bacteria
The NADS protocol can be achieved in <1 hr and yields both quantitative (i.e., cell abundance) and qualitative (i.e., viable, membranedamaged, and membrane-compromised bacteria; nucleic acid content) information. This protocol is compatible with routine sampling and analyses and with the high-frequency sampling required to perform spatial or temporal monitoring of bacteria in aquatic environments. High-frequency sampling is not feasible with other methods much more time-consuming
than flow cytometry, such as epifluorescence microscopy and acridine-orange or DAPI staining on membrane filters (Zimmerman and Meyer-Reil, 1974; Hobbie et al., 1977).
Literature Cited Barbesti, S., Citterio, S., Labra, M., Baroni, M.D., Neri, M.G., and Sgorbati, S. 2000. Two and three color fluorescence flow cytometric analysis of immunoidentified viable bacteria. Cytometry 40:214-218. Gasol, J.M. and Del Giorgio, P.A. 2000. Using flow cytometry for counting natural planktonic bacteria and understand the structure of planktonic bacterial communities. Scientia Marina 64:197224. Gasol, J.M., Zweifel, U.L., Peters, F., Fuhrman, J.A., and Hagström, A. 1999. Significance of size and nucleic acid content heterogeneity as measured by flow cytometry in natural planktonic bacteria. Appl. Environ. Microbiol. 65:44754483. Grégori, G., Citterio, S., Ghiani, A., Labra, M., Sgorbati, S., Brown, S., and Denis, M. 2001. Resolution of viable and membrane-compromised bacteria in freshwater and marine waters based on analytical flow cytometry and nucleic acid double staining. Appl. Environ. Microbiol. 67:4662-4670. Haugland, R.P. 1998. Handbook of fluorescent probes and research chemicals. Molecular Probes. Eugene, Ore. Hobbie, J.E., Daley, R.J., and Jasper, S. 1977. Use of nuclepore filters for counting bacteria by fluorescence microscopy. Appl. Environ. Microbiol. 33:1225-1228. Jones, K.H. and Senft, J.A. 1985. An improved method to determine cell viability by simultaneous staining with fluorescein diacetatepropidium iodide. J. Histochem. Cytochem. 33:77-79. Joux, F. and Lebaron, P. 2000. Use of fluorescent probes to assess physiological functions of bacteria at single-cell level. Microbes and Infection 2:1523-1535. Kirchman, D.L., Suzuki, Y., Garside, C., and Ducklow, H.W. 1991. High turnover rates of dissolved organic carbon during a spring phytoplankton bloom. Nature 352:612-614. Kogure, K., Simidu, U., and Taga, N. 1979. A tentative direct microscopic method for counting living marine bacteria. Can. J. Microbiol. 25:415420. Lebaron, P., Parthuisot, N., and Catala, P. 1998. Comparison of blue nucleic acid dyes for flow cytometric enumeration of bacteria in aquatic systems. Appl. Environ. Microbiol. 64:17251730. Lloyd, D. and Hayes, A.J. 1995. Vigor, vitality and viability of microorganisms. FEMS Microbiol. Lett. 133:1-7.
11.15.6 Supplement 23
Current Protocols in Cytometry
López-Amorós, R., Castel, S., Comas-Riu, J., and Vives-Rego, J. 1997. Assessment of E. coli and Salmonella viability and starvation by confocal laser microscopy and flow cytometry using Rhodamine 123, DiBAC4(3), propidium iodide and CTC. Cytometry 29:298-305. McFeters, G.A., Yu, F.P., Pyle, B.H., and Stewart, P.S. 1995. Physiological assessment of bacteria using fluorochromes. J. Microbiol. Meth. 21:113. Mitchell, P. 1961. Coupling of phosphorylation to electron and hydrogen transfer by chemi-osmotic type mechanism. Nature 191:144-148. Nebe-von Caron, G. and Badley, R.A. 1995. Viability assessment of bacteria in mixed populations using flow cytometry. J. Microsc. 179:55-66. Nicholls, D.G. 1982. Bioenergetics. An introduction to the chemiosmotic theory. Academic Press, London.
Williams, S.C., Hong, Y., Danavall, D.C.A., Howard-Jones, M.H., Gibson, D., Frisher, M.E., and Verity, P.G. 1998. Distinguishing between living and nonliving bacteria evolution of the vital stain propidium iodide and its combined use with molecular probes in aquatic samples. J. Microbiol. Meth. 32:225-236. Zimmerman, R. and Meyer-Reil, L. 1974. A new method for fluorescence staining of bacterial populations on membrane filters. Kiel. Meeresforsch. 30:24-27. Zweifel, U.L. and Hagström, A. 1995. Total counts of marine bacteria include a large fraction of non-nucleoid-containing bacteria (ghosts). Appl. Environ. Microbiol. 61:2180-2185.
Key Reference Barbesti et al. 2000. See above. The original presentation of the NADS protocol.
Pomeroy, L.R. 1984. Significance of microorganisms in carbon and energy flow in marine ecosystems In Current Perspectives in Microbial Ecology (M.J. Klug and C.A. Reddy, eds.) pp. 405-411. American Society for Microbiology, Washington, D.C.
Grégori et al. 2001. See above
Porter, J., Diaper, J., Edwards, C., and Pickup, R. 1995. Direct measurements of natural planktonic bacterial community viability by flow cytometry. Appl. Environ. Microbiol. 61:2783-2786.
Internet Resources
Porter, J., Deere, D., Pickup, R., and Edwards, C. 1996. Fluorescent probes and flow cytometry new insights into environmental bacteriology. Cytometry 23:91-96. Sgorbati, S., Barbesti, S., Citterio, S., Bestetti, G., and De Vecchi, R. 1996. Characterization of number, DNA content, viability and cell size of bacteria from natural environments using DAPI PI dual staining and flow cytometry. Minerva Biotec. 8:9-15. Trousselier, M., Courties, C., and Vaquer, A. 1993. Recent applications of flow cytometry in aquatic microbial ecology. Biol. Cell 78:111-121.
This article describes the adaptation of the NADS protocol described by Barbesti et al. (2000) to natural fresh and marine water samples.
http://www.probes.com/servlets/pis This Web site displays technical bulletins on Molecular Probes fluorochromes with useful storage and handling information and suggestions for their use.
Contributed by Gérald Grégori Purdue University Cytometry Laboratories West Lafayette, Indiana Michel Denis Laboratoire d’Océanographie et de Biogéochimie Parc Scientifique et Technologique de Luminy Marseille Cedex, France Sergio Sgorbati and Sandra Citterio Università di Milano-Bicocca Milan, Italy
The authors wish to acknowledge the support of the French-Italian Integrated Action Program “GALILEE” and of the Council of the Provence-Alpes-Côte d’Azur Region. They also acknowledge Sam Duncan, who provided E. coli cultures.
Microbiological Applications
11.15.7 Current Protocols in Cytometry
Supplement 23
Functional Assays of Oxidative Stress Using Genetically Engineered Escherichia coli Strains
UNIT 11.16
Oxidative stress is the result of cellular imbalance between reactive oxygen species (ROS) and antioxidant defenses, in favor of the former. Oxidative stress may be induced in bacteria by many different exogenous biocidal agents and is also involved in endogenous oxygen metabolism. The oxyR operon is a main sensor of oxidative stress in bacteria and regulates the expression of several genes coding for antioxidant enzymes. Compared to wild-type cells, oxyR-deficient bacteria show enhanced sensitivity to oxidative stress and increased accumulation of intracellular ROS when exposed to prooxidants. Intracellular levels of ROS may be determined by flow cytometry using chemically reduced fluorogenic substrates, such as dihydroethidium (hydroethidine, HE) or 2′,7′-dichlorofluorescin diacetate (DCFH-DA). Flow cytometric functional assays in live bacteria are limited in part by the cell wall, which impairs penetration of vital dyes, thus imposing a need for permeabilization procedures. These manipulations are time consuming, may affect cell physiology, and provoke cell aggregation or lysis. Escherichia coli B WP2 strains have been used for mutagenic assays because they possess an altered cell-wall lipopolysaccharide that leads to increased membrane permeability. Flow cytometric analysis of WP2 strains has been shown to be a convenient alternative for cytometric assays of bacterial function (Herrera et al., 2002). This unit presents Basic Protocols and a Support Protocol convenient for flow cytometric studies of intracellular oxidative stress in E. coli B WP2 strains, wild-type (strain IC188) or deficient in the oxyR (∆oxyR, strain IC203) function, using ROS-sensitive fluorogenic substrates. The Basic Protocols are designed for flow cytometric analysis of oxidative stress in both WP2 strains, where IC188 should be considered as a control strain for IC203. Basic Protocols 1 and 2 describe the flow cytometric assay of intracellular superoxide anion and peroxides, respectively, using both WP2 strains. The Support Protocol describes the preparation of phage C21 stock for bacterial verification, the procedure for verifying the E. coli B WP2 phenotype, and the procedure for verifying the deficiency in oxyR function. DETECTION OF INTRACELLULAR SUPEROXIDE ANION IN E. COLI B WP2 STRAINS OxyR deficiency decreases the capacity for detoxifying reactive oxygen species (ROS) produced by endogenous metabolism and by exogenous oxidants. Thus, the IC203 strain generates higher levels of superoxide than the IC188 strain when cells are exposed to a source of superoxide anion. This process may be revealed using the fluorogenic probe dihydroethidium.
BASIC PROTOCOL 1
Materials Overnight bacterial culture (see recipe) LB medium (see recipe) Test substance(s) 1 mg/ml plumbagin (see recipe) PBS (APPENDIX 2A) 1 mg/ml HE (dihydroethidium; see recipe) Microbiological Applications Contributed by Guadalupe Herrera, Alicia Martínez, José-Enrique O’Connor, and Manuel Blanco Current Protocols in Cytometry (2003) 11.16.1-11.16.9 Copyright © 2003 by John Wiley & Sons, Inc.
11.16.1 Supplement 24
37°C incubator (Heraeus Sepatech) Spectrophotometer Flow cytometer with an air-cooled argon laser (488 nm, 15 mW), equipped with light-scatter detectors that measure forward (FS) and side scatter (SS) and with filter set to collect orange fluorescence (575 to 625 nm) Obtain single-cell suspension and expose to test substance(s) 1. Dilute the overnight bacterial culture in fresh LB medium and keep at 37°C until the optical density reading at 600 nm (OD600) is 0.6. 2. Incubate with test substance. As a positive control, incubate 500 µl cell culture for 1 hr at 37°C with different concentrations (20, 30, 40, and 50 µg/ml) of 1 mg/ml plumbagin, a generator of superoxide anion. 3. Centrifuge 10 min at 2800 × g, room temperature, and resuspend in 500 µl PBS. 4. Dilute the sample ten-fold in PBS. Load cells with superoxide-sensitive fluorescent dye 5. Incubate 500-µl aliquots of diluted samples with 5 µl of 1 mg/ml HE (10 µg/ml, final concentration) 10 min at room temperature in the dark. Acquire fluorescence data on flow cytometer 6. Set up flow cytometer. Use logarithmic amplifications and low flow rate settings. Collect orange ethidium fluorescence emission (575 to 625 nm). Exclude large aggregates by gating on the histogram of log forward light scatter versus log side light scatter (see Fig. 11.16.1A). BASIC PROTOCOL 2
DETECTION OF INTRACELLULAR PEROXIDES IN E. COLI B WP2 STRAINS OxyR deficiency decreases the capacity for detoxifying reactive oxygen species (ROS) produced by endogenous metabolism and by exogenous oxidants. Thus, the IC203 strain generates higher levels of peroxides than the IC188 strain when cells are exposed to a source of oxidative stress. This process may be revealed using the fluorogenic probe 2′,7′-dichlorofluorescin diacetate. Materials Overnight bacterial culture (see recipe) LB medium (see recipe) Test substance(s) 1 M hydrogen peroxide (see recipe) 100 mM t-butyl hydroperoxide (see recipe) PBS (APPENDIX 2A) 1 mg/ml DCFH-DA (2′,7′-dichlorofluorescin diacetate; see recipe) 37°C incubator (Heraeus Sepatech) Spectrophotometer Flow cytometer with an air-cooled argon laser (488 nm, 15 mW), equipped with light-scatter detectors that measure forward (FS) and side scatter (SS) and with filter set to collect green fluorescence (∼525 nm)
Assays of Oxidative Stress Using Genetically Engineered E. coli
11.16.2 Supplement 24
Current Protocols in Cytometry
A
B 10
4
100
Cell count
Side scatter
103
102
101
0
100 100
101
102 103 Forward scatter
104
101
102
103
104
Ethidium fluorescence
D 30 25 20 15 10 5 0
IC188 IC203
0
25 Plumbagin (µg/ml)
50
Fluorescein fluorescence
Ethidium fluorescence
C
100
10
IC188
8
IC203
6 4 2 0
0
7.5 10 1 2.5 5 t -Butyl hydroperoxide (mM)
Figure 11.16.1 (A) Flow cytometric features of morphology of E. coli B. Biparametric dotplots of forward scatter versus side scatter of IC188 strain. The rectangular boxes show the scatter-based gate used for aggregate exclusion and fluorescence acquisition. IC203 shows a similar pattern. (B) Flow cytometric assessment of increased sensitivity to oxidative stress of IC203, deficient in oxyR, compared to IC188. Both strains were treated for 1 hr with 50 µg/ml plumbagin, a superoxide-anion donor, and then stained for intracellular superoxide with dihydroethidium, following Basic Protocol 1. Solid histogram indicates IC188 and empty histogram indicates IC203. (C) Dose-response analysis of exogenous superoxide donor plumbagin for intracellular generation of superoxide by strains IC188 (black bars) and IC203 (open bars). E. coli cultures were treated with plumbagin and stained with dihydroethidium as described in Basic Protocol 1. The results are the mean ±S.D. of three experiments. (D) Dose-response analysis of exogenous t-butyl hydroperoxide for intracellular generation of peroxides by strains IC188 (black bars) and IC203 (open bars). E. coli cultures were treated with t-butyl hydroperoxide and stained with DCFH-DA as described in Basic Protocol 2. The results are the mean ±S.D. of three experiments.
Obtain single-cell suspension and expose to test substance(s) 1. Dilute the overnight culture in fresh LB medium and keep at 37°C until OD600 is 0.6. 2. Incubate with test substance. As a positive control, incubate 500 µl cell culture for 60 min at 37°C with different concentrations of hydrogen peroxide (0 to 100 mM) or t-butyl hydroperoxide (0 to 10 mM). 3. Centrifuge 10 min at 2800 × g, room temperature, and resuspend in 500 µl PBS. 4. Dilute the sample ten-fold in PBS.
Microbiological Applications
11.16.3 Current Protocols in Cytometry
Supplement 24
Load cells with peroxide-sensitive fluorescent dye 5. Incubate 500-µl aliquots of diluted samples with 5 µl of 1 mg/ml DCFH-DA (10 µg/ml, final concentration) 10 min at room temperature in the dark. Acquire fluorescence data on flow cytometer 6. Set up flow cytometer. Use logarithmic amplifications and low flow-rate settings. Collect fluorescein fluorescence emission at 525 nm. Exclude large aggregates by gating on the histogram of log forward light scatter versus log side light scatter (see Fig. 11.16.1 A). SUPPORT PROTOCOL
This Support Protocol describes how to obtain bacterial cultures in optimal conditions and to verify their correct phenotype. Materials Bacterial strains IC188 (WP2 uvrA/pKM101; ATCC), IC203 (WP2 uvrA∆oxyR30/pKM101; obtain from Manuel Blanco; [email protected]), and GY783 (C21-sensitive; obtain from Manuel Blanco) Phage C21 (obtain from Manuel Blanco; [email protected]) LB medium (see recipe) Molten top agar (see recipe) LA medium dishes (see recipe) 1 M hydrogen peroxide (see recipe) 37°C incubator (Heraeus Sepatech) 6-mm Whatman paper disks To prepare suspension of phage C21 for E. coli B verification 1a. Incubate 0.2 ml E. coli GY783 strain (sensitive to phage C21) with 0.2 ml of phage C21 suspension 20 min at 37°C. 2a. Add 7 ml of preheated (37°C) LB medium and incubate with shaking at 37°C until bacterial lysis is produced. For obtaining a good phage C21 stock, the inoculated culture must grow until an OD600 = 0.4 before the culture lysis is observed (decreased culture turbidity and appearance of cellular debris).
3a. Centrifuge 10 min at 12,800 × g, 4°C, and collect the supernatant. To verify bacterial E. coli B WP2 phenotype 1b. Mix 100 µl of each overnight culture with 3 ml molten top agar. 2b. Pour this mixture into an LA medium dish. 3b. Inoculate the cell lawn with a 20-µl drop of phage C21. 4b. Incubate overnight at 37°C. The phenotype determined by the outer membrane of the two E. coli B WP2 strains is verified by its sensitivity to lysis by the phage C21. A lysis zone where the phage was placed must be visible in E. coli B bacterial cultures after 12 hr.
Assays of Oxidative Stress Using Genetically Engineered E. coli
11.16.4 Supplement 24
Current Protocols in Cytometry
To verify bacterial ∆oxyR phenotype 1c. Mix 100 µl of each overnight culture with 3 ml molten top agar. 2c. Pour this mixture into an LA medium dish. 3c. Place a sterile 6-mm Whatman paper disk soaked with 10 µl of 1 M hydrogen peroxide on the cell lawn. 4c. Incubate overnight at 37°C. The phenotype determined by the oxyR deficiency is verified by the sensitivity to oxidative stress. IC203 is more sensitive to hydrogen peroxide than IC188 and under these conditions, a 25-mm diameter inhibition zone in IC203 versus an 11-mm diameter inhibition zone in IC188 is expected.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
t-Butyl hydroperoxide, 100 mM Dilute t-butyl hydroperoxide (70% w/v, mol. wt. 90; Sigma-Aldrich) to 100 mM in sterile distilled water. Keep at 4°C until use. Prepare this solution daily and discard unused portion.
DCFH-DA, 1 mg/ml Dissolve DCFH-DA (2′,7′-dichlorofluorescin diacetate; mol. wt. 487; Molecular Probes) at 1 mg/ml in DMSO. Store dry and protected from light up to six months at −80°C. Dye should be stored in 50-ìl aliquots to avoid refreezing
Dihydroethidium (HE), 1 mg/ml Dissolve dihydroethidium (mol. wt. 315; Molecular Probes) at 1 mg/ml in DMSO. Store dry and protected from light in 50-µl aliquots up to six months at −80°C. Do not refreeze. Hydrogen peroxide, 1M Dilute hydrogen peroxide (30% w/w, mol. wt. ∼34; Sigma-Aldrich) to 1 M in sterile distilled water. Prepare this solution daily and discard unused portion.
LA (LB agar) medium 10 g bacto tryptone (Difco) 5 g yeast extract (Difco) 5 g NaCl 20 g agar (Difco) Distilled water to 1 liter Mix and sterilize by autoclaving 30 min on liquid cycle. Store up to 6 months in the dark at room temperature. For preparing LA dishes: Heat the autoclaved LA medium in a microwave until liquid and pour into 10-cm petri dishes. Store the solidified plates inverted up to 2 weeks at 4°C.
Microbiological Applications
11.16.5 Current Protocols in Cytometry
Supplement 24
LB (Luria broth) liquid medium 10 g bacto tryptone (Difco) 5 g yeast extract (Difco) 5 g NaCl Distilled water to 1 liter Dissolve and sterilize by autoclaving 30 min on liquid cycle. Store up to 3 months in the dark at room temperature Molten top agar 5 g NaCl 6 g agar (Difco) Distilled water to 1 liter Dissolve and sterilize by autoclaving 30 min on liquid cycle. Store up to 3 months in the dark at room temperature. Overnight cultures Inoculate 10 ml LB medium (see recipe) with 100 µl of a frozen permanent culture and incubate 12 hr at 37°C. These cultures can be stored up to 3 days at 4°C. To store frozen permanent cultures: Add 90 µl DMSO to 1 ml of an overnight culture and keep at −80°C. Under these conditions, the strains can be kept for several years.
Plumbagin, 1 mg/ml Dissolve plumbagin (5-hydroxy-2-methyl-1,4-naphthoquinone; mol. wt. ∼188; Sigma-Aldrich) at 1 mg/ml in DMSO. Store 500-µl aliquots protected from light up to 4 weeks at −20°C. Do not refreeze. COMMENTARY Background Information
Assays of Oxidative Stress Using Genetically Engineered E. coli
Because of the key role of oxidative stress in a multiplicity of physiological, pathological, and toxicological issues (Droge, 2002), the analysis of intracellular ROS is one of the most relevant applications of current functional flow cytometry (Gorman et al., 1997; O’Connor et al., 2001). Intracellular levels of ROS may be determined by flow cytometry using chemically reduced fluorogenic substrates, like HE or DCFH-DA. HE is cell permeant and can undergo a twoelectron oxidation to form the DNA-binding fluorophore ethidium bromide. The reaction is relatively specific for superoxide anion, with minimal oxidation induced by hydrogen peroxide, peroxynitrite, or hypochlorous acid. The intracellular oxidation of HE to ethidium by superoxide anion has been analyzed by flow cytometry (Carter et al., 1994). The oxidation of DCFH to the fluorescent compound dichlorofluorescein has been used extensively in the detection of oxidants produced during the respiratory burst in intact cells using flow cytometry (Rothe and Valet, 1990). The early interpretation of dichlorofluorescein
fluorescence as a specific marker for quantitative intracellular hydrogen peroxide formation is currently enlarged to a general marker of peroxidative activity (Tarpey and Fridovich, 2001). The regulation of antioxidant responses in E. coli has been studied extensively (Harvey et al., 2000; Zengh and Storz, 2000; Pomposiello and Demple, 2002). The regulon oxyR is one of the main sensors of oxidative stress in E. coli, in addition to other gene loci like soxR and soxS (Zengh et al., 2001; Droge, 2002). At least nine genes coding for proteins that are induced by low concentrations of hydrogen peroxide are under the control of oxyR and include hydroperoxidase I (katG), alkyl hydroperoxide reductase (ahpCF), glutathione reductase (gorA), and glutaredoxin I (grxA) as well as a non-specific DNA-binding protein (dps) and a small untranslated regulatory RNA (oxyS). OxyR is a transcriptional activator existing in a reduced and an oxidized form, the latter being able to activate transcription (Storz et al., 1990). Exposure to hydrogen peroxide or induction of oxidative shift in the thiol/disulfide redox status results in oxidized oxyR binding to all oxyR-
11.16.6 Supplement 24
Current Protocols in Cytometry
regulated promoters. In this way, oxyR protects E. coli against normally lethal concentrations of hydrogen peroxide or against thermal killing. The oxyR deficiency prevents the synthesis of antioxidant enzymes induced by oxidative stress and results in increased intracellular content of ROS (Fig. 11.16.1 B and C) and enhanced mutagenesis arising from DNA lesions caused by ROS (Blanco et al., 1998a,b; Martínez et al., 2000). Flow cytometry has become a choice methodology for bacterial research (Davey and Kell, 1996; Alvarez-Barrientos et al., 2000). A main drawback for flow cytometric studies of bacterial function is the structural barrier of the cell wall that limits the penetration of vital dyes and imposes procedures for transient or permanent permeabilization. These manipulations are time consuming and may affect the physiology of the bacterial cell, and/or result in cell aggregation or lysis. The outer membrane in gram-negative bacteria is fairly impermeable to hydrophilic compounds and moderately impermeable to hydrophobic ones. The WP2 strains used in these protocols are derived from E. coli B and thus lack some portions of the lipopolysaccharide molecule (Janson et al., 1981; Nikaido and Vaara, 1987; Herrera et al., 1993). Such intrinsic changes determine enhanced permeability of these strains to many molecules sharing structural features with common fluorescent dyes; e.g., one or more aromatic rings, heterocyclic rings, polar groups, low and medium molecular weight (Herrera et al., 1993; Martínez et al., 2000). E. coli WP2 are widely used in mutagenesis and toxicity assays (Blanco et al., 1998a,b; Martínez et al., 2000) and have been included in official guidelines for the testing of chemicals (OECD, 1998). Additional advantages of WP2 strains derive from the simplicity of its genetic manipulation. IC188 and IC203 are WP2 uvrA/pKM101 and IC203 carries the del(oxyR30) mutation. Both tester strains carry the trpE65 ochre mutation and can be reverted to Trp+ by base substitutions at the A:T base pair at the trpE65 site or at extragenic ochre suppressor loci containing A:T or G:C base pairs at the reversion site. The two strains contain the SOS mutagenesis proteins UmuDC, encoded in the chromosome, and MucAB, encoded by the pKM101 plasmid. Strain IC203 is constructed by first introducing argE btuB::Tn 10 into the WP2 strain by P1 transduction, selecting for Tcr transductants, and then introducing argE+ del(oxyR30) by another P1
transduction, selecting for Arg+ Tcs (Herrera et al., 1993; Blanco et al., 1998a,b; Martínez et al., 2001). The authors have shown the usefulness of such strains as an advantageous alternative for flow cytometric functional assays of bacterial cell function on the basis of an adequate penetration of fluorochromes, making permeabilization steps unnecessary (Herrera et al., 2002). Thus, IC188 and IC203 might be used for bacterial studies in which classical parameters of mutagenesis can be correlated with intracellular levels of ROS, or for specific studies on the functions of oxyR. From another point of view, IC203 may be the basis for sensitive bioassays of oxidant or antioxidant processes in different experimental settings.
Critical Parameters An essential step is to verify periodically the strain phenotype, especially when bacterial cultures are started from stocks. Strain verification for cell wall permeability and oxyR deletion are described in the Support Protocol. In order to assure optimal and homogeneous metabolic activity of bacterial cells, the experiments should be performed on exponentialphase cultures. A correct setup of the flow cytometer is essential, as it is in general for all microbiological procedures. The small size of bacterial particles demands sensitive optics and appropriate settings to discriminate them from background noise. Calibration of the cytometer with small-size beads is recommended and a low flow rate or a sufficient sample dilution should be used to avoid particle coincidence at the laser interrogation point. To prevent both contamination of the sample and background noise, it is strongly recommended that all susceptible material be autoclaved. In addition, sheath fluid should be used fresh and filtered through a 0.22-µm filter before use. Since living bacteria will end up in the waste, appropriate disinfectant solution should be added to the waste container. The controls that should be performed in each experiment include positive controls of prooxidant substances, as suggested in the Basic Protocols, as well as general viability controls when experimental treatments may result in cell death.
Troubleshooting Oxidant solutions should be freshly prepared and properly stored, especially the unstable compounds like hydrogen peroxide.
Microbiological Applications
11.16.7 Current Protocols in Cytometry
Supplement 24
Prooxidant decay is usually detected by lower fluorescent staining with ROS-sensitive fluorescent probes. Culture contamination with other microorganisms is usually found by changes in the forward- and side-scatter properties of the sample. Confirmation of culture contamination can be performed by standard plate techniques. Contamination of the cytometer hydraulic system (sheath container, tubing) with other microorganisms is indicated by changes in the light-scatter distributions, as mentioned above, and also by increased data rate under the usual flow setup. Confirmation of cytometer contamination can be performed by standard plate techniques using the suspected fluid. Inadequate cleaning of the flow cytometer, especially when applications with particles larger than bacteria (e.g., eukaryotic cells) are run in the same equipment, is clearly evidenced by the appearance of events with high forwardand side-scatter intensities. In general, it is highly recommended that frequent cleaning be implemented in all flow cytometry systems.
Anticipated Results Basic Protocols 1 and 2 provide a simple and rapid system for detecting intracellular ROS in strains IC188 and IC203. Figure 11.16.1 exemplifies the use of these strains for studies related to oxidative stress. Panel A shows the morphological gating to include single bacteria in the assay. Panels B, C, and D exemplify the higher sensitivity to oxidative stress of IC203 compared to IC188 when Basic Protocols 1 and 2 are applied. Under these conditions, IC188 may be used as a negative control for IC203 and thus the implication of oxyR functions, or more generally, the involvement of oxidative stress in a given process, can be studied. Also, panels C and D are examples of the application of IC188 and IC203 to determine the capacity of ROS generation by a given molecule or condition. In this case, the application is exemplified by dose-response analysis using model prooxidants, such as plumbagin, a superoxide donor (panel C), and t-butyl hydroperoxide, a wellcharacterized lipid peroxide (panel D).
Time Considerations
Assays of Oxidative Stress Using Genetically Engineered E. coli
The duration of the biological treatment steps is obviously variable. Time of exposure to prooxidants may be acute or sustained and one might choose, instead, to determine endogenous ROS levels over a prolonged period of time under given culture conditions. The sam-
ple processing time is ∼30 min, including sample running in the flow cytometer. There are some experimental steps where one may stop the experiment. Overnight cultures can be kept 2 to 3 days at 4°C. Bacterial cell suspensions following treatment can be pelleted and kept ∼1 hr at 4°C before further processing for fluorescent staining and flow cytometry.
Literature Cited Alvarez-Barrientos, A.M., Arroyo, J., Cantón, R., Nombela, C., and Sánchez-Pérez, M. 2000. Applications of flow cytometry to clinical microbiology. Clin. Microbiol. Rev. 13:167-195. Blanco, M., Urios, A., and Martínez, A. 1998a. New Escherichia coli WP2 tester strains highly sensitive to reversion by oxidative mutagens. Mutat. Res. 413:95-101. Blanco, M., Martínez, A., Urios, A., Herrera, G., and O’Connor, J.E. 1998b. Detection of oxidative mutagenesis by isoniazid and other hydrazine derivatives in Escherichia coli WP2 tester strain IC203, deficient in OxyR: Strong protective effects of rat liver S9. Mutat. Res. 417:39-46. Carter, W.O., Narayanan, P.K., and Robinson, J.P. 1994. Intracellular hydrogen peroxide and superoxide anion detection in endothelial cells. J. Leukoc. Biol. 55:253–258. Davey, H.M. and Kell, D.B. 1996. Flow cytometry and cell sorting of heterogeneous microbial populations: The importance of single-cell analysis. Microbiol. Rev. 60:641-696. Droge, W. 2002. Free radicals in the physiological control of cell function. Physiol. Rev. 82:47-95. Gorman, A.M., Samali, A., McGowan, A.J., and Cotter, T.G. 1997. Use of flow cytometry techniques in studying mechanisms of apoptosis in leukemic cells. Cytometry 29:97-105. Harvey, E.M., Merchant, K., and Stamler, J.S. 2000. Nitrosation and oxidation in the regulation of gene expression. FASEB J. 14:1889-1900. Herrera, G., Urios, A., Aleixandre, V., and Blanco, M. 1993. Mutability by polyciclic hydrocarbons is improved in derivatives of Escherichia coli WP2 uvrA with increased permeability. Mutat. Res. 301:1-5. Herrera, G., Martínez, A., Blanco, M., and O’Connor, J.E. 2002. Assessment of Escherichia coli B with enhanced permeability to fluorochromes for flow cytometric assays of bacterial cell function. Cytometry 49:62-69. Janson, P.E., Lindberg, B., and Wollin, R. 1981. Structural studies on the hexose region of the core in lipopolysaccharides from Enterobacteriaceae. Eur. J. Biochem. 115:571-577. Jernaes, M.W. and Steen, H.B. 1994. Staining of Escherichia coli for flow cytometry: Influx and efflux of ethidium bromide. Cytometry 17:302309.
11.16.8 Supplement 24
Current Protocols in Cytometry
Martínez, A., Urios, A. and Blanco, M. 2000. Mutagenicity of 80 chemicals in Escherichia coli tester strains IC203, deficient in OxyR, and its oxyR+ parent WP2 uvrA/pKM101: Detection of 31 oxidative mutagens. Mutat. Res. 467:41-53. Nikaido, H. and Vaara, M. 1987. Outer membrane. In Escherichia coli and Sa lmo nella typhimurium: Cellular and Molecular Biology. (F.C. Neidhardt, J.L. Ingraham, K.B. Low, B. Magasanik, M. Schaecchter, and H.E. Umbarger, eds.) pp. 7-22. American Society for Microbiology, Washington, D.C. O’Connor, J.E., Callaghan, R.C., Escudero, M., Herrera, G., Martínez, A., Monteiro, M.C., and Montolíu, H. 2001. The relevance of flow cytometry for biochemical analysis. IUBMB Life 51:231-239. Organization for Economic Cooperation and Development. 1998. OECD Guideline for Testing of Chemicals: Bacterial Reverse Mutation Test. In Ninth Addendum to the OECD Guidelines for the Testing of Chemicals. pp. 1-11. OECD, Paris. Pomposiello, P.J. and Demple, B. 2002. Global adjustment of microbial physiology during free radical stress. Adv. Microb. Physiol. 46:319-341. Rothe, G. and Valet, G. 1990. Flow cytometric analysis of respiratory burst activity in phagocytes with hydroethidine and 2′,7′-dichlorofluorescin. J. Leukoc. Biol. 47:440-448.
Storz, G., Tartaglia, L.A., and Ames, B.N. 1990. Transcriptional regulator of oxidative stress–inducible genes: Direct activation by oxidation. Science 248:189-194. Tarpey, M.M. and Fridovich, I. 2001. Methods of detection of vascular reactive species nitric oxide, superoxide, hydrogen peroxide, and peroxynitrite. Circ. Res. 89:224-236. Zheng, M. and Storz, G. 2000. Redox sensing by prokaryotic transcription factors. Biochem. Pharmacol. 59:1-6. Zheng, M., Wang, X., Templeton, L.J., Smulski, D.R., LaRosa, R.A., and Storz, G. 2001. DNA microarray-mediated transcriptional profiling of the Escherichia coli response to hidrogen peroxide. J. Bacteriol. 183:4562-4570.
Contributed by Guadalupe Herrera, Alicia Martínez, and José-Enrique O’Connor Universidad de Valencia Valencia, Spain Manuel Blanco Instituto de Investigaciones Citológicas Fundación Valenciana de Investigaciones Biomédicas Valencia, Spain
The authors wish to acknowledge the support by grants from IZASA, S.A., Generalitat Valenciana (GV-DV-20-125-96) and Fondo de Investigaciones Sanitarias (01/0151).
Microbiological Applications
11.16.9 Current Protocols in Cytometry
Supplement 24
Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
UNIT 11.17
With the advent of new cytometric technologies and the development of an increasing number of available pathogen-specific antibodies, as well as the accumulation of a large body of microbial sequence information, there have been a rising number of reports in the literature pertaining to pathogen detection by flow cytometry. The food and pharmaceutical industries each have a “Big Four,” i.e., the four primary pathogens of concern to the particular industry. The food industry tests for the presence of Escherichia coli O157:H7, Salmonella, Listeria, and Campylobacter, while the pharmaceutical industry focuses on Staphylococcus aureus, Pseudomonas aeruginosa, Salmonella, and E. coli. Environmental water-related testing targets fecal-contamination indicators like E. coli or Enterococcus and parasites such as Cryptosporidium and Giardia oocysts. All these tests focus on the prevention of disease. This unit presents a series of protocols that have been used in the detection and enumeration of a variety of specific pathogens. It concentrates on designing assays that use pure cultures initially to validate the detection and enumeration capabilities of the cytometer. The Commentary provides recommendations on how to adapt these methods to analyze real-world samples. Basic Protocol 1 outlines a procedure for setting up cytometer instruments for microbial analysis. Labeling methods presented include a direct no-wash approach (Basic Protocol 2), a direct wash approach (Alternate Protocol 1), indirect labeling (Alternate Protocol 2), and fluorescence in situ hybridization (Basic Protocol 3) using peptide nucleic-acid (PNA) probes. A Support Protocol outlines the steps to establish a linear correlation between traditional microbial plate counts and cytometric enumeration. Data are presented on many of the pathogens of concern in food, pharmaceuticals, and the environment. CAUTION: Take appropriate precautions when working with pathogenic organisms. Analyze only in a closed system and handle waste accordingly. Reference the National Institute of Health publication on Biosafety in Microbiological and Biomedical Laboratories (see Internet Resources) for appropriate biosafety-level handling procedures.
PREPARATION OF CYTOMETER FOR DETECTION OF MICROORGANISMS
BASIC PROTOCOL 1
Because bacteria are so much smaller than mammalian cells, the detection and analysis of these microorganisms pose special challenges for cytometry. If the instrument used is not specifically designed for microbial analysis (UNIT 1.11), one must determine that it is capable of such measurements and optimize its performance. This protocol focuses on instrumentation with red laser excitation capabilities (635 nm) and a closed quartz flowcell option to minimize aerosol formation and enhance the sensitivity of detection. The protocol is designed to assist the operator in setting up and validating the system’s ability to detect and enumerate specific pathogens. UNIT 11.4 provides complementary information on how to sterilize a system and verify small particle detection prior to bacteria analysis. UNIT 11.6 provides additional information on how to utilize a flow cytometer to enumerate bacteria as a function of the ratio of the number of events in a bead region versus the bacteria counts and directly via metered sample flow. Deep Red beads (Molecular Probes) are used as standards to provide a reference point allowing the operator to optimize the instrument for pathogen detection. Microbiological Applications Contributed by Kristi R. Harkins and Kelley Harrigan Current Protocols in Cytometry (2004) 11.17.1-11.17.20 C 2004 by John Wiley & Sons, Inc. Copyright
11.17.1 Supplement 29
Materials Sheath fluid: 25 µM cetyltrimethylammonium bromide (CTAB) in deionized water (or prepared as recommended by flow cytometer manufacturer), filtered through 0.22-µm filter 2.5-µm Deep Red beads (Molecular Probes): 0.04%, 0.2%, and 0.8% intensities Flow cytometer with 635-nm excitation and filters for detection of far-red fluorescence at 700 nm Luer-lock fitting (male and female luer-to-tubing adapters) Sterivex GP filter (Millipore) 1. Fill sheath tank of flow cytometer with appropriate sheath fluid. Cut the sheath fluid line, insert the male and female luer-lock fittings, and install the in-line Sterivex filter. Run sheath fluid through the lines for 1 hr with a first installation of the filter. The CTAB sheath solution is recommended for the RBD cytometer used by the authors. Other vendors will have their own sheath formulations.
2. Prepare a mixture of Deep Red beads by adding 50 µl of each intensity bead stock to 500 ml filtered sheath fluid. According to Molecular Probes, the stock concentration of the beads is 6 × 107 beads/ml. The specified dilution will yield ∼103 to 104 beads/ml for each intensity population.
3. Run the mixed bead sample on the cytometer. Use the beam-shaping lens, which provides a small round beam spot, and trigger on fluorescence. Collect side scatter at 635 ± 15 nm and far-red fluorescence at 700 ± 15 nm, using logarithmic scales for both. 4. Adjust the PMT high-voltage settings to bring the side-scatter population into the third decade of the log scale and all three bead intensitites onto the log fluorescence scale
Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
Figure 11.17.1 Fluorescence and side scatter from the 2.5-µm mixed Deep Red beads used for optical alignment purposes. Beads were analyzed at 635 nm (25 mW from diode laser) with fluorescence mean channels of 11, 88, and 443 (R1, R2, and R3, respectively). The combined side-scatter mean was 507. The trigger was set on fluorescence with a discriminator setting up to channel 7. Note the good separation between the fluorescence of the dim 0.04% fluorescence mean and the noise level (<7) and the absence of debris signal below the three bead populations. Cytometric counts in each box were 1334, 990, and 1261 (R1, R2, and R3, respectively) and based upon analysis of 0.25 ml of the mixed bead solution. Counts outside the boxes were equal to 13/0.25 ml. This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www.interscience.wiley.com/c p/colorfigures.htm.
11.17.2 Supplement 29
Current Protocols in Cytometry
(Fig. 11.17.1). Adjust the fluorescence threshold to just below the 0.04% intensity bead distribution. These instrument settings will provide a sufficient dynamic range of detection for a broad array of probe-labeled microbes.
5. Create analysis windows around each population and record the mean fluorescence, side-scatter value, and particle count for each bead population with time. This provides a starting instrument protocol for analyzing the labeled pathogens.
DIRECT LABELING OF A PATHOGEN (DILUTION) Initial experiments should use pure cultures of the pathogen of interest to set up the instrumentation and validate the labeling and enumeration method. Although the following protocol is presented specifically for Salmonella detection, other pathogens can be and have been tested with this method by using the antibody information (source, dye/protein ratio, and concentration) and incubation times provided in Table 11.17.1. Fluorescence and side-scatter ratio information is also provided relative to the 0.2% intensity beads, which will allow the operator to anticipate where the labeled pathogens might fall on the two-parameter histogram.
BASIC PROTOCOL 2
Materials Goat anti-Salmonella antibody (Kirkegaard & Perry) Alexa Fluor 647 (Molecular Probes) or Cy5 (Amersham Biotech) dye conjugation kit Glycerol PBS-BSA: 0.05% (w/v) bovine serum albumin in PBS (APPENDIX 2A), filtered through 0.22-µm filter Overnight culture (APPENDIX 3F) of Salmonella typhimurium (ATCC #14028) Tryptic soy broth (e.g., Difco) Bacterial sample for analysis Millex-GV 0.2-µm syringe filter unit (Millipore) Flow cytometer with 635-nm excitation and filters for collection of far-red fluorescence at 700 nm 12 × 75–mm sterile test tubes Prepare the antibody conjugate 1. Perform the conjugation of the goat anti-Salmonella antibody with the Alexa Fluor 67 or Cy5 fluorochrome, purify the conjugate from the free fluorochrome, and quantitate the dye-to-protein ratio and recovery of the antibody according to manufacturer’s instructions for the conjugation kit. Store the conjugate at a concentration of 0.5 to 1 mg/ml in 50% (v/v) glycerol up to 1 year at −20◦ C. The dye-to-protein ratio should range from 2 to 4 for optimal labeling without affecting the antigen binding site of the antibody.
Prepare the antibody working stock 2. Dilute the antibody conjugate in PBS-BSA to a concentration of 2 µg/ml. Filter the antibody through a 0.2-µm filter to remove antibody aggregates and store in a dark container at room temperature until ready for flow cytometry. All antibodies should be filtered just before setting up the assays. Aggregates will cause high backgrounds in the negative samples. Microbiological Applications
11.17.3 Current Protocols in Cytometry
Supplement 29
Supplement 29
1 to 0.5
Listeria 1 to 0.24 monocytogenes (Alexa Fluor 647)
Enterococcus (Alexa Fluor 647)
1 to 0.04
1 to 0.49
Salmonella worthington (Cy5)
1 to 13.05
1 to 0.02
1 to 5.25
Salmonella typhimurium (Cy5)
Cryptosporidium oocysts (Alexa Fluor 647)
1 to 0.01
1 to 2.56
Salmonella typhimurium (Alexa Fluor 647)
1 to 8.88
1 to 0.03
Staphylococcus aureus 1 to 1.39 (Alexa Fluor 647)
E. coli O157:H7 (Alexa Fluor 647)
1 to 0.03
1 to 0.71
Pseudomonas aeruginosa (Cy5)
1 to 0.05
1 to 0.65
1 to 0.01
1 to 0.02
Fluorescence Scatter
Organism (Dye)
Mean ratios (0.2% beads:bacteria)
Advanced Analytical
Waterborne, Inc.
Kirkegaard & Perry
Kirkegaard & Perry
Kirkegaard & Perry
Kirkegaard & Perry
Kirkegaard & Perry
Virostat
Cygnus
Primary Ab
—
—
—
— —
—
4.65 µg/ml
2 µg/ml
—
1 µg/ml
—
2.5 µg/ml
1 µg/100 µl
Jackson Immunoresearch
2 µg/100 µl
—
1 µg/100 µl
1 µg/ml
Jackson 2 µg/100 µl ImmunoResearch
—
2 µg/ml
Molecular Probes 10 µg/ml
Secondary Ab 1 µg/ml
Primary Ab
Antibody concentration(s)
Jackson 2 µg/ml ImmunoResearch
Secondary Ab
Antibody source(s)
Table 11.17.1 Relative Fluorescence Intensities and Key Labeling Conditions for Antibody-Based Pathogen Detection
Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
11.17.4
Current Protocols in Cytometry
1.83
1.83
3.5
2.3
2.6
2.6
4.0
5.1
2.6
Dye/protein ratios
30 min
30 min
15 min
15 min
20 min
20 min
30 min
20 min
10 min
Primary Ab
—
—
—
—
20 min
20 min
—
20 min
10 min
Secondary Ab
Incubation time(s)
Process samples for flow cytometry 3. Set up the flow cytometer for bacteria detection (see Basic Protocol 1). 4. Starting with an overnight culture of S. typhimurium in tryptic soy broth, perform 1:10 serial dilutions in PBS-BSA, to achieve a concentration of 106 bacteria/ml (which would typically involve a 10−3 dilution of a static overnight culture). Use 12 × 75–mm sterile test tubes for this and all subsequent dilutions. This dilution of a pure culture can be used as a positive control when creating a new protocol, and later for testing real-world samples.
5. Perform 1:10 serial dilutions of the bacterial growth medium alone in PBS-BSA, to achieve the same final end dilution as in step 4 (e.g., 10−3 ). This is the negative control. The 10−3 dilution is typically used for analysis of a pure culture. A minimum dilution of 1:10 is typically required to reduce refractive index effects by the medium on the sample core. The dilution necessary for detection and enumeration using non-pure cultures from environmental, food, or pharmaceutical samples on the cytometer will depend upon the level of contamination in the original sample and the growth kinetics in the enrichment medium. The upper limit of enumeration is 106 bacteria/ml. Most pure bacterial cultures will reach 109 cells/ml before becoming static, so a 10−3 to 10−4 dilution may be required after a long enrichment period (e.g., an overnight pure culture).
6. Perform a final 1:10 dilution of bacteria (106 cells/ml) and negative control into the 2 µg/ml antibody working solution prepared in step 2 (e.g., 0.1 ml of each diluted sample or control into 0.9 ml antibody working solution). 7. Incubate bacteria and negative controls with antibody 30 min in the dark at room temperature. The labeled samples can be fixed prior to analysis with 0.25% (v/v) formaldehyde for sample preservation.
8. Analyze the bacteria on the flow cytometer using the instrument settings from the mixed beads (see Basic Protocol 1).
Figure 11.17.2 Dot-density display of the log fluorescence versus log side scatter for Salmonella typhimurium bacteria labeled using the direct method (Basic Protocol 2). S. typhimurium counts were 12624 (box 1) per 0.25 ml with a fluorescence mean of 42. This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www.interscience.wiley.com/c p/colorfigures.htm.
Microbiological Applications
11.17.5 Current Protocols in Cytometry
Supplement 29
9. Modify the instrument settings as needed to move and identify the bacteria population above the threshold. Create a box around the population and save the protocol. See Figure 11.17.2 for two-parameter histogram of S. typhimurium directly labeled with an Alexa Fluor 647 antibody conjugate.
10. Analyze the negative control on the flow cytometer using the saved protocol. Analysis of the negative control samples should yield no or very low counts in the pathogen analysis box. The negative control background signal in the box defines the lower limit of enumeration. SUPPORT PROTOCOL
CORRELATION OF FLOW CYTOMETRIC DATA WITH TRADITIONAL PLATE-COUNT ENUMERATION When new methods are being developed for pathogen detection, it is important that the researcher validate the correlation between traditional microbial counting methods and cytometric enumeration. This is accomplished by verifying the linear correlation between standard plate counts (SPCs) and cytometric counts from a serial dilution study. This protocol outlines how to verify this correlation.
Additional Materials (also see Basic Protocol 2) Solid medium (APPENDIX 3E) appropriate for bacteria of interest (use selective media for non-pure culture or nonselective media for pure cultures; for pure Salmonella culture use nonselective tryptic soy broth) Petri dishes, 100 × 15 mm Additional reagents and equipment for plating bacteria (APPENDIX 3G) 1. Using an overnight pure culture, perform 1:10 serial dilutions in PBS-BSA from 10−2 to 10−7 . Prepare enough sample volume to perform both plating and flow cytometric analysis. 2. Dilute a portion of each sample dilution from step 1 1:10 in filtered antibody solution (e.g., 0.1 ml diluted sample in 0.9 ml antibody working stock). Incubate and analyze as in Basic Protocol 2. 3. While samples for cytometry are incubating, plate the diluted samples from step 1. Either the spread plate (APPENDIX 3G) or the pour plate technique may be used. Samples should be plated without delay, as some bacterial growth may occur in the diluted state. Do not plate antibody-containing samples, because an antibody killing effect has been observed for some species. The addition of formaldehyde (0.25% v/v) to the labeled samples will preserve sample state prior to analysis.
4. Incubate the plates at the appropriate temperature until the colonies are large enough to count. 5. Make the appropriate calculations to account for dilution factor, volume plated, and volume analyzed. Compare the box counts on the flow cytometer with plate counts. Count-correlation data for Salmonella typhimurium are presented in Figure 11.17.3. Note the linear regression analysis correlation coefficent of 0.9664 with a lower limit of detection at 102 cells/ml. Microbiologists typically look for counts that are within 0.5 log of plate count values as acceptable for alternative enumeration methods. Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
11.17.6 Supplement 29
Current Protocols in Cytometry
Figure 11.17.3 Ten-fold serial dilution series and count correlation analysis of Salmonella typhimurium (Basic Protocol 2). The linear regression R-value is equal to 0.9664 (n = 13). Cytometric lower limit of detection is 102 cells/ml.
DIRECT ANTIBODY LABELING (CELL PELLET) This approach is recommended when the sample requires a preliminary centrifugation step either to concentrate the bacteria or to remove small particulates in the culture medium prior to labeling with antibody conjugate.
ALTERNATE PROTOCOL 1
Additional Materials (also see Basic Protocol 2) Microcentrifuge capable of 20,800 × g 2-ml microcentrifuge tubes, sterile 1. Prepare antibody and control samples (negative and positive) as in Basic Protocol 2, steps 1 to 5. In the case of real-world samples, variability in growth kinetics may or may not make it necessary to dilute the samples prior to centrifugation. Samples with very low growth kinetics (low cell number) make it possible to use centrifugation to remove the medium and allow antibody labeling of all the cells in the sample. Samples with very high growth kinetics (high cell number) require a dilution series prior to centrifugation to meet the 106 maximum cell concentration per ml.
2. Microcentrifuge the bacterial samples 5 min at 10,800 to 20,800 × g. The viability of some bacteria species is affected by high g-force centrifugation. It may be necessary to evaluate centrifugation speed versus recovery for each new pathogen genus. The addition of 0.05% BSA to the diluent improves both recovery and population viability. For samples with high particulate content, it may be possible to microcentrifuge at significantly lower g-force (e.g., 960 × g) with effective recovery in the pellet (the large particles pull the bacteria into the pellet).
3. Resuspend the pellet in 1 ml filtered antibody working stock solution. 4. Incubate all samples with antibody 30 min in the dark at room temperature. 5. Analyze on the flow cytometer using the instrument settings created for the mixed beads (see Basic Protocol 1). 6. Modify the instrument settings as needed to move and identify the bacteria population above the threshold. Create a box around the population and save the protocol. Analysis of the negative control samples should yield no or very low counts in the pathogen analysis box. The negative control background signal in the box defines the lower limit of enumeration.
Microbiological Applications
11.17.7 Current Protocols in Cytometry
Supplement 29
ALTERNATE PROTOCOL 2
INDIRECT PATHOGEN LABELING This protocol is recommended to increase the fluorescence intensity of the microbial population by using a primary/secondary labeling approach. When bacteria are labeled using antibodies, a variety of factors can affect surface antigen density (growth condition, species, serovars, and strain variation). Therefore, a secondary labeling method has been developed to increase the fluorescence of bacteria with lower antigen densities. Some Salmonella serovars for which this has been effective include S. worthington and S. schalwijk.
Additional Materials (also see Basic Protocol 2): Goat anti-Salmonella primary antibody (Kirkegaard & Perry), unconjugated Rabbit anti-goat (or any other anti-goat) secondary antibody labeled with Cy5 (Jackson ImmunoResearch) Microcentrifuge capable of 20,800 × g 2-ml microcentrifuge tubes, sterile Prepare the antibodies 1. Dilute the goat anti-Salmonella primary antibody to a concentration of 2 µg/0.1 ml in PBS-BSA. 2. Filter this primary antibody through a 0.2-µm filter to remove antibody aggregates and store in dark container at room temperature. 3. Dilute the rabbit anti-goat secondary antibody conjugate to 1 µg/0.1 ml in PBS-BSA. 4. Filter the secondary antibody conjugate through a 0.2-µm filter to remove antibody aggregates and store in dark container at room temperature. All antibodies should be filtered just before setting up the assays for the day. Aggregates will cause high backgrounds in the negative samples.
Process samples for flow cytometry 5. Set up the flow cytometer for bacteria detection (see Basic Protocol 1). 6. Using the overnight culture of S. typhimurium, perform a 1:10 dilution in PBS-BSA. Make a 1:10 dilution in PBS-BSA of bacterial growth medium alone as a negative control. 7. Add 5 µl diluted sample for analysis to 95 µl filtered primary antibody solution from step 2. Also add 5 µl diluted medium to a separate 95 µl of antibody solution as a negative control. For non-pure cultures, the dilution necessary for detection and enumeration on the cytometer will depend the level of contamination in the original sample and the growth kinetics in the enrichment medium.
8. Incubate the sample 20 min in the dark at room temperature. 9. Add 0.9 ml PBS-BSA to bring the sample volume to 1 ml. Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
10. Microcentrifuge samples 5 min at 10,800 to 20,800 × g and remove the supernatants. Resuspend each of the pellets in 0.1 ml filtered secondary antibody conjugate solution from step 4. 11. Incubate the sample 20 min in the dark at room temperature. 12. Add 0.9 ml PBS-BSA to bring the sample volume to 1 ml.
11.17.8 Supplement 29
Current Protocols in Cytometry
Figure 11.17.4 Dot-density plots of the log fluorescence versus log side scatter for (A) Salmonella typhimurium (14395/0.25 ml with fluorescence mean 70) and (B) Salmonella worthington (18931/0.25 ml with fluorescence mean 7) labeled using the indirect method (Alternate Protocol 2). This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www.interscience.wiley.com/c p/colorfigures.htm.
13. Analyze on the flow cytometer using the instrument settings from the mixed beads (see Basic Protocol 1). 14. Modify the instrument settings as needed to move and identify the bacteria population above the threshold. Create a box around the population and save the protocol. Analysis of the control samples should yield no or very low counts in the pathogen analysis box. The background signal counts within the box will define the lower limit of enumeration. The fluorescence intensity obtained by the indirect method will typically be a minimum of 2-fold brighter than that of the direct method. Note the Salmonella data presented in Table 11.17.1. Salmonella worthington has almost 10-fold less available surface antigen than S. typhimurium, and the indirect method allows the researcher to bring the population on scale above the fluorescence threshold (see Fig. 11.17.4, panels A and B).
IN SITU HYBRIDIZATION OF rRNA The small subunit ribosomal RNA (rRNA) provides an abundant intracellular target that is commonly used as a phylogenetic marker in bacteria classification. This protocol provides a FISH procedure for human pathogen detection using peptide nucleic-acid (PNA) probes in place of nucleic-acid probes. Because PNA probes have a neutral backbone, the kinetics of binding and the stability of the hybridization are significantly better than what is obtained with standard nucleic-acid probes. The following method is presented specifically for Salmonella detection. Other pathogens can be and have been tested with this method using the information from Table 11.17.2, which provides referenced probe sequence data. The fluorescence and side scatter ratios relative to the 0.2% intensity beads are also provided. This information will allow the operator to anticipate where the labeled pathogens might fall on the two-parameter histogram.
BASIC PROTOCOL 3
Materials Lyophilized Cy5-PNA Sal probe (Applied Biosystems; store at −20◦ C; see probe sequence in Table 11.17.2) Dimethylformamide (DMF)
Microbiological Applications
11.17.9 Current Protocols in Cytometry
Supplement 29
Table 11.17.2 Relative Fluorescence Intensities and Sequence Information for rRNA-Based Cy5 5 -End-Labeled Fluorescence In Situ Hybridization of Various Pathogens
Organism
Mean ratios (0.2% beads:bacteria)
Probe sequence
Salmonella species
1 to 0.81
5 TAA GCC GGG ATG GC 3
Listeria species
1 to 0.68
E. coli
1 to 0.1
Pseudomonas aeruginosa Staphylococcus aureus
Not done Not done
Enterococcus faecalis
1 to 0.16
Cryptosporidium oocysts
Not done
Sequence reference Perry-O’Keefe et al. (2001)
5 AAG GGA CAA GCA GT 3
Brehm-Stecher et al. (2002)
Perry-O’Keefe et al. (2001)
Perry-O’Keefe et al. (2001)
5 TCA ATG AGC AAA GGT 3
5 CTG AAT CCA GGA GCA 3
5 GCT TCT CGT CCG TTC 3
5 CCT CTG ATG GGT AGG TT 3
5 CGG TTA TCC ATG TAA GTA AAG 3
Perry-O’Keefe et al. (2001)
Frahm et al. (1998) Vesey et al. (1998)
16% (w/v) formaldehyde (methanol-free, Polysciences) Log-phase culture (APPENDIX 3F) of Salmonella typhimurium (ATCC #14028) PBS-BSA: 0.05% (w/v) bovine serum albumin in PBS (APPENDIX 2A), filtered through 0.22-µm filter Hybridization buffer (see recipe), optimized for stringency (see Critical Parameters) Wash buffer (see recipe), optimized for stringency (see Critical Parameters) Temperature-controlled heat block Microcentrifuge capable of 20,800 × g Flow cytometer with 635-nm excitation and filters to collect emission at 700 nm Prepare probe 1. Rehydrate the lyophilized PNA probe to a 25 µM stock in 50% dimethylformamide (DMF) in deionized water. Store stock up to 6 months at −20◦ C. Applied Biosystems provides volume information for creating a 100-µM stock; add four times the recommended volume 50% DMF to achieve a 25-µM stock. Cy5–end labeled PNA probes tend to form aggregates with time in DMF; therefore, long-term storage in a lyophilized state is recommended.
2. Heat the 25-µM probe stock 10 min at 55◦ C in a temperature-controlled heat block. Microcentrifuge probe 3 min at 20,800 × g to remove aggregates. Transfer the supernatant to a fresh tube for use and re-storage. Aggregates will cause high background signal in the negative control samples.
Process samples for flow cytometry 3. Add 0.25 ml of 16% formaldehyde to 0.75 ml fresh log-phase S. typhimurium culture (4% final formaldehyde). Let sit 15 min. Some form of fixation is required to preserve cell structure. Formaldehyde is typically used, but it has been recently observed that gram-positive bacteria in FISH assays respond better to a fixation step using ∼50% (v/v) ethanol.
Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
4. Dilute fixed cells down to a concentration of 10−2 to 10−3 of the original concentration (final concentration <107 /ml) in PBS-BSA. Microcentrifuge 0.1 ml of this dilution 3 min at 20,800 × g, and remove supernatant. Resuspend pellet in 100 µl hybridization buffer.
11.17.10 Supplement 29
Current Protocols in Cytometry
5. Add 1 µl Sal PNA probe and incubate 30 min at 55◦ C protected from light. 6. Add 500 µl wash buffer and incubate 30 min at 55◦ C protected from light. 7. Microcentrifuge 5 min at 20,800 × g and remove supernatant. Resuspend pellet in 1 ml wash buffer and analyze on the flow cytometer using the mixed beads protocol (see Basic Protocol 1). 8. Modify instrument setting as needed to move and identify the bacterial population above the threshold. Create a box around the population and save the protocol. Figure 11.17.5 provides two-parameter histograms and information on the mean fluorescence values of multiple Salmonella serovars labeled by this protocol. Note the uniform intensity levels of each of the variants.
Figure 11.17.5 Dot-density plots for multiple Salmonella serovars using rRNA PNA probes (Basic Protocol 3). Note the uniformity in the labeling process based upon fluorescence intensity (all populations are within the box). (A) S. typhimurium: in-box count 25298, fluorescence mean 18; (B) S. enteritidis: in-box count 12607, fluorescence mean 16; (C) S. choleraesuis: in-box count 7570, fluorescence mean 12; and (D) S. paratyphi A: in-box count 8123, fluorescence mean 18. This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www.interscience.wiley.com/c p/colorfigures.htm. Microbiological Applications
11.17.11 Current Protocols in Cytometry
Supplement 29
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Hybridization buffer 10% (w/v) dextran sulfate 0% to 50% (v/v) formamide 0.1 M NaCl 15 mM sodium citrate 0.05% (w/v) sodium dodecyl sulfate (SDS) 20 mM Tris·Cl, pH 7 to 9 (APPENDIX 2A) See Critical Parameters for optimization of hybridization conditions.
Wash buffer 5 mM EDTA 100 mM NaCl 0.15% (w/v) sodium dodecyl sulfate (SDS) 20 mM Tris·Cl, pH 7 to 9 (APPENDIX 2A) See Critical Parameters for optimization of stringency conditions.
COMMENTARY Background Information
Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
With today’s available technologies, the small size and lower nucleic-acid content (approximately three orders of magnitude lower than that of mammalian cells) of bacterial cells are becoming less of a problem for flow cytometric analysis. In 2000 a special issue of the Journal of Microbiological Methods (Vol. 42) summarized select presentations from a conference entitled “Analysis of Microbial Cells at the Single Cell Level— Why, How, and When?” discussing cytometric techniques (flow and image). In that issue are articles demonstrating the adaptation of methods and technologies originally designed for mammalian cell analysis to the field of microbiology. In addition, there is today a broad array of materials available for specific pathogen labeling that includes commercially available antibodies and extensive sequence database information for designing molecular probes. The discussion in this unit focuses on the application of such methods to disease prevention through food, water, and raw-material screening prior to human consumption (also see UNIT 11.6). Other applications that may include the use of flow cytometry as a diagnostic tool for human disease are not covered in this unit. Historically, most microbial testing has been performed on the final product. In recent years, both food and pharmaceutical industries have implemented a hazard analysis
critical control points (HACCP) system for raw-material and in-process testing for pathogens. Why is this important? The CDC estimates that in the United States, 76 million people become ill each year, with 300,000 hospitalizations and 5,000 deaths associated with foodborne illnesses (http://www.cdc.gov/foodsafety/). Tortorello et al. (1998) used flow cytometry to detect a single E. coli O157:H7 pathogen in apple juice following a 10-hr enrichment period. From 1999 to 2000, 39 outbreaks of drinking water–associated gastroenteritis were reported by 25 states, resulting in an estimated 2,068 illnesses. Approximately half of these outbreaks were caused by known infectious etiology: 35% by parasites, 45% by bacteria, and 20% by viruses (http://www.cdc.gov/ mmwr/PDF/SS/SS5108.pdf). The flow cytometric detection of Bacillus anthracis spores (Dang et al., 2001) demonstrates the utility of antibody-based labeling for flow cytometric detection of pathogens that are of key concern in the environment today. There is also a body of literature on the detection of Cryptosporidium and Giardia oocysts using flow cytometry (Hoffman et al., 1997; Vesey et al., 1993, 1994). In the clinical realm, Alvarez-Barriento et al. (2000) published an article entitled “Applications of Flow Cytometry to Clinical Microbiology.” Fluorescent antibodies have
11.17.12 Supplement 29
Current Protocols in Cytometry
been used for the flow cytometric detection of Haemophilus (Srikumar et al., 1992), Salmonella (Clark and Pinder, 1998), Mycobacterium (Ozanne et al., 1996), Brucella (Bowden et al., 1995), Pseudomonas aeruginosa (Hughes et al., 1996), Legionella (Ingram et al., 1982), and Bacteroides fragilis (Lutton et al., 1991). The concept of using flow cytometry to analyze a natural cell suspension, e.g., bacteria, is not new, but instruments devoted entirely to bacteria detection are rare. The RBD flow cytometer used to generate the example data presented in this unit is designed with the sole purpose of accurately measuring low numbers of bacteria and parasite oocysts. A red diode laser provides 635-nm excitation light, optimal for the excitation of Cy5 (Amersham) and Alexa Fluor 647 (Molecular Probes). The selection of a longer excitation wavelength reduces the natural fluorescence of proteins and pigments that may be found in food and pharmaceutical samples. An increasing number of vendors offer pathogen-specific antibodies. Many of the vendors are listed in Table 11.17.1, with contact information in the SUPPLIERS APPENDIX at the end of this volume. Both polyclonal and monoclonal antibodies to a variety of different microbes can be found. However, polyclonal antibodies are typically needed to provide sufficient signal for flow cytometric detection. Antibodies can be purchased in pure unconjugated form, as enzyme conjugates for ELISA assay, and in some cases conjugated to fluorescein or biotin, depending upon the vendor. The use of 635-nm excitation will require the researcher to perform a conjugation to either Cy5 or Alexa Fluor 647, or to employ the indirect labeling method and purchase a commercially available secondary antibody conjugate. As a potential secondary approach to labeling specific pathogens, rRNA-specific (PNA) probes have been shown to be sensitive and specific. Stender et al. (2002) recently published a good method article on the use of PNA for rapid microbiology. Techniques for FISH (fluorescence in situ hybridization) labeling of bacteria cells have been reported in the literature (Amann et al., 1995), including papers on flow cytometric detection of FISH-labeled microbes (Wallner et al., 1993, Zoetendal et al., 2002). This method involves a formaldehyde fixation followed by permeabilization with solvent or detergent during incubation of the cells with probe in a hybridization buffer. The primary advantage to this approach is
uniform labeling of log-phase cells, independent of species or serotype difference for a genus-specific probe. It is difficult to find a surface protein that provides similar characteristics in labeling intensity and uniformity across species. The primary disadvantages of this method are the additional preparation time and additional centrifugation steps, which increase the risk of cell loss during sample processing. The advantage of using flow cytometry for assays of this nature stems from the single-cell analysis and enumeration capabilities. From a practical standpoint, assays may be completed within a shorter enrichment time frame because of the potential ability of the flow cytometer to detect lower numbers of cells. The time saved can be translated into significant savings for the manufacturer or producer. Typically, during the testing time, the product either remains in inventory or may be shipped at risk, which can lead to costly recalls. For example, in July 2002, the USDA (United States Department of Agriculture) recalled 19 million pounds of ground beef products because of the potential E. coli O157:H7 health risk to the population. From a research standpoint, flow cytometry offers the researcher the option to analyze other cellular parameters simultaneously within specific pathogens to better understand the physiology of the cell. Real-world sample issues The focus of this unit is on the detection of microbes from cultures growing in specific pathogen enrichment media. Bacteria must be viable and culturable to be detected in this assay process. Additionally, it has been noted that slow-growing or static microbes have a lower level of rRNA per cell, which leads to a dimmer fluorescence signal (Amann et al., 1995). The authors have observed that the FISH fluorescence detection levels are significantly higher in cells that are in log-phase growth and that transferring static cells to an enrichment medium for as little as 20 min prior to fixation significantly augments the fluorescence signal. The RBD 2100 cytometer (Advanced Analytical Technologies) has been used to detect low-level contamination—1 to 4 colonyforming units (cfu)—on sponges used to sample hog carcasses following an 8-hr enrichment of Salmonella in buffered peptone water. Similar detection results have been achieved for E. coli O157:H7 in 65 g of ground beef enriched for 7 hr. While the specific details of preparing matrix-intense samples for analysis
Microbiological Applications
11.17.13 Current Protocols in Cytometry
Supplement 29
were not covered in these protocols, food or pharmaceutical samples are typically diluted into a defined volume of enrichment medium for a specific pathogen following USDA/FSIS or USP guidelines (e.g., 65 g beef per 585 ml of pathogen-specific enrichment medium for E. coli O157:H7 testing), followed by a culture period that can range from 18 to 72 hr depending upon the microorganism. The large particulates remaining in these samples can be removed via a series of approaches that may include step filtration first through a 40-µm mesh and subsequently through a smaller filter mesh (e.g., 3-µm). Fine particles can also cause analysis problems, in which case centrifugation can be used to pellet the bacteria, leaving the fine particles in the supernatant (also see UNIT 11.6). Some assay formats will require the exact enumeration of pathogens within a specific sample volume and at levels significantly lower than those presented in this unit’s data sets. In the case of environmental water samples, it is typical to have an action level defined. Samples that have a specific bacteria concentration above this level may require the implementation of an action plan such as closing a recreational beach owing to the presence of high numbers of organisms (>400 E. coli per 100 ml or >10 Enterococcus per 100 ml) indicative of fecal contamination in ocean water. In these scenarios, it may be necessary to concentrate and remove matrices from the sample prior to labeling, analysis, and enumeration. Concentration steps may involve filtration and recovery, centrifugation, or immunomagnetic separation. In some cases (e.g., Cryptosporidium oocysts), large volumes (10 liters) of raw water sample must be concentrated into a 10-ml volume. This process is followed by immunomagnetic separation, removal of the beads from the surface of the oocysts by acid treatment, and labeling with a fluorescent antibody for microscope enumeration.
Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
Alternative methods A variety of rapid microbial testing technologies exist in the marketplace today. Polymerase chain reaction (PCR)–based assays and antibody- or probe capture–based systems continue to require an enrichment time of 24 to 48 hr prior to the rapid assay of a minimum of 104 bacteria per ml. Most of the rapid pathogen detection methods range in cost from $5 to $10 per sample. Expenditures for the capital equipment required to interface with these test kits range from $0 (lateral flow devices) to more than $100,000 (PCR system). In contrast, tra-
ditional microbial tests can be conducted for less than $1 per test, if one does not take into account the value of the time involved in waiting for the colonies to grow on the culture plates. Culture-based: Traditional culture-based methods may require exact enumeration. For example, a set volume of drinking water or environmental water is membrane-filtered, followed by culture of the collection membrane on selective media and visual colony enumeration 24 to 48 hr later. In the case of oocysts, visual enumeration is done under the microscope from the immunomagnetic separation (IMS)–treated filtrate. The majority of these methods are outlined as established EPA methods (see http://www.epa.gov/nerlcwww/). In the case of presence/absence assays, there is a zero tolerance for the specific pathogen. This requires that a defined weight or volume of the test material be cultured in selective broths for 24 to 48 hr, followed by plating onto solid selective media for another 24 hr, and finally secondary verification of the pathogen presence using PCR or antibody-based latex agglutination assays (see http://www.fsis.usda. gov/Ophs/Microlab/Mlgchp15.Pdf). Antibody-based: The majority of the rapid pathogen detection test kits on the market today involve some form of antibody labeling of antigens associated with each pathogen, followed by an indirect measurement of color, fluorescence, or chemiluminescence indicator process. A list of some of these assays has been published for the food industry (Feng, 1998) and is available at http://www.cfsan.fda.gov/∼ebam/bam-a1. html. Molecular-based: PCR-based technology requires selective enrichment and isolation/purification of the nucleic-acid material from the sample prior to assay to eliminate materials that can interfere with the assay performance. Probe-capture technologies involve cell lysis and incubation with a nucleicacid probe for capture. A secondary probe is then added that is specific for a different sequence. This secondary probe contains a reactive molecule, providing an assay format that can be indirectly detected through chemiluminescence, for example. Both PCR and probe-capture assays represent an indirect approach to pathogen detection. Fluorescence microscopy has been used in conjunction with FISH assays for the direct detection of Staphylococcus aureus from blood cultures (Oliveira et al., 2002). Chemiluminescent in situ hybridization (CISH) was used to detect
11.17.14 Supplement 29
Current Protocols in Cytometry
Pseudomonas aeruginosa in bottled water (Stender et al., 2002) and E. coli in municipal water samples (Stender et al., 2001) using a 5-hr to 24-hr enrichment time frame.
Critical Parameters and Troubleshooting Antibody assays Specificity is defined here as the absence of false positives in an absence/presence assay format. The presence of nonpathogenic organisms that bind to a pathogen-specific polyclonal antibody may generate a false positive, depending upon the degree of binding that occurs. The presence of such organisms will also depend upon the sample matrix tested. Uncooked meat typically contains a fairly high and varied natural microbial count, while surface swab samples in recently cleaned processing areas will have a fairly low microbial count. The assay specificity is thus defined within the matrix to be tested. When a higher than acceptable level of false positives exists within an assay format, the growth conditions can be altered in an attempt to reduce the growth of the unknown, the unknown organism can be isolated and used to adsorb out the nonspecific antibody component, or additives that can selectively suppress a subset of bacteria types may be used, if available. Another option would be to test antibody sources from alternative vendors or consider the use of a mixture of multiple monoclonal antibodies to achieve improved specificity and sufficient fluorescence intensity. Sensitivity is the absence of false negatives, which can occur for a variety of reasons. First, the assay conditions may not be optimal for recovery of the target organism. Make sure that the medium used is correct and that the enrichment time is sufficient for the target organism to grow to a detectable level. Inoculating the matrix sample with one to two cfu of staticphase cells typically provides a sufficient test of enrichment recovery time frames. In some cases, the microbes are subjected to stresses that would be typical for the matrix sample (e.g., heat in the case of steam-cleaned carcasses, cold in the case of stored perishables, or high salt in the case of preserved samples). Secondly, when working with complex matrices (e.g., ground raw meat), it is important to make sure that the antibody conjugate does not become the limiting reagent as a result of nonspecific binding to debris in the sample. A symptom of this would be high background signal in a negative sample. The addition of
protein-based blocking agents (e.g., BSA, casein, or gelatin) may reduce this nonspecific binding. The addition of more antibody may also ameliorate this problem. When testing for the sensitivity of an assay, it is important to conduct side-by-side comparisons with a traditional method (i.e., a USDA/FSIS- or FDArecommended protocol) to make sure that all positive samples are being detected, until confidence in the assay is achieved.
PNA FISH assays The design of the PNA probe is an important first step in developing a FISH assay. A large body of searchable microbial rRNA sequence information can be found at http://www.psb.ugent.be/rRNA/ and a number of software packages can be used to assist in the probe design—e.g., Advanced Blast (http://www.ncbi.nlm.nih.gov/BLAST/) and DNASTAR. Applied Biosystems (http://www.appliedbiosystems.com/support/ pnadesigner.cfm) provides design guidelines and software for evaluating PNA probe sequence suitability. In general the sequence length should be 12 to 17 bases and the purine percentage should be <60% of the total bases. Purine-rich PNA probes tend to aggregate and exhibit low solubility in an aqueous phase. Also, keep the maximum purine stretch to four in a row, three in a row for guanines. If this is an issue, consider probing the complementary strand. Avoid self-complementary sequences with inverse repeats, hairpins, and palindromes. These types of probes are prone to aggregate because PNA/PNA interactions are stronger than PNA/NA (nucleic acid) interactions. Because of the PNA probe’s uncharged state, the melting temperature of a PNA/NA duplex will be higher than that of the corresponding NA-NA duplex. The use of lower salt concentrations in the hybridization conditions further increases the melting temperature differences, providing for very stable hybridization formations. Hybridization buffers are typically composed of a buffered (pH 7 to 9) low-salt (100 mM) solution with denaturant present (SDS and/or formamide). Buffer composition and incubation time and temperature are evaluated to achieve the brightest specific fluorescence intensity labeling on the cytometer. The hydrophobic nature of the probe enhances uptake through the hydrophobic bacterial cell wall, and the low-salt conditions reduce the rRNA secondary structure (Perry-O’Keefe et al., 2001). These conditions act to reduce the time
Microbiological Applications
11.17.15 Current Protocols in Cytometry
Supplement 29
typically required for PNA/RNA hybridization in comparison to DNA/RNA hybridization. FISH assays can be made more specific by adjusting the wash step conditions. After hybridization, it is important to remove all the PNA that may be partially hybridized to nontarget sequences. This can be achieved in the wash step by increasing the temperature, pH, and/or denaturant concentration to increase the binding stringency. Under these conditions, the tolerance for mismatches will be low, and the specificity will be high. The absence of signal from populations can sometimes be attributed to slow or static growth of the population. This can lead to sensitivity issues in presence/absence assays as defined by the presence of false negatives. In this case, the microbes may be nonviable, or viable but not culturable. Specialized media have been developed to assist microbes in recovery and growth for certain assay types. Detection limits and count correlations The key advantage of this technology lies in early pathogen detection at lower levels than the alternative methods. It is important to analyze the appropriate negative control samples containing medium and/or environmental matrix to determine the degree of background signal within the pathogen analysis windows established on the cytometer. Background signal can be generated as a result of antibody or probe aggregates, nonspecific binding of the antibody or probe to environmental debris, or, in some cases, poor system cleaning. Reagent aggregation and system cleaning can be evaluated by analyzing the reagents in buffer only, without the real-world sample, using the same protocol as for the assay. Several factors can affect count correlations. If the antibody binds to dead cells, then the cultured numbers will be lower than the cytometric counts. If the rRNA probe does not bind to sufficient rRNA in the case of static or slow-growing cells, then cultured numbers may be higher with time than the cytometric counts. If the protocol involves centrifugation steps where cell losses may occur, it will be important to determine the degree of error associated with cell losses from multiple preparations of the same sample by different individuals using different equipment in different laboratory settings. Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
Anticipated Results The success of each potential application will depend on the instrument cleanliness (large and small particulates removed and no
evidence of bacteria growth in the sheath and sample lines) and the lower limit of fluorescence sensitivity for the flow cytometer in use. Figure 11.17.1 depicts the pattern that should be observed from the Deep Red bead mixture, with the 2.5-µm beads well into the third decade and the 0.04% intensity beads completely off the threshold. The analysis of a negative control sample will identify the lower limit of enumeration. This limit will be determined by the degree of fluorochromeconjugated probe aggregation that may be present in the sample, as well as the degree of nonspecific binding the probe may experience with environmental materials and other microbes that may be found in the sample. These data have been generated from various count correlation studies from pure organism preparations, which depict lower detection limits from 102 to 103 cells/ml. The accuracy of presence/absence detection will depend upon the antibody sensitivity and specificity and the experimental design for the rRNA detection. Selective enrichment will enhance the growth of the pathogen and suppress the growth of some other organisms. However, other organisms may exist that create the potential for a false positive to occur. The reliability of the assay will depend upon thorough specificity testing of the probe or antibody in a pathogen-free sample of interest. The Salmonella and E. coli O157:H7 antibodybased assays have been used on samples of hog carcasses and on ground beef, respectively. These represent relatively low (sponges) and high (ground beef) matrix samples. Antibody methods: The anti-Salmonella antibody was tested for sensitivity (absence of false negatives) using S. typhimurium and S. worthington, as shown in Figures 11.17.2 and 11.17.4. Other serovars have also been detected with this technique, including S. choleraesuis, S. enteritidis, S. heidelberg, S. adelaide, S. anatum, S. dublin, S. schalwijk, S. paratyphi A, and S. hadar. In the case of enteric pathogens, specificity testing (absence of false positives) may include testing against pure cultures of other enterics. The Salmonella antibody has also been tested against Shigella boydii, Citrobacter freundi, Enterobacter cloacae, Klebsiella pneumoniae, and Proteus vulgaris with no cross-reactivity occurring. Figure 11.17.6, panels A to E, provides two-parameter dot-density displays of antibody-based labeling and cytometric detection of Listeria innocua, E. coli O157:H7, Pseudomonas aeruginosa, Enterococcus, and Cryptosporidium oocysts to demonstrate how
11.17.16 Supplement 29
Current Protocols in Cytometry
Figure 11.17.6 Dot density plot for antibody-labeled bacteria. (A) Listeria innocua: 12553/ 0.25 ml (Basic Protocol 2); (B) E. coli O157:H7: 6270/0.25 ml (Basic Protocol 2); (C) Pseudomonas aeruginosa: 5662/0.25 ml (Alternate Protocol 2); (D) Enterococcus faecalis: 29891/0.25 ml (Basic Protocol 2); and (E) Cryptosporidium oocysts: 1537/0.25 ml (Basic Protocol 2). This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www.interscience.wiley.com/c p/colorfigures.htm.
Microbiological Applications
11.17.17 Current Protocols in Cytometry
Supplement 29
these protocols can be used to detect and enumerate other pathogens. PNA methods: The Salmonella PNA probe has been tested for sensitivity using S. typhimurium, S. paratyphi A, S. choleraesuis, and S. enteritidis as shown in Figure 11.17.5. Other serovars also were detected with similar results, including S. hadar, S. worthington, S. heidelberg, S. adelaide, S. anatum, and S. dublin. No cross-reactivity was observed in pure cultures of Shigella boydii, Citrobacter freundi, Klebsiella pneumoniae, Proteus vulgaris, or E. coli. Figure 11.17.7 shows dotdensity histograms typical for specific FISH labeling and cytometric analysis of pure cultures of Listeria and E. coli. This Listeria PNA probe uniformly labels all six species: L. monocytogenes, L. welshimeri, L. seeligeri, L. ivanovii, L. innocua, and L. grayi.
Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
Count correlations: Using antibody-based assays, count correlations similar to the Salmonella data presented in Figure 11.17.3 and ranging from 102 to 106 bacteria per ml with linear regression correlations R2 >0.95 have been achieved for a variety of microbes. The lower limit of detection is typically >100 bacteria per ml. Microbiologists have indicated that counts should be within a 0.5 log of the traditional methodology.
Time Considerations The initial instrument setup process will take ∼2 hr with the installation of the new in-line filter, system flush, and mixed bead analysis and protocol adjustments. The system should not require a new filter or system flush, unless the negative control background counts appear higher than normal. The routine
Figure 11.17.7 Dot-density plots for PNA FISH-labeled bacteria (Basic Protocol 3). (A) Listeria monocytogenes (ATCC 19111): 11997/0.25 ml; (B) listeria grayi: 15496/0.25 ml; (C) E. coli (Strain ECOR 58): 12548/0.25 ml; and (D) E. coli (Strain ECOR 72): 19471/0.25 ml. This black and white facsimile of the figure is intended only as a placeholder; for full-color version of figure go to http://www.interscience.wiley.com/c p/colorfigures.htm.
11.17.18 Supplement 29
Current Protocols in Cytometry
analysis of mixed beads and protocol adjustment on a system with clean lines should take only 20 min. Antibody-based analysis will typically take ∼30 min for direct, 1 hr for indirect, and up to 2 hr for the rRNA-based labeling method. Plate culture for enumeration comparison assays will take from 18 to 48 hr, depending upon the bacterial genus.
Literature Cited Amann, R.I., Ludwig, W., and Schleifer, K.H. 1995. Phylogenic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59:143-169. Alvarez-Barrientos, A., Arroyo, J., Canton, R., Nombela, C., and Sanchez-Perez, M. 2000. Applications of flow cytometry to clinical microbiology. Clin. Microbiol. Rev. 13:167-195. Bowden, R.A., Cloeckaert, A., Zygmunt, M.S., Bernard, S., and Dubray, G. 1995. Surface exposure of outer membrane protein and lipopolysaccharide epitopes in Brucella species studied by enzyme-linked immunosorbant assay and flow cytometry. Infect. Immun. 63:3945-3952. Brehm-Stecher, B.F., Hyldig-Nielsen, J.J., and Johnson, E.A. 2002. Design and evaluation of 16S rRNA-targeted peptide nucleic acid probes for whole cell detection of the genus Listeria. Abstract P-41. 102nd American Society for Microbiology General Meeting, Salt Lake City, Utah. Clark, R.G. and Pinder, A.C. 1998. Improved detection of bacteria by flow cytometry using a combination of antibody and viability markers. J. Appl. Microbiol. 84:577-584. Dang, J.L., Heroux, K., Kearney, J., Arasteh, A., Gostomski, M., and Emanuel, P.A. 2001. Bacillus spore inactivation methods affect detection assays. Appl. Environ. Microbiol. 67:3665-3670. Feng, P. 1998. Rapid methods for detecting foodborne pathogens. Appendix I. In Bacteriological Analytical Manual, 8th ed. AOAC International, Gaithersburg, Md. Frahm, E., Heiber, I., Hoffman, S., Koob, C., Meier, H., Ludwig, W., Amann, R., Schleifer, K.H., and Obst, U. 1998. Application of 23S rDNA targeted oligonucleotide probes specific for enterococci to water hygiene control. Syst. Appl. Microbiol. 21:450-453. Hoffman, R.M., Standridge, J.H., Prieve, A.F., Cucunato, J.C., and Bernhardt, M. 1997. The use of flow cytometry for detection of Cryptosporidium and Giardia in water samples. J. Am. Water Works Assoc. 89:104-111. Hughes, E.E., Matthews-Greer, J.M., and Gilleland, H.E.J. 1996. Analysis by flow cytometry of surface-exposed epitopes of outer membrane protein F of Pseudomonas aeruginosa. Can. J. Microbiol. 42:859-862. Ingram, M.T., Cleary, T.J., Price, B.J., and Castro, A. 1982. Rapid detection of Legionella pneumophila by flow cytometry. Cytometry 3:134147.
Lutton, D.A., Patrick, S., Crockard, A.D., Stewart, L.D., Larkin, M.J., Dermott, E., and McNeill, T.A. 1991. Flow cytometric analysis of withinstrain variation in polysaccharide expression by Bacteroides fragilis by use of murine monoclonal antibodies. J. Med. Microbiol. 35:229237. Oliveira, K., Procop, G., Wilson, D., Coull, J., and Stender, H. 2002. Rapid identification of Staphylococcus aureus directly from blood cultures by fluorescence in situ hybridization with peptide nucleic acid probes. J. Clin. Microbiol. 40:247251. Orzanne, V., Ortalo-Magne, A., Vercellone, A., Fournie, J.J., and Daffe, M. 1996. Cytometric detection of mycobacterial surface antigens; Exposure of mannosyl epitopes and of the arabinan segment of arabinomannans. J. Bacteriol. 178:7254-7259. Perry-O’Keefe, H., Rigby, S., Oliveira, K., Sorensen, D., Stender, H., Coull, J., and HyldigNielsen, J.J. 2001. Identification of indicator microorganisms using a standardized PNA FISH method. J. Microbiol. Methods 47:281-292. Srikumar, R., Chin, A.C., Vachon, V., Richardson, C.D., Ratcliffe, M.J. Saarinen, L., Keyhty, H., and Coulton, J.W. 1992. Monoclonal antibodies specific to porin of Haemophilus influenzae type b: Localization of the cognate epitopes and test for their biological activities. Mol. Microbiol. 6:665-676. Stender, H., Broomer, A., Oliveira, K., PerryO’Keefe, H., Hyldig-Nielsen, J.J., Sage, A., Young, B., and Coull, J. 2000. Rapid detection, identification and enumeration of Pseudomonas aeruginosa in bottled water using peptide nucleic acid probes. J. Microbiol. Methods 42:245253. Stender, H., Broomer, A.J., Oliveira, K., PerryO’Keefe, H., Hyldig-Nielsen, J.J., Sage, A., and Coull, J. 2001. Rapid detection, identification and enumeration of E. coli cells in municipal water by chemiluminescent in situ hybridization. Appl. Environ. Microbiol. 67:142-147. Stender, H., Fiandaca, M., Hyldig-Nielsen, J.J., and Coull, J. 2002. PNA for rapid microbiology. J. Microbiol. Methods 48:1-17. Tortorello, M.L., Reineke, K.F., Stewart, D.S., and Raybourne, R.B. 1998. Comparison of methods for determining the presence of Escherichia coli O157:H7 in apple juice. J. Food Prot. 61:14251430. Vesey, G., Slade, J.S., Byrne, M., Shepherd, K., Dennis, P.J., and Fricker, C.R. 1993. Routine monitoring of Cryptosporidium oocyst in water using flow cytometry. J. Appl. Bacteriol. 75:8790. Vesey, G., Narai, J., Ashbolt, N., Williams, K.L., and Veal, D.A. 1994. Detection of specific microorganisms in environmental samples by flow cytometry. In Methods in Cell Biology, Vol. 42 (Z. Darzynkiewicz,J.P. Robinson, and H.A. Crissman, eds.) pp. 482-522. Academic Press, New York, N.Y.
Microbiological Applications
11.17.19 Current Protocols in Cytometry
Supplement 29
Vesey, G., Ashbolt, N., Fricker, E.J., Deere, D., Williams, K.L., Veal, D.A., and Dorsch, M. 1998. The use of a ribosomal RNA targeted oligonucleotide prove for fluorescent labeling of viable Cryptosporidium parvum oocyst. J. Appl. Microbiol. 85:429-440.
http://www.epa.gov/nerlcwww/
Wallner, G., Amann R., and Beisker, W. 1993. Optimizing fluorescent in situ hybridization with rRNA-targeted oligonucleotide probes for flow cytometric identification of microorganisms. Cytometry 14:136-143.
http://www.psb.ugent.be/rRNA/
Zoetendal, E.G., Ben-Amor, K., Harmsen, H.J.M, Schut, F., Akkermans, A.D.L., and de Vos, W.M. 2002. Quantification of uncultured Ruminococcus obeum-like bacteria in human fecal samples by fluorescent in situ hybridization and flow cytometry using 16S rRNA-targeted probes. Appl. Environ. Microbiol. 68:4225-4232.
Appendix of rapid methods for detecting food pathogens from the on-line Bacteriological Analytical Manual.
Internet Resources http://www.orcbs.msu.edu/biological/BMBL/ BMBL-1.htm United States Department of Health and Human Services, National Institute of Health: Biosafety in Microbiological and Biomedical Laboratories, 3rd ed. Publication No. (CDC) 93-8395, 993.
This EPA method 1623 can be found here. http://www.epa.gov/nerlcwww/1623ap01.pdf EPA method for concentrating and enumerating Cryptosporidium oocysts. Searchable rRNA sequence data base for a variety of bacteria types. http://www.cfsan.fda.gov/∼ebam/bam-a1.html
http://www.linscottsdirectory.com A service-for-fee site that allows the user to search for vendors that produce antibodies to pathogens.
Contributed by Kristi R. Harkins and Kelley Harrigan Advanced Analytical Technologies, Inc. Ames, Iowa
Labeling of Bacterial Pathogens for Flow Cytometric Detection and Enumeration
11.17.20 Supplement 29
Current Protocols in Cytometry
CHAPTER 12 Cellular and Molecular Imaging INTRODUCTION iological imaging is undergoing a renaissance in the world of science. New technologies are becoming available almost monthly and old technologies are being reworked from analog to digital, from low-resolution to high-resolution, from slow to fast, and from monochrome to multispectral.
B
In the past, Current Protocols in Cytometry has focused heavily on the technology that embodies cytometry—flow cytometry. Flow cytometry has been a mainstay in immunology and hematology because it provides something that is very difficult, if not impossible, for other technologies to perform—quantitation. In fact, it could be said that, of all measurement technologies in the biological sciences, flow cytometry is the most quantitative, despite the efforts of many investigators to develop true quantitative assays for other approaches. However, it is also true that a considerable number of quantitative tools initially developed for flow cytometry have found their way into other fields. Among these tools are, certainly, fluorescent microbeads, which owe their very existence to the needs and demands of the flow cytometry field. There are now beads for every possible need, and then some. Clearly flow cytometry has had a significant effect on the rather complex area of quantitative fluorescence. With the emergence of readily available and affordable digital imaging technology, many individuals have jumped at the opportunity to attempt quantitative fluorescence analyses with fluorescence images. Most of these measurements are marginally better than guesswork, and the entire area of quantitative measurements in fluorescence imaging has been relegated to the “too hard” area. There is certainly a great opportunity here for some valuable developments! Biological imaging has grown for several reasons: lower cost of camera technology, higher resolution of cameras, better and cheaper monitoring devices, the evolution of confocal microscopy from an advanced technological toy into a standard lab tool, and the availability of low-cost, high-power computers and a significant amount of imageprocessing software. Thus, the availability of many inexpensive imaging systems brings an equal increase in use of imaging across the entire biological research spectrum. This chapter opens a new era for Current Protocols in Cytometry. To accommodate this growth in the field, and to ensure a link between traditional cytometry and next-generation cytometry, i.e., what is currently being referred to as cytomics, we have begun to address cellular as well as molecular imaging. UNIT 12.1 considers the issues of which tools to use—flow cytometry or imaging—and under what conditions. In particular, it compares the advantages and disadvantages of flow and image and provides examples illustrating the proper choice of each technology. The result is a better understanding of why both technologies are complementary in many applications. It is clear that many scientists use the tools that are familiar to them, often in preference to the best tool. In cases where very advanced and rather expensive technologies are concerned, this is not surprising. However, there are clearly times when
Cellular and Molecular Imaging Contributed by J. Paul Robinson and Jurek Dobrucki Current Protocols in Cytometry (2005) 12.0.1-12.0.2 C 2005 by John Wiley & Sons, Inc. Copyright
12.0.1 Supplement 31
one form of cytometry is definitely superior to another. What then constitute the criteria for a decision when both are available? This unit addresses some of these concerns. UNIT 12.2 is an introductory study of digital imaging. As we approach cellular and molecular imaging, a good understanding of and background in digital imaging is vital. Even before beginning image collection it is highly desirable to understand the implications of the decisions made for both initial acquisition and subsequent manipulation of images. How does one appropriately sample an object to be imaged? What are some of the technical details of camera image storage? Further, what is the impact of noise? How should images be stored? What storage solutions are acceptable in biological imaging? How much data loss is acceptable and under what conditions? Which file formats should be used and when? This very straightforward unit answers all these questions and shows how best to meet one’s imaging needs.
re-examines confocal microscopy from a current perspective. It outlines many of the most modern applications of confocal microscopy and the surrounding issues. The expanding applications of confocal microscopy have been creating a demand for both minor and major modifications of the technology to enhance imaging capabilities for a growing variety of samples. Techniques of interest such as FRET, FLIP, FRAP, and IFRAP are described. In addition, the unit includes a discussion on multispectral imaging, perhaps the potentially most exciting innovation in the field of biological imaging. The other area that is really beginning to have considerable impact on biology, medicine, and pharmacology is high-content screening (HCS) and high-throughput screening (HTS). Most of these technologies claim at least some dependence on confocal microscopy in their ability to obtain high-content, high-resolution images of live cells at very high speeds. This unit also introduces technology such as programmable array microscopes (PAMs), a technique in which spatial modulators are placed in the imaging plane of a microscope and used to generate patterns of illumination and/or detection.
UNIT 12.3
UNIT 12.4 deals with the kinetics of cellular change over time in cell-culture environments. Time-lapse monitoring poses a number of challenges. First is the ability to maintain cells in a sterile environment over a long period of time. This is not a trivial problem, but the solution will lead to even knottier ones. How do you track cells that change location? What software do you use to capture images without damaging cell by excessive excitation? How do you translate the images into data points of value? The answers can be found in this unit, the product of experience from this group working in time-lapse imaging over time periods of minutes to days.
Future units in this chapter will address the entire gamut of biological imaging as well as advanced image processing and analysis, 3-D and 4-D reconstruction, and influential aspects of imaging such as shape analysis and segmentation. J. Paul Robinson and Jurek Dobrucki
Introduction
12.0.2 Supplement 31
Current Protocols in Cytometry
Comparative Overview of Flow and Image Cytometry Cytometry is a study of cell measurement and may be divided into two basic types: flow and image. Each has its strengths and weaknesses, its capabilities and shortcomings, and the data obtained by the two methods are often complementary. In general, researchers and clinicians tend to do one or the other—but seldom both. This unit aims to introduce those familiar with one cytometry technology to the other approach and to help them identify the potential applications to their research or studies. The common feature shared by flow and image systems is measurement of fluorescencebased, and to some extent, scatter-based information. Both technologies use almost identical excitation and detection technologies and both systems collect data in a manner that is considered relatively high content. This unit will identify important differences between flow and image cytometry in measuring fluorescence, as well as a number of commonalities. There are three essential differences between the application of flow and imaging: 1. Flow cytometry makes a single highcontent measurement in a short time; once analyzed, a cell is lost and cannot be re-analyzed. Imaging may collect information on the same cells over an extended period of time, either continuously or at intervals. 2. Imaging contains spatial information, which flow cannot provide.
UNIT 12.1
3. Many flow systems can collect a far higher number of fluorescence variables than can imaging systems. A major difference between flow cytometry and imaging is that the latter can collect spatial information for materials larger than single cells, such as tissue sections, and correlate that with fluorescence. In the case of single cells, flow cytometry is able to identify each cell as a function of time as it passes through the laser beam; an imaging system must first identify a target within a large spatial location, segment the object, and then analyze it.
INTRODUCTION TO FLOW CYTOMETRY Flow cytometry is a technology that uses a laser to interrogate a cell flowing past in a stream of liquid. As each cell passes through the laser beam, a series of simultaneous measurements are made and the data recorded in listmode format. Three primary measurements are made: forward light scatter (usually representative of cell size); excitation light scatter orthogonal to the beam (mostly related to cell refractivity or granularity); and finally a series of spectral measurements of fluorescence emission. This is shown diagrammatically in Figure 12.1.1. The units in Chapter 1 of this manual provide detailed description of flow cytometry instrumentation.
Figure 12.1.1 Typical flow chamber in which fluorochrome-labeled cells are injected into a flowing sheath of saline and directed through the laser beam. Left: typical fluorescence histograms. Right: forward-scatter histogram. Diagram is not drawn to scale.
Cellular and Molecular Imaging
Contributed by J. Paul Robinson
12.1.1
Current Protocols in Cytometry (2005) 12.1.1-12.1.11 C 2005 by John Wiley & Sons, Inc. Copyright
Supplement 31
INTRODUCTION TO IMAGING Imaging is a method for creating twodimensional (2-D) or three-dimensional (3-D) visualizations of cells or tissue by presenting scatter or fluorescence information in the context of spatial relationships of all components and features. Imaging systems use light sources similar to those of flow cytometry (e.g., lasers) and can use PMTs, cameras, or other detectors to extract essential information from the sample. It is possible to measure the scatter (bright-field or back-scattered light) simultaneously with fluorescence signatures. All data points will be identified within the spatial confines of the area of view. Traditionally, imaging is performed using microscopes with magnifications from 0.5× to over 100×. Imaging covers multiple technologies, from bright-field and dark-field microscopy to fluorescence and multiphoton microscopy and back scatter, as well as second harmonic generation. The units in Chapter 2 of this manual provide detailed description of image cytometry instrumentation. Typically, imaging systems are preferable when either repeated measurements on a particular cell or group of cells are required over a time period (time lapse) or retention of the spatial relationship of a cell system is necessary. Imaging on flow systems is possible, but uncommon, although commercial instruments are now available that do some imaging in flow (discussed below). It is entirely possible to obtain from imaging systems information similar to that from flow cytometry, but the time required to obtain these data is usually increased by an order of magnitude or more. The clear advantage of imaging systems is the ability to visualize the sample, something that is not possible with traditional flow cytometry technologies. Emerging technologies may provide instruments having some advantages of both systems, as will be discussed briefly later in this unit.
INTRODUCTION TO SCANNING LASER CYTOMETRY Scanning laser cytometry (also see is to some extent a hybrid of flow cytometry techniques and image analysis. For this reason, it is considered independently from flow and imaging. Scanning laser cytometry systems use multiple lasers to scan cells attached to slides or dishes. The cells can be stained or unstained, and frequently they are stained and destained more than once to gain a very large array of variables. The key feature separating scanning laser cytometry from
UNIT 2.10)
Comparative Overview of Flow and Image Cytometry
imaging is that it identifies every cell in a selected field on a slide and places those cells into a correlated listmode file like a flow cytometer. The slide can be removed from the instrument in order to destain, restain, or modify the cells in some way; it can then be placed back on the instrument for re-analysis. The technology is so accurate that every cell can be relocated and an additional set of variable values can be placed within the correlated listmode data set. Once this is accomplished, all of the data points are correlated and some very impressive multivariate analysis can be performed. Recently, there have been reports of scanning laser analysis using about 20 colors, significantly more than even the most advanced flow cytometers can handle (Lenz et al., 2003).
ADVANTAGES OF FLOW CYTOMETRY Flow cytometry possesses several clear, even unique advantages. This technology can create correlated listmode data that allow complex relationships to be made very quickly and efficiently. Given appropriate standards, it is possible to perform very accurate quantitative measurements on fluorescence. In addition, the fluorescence components can be correlated with the size, structure, and other physical aspects of particles. Further, because of the very nature of measurement, there is a high signal-to-noise ratio, with essentially no background from the noncellular or particulate environment. In essence, the area of measurement in a flow cytometer, commonly termed the sample excitation volume, provides a signal only from the passing cell or particle, not the suspending fluid. This, of course, is quite different from trying to select cellular or particulate components in an image array. Finally, the more complex instruments known as cell sorters can recover specific cells after measurement for further biochemical analyses. Suspension cell systems such as HL-60 cells, blood, bone marrow, and microbial cultures are ideal for flow cytometric analysis. All these samples are already in single-cell suspension form and require only appropriate bioassay processing to allow analysis by flow cytometry. It is then very easy to supply the cell environments with high concentrations of labeling antibodies, either polyclonal or monoclonal, against surface antigens or receptors on the cells. With or without washing, it is possible to gain high-quality information. Another clear advantage of flow cytometry is that with the use of multiple excitations
12.1.2 Supplement 31
Current Protocols in Cytometry
and emissions, it is possible to extract significantly more spectral information than most image systems can provide. While this is perhaps one of the most significant areas of change in technology with the advent of multispectral imaging, it is still clear that flow cytometry has greater detection capability (sensitivity) and spectral separation capability. This is because of the lower background of measurements, and because every measurement is of a single cell rather than of an accumulation of cells or a region of tissue. An example of this is the current work being carried out on functional T cell subsets by De Rosa and Roederer (2001). A fourth advantage of flow cytometry is that it is possible to re-collect very specific subsets of identified cells using the technology of electrostatic-based cell sorting (also see UNIT 1.7). Once cells are identified, regardless of the complexity of the identification process, the stream is pulsed by a piezoelectric crystal and broken into droplets. Using a series of algorithms based on knowledge about the identified cells, the time after their identification, and an agreement between software and hardware as to their exact location, a specific droplet containing a cell of interest can be given an electric charge, pulled out of the stream, and deflected into a container. This can be done repeatedly at very high speed to recover sufficient rare cells to perform a significant array of further
testing if necessary. If desired, sorting can be performed under aseptic conditions to ensure sterility of the environment containing the cell. Most manufacturers of cell sorters allow for sorting of single cells into microtiter plates for subsequent culture of individual cells. A final advantage of flow cytometry is that with this technique it is more convenient and possibly easier to utilize automated classification routines on subpopulations of cells of like nature because of the large number of variables collected, including forward and side scatter and 5 to 12 wavelength bands. While there are clearly many classification opportunities using imaging (discussed later), the imaging setup also has more disadvantages. Clearly, a flow cytometer cannot evaluate a piece of tissue such as a tumor biopsy. However, a number of technologies facilitate transforming tissue into single-cell suspensions suitable for flow cytometry. Normal tissue-disaggregating techniques such as grinding are adequate, but some fast and very efficient available technologies make flow cytometry very achievable on biopsy materials. One is the Medimachine (Fig. 12.1.2) in which relatively large pieces (grams) of tissue can be very quickly disaggregated (in <1 min) into a single-cell suspension in any medium desired. While disaggregation is never as good as using cells already in suspension, this technique
Figure 12.1.2 The Medimachine automated mechanical disaggregation system is a very useful device that allows disaggregation of biopsy materials, providing single-cell suspensions of excellent quality and making it possible to do high-quality flow cytometry on tissue specimens. The principle of this instrument is the high-speed cutting tool that spins within the small sampling unit. Single cells pass through the holes in the blades and into a reservoir below, where they can be rescued using a pipet from a side port. Once cells are removed from the vessel they can be stained, e.g., for cell-cycle analysis or for specific antigens of interest, using traditional flow cytometry staining techniques.
Cellular and Molecular Imaging
12.1.3 Current Protocols in Cytometry
Supplement 31
Figure 12.1.3 Typical images from image systems. (A) Bright-field image of tongue histology. (B) Brightest pixel reconstruction of pine tree pollen. (C) Standard fluorescence image of a human neutrophil. (D) A human neutrophil which has phagocytosed yeast. (E) Reconstruction of bacteria growing in a biofilm. B and D are confocal images. All of these are typical images where the spatial relationship of components is crucial for understanding the processes of interest.
creates a real opportunity for the application of flow cytometry to whole tissue samples. There is an excellent alternative that falls within the discipline of cytometry, but not flow cytometry, i.e., the laser scanning cytometer, discussed elsewhere in this unit (also see UNIT 2.10).
ADVANTAGES OF IMAGING
Comparative Overview of Flow and Image Cytometry
It is not difficult to identify the advantages of imaging for evaluating tissue or aggregates. The relationships among cells are retained when entire tissue is imaged. From the perspective of pathology, this is one of the reasons that imaging provides uniquely valuable data. As shown in Figures 12.1.3 and 12.1.4, traditional microscopic imaging provides organized information that preserves the relationship of cells. Given a variety of antigenic determinants, considerably more information may be available on a per-cell basis, but none of these data provides the relationship information that imaging does. For a pathologist evaluating abnormalities in tissue, imaging is clearly the best mechanism.
COMPARISON OF TRADITIONAL IMAGING AND FLOW CYTOMETRY The goal of traditional imaging is to create a high-quality representation of an object in order to identify components within the cell. Resolution becomes important when hard-to-see objects such as organelles need to be segmented or identified. This frequently means the use of reasonably long exposure times or averaging of multiple images. Figure 12.1.3 shows a very good series of images that demonstrate the usefulness and even superiority of imaging. In fact, imaging in one of its many forms would be the only reasonable method for evaluating these samples. In several of these images, ordinary bright-field fluorescence microscopy is all that is necessary to obtain the relevant information. However, it is also true that additional imaging techniques such as confocal microscopy enhance the data by providing a 3-D dataset from which internal structure can be ascertained. In Figure 12.1.3, this is shown in the pollen as well as in the
12.1.4 Supplement 31
Current Protocols in Cytometry
Figure 12.1.4 (A) Cells organized in an ordered matrix. The abnormal cell (x) is in a specific position. From a pathology perspective, the location of this cell and the relationship to others provide important information. (B) Cells as they might be in suspension, totally randomly arranged.
human neutrophil in which the phagocytosed yeast particles are very clearly visible. In some cases, it is possible to utilize both flow and imaging to enhance the data significantly. For the data shown in Figure 12.1.3, there is no apparent advantage in using flow cytometry to extract additional information. In contrast, Figure 12.1.5 illustrates a case where flow cytometry can be a very valuable complement to the image analysis. These histograms demonstrate the most significant component that impacts side scatter. The images represent typical blood cells stained with hematoxylin and eosin. Beside each image is a cartoon representation of what the side scatter might look like if a population of the cells associated with each image were run on a flow cytometer. Side scatter, frequently referred to as orthogonal scatter, is most affected by the multiple components within a cell that cause light refraction. Examples of these components are cytoplasmic granules, small organelles, and sometimes polymorphic DNA as found in neutrophils. The histogram of the lymphocyte in the very center, shown in the top right corner, represents a classic standard. Lymphocytes are morphologically reasonably homogeneous and consist mostly of nucleus with a minimal amount of cytoplasm and essentially no granularity. Thus the distribution is very tight, the coefficient of variation (CV) is small, and the side scatter is very low. Data such as these are difficult to obtain by imaging. Flow cytometry can very successfully identify differences in cellular composition using combinations of forward and side scatter. Indeed, it is a very
accurate technology for separating populations that have different composition, even though there is usually no way of visualizing the cells themselves (see discussion below of the Amnis imaging cytometer and the LSC cytometer systems). Quantitative measurement is another area in which flow cytometry offers some serious advantages. However, there are clearly times when imaging must be used to support the conclusions being drawn from the flow data. For example, mitochondrial dysfunction, including the mitochondrial permeability transition and a decrease of mitochondria membrane potential, can be studied by both flow cytometry and imaging, using drugs such as rotenone and diphenyliodonium (IDP). Loss of mitochondrial membrane potential can be demonstrated by flow histograms as shown in Figure 12.1.6, which provides a quantitative value for the mitochondrial population. However, to confirm the location of these changes, imaging is very helpful. Salvioli and coworkers studied the mitochondria permeability transition (MPT) in HL-60 cells loaded with calcein-AM and tetramethylrhodamine methyl ester (TMRM). Calcein is a green fluorescent dye that normally distributes in the cytosol and is excluded from mitochondria. When the mitochondrial membrane permeability increases, calcein is able to redistribute from the cytosol into the mitochondria. With the combination of calcein and the cationic red fluorescent dye TMRM, it is possible to monitor MPT and mitochondrial membrane potential simultaneously by
Cellular and Molecular Imaging
12.1.5 Current Protocols in Cytometry
Supplement 31
Figure 12.1.5 A comparison of information from traditional imaging and flow cytometry. Cartoons represent side-scatter histograms of the cell populations in the images. In the center of the middle image is a lymphocyte. The side-scatter histogram for this cell is in the top right corner and reflects the homogeneity and low granularity of this cell type. Note that some cells, such as macrophages (top center, right center), have significant variation in shape and granule composition. This increases the CV of the histogram. All images were collected on a COSMIC digital microscope.
confocal microscopy (Salvioli et al., 1997). In contrast to the previous analysis, in this case we are more interested in the location of the fluorescent products within the cellular environments. This necessitates the use of imaging, as it is vital to determine the exact location of the fluorescence signal in the cell (Fig. 12.1.7). Flow cytometry will give an accurate measure of what is in the cell, and even how much, but is unable to provide any location information. This does raise the issue of alternative cytometry tools such as the laser scanning cytometer and the imaging cytometry technologies.
COMBINATION STUDIES WITH FLOW AND IMAGING
Comparative Overview of Flow and Image Cytometry
The combination of flow cytometry and image analysis creates some very interesting possibilities. Consider studies involving microorganisms. Although flow cytometry would appear to be the obvious choice, there are times when it is crucial to identify relationships—
for example, between live and dead organisms within colonies. The only way to do this is to image the colonies with a multicolor imaging method such as confocal microscopy, so that it is possible to distinguish the location of dead organisms, and subsequently to evaluate the number of live versus dead using flow cytometry. This is shown in Figure 12.1.8, in which live and dead organisms can be differentiated using the same criteria—red fluorescence indicating dead cells and green fluorescence indicating live cells. One of the true advantages of flow cytometry is very clearly seen in this example. Using traditional gating strategies, the live and dead cells are rapidly separated and quantified. However, it is much more difficult to quantitate the live and dead cells in the image. It is possible to quantitate the number of red and green pixels, but there are many variables involved in doing actual organism quantitation. Clearly a combination of both technologies is the best approach for such studies.
12.1.6 Supplement 31
Current Protocols in Cytometry
Figure 12.1.6 Flow histograms demonstrating mitochondrial membrane transitions. Numbers above each histogram represent median channel number. Mitochondrial populations in panels (A) to (D) show decrease in membrane potential, while those in panels (E) to (H) show very little change. Abbreviation: TMRM, tetramethylrhodamine methyl ester.
Figure 12.1.7 Cell images demonstrating the ability of imaging to evaluate the presence of certain fluorescence probes and their location within organelles in the cell environment.
IMAGING CYTOMETRY Two particular combinations of imaging cytometry are especially relevant to this unit. The first is based on the laser scanning cytometer (LSC), which has a unique way of scanning cells and converting the data into a location
index together with a correlated listmode file with all the variables collected. This is in addition to the opportunity that the LSC provides for acquiring images of cells. The unique idea generated by Kamentsky when he developed the LSC technology was to use the traditional
Cellular and Molecular Imaging
12.1.7 Current Protocols in Cytometry
Supplement 31
Figure 12.1.8 For a bacterial study, a composite of flow cytometry and image analysis is required both to evaluate how many microorganisms are present and to differentiate between live and dead organisms. A live-dead stain was used so that both image and flow systems were able to make identical measurements under completely different circumstances. Image data were collected from a biofilm, while the flow data were collected from individual organisms in suspension.
Comparative Overview of Flow and Image Cytometry
flow cytometry approach to data display along with many of the data analysis techniques that are fundamental to flow cytometry. The result was a significantly successful integration of basic cytometry and imaging. The technology is not based on flow, so it is quite different from flow cytometry, but there are many similarities, mainly in respect to image processing. Single cells attached to slides are scanned and their locations identified and recorded. Images are taken of each cell and a series of analyses are made, such as fluorescence intensity at multiple wavelengths, morphologic analyses, and various morphometric and densitometric measurements. More recently an inverted system has been developed which can analyze adherent cells in culture. The real advantage of the LSC technology is that it is possible to maintain morphometric relationships in tissue so that all data collected are within the context of the tissue structure. This is shown in a recently published data set (Fig. 12.1.9) that demonstrates the true value of this technology. It is clear that imaging alone or cytometry alone cannot extract as many features as the combination of both technologies. First, the ability to
identify single cells, even in a complex tissue, is an important component of the additional value of LSC technology. What is absolutely unique about this technology is that it can not only evaluate multiple colors of fluorescence (as many as 10 to 15) but also correlate them within a tissue section, and even within multiple tissue sections taken sequentially, to create a pseudo-3-D relationship within the tissue section. This technology was the first to create quantitative relationships between individual cells and their spatial position within a tissue section. However, another recent development in technology is the imaging cytometer built by Amnis Corp. (Fig. 12.1.10). In this technology, cells are imaged as they are analyzed in a flowing stream at reasonably high speed, at least close to that of traditional flow cytometers. The major difference between this imaging cytometer and the traditional cytometer is the ability to identify each cell by imaging technology and display it with a list of images that represent cells with a particular set of parameters. The technology involved is quite different from traditional imaging systems and uses
12.1.8 Supplement 31
Current Protocols in Cytometry
Figure 12.1.9 An example of LSC data in which each cell has been segmented using imageprocessing algorithms. A large number of parameters have been collected into a correlated listmode file that can be further analyzed by comparing the positional information on histograms and dual-parameter dot plots with actual images of cells and with their location on a tissue array. By collecting multiple tissue sections, it is even possible to identify 3-D relationships of cells within the organ or part of organ evaluated. Data from Gerstner et al. (2004).
some recent developments in imaging known as time delay integration (TDI) technology.
CONCLUSION Flow cytometry is a technology that has matured over a 40-year period of application in many areas of biology and materials science. The key value of flow cytometry, while not absolutely unique, is the capacity to provide quantitative fluorescence measure-
ments. Flow cytometry is perhaps the best technology for quantitation in the biological domain. The disadvantage of flow cytometry is that in its traditional form it can evaluate only single cells. A compromise is the LSC technology, which can evaluate single cells within tissue and provide quantitative information within the spatial context of the whole tissue. A further feature of this technology is the ability to provide images, albeit at relatively
Cellular and Molecular Imaging
12.1.9 Current Protocols in Cytometry
Supplement 31
Figure 12.1.10 Layout of the imaging cytometer. This system allows for basic flow cytometry measurements as well as the ability to make a reasonably good image of each cell. While the images are not at a resolution that allows detailed analysis, they are nevertheless particularly useful since they do provide sufficient information to identify morphological differences in cells. Figure courtesy of Amnis Corp.
Figure 12.1.11 Automated segmentation algorithms allow critical analysis of single cells from image-based systems. Once a cell area has been classified, a number of analyses can be performed.
Comparative Overview of Flow and Image Cytometry
low resolution. The other cytometry technology based on flow, the Amnis technology, allows the same complex multivariate analysis while providing simultaneous images for morphological evaluation. An additional issue with imaging is the ability to automatically segment cells (see Fig. 12.1.11) and then perform quantitative measurements. This is usually done using a scanning laser cytometer, but is
being done more frequently with high-content screening systems evaluating large numbers of single cells. Naturally, none of these technologies is capable of the high resolution provided by fluorescence microscopy or confocal microscopy. These imaging technologies are able to provide very good indications of true spatial co-localization. When resolution is absolutely required to identify subcellular organelles, or
12.1.10 Supplement 31
Current Protocols in Cytometry
to locate areas of co-localization, confocal or multiphoton microscopy is absolutely the best technology available. Frequently, one technology leads an investigator to utilize another. Over the past decade, both flow and image have been heavily utilized in the author’s laboratory, and it has become clear that the vast majority of biological problems require multiple technologies to identify solutions. It would not be unreasonable to suggest that many investigators use only the technologies that are readily available to them. If they have extensive experience with imaging but not with flow cytometry, they will utilize the imaging capabilities to the best of their ability. Clearly, with the current focus on the expansion of multidisciplinary studies across biology, there will be a significant increase in utilization of tools in combination, such as flow cytometry and advanced imaging.
Gerstner, A.O.H., Trumpfheller,C., Racz, P., Osmancik, P., Tenner-Racz, K., and T´arnok, A. 2004. Quantitative histology by multicolor slidebased cytometry. Cytometry 59A:210-219. Lenz, D., Gerstner, A., Laffers, W., Steinbrecher, M., Bootz, F., and T´arnok, A. 2003. Six and more color immunophenotyping on the slide by Laser Scanning Cytometry (LCS). Proc. SPIE 4962:364-374. Salvioli, S., Ardizzoni, A., Franceschi, C., and Cossarizza, A. 1997. JC-1, but not DiOC6(3) or rhodamine, is a reliable fluorescent probe to assess delta psi changes in intact cells: Implications for studies on mitochondrial functionality during apoptosis. FEBS Lett. 411:77-82.
Contributed by J. Paul Robinson Purdue University West Lafayette, Indiana
LITERATURE CITED De Rosa, S.C. and Roederer, M. 2001. Eleven-color flow cytometry. A powerful tool for elucidation of the complex immune system. Clin. Lab. Med. 21:697-712.
Cellular and Molecular Imaging
12.1.11 Current Protocols in Cytometry
Supplement 31
Basics of Digital Microscopy Optical microscopy is a well established tool in biological research. In modern microscopy, analog light detectors like the human eye or photographic film are replaced with a digital camera. Digital microscopy comprises image formation by optics, image registration by a digital camera, and saving of image data in a computer file. Making optimal use of digital microscopy requires taking into account limitations that are particular to each of these processes. The discussion presented here is limited to wide-field light microscopy. Confocal, multiphoton, and other specialized microscopy techniques are described in other units (see Chapter 2).
OPTICAL IMAGING Resolution in the Spatial Domain The laws of geometrical optics predict no limits on the size of the object that may be observed in a microscope. However, when light passes through an objective, diffraction occurs at the pupil owing to the wave nature of light. Thus, the image formed by a point light source at the focal plane is not a single point. Instead, the light intensity is distributed (spread) over
an area on the image plane. This distribution is given by the point spread function (PSF). The pattern (Airy disc) features a central highintensity spot and many concentric rings. The actual shape of the PSF (degree of spreading) depends on the wavelength of light used for imaging and on the numerical aperture (NA) of the microscope objective (Fig. 12.2.1).
Spatial Resolution Criteria Spatial resolution refers to the ability to distinguish two small or point-size objects separated by a given distance. Resolution in optical microscopy is often assessed by means of the Rayleigh criterion (Rayleigh, 1879), which was originally formulated for determining the resolution of two-dimensional telescope images observed with the human eye. According to this criterion, two closely spaced Airy disks are considered distinct if the distance d between them is greater than the radius of the Airy disk (Fig. 12.2.1). This critical distance can be estimated using the formula: d = 0.6100 λ/NA = 0.6100λ/nsin(α/2) where λ is the imaging wavelength, α the full aperture angle, and n the light refraction
Figure 12.2.1 Dependence of microscope PSF on numerical aperture (NA, in rows) and wavelength (λ, in columns). Plot coordinates are indicated in the top left corner: x,y = position (from −0.75 to 0.75 µm); i = intensity. Width of the PSF decreases with increasing numerical aperture and increases with wavelength. Contributed by Tytus Bernas Current Protocols in Cytometry (2005) 12.2.1-12.2.14 C 2005 by John Wiley & Sons, Inc. Copyright
UNIT 12.2
Cellular and Molecular Imaging
12.2.1 Supplement 31
Figure 12.2.2 Summed image intensities (PSFs) of two points resolved according to (from left to right) Rayleigh, Sparrow, and FWHM (Houston) criteria. Plot coordinates are indicated in the top left corner: x,y = position (from −0.75 to 0.75 µm); i = intensity. Resolution (critical distance) depends on what intensity contrast is considered sufficient.
coefficient. It is assumed that the numerical apertures of condenser and objective are equal. Note that the smaller the value of d, the better the resolution. Sparrow (1916) suggested an alternative, which depends on the property of the summed intensity profiles. As one moves two initially overlapping images of points apart, an intensity minimum develops. Sparrow suggested that resolution occurs at the line separation where this saddle point first develops (i.e., the gradient at the peak of the summed profile is zero). The critical distance for the Sparrow criterion is: d = 0.4750λ/NA Another method is to adopt the size of a point image to define resolution. Houston (1926) suggested that two points may be resolved if the distance between them is equal to the width of the Airy disk measured at half the maximum intensity (full width half maximum, FWHM). Thus, the distance for the Houston criterion is given by: d = 0.5015λ/NA
Basics of Digital Microscopy
Figure 12.2.2 shows a comparison of the three criteria. It is clear that the choice of resolution criterion depends on what intensity difference (between maxima and the separating minimum) is considered sufficient. These resolution criteria were based on the sensitivity of the eye of a human observer. In modern digital instruments, image-processing techniques are used to improve contrast (and thus perceived resolution) and to analyze images in an automated way. Hence, spatial resolution is usually described in the frequency domain using the Fourier transform of the PSF, the optical transfer function (OTF).
Resolution in the Frequency Domain A description based on the OTF states that any “optical object” can be described as a weighted sum of sinusoidal distributions of light, where the sinusoids have varying spatial frequencies. The way an object “propagates” through an optical system depends on the OTF of that system. Thus, by knowing the OTF, one can also predict how the image of an object will look. Formally, the spatial frequency amplitudes of the image will be given by the spatial frequency amplitudes of the object multiplied by the OTF, which is itself a function of the spatial frequencies. The OTF is a two-dimensional Fourier transform of the PSF: OTF( fx , fy ) = F {PSF(x,y)} = F {PSF(r)} = OTF( fr ) where F(x) is the two-dimensional Fourier transform operation, x,y are the spatial coordinates, r is the radial coordinate, fx ,fy are the spatial frequencies in the x and y direction, respectively, and fr is the radial spatial frequency. The OTF (shown in Fig. 12.2.3) is 0 for frequencies greater than the cutoff frequency given by the formula fc = 2NA/λ, where NA is the numerical aperture, λ the imaging wavelength, and, fc the cutoff frequency. In other words, spatial frequencies greater than fc can not be transferred by a given optical system, which therefore is said to be bandlimited. More detailed discussion of OTF and PSF can be found in UNIT 2.6.
DIGITAL REGISTRATION OF AN OPTICAL IMAGE Architecture of the Light Sensor The majority of light sensors (cameras) used in microscopy are based on silicon. This
12.2.2 Supplement 31
Current Protocols in Cytometry
Figure 12.2.3 Optical transfer function (OTF) corresponding to PSF from Figure 12.2.1 (lower part, NA = 0.95; λ = 0.53 µm). Plot coordinates are indicated in the top left corner: fx ,fy = spatial frequencies in x and y direction (from −4.4 to 4.4 µm−1 ); i = intensity. OTF is 0 above cutoff frequency (i.e., is band-limited).
Figure 12.2.4 Schematic representation of a light-sensitive element (pixel) of a CMOS camera. The photons incident on the light-sensitive area and photoelectrons (e− ) trapped in the potential well are indicated.
element can form large crystals in which each atom is bound to its six neighbors, forming a rectangular, three-dimensional lattice structure. An incident photon may break one of these bonds, releasing an electron (photoelectric effect). A thin metal layer deposited on the surface of the silicon and charged with a positive voltage creates a potential well that collects and stores the electrons (Fig. 12.2.4). Each potential well corresponds to a lightsensitive element of the camera. Depending on the element design, several types of cameras are distinguished: CCD (charge-coupled
device), CID (charge-injection device), CMOS APS (complementary metal-oxide semiconductor active pixel sensor), CMOS PPS (complementary metal-oxide semiconductor passive pixel sensor), and others. Nonetheless, all cameras considered here operate as a rectangular array of light-sensitive elements (chip). The performance of such a device is characterized by the signal-to-noise ratio, SNR, which determines the quality of measurement. SNR is usually expressed in decibels, dB, using the formula SNR = 20log{signal/noise}. The following section briefly discusses the sensor
Cellular and Molecular Imaging
12.2.3 Current Protocols in Cytometry
Supplement 31
Basics of Digital Microscopy
parameters, which influence both terms of this quotient. The amount of signal is limited by the well depth, which is a measure of how much charge an individual light-sensitive element on a camera chip can accumulate. This parameter is generally measured in electrons and depends on the fraction of the chip that is light sensitive (fill factor). Often, half or more of the available sensor area is covered by opaque charge-transfer circuitry, leaving gaps between the pixels and reducing the fill factor below the ideal of 100%. The chip can be coated with a thin layer of microlenses, each of which focuses the incoming light it receives onto the sensitive areas of one pixel. Well depth is proportional to the size of the element (pixel). For example, a Kodak KAF-3200E CCD chip (6.8µm pixels) has a well depth of 55,000 electrons, whereas a Kodak KAF-6303E chip (9.0µm pixels) features a well depth of 100,000. When the well depth is exceeded, electrons penetrate to adjacent wells, resulting in a bright streak extending vertically from a saturated spot. This effect, called blooming, is prevented if wells in a camera chip have drains to remove excess electrons. However, chips equipped with this antiblooming protection have much lower well depths, and thus are less sensitive than their counterparts without this feature. An area of adjacent pixels can be combined into one larger pixel in a process called binning. For instance, 2×2 binning means that the electric charge from 4 adjacent pixels is pooled together. This increases the sensitivity to light by a factor of 4. However, the effective width and height (in pixels) of the chip, the resolution, is correspondingly halved. Only a fraction of incident photons are converted into electrons (photoelectrons), which are stored and then read out at the end of the exposure. The number of photons converted depends on the camera’s quantum efficiency (QE), which is a function of wavelength. Standard cameras are most sensitive to green and red wavelengths in the region between 550 and 900 nm. Most midrange cameras have maximum QE in the range of 25% to 50%, whereas high-grade scientific CCDs may have a QE close to 98%. All acquired images are contaminated by noise, a stochastic phenomenon that can neither be compensated for nor eliminated. The noise sources that play a role in scientific cameras are photon noise, thermal noise (dark current and hot pixels), readout noise (amplifier noise and on-chip electronic noise), and quantization noise.
Photon noise is a result of inherent variation in the arrival rate of photons incident on a camera chip owing to the quantum nature of light. The number of photoelectrons fluctuates randomly with photon incidence at each element (pixel) on a camera. Since the interval between photon arrivals is governed by Poisson statistics, the photon noise is equivalent to the square root of the signal. Thus, even if the photon noise were the only noise source, SNR would still be finite. Dark noise arises from electrons thermally released from the silicon structure of a camera chip and subsequently accumulated by lightsensitive elements. Dark noise is not affected by incident light, but is highly dependent on device temperature. Cooling radically reduces the dark noise. The generation rate of thermal electrons at a given temperature is referred to as dark current. Like photon noise, dark noise follows Poisson statistics and is equivalent to the square root of the number of thermal electrons. Readout noise originates in the process of reading the signal from the sensor and is caused by the on-chip electronics. This noise depends on the readout rate: it is inversely proportional to very low readout rates, approximately constant (and minimal) for moderate readout rates (20 to 500 kHz), and increases again for high readout rates. The readout noise is additive, Gaussian distributed, and independent of the signal. It is therefore expressed by its standard deviation (root mean square, rms) in number of electrons. Quantization noise is a result of round-off errors caused by conversion of continuous values of electric charge accumulated by lightsensitive elements to a finite number of discrete intensity levels. This noise is additive, uniformly distributed, and independent of the signal. Quantization noise increases as the number of levels (bit depth) of the analog-to-digital converter (ADC) decreases. However, even for an 8-bit ADC (minimum for a scientific camera) the noise does not exceed 0.5 electrons per pixel. Thus, quantization noise is usually ignored. Overall system signal-to-noise is commonly calculated using the following formula for camera system signal-to-noise ratio: SNR = PQe t/[PQe t + Dt + Nr 2 ]1/2 where P is the incident photon flux (photons/pixel/sec), Qe represents the CCD quantum efficiency, t is the integration time (sec), D is the dark current value (electrons/ pixel/sec), and Nr represents readout noise
12.2.4 Supplement 31
Current Protocols in Cytometry
Figure 12.2.5 Influence of camera chip temperature (x axis, ◦ C) and incident light intensity (y axis, arbitrary units) on different components of camera noise. Signal-to-noise ratio (SNR) is indicated by pie size, whereas relative contributions of photon, readout, and dark noise are depicted by pie slices. The parameters were estimated for a Sony ICX085 chip and full well capacity.
[rms, (electrons/pixel)1/2 ]. Influence of the photon flux and the dark current on SNR is illustrated in Figure 12.2.5.
Color Versus Monochrome Cameras The operation of camera sensors discussed so far relies on measurement of total light intensity, but not wavelength spectrum. Thus, only monochrome images are obtained. Registration of color images requires measurement of light intensity at several wavelengths. Typically red, green, and blue ranges are chosen, to mimic human vision. Two main types of color cameras are used: either a single sensor with a wavelength-selection filter or a threesensor system. Single-sensor cameras utilize a set of three filters, a single liquid-crystal tunable filter, or an adherent filter matrix to register the red, green, and blue images. When the set of filters (placed in a slide or filter wheel) or the tunable filter is used, the three images are obtained in sequence. High transmission (>95% in the selected range) is an advantage of using three separate filters. However, changing the filters is a relatively slow process and may be a source of vibrations. Tunable filters (Fig. 12.2.6), while free from these disadvantages, are characterized by relatively low maximum transmittance, which does not
exceed 50% with unpolarized incident light (Reichman, 2000). The adherent filter matrix differs from systems described previously in that each light-sensitive element is coupled with its own filter (red, green, or blue; see Fig. 12.2.7). This design combines high transmittance with speed, as the color image is acquired in one pass. However, actual sampling intervals and resolution are decreased in comparison with the same sensor without the filter matrix. Software interpolation is used to match component images and increase the number of pixels. Some color cameras employ a piezo-controlled translocation mechanism to increase the sampling frequency, at the cost of reduced sensitivity or increased acquisition time. An interesting multilayer color sensor (the Foveon X3; Fig. 12.2.8) was recently introduced to ameliorate the problem of decrease in spatial frequency. This sensor features three layers of light-sensitive material (blue, green, and red), a construction which permits simultaneous registration of three colors with full resolution. However, its quantum efficiency is lower (<50%) than that of the majority of scientific-grade cameras of classic construction. Simultaneous acquisition of three colors with high sensitivity and resolution is possible using a three-sensor camera, which has a beam-splitting prism and trim
Cellular and Molecular Imaging
12.2.5 Current Protocols in Cytometry
Supplement 31
Figure 12.2.6
Schematic representation of acousto-optical tunable filter (from Brimrose Corp.).
Figure 12.2.7 Adherent filter matrix (bayer pattern) comprising red (R), green (G), and blue (B) filters. Top row: the whole matrix. Bottom row: decomposition into color components. Increased sampling intervals in the component images result in lower resolution.
Figure 12.2.8 Diagram of a Foveon X3 chip (From Foveon Inc.; http://www.foveon.com), which filters the color components (RGB) by wavelength-dependent absorption via silicon layers.
Basics of Digital Microscopy
filters that enable each sensor to image the appropriate color. However, these cameras are far more expensive than single-sensor ones. In general, color cameras are less sensitive than their monochrome counterparts because
of the additional beam-splitting and wavelength selection components. Nonetheless, this disadvantage may be offset by the ability to image a specimen at multiple wavelengths simultaneously.
12.2.6 Supplement 31
Current Protocols in Cytometry
Table 12.2.1 Microscope Magnification Requirements for Maximum Optical Resolution (λ = 0.515 µm)
Objective (NA)
Nyquist sampling Sampling interval in the interval in the object image plane when no plane (µm) relay lens is used
Relay lens needed (total magnification), 6.8-µm pixels
Relay lens needed (total magnification), 9-µm pixels
10× (0.45)
0.286
2.86
2.4× (24×)
3.2× (32×)
20× (0.75)
0.172
3.44
2.0× (40×)
2.6× (52×)
40× (0.75)
0.172
6.88
1.0× (40×)
1.3× (52×)
40× (1.15)
0.112
4.48
1.5× (60×)
2.0× (80×)
60× (0.95)
0.136
8.16
0.8× (50×)
1.1× (66×)
60× (1.40)
0.092
5.52
1.2× (74×)
1.6× (97×)
100× (1.25)
0.103
10.3
0.7× (66×)
0.9× (87×)
100× (1.40)
0.092
9.2
0.8× (74×)
1.0 (97×)
Sampling and Quantization The microscopic image of a specimen is continuous with respect to x and y coordinates (space) and intensity (amplitude). Digital imaging relies on the projection of such an image onto a 2-D array of light-sensitive elements. The output of each element (electric charge accumulated in the potential well) is proportional to the incident light intensity. Hence, the intensity is measured (probed) at a finite number of points (pixels) in the process called sampling. Similarly, the intensity is converted to a discrete scale (usually integer-based) during quantization. Accurate transition between optical and digital images requires that sensor characteristics be properly matched to the optical system performance. Rules for such optimization are discussed briefly in the following section.
Spatial Sampling A fundamental principle of sampling is expressed by the so-called Whittaker-Shannon sampling theorem, which is useful for understanding the information loss resulting from discrete sampling. The theorem states that if a two-dimensional function f(x,y) is bandlimited to spatial frequencies below fcx cycles per unit value of x (in the x direction), and to frequencies below fcy cycles per unit value of y (in the y direction), then the function can be completely reconstructed by taking four fcx fcy samples per unit area on the x,y plane. In the absence of noise, fcx and fcy are equal to the OTF cutoff frequency. Thus, the maximum sampling interval (in the x and y directions of the object plane) is given by the Nyquist criterion, Dso ≤ 0.25λ/NA.
Since the optical system (microscope) projects a magnified image of the object onto the sensor array, Ds in the image plane is given by Dsi ≤ (0.25λ/NA) × M where M is the total magnification. An alternative approach involves digitizing with a sampling interval that is no greater than one-half the size of the smallest resolvable feature of the optical image. If this method is adopted, Dso = 0.5d and depends on the resolution criterion. Hence, Ds in the image plane is given by Dso ≤ 0.5d × M. The pixel size of the sensor should not exceed Dso in order to achieve optimum sampling. This optimal sampling is usually obtained by varying the total magnification (M) of the microscope using a relay lens (see examples in Table 12.2.1). When the sampling frequency is lower than the Nyquist limit, then frequencies in the original signal that are higher than half the sampling rate are aliased (i.e., appear in the image as lower spatial frequencies). Conversely, if too many pixels are gathered by the camera, no additional spatial information is afforded, and the image is oversampled (see Fig. 12.2.9). The extra pixels theoretically do not contribute to the spatial resolution, but can often help improve the accuracy of feature measurements taken from a digital image (Young, 1996).
Quantization After an object has been imaged and sampled, each of the continuous intensities represented within the specimen is converted into a digital brightness value (level). The accuracy of the digital value is directly proportional to the bit depth of the digitizing device. For
Cellular and Molecular Imaging
12.2.7 Current Protocols in Cytometry
Supplement 31
Figure 12.2.9 Image of AO-stained fibroblast nucleus sampled optimally at the Nyquist rate (2x binning, A) and oversampled two times (no binning, B).
instance, if one bit is utilized, the image can represent only two brightness levels. Four or eight bits are sufficient to express 16 or 256 levels, respectively. The number of bits needed to represent a signal accurately depends on the SNR of the sensor: 20Log(Nbit ) ≥ SNR.
Basics of Digital Microscopy
The SNR of the sensor depends on the imaging conditions (see the equation above for calculation of camera system overall SNR). The bit depth should be chosen so that the maximum SNR can be accurately represented. The upper limit of precision may be estimated by SNRmax = well depth/readout noise. However, owing to the presence of dark and photon noise, this maximum SNR is not achieved in practice. On the other hand, obtaining real Nbit data requires more than simply the use of an N-bit ADC: illumination stability, freedom from digital electronic interference, and proper bandwidth for sampling rate are additional considerations. Hence the above equation gives a safe estimate of necessary bit depth. The quantization scheme discussed so far is linear; it assumes that all the brightness levels are separated by the same value (minimal noise). However, in the case of scientificgrade CCD signals, degradation comes mainly through photon noise, which is proportional √ to N. In such a system, if one demands that brightness levels be separated by an amount at least equal to their standard deviation, the difference between the first two such levels (1 events/pixel and 4 events/pixel) is 3 events/pixel, whereas that between the 15th and 16th levels is 31 events/pixel. In other words, the number of “meaningful” gray lev√ els is proportional to N. Hence, only 0.5Nbit is necessary to preserve the information.
The resolution limit of an optical microscope can be extended by deconvolution (Holmes et al., 1995). However, the presence of noise results in a decrease of cutoff frequency (Stelzer, 1997) and imposes a limit on deconvolution efficiency (Holmes and Liu, 1991; Holmes, 1992). Several methods for estimating the achievable resolution have been developed (van Dekker and Bos, 1997). As SNR is affected by pixel size, larger pixels (and sampling below optimum rate) may bring an increase in actual resolution when SNR is very low (Neifeld, 1998).
STORAGE OF DIGITAL IMAGE DATA Following sampling and quantization, digital images are stored as files in permanent memory (hard drives, CDs, DVDs, Flash ROM modules, and others). This section contains an overview of standards of color coding and compression, as these factors have significant impact on the fidelity and efficiency of digital image storage. A description of the most popular file formats is included as well.
Color Spaces Color spaces are schemes to code colors using components. Although several are found in digital applications (Gonzalez and Wood, 2002), images are usually stored using RGB, CMYK, or YUV. RGB (Red-Green-Blue) is a method of generating colors in a video system (monitors and cameras) that uses the additive primaries method. Percentages of red (R), blue (B), and green (G) are mixed to form the colors. Zero percent of the colors creates black, 100 percent of the colors creates white.
12.2.8 Supplement 31
Current Protocols in Cytometry
CMYK (or sometimes YMCK) is a color model used in color printing and is based on mixing pigments of the following colors: cyan (C), magenta (M), yellow (Y), and black (K, Key). The mixture of ideal CMY color is subtractive (cyan, magenta, and yellow printed together on white result in black). CMYK works through light absorption. The colors that are seen are from the wavelengths of light that are not absorbed. In CMYK, magenta plus yellow produces red, magenta plus cyan makes blue, cyan plus yellow generates green, and the combination of cyan, magenta, and yellow forms black. Because the “black” generated by mixing the subtractive primaries is not as dense as that of a genuine black ink (one that absorbs throughout the visible spectrum), fourcolor printing uses black in addition to the three primaries. YUV (also known as YCbCr and YPbPr) is a color space in which the Y stands for the luminance (brightness) component and U (Cb, Pb) and V (Cr, Pr) are chrominance (color) components. It is commonly used in video applications, where it is also referred to as component video. YUV signals are created from an original RGB source. The weighted values of R, G, and B are added together to produce a single Y signal, representing the overall brightness, or luminance, of that spot. The U signal is then created by subtracting the Y from the blue signal of the original RGB, and V by subtracting the Y from the red.
Compression Algorithms An array of algorithms have been developed in order to decrease the space occupied by a stored image file. These algorithms differ with respect to compression efficiency, speed, and complexity. Lossless algorithms introduce no distortion, in contrast to lossy compression schemes. However, the former offer smaller compression ratios than the latter. LZW is a lossless compression algorithm named after its developers, Lempel and Ziv, with later modifications by Welch. Typically, LZW compresses image files to about onehalf their original size. Compression ratios as high as 5:1 are also obtainable when the image has long runs or a lot of solid-color areas. LZW is referred to as a substitutional or dictionary-based encoding algorithm. The algorithm builds a data dictionary (also called a translation table or string table) of data occurring in an uncompressed data stream. Patterns of data (substrings) are identified in the data stream and are matched to entries in the
dictionary. If the substring is not present in the dictionary, a code phrase is created based on the data content of the substring, and is stored in the dictionary. The phrase is then written to the compressed output stream. When a reoccurrence of a substring is identified in the data, the phrase of the substring already stored in the dictionary is written to the output. Because the phrase value has a physical size that is smaller than the substring it represents, data compression is achieved. RLE (run-length encoding) is a lossless data compression algorithm that is supported by most image bitmap file formats. RLE works by reducing the physical size of a repeating string of characters. This repeating string, called a run, is typically encoded into two bytes. The first byte represents the number of characters in the run and is called the run count. In practice, an encoded run may contain 1 to 128 or 256 characters; the run count is usually expressed as the number of characters minus one (a value in the range of 0 to 127 or 255). The second byte is the value of the character in the run, which is in the range of 0 to 255, and is called the run value. RLE schemes are simple and fast, but their compression efficiency depends on the type of image data being encoded. A black-and-white image (two intensity levels) that is mostly white, such as the page of a book, will encode well (with ratios of 6:1 or better), owing to the large amount of contiguous identical data. An image with multiple intensity levels will not encode efficiently (ratios lower than 2:1 may be expected). Huffman encoding is a simple compression algorithm introduced by David Huffman in 1952 and supported today by several imaging file formats. This lossless algorithm uses a predefined dictionary of commonly occurring image byte patterns (strings), which are given low (short) indices (codes) in the dictionary. Data are encoded by replacing each image string that occurs in the dictionary with its index number. The dictionary is not part of the compressed file. Compression efficiency is dependent on whether the dictionary entries encode common byte patterns occurring in a particular image. Therefore, ratios between 1.3 and 2.5 are most common in practice. Deflation is a lossless compression algorithm that uses a combination of LZ77 (which is the basis of LZW) and Huffman coding to achieve better compression than is possible with either one alone. It was originally defined by Phillip W. Katz for his version of ZIP (PKZIP).
Cellular and Molecular Imaging
12.2.9 Current Protocols in Cytometry
Supplement 31
JPEG is a lossy compression scheme introduced by the Joint Photographic Experts Group to compress photographs. This scheme is based on the discrete cosine transform (DCT) and divided into the following stages: 1. Transformation of the image into an optimal color space. The best compression ratios are achieved when a luminance/chrominance color space, such as YUV, is used. 2. Down-sampling (up to 2× horizontal and vertical) of chrominance components (U and V) by averaging groups of pixels together. 3. Transformation with DCT performed on 8 × 8 blocks of pixels to give 64 frequency coefficients for a block. 4. Quantization of DCT coefficients in each block using weighting functions optimized for the human eye. Each of the 64 positions of the DCT block has its own coefficient, with the higher-order (frequency) terms being quantized more heavily than the lower-order (frequency) terms (that is, the higher-order terms have larger quantization coefficients). Furthermore, separate quantization tables are employed for luminance and chrominance data, with the chrominance data being quantized more heavily than the luminance data. This allows JPEG to discard data exploiting further differing sensitivity of the human eye to luminance and chrominance. It is this step that is controlled by the “quality” setting of most JPEG compressors. 5. Encoding the resulting coefficients (image data) using a Huffman variable-wordlength algorithm to remove redundancies in the coefficients.
Basics of Digital Microscopy
JPEG may achieve high (15:1 or better) compression ratios. The majority of JPEG encoders allow adjustment of the ratios via a setting for desired “image quality.” It should be emphasized that artifacts (distortions) are introduced even with the lowest possible compression (highest quality). JPEG2000 is a new image compression standard being developed by the Joint Photographic Experts Group. It is designed for different types of still images (bi-level, graylevel, color, multicomponent) and provides high-quality images at low bit rates, overcoming many of the limitations of the original JPEG standard. This scheme is based on the discrete wavelet transform (DWT) and divided into the following stages:
1. Splitting of the image into a set of tiles (a tile can occupy the whole image). Optional transformation into YUV (luminance/ chrominance ) color space. 2. Decomposition of the image with DWT into the low-resolution (low-pass) and high-resolution (high-pass) component images (sub-bands). 3. Scalar quantization of the DWT coefficients. This is a lossy step unless integer DWT was used. 4. Grouping of DWT coefficients from corresponding sub-bands. Decomposition of the grouped coefficients into bit planes. 5. Truncation of the resulting bit streams to achieve demanded quality (distortion). This is a lossy step. 6. Compression with a lossless algorithm (e.g., Huffman coding). JPEG2000 may compress images in a lossless manner (ratios up to 2:1). Better compression (ratios comparable to JPEG) may be obtained at the cost of introducing distortions. Nonetheless the distortions are less severe compared with those produced in JPEG.
File Formats Implementation of various color-coding and compression schemes has led to several image file formats. The choice of format depends on the type and intended use of the image data. This section includes a short overview of the most common formats. BMP (basic multilingual plane, bitmap) is the standard Microsoft Windows raster file format, which is not recognized by other computer systems (with the exception of a few programs). The format supports pixels represented by 1 bit (monochrome image), 4 bits (16 possible values), or 8 bits (256 values). In this case, a color table is used to assign displayed colors to these pixel values. The colors are encoded in an RGB scheme. Alternatively, pixels may be represented by 24 bits. In this case, each RGB component is represented by three 8-bit segments (true color). Thus, no color table is used. The format supports lossless (run-length) compression for 4- or 8-bit variants. GIF (graphics interchange format) is a service mark of CompuServe Incorporated introduced in 1987 (GIF87a). In 1989 a revised specification (GIF89a) added some features to the format, namely the capacity to store animations and textual comments. Pixels are encoded using 8 bits (256 possible values). A color
12.2.10 Supplement 31
Current Protocols in Cytometry
table is used to assign displayed color (RGB scheme) to a pixel value. The colors include binary transparency. Every pixel in the graphic that has the value assigned as transparent becomes invisible. The GIF file format also supports interlacing, i.e., saving a file using four passes instead of just one. On each pass, only certain lines of the image are saved to the file. To display an interlaced image, GIF loads progressively so that a lower-resolution preview can be seen before the full image is shown. The format supports lossless compression with the LZW algorithm. However, since the algorithm was patented (by Unisys) several programs write only uncompressed files (which nonetheless are compliant with LZW specification). JPEG implements the lossy JPEG compression algorithm. The format supports pixel encoding using only 24 bits divided into three color components (8 bits each). These are luminance (Y) and two chrominance components (U and V, respectively). A JPEG file can be saved as a progressive JPEG, which is very similar to the interlaced GIF. As with GIF, this presents a low-quality image first, and over several passes improves the quality. JPEG does not support transparency, however. JPEG2000 was created by the Joint Photographic Experts Group committee with the intention of superseding their original JPEG standard. The JPEG2000 algorithm may be used to obtain either lossy or lossless compression. The specification supports up to 16,384 components with 38-bit precision each. However, currently pixels are encoded as 1-, 4-, or 8-bit indexed color (with color table), 8- or 16-bit grayscale, 24-bit RGB (3 color components), or 32-bit CMYK (4 color components). Some implementations offer the alpha (transparency) channel (component). PNG (portable network graphics) is a relatively new image format that is becoming popular on the World Wide Web and elsewhere. The format was developed largely to deal with some of the shortcomings of GIF. PNG stores image data using 1,2, 4, or 8 (PNG8) bits per pixel and a color table to assign displayed color (RGB scheme) to a pixel value. The table may be omitted in 8-bit images, which are saved as grayscale (i.e., with pixel values representing intensities only but not colors). Alternatively, pixels may be encoded using 24 bits to represent the three components of RGB color (PNG24). Both grayscale and RGB PNGs can have 16-bit precision, that is, 16-bit and 48bit pixels, respectively. Furthermore, both can also have an 8-bit alpha channel to represent
256 levels of transparency for a pixel. Other image attributes that can be stored in PNG files include gamma values (see Gonzalez and Woods, 2002), background color, and textual information. PNG uses a lossless compression algorithm known as deflation. Tagged Image File Format (abbreviated TIFF) was created by Aldus for use with PostScript printing. Now controlled by Adobe, TIFF has become the standard graphics format for high-bit-depth (32-bit) graphics, and can be directly manipulated using PostScript. TIFF features an array of options that can be used to include all sorts of image formats in the file. The actual characteristics of a given TIFF file, including simple geometry of the image, data arrangement, and compression type, are defined using specific tags in the file header. Pixels represent indexed colors (with color table) or grayscale values (without the table) with 1, 4, or 8 bits. The format also supports full-color images in which color is stored using components encoded with 8- or 16-bit precision. The components may belong to RGB, YUV, or CMYK color spaces. A separate alphacomponent value may be assigned to pixels to describe transparency. Therefore, TIFF permits using up to 64 bits per pixel. The format may be used as a container for images compressed in a lossless manner with Huffman, RLE, LZW, or ZIP (deflate) algorithms. Lossy compression is achieved with JPEG. Nonetheless, despite the flexibility of this format, the vast majority of TIFF files, and the code that reads them, are based on a simple 32-bit (RGB and alpha) uncompressed image. Table 12.2.2 summarizes the main features of file formats discussed in this unit. File formats are compared in Figure 12.2.10 as well. One may note that most efficient lossless compression algorithms are implemented in JPEG2000, TIFF, and PNG. The first format also permits lossy compression with relatively few distortions. However, JPEG2000 is not yet widely supported, having been introduced only recently. The basic standard of TIFF, on the other hand, is widely supported and highly versatile. However, owing to their complex structure, TIFFs tend to be rather large. PNG lies in between as far as size and support are concerned.
CONCLUSIONS The usefulness of digital microscopy depends on whether or not image registration and transfer into digital form are performed in an optimal way. The first step requires finding a proper balance between maximizing sampling
Cellular and Molecular Imaging
12.2.11 Current Protocols in Cytometry
Supplement 31
Figure 12.2.10 Comparison of image file formats. (A) Original image of tubulin fibers in FITCstained fibroblast and (B) its magnified fragment. This image (stored as an uncompressed TIFF, 8-bit grayscale) was used as the standard for calculation of the compression ratios (Cr ). The image was compressed using: (C) lossless LZW (TIFF); (D) lossless deflation (PNG); JPEG2000 using either lossless algorithm (E) or lossy algorithm (F); and (G) lossy JPEG (JPG). Compression errors (lossy JPEG2000 and JPEG) are illustrated using difference images (pixel values were multiplied by 8).
Basics of Digital Microscopy
12.2.12 Supplement 31
Current Protocols in Cytometry
Table 12.2.2 Comparison of Common Image File Formats
Compression
File type
Bit depth
Color coding
Transparency (bits)
BMP
1/4/8
Grayscale
—
RLE
—
Poor Inefficient compressiona (Microsoft format)
24
RGB
—
—
—
—
—
GIF
8
RGB color table
1
LZW
—
Excellent
No grayscale support
JPEG
24
YUV
—
—
JPEG
Good
Very efficient compressiona (with distortion). No RGB and grayscale support.
Color table
38b
JPEG200
JPEG200
Fair
Very efficient compression.a Relatively new format intended to replace JPEG.
8/16
Grayscale
—
—
—
—
—
24/48
RGB
—
—
—
—
—
36/64
CMYK
—
—
—
—
—
1/2/4/8
Grayscale or RGB color table
8
Deflation
—
Good
Efficient compression. Format developed to replace GIF on the World Wide Web.
24/48
RGB
—
—
—
—
—
1/4/8
RGB color table
8
JPEG Deflation (ZIP), RLE, Huffman, LZW
Good
Format offers large array of features, which may be extended further using specific tags. Only a limited number of these are supported universally. Efficient compression.a
1/4/8/16
Grayscale
—
—
—
—
—
24/48
RGB, YUV
—
—
—
—
—
32/64
CMYK
—
—
—
—
—
JPEG2000 1/4/8
PNG
TIFF
Lossless
Lossy
Portability Remarks
a Microscopic images. b As specified by the standard. However, this feature is currently not supported by any program.
density (and thus precision) and SNR (and accuracy). If the balance is found, then imaging is performed with maximum attainable resolution. The second step requires matching characteristics of the image file with generated image data so that the file occupies minimum space and can be easily transferred with no loss of necessary information.
LITERATURE CITED Gonzalez, R.C. and Woods, R.F. 2002. Digital Image Processing, 2nd ed. Prentice Hall, Upper Saddle River, N.J. Holmes, T.J. 1992. Blind deconvolution of quantum-limited incoherent imagery: Maximum-likelihood approach. J. Opt. Soc. Amer. A-9:1052-1061.
Cellular and Molecular Imaging
12.2.13 Current Protocols in Cytometry
Supplement 31
Holmes, T.J., Bhattacharyya, S., Cooper, J.A., Hanzel, D., Krishnamurti, V., Lin, W.C., Roysam, B., Szarowski, D.H., and Turner, J.N. 1995. Light microscopic images reconstructed by maximum-likelihood deconvolution. In Handbook of Biological Confocal Microscopy (J.B. Pawley, ed.) pp. 389-402. Plenum Press, New York. Holmes, T.J. and Liu, Y.-H. 1991. Acceleration of maximum-likelihood image restoration for fluorescence microscopy and other incoherent imagery. J. Opt. Soc. Amer. A-8:893-907. Houston, W. V. 1926. The fine structure and wavelength of the Balmer Lines. Astrophys. J. 64:81. Neifeld, M.A. 1998. Information, resolution and space-bandwidth product. Opt. Lett. 15:14771479. Rayleigh, J.W.S. 1879. Investigations in Optics with special reference to the Spectroscope. Phil. Mag. 8:261-274.
Reichman, J. 2000. Handbook of Optical Filters for Fluorescence Microscopy. Chroma Technology Corp., Rockingham, Vt. Sparrow, C. M. 1916. On spectroscopic resolving power. Astrophys. J. 44:76. Stelzer, E.H.K. 1997. Contrast, resolution, pixelation, dynamic range and signal-to-noise ratio: Fundamental limits of resolution in fluorescence light microscopy. J. Microsc. 189:15-24. Van Dekker, A.J. and Bos, A. 1997. Resolution: A survey. J. Opt. Soc. Amer. A-14:547-557. Young, I.T. 1996. Quantitative microscopy. IEEE Eng. Med. Biol. 15:59-66.
Contributed by Tytus Bernas Purdue University West Lafayette, Indiana
Basics of Digital Microscopy
12.2.14 Supplement 31
Current Protocols in Cytometry
Modern Confocal Microscopy
UNIT 12.3
The idea of confocal microscopy was first patented by Marvin Minsky in 1957. However, the origins of confocal optics go back to a microscopic spectrophotometer made by Hiroto Naora in the immediate postwar years. His first publication on this subject in English appeared in Science over 50 years ago (Naora, 1951). The first purely analog mechanical confocal microscope was designed and built by Mojmir Petran and Maurice D. Egger (Egger and Petran, 1967). Two years later Egger and Davidovits introduced the first confocal instrument utilizing a laser light source (Davidovits and Egger, 1969). In 1983 Cox and Sheppard reported the use of a computer to collect and store confocal images (Cox and Sheppard, 1983), and finally in 1987 the first commercial confocal microscope, the Bio-Rad MRC500 based on the design by Brad Amos, appeared. This instrument permitted researchers to collect, by means of optical sectioning, 3-D images of a living, thick specimen, with temporal and spatial resolutions as good as or superior to those of video microscopy. The term confocal in the context of biological microscopy was probably used for the first time by Brakenhoff et al. (1979). It describes an optical platform in which illumination is confined to a diffraction-limited spot in the specimen and detection is similarly confined by placing an aperture (a pinhole) in front of the detector in a position optically conjugate to (i.e., simultaneously focused in the same plane as) the focused spot (Amos and White, 2003; also see UNIT 2.8). The most commonly used method of image collection in confocal fluorescence microscopy is based on point scanning. In the laser scanning confocal instrument, the scanning point light source and the detector aperture share a common focus at the level of the specimen. The laser beam, guided by galvanometer-driven mirrors, is deflected in the direction of a microscope objective and focused into the specimen. The fluorescent light is emitted from the specimen in all directions, but only a part of it is collected by the microscope objective, focused onto the confocal pinhole, and converted into an electrical signal by the detector. Light that is emitted from locations either in front of or behind the focal point in the object is focused either in front of or behind the detector pinhole (Stelzer et al., 1991; see Fig. 12.3.1). Instead of scanning the sample with a laser beam, similar effects can be achieved using multiple pinholes arranged in a raster pattern. The most commonly used pattern— the Nipkow disk—consists of a series of rectangular perforated holes arranged in an Archimedes spiral. Using a Nipkow disk in a scanning confocal microscope, one can observe the sample in real time through the eyepiece. However, the apparent drawback of the spinning-disk microscope is its low illumination efficiency—usually less than a few percent. The light throughput, and consequently the system sensitivity, can be improved by the use of closely spaced, large pinholes. This approach is, however, impractical, as the confocal-imaging properties of such an instrument would be degraded by the increase in cross-talk between neighboring pinholes. Recent developments in the Nipkow disk–based system have solved some of these limitations. Detailed discussion of various methods of scanning can be found in UNIT 2.8. Confocal technology offers the following advantages: 1. Improved resolution in the “xy” plane: Achievable resolution of optical instruments is limited by the diffraction of light. In the case of an optical microscope, Fraunhofer diffraction on a circular aperture faithfully models the imaging process (Born and Wolf,
Contributed by Bartek Rajwa Current Protocols in Cytometry (2005) 12.3.1-12.3.12 C 2005 by John Wiley & Sons, Inc. Copyright
Cellular and Molecular Imaging
12.3.1 Supplement 31
Figure 12.3.1 Conventional (A) and confocal (B) fluorescence microscopes: (a) light source; (b) dichroic mirror; (c) objective lens; (d) plane of focus; (e) biological sample; (f) light detector (usually a CCD camera for a conventional fluorescence microscope, and a photomultiplier for a point-scanning confocal microscope); and (g) confocal aperture, blocking out-of-focus light.
1999). The intensity distribution I(ν) produced by diffraction is proportional to:
Equation 12.3.1
where J1 (x) is a Bessel function of the first kind, and ν is a coordinate which is related to the transverse distance in the focal plane d:
Equation 12.3.2 Modern Confocal Microscopy
This intensity distribution is also called the Airy diffraction pattern or Airy disk and describes the two-dimensional intensity point spread function (PSF) of the system; θ is the half angle of the cone of light converging to an illuminated spot or diverging from
12.3.2 Supplement 31
Current Protocols in Cytometry
one; and λ is the wavelength. The resolution r defined by Rayleigh is the necessary separation of two point objects such that their diffraction pattern shows a detectable drop in intensities between them. When the distance to the first dark fringe in the diffraction pattern is treated as a measurement of the resolution in a Rayleigh sense, this detectable drop is about 26%. The simpler full width half maximum (FWHM) criterion relies on the measurement of the Airy disk at half its maximum height. For the conventional microscope this is about the same as the radius of the first minimum (Amos, 2000):
Equation 12.3.3
where n is the index of refraction of the medium; NA is the product n sin θ and is called the numerical aperture. In a confocal instrument, the use of a small confocal pinhole causes the photomultiplier to function as a point detector. Therefore, if identical optics are used for illumination and observation, the function describing intensity distribution becomes a product of two identical Airy disks, and is proportional to:
Equation 12.3.4
The resolution limit r derived from the function above, defined again as the width of the Airy disk at half its maximum height, differs from the resolution limit for a classical microscope and is expressed by the following equation:
Equation 12.3.5
The confocal system has an amplitude point spread function (PSF) that is narrower than that of the corresponding nonconfocal system by factor of 0.72, as measured between the half-power points (the FWHM criterion; see Fig. 12.3.2). Indeed, comparing Equation 12.3.2 and Equation 12.3.4 it is evident that confocal resolution in the “xy” plane (measured using the FWHM criterion) is increased over that achievable by conventional optical microscopy. However, the distance from peak to first zero (the conventional Rayleigh criterion) is not changed (Sheppard and Choudhury, 1977). In practice, the resolution improvement varies for different imaging modes (fluorescence, backscattered light, and transmitted light). 2. Improvement of signal-to-noise ratio through rejection of out-of-focus light from planes other than the plane of focus: Confocal microscopes use only light that comes from the volume of the object that is conjugate to the detector and the source. Once the background light has been reduced, the full resolution available from the optics may be utilized. The confocal diffraction pattern has much less energy outside the central peak than does the wide-field pattern. Hence, a bright object near a dim one is less likely to contribute background light to spoil the contrast. This means that the dim object resolved in a confocal laser scanning microscope in the Rayleigh sense can be really seen by an observer as resolved. Therefore, rejection of out-of-focus background results in an improved signal-to-noise ratio. Sandison and Webb showed that the reduction of background in a
Cellular and Molecular Imaging
12.3.3 Current Protocols in Cytometry
Supplement 31
Figure 12.3.2
Axial and lateral point spread functions of confocal and conventional microscopes.
point-scanning confocal system could provide signal-to-noise ratio 10 times better than that of a conventional microscope. The spinning-disk confocal instruments have a signalto-noise ratio that is greater than that of the conventional microscope by a factor of 2 to 3 (Sandison and Webb, 1994). Confocal optics also discriminate against diffuse scattering from planes away from the focal plane (Sheppard and Wilson, 1978). 3. Three-dimensional imaging capability—optical sectioning through thick specimens: For a conventional microscope the intensity variation as a point object is displaced along the axis is given by:
Equation 12.3.6
where u is a normalized optical coordinate related to the axial distance z by:
Equation 12.3.7
In a confocal microscope this intensity is squared (Sheppard and Wilson, 1978):
Modern Confocal Microscopy
Equation 12.3.8
12.3.4 Supplement 31
Current Protocols in Cytometry
In consequence the axial extent of the confocal microscope PSF is about 30% smaller compared to that of the conventional microscope PSF (Fig. 12.3.2). Therefore axial resolutions for conventional and confocal microscopes can be expressed as:
Equation 12.3.9
respectively (Jonkman and Stelzer, 2002). If depth of field is defined in terms of the falloff in maximum intensity for a point image, then one can see that the depth of field for a confocal microscope is only slightly reduced relative to that of a conventional microscope (Sheppard and Matthews, 1987; Sheppard, 1988). Therefore, the improvement in axial resolution (Equation 12.3.9) does not explain the optical sectioning capabilities of confocal microscopy. However, if one considers the variation in the integrated intensity for the image of a point source (which shows the total power in the image), one will see why the confocal microscope discriminates against parts of the object not in the focal plane. In a conventional microscope, the integrated intensity of the light emanating from any one point in the specimen is almost unchanged as one moves away from the focus. However, the confocal integrated intensity PSF has a maximum in the focal plane. Therefore, in contrast to conventional fluorescent microscopes, confocal systems collect signal only from fluorochromes located in the neighborhood of the focus (Fig. 12.3.3; Sheppard and Wilson, 1978; Jonkman and Stelzer, 2002). It is important to remember that the axial imaging properties of a confocal system are degraded by the presence of spherical
Figure 12.3.3 Integrated intensities of conventional and confocal microscopes (adapted from Sheppard and Wilson, 1978).
Cellular and Molecular Imaging
12.3.5 Current Protocols in Cytometry
Supplement 31
aberration, which may happen when focusing with a high-NA, oil-immersion objective into an aqueous biological object. Confocal instruments faithfully image structures ranging from those with dimensions as small as subcellular organelles up to whole tissue preparations. However, a variety of technical constraints limit the maximal thickness of the objects that can be imaged. To obtain 3-D images that closely represent the geometry of the sample, the light path through the sample must be as short as possible, since imaging artifacts like astigmatism, spherical aberration, and intensity attenuation increase with path length (Hell et al., 1993). Coupled with digital reconstruction techniques, confocal microscopy can extract image information which, while present in the data, is not easily accessible through simple presentation of the individual sections.
MODERN APPLICATIONS OF CONFOCAL MICROSCOPY Although detailed discussion of the application of confocal microscopy is beyond the scope of this short commentary, it must be emphasized that the impressive development of this field has been driven mostly by modern research methods and applications such as fluorescence resonance energy transfer (FRET) and photobleaching-based techniques (FRAP, IFRAP, FLIP) as well as by the increasing popularity of fluorescent protein markers and, finally, the recent introduction of nanocrystal-linked fluorochromes (“quantum dots”). Below, two of these important techniques will be briefly described.
Fluorescence Resonance Energy Transfer (FRET) FRET is a distance-dependent photophysical interaction between two molecules in which the excitation energy of one molecule (the donor) can be transferred nonradiatively to the other (the acceptor). For FRET to occur, molecules must be in close proximity (no greater than 100 Å), and the absorption spectrum of the recipient molecule must overlap the emission spectrum of the donor (Emptage, 2001; Wouters et al., 2001). FRET has been used in both spectrophotometry and microscopy for a number of years; however, only recently, with the introduction of fluorescent proteins that form convenient donor– acceptor FRET pairs, has the method become widespread. For detailed discussion of FRET, see UNIT 1.12. Photobleaching-Based Methods (FLIP/FRAP/IFRAP) Photobleaching-based techniques have been known since the mid 70’s. However, only with the proliferation of confocal microscopy and the introduction of fluorescent proteins have these methods been made available to a wider group of cell biologists. Three basic photobleaching strategies are used in biological research. In fluorescence recovery after photobleaching (FRAP), a small region of interest is bleached by scanning it with high laser power, and the recovery of fluorescence is monitored using low laser power. In reverse FRAP (IFRAP), the entire sample except the region of interest is bleached, then loss of fluorescence from the area of interest is observed. Finally, the fluorescence loss in photobleaching (FLIP) technique requires monitoring of one site and bleaching of another site in the sample. If there is an exchange of fluorescent molecules between two regions, an increase of fluorescence in the bleached site will occur, whereas the other region should gradually exhibit loss of fluorescence (McNally and Smith, 2002). For detailed discussion of FRAP please see UNIT 2.12.
Modern Confocal Microscopy
12.3.6 Supplement 31
Current Protocols in Cytometry
PROGRESS IN CONFOCAL MICROSCOPY HARDWARE The confocal instruments available on the market today are more complex and sophisticated than their early counterparts. Here is a brief outline of some recent developments.
Spectral Imaging Instruments Multispectral imaging is a well known tool widely used for remote sensing. However, only recently has it been introduced to standard confocal scanning instruments, becoming one of the most important advances in the field of biological imaging. There are a number of reasons why multispectral fluorescence microscopy has become such an in-demand tool among cell biologists. Firstly, spectral overlap or cross-talk can be difficult to eliminate even in the simplest systems where two or three fluorophores are used simultaneously. Secondly, spectral fingerprints of intrinsic or introduced fluorophores have the potential to reveal information about the physiological processes inside live cells. Thirdly, spectral imaging can enhance other sophisticated techniques like FRET or dye-ratio imaging (Dickinson et al., 2001; Berg, 2004). Today almost all producers of confocal systems offer multispectral imaging capability. Various technologies are available. The now defunct Bio-Rad Microscience Division proposed the simplest but also most robust design (Fig. 12.3.4). Multispectral information is collected by three conventional PMTs after passing through dual filter wheels positioned in front of each of them. The first filter wheel contains long-pass filters and the second one short-pass filters, providing the means for selecting a narrow detection bandwidth. To obtain the full spectrum of the sample, multiple images are collected and the spectra are built only after acquisition (Berg, 2004). This approach can limit severely the capability to collect multispectral information from rapidly changing samples. Leica instruments (Leica Microsystems AG; http://www.leica-microsystems.com) also use the conventional PMT to acquire multispectral data. Narrow bands are selected by a slit mounted in front of a dispersing prism (Fig. 12.3.5). The system offers very high spectral resolution; however, it requires very high precision of the slit movement. This may lead to various problems caused by vibration, changes in humidity, or temperature. The Zeiss design (Carl Zeiss; http://www.zeiss.com) is the most radical departure from the classical optical arrangements of confocal microscopy (Fig. 12.3.6). The Zeiss
Figure 12.3.4 The multispectral detection system of Bio-Rad Rainbow instruments: (a) fluorescence signal; (b) filter wheel containing long-pass filters; (c) filter wheel containing short-pass filters; (d) photomultiplier.
Cellular and Molecular Imaging
12.3.7 Current Protocols in Cytometry
Supplement 31
Figure 12.3.5 The multispectral detection systems of Leica confocal instruments: (a) fluorescence signal; (b) prism; (c) dispersed light; (d) mechanical aperture selecting the band of interest; (e) photomultiplier.
Modern Confocal Microscopy
Figure 12.3.6 The multispectral detection system of Zeiss Meta instruments: (a) fluorescence signal from the biological sample; (b) diffraction grating; (c) dispersed fluorescence light; (d) 32channel multianode photomultiplier.
12.3.8 Supplement 31
Current Protocols in Cytometry
META system uses a multianode PMT from Hamamatsu (Hamamatsu Photonics K.K.; http://www.photonicsonline.com) to collect signal dispersed by a grating. Because Zeiss instruments can collect multiple spectral points simultaneously, the full spectral scan requires less time than in other systems. However, the sensitivity of the META system is often questioned. The Hamamatsu 32-anode PMT operates between 185 and 650 nm, with a peak response at 420 nm, so most of the biologically relevant measurements are made on the top fringe of its sensitivity region. Another problem with the system is that the electronics of the current version allows simultaneous collection of only 8 channels at a time.
New Nipkow-Disk Systems Until recently, confocal microscopy technology did not offer reliable and inexpensive systems for real-time observations. Designs introduced by Bio-Rad (DVC line scanner built by Amos), Nikon (ultrafast point scanner built by Tsien and Backsai), and Noran did not gain popularity, owing partially to the complicated design and partially to the high cost. On other hand, the spinning disk–based designs suffered from a very weak light budget (i.e., the number of photons reaching the detector was very low), making the instruments almost unsuitable for work with weakly fluorescent samples. This problem was solved by Tanaami and colleagues, who improved the original design by Petran. The new spinning-disk instruments utilize two disks instead of one. The upper disk consists of several thousand tiny microlenses. When light illuminates this disk, the microlenses focus the light onto the lower disk, which has several thousand pinholes arranged in the same pattern. The light passing through each pinhole is aimed by the objective lens at a spot on the specimen. Light from the specimen passes back through the objective lens and pinholes of the first disk, and is reflected by a beam splitter to a CCD camera. The upper disk containing the microlenses and the lower disk containing the pinholes are physically connected and rotated together by an electrical motor, thus raster-scanning the specimen (Tanaami et al., 2002). This design, marketed by Yokogawa, found its way to many confocal instruments of the market. Among these the integrated system offered by Perkin-Elmer has become especially popular. High-Throughput Screening/High-Content Screening Recently, confocal optics found its way to high-throughput (HTS)/high-content screening (HCS) instruments (Eggeling et al., 2003). The major limitation of the traditional confocal technique in the context of high throughput is, of course, speed, or rather the lack thereof, so HTS/HCS instruments adopt line-scanning or spinning-disk technology. For instance, the ArrayScan HTS reader (Cellomics Inc.; http://www.cellomics.com) uses a full line of the sample illuminated and mapped onto an appropriate pixel line on the charge-coupled device (CCD) camera (Eggeling et al., 2003). For the sake of higher throughput, the Opera (Evotec Technologies; http://www.evotec-technologies.com) and the Pathway HT (Atto Bioscience; http://www.atto.com) employ a Nipkow disk to project fluorescence from several confocal volumes in parallel to a CCD camera (Zemanova et al., 2003). Programmable Array Microscopes (PAMs), Structured Illumination: Alternative Confocal Technologies Progress in confocal hardware development did not stop with the well established technologies of point scanning, line scanning, and spinning disk–based scanning. A number of new ideas have been introduced in recent years, promising simpler, more robust, and less expensive confocal microscopy. Programmable array microscopes (PAMs) are a family of microscope systems in which a spatial light modulator is placed in an image plane of the microscope and used to generate patterns of illumination and/or detection (Hanley et al., 1999). A good example of spatial
Cellular and Molecular Imaging
12.3.9 Current Protocols in Cytometry
Supplement 31
light modulators is the digital micromirror device (DMD) designed by Texas Instruments. The DMD is a semiconductor-based “light switch” array of thousands of individually addressable, tiltable mirror pixels. DMD technology is widely used as a spatial light modulator for projectors. In a microscope, micromirrors of DMDs can be used to create a pattern of reflection pinholes for highly parallel light collection. Microscopes using DMDs to create confocality have been reported, but the idea still awaits commercialization (Liang et al., 1997; Hanley et al., 1998). Another interesting technique suggests utilization of so-called structured illumination. This concept proposes to change the illumination system of the microscope to project a single spatial-frequency grid pattern onto the object. A microscope utilizing such an illumination model would efficiently image only that portion of the object for which the grid pattern is in focus. Obviously, the resultant image would have the unwanted grid pattern superimposed. However, a simple yet ingenious method has been proposed to remove the grid pattern in real time, permitting acquisitions of optically sectioned images from a conventional wide-field microscope (Neil et al., 1997). The concept of structured illumination has been commercialized and the product based on it, called OptiGrid, is available from Thales Optem (http://www.thales-optem.com) and Klughammer (http://www.klughammer.de).
Light Sources: Gas and Semiconductor Lasers Historically noble-gas gas lasers have been the most commonly used light sources in scanning confocal microscopy. In fact, they were the only reliable lasers available when the first commercial scanning confocal systems were introduced. Until now, the most commonly employed lasers included argon-ion and krypton-argon lasers. Argon lasers produce very intense lines at 488 and 514.5 nm. Other lines have at most one-third the intensity of these principal lines. The small argon-ion lasers used in confocal microscopy produce 10- to 500-mW TEM00 mode at 488 nm. They are compact and air cooled, stabilize in <15 min after being turned on, and exhibit low amplitude noise. Kryptonargon lasers contain a mixture of argon and krypton gases. They provide both the strong blue and green emissions of argon lasers, and in addition, they produce the red and yellow lines of the krypton-ion transitions at 647.1 and 568.2 nm. Helium-neon lasers (the first gas lasers designed), which are available now with 543.5-nm lines, have also gained popularity among confocal microscopists thanks to their low cost, reliability, and compactness. Although gas lasers are probably still the most popular lasers among users of confocal system, undoubtedly the area of laser technology that is of greatest interest to confocal microscopists is semiconductor diode laser development. Even though diode lasers have been commercially available since 1962, only recently have they offered sufficient output power and beam quality to be used in imaging systems. The most attractive feature of diode lasers is believed to be their price. Unfortunately, the cheap mass-produced diodes known from CD and DVD players emit in red and have found little use in biological confocal microscopy. Among red diodes only the 635-nm AlGaInP lasers can be easily used in standard confocal systems.
Modern Confocal Microscopy
Currently the majority of diode lasers emitting green or blue light use the technology of solid-state diode pumping and frequency doubling (DPSSFD). The most common DPSSFD lasers use a crystal doped with neodymium (Nd)—either Nd:YAG (yttrium aluminum garnet) or Nd:YVO4 (Nd:yttrium orthovanadate). These lasers can then be frequency doubled to produce violet (405-nm), blue (475-nm), and green (532-nm) lines. DPSSFD lasers are inexpensive but unfortunately often demonstrate significant intensity fluctuations.
12.3.10 Supplement 31
Current Protocols in Cytometry
Recently an alternative approach for providing shorter wavelengths (440 nm, 405 nm, 375 nm) became attainable with the introduction of the first commercial blue diode lasers. Currently blue and violet gallium nitride (GaN) laser diodes are manufactured by only a few companies, using two main technologies: by growing GaN crystals on dissimilar materials like sapphire (Nichia Chemicals, Japan) and silicon carbide (Cree), or by utilizing the unique approach of extremely high pressure to grow GaN crystals on GaN substrates (Unipress Top-GaN). Although both technologies are still very complicated and expensive, the blue diode lasers found their way to confocal microscopy almost immediately after being introduced (Girkin and Ferguson, 2000). The Coherent Vioflame/Radius (Coherent Laser Inc.; http://www.cohr.com) and the iFLEX-2000 (Point Source; http://www.point-source.com) are examples of commercial lasers utilizing the new GaN diodes. Sadly, the 488-nm wavelength is outside the range currently available from GaN semiconductor lasers. The closest line offered by the DPSSFD Nd:YAG lasers mentioned earlier is 475 nm. Direct doubling of semiconductor lasers remains the most promising option despite the difficulties and complexity of such a solution. The situation improved with the introduction of VECSEL (Vertical External Cavity Surface Emitting Laser) systems (Tropper et al., 2004). Sapphire—an optically pumped VECSEL from Coherent Laser—is pumped by a high-power 808-nm laser diode and lases at 946 nm, which is intracavitydoubled to 488 nm. A competitive product from Novalux (http://www.novalux.com) uses electrical pumping. An excellent detailed discussion of the latest developments in laser technology can be found in UNIT 1.9.
CONCLUSION Despite the fact that more sophisticated techniques like 2-photon microscopy, 4-pi microscopy, and others have become available, confocal microscopy, owing to its simplicity and affordable price, and to the progress in laser technology and computing associated with it, remains the most widely used tool for three-dimensional optical imaging of cells and tissues. The future of confocal microscopy will most likely be shaped by two trends—introduction of small (“personal”) confocal instruments, and expansion of highly automated HTS/HCS instruments utilizing confocal optics.
Literature Cited Amos, W.B. 2000. Instruments for fluorescence imaging. In Protein Localization by Fluorescent Microscopy (V.J. Allan, ed.) pp. 67-108. Oxford University Press, Oxford. Amos, W.B. and White, J.G. 2003. How the confocal laser scanning microscope entered biological research. Biol. Cell 95:335-342. Berg, R.H. 2004. Evaluation of spectral imaging for plant cell analysis. J. Microsc. 214:174-181. Born, M. and Wolf, E. 1999. Principles of Optics: Electromagnetic Theory of Propagation, Interference and Diffraction of Light, 7th expanded ed. Cambridge University Press, Cambridge. Brakenhoff, G.J., Blom, P., and Barends, P. 1979. Confocal scanning light-microscopy with high aperture immersion lenses. J. Microsc. (Oxf.) 117:219-232. Cox, I.J. and Sheppard, C.J.R. 1983. Scanning optical microscope incorporating a digital framestore and microcomputer. Appl. Optics 22:14741478.
Davidovits, P. and Egger, M.D. 1969. Scanning laser microscope. Nature 223:831-832. Dickinson, M.E., Bearman, G., Tille, S., Lansford, R., and Fraser, S.E. 2001. Multi-spectral imaging and linear unmixing add a whole new dimension to laser scanning fluorescence microscopy. Biotechniques 31:1272, 1274-1276, 1278. Eggeling, C., Brand, L., Ullmann, D., and Jager, S. 2003. Highly sensitive fluorescence detection technology currently available for HTS. Drug Discov. Today 8:632-641. Egger, M.D. and Petran, M. 1967. New reflectedlight microscope for viewing unstained brain and ganglion cells. Science 157:305-307. Emptage, N.J. 2001. Fluorescent imaging in living systems. Curr. Opin. Pharmacol. 1:521-525. Girkin, J.M. and Ferguson, A.I. 2000. Confocal microscopy using an InGaN violet laser diode at 406 nm. Optics Express 7:336-341. Hanley, Q.S., Verveer, P.J., Gemkow, M.J., ArndtJovin, D., and Jovin, T.M. 1999. An optical sectioning programmable array microscope
Cellular and Molecular Imaging
12.3.11 Current Protocols in Cytometry
Supplement 31
implemented with a digital micromirror device. J. Microsc. 196(part 3):317-331.
Sheppard, C.J.R. 1988. Super-resolution in confocal imaging. Optik 80:53-54.
Hanley, Q.S., Verveer, P.J., and Jovin, T.M. 1998. Optical sectioning fluorescence spectroscopy in a programmable array microscope. Appl. Spectrosc. 52:783-789.
Sheppard, C.J.R. and Choudhury, A. 1977. Imageformation in scanning microscope. Optica Acta 24:1051-1073.
Hell, S., Reiner, G., Cremer, C., and Stelzer, E.H.K. 1993. Aberrations in confocal fluorescence microscopy induced by mismatches in refractive index. J. Microsc. (Oxf.) 169:391-405. Jonkman, J.E.N. and Stelzer, E.H.K. 2002. Resolution and contrast in confocal and two-photon microscopy. In Confocal and Two-Photon Microscopy. Foundations, Applications, and Advances. (A. Diaspro, ed.) pp. 101-126. WileyLiss, New York. Liang, M.H., Stehr, R.L., and Krause, A.W. 1997. Confocal pattern period in multiple-aperture confocal imaging systems with coherent illumination. Optics Lett. 22:751-753. McNally, J. and Smith, C.L. 2002. Photobleaching by confocal microscopy. In Confocal and TwoPhoton Microscopy. Foundations, Applications, and Advances. (A. Diaspro, ed.) pp. 525-538. Wiley-Liss, New York. Naora, H. 1951. Microspectrophotometry and cytochemical analysis of nucleic acids. Science 114:279-280. Neil, M.A.A., Juskaitis, R., and Wilson, T. 1997. Method of obtaining optical sectioning by using structured light in a conventional microscope. Optics Lett. 22:1905-1907. Sandison, D.R. and Webb, W.W. 1994. Background rejection and signal-to-noise optimization in confocal and alternative fluorescence microscopes. Appl. Optics 33:603-615.
Sheppard, C.J.R. and Matthews, H.J. 1987. Imaging in high-aperture optical systems. J. Opt. Soc. Am. A 4:1354-1360. Sheppard, C.J.R. and Wilson, T. 1978. Depth of field in scanning microscope. Optics Lett. 3:115-117. Stelzer, E.H., Wacker, I., and De Mey, J.R. 1991. Confocal fluorescence microscopy in modern cell biology. Semin. Cell Biol. 2:145-152. Tanaami, T., Otsuki, S., Tomosada, N., Kosugi, Y., Shimizu, M., and Ishida, H. 2002. Highspeed 1-frame/ms scanning confocal microscope with a microlens and Nipkow disks. Appl. Opt. 41:4704-4708. Tropper, A.C., Foreman, H.D., Garnache, A., Wilcox, K.G., and Hoogland, S.H. 2004. Vertical-external-cavity semiconductor lasers [Review]. J. Phys. D 37:R75-R85. Wouters, F.S., Verveer, P.J., and Bastiaens, P.I.H. 2001. Imaging biochemistry inside cells. Trends Cell Biol. 11:203-211. Zemanova, L., Schenk, A., Valler, M.J., Nienhaus, G.U., and Heilker, R. 2003. Confocal optics microscopy for biochemical and cellular high-throughput screening. Drug Discov. Today 8:1085-1093.
Contributed by Bartek Rajwa Purdue University Cytometry Laboratories West Lafayette, Indiana
Modern Confocal Microscopy
12.3.12 Supplement 31
Current Protocols in Cytometry
Time-Lapse Microscopy Approaches to Track Cell Cycle Progression at the Single-Cell Level
UNIT 12.4
The application of time-lapse microscopy or video microscopy (Weiss et al., 1989) underlies the fundamental basis of cell-based assays where time is the quality parameter. The time-lapse microscopy concept can be described as the repeated collection of a field of view from a microscope (x,y) at discrete time intervals (t) (Allen, 1985). The information contained in the field of view is determined by the contrast mode, which includes phase, differential interference contrast (Allen et al., 1981; Brugmans et al., 1982), dark field, and fluorescence (see review in White and Errington, 2004). Transmission phase offers a probeless and relatively non-perturbing contrast mode, which provides information on cell morphology (essentially cell shape and cell position); the changes in these two basic features facilitate assays describing dynamic cell behavior interacting on a 2-D substrate. The duration of the time interval defines the temporal resolution, which in turn characterizes the type of event detected. At the core of the assays is the implementation of time-lapse microscopy to link the initial cell cycle position during acute exposures to anti-cancer agents with the anti-proliferative consequences for individual cells. The focus of this unit is to describe a systematic protocol for the acquisition and analysis of time-lapse sequences for the purposes of understanding single cell cycle dynamics (Dover and Potten, 1988; Marquez et al., 2003). The approach incorporates fundamental concepts arising from the ability to capture simple video sequences of cells over 48 to 72 hr, from which it is possible to extract kinetic descriptors that reflect the interplay of mitosis and cell death in the growth of an unsynchronized tumor population (Allman et al., 2003; Chu et al., 2004). The benefits of such an approach are also described in a screening context by employing a multi-well format which enables the user to test different derivatives of drugs, multiple dose ranges (Feeney et al., 2003), or cell cultures with unique genetic backgrounds (Marquez et al., 2004). The fact that single cells are monitored in situ means that proliferation assays can be multiplexed with cell motility measurements, as required for wound-healing models (Stephens et al., 2004). This unit aims to present a generic methodology for capturing time-lapse sequences and subsequently mining the data for comprehensive event analysis. The time-lapse approach enables users to determine the kinetics of single-cell cycle traverse, delay, arrest, and checkpoint breaching in response to drug perturbations and can be considered as the basis for time-encoded cell-based assays.
System Setup In essence, time-lapse microscopy is very simple to implement on any research microscope. Many manufacturers are able to deliver an integrated or complete system (see Internet Resources); however, it is recommended to understand the role of the principal components that make up a time-lapse instrument, ensuring that the instrument is capable of delivering the experimental design. The instrument description given below is for very specific assays that consist of single-channel phase transmission with no fluorescent probes. The time-lapse station should consist of an inverted microscope fitted with an incubator system for 37◦ C, 5% CO2 maintenance and a CCD camera, with the entire system sitting on a vibration-free table. Illumination must be controlled by a shutter in front of
Contributed by Rachel J. Errington, Nuria Marquez, Sally C. Chappell, Marie Wiltshire, and Paul J. Smith Current Protocols in Cytometry (2005) 12.4.1-12.4.11 C 2005 by John Wiley & Sons, Inc. Copyright
Cellular and Molecular Imaging
12.4.1 Supplement 31
the transmission. The system should have an x, y positioning stage capable of holding multi-well dishes with separate z-focus to control multi-field acquisition. Image-capture software must be able to control these peripherals in a flexible way so that users can compile a bespoke acquisition regimen. A 10× (PH1) apochromat objective lens for image collection gives a field size of 500 × 500 µm (at a resolution of 1 µm per pixel), with ∼40 to 70 cells per field. For discussion of optimal instrument settings, see Critical Parameters. BASIC PROTOCOL 1
TIME-LAPSE ACQUISITION USING ADHERENT CELLS Many of the cell lines used for cell cycle analysis are adherent populations gowing on a two-dimensional substrate and many are very compatible with tissue culture plastic. Time-lapse microscopy is an approach allowing single-cell tracking without perturbing the adherent system. If cells are growing on the tissue culture plastic, it is easy to locate the cell relative to the dish surface. The procedure can be adapted for a multiwell dish. This protocol is for a multi-well plate preparation and the time-lapse setup is for a 24 to 48 hr sequence.
Materials Adherent mammalian cells of interest or cultures Cell-specific tissue culture medium Drug of choice Humidified 5% CO2 supply 6-well tissue culture dish Humidified 37◦ C, 5% CO2 incubator Time-lapse microscope Prepare cells 1. Seed attached mammalian cells in a 6-well tissue culture dish at a density of ∼1 × 104 cells/ml in cell cuture medium. Incubate 24 hr in a humidified 37◦ C, 5% CO2 incubator. 2. After 24 hr, treat the cells with the appropriate dose of drug. For a pulse treatment, wash at least three times, each time by aspirating the medium from each well, replacing it with 37◦ C medium, and incubating 2 min at room temperature. For a continuous treatment, ensure that sufficient medium is originally placed in each well and that the concentration of drug is correct. A final volume of 4 ml per well is needed.
3. Mark the top left-hand well with a cross scratch; this will become the zero position. If the plate is removed from the holder and then replaced, this position is easily found.
Prepare instrument 4. Place the plate on the automated stage. Attach the CO2 supply (see Troubleshooting for hints and tips). 5. Set up with 10× phase transmission and set lamp to minimal setting.
Time-Lapse Microscopy to Track Cell Cycle Progression
6. Check that the image-acquisition software can communicate with the microscope and the camera. Make sure that the system can capture a digital image (exposure should be ∼10 to 30 msec).
12.4.2 Supplement 31
Current Protocols in Cytometry
Collect images 7. Starting with the first well, move to position x1 ,y1 . Open the shutter and capture the image. Move to position x2 ,y2 and repeat. Continue in this fashion until all wells have been imaged. Usually each field should contain 40 to 70 cells.
8. Wait 5 to 30 min and sample the entire plate again. Acquire images for 24 to 48 hr. Sequences are typically captured in triplicate for each treatment regime at 10-min intervals for a period of 24 to 48 hr.
9. Save image sequence onto a CD or DVD and make a backup master copy. For discussion of optimal experimental acquisition, see Critical Parameters.
TIME-LAPSE ACQUISITION WITH ENDPOINT ASSAY TO MARK S-PHASE CELLS
ALTERNATE PROTOCOL 1
Incubation of cells containing halogenated DNA precursors such as 5-bromodeoxyuridine (BrdU) leads to the incorporation of thymidine analogs into strands of DNA during replication (S-phase). The incorporation can be detected immunocytochemically (UNIT 7.7). Critical to the experimental design for time-lapse imaging is determining the time window of exposure to the analogs. For example, to identify the cells in S-phase during drug exposure, it is recommended to pulse-label prior to drug treatment. Treat the cultures with drug, remove excess BrdU and drug, and place on the time-lapse instrument to acquire the time series. The BrdU label will be carried through the cell lineages and therefore the original S-phase cells can be retro-identified after the 48-hr time lapse using immunocytochemistry methods (Feeney et al., 2003). Once the S-phase cells are identified, G2 and G1 cells are classified by virtue of their mitotic timing, before or after the S-phase division, respectively. BrdU solutions are light sensitive and together with illumination light can damage DNA. Therefore, it is not recommended to label the cells for extended periods during time-lapse collection.
Additional Materials (also see Basic Protocol 1) 10 mg/ml BrdU stock in distilled water (32 mM) 70% ethanol Phosphate-buffered saline (PBS; APPENDIX 2A) 2 M HCl PBS/0.5% Tween 0.1 M sodium tetraborate Phosphate-buffered saline with 0.6% bovine serum albumin (PBS/BSA) Primary anti-BrdU antibody Secondary peroxidase anti-mouse antibody 10 × diaminobenzidene (DAB) stock solution (sigma, cat. no. 5637) 30% hydrogen peroxide (H2 O2 ) 1. Pulse-label cells with 40 µM bromodeoxyuridine (BrdU) 15 min at 37◦ C. Immediately prior to the drug treatment wash at least three times, each time by aspirating the medium from each well, replacing it with 37◦ C medium, and incubating 2 min at room temperature. 2. Add treatment to the appropriate wells and acquire time-lapse sequence following Basic Protocol 1, steps 2 through 9. Record and save the field coordinates. Remove the multi-well plate and process for BrdU detection.
Cellular and Molecular Imaging
12.4.3 Current Protocols in Cytometry
Supplement 31
Fix and process to detect BrdU label (Bond et al., 1996) 3. Fix cells in 70% ethanol 30 min at 4◦ C. Wash two times using the technique described in step 1, except substitute room temperature PBS for the medium. 4. Denature in 2 M HCl 5 min at 37◦ C. Wash once using the technique described in step 1, except substitute room temperature PBS/0.5% Tween for the medium. 5. Neutralize in 0.1 M sodium tetraborate 5 min at room temperature. Wash two times using the technique described in step 1, except substitute room temperature PBS for the medium. 6. Dilute primary anti-BrdU antibody 1:200 in PBS/BSA, add to cells (enough to cover bottom of well, 0.5 to 1 ml), and incubate 1 hr at room temperature. Wash two times using the technique described in step 1, except substitute room temperature PBS for the medium. 7. Dilute secondary peroxidase anti-mouse antibody 1:100 in PBS/BSA, add to cells (enough to cover bottom of well, 0.5 to 1 ml), and incubate 1 hr at room temperature. Wash two times using the technique described in step 1, except substitute room temperature PBS for the medium. 8. Add 1 ml of substrate per well: DAB stock diluted 1:10 in PBS along with a 1:3 dilution of 30% H2 O2 . Incubate 5 min at room temperature and wash using the technique described in step 1, except substitute room temperature PBS for the medium. 9. Place the plate onto the microscope stage and recall coordinate settings from timelapse recording. 10. Record bright-field images of the BrdU-labeled cells and mark the BrdU-labeled cells using region-of-interest (ROI) overlays. 11. Run the time-lapse sequence backwards to identify the original S-phase cells. 12. Monitor the behavior of the S-phase cells by time to event (see Data Mining). 13. Classify the G1 and G2 cells by recording the mitotic event relative to the S-phase cells. ALTERNATE PROTOCOL 2
TIME-LAPSE ACQUISITION USING SUSPENSION CELLS Suspension cells provide a unique logistical problem as they move around in the medium. As a result, tracking single-cell lineages is difficult under the basic conditions. Placing a thin layer of agarose onto the suspension cells immobilizes the cells in one place but at the same time does not compromise their growth characteristics. In addition, by placing the exipients or drugs directly in the gel phase, it is possible to characterize drug effects in a multi-well format.
Additional Materials (also see Basic Protocol 1) 10% RPMI SeaPlaque agarose (FMC BioProducts, Flowgen Instruments) Microwave oven 37◦ C water bath 5-ml stripette Time-Lapse Microscopy to Track Cell Cycle Progression
1. Spin down suspension cell line and resuspend in pre-warmed medium (10% RPMI). 2. Seed the cell sample at a concentration of 2.5 × 105 cells/ml. Add 0.5 ml to each well of a 12-well plate and incubate 30 min in a humidified 37◦ C, 5% CO2 incubator.
12.4.4 Supplement 31
Current Protocols in Cytometry
3. Prepare agarose holding gel by dissolving 0.08 g agarose in 5 ml of 10% RPMI. Heat a maximum of 10 sec in a microwave oven at low power. Add 5 ml of 10% RPMI at 37◦ C and place agarose preparation in a 37◦ C water bath to cool slightly (do not allow to solidify). Add appropriate drug concentrations to the agarose solution as required. 4. Carefully overlay 1.5 ml agarose preparation in each well using a 5-ml stripette. 5. Fill inter-well space with water- or saline-soaked tissue (Kimwipes) to maintain the dish humidity. 6. Place the dish onto the holder of the microscope station. 7. Capture time-lapse images up to 48 hr at 5-min intervals.
SEQUENCE ANALYSIS FOR MITOSIS EVENT OR CELL DEATH The time-lapse sequence is very rich in information. It contains details of cellular behavior, which includes cell spreading, cell motility, cell division, cell death, and cell-cell interactions. All these activities can be quantified providing they have been sampled at the Nyquist frequency (see Critical Parameters). This protocol focuses on behavior that describes the cell cycle. The basic requirement is to record the time-dependent mitotic events within a field of cells. It is very difficult to segment a cell based only on the transmission image; therefore, this protocol is a manual one, although, depending on the software program, some elements of automation can be implemented. The raw method is described for mitosis so that users can understand the fundamental process.
BASIC PROTOCOL 2
For cell-death event analysis, record time to cell death. Cell death has many guises: the cells can undergo lysis, or shrink, bleb, and disintegrate. The latter description would probably be apoptosis, which would have to be confirmed using specific assays.
Materials Cells to be viewed Video playback software 1. Find some suitable video playback software. 2. Mark each cell at time point zero using an ROI in the field and give it a reference name: e.g., cell1, cell2, cell3, and so on. The total number of cells provides the field with the mitotic potential, usually 25 to 60 cells per field.
3. Follow cell1 until it reaches mitosis. The cell morphology changes from flat to round, which is highly refractive. This is called the first event for cell1 and has some associated parameters.
4. Record the time to mitosis for cell1. Record the time to resolve the mitosis for cell1. Record the quality of the resolution for cell1. In normal cells, a successful mitosis leads to two daughters. However, in perturbed conditions sometimes the chromosomes do not segregate properly. Then the cell becomes polyploid, or in some circumstances, spindle perturbation leads to three daughters.
5. Extend the nomenclature for the resultant cells—for example: a. Mitotic outcome: cell1.1 and cell1.2 b. Polyploid outcome: cell1.1 c. Three-daughter outcome: cell1.1, cell1.2, and cell1.3
Cellular and Molecular Imaging
12.4.5 Current Protocols in Cytometry
Supplement 31
6. Track the complete lineage for cell1. 7. Repeat the analysis for all the cells in the field. 8. Place all the information extracted from the sequence into a spreadsheet for further manipulation and data mining. BASIC PROTOCOL 3
DATA MINING—NORMALIZED EVENT DISTRIBUTION There are very simple ways of summarizing the data acquired from the analysis phase, which indicate the behavior of cells in each treatment condition obtained from a 24- or 48-hr sequence. The first method provides a general view of event distribution (Marquez et al., 2003, 2004). This approach allows for screening and comparisons from well to well and provides a good overview of drug effects on cells (see Fig. 12.4.1). Obtain the total number of events throughout the sequence for each treatment condition. Then break down to sub-categories by counting the number of events associated with each event type, e.g., daughters, polyploidy, and cell death. Normalize the total events to 100 and convert the sub-categories by the same factor. These will then represent a percentage fraction of the normalized total events.
BASIC PROTOCOL 4
DATA MINING—TIME-TO-EVENT CURVES Time-to-event curves represent the dynamics of cumulative events occurring in a treatment regime. They are usually processed and presented in plots so that quantitative comparisons from well to well are possible (Marquez et al., 2003). Present data in a spread sheet in the form of cumulative event data versus time (hr). The plot of cumulative events over time (Fig. 12.4.2A) provides a basic graphical display. Normalizing this against the number of cells (or the potential for events) in the original field derives a cumulative event potential parameter, enabling quantitative comparisons of drug effects (Fig. 12.4.2B).
Time-Lapse Microscopy to Track Cell Cycle Progression
Figure 12.4.1 Typical event distribution cumulated over 40 hr for different drug concentrations. Under control conditions, all the events during the course of the sequence are successful cell divisions leading to two daughters. At the low dose, the number of daughter events decreases and mitosis leads to polyploid cells. At the high dose, there are no successful cell divisions and cell death (apoptosis) becomes the predominant event.
12.4.6 Supplement 31
Current Protocols in Cytometry
Figure 12.4.2 Time-to-event curves to demonstrate the dynamics of event delivery in treated and untreated conditions. (A) Plot of all cumulated events for control, low dose– and high dose–treated cultures. (B) After correction for the number of cells (or the potential for events) in the original field, the low-dose plot alters respective to the control conditions, because the low-dose conditions experienced fewer events and contained fewer cells in the original field. Therefore, by normalizing for the number of cells in the field, a true effect of the drug is depicted. (C) By normalizing against the control for each experimental run, experiments from different culture days can be pooled together, since this represents the perturbation effect relative to the control conditions. (D through F) Time-to-event curves for control and drug-treated conditions. The dotted line represents the overall event curve. However, each condition has a very different event type contribution. (D) The control conditions consist primarily of daughter events with an insignificant background of cell death. (E) Low-dose treatment does not demonstrate a delay, but the mitotic event results in both daughter and polyploid outcomes. (F) At the high dose, there is a significant event delay and cell death becomes the predominant event. The kinetics of time-dependent and dose-dependent drug effect has been extracted from simple video sequences.
Finally, normalizing the cumulative event potential of the controls to represent 100% at the end time point enables the pooling of independently derived experiments (Fig. 12.4.2C), a normalized cumulative event potential. This procedure can be repeated for the individual sub-categories of events such as time to mitosis, time to cell death, and time to polyploidy (Fig. 12.4.2D through F).
DATA MINING—DURATION OF MITOTIC EVENT The duration of the mitotic event gives invaluable information on the perturbation of molecular events involved in the mitotic assembly checkpoint (Marquez et al., 2003). Subtract time to mitosis from time to resolve mitosis. This provides a parameter describing mitotic duration. The plot of mitotic duration versus time to mitotic event provides a basic graphical display that indicates whether drug exposure perturbs mitotic traverse (Fig. 12.4.3).
BASIC PROTOCOL 5
Cellular and Molecular Imaging
12.4.7 Current Protocols in Cytometry
Supplement 31
Figure 12.4.3 The duration of each mitotic event can be derived during the course of the sequence by subtracting the time to mitosis from the time to resolve mitosis. The plot demonstrates the effect of a microtubule-disrupting agent (colcemid; closed circles), exhibiting a mean mitotic duration of 5.1 ± 1.1 hr. It also demonstrates that this drug has an immediate effect that is sustained throughout the time course. Control or untreated cultures exhibit a mean duration of 1.2 ± 0.57 hr.
Figure 12.4.4 (A) Marking of S-phase cells using BrdU to classify the cell cycle position of all cells in the field at the start of the time sequence. By pulsing the culture with 5-bromodeoxyuridine (BrdU) according to Alternate Protocol 1, users can identify S-phase cells and the subsequent time to event for these cells. In this cell type (human breast tumor MCF-7 cells), 95% of the marked cells deliver between 5 and 16 hr. Therefore, the cells that deliver between 0 and 5 hr can be classified as G2 cells and those that deliver immediately after the S-phase zone are designated G1 cells. (B) Determining the extent of G2 checkpoint breaching. A hypothetical plot shows that by plotting the normalized event potential for cell type a in untreated (solid line) and treated (dotted line) conditions, it can be shown that the treated conditions lead to an abrogated mitotic delivery originating in G2, a G2 checkpoint. However, type b cells breach this checkpoint because they exhibit a normal response in both untreated (solid line) and treated conditions (dotted line).
BASIC PROTOCOL 6
Time-Lapse Microscopy to Track Cell Cycle Progression
DATA MINING—G2 CHECKPOINT BREACHING Late cell cycle (G2) effects can be dissected in detail from the derived event analysis data. Alternate Protocol 1 provides critical information that allows users to identify G2 cells within the original population, cells that deliver to mitosis before the S-phase-marked cells (Fig. 12.4.4A). The first cohort of cells (4 to 8 hr) to deliver to mitosis will be in G2 during the drug treatment (Marquez et al., 2003).
12.4.8 Supplement 31
Current Protocols in Cytometry
Present data in the form of cumulative event data versus time (hr) in a spread sheet for G2 cells only. Present data to compare the G2 response in treated and untreated conditions (Fig. 12.4.4B) to determine if the drug used induces a G2 checkpoint in cell type A. Present data to detemine the G2 response in cell type B, which could be a genetic variant of cell type A. For discussion of data statistics, see Critical Parameters.
COMMENTARY Background Information Time-lapse microscopy is a simple tool that has been used to study basic mechanisms of cell motility, wound healing (Weiss et al., 1989), and, recently, aspects of cell cycle regulation (Marquez et al, 2003). The need to understand cellular architecture and dynamics during cell migration and as a result of interacting with a two-dimensional surface has really increased the application of time-lapse microscopy (Allen, 1985). However the basic protocols outlined in this unit allow the well established methods to be adapted for understanding cell cycle regulation for a single cell (Feeney et al., 2003), which, compared to flow cytometry, is limited by the total number of cells, but provides unique data as it enables the tracking of the cell through the cell cycle journey, potentially leading to extensive lineages.
The shutter that is located between the transmission lamp and the condenser should have a minimum aperture of 25 mm, and temperature heat deflectors. However, the opening time is not a critical parameter, and 100 to 400 msec is sufficient. The speed of a motorized stage is defined by its acceleration and the sequential movement. The resolution of the x,y positioning stage does not have to be of the highest standard; ±1 µm is sufficient. The software controlling the acquisition instrument has to be reliable and communication between the PC and peripherals robust to collect multi-well or field data. The handling of the sequence data is also crucial. Typical data sets can reach sizes of 2 to 3 GB; hence, the data can become cumbersome. It is wise for users to determine how specific packages handle the multi-well sequences and in which file format the data are acquired.
Critical Parameters Optimal instrument settings The microscope incubator system is a critical component of the time-lapse instrument. Users are advised to choose this with care as it will ultimately determine the stability and robustness of the time-lapse system. It is recommended that the incubator system encompass a large volume of the microscope, the stage area in particular as well as the underside of the stage, including the nose pieces. Otherwise, airflow around the instrument will result in thermal fluxes and continual focus drift. CO2 regulation using technology analogous to that of tissue-culture incubators is available, but expensive, and has limited success. A more pragmatic approach using regulated delivery of 5% CO2 directly to the multi-well dish is both reliable and cost effective once it has been calibrated across the whole dish. The CCD camera used for this type of time-lapse system can be considered at a low specification. Monochrome (8-bit) and color CCD cameras are sufficient since sensitivity required for fluorescence detection is not required.
Optimal experimental acquisition The danger with time-lapse microscopy is to always oversample. This means to collect data that have an optical resolution that is too high and a sequence that consists of images collected at a high interval rate. This leads to sequences which are large and cumbersome to manipulate and to archive. The experimental aim is to be as non-invasive as possible; therefore, the total exposure to light must be minimized to ensure normal culture growth. The assays described in this unit can be achieved using a 10× phase objective lens, although some cell types (e.g., lymphomas) require a slightly higher magnification (20×). Eight-bit rather than 12-bit digital frames are more than adequate for transmission imaging. The interval time depends on the type of cells being used in the assay, but constraints will be dictated by two key parameters: the mitotic duration and the cell motility. Nyquist sampling (Pawley, 1995) shows that to guarantee the ability to detect every mitotic event in a population of unsynchronized cells, sampling must be performed at a rate of half the mitotic
Cellular and Molecular Imaging
12.4.9 Current Protocols in Cytometry
Supplement 31
duration (which for most cell types is in a range of 30 to 80 min). To track a cell from frame to frame, the cell motility must also be taken into account. Astrocytes, fibroblasts, and neutrophils exhibit a range of speeds from 0.1 to 1 to 10 µm/min, respectively, requiring the users to sample adequately for single-cell tracking. Therefore, for primary fibroblasts (with a 40min mitotic duration) a 10-min time interval would be adequate.
Time-Lapse Microscopy to Track Cell Cycle Progression
encompassing a large volume around the microscope always works well. Placing the instrument under an air-conditioning unit or by the laboratory door is not conducive to a stable system.
Anticipated Results
Data statistics Cox regression and log-rank methods can be used to analyze the time to a given end point. The use of Cox regression allows the effect of the variation between cell lines to be included in the model and, in general, a stratified analysis can be performed to test the effect of a drug within a given cell line. Cells that move out of the field during the period of observation, and those that do not divide or die, are considered to be censored at a given time point (Therneau and Grambsch, 2000).
The approach to time-lapse microscopy presented in this unit has three distinct phases: (1) The acquisition phase, in which multidimensional images consisting of digital frames (x,y) are obtained in time (t), for multiple fields per treatment regime. (2) The image-analysis phase, which involves the extraction and time-stamping of cellular events such as mitosis, cell death, and cell fusion from the original image data. (3) The data-mining phase, which involves the systematic presentation and visualization of normalized event curves based on robust and hypothesis-driven data manipulation. Typcial results are shown in Figures 12.4.1 to 12.4.4.
Troubleshooting
Time Considerations
Maintaining adequate pH and humidity across the entire multi-well plate is a matter of trial and error. A CO2 -enriched environment extends the life of the growth medium during prolonged experiments. In some systems, the supply of gas can be via a regulated supply to the whole microscope incubator; these incubators are expensive and sometimes inappropriate for certain dishes. Customizing a supply directly to the dish provides a flexible and adaptable system for environment regulation. For example, providing an attachment with a precision needle–valve controller for the flow of CO2 or CO2 mixture through a water reservoir results in a humidified gas supply and provides a simple visual indication of flow. Calibrating all the typical multi-well plates to be used in a laboratory setup and introducing some element of standardization is worthwhile. The total volume of medium per well and across the plate will change the environment dramatically, and each multi-well plate has distinct configurations. Focus drift is a common problem during long-term imaging. Assuming that the plate has been adequately clamped into the plate holder, focus drift has two main causes. The first is inadequate vibration isolation. It is recommended that long-term time-lapse instruments be placed on vibration isolation tables. The second is thermal fluctuations and airflow across the microscope. Hence, an incubator
Each protocol has a very similar time requirement. The acquisition may take 24 to 48 hr depending on the length of the sequence required. Preparing the cells for the experiment within a multiwell format has similar time constraints as for normal tissue culture protocols. However all plates must be prepared 24 hr in advance to allow the cultures reach normal growth kinetics before manipulation. Image analysis, however, is the rate-limiting step for information extraction. Some of the analysis can be automated; however, much of it is done by observation and manual input, thereby establishing a bottleneck of data flow. For every multiwell experiment performed, at least 2 to 3 weeks are required to extract the information before embarking on the data-mining phase.
Literature Cited Allen, R.D. 1985. New observation on cell architecture and dynamics by video-enhanced contrast optical microscopy. Annu. Rev. Biophys. Chem. 14:265-290. Allen, R.D., Allen, N.S., and Travis, J.L. 1981. Video-enhanced contrast, differential interference contrast (AVEC-DIC) microscopy: A new method capable of analyzing microtubulerelated motility in the reticulopodial network of Allogromia laticollaris. Cell Motil. 1:291-302. Allman, R., Errington, R.J., and Smith, P.J. 2003. Delayed expression of apoptosis in human lymphoma cells undergoing low-dose taxol-induced mitotic stress. Br. J. Cancer 88:1649-1658.
12.4.10 Supplement 31
Current Protocols in Cytometry
Brugmans, N., Cassiman, J.J., Van der Heydt, L., Oosterlinck, A.J.J., Vlietinck, R., and Vanden Berghe, H. 1982. Quantification of the degree of cell spreading of human fibroblasts by semi-automated analyses of the cell perimeter. Cytometry 3:262-268. Bond, J., Haughton, M., Blaydes, J., Gire, V., Wynford-Thomas, D., and Wyllie, F. 1996. Evidence that transcriptional activation by p53 plays a direct role in the induction of cellular senescence. Oncogene 13:2097-2104. Chu, K., Teele, N., Dewey, M.W., Albright, N., and Dewey, W.C. 2004. Computerized video time lapse study of cell cycle delay and arrest, mitotic catastrophe, apoptosis and clonogenic survival in irradiated 14-3-3sigma and CDKN1A (p21) knockout cell lines. Radiat. Res. 162:270-286. Dover, R. and Potten, C.S. 1988. Heterogeneity and cell cycle analyses from time-lapse studies of human keratinocytes in vitro. J. Cell Sci. 89:359364. Feeney, G.P., Errington, R.J., Wiltshire, M., Marquez, N., Chappell, S.C., and Smith, P.J. 2003. Tracking the cell cycle origins for escape from topotecan action by breast cancer cells. Br. J. Cancer 88:1310-1317. Marquez, N., Chappell, S.C., Sansom, O.J., Clarke, A.R., Court, J., Errington, R.J., and Smith, P.J. 2003. Single cell tracking reveals that Msh2 is a key component of an early-acting DNA damageactivated G2 checkpoint. Oncogene 22:76427648. Marquez, N., Chappell, S.C., Sansom, O.J., Clarke, A.R., Teesdale-Spittle, P., Errington, R.J., and Smith, P.J. 2004. Microtubule stress modifies intra-nuclear location of Msh2 in mouse embryonic fibroblasts. Cell Cycle 3:662-671. Pawley, J.B. (ed.) 1995. Handbook of Biological Confocal Microscopy. Plenum, New York.
Stephens, P., Grenard, P., Aeschlimann, P., Langley, M., Blain, E., Errington, R., Kipling, D., Thomas, D., and Aeschlimann, D. 2004. Crosslinking and G-protein functions of transglutaminase 2 contribute differentially to fibroblast wound healing responses. J. Cell Sci. 117:3389-3403. Therneau, T.M. and Grambsch, P.M. 2000. Modeling Survival Data. Extending the Cox Model. Springer-Verlag, New York. Weiss, D.G., Maile, W., and Wick, R.A. 1989. Video microscopy. In Light Microscopy in Biology. A Practical Approach. (A.J. Lacey, ed.) pp. 221278. IRL Press, London. White, N.S. and Errington, R.J. 2004. Fluorescence techniques for drug delivery research: Theory and practice. Adv. Drug Deliv. Rev. In press.
Internet Resources http://www.kineticimaging.com/ http://www.image1.com/ http://www.solentsci.com/ http://www.zeiss.com/ http://www.nikon.com/ http://www.olympus-global.com/en/global/ Web sites for Kinetic Imaging, Universal Imaging Corporation, Solent Scientific, Carl Zeiss, Nikon, and Olympus, manufacturers of integrated or complete systems for time-lapse microscopy.
Contributed by Rachel J. Errington, Nuria Marquez, Sally C. Chappell, Marie Wiltshire, and Paul J. Smith School of Medicine Heath Park, Cardiff, United Kingdom
Cellular and Molecular Imaging
12.4.11 Current Protocols in Cytometry
Supplement 31
Three-Dimensional Visualization of Blood and Lymphatic Vasculature in Tissue Whole Mounts Using Confocal Microscopy
UNIT 12.5
This unit presents protocols (optimized for murine ear skin) for labeling blood vasculature with fluorescein isothiocyanate (FITC)–conjugated Lycopersicon esculentum (tomato) lectin, as well as protocols for detecting vascular leakage sites, lymphatic vasculature, and α-smooth muscle actin–positive cells lining vascular walls. The Basic Protocol is based upon the classic protocol published by McDonald and colleagues (Thurston et al., 1996), which includes pretreatment of blood vasculature with fixative and bovine serum albumin (BSA) prior to intravenous perfusion of FITC-conjugated L. esculentum lectin. In Alternate Protocol 1, FITC-conjugated L. esculentum lectin is injected intravenously and allowed to circulate under “normal” flow and hemodynamic pressure prior to perfusion with fixative. Either protocol can be combined with rhodamine-conjugated Ricinus communis (castor bean) agglutinin I labeling, as described in Alternate Protocol 2, to reveal sites of subendothelial basement membrane exposure to luminal surfaces, a finding that is characteristic of reactive vasculature. Alternate Protocol 3 describes how to visualize lymphatic vasculature with LYVE-1 antiserum while simultaneously visualizing blood vasculature. Visualization of perivascular support cells along with the blood vasculature is described in Alternate Protocol 4. NOTE: All protocols using live animals must first be reviewed and approved by an institutional animal care and use committee (IACUC) and must follow officially approved procedures for the care and use of laboratory animals.
DIRECT DETECTION OF BLOOD VASCULATURE IN TISSUE WITH PREFIXATION OF VASCULATURE
BASIC PROTOCOL
This protocol is based on the classic protocol described by McDonald and colleagues (Thurston et al., 1996), which includes pretreatment of blood vasculature with fixative and bovine serum albumin (BSA) prior to intravenous (i.v.) perfusion of Lycopersicon esculentum lectin.
Materials Mouse from which tissue is to be harvested Anesthetic for inhalation: 2% (v/v) isoflurane in oxygen 1% (w/v) paraformaldehyde/0.5% (v/v) glutaraldehyde fixative solution (see recipe) PBS (APPENDIX 2A) PBS (APPENDIX 2A)/1.0% (w/v) BSA, filtered (0.44-µm filter) Fluorescein isothiocyanate (FITC)–conjugated Lycopersicon esculentum (tomato) lectin (Vector Laboratories; 2.0 mg/ml stock) 1% (w/v) paraformaldehyde fixative solution (see recipe) Antifade mounting medium (e.g., Molecular Probes ProLong Gold antifade reagent) Nitrocellulose-based lacquer (nail polish) or rubber cement Wide-bore blunt needle (cannula) or blunted 16-G needle connected to tubing Hand-held pump with attached manometer Dissecting equipment, including dissecting microscope 24-well tissue culture plates. Glass microscope slides and coverslips Contributed by Alexandra Eichten, H.-C. Jennifer Shen, and Lisa M. Coussens Current Protocols in Cytometry (2005) 12.5.1-12.5.11 C 2005 by John Wiley & Sons, Inc. Copyright
Cellular and Molecular Imaging
12.5.1 Supplement 32
Laser scanning confocal microscope (Zeiss LSM 510 or a similar confocal system) 500-to-550-nm band-pass filter Zeiss LSM Image Examiner or similar three-dimensional analysis software Additional reagents and equipment for administering inhalation anesthesia (Donovan and Brown, 1998) Perfuse mouse with fixative and FITC-labeled lectin 1. Anesthetize the mouse using inhaled 2% isoflurane in oxygen (Donovan and Brown, 1998). 2. Open the mouse’s chest, use scissors to cut the right atrium of the heart, and use a wide-bore blunt needle (cannula) or a blunted 16-G needle connected to tubing to cannulate the left ventricle. Cutting of the right atrium creates an opening through which perfusates and blood can be released after circulation through the systemic and pulmonary circuits.
3. Fix the vasculature by perfusing mouse 3 min with 1% paraformaldehyde/0.5% glutaraldehyde fixative solution (under a constant pressure of 120 mm Hg, applied using a hand-held pump with an attached manometer) through the left ventricle (and thus through the aorta). 4. Perfuse the mouse 1 min with PBS under a constant pressure of 120 mm Hg. 5. Perfuse the mouse with 25 ml PBS/1% BSA under a constant pressure of 120 mm Hg. 6. Perfuse the mouse with 25 ml PBS/1% BSA containing 2.0 µg/ml FITC-conjugated L. esculentum lectin under a constant pressure of 120 mm Hg. 7. Perfuse the mouse with 25 ml PBS/1% BSA under a constant pressure of 120 mm Hg. 8. Perfuse the mouse 1 min with PBS under a constant pressure of 120 mm Hg.
Harvest and mount tissue 9. Harvest the ear tissue (or tissue from some other site) by dissection. Transfer the harvested tissue to a 24-well tissue culture plate, add 1 to 2 ml of 1% paraformaldehyde fixative solution, and then incubate 4 hr (or overnight) at 4◦ C, protected from light and with constant rotation. 10. Remove the paraformaldehyde solution from the tissue and then wash the tissue three times, each time by adding 2.0 ml PBS; incubating 3 min at 4◦ C, protected from light; and then removing the PBS. 11. If using ear tissue, separate the dorsal and ventral aspects of the ears and remove all cartilage under a dissecting microscope. Discard all dorsal-side ear tissue. The dorsal side of the ear is not used because it contains muscle tissue, which complicates the imaging process.
12. Carefully mount the ventral aspect of the ear in antifade mounting medium on a glass slide, with the dermal surface of the tissue facing up. Add a coverslip and seal all edges with nail polish or rubber cement. 13. Store the mounted tissue at 4◦ C, protected from light. 3-D Visualization of Blood and Lymphatic Vasculature in Tissue
For optimal results, tissue should be imaged as soon as possible after mounting.
12.5.2 Supplement 32
Current Protocols in Cytometry
Collect and analyze confocal data 14. Perform confocal microscopy (UNIT 2.8) on the mounted tissue using a Zeiss LSM 510 META microscope (or a similar confocal system) operating in channel mode with 488-nm excitation, and collect the resulting FITC fluorescence emission (green fluorescence) using a 500-to-550-nm band-pass filter. Acquire a z-stack at a resolution of 1 Airy unit to allow three-dimensional reconstruction. 15. Digitally reconstruct the imaged tissue section and analyze the reconstructed image using the Zeiss LSM Image Examiner program or similar three-dimensional analysis software.
DIRECT DETECTION OF BLOOD VASCULATURE IN TISSUE WITHOUT PREFIXATION OF VASCULATURE
ALTERNATE PROTOCOL 1
This protocol is a modified version of the Basic Protocol and omits pretreatment of the vasculature with fixative prior to perfusion with FITC-conjugated Lycopersicon esculentum lectin. Instead, the fluorescently labeled lectin is injected intravenously into the animal before perfusion with paraformaldehyde. In addition, perfusion with paraformaldehyde is performed at a constant flow rate, rather than at a constant pressure.
Additional Materials (also see Basic Protocol) 4% (w/v) paraformaldehyde fixative solution (see recipe) 0.5-ml insulin syringe or 1.0-ml syringe equipped with a 29-G needle 60-ml syringe equipped with a blunted 16-G needle Infusion pump for cardiac perfusion (KDS100; KD Scientific) 1. Anesthetize the mouse using inhaled 2% isoflurane in oxygen (Donovan and Brown, 1998). 2. Using a 0.5-ml insulin syringe (or, alternatively, a 1-ml syringe) equipped with a 29-G needle, inject 100 µl of 2.0 mg/ml FITC-conjugated L. esculentum lectin solution into the mouse’s tail vein over the course of 20 to 30 sec. A 0.5-ml insulin syringe is recommended for this step, as this will minimize the void volume.
3. Allow the injected FITC-conjugated lectin to circulate for 3 min. 4. Open the mouse’s chest, cut the right atrium of the heart, and use a 60-ml syringe equipped with a blunted 16-G needle to cannulate the left ventricle. 5. Using a KDS100 infusion pump, perfuse mouse with 35 ml of 4% paraformaldehyde fixative solution through the left ventricle (and thus through the aorta) at a rate of 420 ml/hr (35 ml in 5 min). 6. Harvest the ear tissue (or tissue from some other site) by dissection. Transfer the harvested tissue to a 24-well tissue culture plate, add 1 to 2 ml of 4% paraformaldehyde fixative solution, and then incubate 4 hr (or overnight) at 4◦ C, protected from light and with constant rotation. 7. Wash and mount the tissue of interest as described in Basic Protocol, steps 10 to 13. 8. Perform confocal microscopy on the tissue of interest and analyze the resulting images as described in Basic Protocol, steps 14 to 15. Cellular and Molecular Imaging
12.5.3 Current Protocols in Cytometry
Supplement 32
ALTERNATE PROTOCOL 2
DIRECT DETECTION OF BLOOD VASCULATURE AND EXPOSED REGIONS OF SUBENDOTHELIAL BASEMENT MEMBRANE This protocol combines Alternate Protocol 1 with rhodamine-conjugated Ricinus communis (castor bean) agglutinin I labeling to visualize luminal surfaces of blood vasculature and exposed regions of subendothelial basement membrane at vascular leakage sites.
Additional Materials (also see Basic Protocol) Rhodamine-conjugated R. communis (castor bean) agglutinin I (Vector Laboratories; 5.0 mg/ml stock) 4% (w/v) paraformaldehyde fixative solution (see recipe) 0.5-ml insulin syringe or 1.0-ml syringe equipped with a 29-G needle 60-ml syringe equipped with a blunted 16-G needle Infusion pump for cardiac perfusion (KDS100; KD Scientific) 565-to-615-nm band-pass filter or 650-nm long-pass filter Inject labeled compounds and perfuse mouse with fixative 1. Anesthetize the mouse using inhaled 2% isoflurane in oxygen (Donovan and Brown, 1998). 2. Using a 0.5-ml insulin syringe (or, alternatively, a 1.0-ml syringe) equipped with a 29-G needle, inject a mixture containing 50 µl of 5.0 mg/ml rhodamine-conjugated R. communis agglutinin I solution and 100 µl of 2.0 mg/ml FITC-conjugated Lycopersicon esculentum lectin solution into the mouse’s tail vein over the course of 20 to 30 sec. 3. Allow the injected compounds to circulate for 3 min. 4. Open the mouse’s chest, cut the right atrium of the heart, and use a 60-ml syringe equipped with a blunted 16-G needle to cannulate the left ventricle. 5. Using a KDS100 infusion pump, perfuse mouse with 35 ml of 4% paraformaldehyde fixative solution through the left ventricle (and thus through the aorta) at a rate of 420 ml/hr (35 ml in 5 min).
Harvest and mount tissue 6. Harvest the ear tissue (or tissue from some other site) by dissection. Transfer the harvested tissue to a 24-well tissue culture plate, add 1 to 2 ml of 4% paraformaldehyde fixative solution, and then incubate 4 hr (or overnight) at 4◦ C, protected from light and with constant rotation. 7. Wash and mount the tissue of interest as described in Basic Protocol, steps 10 to 13.
Collect and anlayze confocal data 8. Perform confocal microscopy (UNIT 2.8) on the mounted tissue using a Zeiss LSM 510 microscope (or a similar confocal system) operating in channel mode with 488-nm excitation. Collect the resulting FITC fluorescence emission (green fluorescence) using a 500-to-550-nm band-pass filter, and collect the resulting rhodamine fluorescence emission (red fluorescence) using a 565-to-615-nm band-pass filter. Acquire a z-stack at a resolution of 1 Airy unit to allow three-dimensional reconstruction.
3-D Visualization of Blood and Lymphatic Vasculature in Tissue
9. Digitally reconstruct the imaged tissue section and analyze the reconstructed image using the Zeiss LSM Image Examiner program or similar three-dimensional analysis software.
12.5.4 Supplement 32
Current Protocols in Cytometry
DIRECT DETECTION OF BLOOD AND LYMPHATIC VASCULATURE IN TISSUE
ALTERNATE PROTOCOL 3
This protocol combines Alternate Protocol 1 with an antibody-antigen-based immunodetection procedure. When an antibody with immunoreactivity against the lymphaticspecific cell surface receptor LYVE-1 is used, this combined approach allows one to simultaneously visualize blood and lymphatic vasculature in tissue.
Additional Materials (also see Basic Protocol) 4% (w/v) paraformaldehyde fixative solution (see recipe) PBS (APPENDIX 2A)/0.3% (v/v) Triton X-100 Normal goat serum 1:2000 rabbit anti-mouse LYVE-1 antibody (Upstate Biotechnology) in PBS (APPENDIX 2A)/0.3% (v/v) Triton X-100 1:500 goat anti-rabbit Alexa Fluor 594–conjugated secondary antibody or similar fluorochrome-conjugated secondary antibody (Molecular Probes) in PBS (APPENDIX 2A)/0.3% (v/v) Triton X-100 0.5-ml insulin syringe or 1.0-ml syringe equipped with a 29-G needle 60-ml syringe equipped with a blunted 16-G needle Infusion pump for cardiac perfusion (KDS100; KD Scientific) Platform rocker 650-nm long-pass filter Inject FITC-labeled lectin and perfuse mouse with fixative 1. Anesthetize the mouse using inhaled 2% isoflurane in oxygen (Donovan and Brown, 1988). 2. Inject FITC-labeled Lycopersicon esculentum lectin and perfuse mouse with 4% paraformaldehyde fixative solution as described in Alternate Protocol 1, steps 2 to 5.
Harvest tissue 3. Harvest the ear tissue (or tissue from some other site) by dissection. Transfer the harvested tissue to a 24-well tissue culture plate, add 1 to 2 ml of 4% paraformaldehyde fixative solution, and then incubate 4 hr (or overnight) at 4◦ C, protected from light and with constant rotation. 4. Remove the paraformaldehyde solution from the tissue and then wash the tissue three times, each time by adding 2.0 ml PBS, incubating 3 min at 4◦ C, protected from light, and then removing the PBS. 5. If using ear tissue, separate the dorsal and ventral aspects of the ears and remove all cartilage under a dissecting microscope. Discard all dorsal-side ear tissue. The dorsal side of the ear is not used because it contains muscle tissue, which complicates the imaging process.
Perform immunostaining NOTE: All immunostaining incubations are performed at 4◦ C on a platform rocker, with care being taken to protect samples from light. (Wrapping aluminum foil around the 24well tissue culture plates in which the incubations take place works well for this purpose.) 6. Transfer the ventral aspect of the ear to a 24-well tissue culture plate. 7. Prepare a fresh solution of 3% (v/v) goat serum in PBS/0.3% Triton X-100. Block the harvested tissue by adding 2.0 ml of this solution and incubating overnight. Cellular and Molecular Imaging
12.5.5 Current Protocols in Cytometry
Supplement 32
8. Remove the blocking solution from the tissue. Add at least 100 µl primary rabbit anti-mouse LYVE-1 antibody (1:2000 dilution in PBS/0.3% Triton X-100) to the tissue and incubate overnight. 9. Remove the primary antibody solution from the tissue. Wash the tissue five times (total washing time, 3 to 4 hr), each time by adding 2.0 ml PBS/0.3% Triton X-100, incubating, and then removing the PBS/0.3% Triton X-100 solution. 10. Prepare a fresh solution of 3% (v/v) goat serum in PBS/0.3% Triton X-100. Reblock the tissue by adding 2.0 ml of this solution and incubating 2 hr. 11. Remove the blocking solution from the tissue. Add 200 to 250 µl goat anti-rabbit Alexa Fluor 594–conjugated secondary antibody (1:500 dilution in PBS/0.3% Triton X-100) to the tissue and incubate overnight. 12. Remove the secondary antibody solution from the tissue. Wash the tissue ten times (total washing time, 3 to 4 hr), each time by adding 2.0 ml PBS/0.3% Triton X-100, incubating, and then removing the PBS/0.3% Triton X-100 solution.
Mount tissue 13. Carefully mount the immunostained tissue in antifade mounting medium on a glass slide, with the dermal surface of the tissue facing up. Add a coverslip to the slide and seal all edges with nail polish or rubber cement. 14. Store the mounted tissue at 4◦ C, protected from light. For optimal results, tissue should be imaged as soon as possible after mounting.
Collect and analyze confocal data 15. Perform confocal microscopy (UNIT 2.8) on the mounted tissue using a Zeiss LSM 510 META microscope (or a similar confocal system) operating in channel mode with 488-nm (for FITC) and 543-nm (for Alexa Fluor 594) excitation. Collect the resulting FITC fluorescence emission (green fluorescence) using a 500-to-550-nm band-pass filter, and the resulting Alexa Fluor 594 fluorescence emission (red fluorescence) using a 650-nm long-pass filter. Acquire a z-stack at a resolution of 1 Airy unit to allow three-dimensional reconstruction. 16. Digitally reconstruct the imaged tissue section and analyze the reconstructed image using the Zeiss LSM Image Examiner program or similar three-dimensional analysis software. ALTERNATE PROTOCOL 4
DIRECT DETECTION OF PERIVASCULAR SUPPORT CELLS IN BLOOD VASCULATURE Combining Alternate Protocol 1 or Alternate Protocol 2 with immunodetection using an antibody against α-smooth muscle actin allows visualization of the morphology of vascular smooth muscle cells lining the vasculature (Alternate Protocol 1) or assessment of the localization of these cells relative to vascular leakage sites (Alternate Protocol 2).
Additional Materials (also see Basic Protocol)
3-D Visualization of Blood and Lymphatic Vasculature in Tissue
Fluorescein isothiocyanate (FITC)–conjugated Lycopersicon esculentum (tomato) lectin (Vector Laboratories; 2.0 mg/ml stock) or FITC-conjugated Ricinus communis (castor bean) agglutinin I (Vector Laboratories; 5.0 mg/ml stock) 4% (w/v) paraformaldehyde fixative solution (see recipe) 1:1000 Cy3-conjugated anti–α-smooth muscle actin mouse monoclonal antibody (Sigma) in PBS (APPENDIX 2A)/0.3% (v/v) Triton X-100 Normal goat serum PBS (APPENDIX 2A)/0.3% (v/v) Triton X-100
12.5.6 Supplement 32
Current Protocols in Cytometry
0.5-ml insulin syringe or 1.0-ml syringe equipped with a 29-G needle 60-ml syringe equipped with a blunted 16-G needle Infusion pump for cardiac perfusion (KDS100; KD Scientific) 565-to-615-nm band-pass filter Inject labeled compound and perfuse fixative into mouse 1. Anesthetize the mouse using inhaled 2% isoflurane in oxygen (Donovan and Brown, 1988). 2. Using a 0.5-ml insulin syringe (or, alternatively, a 1.0-ml syringe) equipped with a 29-G needle, inject 100 µl of 2.0 mg/ml FITC-conjugated L. esculentum lectin solution or 50 µl of 5.0 mg/ml FITC-conjugated R. communis agglutinin I solution into the mouse’s tail vein over the course of 20 to 30 sec. A 0.5-ml insulin syringe is recommended for this step, as this will minimize the void volume.
3. Allow the injected FITC-conjugated compound to circulate for 3 min. 4. Open the mouse’s chest, cut the right atrium of the heart, and use a 60-ml syringe equipped with a blunted 16-G needle to cannulate the left ventricle. 5. Using a KDS100 infusion pump, perfuse mouse with 35 ml of 4% paraformaldehyde fixative solution through the left ventricle (and thus through the aorta) at a rate of 420 ml/hr (35 ml in 5 min).
Harvest tissue 6. Harvest the ear tissue (or tissue from some other site) by dissection. Transfer the harvested tissue to a 24-well tissue culture plate, add 1 to 2 ml of 4% paraformaldehyde fixative solution, and then incubate 4 hr (or overnight) at 4◦ C, protected from light and with constant rotation. 7. Remove the paraformaldehyde solution from the tissue and then wash the tissue three times, each time by adding 2.0 ml PBS; incubating 3 min at 4◦ C, protected from light; and then removing the PBS. 8. If using ear tissue, separate the dorsal and ventral aspects of the ears and remove all cartilage under a dissecting microscope. Discard all dorsal-side ear tissue. The dorsal side of the ear is not used because it contains muscle tissue, which complicates the imaging process.
Perform immunostaining NOTE: All immunostaining incubations are performed at 4◦ C on a platform rocker, with care being taken to protect samples from light. (Wrapping aluminum foil around the 24-well tissue culture plates in which the incubations take place works well for this purpose.) 9. Transfer the ventral aspect of the ear to a 24-well tissue culture plate. 10. Prepare a fresh solution of 3% (v/v) goat serum in PBS/0.3% Triton X-100. Permeabilize and block tissue by adding 2 ml of this solution and incubating overnight. 11. Remove the blocking solution from the tissue. Add 100 to 250 µl Cy3-conjugated antiα-smooth muscle actin mouse monoclonal antibody (1:1000 dilution in PBS/0.3% Triton X-100) to the tissue and incubate overnight. 12. Remove the antibody solution from the tissue. Wash the tissue ten times (total washing time, 3 to 4 hr), each time by adding 2.0 ml PBS/0.3% Triton X-100, incubating, and then removing the PBS/0.3% Triton X-100 solution.
Cellular and Molecular Imaging
12.5.7 Current Protocols in Cytometry
Supplement 32
Mount tissue and collect and analyze confocal data 13. Mount the immunostained tissue as described in Alternate Protocol 3, steps 13 to 14. 14. Perform confocal microscopy (UNIT 2.8) on the mounted tissue using a Zeiss LSM 510 META microscope (or a similar confocal system) operating in channel mode with 488-nm (for FITC) and 543-nm (for Cy3) excitation. Collect the resulting FITC fluorescence emission (green fluorescence) using a 500-to-550-nm band-pass filter, and collect the resulting Cy3 fluorescence emission (orange fluorescence) using a 565-to-615-nm band-pass filter. Acquire a z-stack at a resolution of 1 Airy unit to allow three-dimensional reconstruction. 15. Digitally reconstruct the imaged tissue section and analyze the reconstructed image using the Zeiss LSM Image Examiner program or similar three-dimensional analysis software.
REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Paraformaldehyde (4% and 1% w/v) fixative solution, pH 7.4 For 4% solution: Heat 100 ml of 1× PBS (APPENDIX 2A), pH 7.4, to 55◦ C. Add 4.0 g paraformaldehyde and stir to dissolve. Cool solution to room temperature and pass through a 0.44-µm filter. Store up to 1 week at 4◦ C. For 1% solution: Dilute 4% paraformaldehyde fixative solution fourfold in PBS (APPENDIX 2A). CAUTION: Paraformaldehyde is a carcinogen and may also cause allergic reactions. Use in a fume hood and dispose of paraformaldehyde-containing waste in accordance with institutional guidelines. Although they can be stored up to 1 week at 4◦ C, these solutions work best when made fresh before each use.
Paraformaldehyde (1% w/v)/glutaraldehyde (0.5% v/v) fixative solution, pH 7.4 Heat 100 ml of 1× PBS (APPENDIX 2A), pH 7.4, to 55◦ C. Add 1.0 g paraformaldehyde and stir to dissolve. Cool to room temperature, and then add 2.0 ml of 25% (v/v) glutaraldehyde and stir into solution. Pass solution through a 0.44-µm filter and store up to 1 week at 4◦ C. CAUTION: Glutaraldehyde and paraformaldehyde are carcinogens and may also cause allergic reactions. Use in a fume hood and dispose of glutaraldehyde- and paraformaldehyde-containing waste in accordance with institutional guidelines. Although it can be stored up to 1 week at 4◦ C, this solution works best when made fresh before each use. Glutaraldehyde is typically supplied as a 25% (v/v) solution and can be stored in aliquots at −20◦ C. Freeze/thaw cycles do not affect the stability of glutaraldehyde, but it is best to thaw frozen aliquots immediately before use. COMMENTARY 3-D Visualization of Blood and Lymphatic Vasculature in Tissue
Background Information In mammals, two vascular systems—the cardiovascular system and the lymphatic system—interact to monitor and maintain appropriate interstitial and intravascular fluid
volume. The cardiovascular system consists of endothelial cell–lined arteries, arterioles, capillaries, postcapillary venules, and veins. Together, these vessels make up the systemic and pulmonary circuits, which intersect at the
12.5.8 Supplement 32
Current Protocols in Cytometry
heart. Oxygen-rich blood leaves the left ventricle and flows through the aorta, smaller arteries, and arterioles into capillaries, where it supplies tissues with oxygen and nutrients while simultaneously removing toxic waste products, such as carbon dioxide. Oxygenpoor blood subsequently flows back toward the heart via postcapillary venules, veins, and the vena cava; enters the right atrium; flows into the right ventricle; and is pumped through the pulmonary artery to lung capillaries, where waste products are exchanged for oxygen. The resulting oxygen-rich blood is then transported to the left ventricle of the heart via the pulmonary veins, thus completing the circuit. When blood flows under high hydrostatic pressure through small capillaries and postcapillary venules, fluid, macromolecules, and blood cells extravasate (i.e., leak) into interstitial tissue compartments. Excessive interstitial fluid and blood cells return to the blood circulation via the lymphatic system, which is an open-ended, endothelial-lined vessel system consisting of lymphatic capillaries, lymphatic vessels, lymph nodes, and lymphoid organs. Interstitial fluid enters the highly permeable initial lymphatic capillaries, a process known as lymph formation. Lymph plays an important role in transporting immune cells and absorbing intestinal fats and thus contains lymphocytes, proteins, and fats. Lymph is transported via the larger valve-containing lymph vessels back into the blood circulation, entering through the subclavian veins. Together, these two fluid and cellular transport systems (the cardiovascular system and the lymphatic system) maintain appropriate hemodynamic and interstitial pressure, which is essential for proper tissue and organ function. Upon tissue or organ “damage,” each system exhibits innate response capabilities that, under physiologic conditions, initiate healing and the reestablishment of homeostasis (through menstruation, mammary gland lactation, or wound healing, for example); however, when chronically activated following pathologic assault (e.g., diabetic retinopathy, chronic inflammation, cancer), these responses can contribute to disease pathogenesis (McDonald, 1988a,b; McDonald, 1990; McDonald, 1994; Folkman, 1995; Thurston et al., 1996; Coussens et al., 1999; Hashizume et al., 2000; Beasley et al., 2002; Coussens and Werb, 2002; van Kempen et al., 2002). Three-dimensional visualization of vasculature in whole tissues using confocal
microscopy has proved to be a valuable tool for examining cellular and architectural changes accompanying altered blood and lymphatic vascular function (Thurston et al., 1996; Hashizume et al., 2000; Prevo et al., 2001). The protocols described here, which use murine skin as a prototype, have been adapted from McDonald and colleagues, who evaluated the relative affinities of various lectins for carbohydrate moieties present on the luminal surfaces of homeostatic versus inflamed blood endothelium. These pioneering studies revealed differential binding of lectins to carbohydrate moieties, making it possible to characterize endothelium in distinctive physiologic and pathologic states (Thurston et al., 1996). For example, lectins (including Lycopersicon esculentum lectin) with a primary specificity for N-acetyl-D-glucosamine oligomer bind uniformly to the luminal surface of blood endothelium in homeostatic and inflamed arterioles, capillaries, and venules (McDonald, 1994; Thurston et al., 1996). In contrast, lectins that bind N-acetyl-D-galactosamine, fucose, and sialic acid have affinity only for venules, whether homeostatic or inflamed. A third group of lectins exhibits selectivity for the mannose oligomers that are common in inflamed venules, but not in homeostatic ones (Thurston et al., 1996). Lectins that have primary specificity for galactosyl oligomers (e.g., Ricinus communis agglutinin I) bind to endothelium in inflamed venules at focal sites of subendothelial basement membrane exposure to luminal surfaces (Wadsworth et al., 1993; Thurston et al., 1996). Thus, when injected intravenously (i.v.) into live organisms, lectins harboring distinctive regional affinities for various types of blood vascular endothelium allow functional and morphological analysis of blood vasculature separate from the endothelial cells contained within the lymphatic vascular network. The hyaluronan receptor LYVE-1 (Prevo et al., 2001) is expressed by lymphatic endothelial cells but not by endothelial cells in blood vasculature; thus, combining the use of the i.v.-injected lectins with the immunohistochemical detection of LYVE-1 in tissue (Karkkainen et al., 2002) allows distinctive visualization of each vascular system. Similarly, using an antibody immunoreactive to the αchain of smooth muscle actin allows morphological visualization of vascular support cells embedded in the basement membrane around blood vasculature (Skalli et al., 1986; Schor et al., 1995).
Cellular and Molecular Imaging
12.5.9 Current Protocols in Cytometry
Supplement 32
Critical Parameters and Troubleshooting The efficiency of uniformly labeling blood vasculature with fluorescein isothiocyanate (FITC)–conjugated L. esculentum lectin and focal vascular leakage sites with rhodamineconjugated R. communis agglutinin I strongly depends on the ability of the researcher to perform tail-vein injections. The signal-to-noise ratio for immunohistochemical detection of the lymphatic endothelial cell–specific marker LYVE-1 strongly depends on the specificity of the fluorophoreconjugated secondary antibody used, as well as on the thoroughness of the washes. Successful cardiac perfusion of fluids into the vasculature is heavily dependent on the ability of the researcher not to puncture the ventricular septum (which separates the left and right ventricles) or the left atrium.
Anticipated Results
3-D Visualization of Blood and Lymphatic Vasculature in Tissue
These protocols are designed to detect and visualize blood vasculature using FITCconjugated L. esculentum lectin, which uniformly stains the blood vasculature without causing extensive background fluorescence. Injection of rhodamine-conjugated R. communis agglutinin I allows focal detection of vascular leakage sites with very low background fluorescence. Under homeostatic conditions, when murine tissue is unchallenged by inflammatory agents, few vascular leakage sites are detected in cutaneous tissues. Positive controls that readily induce vascular leakage through various cellular pathways should be included in all extravasation experiments; such controls can be generated by topically applying inflammatory agents such as mustard oil to mouse ears (Inoue et al., 1997). Vascular leakage sites are often found at the peripheral edge of the ear (Thurston et al., 1999). Detection of lymphatic vasculature is based on imunohistochemical staining for the hyaluronan receptor LYVE-1, which is specific to lymphatic endothelial cells (Prevo et al., 2001; Karkkainen et al., 2002). Lymphatic vasculature is typically wider in diameter compared to homeostatic blood vasculature, and the LYVE-1 antibody reveals that this vasculature possesses a latticelike structure. Importantly, although the immunohistochemical detection of α-smooth muscle actin–positive cells allows visualization of blood vascular support cells, it does not distinguish pericytes from smooth muscle cells lining the blood vasculature. Nevertheless, one
should expect to discern smooth muscle actin– positive cells tightly wrapped around arteriole vessels, whereas such cells are loosely connected in postcapillary venules (McDonald and Choyke, 2003).
Time Considerations Detection of blood vasculature via intravenous injection of FITC-conjugated L. esculentum lectin or detection of vascular leakage sites using rhodamine-conjugated R. communis agglutinin I can be performed in one to two days, whereas immunohistochemical detection of the lymphatic vasculature or of pericytes/smooth muscle actin–positive cells requires treatment of tissue for up to five days.
Literature Cited Beasley, N.J., Prevo, R., Banerji, S., Leek, R.D., Moore, J., van Trappen, P., Cox, G., Harris, A.L., and Jackson, D.G. 2002. Intratumoral lymphangiogenesis and lymph node metastasis in head and neck cancer. Cancer Res. 62:1315-1320. Coussens, L.M. and Werb, Z. 2002. Inflammation and cancer. Nature 420:860-867. Coussens, L.M., Raymond, W.W., Bergers, G., Laig-Webster, M., Behrendtsen, O., Werb, Z., Caughey, G.H., and Hanahan, D. 1999. Inflammatory mast cells up-regulate angiogenesis during squamous epithelial carcinogenesis. Genes Dev. 13:1382-1397. Donovan, J. and Brown, P. 1998. Anesthesia. In Current Protocols in Immunology (J.E. Coligan, A.M. Kruisbeek, D.H. Margulies, E.M. Shevach, and W. Strober, eds.) pp. 1.4.1-1.4.5. John Wiley & Sons, Hoboken, N.J. Folkman, J. 1995. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat. Med. 1:2731. Hashizume, H., Baluk, P., Morikawa, S., McLean, J.W., Thurston, G., Roberge, S., Jain, R.K., and McDonald, D.M. 2000. Openings between defective endothelial cells explain tumor vessel leakiness. Am. J. Pathol. 156:1363-1380. Inoue, H., Asaka, T., Nagata, N., and Koshihara, Y. 1997. Mechanism of mustard oil-induced skin inflammation in mice. Eur. J. Pharmacol. 333:231-240. Karkkainen, M.J., Makinen, T., and Alitalo, K. 2002. Lymphatic endothelium: A new frontier of metastasis research. Nat. Cell Biol. 4:E2-E5. McDonald, D.M. 1988a. Respiratory tract infections increase susceptibility to neurogenic inflammation in the rat trachea. Am. Rev. Respir. Dis. 137:1432-1440. McDonald, D.M. 1988b. Neurogenic inflammation in the rat trachea: I: Changes in venules, leucocytes and epithelial cells. J. Neurocytol. 17:583603. McDonald, D.M. 1990. The ultrastructure and permeability of tracheobronchial blood vessels in health and disease. Eur. Respir. J. 12:572s-585s.
12.5.10 Supplement 32
Current Protocols in Cytometry
McDonald, D.M. 1994. Endothelial gaps and permeability of venules in rat tracheas exposed to inflammatory stimuli. Am. J. Physiol. 266:L61L83.
venules by lectin binding. Am. J. Physiol. 271:H2547-H2562.
McDonald, D.M. and Choyke P.L. 2003. Imaging of angiogenesis: From microscope to clinic. Nat. Med. 9:713-725.
Thurston, G., Suri, C., Smith, K., McClain, J., Sato, T.N., Yancopoulos, G.D., and McDonald, D.M. 1999. Leakage-resistant blood vessels in mice transgenically overexpressing angiopoietin-1. Science 286:2511-2514.
Prevo, R., Banerji, S., Ferguson, D.J., Clasper, S., and Jackson, D.G. 2001. Mouse LYVE-1 is an endocytic receptor for hyaluronan in lymphatic endothelium. J. Biol. Chem. 276:1942019430.
van Kempen, L.C., Rhee, J.S., Dehne, K., Lee, J., Edwards, D.R., and Coussens, L.M. 2002. Epithelial carcinogenesis: Dynamic interplay between neoplastic cells and their microenvironment. Differentiation 70:610-623.
Schor, A.M., Canfield, A.E., Sutton, A.B., Arciniegas, E., and Allen, T.D. 1995. Pericyte differentiation. Clin. Orthop. 313:81-91.
Wadsworth, J.D., Okuno, A., and Strong, P.N. 1993. Assignment of laminin heavy chains using the lectin Ricinus communis agglutinin-1. Biochem. J. 295:537-541.
Skalli, O., Ropraz, P., Trzeciak, A., Benzonana, G., Gillessen, D., and Gabbiani, G. 1986. A monoclonal antibody against alpha-smooth muscle actin: A new probe for smooth muscle differentiation. J. Cell Biol. 103:2787-2796. Thurston, G., Baluk, P., Hirata, A., and McDonald, D.M. 1996. Permeability-related changes revealed at endothelial cell borders in inflamed
Contributed by Alexandra Eichten, H.-C. Jennifer Shen, and Lisa M. Coussens University of California at San Francisco San Francisco
The authors would like to acknowledge support from the Serono Foundation for the Advancement of Medial Sciences, as well as support from the Department of Defense, the National Cancer Institute, and the UCSF Comprehensive Cancer Center.
Cellular and Molecular Imaging
12.5.11 Current Protocols in Cytometry
Supplement 32
ABBREVIATIONS AND USEFUL DATA
APPENDIX 1
Abbreviations Used in This Manual
APPENDIX 1A
3D IA three-dimensional image analysis A adenine A23187 calcium ionophore A280 absorbance at 280 nm 7-AAD 7-aminoactinomycin D AAB biotinylated anti-avidin Ab antibody ABC antibody binding capacity ABDM anti-bromodeoxyuridine antibody AC alternating current ACD acid citrate dextrose AD actinomycin D ADB 1,4-diacetoxy-2,3-dicyanobenzene ADC analog-to-digital converter ADCC antibody-dependent cellular cytotoxicity AES acetic acid/ethanol/sorbitol (fixative) or aminopropyltriethoxysilane AET 2-aminoethylisothiouronium hydrobromide Ag antigen AlGaInP aluminum gallium indium phosphide át ratio of red to red plus green fluorescence AM acetoxymethyl (ester derivative) AMCA aminomethylcoumarin acetic acid AML acute myelogenous leukemia ANEPPS 1-(3-sulfonatopropyl)-4-[β-(2-{din-butylamino}-6-naphthyl)vinyl] pyridinium betaine AO acridine orange AOD acousto-optical deflector AOTF acoustico-optical tunable filter AP alkaline phosphatase APC allophycocyanin APC antigen-presenting cell API application program interface APS active pixel sensor APTRA 2-aminophenol-N,N,O-triacetic acid BAC bacterial artificial chromosome BAPTA 1,2-bis-(2-aminophenoxy)ethaneN,N,N′,N′-tetraacetic acid BCECF 2′,7′-bis-carboxyethyl-5(and 6)-carboxyfluorescein BCI 5-bromo-4-chloroindole BCIP 5-bromo-4-chloro-3-indolyl phosphate BCR B-cell receptor â-gal β-galactosidase BfISH bright-field in situ hybridization BFP blue fluorescent protein
BH bivariate histogram BODIPY 4,4-difluoro-4-bora-3a,4a-diaza-sindacene (basic structure) bp base pair BM bone marrow BMMC bone marrow mast cell BrdU 5-bromodeoxyuridine BrdUTP 5-bromodeoxyuridine triphosphate BRET bioluminescent resonance energy transfer BrU 5′-bromouridine BSA bovine serum albumin C cytosine CA chromomycin A3 CCCP carbonyl cyanide chlorophenyl hydrazone CARD catalyzed reporter deposition CCD charge-coupled device CCIR Comité Consultatif International des Radiocommunications CCS Clinical Cytometry Society CDF cumulative distribution function CDK cyclin-dependent protein kinase CDR recordable compact disk CD-ROM compact disk read-only memory CEP centromere enumeration probe CFDA carboxyfluorescein diacetate CFSE carboxyfluorescein diacetate succinimidyl ester cfu colony-forming units CGH comparative genomic hybridization CI chromosome index CID charge injection device CM-DiD chloromethyl derivative of DiIC18(3) CMFDA 5-chloromethylfluorescein diacetate CML chronic myelogenous leukemia CMOS complementary metal-oxide semiconductor CMTMRos 4-chloromethyl tetramethylrosamine CMXRos chloromethyl-X-rosamine COM center of mass cpd channels per decade CPS cluster pattern similarity CPT camptothecin CRBC chicken red blood cells CRC cyclic redundancy check Cr-LiSAF chromium-doped lithium strontium aluminum fluoride CsA cyclosporin A
Abbreviations and Useful Data
A.1A.1 Copyright © 2002 by John Wiley & Sons, Inc.
Supplement 20 CPCY
Abbreviations Used in This Manual
CTAD citrate theophylline adenosine dipyridimole CTC 5-cyano-2,3-ditolyltetrazolium chloride CTN calf thymocyte nuclei CV coefficient of variation CW continuous wave (laser) DAB 3,3′-diaminobenzidine DABCO 1,4-diazobicyclo-(2,2,2)-octane DAG diacylglycerol DAPI 4′,6-diamidino-2-phenylindole DB dilution buffer D/B dye-to-DNA-base (ratio) DBMS database management system DCF dichlorofluorescein DCFH-DA dichlorofluorescin diacetate DCH 2,3-dicyanohydroquinone DCI 4′,6-dicarboxyamide-2-phenylindole ∆ψ membrane potential DFM digital fluorescence microscopy DHPN 1,4-dihydroxyphthalonitrile DHR123 dihydrorhodamine 123 DI DNA (ploidy) index DiBAC4(3) bis-(1,3-dibutylbarbituric acid) trimethine oxonol DIC differential interference contrast Dig digoxygenin DiIC1(3) 1,1′-dimethyl-3,3,3′,3′-tetramethylindocarbocyanine DiIC1(5) 1,1′-dimethyl-3,3,3′,3′-tetramethylindodicarbocyanine DiIC18(3) 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine DiOC5(3) 3,3′-dipentyloxacarbocyanine DiOC6(3) 3,3′-dihexyloxacarbocyanine DiOC18(3) 3,3′-dioctadecyloxacarbocyanine DIRVISH direct visual hybridization DiTBAC4(3) bis-(1,3-dibutylthiobarbituric acid) trimethine oxonol DiTBAC4(5) bis-(1,3-dibutylthiobarbituric acid) pentamethine oxonol DLS dodecyl lithium sulfate DMC dynamic molecular combing DMF dimethylformamide DMSO dimethyl sulfoxide DMSP dichroic mirror short-pass DNA deoxyribonucleic acid DPI dots per inch dpm disintegrations per minute dNTP deoxyribonucleoside triphosphate DOP-PCR degenerate nucleotide–primed polymerase chain reaction DPBS Dulbecco’s phosphate-buffered saline DNP dinitrophenyl DPSS dye-pumped solid-state (laser) DRT dynamic regional thresholding
ds double-stranded (DNA) DSens diagnostic sensitivity DSpec diagnostic specificity DTT dithiothreitol dUTP deoxyuridine triphosphate dWB diluted whole blood EB ethidium bromide EBFP enhanced blue fluorescent protein ECD energy-coupled dye (fluorochrome) EDAC 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide EDTA ethylenediaminetetraacetic acid EFLP edge filter long-pass EGFP enhanced green fluorescent protein (product name) EIA Electronic Industries Association EMA ethidium monoazide ER endoplasmic reticulum ERK extracellular signal-regulatory kinase ESP EDTA/sarcosyl/proteinase K (buffer) E:T effector/target ratio EYFP enhanced yellow fluorescent protein F(ab′)2 antigen-binding fragment (of immunoglobulin) FACS fluorescence-activated cell sorting (Becton Dickinson trade name) FADH2 reduced flavin adenine dinucleotide FALS forward-angle light scatter FBS fetal bovine serum Fc complement-binding fragment (of immunoglobulin) FCCP carbonyl cyanide p-(trifluoromethoxy)phenylhydrazone FcεRI high-affinity IgE receptor FCS Flow Cytometry Standard f(D) grayscale transformation function FDA fluorescein diacetate FDG fluorescein di-β-D-galactoside FDGlcU fluorescein di-β-D-glucuronide (β-glucuronidase substrate) FI Fourier interference FISH fluorescence in situ hybridization FITC fluorescein isothiocyanate FLFCM fluorescence lifetime flow cytometry FM464 dye used as a marker of cell outline fMLP formyl-methionyl-leucyl-phenylalanine FN false negatives F/P fluorescence to protein ratio FPCV full peak coefficient of variation FRAP fluorescence recovery after photobleaching FRET fluorescent resonant energy transfer FLIP fluorescence loss in photobleaching FS forward scatter
A.1A.2 Supplement 20
Current Protocols in Cytometry
FST fostriecin FT Fourier transform G gauge or guanine G1 phase of cell cycle G1Q phase of cell cycle for arrested quiescent cells G2M phase of cell cycle G2Q phase of cell cycle for arrested quiescent cells GaAlAs gallium aluminum arsenide GaInP gallium indium phosphide Gb gigabyte gCRBC glutaraldehyde-fixed chicken red blood cells GFP green fluorescent protein GM Gompert Meyer (unit of molecular absorbance for the two-photon case) GMBS γ-maleimidobutyric acid N-hydroxysuccinimide ester GMD gel microdrop GPI glycosylphosphatidylinositol GPRP glycine-proline-arginine-proline (tetrapeptide) GSH glutathione (reduced form) GST glutathione-S-transferase GUI graphical user interface HBS HEPES-buffered saline HBSS Hanks’ balanced salt solution HDTV high-definition television HE hydroethidine HeCd helium/cadmium (laser) HeNe helium/neon (laser) HEPES N-2-hydroxyethylpiperazine-N′-2ethanesulfonic acid HIC hydrophobic interaction chromatography HIV human immunodeficiency virus HLF human lung fibroblast HO Hoechst 33258 (DNA stain) HO342 Hoechst 33342 HPCV half peak coefficient of variation HRP horseradish peroxidase HSC hematopoietic stem cell HSI hue, saturation, intensity HSR homogeneously stained region HT HEPES/Tyrode’s (buffer) HTB heat transfer block Hz hertz IA image analysis IC intracellular antigens IdU 5-iododeoxyuridine Ig immunoglobulin I/O input/output IOD integrated optical density IP3 inositol 1,4,5-triphosphate IR infrared
IRES internal ribosomal entry site ISAC International Society for Analytical Cytology ISH in situ hybridization ITAM immunoreceptor tyrosine-based activation motif ITIM immunoreceptor tyrosine-based inhibitory motif ITRU inline thermoregulation unit JC-1 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolylcarbocyanine iodide K562 human myeloid leukemia cell line KAR killer cell–activating receptor kb kilobase kbp kilobase pair KDD knowledge discovery in databases kHz kilohertz KIR killer cell–inhibitory receptor Ltot sum of red and green luminiscence intensities LAT linker for the activation of T cells LB Luria-Bertani (medium) LCF linear classification function LCTF liquid crystal tunable filter LDS-751 laser dye styryl 751 LED light-emitting diode LI labeling index LIBS ligand-induced binding sites LMP low-melting point LWD long working distance M phase of cell cycle MAb monoclonal antibody MAPK mitogen-activated protein kinase Mb megabase, megabyte mBBr monobromobimane mBCl monochlorobimane Mbp megabase pair MC mast cell MCF mean channel fluorescence MDS myelodysplastic syndrome MEM minimal essential medium MES 2-(N-morpholino)ethanesulfonic acid MESF molecules of equivalent soluble fluorochrome MFI mean fluorescence intensity MHz megahertz MI mithramycin MP membrane potential MPN most probable number MPO myeloperoxidase MTF modulation transfer function 6-MP 6-methylpurine 6-MPDR 6-methylpurine deoxyriboside 6-MPR 6-methylpurine riboside 4-MU 4-methylumbelliferone
Abbreviations and Useful Data
A.1A.3 Current Protocols in Cytometry
Supplement 20
Abbreviations Used in This Manual
4-MU 4-methylumbelliferone MUG 4-methylumbelliferyl-β-D-galactopyranoside MUGlcU 4-methylumbelliferyl-β-glucuronide MTX methotrexate MWCO molecular weight cutoff n refractive index N noise N degree of ploidy of cells (e.g., 1N = haploid, 2N = diploid, 4N = tetraploid) NA numerical aperture NADH2 reduced nicotinamide adenine dinucleotide NADPH reduced nicotinamide adenine dinucleotide phosphate NAO nonyl acridine orange NBD 7-nitrobenz-2-oxa-1,3-diazole NBM normal bone marrow NBT nitroblue tetrazolium Nc number of pixels in chess-knight move (orthogonal-diagonal or diagonal-orthogonal) NC nucleated cell NCCLS National Committee for Clinical Laboratory Standards Nd number of pixels in diagonal direction Nd-YLF neodymium-doped yttrium lanthanum fluoride Nd-YVO4 neodymium-doped yttrium vanadate ND neutral density (filter) NFAT nuclear factor of activated T cells NHS N-hydroxysuccinimide NIC Nomarski interference contrast NIH National Institutes of Health NK natural killer (cells) No number of pixels in orthogonal direction NP-40 Nonidet P-40 NSB nonspecific binding NTSC National Television Standards Committee OD optical density Op Amp operational amplifier OTF optical transfer function PAR1 protease-activated receptor 1 PBS phosphate-buffered saline PCR polymerase chain reaction pdf probability density function PE phycoerythrin or processing element PerCP peridinin chlorophyll protein (fluorochrome) PE-SA phycoerythrin-streptavidin PETG phenylethyl-β-D-thiogalactopyranoside PFGE pulsed-field gel electrophoresis
Pgp P-glycoprotein pHe extracellular pH pHi intracellular pH PI propidium iodide PIPES piperazine-N,N′-bis(2-ethanesulfonic acid) PKH2 green fluorescent cell linker developed by Paul K. Horan PKH26 red fluorescent cell linker developed by Paul K. Horan PLM probe length measurement PMA phorbol 12-myristate 13-acetate PMSF phenylmethylsulfonyl fluoride PMT photomultiplier tube PNH paroxysmal nocturnal hemoglobinuria poly(rA) polyriboadenosine PPACK D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone ppi pixels per inch PPU pulse pileup PRINS primed in situ (labeling) PRINSES primed in situ labeling en suspension PS power splitter PSD phase-sensitive detector psf point-spread function PTK protein tyrosine kinase PX membrane permeability of ion X PY pyronin Y QC quality control QFCM quantitative flow cytometry QIFI quantitative indirect immunofluorescence QSC quantum simply cellular R110 rhodamine 110 R123 rhodamine 123 RAM-TRITC TRITC-conjugated rabbit anti–mouse IgG RAM random access memory RBC red blood cell RET resonance energy tandem (fluorogenic substrate) RF radio frequency RFLP restriction fragment length polymorphism RGB red, green, blue RIBS receptor-induced binding sites RMS root mean square RNA ribonucleic acid R-PE R-phycoerythrin RPF resuscitation-promoting factor S phase of cell cycle SA streptavidin SAS software package for multivariate analysis SATA N-succinimidyl-S-acetylthioacetate
A.1A.4 Supplement 20
Current Protocols in Cytometry
SA-TR streptavidin-Tricolor SBFI sodium binding benzofuran isophthalate SBIP strand breaks induced by photolysis SCE sorbitol/citrate/EDTA (buffer) SD standard deviation SE succinimidyl ester SH2 Src homology 2 1,4-SL 1,4-saccharolactone SMCD systemic mast cell disease S/N signal-to-noise ratio SNARF-1 SemiNaphthoRhodaFluor-1 SP side population SPT single-platform technology; single-particle tracking ss single-stranded (DNA) SS side scatter SSC saline/sodium citrate SSCA single-cell cytotoxicity assay SSQ sum of squared differences SSR solid-state relay STAT signal transducer and activator of transcription STS sequence-tagged site T4 PNK T4 polynucleotide kinase T thymine TAE Tris/acetate/EDTA (buffer) Taq Thermus aquaticus (DNA polymerase) TBE Tris/borate/EDTA (buffer) TC1, TC2 T suppressor cell phenotypes (CD8+) TCC temperature control circuit TCR T cell receptor TdT terminal deoxynucleotide transferase TE Tris/EDTA (buffer) TEM transverse excitation mode
TFFDA tetrafluorofluorescein diacetate TH1, TH2 T helper cell phenotypes (CD4+) THF tetrahydrofuran Tm melting temperature (of DNA) TMA trimethylamine TMB tetramethylbenzidine TMRE tetramethylrhodamine ethyl ester TNB Tris/NaCl blocking reagent TNT Tris/NaCl/Tween buffer TO thiazole orange TP true positives Tpot tumor potential doubling time TPP+ tetraphenylphosphonium ion TR Texas Red TRAP thrombin receptor–activating peptide Ts duration of S phase of cell cycle TSA tyramide signal amplification TSB trypticase soy broth U units (e.g., of enzyme) ULWD ultralong working distance UV ultraviolet UVID ultraviolet-induced detection VCC volumetric capillary cytometry VRC vanadyl ribonucleoside complex v/v volume/volume WB whole blood WBC white blood cell WORM write-only, read many times w/v weight/volume Xgal 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside XHIM X-linked hyper IgM (syndrome) YAC yeast artificial chromosome YAG yttrium aluminum garnet (laser) YG yellow-green (beads)
Abbreviations and Useful Data
A.1A.5 Current Protocols in Cytometry
Supplement 20
Common Conversion Factors
APPENDIX 1B
Table A.1B.1 lists some of the more common conversion factors for units of measure used throughout Current Protocols manuals, while Table A.1B.2 gives prefixes indicating powers of ten for SI units. Table A.1B.3 gives conversions between temperatures on the Celsius (Centigrade) and Fahrenheit scales. Celsius temperatures are converted to Fahrenheit temperatures by multiplying the Celsius figure by 9, dividing by 5, and adding 32, or by multiplying the Celsius figure by 1.8 and adding 32. Fahrenheit is converted to Celsius by subtracting 32 from the Fahrenheit figure, multiplying by 5, and dividing by 9. In Table A.1B.3, the center figure represents the temperature one has read on one of the scales; the figure to the left is the conversion of that figure into Celsius if read in Fahrenheit, while that to the right represents the conversion to Fahrenheit if read in Celsius: e.g., the temperature 88 Fahrenheit converts to 31.1°C, while the temperature 88 Celsius converts to 190.4°F.
Table A.1B.1 Unit of Measurement Conversion Chart
To convert:
Into:
Multiply by:
amperes per square centimeter (amp/cm2)
amperes per square inch (amp/in.2) amperes per square meter (amp/m2)
6.452 104
amperes per square inch (amp/in.2)
amperes per square centimeter (amp/cm2) amperes per square meter (amp/m2)
0.1550 1.55 × 103
ampere-hours (amp-hr)
coulombs (C) faradays
3.6 × 103 3.731 × 10−2
atmospheres (atm)
bar millimeters of mercury (mmHg) or torr tons per square foot (tons/ft2)
1.01325 760 1.058
bar
atmospheres (atm) dynes per square centimeter (dyn/cm2) kilograms per square meter (kg/m2) pounds per square foot (lb/ft2) pounds per square inch (lb/in.2 or psi)
0.9869 106 1.020 × 104 2,089 14.50
British thermal units (Btu)
ergs gram-calories (g-cal) horsepower-hours (hp-hr) joules (J) kilogram-calories (kg-cal) kilogram-meters (kg-m) kilowatt-hours (kW-hr)
1.0550 × 1010 252.0 3.931 × 104 1,054.8 0.2520 107.5 2.928 × 10−4
British thermal unit per minute (Btu/min)
foot-pounds per second (ft-lb/sec) horsepower (hp) watts (W)
12.96 2.356 × 10−2 17.57
bushels
cubic feet (ft3) cubic inches (in.3) cubic meters (m3) liters quarts, dry
1.2445 2,150.4 3.524 × 10−2 35.24 32.0 continued
Current Protocols in Cytometry (2001) A.1B.1-A.1B.8 Copyright © 2001 by John Wiley & Sons, Inc.
A.1B.1 Supplement 15
Table A.1B.1 Unit of Measurement Conversion Chart, continued
To convert:
Into:
Multiply by:
centimeters (cm)
feet (ft) inches (in.) kilometers (km) meters (m) miles millimeters (mm) mils yards
3.281 × 10−2 0.3937 10−5 10−2 6.214 × 10−6 10.0 393.7 1.094 × 10−2
centimeters per second (cm/sec)
feet per minute (ft/min) feet per second (ft/sec) kilometers per hour (km/hr) meters per minute (m/min) miles per hour (miles/hr) miles per minute (miles/min)
1.1969 3.281 × 10−2 3.6 × 10−2 0.6 2.237 × 10−2 3.728 × 10−4
coulombs (C)
faradays
1.036 × 10−5
coulombs per square centimeter (C/cm2)
coulombs per square inch (C/in.2) coulombs per square meter (C/m2)
64.52 104
coulombs per square inch (C/in.2)
coulombs per square centimeter (C/cm2) coulombs per square meter (C/m2)
0.1550 1.55 × 103
cubic centimeters (cm3)
cubic feet (ft3) cubic inches (in.3)
3.531 × 10−5 6.102 × 10−2
cubic meters (m3) cubic yards gallons, U.S. liquid liters pints, U.S. liquid quarts, U.S. liquid
10−6 1.308 × 10−6 2.642 × 10−4 10−3 2.113 × 10−3 1.057 × 10−3
days
hours (hr) minutes (min) seconds (sec)
24.0 1.44 × 103 8.64 × 104
degrees (of angle; °)
minutes (min) quadrants, of angle radians (rad) seconds (sec)
60.0 1.111 × 10−2 1.745 × 10−2 3.6 × 104
drams
grams (g) grains ounces, avoirdupois (oz)
1.7718 27.3437 6.25 × 10−2
dynes (dyn)
joules per centimeter (J/cm) joules per meter (J/m) or newtons (N) kilograms (kg) pounds (lb)
10−7 10−5 1.020 × 10−6 2.248 × 10−6
faradays
ampere-hours (amp-hr) coulombs (C)
26.80 9.649 × 10−4 continued
A.1B.2 Supplement 15
Current Protocols in Cytometry
Table A.1B.1 Unit of Measurement Conversion Chart, continued
To convert:
Into:
Multiply by:
foot-pounds per minute (ft-lb/min)
British thermal units per minute (Btu/min) foot-pounds per second (ft-lb/sec) horsepower (hp) kilogram-calories per minute (kg-cal/min) kilowatts (kW)
1.286 × 10−3 1.667 × 10−2 3.030 × 10−5 3.24 × 10−4 2.260 × 10−5
grams (g)
decigrams (dg) dekagrams (dag) dynes (dyn) grains hectograms (hg) kilograms (kg) micrograms (µg) milligrams (mg) ounces, avoirdupois (oz) ounces, troy pounds (lb)
10 0.1 980.7 15.43 10−2 10−3 106 103 3.527 × 10−2 3.215 × 10−2 2.205 × 10−3
horsepower (hp)
horsepower, metric
1.014
inches (in.)
centimeters (cm) feet (ft) meters (m) miles millimeters (mm) yards
2.540 8.333 × 10−2 2.540 × 10−2 1.578 × 10−5 25.40 2.778 × 10−2
inches of mercury (in. Hg)
atmospheres (atm) kilogram per square centimeter (kg/cm2) kilograms per square meter (kg/m2) pounds per square foot (lb/ft2) pounds per square inch (lb/in.2 or psi)
3.342 × 10−2 3.453 × 10−2 345.3 70.73 0.4912
joules (J)
British thermal units (Btu) ergs foot-pounds (ft-lb) kilogram-calories (kg-cal) kilogram-meters (kg-m) newton-meter (N-m) watt-hours (W-hr)
9.480 × 10−4 107 0.7376 2.389 × 10−4 0.1020 1 2.778 × 10−4
kilolines
maxwells (Mx)
103
kilometers (km)
centimeters (cm) feet (ft) inches (in.) meters (m) miles yards
105 3,281 3.937 × 104 103 0.6214 1,094
kilowatts (kW)
British thermal units per minute (Btu/min) foot-pounds per minute (ft-lb/min) horsepower (hp) kilogram-calories per minute (kg-cal/min)
56.92 4.426 × 104 1.341 14.34
continued
A.1B.3 Current Protocols in Cytometry
Supplement 15
Table A.1B.1 Unit of Measurement Conversion Chart, continued
To convert:
Into:
Multiply by:
liters
bushels, U.S. dry cubic centimeters (cm3) cubic feet (ft3) cubic inches (in.3) cubic meters (m3) cubic yards gallons, U.S. liquid gallons, imperial kiloliter (kl) pints, U.S. liquid quarts, U.S. liquid
2.838 × 10−2 103 3.531 × 10−2 61.02 10−3 1.308 × 10−3 0.2642 0.21997 10−3 2.113 1.057
maxwells (Mx)
webers (Wb)
10−8
micrograms (µg)
grams (g)
10−6
microliters (µl)
liters
10−6
milligrams (mg)
grams (g)
10−3
milligrams per liter (mg/liter)
parts per million (ppm)
1.0
millihenries (mH)
henries (H)
10−3
milliliters (ml)
liters
10−3
millimeters (mm)
centimeters (cm) feet (ft) inches (in.) kilometers (km) meters (m) miles
0.1 3.281 × 10−3 3.937 × 10−2 10−6 10−3 6.214 × 10−7
millimeters of mercury (mmHg) or torr
atmospheres (atm) kilograms per square meter (kg/m2) pounds per square foot (lb/ft2) pounds per square inch (lb/in.2 or psi)
1.316 × 10−3 136.0 27.85 0.1934
nepers (Np)
decibels (dB)
8.686
newtons (N)
dynes (dyn) kilograms, force (kg) pounds, force (lb)
105 0.10197162 4.6246 × 10−2
ohms (Ω)
megaohms (MΩ) microhms (µΩ)
106 10−6
ounces, avoirdupois
drams grains grams (g) pounds (lb) ounces, troy tons, metric
16.0 437.5 28.349527 6.25 × 10−2 0.9115 2.835 × 10−5
ounces, fluid
cubic inches (in.3) liters
1.805 2.957 × 10−2 continued
A.1B.4 Supplement 15
Current Protocols in Cytometry
Table A.1B.1 Unit of Measurement Conversion Chart, continued
To convert:
Into:
Multiply by:
ounces, troy
grains grams (g) ounces, avoirdupois (oz) pounds, troy
480.0 31.103481 1.09714 8.333 × 10−2
pascal (P)
newton per square meter (N/m2)
1
pounds, force (lb)
newtons (N)
21.6237
pounds per square foot (lb/ft2)
atmospheres (atm) inches of mercury (in. Hg) kilograms per square meter (kg/m2) pounds per square inch (lb/in.2 or psi)
4.725 × 10−4 1.414 × 10−2 4.882 6.944 × 10−3
pounds per square inch (lb/in.2 or psi)
atmospheres (atm) inches of mercury (in. Hg) kilograms per square meter (kg/m2) pounds per square foot (lb/ft2) bar
6.804 × 10−2 2.036 703.1 144.0 6.8966 × 10−2
quadrants, of angle
degrees (°) minutes (min) radians (rad) seconds (sec)
90.0 5.4 × 103 1.571 3.24 × 105
quarts, dry
cubic inches (in.3)
67.20 (cm3)
quarts, liquid
cubic centimeters cubic feet (ft3) cubic inches (in.3) cubic meters (m3) cubic yards gallons liters
radians (rad)
degrees (°) minutes (min) quadrants seconds (sec)
57.30 3,438 0.6366 2.063 × 105
watts (W)
British thermal units per hour (Btu/hr) British thermal units per min (Btu/min) ergs per second (ergs/sec)
3.413 5.688 × 10−2 107
webers (Wb)
maxwells (M) kilolines
108 105
946.4 3.342 × 10−2 57.75 9.464 × 10−4 1.238 × 10−3 0.25 0.9463
torr see millimeter of mercury
Abbreviations and Useful Data
A.1B.5 Current Protocols in Cytometry
Supplement 15
Table A.1B.2 Power of Ten Prefixes for SI Units
Prefix
Factor
Abbreviation
atto femto pico nano micro milli centi deci deka hecto kilo myria mega giga tera peta exa
10−18 10−15 10−12 10−9 10−6 10−3 10−2 10−1 101 102 103 104 106 109 1012 1015 1018
a f p n µ m c d da h k my M G T P E
Common Conversion Factors
A.1B.6 Supplement 15
Current Protocols in Cytometry
Table A.1B.3 Celsius/Fahrenheit Temperature Conversion Charta
Degrees Celsius (°C)
Temperature
Degrees Fahrenheit (°F)
−17.8 −17.2 −16.7 −16.1 −15.6 −15.0 −14.4 −13.9 −13.3 −12.8 −12.2 −11.7 −11.1 −10.6 −10.0 −9.4 −8.9 −8.3 −7.8 −7.2 −6.7 −6.1 −5.6 −5.0 −4.4 −3.9 −3.3 −2.8 −2.2 −1.7 −1.1 −0.6 0.0 0.6 1.1 1.7 2.2 2.8 3.3 3.9 4.4 5.0 5.6 6.1 6.7 7.2 7.8 8.3 8.9 9.4 10.0 10.6
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
32.0 33.8 35.6 37.4 39.2 41.0 42.8 44.6 46.4 48.2 50.0 51.8 53.6 55.4 57.2 59.0 60.8 62.6 64.4 66.2 68.0 69.8 71.6 73.4 75.2 77.0 78.8 80.6 82.4 84.2 86.0 87.8 89.6 91.4 93.2 95.0 96.8 98.6 100.4 102.2 104.0 105.8 107.6 109.4 111.2 113.0 114.8 116.6 118.4 120.2 122.0 123.8 continued
Abbreviations and Useful Data
A.1B.7 Current Protocols in Cytometry
Supplement 15
Table A.1B.3 Celsius/Fahrenheit Temperature Conversion Charta, continued
Degrees Celsius (°C)
Temperature
Degrees Fahrenheit (°F)
11.1 11.7 12.2 12.8 13.3 13.9 14.4 15.0 15.6 16.1 16.7 17.2 17.8 18.3 18.9 19.4 20.0 20.6 21.1 21.7 22.2 22.8 23.3 23.9 24.4 25.0 25.6 26.1 26.7 27.2 27.8 28.3 28.9 29.4 30.0 30.6 31.1 31.7 32.2 32.8 33.3 33.9 34.4 35.0 35.6 36.1 36.7 37.2 37.8
52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
125.6 127.4 129.2 131.0 132.8 134.6 136.4 138.2 140.0 141.8 143.6 145.4 147.2 149.0 150.8 152.6 154.4 156.2 158.0 159.8 161.6 163.4 165.2 167.0 168.8 170.6 172.4 174.2 176.0 177.8 179.6 181.4 183.2 185.0 186.8 188.6 190.4 192.2 194.0 195.8 197.6 199.4 201.2 203.0 204.8 206.8 208.4 210.2 212.0
a°F = 9⁄ (°C) + 32; °C = 5⁄ (°F − 32); °K = °C + 273.15. 5 9
Common Conversion Factors
A.1B.8 Supplement 15
Current Protocols in Cytometry
STOCK SOLUTIONS, EQUIPMENT, AND LABORATORY GUIDELINES
APPENDIX 2
Common Stock Solutions, Buffers, and Media
APPENDIX 2A
Use distilled, deionized water in all recipes. Ammonium chloride lysing solution, 10× 80.2 g NH4Cl (1.5 M) 8.4 g NaHCO3 (100 mM) 3.7 g disodium EDTA (10 mM) H2O to 900 ml Adjust pH to 7.4 with 1 N HCl or 1 N NaOH Add H2O to 1 liter Store ≤6 months at 4°C Working solution: Dilute 1:10 to make working lysing solution fresh before use. Keep working solution cold and discard any unused portion. This solution is used to lyse erythrocytes. Never store lysing solution at <10× concentration, as it will form ammonium carbonate, which is ineffective. Sodium bicarbonate may be replaced by 10 g KHCO3 (100 mM) and disodium EDTA by 3.66 g tetrasodium EDTA (8.2 mM). If desired, lysing solution may be adjusted to a specific pH using 1 N HCl or 1 N NaOH. Some researchers prefer to use a 10-fold lower concentration of EDTA (1 mM disodium EDTA or 0.82 mM tetrasodium EDTA in the 10× recipe).
Complete DMEM Dulbecco’s modified Eagle medium, high-glucose formulation (e.g., Life Technologies), containing: 5%, 10%, or 20% (v/v) FBS, heat-inactivated (optional; see recipe) 1% (v/v) nonessential amino acids 2 mM L-glutamine (see recipe) 100 U/ml penicillin 100 µg/ml streptomycin sulfate Filter sterilize and store ≤1 month at 4°C Throughout this manual, the percentage of serum (usually fetal bovine serum, FBS) used in a protocol step is indicated by a numeral hyphenated to the base medium name. Thus, “complete DMEM-10” indicates that 10% FBS is used. Absence of a numeral indicates that no serum is used. DMEM containing 4500 mg/liter D-glucose can be obtained from Life Technologies. DMEM is also known as Dulbecco’s minimum essential medium.
Complete RPMI RPMI 1640 medium (e.g., Life Technologies) containing: 2%, 5%, 10%, 15%, or 20% FBS, heat-inactivated (optional; see recipe) 2 mM L-glutamine (see recipe) 100 U/ml penicillin 100 µg/ml streptomycin sulfate Filter sterilize and store ≤1 month at 4°C
Stock Solutions, Equipment, and Laboratory Guidelines Current Protocols in Cytometry (1999) A.2A.1-A.2A.4 Copyright © 1999 by John Wiley & Sons, Inc.
A.2A.1 Supplement 9
DTT (DL-dithiothreitol), 0.1 M 0.154 g DTT 10 ml H2O Store in 100-ml aliquots ≤2 years at −20°C Prior to use, thaw an aliquot and place at 0°C Aliquots can be thawed and frozen repeatedly.
EDTA (ethylenediamine tetraacetic acid), 0.5 M, pH 8 Dissolve 186.1 g Na2EDTA⋅2H2O in 700 ml water with stirring. Adjust the pH to 8.0 with 10 M NaOH, then adjust volume to 1 liter, and autoclave. Store ≤1 year at room temperature. The EDTA will not dissolve fully until the pH is adjusted to 8.
Ethidium bromide staining solution Concentrated stock solution: Dissolve 1 mg ethidium bromide (e.g., Molecular Probes) in 1 ml distilled water. Store at 4°C in the dark or in foil-wrapped bottle; stable for months. Staining solution: For staining gels, add 20 µl concentrated stock to 200 ml electrophoresis buffer (e.g., 1× TBE or TAE; see recipes) or water. Alternatively, 0.5 ìg/ml ethidium bromide may be incorporated into gels directly as they are being cast. CAUTION: Ethidium bromide, a nucleic acid–intercalating dye, is mutagenic and must be handled and disposed of with appropriate care.
FBS (fetal bovine serum) Thaw purchased fetal bovine serum (shipped on dry ice and kept frozen until needed). Store 3 to 4 weeks at 4°C. If FBS is not to be used within this time, aseptically divide into smaller aliquots and refreeze until used. Store ≤1 year at −20°C. Repeated thawing and refreezing should be avoided as it may cause denaturation of the serum.
Inactivated FBS (FBS that has been treated with heat to inactivate complement protein and thus prevent an immunological reaction against cultured cells) is useful for a variety of purposes. It can be purchased commercially or made in the lab. To inactivate FBS, heat serum 30 min to 1 hr in a 56°C water bath. Alternatively, FBS may be inactivated through radiation treatment.
Formamide, deionized Stir 100 ml formamide with 5 g Serdolit MB-1 ion exchanger (Serva) for 2 hr at room temperature. Filter twice through Whatman no.1 filter paper. Store in 1-ml aliquots ≤5 years at −20°C. Prior to use, thaw an aliquot and place at 0°C. Aliquots can be thawed and frozen repeatedly. IMPORTANT NOTE: Formamide used in large-volume washing solutions does not need to be deionized. However, check the pH and adjust to 7 with a few drops of 5 M HCl as needed.
Gel loading buffer, 6× 0.25 g bromphenol blue (e.g., Sigma) 0.25 g xylene cyanol (e.g., Sigma) 70 ml distilled H2O 30 ml glycerol (e.g., Sigma) Store indefinitely at 4°C Common Stock Solutions, Buffers, and Media
continued
A.2A.2 Supplement 9
Current Protocols in Cytometry
Alternatively, 40% (w/v) sucrose (molecular biology grade) or 15% (w/v) Ficoll 400 may be used instead of the glycerol; these formulations are essentially interchangeable. The solution may be stored at room temperature if Ficoll is used. This buffer does not need to be sterilized. Other concentrations can be prepared if more convenient. L-Glutamine,
0.2 M (100×) Thaw commercially prepared frozen L-glutamine or prepare an 0.2 M solution in water, aliquot aseptically into usable portions, then refreeze. For convenience, L-glutamine can be stored in 1-ml aliquots if 100-ml bottles of medium are used and in 5-ml aliquots if 500-ml bottles are used. Store ≤1 year at −20°C. Many laboratories supplement medium with 2 mM L-glutamine—1% (v/v) of 100× stock— just prior to use.
HBSS (Hanks’ balanced salt solution) 0.40 g KCl (5.4 mM) 0.09 g Na2HPO4⋅7H2O (0.3 mM) 0.06 g KH2PO4 (0.4 mM) 0.35 g NaHCO3 (4.2 mM) 0.14 g CaCl2 (1.3 mM) 0.10 g MgCl2⋅6H2O (0.5 mM) 0.10 g MgSO4⋅7H2O (0.6 mM) 8.0 g NaCl (137 mM) 1.0 g D-glucose (5.6 mM) Add H2O to 1 liter and adjust pH to 7.4 Filter sterilize Store indefinitely at 4°C HBSS can also be purchased from a number of commercial suppliers. It is sometimes desirable to use HBSS without Ca2+ and Mg2+, which may be obtained commercially or prepared by omitting CaCl2, MgCl2, and MgSO4 from the above recipe. These components are optional and usually have no effect on an experiment; in a few cases, however, their presence may be detrimental. Consult individual protocols for recommendations.
PBS (phosphate-buffered saline) 0.23 g NaH2PO4 (anhydrous; 1.9 mM) 1.15 g Na2HPO4 (anhydrous; 8.1 mM) 9.00 g NaCl (154 mM) Add H2O to 900 ml If needed, adjust to desired pH (usually 7.2 to 7.4) with 1 M NaOH or 1 M HCl Add H2O to 1 liter Filter sterilize Store indefinitely at 4°C Without pH adjustment, pH will generally be ~7.3. Some laboratories find it convenient to prepare PBS at 10× concentration and dilute this to make the working solution. PBS can also be purchased commercially in powder form.
RNase A stock solution (DNase-free), 2 mg/ml Dissolve 2 mg RNase A (e.g., Sigma) in 1 ml distilled water. If the RNase is not DNase free, heat the solution 5 min at 100°C to inactivate any traces of DNase. Store ≤1 year at 4°C. Stock Solutions, Equipment, and Laboratory Guidelines
A.2A.3 Current Protocols in Cytometry
Supplement 28
SSC, 20× 175.3 g NaCl 88.24 g sodium citrate Add 950 ml H2O Adjust pH to 7 with 5 M HCl Dilute to 1 liter with H2O Store ≤1 month at room temperature TAE buffer, 50× 242 g Tris base 57.1 ml glacial acetic acid 10 ml 0.5 M EDTA, pH 8.0 (see recipe) H2O to 1 liter Store ≤1 year at room temperature TBE buffer, 10× 108 g Tris base (890 mM) 55 g boric acid (890 mM) 960 ml H2O 40 ml 0.5 M EDTA, pH 8.0 (20 mM; see recipe) Store indefinitely at room temperature TE buffer 10 mM Tris⋅Cl (see recipe) at desired pH 1 mM EDTA, pH 8.0 Autoclave Store indefinitely at room temperature Tris⋅Cl, 1 M Dissolve 121 g Tris base [tris(hydroxymethyl)aminomethane] in 800 ml H2O Adjust to desired pH with concentrated HCl Mix and add H2O to 1 liter Approximately 70 ml HCl is needed to achieve a pH 7.4 solution and ~42 ml for a solution that is pH 8.0. IMPORTANT NOTE: The pH of Tris buffers changes significantly with temperature, decreasing 0.028 pH units per 1°C. Tris-buffered solutions should be adjusted to the desired pH at the temperature at which they will be used. Because the pKa of Tris is 8.08, Tris should not be used as a buffer below pH ~7.2 or above pH ~9.0.
Trypsin/EDTA solution Prepare in sterile HBSS (see recipe) or 0.9% (w/v) NaCl: 0.25% (w/v) trypsin 0.2% (w/v) EDTA Store ≤1 year (until needed) at −20°C Specific applications may require different concentrations of trypsin. Trypsin/EDTA solution is commercially available in various concentrations including 10×, 1×, and 0.25% (w/v). It is received frozen from the manufacturer and can be thawed and aseptically divided into smaller volumes. Preparing trypsin/EDTA from powdered stocks may reduce its cost; however, most laboratories prefer commercially prepared solutions for convenience. EDTA (disodium ethylenediamine tetraacetic acid) is added as a chelating agent to bind Ca2+ and Mg2+ ions that can interfere with the action of trypsin. Common Stock Solutions, Buffers, and Media
A.2A.4 Supplement 28
Current Protocols in Cytometry
COMMONLY USED TECHNIQUES
APPENDIX 3
Cell Counting
APPENDIX 3A
COUNTING CELLS USING A HEMACYTOMETER Determining the number of cells in a preparation or a culture may at times be necessary for standardization of conditions, estimation of staining reagent required, or accurate quantitation experiments. Cells can be counted directly under the microscope with the aid of the hemacytometer. Cell viability can also be determined by staining cells with trypan blue and counting (see APPENDIX 3B, Support Protocol 3).
BASIC PROTOCOL 1
The hemacytometer is a thick glass slide with a central area designed as a counting chamber. The exact design may vary; the one described here is the Improved Neubauer (Fig. A.3A.1). The central portion of the slide is the counting platform, which is surrounded by a 1-mm groove. A transverse groove divides this platform into two counting chambers, each consisting of a silver footplate on which is etched a 3 × 3–mm grid. This grid is subdivided into nine 1 × 1–mm squares. Cells are usually counted in the four corner squares and the center square. The former are further divided into 16 tertiary squares and the latter into 25 tertiary squares to aid in counting. Accompanying the hemacytometer slide is a thick, even-surfaced coverslip. Ordinary coverslips may have uneven surfaces, which can introduce errors in cell counting; therefore, it is imperative that the coverslip provided with the hemacytometer be used in determining cell number. Cell suspension is applied to a defined volume and counted so cell density can be calculated. Materials 70% (v/v) ethanol Cell suspension 0.4% (w/v) trypan blue or 0.4% (w/v) nigrosin, prepared in HBSS (APPENDIX 2A) Hemacytometer with coverslip (e.g., Improved Neubauer, VWR) Hand-held counter Prepare hemacytometer 1. Clean surface of hemacytometer slide and coverslip with 70% alcohol. Coverslip and slide should be clean, dry, and free from lint, fingerprints, and watermarks.
2. Place the coverslip over the central area of the hemacytometer and press firmly into position. The coverslip should rest evenly over the silver counting area. Some workers place a small amount of water on the ground glass area of the hemacytometer, on which the coverslip rests, to improve the seal.
Prepare cell suspension 3. For cells grown in monolayer cultures, detach cells from surface of dish using trypsin (see APPENDIX 3B, Basic Protocol). 4. Dilute cells as needed to obtain a uniform suspension. Disperse any clumps. When using the hemacytometer, a maximum cell count of 20 to 50 cells per 1-mm square is recommended. Contributed by Mary C. Phelan and Gretchen Lawler Current Protocols in Cytometry (1997) A.3A.1-A.3A.4 Copyright © 1997 by John Wiley & Sons, Inc.
Commonly Used Techniques
A.3A.1
Load hemacytometer 5. Use a sterile Pasteur pipet to transfer cell suspension to edge of hemacytometer counting chamber. Hold tip of pipet under the coverslip and dispense one drop of suspension. Suspension will be drawn under the coverslip by capillary action. Hemacytometer should be considered nonsterile. If cell suspension is to be used for cultures, do not reuse the pipet and do not return any excess cell suspension in the pipet to the original suspension.
6. Fill second counting chamber in the same manner.
coverslip
grid load cell suspension
1mm
2
1 3 4
Cell Counting
1mm
5
Figure A.3A.1 Hemacytometer slide (Improved Neubauer) and coverslip. Coverslip is applied to slide and cell suspension is added to counting chamber using a Pasteur pipet. Each counting chamber has a 3 × 3–mm grid (enlarged). The four corner squares (1, 2, 4, and 5) and the central square (3) are counted on each side of the hemacytometer (numbers added).
A.3A.2 Current Protocols in Cytometry
Count cells 7. Allow cells to settle for a few minutes before beginning to count. Blot off excess liquid. 8. View slide on microscope with 100 magnification. A 10× ocular with a 10× objective = 100× magnification. Position slide to view the large central area of the grid (section 3 in Fig. A.3A.1); this area is bordered by a set of three parallel lines. The central area of the grid should almost fill the microscope field. Subdivisions within the large central area are also bordered by three parallel lines and each subdivision is divided into sixteen smaller squares by single lines. Cells within this area should be evenly distributed without clumping. If cells are not evenly distributed, wash and reload hemacytometer.
9. Use a hand-held counter to record cell counts in each of the four corner and central squares (Fig. A.3A.1, squares numbered 1 to 5). Repeat counts for other counting chamber. Five squares (four corner and one center) are counted from each of the two counting chambers for a total of ten squares counted. Count cells touching the middle line of the triple line on the top and left of the squares. Do not count cells touching the middle line of the triple lines on the bottom or right side of the square.
Calculate cell number 10. Determine cells per milliliter by the following calculations: cells/ml = average count per square × dilution factor × 104 total cells = cells/ml × total original volume of cell suspension from which sample was taken. 104 is the volume correction factor for the hemacytometer: each square is 1 × 1 mm and the depth is 0.1 mm.
11. Decontaminate coverslip and hemacytometer by rinsing with 70% ethanol and then deionized water. Air dry and store for future use. COUNTING CELLS USING A COULTER COUNTER The Coulter counter is an electronic particle counter that provides an alternative to the hemacytometer for counting cells. Cells in suspension pass through an aperture across which flows an electric current. Their presence alters the electrical resistance of the medium and induces changes in the current flow and voltage. The magnitude of these changes is directly proportional to cell size and is electronically converted to a particle count. Up to 500 particles per second can be individually counted and sized, regardless of shape or orientation. On most Coulter counters a threshold control can be set to eliminate counting of particles smaller than a selected size, thereby reducing inaccuracies introduced by debris and other materials not of interest. Operation and maintenance of the Coulter counter is fully explained in the manufacturer’s manual.
BASIC PROTOCOL 2
Materials Cell suspension 20 ml counting vials Zap-O-globin (Coulter), lysing and hemoglobin reagent (VWR), or equivalent (for counting white blood cells) 1. Pipet 10 ml PBS into a counting vial and add 20 µl well-mixed cell suspension. This is a 1/500 dilution, suitable for most cells. To count erythrocytes, make an additional 1/100 dilution (1/50,000 final).
Commonly Used Techniques
A.3A.3 Current Protocols in Cytometry
2. For a white cell count on peripheral blood: Add 3 drops of Zap-O-globin to the vial. Cap the vial and invert once or twice to mix. The Zap-O-globin will lyse any residual erythrocytes, which otherwise distort the count because of their overwhelming numbers. This step is unnecessary with cell cultures. Do not use Zap-O-globin for erythrocyte counts. Once opened, Zap-O-globin does not keep indefinitely. It should be protected from light and discarded as soon as it shows a green tint or formation of hair-like crystals.
3. Place the vial under the aperture tube of the counter. Open the vacuum control stopcock to rezero the instrument and close again to begin the count. 4. Record the result and repeat the count for verification. For the 1/500 dilution, the display reads directly in cells/mm3; multiply the reading by 103 to obtain cells/ml. Results should be within 50 counts of each other, and a background reading made on PBS alone should be less than 50. High backgrounds indicate contamination of the PBS, air bubbles in the system, or a dirty instrument.
COMMENTARY Cell counting with a hemacytometer is tedious and time-consuming, but does permit effective discrimination of live from dead cells using trypan blue exclusion (APPENDIX 3B). In addition, the procedure is less subject to errors arising from cell clumping or size heterogeneity. Significant inaccuracies can occur in two situations, when a nonrepresentative sample is counted and when too few cells are counted; at least 100, and preferably 300, should be counted. Counting cells is more quickly and easily performed by an electronic counter, but livedead discrimination is unreliable. Cell populations containing large numbers of dead cells and/or cell clumps are difficult to count accurately. In addition, electronic counting requires
resetting of the instrument for cell populations of different sizes; heterogeneous populations can give rise to inaccurate counts, and resting and activated cells may require counting at separate settings. In general, electronic cell counting is best performed on fresh peripheral blood cells.
Contributed by Mary C. Phelan (hemacytometer) Thompson Children’s Hospital Chattanooga, Tennessee Gretchen Lawler (Coulter counter) Purdue University West Lafayette, Indiana
Cell Counting
A.3A.4 Current Protocols in Cytometry
Techniques for Mammalian Cell Tissue Culture
APPENDIX 3B
Tissue culture technology has found wide application in biological research. Monolayer cell cultures are utilized in cytogenetic, biochemical, and molecular laboratories for research as well as diagnostic studies. In most cases, cells or tissues must be grown in culture for days or weeks to obtain sufficient numbers of cells for analysis. Maintenance of cells in long-term culture requires strict adherence to aseptic technique to avoid contamination and potential loss of valuable cell lines. The first section of this appendix discusses basic principles of sterile technique. An important factor influencing the growth of cells in culture is the choice of tissue culture medium. Many different recipes for tissue culture media are available and each laboratory must determine which medium best suits their needs. Individual laboratories may elect to use commercially prepared medium or prepare their own. Commercially available medium can be obtained as a sterile and ready-to-use liquid, in a concentrated liquid form, or in a powdered form. Besides providing nutrients for growing cells, medium is generally supplemented with antibiotics, fungicides, or both to inhibit contamination. The second section of this appendix discusses medium preparation. As cells grown in monolayer reach confluency, they must be subcultured or passaged. Failure to subculture confluent cells results in reduced mitotic index and eventually in cell death. The first step in subculturing is to detach cells from the surface of the primary culture vessel by trypsinization or mechanical means. The resultant cell suspension is then subdivided, or reseeded, into fresh cultures. Secondary cultures are checked for growth, fed periodically, and may be subsequently subcultured to produce tertiary cultures, etc. The time between passaging cells varies with the cell line and depends on the growth rate. The Basic Protocol describes subculturing of a monolayer culture grown in petri dishes or flasks. Support Protocols 1 to 4 describe freezing of monolayer cells, thawing and recovery of cells, determining cell viability by trypan blue extraction, and preparing cells for transport. The Alternate Protocol describes freezing suspension cells. STERILE TECHNIQUE It is essential that sterile technique be maintained when working with cell cultures. Aseptic technique involves a number of precautions to protect both the cultured cells and the laboratory worker from infection. The laboratory worker must realize that cells handled in the lab are potentially infectious and should be handled with caution. Protective apparel such as gloves, lab coats or aprons, and eyewear should be worn when appropriate (Knutsen, 1991). Care should be taken when handling sharp objects such as needles, scissors, scalpel blades, and glass that could puncture the skin. Sterile disposable plastic supplies may be used to avoid the risk of broken or splintered glass (Rooney and Czepulkowski, 1992). Frequently, specimens received in the laboratory are not sterile, and cultures prepared from these specimens may become contaminated with bacteria, fungus, or yeast. The presence of microorganisms can inhibit growth, kill cell cultures, or lead to inconsistencies in test results. The contaminants deplete nutrients in the medium and may produce substances that are toxic to cells. Antibiotics (penicillin, streptomycin, kanamycin, or gentamycin) and fungicides (amphotericin B or mycostatin) may be added to tissue Contributed by Mary C. Phelan Current Protocols in Cytometry (1997) A.3B.1-A.3B.10 Copyright © 1997 by John Wiley & Sons, Inc.
Commonly Used Techniques
A.3B.1
Table A.3B.1 Working Concentrations of Antibiotics and Fungicides for Mammalian Cell Culture
Additive Penicillin Streptomycin sulfate Kanamycin Gentamycin Mycostatin Amphotericin B
Final concentration 50–100 U/ml 50–100 µg/ml 100 µg/ml 50 µg/ml 20 µg/ml 0.25 µg/ml
culture medium to combat potential contaminants (see Table A.3B.1). An antibiotic/antimycotic solution or lyophilized powder that contains penicillin, streptomycin, and amphotericin B is available from Sigma. The solution can be used to wash specimens prior to culture and can be added to medium used for tissue culture. Similar preparations are available from other suppliers. All materials that come into direct contact with cultures must be sterile. Sterile disposable dishes, flasks, pipets, etc., can be obtained directly from manufacturers. Reusable glassware must be washed, rinsed thoroughly, then sterilized by autoclaving or by dry heat before reusing. With dry heat, glassware should be heated 90 min to 2 hr at 160°C to ensure sterility. Materials that may be damaged by very high temperatures can be autoclaved 20 min at 120°C and 15 psi. All media, reagents, and other solutions that come into contact with the cultures must also be sterile; medium may be obtained as a sterile liquid from the manufacturer, autoclaved if not heat-sensitive, or filter sterilized. Supplements can be added to media prior to filtration, or they can be added aseptically after filtration. Filters with 0.20- to 0.22-µm pore size should be used to remove small gram-negative bacteria from culture media and solutions. Contamination can occur at any step in handling cultured cells. Care should be taken when pipetting media or other solutions for tissue culture. If the pipet tip comes into contact with the benchtop or any other nonsterile surface, it should be discarded and a fresh pipet obtained. Forceps and scissors used in tissue culture can be rapidly sterilized by dipping in 70% alcohol and flaming. Although tissue culture work can be done on an open bench if aseptic methods are strictly enforced, many labs prefer to perform tissue culture work in a room or low-traffic area reserved specifically for that purpose. At the very least, biological safety cabinets are recommended to protect the cultures as well as the laboratory worker. In a laminar flow hood, the flow of air protects the work area from dust and contamination and acts as a barrier between the work surface and the worker. Many different styles of safety hoods are available and the laboratory should consider the types of samples being processed and the types of potential pathogenic exposure in making their selection. Manufacturer recommendations should be followed regarding routine maintenance checks on air flow and filters. For day-to-day use, the cabinet should be turned on for at least 5 min prior to beginning work. All work surfaces both inside and outside of the hood should be kept clean and disinfected daily and after each use. Techniques for Mammalian Cell Tissue Culture
Some safety cabinets are equipped with ultraviolet (UV) lights for decontamination of work surfaces. However, UV lamps are generally ineffective and their use is no longer recommended (Knutsen, 1991). Effectiveness diminishes over time as the glass tube
A.3B.2 Current Protocols in Cytometry
gradually loses its ability to transmit short UV wavelengths, and it may also be reduced by dust on the glass tube, distance from the work surface, temperature, and air movement. Even when the UV output is adequate, the rays must strike directly in order to kill, and will destroy only microorganisms such as bacteria, virus, and mold spores, not insects or other large organisms (Westinghouse Electric Company, 1976). The current recommendation is that work surfaces be wiped down with ethanol for decontamination, although some labs use the lamps in addition to ethanol. A special metering device is available to measure the output of UV lamps, and the lamps should be replaced when they fall below the minimum requirements for protection (Westinghouse Electric Company, 1976). Cultures should be checked routinely for contamination, which may be signaled by cloudiness and turbidity. Once contamination is confirmed with a microscope, infected cultures are generally discarded. Keeping contaminated cultures increases the risk of contaminating other cultures. PREPARATION OF CULTURE MEDIUM Choice of tissue culture medium comes from experience. An individual laboratory must select the medium that best suits the type of cells being cultured. Chemically defined media are available in liquid or powdered form from a number of suppliers. Sterile, ready-to-use medium has the advantage of being convenient, although it is more costly than other forms. Powdered medium must be reconstituted with tissue culture grade water according to manufacturer’s directions. Distilled or deionized water is not of sufficiently high quality for medium preparation; double- or triple-distilled water or commercially available tissue culture water should be used. The medium should be filter-sterilized and transferred to sterile bottles. Prepared medium can generally be stored 1 month in a 4°C refrigerator. Laboratories using large volumes of medium may choose to prepare their own medium from standard recipes. This is the most economical approach, but it is time-consuming. A basic medium is composed of amino acids, glucose, salts, vitamins, and other nutrients and is supplemented by addition of L-glutamine, antibiotics (typically penicillin and streptomycin sulfate), and usually serum to formulate a “complete medium.” Where serum is added, the amount is indicated as a percentage of fetal bovine serum (FBS) or other serum. Some media are also supplemented with antimycotics, nonessential amino acids, various growth factors, and/or drugs that provide selective growth conditions. Supplements should be added to medium prior to sterilization or filtration, or added aseptically just before use. The optimum pH for most mammalian cell cultures is 7.2 to 7.4. Adjust pH of the medium as necessary after all supplements are added. Buffers such as bicarbonate and HEPES are routinely used in tissue culture medium to prevent fluctuations in pH that might adversely affect cell growth. HEPES is especially useful in solutions used for procedures that do not take place in a controlled CO2 environment. Fetal bovine serum (FBS; sometimes known as fetal calf serum, FCS) is the most frequently used serum supplement. Calf serum, horse serum, and human serum are also used; some cell lines are maintained in serum-free medium (Freshney, 1993). Complete medium is supplemented with 5% to 30% (v/v) serum, depending on the requirements of the particular cell type being cultured. Serum that has been heat-inactivated (30 min to 1 hr at 56°C) is generally preferred, because heat treatment inactivates complement and is thought to reduce the number of contaminants. Serum is obtained frozen, then is thawed, divided into smaller portions, and refrozen until needed. The ability of serum to promote cell growth varies from lot to lot; most suppliers will reserve a supply of a given lot upon request.
Commonly Used Techniques
A.3B.3 Current Protocols in Cytometry
Commercially prepared media containing L-glutamine are available, but many laboratories choose to obtain medium without L-glutamine, and then add it to a final concentration of 2 mM just before use. L-glutamine is an unstable amino acid that, upon storage, converts to a form cells cannot use. Breakdown of L-glutamine is temperature- and pH-dependent. To prevent degradation, 100× L-glutamine should be stored frozen in aliquots until needed. As well as practicing good aseptic technique, most laboratories add antimicrobial agents to medium to further reduce the risk of contamination. A combination of penicillin and streptomycin is the most commonly used antibiotic additive; kanamycin and gentamycin are used alone. Mycostatin and amphotericin B are the most commonly used fungicides (Rooney and Czepulkowski, 1992). Table A.3B.1 lists the final concentrations for the most commonly used antibiotics and antimycotics. Combining antibiotics in tissue culture media can be tricky, as some antibiotics are not compatible, and one may inhibit the action of another. Furthermore, combined antibiotics may be cytotoxic at lower concentrations than is true for the individual antibiotics. In addition, prolonged use of antibiotics may cause cell lines to develop antibiotic resistance. For this reason, some laboratories add antibiotics and/or fungicides to medium when initially establishing a culture but eliminate them from medium used in later subcultures. All tissue culture medium, whether commercially prepared or prepared within the laboratory, should be tested for sterility prior to use. A small aliquot from each lot of medium is incubated 48 hr at 37°C and monitored for evidence of contamination such as turbidity (infected medium will be cloudy) and color change (if phenol red is the indicator, infected medium will turn yellow). Any contaminated medium should be discarded. BASIC PROTOCOL
TRYPSINIZING AND SUBCULTURING CELLS FROM A MONOLAYER A primary culture is grown to confluency in a 60-mm petri dish or 25-cm2 tissue culture flask containing 5 ml tissue culture medium. Cells are dispersed by trypsin treatment and then reseeded into secondary cultures. The process of removing cells from the primary culture and transferring them to secondary cultures constitutes a passage or subculture. Materials Primary cultures of cells HBSS (APPENDIX 2A) without Ca2+ and Mg2+, 37°C 0.25% (w/v) trypsin/0.2% EDTA solution (APPENDIX 2A), 37°C Complete medium with serum: e.g., DMEM supplemented with 10% to 15% (v/v) fetal bovine serum (complete DMEM-10; APPENDIX 2A), 37°C Sterile Pasteur pipets 37°C warming tray or incubator Tissue culture plasticware or glassware including pipets and 25-cm2 flasks or 60-mm petri dishes, sterile NOTE: All incubations are performed in a humidified 37°C, 5% CO2 incubator unless otherwise specified. 1. Remove all medium from primary culture with a sterile Pasteur pipet. Wash adhering cell monolayer once or twice with a small volume of 37°C HBSS without Ca2+ and Mg2+ to remove any residual FBS, which may inhibit the action of trypsin. Use a buffered salt solution that is Ca2+- and Mg2+-free to wash cells. Ca2+ and Mg2+ in the salt solution can cause cells to stick together.
Techniques for Mammalian Cell Tissue Culture
If this is the first medium change, rather than discarding medium that is removed from primary culture, put it into a fresh dish or flask. The medium contains unattached cells that may attach and grow, thereby providing a backup culture.
A.3B.4 Current Protocols in Cytometry
2. Add enough 37°C trypsin/EDTA solution to culture to cover adhering cell layer. 3. Place plate on a 37°C warming tray 1 to 2 min. Tap bottom of plate on the countertop to dislodge cells. Check culture with an inverted microscope to be sure that cells are rounded up and detached from the surface. If cells are not sufficiently detached, return plate to warming tray for an additional minute or two.
4. Add 2 ml of 37°C complete medium. Draw cell suspension into a Pasteur pipet and rinse cell layer two or three times to dissociate cells and to dislodge any remaining adherent cells. As soon as cells are detached, add serum or medium containing serum to inhibit further trypsin activity that might damage cells. If cultures are to be split 1/3 or 1/4 rather than 1/2, add sufficient medium such that 1 ml of cell suspension can be transferred into each fresh culture vessel.
5. Add an equal volume of cell suspension to fresh dishes or flasks that have been appropriately labeled. Alternatively, cells can be counted using a hemacytometer or Coulter counter (APPENDIX 3A) and diluted to the desired density so a specific number of cells can be added to each culture vessel. A final concentration of ∼5 × 104 cells/ml is appropriate for most subcultures. For primary cultures and early subcultures, 60-mm petri dishes or 25-cm2 flasks are generally used; larger petri dishes or flasks (e.g., 150-mm dishes or 75-cm2 flasks) may be used for later subcultures. Cultures should be labeled with cell line or patient name, lab number, date of subculture, and passage number.
6. Add 4 ml fresh medium to each new culture. Incubate in a humidified 37°C, 5% CO2 incubator. If using 75-cm2 culture flasks, add 9 ml medium per flask. Some labs now use incubators with 5% CO2 and 4% O2. The low oxygen concentration is thought to simulate the in vivo environment of cells and to enhance cell growth.
7. If necessary, feed subconfluent cultures after 3 or 4 days by removing old medium and adding fresh 37°C medium. 8. Passage secondary culture when it becomes confluent by repeating steps 1 to 7, and continue to passage as necessary. FREEZING HUMAN CELLS GROWN IN MONOLAYER CULTURES It is sometimes desirable to store cell lines for future study. To preserve cells, avoid senescence, reduce the risk of contamination, and minimize effects of genetic drift, cell lines may be frozen for long-term storage. Without the use of a cryoprotective agent freezing would be lethal to the cells in most cases. Generally, a cryoprotective agent such as dimethylsulfoxide (DMSO) is used in conjunction with complete medium for preserving cells at −70°C or lower. DMSO acts to reduce the freezing point and allows a slower cooling rate. Gradual freezing reduces the risk of ice crystal formation and cell damage.
SUPPORT PROTOCOL 1
Materials Log-phase monolayer culture of cells in petri dish Complete medium (e.g., DMEM or RPMI; APPENDIX 2A) Freezing medium: complete medium supplemented with 10% to 20% (v/v) FBS and 5% to 10% (v/v) DMSO, 4°C Benchtop clinical centrifuge (e.g., Fisher Centrific or Clay Adams Dynac) with 45°C fixed-angle or swinging-bucket rotor
Commonly Used Techniques
A.3B.5 Current Protocols in Cytometry
1. Trypsinize log-phase monolayer culture of cells from plate (see Basic Protocol, steps 1 to 4). It is best to use cells in log-phase growth for cryopreservation.
2. Transfer cell suspension to a sterile centrifuge tube and add 2 ml complete medium with serum. Centrifuge 5 min at 300 to 350 × g (∼1500 rpm in Fisher Centrific rotor), room temperature. Cells from three or more dishes from the same subculture of the same patient can be combined in one tube.
3. Remove supernatant and add 1 ml of 4°C freezing medium. Resuspend pellet. 4. Add 4 ml of 4°C freezing medium, mix cells thoroughly, and place on wet ice. 5. Count cells using a hemacytometer (see APPENDIX 3A). Dilute with more freezing medium as necessary to get a final cell concentration of 106 or 107 cells/ml. To freeze cells from a nearly confluent 25-cm2 flask, resuspend in roughly 3 ml freezing medium.
6. Pipet 1-ml aliquots of cell suspension into labeled 2-ml cryovials. Tighten caps on vials. 7. Place vials 1 hr to overnight in a −70°C freezer, then transfer to liquid nitrogen storage freezer. Alternatively, freeze cells in a Cryo 1°C freezing container (Nalge) according to manufacturer’s instructions. This container is designed to produce a steady 1°C/min drop in temperature and is especially useful with cell lines such as HL-60 that are sensitive to the the freeze/thaw process. Some laboratories place vials directly into the liquid nitrogen freezer, omitting the gradual temperature drop. Although this is contrary to the general recommendation to gradually reduce the temperature, laboratories that routinely use a direct-freezing technique report no loss of cell viability on recovery. Keep accurate records of the identity and location of cells stored in liquid nitrogen freezers. Cells may be stored for many years and proper information is imperative for locating a particular line for future use. ALTERNATE PROTOCOL
FREEZING CELLS GROWN IN SUSPENSION CULTURE Freezing cells from suspension culture is similar in principle to freezing cells from monolayer. The major difference is that suspension cultures need not be trypsinized. 1. Transfer cell suspension to a centrifuge tube and centrifuge 10 min at 300 to 350 × g (∼1500 rpm in Fisher Centrific centrifuge), room temperature. 2. Remove supernatant and resuspend pellet in 4°C freezing medium at a density of 106 to 107 cells/ml. Some laboratories freeze lymphoblastoid lines at the higher cell density because they plan to recover them in a larger volume of medium and because there may be a greater loss of cell viability upon recovery as compared to other types of cells (e.g., fibroblasts).
3. Transfer 1-ml aliquots of cell suspension into labeled cryovials and freeze as for monolayer cultures.
Techniques for Mammalian Cell Tissue Culture
A.3B.6 Current Protocols in Cytometry
THAWING AND RECOVERING HUMAN CELLS When cryopreserved cells are needed for study, they should be thawed rapidly and plated at high density to optimize recovery.
SUPPORT PROTOCOL 2
CAUTION: Protective clothing, particularly insulated gloves and goggles, should be worn when removing frozen vials or ampules from the liquid nitrogen freezer. The room containing the liquid nitrogen freezer should be well-ventilated. Care should be taken not to spill liquid nitrogen on the skin. Materials Cryopreserved cells stored in liquid nitrogen freezer 70% (v/v) ethanol Complete medium (e.g., DMEM or RPMI; APPENDIX 2A) containing 20% FBS (APPENDIX 2A), 37°C Tissue culture dish or flask NOTE: All incubations are performed in a humidified 37°C, 5% CO2 incubator unless otherwise specified. 1. Remove vial from liquid nitrogen freezer and immediately place it into a 37°C water bath. Agitate vial continuously until medium is thawed. The medium usually thaws in <60 sec. Cells should be thawed as quickly as possible to prevent formation of ice crystals, which can cause cell lysis. Try to avoid getting water around the cap of the vial.
2. Wipe top of vial with 70% ethanol before opening. Some labs prefer to submerge the vial in 70% ethanol and air dry before opening.
3. Transfer thawed cell suspension into a sterile centrifuge tube containing 2 ml warm complete medium containing 20% FBS. Centrifuge 10 min at 150 to 200 × g (∼1000 rpm in Fisher Centrific), room temperature. Discard supernatant. Cells are washed with fresh medium to remove residual DMSO.
4. Gently resuspend cell pellet in small amount (∼1 ml) of complete medium/20% FBS and transfer to properly labeled culture dish or flask containing the appropriate amount of medium. Cultures are reestablished at a higher cell density than that used for original cultures because there is some cell death associated with freezing. Generally, 1 ml cell suspension is reseeded in 5 to 20 ml medium.
5. Check cultures after ∼24 hr to ensure that cells have attached to the plate. 6. Change medium after 5 to 7 days or when pH indicator (e.g., phenol red) in medium changes color. Keep cultures in medium with 20% FBS until cell line is reestablished. If recovery rate is extremely low, only a subpopulation of the original culture may be growing; be extra careful of this when working with cell lines known to be mosaic.
Commonly Used Techniques
A.3B.7 Current Protocols in Cytometry
SUPPORT PROTOCOL 3
DETERMINING CELL VIABILITY BY TRYPAN BLUE STAINING Materials Cell suspension 0.4% (w/v) trypan blue prepared in HBSS (APPENDIX 2A) HBSS (APPENDIX 2A) 1. Determine number of viable cells by adding 0.5 ml of 0.4% trypan blue, 0.3 ml HBSS, and 0.1 ml cell suspension to a small tube. Mix thoroughly and let stand 5 min before loading hemacytometer and counting as described in APPENDIX 3A. 0.4% trypan blue can be used to determine the viable cell number. Nonviable cells will take up the dye, whereas live cells will be impermeable to it.
2. Count total number of cells and total number of viable (unstained) cells. Calculate percent viable cells as follows: % viable cells =
SUPPORT PROTOCOL 4
number of unstained cells × 100 total number of cells
PREPARING CELLS FOR TRANSPORT Both monolayer and suspension cultures can easily be shipped in 25-cm2 tissue culture flasks. Cells are grown to near confluency in a monolayer or to desired density in suspension. Medium is removed from monolayer cultures and the flask is filled with fresh medium. Fresh medium is added to suspension cultures to fill the flask. It is essential that the flasks be completely filled with medium to protect cells from drying if flasks are inverted during transport. The cap is tightened and taped securely in place. The flask is sealed in a leak-proof plastic bag or other leak-proof container designed to prevent spillage in the event that the flask should become damaged. The primary container is then placed in a secondary insulated container to protect it from extreme temperatures during transport. A biohazard label is affixed to the outside of the package. Generally, cultures are transported by same-day or overnight courier. Cells can also be shipped frozen. The vial containing frozen cells is removed from the liquid nitrogen freezer and placed immediately on dry ice in an insulated container to prevent thawing during transport. COMMENTARY Background Information
Techniques for Mammalian Cell Tissue Culture
At its inception in the early twentieth century, tissue culture was applied to the study of tissue fragments in culture. New growth in culture was limited to cells that migrated out from the initial tissue fragment. Tissue culture techniques evolved rapidly, and since the 1950s culture methods have allowed the growth and study of dispersed cells in culture (Freshney, 1993). Cells dispersed from the original tissue can be grown in monolayers and passaged repeatedly to give rise to a relatively stable cell line. Four distinct growth stages have been described for primary cells maintained in culture. First, cells adapt to the in vitro environment. Second, cells undergo an exponential growth
phase lasting through ∼30 passages. Third, the growth rate of cells slows, leading to a progressively longer generation time. Finally, after 40 or 50 passages, cells begin to senesce and die (Lee, 1991). It may be desirable to study a particular cell line over several months or years, so monolayer cultures can be preserved to retain the integrity of the cell line. Aliquots of early passage cell suspensions are frozen, then thawed, and cultures reestablished as needed. Freezing monolayer cultures prevents changes due to genetic drift and avoids loss of cultures due to senescence or accidental contamination (Freshney, 1993). Cell lines are commercially available from a number of sources, including the American
A.3B.8 Current Protocols in Cytometry
Type Culture Collection (ATCC) and the Human Genetic Mutant Cell Repository at the Coriell Cell Repository (CCR). These cell repositories are a valuable resource for researchers who do not have access to suitable populations.
Critical Parameters Use of aseptic technique is essential for successful tissue culture. Cell cultures can be contaminated at any time during handling, so precautions must be taken to minimize the chance of contamination. All supplies and reagents that come into contact with cultures must be sterile and all work surfaces should be kept clean and free from clutter. Cultures should be 75% to 100% confluent when selected for subculture. Growth in monolayer cultures will be adversely affected if cells are allowed to become overgrown. Passaging cells too early will result in a longer lag time before subcultures are established. Following dissociation of the monolayer by trypsinization, serum or medium containing serum should be added to the cell suspension to stop further action by trypsin that might be harmful to cells. When subculturing cells, add a sufficient number of cells to give a final concentration of ∼5 × 104 cells/ml in each new culture. Cells plated at too low a density may be inhibited or delayed in entry into growth stage. Cells plated at too high a density may reach confluence before the next scheduled subculturing; this could lead to cell loss and/or cessation of proliferation. The growth characteristics for different cell lines vary. A lower cell concentration (104 cells/ml) may be used to initiate subcultures of rapidly growing cells, and a higher cell concentration (105 cells/ml) may be used to initiate subcultures of more slowly growing cells. Adjusting the initial cell concentration permits establishment of a regular, convenient schedule for subculturing—e.g., once or twice a week (Freshney, 1993). Cells in culture will undergo changes in growth, morphology, and genetic characteristics over time. Such changes can adversely affect reproducibility of laboratory results. Nontransformed cells will undergo senescence and eventual death if passaged indefinitely. The time of senescence will vary with cell line, but generally at between 40 and 50 population doublings fibroblast cell lines begin to senesce. Cryopreservation of cell lines will protect against these adverse changes and will avoid potential contamination.
Cultures selected for cryopreservation should be in log-phase growth and free from contamination. Cells should be frozen at a concentration of 106 to 107 cells/ml. Cells should be frozen gradually and thawed rapidly to prevent formation of ice crystals that may cause cells to lyse. Cell lines can be thawed and recovered after long-term storage in liquid nitrogen. The top of the freezing vial should be cleaned with 70% alcohol before opening to prevent introduction of contaminants. To aid in recovery of cultures, thawed cells should be reseeded at a higher concentration than that used for initiating primary cultures. Careful records regarding identity and characteristics of frozen cells as well as their location in the freezer should be maintained to allow for easy retrieval. For accurate cell counting, the hemacytometer slide should be clean, dry, and free from lint, scratches, fingerprints, and watermarks. The coverslip supplied with the hemacytometer should always be used because it has an even surface and is specially designed for use with the counting chamber. Use of an ordinary coverslip may introduce errors in cell counting. If the cell suspension is too dense or the cells are clumped, inaccurate counts will be obtained. If the cell suspension is not evenly distributed over the counting chamber, the hemacytometer should be washed and reloaded.
Anticipated Results Confluent cell lines can be successfully subcultured in the vast majority of cases. The yield of cells derived from monolayer culture is directly dependent on the available surface area of the culture vessel (Freshney, 1993). Overly confluent cultures or senescent cells may be difficult to trypsinize, but increasing the time of trypsin exposure will help dissociate resistant cells. Cell lines can be propagated to get sufficient cell populations for cytogenetic, biochemical, and molecular analyses. It is well accepted that anyone can successfully freeze cultured cells; it is thawing and recovering the cultures that presents the problem. Cultures that are healthy and free from contamination can be frozen and stored indefinitely. Cells stored in liquid nitrogen can be successfully thawed and recovered in over 95% of cases. Several aliquots of each cell line should be stored to increase the chance of recovery. Cells should be frozen gradually, with a temperature drop of ∼1°C per minute, but thawed rapidly. Gradual freezing and rapid
Commonly Used Techniques
A.3B.9 Current Protocols in Cytometry
thawing prevents formation of ice crystals that might cause cell lysis. Accurate cell counts can be obtained using the hemacytometer if cells are evenly dispersed in suspension and free from clumps. Determining the proportion of viable cells in a population will aid in standardization of experimental conditions.
Time Considerations Establishment and maintenance of mammalian cell cultures require a regular routine for preparing media and feeding and passaging cells. Cultures should be inspected regularly for signs of contamination and to determine if the culture needs feeding or passaging.
Lee, E.C. 1991. Cytogenetic Analysis of Continuous Cell Lines. In The ACT Cytogenetics Laboratory Manual, 2nd ed. (M.J. Barch, ed.) pp. 107-148. Raven Press, New York. Rooney, D.E. and Czepulkowski, B.H. (eds.) 1992. Human Cytogenetics: A Practical Approach, Vol. I. Constitutional Analysis, 2nd ed. IRL Press, Washington, D.C. Westinghouse Electric Company. 1976. Westinghouse sterilamp germicidal ultraviolet tubes. Westinghouse Electric Corp., Bloomfield, NJ.
Key Reference Lee, E.C. 1991. See above. Contains pertinent information on cell culture requirements including medium preparation and sterility. Also discusses trypsinization, freezing and thawing, and cell counting.
Literature Cited Freshney, R.I. 1993. Culture of Animal Cells. A Manual of Basic Techniques, 3rd ed. Wiley-Liss, New York. Knutsen, T. 1991. In The ACT Cytogenetics Laboratory Manual, 2nd ed. (M.J. Barch, ed.) pp. 563-587. Raven Press, New York.
Contributed by Mary C. Phelan Thompson Children’s Hospital Chattanooga, Tennessee
Techniques for Mammalian Cell Tissue Culture
A.3B.10 Current Protocols in Cytometry
Diagnosis and Treatment of Mycoplasma-Contaminated Cell Cultures
APPENDIX 3C
Mycoplasma contamination is a serious and frequent problem in the culture laboratory. Although mycoplasma contamination may be suspected by the failure of cells to thrive, the formal diagnosis rests on the detection of adenosine phosphorylase secretion by infected cell lines. Basic Protocol 1 describes how to test for mycoplasma contamination, while Basic Protocol 2 and the Alternate Protocol present methods for antibiotic treatment of infected cultures. TESTING FOR MYCOPLASMA INFECTION A convenient and accurate method for testing cultured cells for infection with mycoplasma is based on the observation of McGarrity and Carson (1982) that these organisms secrete abundant adenosine phosphorylase, an enzyme capable of converting the nontoxic adenosine analog 6-methylpurine deoxyriboside (6-MPDR) into the potent antimetabolites 6-methylpurine (6-MP) and 6-methylpurine riboside (6-MPR), which are toxic to mammalian cells.
BASIC PROTOCOL 1
This method can also be used to evaluate conditioned medium for the possible presence of mycoplasma. In the instructions accompanying the MycoTect kit obtained from Life Technologies, 3T6 cells are suggested for use as the indicator cell. However, for many immunology laboratories the SP2/0-Ag14 hybridoma cell line—a widely used fusion partner for generating antibody-secreting hybridomas (Shulman et al., 1978)—is more convenient to use. In this protocol, a potential problem of differential susceptibility of cells to the toxic effects of 6-MP is avoided by culturing a highly susceptible cell line overnight with culture supernatants from the cells to be tested. Contamination is detected after addition of 6-MPDR—if growth is not observed, mycoplasma is present. Materials SP2/0-Ag14 hybridoma cells (ATCC #CRL 1581) Cells to be tested in appropriate medium without antibiotics MycoTect (Life Technologies) containing 6-methylpurine deoxyriboside (6-MPDR) 6-methylpurine (6-MP; Sigma) 96-well flat-bottom microtiter plates (Costar, Falcon, or equivalent) 1. Prepare 50-µl triplicate cultures containing 2 × 103 SP2/0-Ag14 cells (4 × 104 cells/ml) in 96-well flat-bottom microtiter plates. 2. Culture cells to be tested in medium without antibiotics in a humidified 37°C, 5% CO2 incubator for ≥24 hr. Allow cells to overgrow, remove supernatant, and add 50 µl culture supernatant to each well of the microtiter plate (from step 1). Routinely, each cell line is tested in triplicate—i.e., cells are cultured in nine microtiter wells—although this may be unnecessary. Each cell line should be tested every 2 months to be certain that contamination has not been inadvertently introduced.
3. Add 100 µl medium to first well (negative control), 100 µl 6-MPDR (40 µM final) to second well, and 100 µl 6-MP (6 µM final) to third well (positive control). Culture cells 2 to 3 days in a humidified 37°C, 5% CO2 incubator. The MycoTect kit also includes adenosine phosphorylase to be used as an additional positive control; use according to the instructions provided. Contributed by Frank W. Fitch, Thomas F. Gajewski, and Wayne M. Yokoyama Current Protocols in Cytometry (2000) A.3C.1-A.3C.4 Copyright © 2000 by John Wiley & Sons, Inc.
Commonly Used Techniques
A.3C.1 Supplement 11
4. Screen cultures visually using an inverted microscope for growth of indicator SP2/0Ag14 cells. If mycoplasma is not present, cell growth is observed in the presence of medium and 6-MPDR but not in the presence of 6-MP. If mycoplasma is present, cell growth is observed in the presence of medium but not in the presence of 6-MPDR or 6-MP (see Basic Protocol 2 and Alternate Protocol for treatment). Staining cells with crystal violet is recommended in the instructions included in the MycoTect kit, but is usually unnecessary. BASIC PROTOCOL 2
TREATMENT OF MYCOPLASMA-CONTAMINATED CELL CULTURES WITH BM-CYCLIN Until recently, the only effective treatment of mycoplasma contamination was to discard the culture. Alternatively, cell lines could be passed in vivo; e.g., by inoculation and rescue from a nude mouse. Direct treatment of mycoplasma-contaminated cell cultures has become possible with the development of BM-cyclin, an antibiotic regimen from Roche Molecular Biochemicals. More recently, the antibiotic ciprofloxacin has been successfully employed (Schmitt et al., 1988; see Alternate Protocol). Modified protocols for the use of both antibiotic regimens are given below. The BM-cyclin regimen is considerably more complicated and BM-cyclin is potentially more toxic to cells. Nevertheless, there is more experience with the efficacy of BM-cyclin over ciprofloxacin, so for now, BM-cyclin should be considered the treatment of choice. The BM-cyclin regimen utilizes two solutions. First, the maximum concentration of each solution tolerated by the cells (i.e., which allows cell growth and does not kill cells) is determined. Next, the cells are cultured in BM-cyclin solution 1 for 3 days, then solution 2 for 4 days. Finally, this cycle is repeated ≥2 times, usually resulting in mycoplasma eradication. Materials BM-cyclin solutions 1 and 2 (see recipe), sterile Contaminated cells (see Basic Protocol 1) 1. Determine the maximum amount of BM-cyclin solutions 1 and 2 that can be tolerated by the contaminated cells by setting up a titration curve for each solution. Prepare several different concentrations of each BM-cyclin solution beginning at 20 µl of each BM-cyclin solution per 10 ml cell culture and then halving the concentration for several dilutions (e.g., use 10 µl/10 ml, 5 µl/10 ml, and 2.5 µl/10 ml). Culture cells at density appropriate for that cell line for 3 days in BM-cyclin solution 1 and 4 days in BM-cyclin solution 2 in a humidified 37°C, 5% CO2 incubator. 2. Screen cultures visually using an inverted microscope to identify the BM-cyclin solution dose that does not kill cells and allows cell growth. Alternatively, an aliquot of cells can be counted using the trypan blue exclusion assay (APPENDIX 3B).
3. Culture cells 3 days in the maximum tolerable dose of BM-cyclin solution 1, then 4 days in BM-cyclin solution 2 at the same dose in a humidified 37°C, 5% CO2 incubator. This represents one cycle of treatment. If the cell density has plateaued, split into fresh growth medium plus the appropriate BM-cyclin solution at the maximal tolerated dose. Diagnosis and Treatment of MycoplasmaContaminated Cell Cultures
4. Freeze an aliquot of the treated cells (APPENDIX 3B). Freeze aliquots of treated cells after each cycle and after the treatment has been completed (step 6). This will prevent loss of the cell line to another contaminant.
A.3C.2 Supplement 11
Current Protocols in Cytometry
5. Repeat steps 3 and 4 twice. Screen cultures visually using an inverted microscope after each cycle for signs of improved cell viability and growth. Split cells as their growth rate and density dictate. Alternatively, count cells using the trypan blue exclusion assay. In most cases, three cycles are sufficient to eradicate the mycoplasma. Occasionally, additional cycles are necessary. If so, treat cells with one cycle of antibiotics beyond eradication. Chronic administration (>4 weeks) of the antibiotics is not recommended since resistant organisms may develop. Treated cells should grow more vigorously and lose symptoms of mycoplasma contamination (see Troubleshooting).
6. After completion of the BM-cyclin treatment, freeze aliquots of the mycoplasma-free cells (APPENDIX 3B). 7. Test cell line for presence of mycoplasma infection (see Basic Protocol 1). 8. Culture cells in absence of BM-cyclin and watch for recurrence of mycoplasma contamination (see Basic Protocol 1). TREATMENT OF MYCOPLASMA-CONTAMINATED CELL CULTURES WITH CIPROFLOXACIN As discussed in the previous protocol, treatment with ciprofloxacin is a more recently developed procedure than the BM-cyclin treatment. It is easier, less toxic, and less time-consuming than the BM-cyclin protocol.
ALTERNATE PROTOCOL
Additional Materials 10 mg/ml ciprofloxacin⋅HCl (see recipe), sterile 1. Add 10 mg/ml ciprofloxacin⋅HCl to contaminated cells to 10 µg/ml final (a 1:1000 dilution). Keep ciprofloxacin⋅HCl in the cultures for 12 days in a humidified, 37°C 5% CO2 incubator. 2. Freeze, test, and culture cells (see Basic Protocol 2, steps 6 to 8). REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
BM-cyclin solutions 1 and 2 Prepare and filter sterilize BM-cyclin (Roche Molecular Biochemicals) solutions 1 and 2 using a 0.45-µm filter. Freeze aliquots indefinitely at −20°C and thaw as needed. Frequent thawing and freezing of the BM-cyclin solutions does not appear to affect the activity of these antibiotics.
Ciprofloxacin⋅HCl, 10 mg/ml Dissolve ciprofloxacin⋅HCl (Sigma) in water to 10 mg/ml. Filter sterilize through a 0.45-µm filter, aliquot, and store in sterile tubes at −20°C. COMMENTARY Critical Parameters When treating cultures with BM-cyclin, the major critical parameter is that the correct dose of the drug be used (see Basic Protocol 2). In limited experience, toxicity of ciprofloxacin has not been a problem.
Another consideration is whether the desired phenotype of the cell will change during the BM-cyclin or ciprofloxacin treatment. It is possible that a monoclonal antibody–producing hybridoma will lose its secretory capacity during the relatively prolonged treatment and
Commonly Used Techniques
A.3C.3 Current Protocols in Cytometry
Supplement 21
may need to be recloned by limiting dilution (Yokoyama, 1991) during treatment.
Troubleshooting Sources of contamination The most common sources of mycoplasma contamination are previously contaminated cell lines, bovine sources (e.g., FBS), and other animal sources. The best method of eradication is prevention. FBS should be heat-inactivated before use in tissue culture. Any animals that will be used to generate cell lines or bioactive products should be purchased from specificpathogen-free (SPF) suppliers and housed in pathogen-free facilities. In general, newly derived cell lines and cell lines from other laboratories should be quarantined until the investigator is satisfied that the cultures and/or products are mycoplasma free. Suspicious cell lines should not be manipulated at the same time as “clean” cultures, as aerosol inoculation appears to be a frequent mode of cross-contamination with mycoplasma. Finally, a separate hood and tissue culture incubator should be set aside for any certified mycoplasma-free cultures. Indications of contamination Mycoplasma contamination is a frequently overlooked but potentially serious problem. The following symptoms may indicate mycoplasma contamination: 1. A once vigorous cell line grows poorly. 2. An adherent cell line is less adherent. 3. Cell line cannot be grown to high density (>1 × 106 cells/ml). 4. B or T hybridomas fail to be generated from cell fusion techniques. 5. Cell lines cannot be cloned by limiting dilution. 6. Large amount of debris is evident in the culture supernatant. Mycoplasma contamination can interfere in bioassays. The following effects may indicate mycoplasma contamination:
Diagnosis and Treatment of MycoplasmaContaminated Cell Cultures
1. Any inhibitory effect of monoclonal antibody or culture supernatant on any proliferation assay. 2. Any costimulatory effect of culture supernatants on B-cell proliferation. 3. Cell lines with higher than expected [3H]thymidine incorporation. 4. Failure of T-cell clones or hybridomas to give usual proliferation or lymphokine responses. If a functional effect of a culture supernatant is eliminated by treatment of the supernatant as
indicated below, its source may be mycoplasma contamination. The following treatments are effective in eliminating mycoplasma but obviously may have an effect on the cell viability: 1. Heat-inactivation 30 min at 56°C. 2. Filtration through a 0.1-µm filter. 3. Ultracentrifugation 24 hr at 100,000 × g.
Anticipated Results Most cultures will be relatively free of mycoplasma contamination after the BM-cyclin treatment. Whether complete eradication has been achieved should be viewed with caution as long as the cell line is used.
Time Considerations BM-cyclin treatment requires one week for each cycle. At least three cycles are required but sometimes more are necessary.
Literature Cited McGarrity, G.J. and Carson, D.A. 1982 Adenosine phosphorylase-mediated nucleoside toxicity: Application towards the detection of mycoplasmal infection in mammalian cell cultures. Exp. Cell Res. 139:199-205. Schmitt, K., Däubeaner, W., Bitter-Suermann, D., and Hadding, U. 1988. A safe and efficient method for elimination of cell culture mycoplasmas using ciprofloxacin. J. Immunol. Methods 109:17-25. Shulman, M., Wilde, C.D., and Köhler, G. 1978. A better cell line for making hybridomas secreting specific antibodies. Nature (Lond.) 276:269270. Yokoyama, W. 1991. Production of monoclonal antibodies. In Current Protocols in Immunology. (J.E. Coligan, A.M. Kruisbeek, D.M. Margulies, E.M. Shevach, and W. Strober, eds.) pp. 2.5.12.5.17. John Wiley & Sons, New York.
Key Reference McGarrity, G.J., Sarama, J., and Vanaman, V. 1979. Factors influencing microbiological assay of cell-culture Mycoplasmas. In Vitro (Rockville) 15:73-81. Classic review of mycoplasma contamination and its evaluation.
Contributed by Frank W. Fitch and Thomas F. Gajewski University of Chicago Chicago, Illinois Wayne M. Yokoyama (antibiotic treatment) University of California School of Medicine San Francisco, California
A.3C.4 Supplement 11
Current Protocols in Cytometry
Wright-Giemsa and Nonspecific Esterase Staining of Cells CYTOSPIN PREPARATION A cytospin preparation allows one to observe the morphology of the cell population being stained. The goal is to collect a representative sample of preserved cells as a randomly distributed, uncrowded monolayer.
APPENDIX 3D
BASIC PROTOCOL 1
Materials 5 × 106 cell/ml suspension of cells of interest 22% (w/v) bovine serum albumin (BSA) Cytocentrifuge (e.g., Cytospin 3; Shandon/Lipshaw) and standard microscope slides 1. Place 1 × 106 cells in the assembled sample cup of the cytocentrifuge. The recommended concentration (5 × 106 cells/ml) is for cells of “average” size. Larger cells should be at a lower concentration, while small cells such as bacteria will require a higher concentration.
2. Add four drops of 22% BSA. 3. Spin 5 min at 50 × g. The cells are now ready for staining.
WRIGHT-GIEMSA STAINING Wright-Giemsa staining is a commonly used technique to study the morphology of lymphoid cells. The method presented here is for use in staining multiple slides; however, it is easily adapted to staining a single slide (see annotation to step 2).
BASIC PROTOCOL 2
Materials 100% methanol Wright stain (Columbia Diagnostics) Working buffer (see recipe) Working Giemsa stain (see recipe) CAUTION: Methanol is toxic and can cause blindness if swallowed. 1. Make a cytospin preparation of the cells of interest (see Basic Protocol 1). Alternatively, smear cell preparation to be stained on standard microscope slides and air dry. 2. To stain the slides, successively submerge slides into containers of: 100% methanol for 1 to 2 min Wright stain for 4 min Working buffer for 4 min Working Giemsa stain for 4 min Distilled water. If only 1 to 3 slides need to be stained, a single slide procedure is preferable. In this case, overlay slides with the same solutions mentioned above using a Pasteur pipet. Remove the solutions from the slide after the indicated time interval by turning the slide on its side and touching it to absorbent paper.
3. Air dry the slides. 4. Examine stained slides under a microscope (see Anticipated Results). Contributed by Warren Strober Current Protocols in Cytometry (2000) A.3D.1-A.3D.4 Copyright © 2000 by John Wiley & Sons, Inc.
Commonly Used Techniques
A.3D.1 Supplement 11
BASIC PROTOCOL 3
NONSPECIFIC ESTERASE STAINING The nonspecific esterase stain can be used to identify cell types containing esterases that have a characteristic ability to split esters under particular conditions (Li et al., 1973). In the staining method given here, the substrate, α-naphthyl butyrate, is incubated with cells under conditions in which esterases present in monocytes/macrophages split the substrate to yield an intermediate that can be coupled with hexazotized pararosaniline to yield a colored precipitate. Thus, this staining reaction can be used to identify monocytes/macrophages in cell preparations. Materials Buffered formaldehyde-acetone fixative (see recipe), ice-cold Phosphate buffer/hexazotized pararosaniline/α-naphthyl butyrate (see recipe) Gill’s hematoxylin (Columbia Diagnostics) 1. Make a cytospin preparation of the cells of interest (see Basic Protocol 1). Alternatively, smear cell preparation to be stained on standard microscope slides and air dry. 2. Fix cells by submerging slides in ice-cold buffered formalin-acetone fixative for 30 sec. 3. Air dry fixed cells 10 to 30 min at room temperature. 4. Incubate fixed cells in 40.2 ml phosphate buffer/hexazotized pararosaniline/αnaphthyl butyrate 45 minutes at room temperature. 5. Rinse slides in distilled water. 6. Counterstain slides by immersing in Gill’s hematoxylin 5 to 10 min. 7. Wash slides in tap water 15 to 30 sec and air dry. 8. Examine stained cell preparation under a microscope; alternatively, mount smear with a suitable mounting medium (e.g., Permount). REAGENTS AND SOLUTIONS Use deionized, distilled water in all recipes and protocol steps. For common stock solutions, see APPENDIX 2A; for suppliers, see SUPPLIERS APPENDIX.
Buffered formaldehyde-acetone fixative 20 mg Na2HPO4 100 mg KH2PO4 30 ml H2O 45 ml acetone 25 ml formaldehyde Dissolve phosphate salts in water, add acetone and formaldehyde, and mix well (final pH should be 6.6). Store at 4°C.
Wright-Giemsa and Nonspecific Esterase Staining of Cells
Phosphate buffer/hexazotized pararosaniline/α-naphthyl butyrate 38 ml 0.15 M phosphate buffer, pH 6.3 (see below) 200 µl hexazotized pararosaniline (see below) 2.0 ml α-naphthyl butyrate in ethylene glycol monomethyl ether (see below) Filter the cloudy white solution before use and if necessary adjust pH to 6.3 using either solution A (to increase) or solution B (to decrease); see below for solution A and B recipes. 0.15 M phosphate buffer, pH 6.3 Solution A: 0.15 M (9.47 g/liter) Na2HPO4 Solution B: 0.15 M (20.4 g/liter) KH2PO4
A.3D.2 Supplement 11
Current Protocols in Cytometry
Mix 1 vol of solution A with 3 vol of solution B to obtain the desired total volume. Adjust pH to 6.3, if necessary, using solution A to increase or solution B to decrease. Hexazotized pararosaniline 0.1 g pararosaniline⋅HCl (Sigma) 2.0 ml H2O 0.5 ml 12 M HCl Mix the pararosaniline, water, and HCl, and dissolve the pararosaniline under hot water. Immediately before use, combine equal volumes of dissolved pararosaniline⋅HCl and 4% (w/v) sodium nitrite (prepare weekly) and mix for 1 to 2 min. α-Naphthyl butyrate in ethylene glycol monomethyl ether µl α-naphthyl butyrate (store in aliquots at –20°C in plastic tubes; allow to liquify before use) 2.0 ml ethylene glycol monomethyl ether (Fisher) Prepare immediately before use Working buffer 24 ml Wright stain 119 ml Sorensen buffer, pH 6.5 (see below) Sorensen buffer, pH 6.5 265 ml solution A: 0.15 M (9.47 g/liter) Na2HPO4 735 ml solution B: 0.15 M (20.4 g/liter) KH2PO4 Adjust pH to 6.5, if necessary, using solution A to increase or solution B to decrease. Working Giemsa stain 10 ml Giemsa stain (Columbia Diagnostics) 80 ml H2O COMMENTARY Background Information The stains used in the Wright-Giemsa protocol, the Romanowsky stains, are a mixture of methylene blue (and other closely related thiazine dyes) and eosin (Wittekind, 1979). The staining protocol is a two-stage method that allows for a more intense staining of the nuclei than would be possible if both Wright and Giemsa stains were mixed together. Esterases present in white blood cells that are capable of hydrolyzing various aliphatic and aromatic short chain esters are known as nonspecific esterases. These esterases are cell specific, provided that selective substrate and pH conditions are used. For instance, either α-naphthyl acetate or α-naphthyl butyrate is split by esterases (under acidic conditions) found only in cells of the monocyte series; nonspecific esterases in differentiated histiocytes in tissues are resistant to fluoride inhibition whereas those in macrophages are not. Thus, histiocytes can be distinguished from macrophages by addition of sodium fluoride (15 mg) to the incubation mixture. Other esterases such as chloroacetate esterases can be used to identify neutrophils and, to a lesser
extent, mast cells; in this case naphthol AS-D chloroacetate is used as the substrate.
Critical Parameters The buffer used in the Wright-Giemsa protocol must be in the pH 6.4 to 6.8 range. If it is too acidic, the stain will be too red and nuclei will be too light; if it is too basic the stain will be too blue and cytoplasmic detail will be indistinct. The stain concentration is also important because an improper concentration leads to stains that are either too pale or too intense; in either case, morphologic detail will be lost. Commercially available Romanowsky stains may vary considerably in composition, even between batches obtained from the same supplier (Marshall et al., 1975). If a stain is unsatisfactory, it may be necessary to obtain a stain source. Staining solutions have an indefinite shelf life. For staining human bone marrow specimens which are rich in fat, the time intervals for each step should be lengthened and slides exposed to the stains twice. Prolonged incubation and high pH (pH >7.0) can lead to a positive reaction due to chloroacetate esterase present in granulocytes;
Commonly Used Techniques
A.3D.3 Current Protocols in Cytometry
Supplement 11
thus, the conditions outlined must be followed more or less precisely to maintain monocyte specificity. The length of the tap water wash after counterstaining with Gill’s hematoxylin should be monitored by examination of the wet slide under a microscope. If the nuclei are still red, wash for a few more seconds; if the nuclei appear to be washed out, restain with Gill’s hematoxylin and wash for less time.
Anticipated Results On Wright-Giemsa-stained slides, red blood cells should appear tan-pink, leukocyte nuclei should appear reddish-purple, neutrophil granules should appear reddish to pink lilac, eosinophil granules should appear bright red to orange, basophil granules should appear dark reddish purple, thrombocytes should appear dark lilac, and monocytes should exhibit a slate-gray cytoplasm and many pink granules. Positive nonspecific esterase staining (presence of enzyme activity) is seen as red-brown granules in the cytoplasm of cells. Hematoxylin stains the nuclei violet.
Time Considerations For the Wright-Giemsa protocol, small numbers of slides can be stained in 10 to 12 min. For the nonspecific esterase protocol, several slides can be stained in 60 min provided all reagents have been assembled.
Literature Cited Li, C.Y., Lam, K.W., and Yam, L.T. 1973. Esterases in human leukocytes. J. Histochem. Cytochem. 21:1-12. Marshall, P.N., Bentley, S.A., and Lewis, S.M. 1975. An evaluation of some commercial Romanowsky stains. J. Clin. Pathol. (Lond.) 28:680. Wittekind, D. 1979. On the nature of Romanowsky dyes and the Romanowsky-Giemsa effect. Clin. Lab. Haemat. 1:247-262.
Contributed by Warren Strober National Institute of Allergy and Infectious Diseases Bethesda, Maryland
Wright-Giemsa and Nonspecific Esterase Staining of Cells
A.3D.4 Supplement 11
Current Protocols in Cytometry
Techniques for Bacterial Cell Culture: Media Preparation and Bacteriological Tools
APPENDIX 3E
Basic information for the culture of bacteria is presented here, using Escherichia coli as a representative organism. The techniques and recipes should be readily transferrable. The supplier of the particular bacterium being cultured will recommend an appropriate medium. Escherichia coli is a rod-shaped bacterium with a circular chromosome about 3 million base pairs (bp) long. It can grow rapidly on minimal medium that contains a carbon compound such as glucose (which serves both as a carbon source and an energy source) and salts which supply nitrogen, phosphorus, and trace metals. E. coli grows more rapidly, however, on a rich medium that provides the cells with amino acids, nucleotide precursors, vitamins, and other metabolites that the cell would otherwise have to synthesize. When E. coli is grown in liquid culture, a small number of cells are first inoculated into a container of sterile medium. After a period of time, called the lag period, the bacteria begin to divide. In rich medium, a culture of a typical strain will double in number every 20 or 30 min. This phase of exponential growth of the cells in the culture is called log phase (sometimes subdivided into early-log, middle-log, and late-log phases). Eventually the cell density increases to a point at which nutrients or oxygen become depleted from the medium, or at which waste products (such as acids) from the cells have built up to a concentration that inhibits rapid growth. At this point, which under normal laboratory conditions occurs when the culture reaches a density of 1 to 2 × 109 cells/ml, the cells stop dividing rapidly. This phase is called saturation, and a culture that has just reached this density is said to be freshly saturated. MEDIA PREPARATION Recipes are provided below for minimal liquid media, rich liquid media, solid media, top agar, and stab agar. Tryptone, yeast extract, agar (Bacto-agar), nutrient broth, and Casamino Acids are from Difco. NZ Amine A is from Hunko Sheffield (Kraft). MINIMAL MEDIA Ingredients for these media should be added to water in a 2-liter flask and heated with stirring until dissolved. The medium should then be poured into separate bottles with loosened caps and autoclaved 15 min at 15 lb/in2, 121°C. Do not add nutritional supplements or antibiotics to any medium until it has cooled to <50°C. After the bottles cool to below 40°C, the caps can be tightened and the concentrated medium stored indefinitely at room temperature. M9 medium, 5×, per liter 30 g Na2HPO4 15 g KH2PO4 5 g NH4Cl 2.5 g NaCl 15 mg CaCl2 (optional)
Commonly Used Techniques Contributed by Karen Lech and Roger Brent Current Protocols in Cytometry (2000) A.3E.1-A.3E.8 Copyright © 2000 by John Wiley & Sons, Inc.
A.3E.1 Supplement 11
M63 medium, 5× Per liter: 10 g (NH4)2SO4 68 g KH2PO4 2.5 mg FeSO4⋅7H2O Adjust to pH 7 with KOH A medium, 5× Per liter: 5 g (NH4)2SO4 22.5 g KH2PO4 52.5 g K2HPO4 2.5 g sodium citrate⋅2H2O Before they are used, concentrated media should be diluted to 1× with sterile water and the following sterile solutions added, per liter: 1 ml 1 M MgSO4⋅7H2O 10 ml 20% carbon source (sugar or glycerol) and, if required: 0.1 ml 0.5% vitamin B1 (thiamine) 5 ml 20% Casamino Acids or L amino acids to 40 µg/ml or DL amino acids to 80 µg/ml Antibiotic (see Table A.3E.1) RICH MEDIA Unless otherwise specified, rich media should be autoclaved for 25 min. Antibiotics and nutritional supplements should be added only after the solution has cooled to 50°C or below. A flask containing liquid at 50°C feels hot but can be held continuously in one’s bare hands. H medium, per liter 10 g tryptone 8 g NaCl Lambda broth, per liter 10 g tryptone 2.5 g NaCl LB medium, per liter 10 g tryptone 5 g yeast extract 5 g NaCl 1 ml 1 N NaOH The original recipe for LB medium (sometimes referred to as Luria or Lenox broth), does not contain NaOH. There are many different recipes for LB that differ only in the amount of NaOH added. Even though the pH is adjusted to near 7 with NaOH, the medium is not very highly buffered, and the pH of a culture growing in it drops as it nears saturation.
Techniques for Bacterial Cell Culture
NZC broth, per liter 10 g NZ Amine A 5 g NaCl 2 g MgCl2⋅6H2O Autoclave 30 min at 15 lb/in2, 121°C. Cool to <50°C and add 5 ml 20% Casamino Acids.
A.3E.2 Supplement 11
Current Protocols in Cytometry
Table A.3E.1
Antibiotics, Their Modes of Action, and Modes of Bacterial Resistancea
Antibioticb
Final Stock conc. Mode conc. of action (mg/ml) (µg/ml)
Ampicillinc
4
50
Chloramphenicol, in methanol
10
20
D-Cycloserine,d in
10
200
Gentamycin
10
15
Kanamycin
10
30
Kasugamycin
10
1000
5
15
34
150
Spectinomycin
10
100
Streptomycin
50
30
Tetracycline,e in 70% ethanol
12
12
0.1 M sodium phosphate buffer, pH 8
Nalidixic acid, pH to 11 with NaOH Rifampicin,e in methanol
Bacteriocidal; only kills growing E. coli; inhibits cell wall synthesis by inhibiting formation of the peptidoglycan cross-link Bacteriostatic; inhibits protein synthesis by interacting with the 50S ribosomal subunit and inhibiting the peptidyltransferase reaction Bacteriocidal; only kills growing E. coli; inhibits cell wall synthesis by preventing formation of D-alanine from L-alanine and formation of peptide bonds involving D-alanine Bacteriocidal; inhibits protein synthesis by binding to the L6 protein of the 50S ribosomal subunit
Bacteriocidal; inhibits protein synthesis; inhibits translocation and elicits miscoding
Bacteriocidal; inhibits protein synthesis by altering the methylation of the 16S RNA and thus an altered 30S ribosomal subunit Bacteriostatic; inhibits DNA synthesis by inhibiting DNA gyrase Bacteriostatic; inhibits RNA synthesis by binding to and inhibiting the β subunit of RNA polymerase; rifampicin sensitivity is dominant. Bacteriostatic; inhibits translocation of peptidyl tRNA from the A site to the P site Bacteriocidal; inhibits protein synthesis by binding to the S12 protein of the 30S ribosomal subunit and inhibiting proper translation; streptomycin sensitivity is dominant Bacteriostatic; inhibits protein synthesis by preventing binding of aminoacyl tRNA to the ribosome A site
Mode of resistance β-lactamase hydrolyzes ampicillin before it enters the cell
Chloramphenicol acetyltransferase inactivates chloramphenicol
Mutations destroy the D-alanine transport system
Aminoglycoside acetyltransferase and aminoglycoside nucleotidyltransferase inactivate gentamycin; mutations in rplF (encodes the L6 protein) prevent the gentamycin from binding Aminoglycoside phosphotransferase, also known as neomycin phosphotransferase, aminoglycoside acetyltransferase, and aminoglycoside nucleotidyltransferase; inactivates kanamycin Mutations prevent kasugamycin from binding to the ribosome; mutations decrease uptake of kasugamycin Mutations in the host DNA gyrase prevent nalidixic acid from binding Mutation in the β subunit of RNA polymerase prevents rifampicin from complexing; rifampicin resistance is recessive Mutations in rpsE (encodes the S5 protein) prevent spectinomycin from binding; spectinomycin sensitivity is dominant and resistance is recessive Aminoglycoside phosphotransferase inactivates streptomycin; mutations in rpsL (encodes the S12 protein) prevent streptomycin from binding; streptomycin resistance is recessive Active efflux of drug from cell
aData assembled from Foster (1983), Gottlieb and Shaw (1967), and Moazed and Noller (1987). bAll antibiotics should be stored at 4°C, except tetracycline, which should be stored at −20°C. All antibiotics should be dissolved in sterile distilled H O 2
unless otherwise indicated. cCarbenicillin, at the same concentration, can be used in place of ampicillin. Carbenicillin can be stored in 50% ethanol/50% water at −20°C. dD-cycloserine solutions are unstable. They should be made immediately before use. eLight-sensitive; store stock solutions and plates in the dark.
A.3E.3 Current Protocols in Cytometry
Supplement 11
Superbroth, per liter 32 g tryptone 20 g yeast extract 5 g NaCl 5 ml 1 N NaOH TB (terrific broth), per liter 12 g Bacto tryptone 24 g Bacto yeast extract 4 ml glycerol Add H2O to 900 ml and autoclave, then add to above sterile solution 100 ml of a sterile solution of 0.17 M KH2PO4 and 0.72 M K2HPO4. Tryptone broth, per liter 10 g tryptone 5 g NaCl 2× TY medium, per liter 16 g tryptone 10 g yeast extract 5 g NaCl TYGPN medium, per liter 20 g tryptone 10 g yeast extract 10 ml 80% glycerol 5 g Na2HPO4 10 g KNO3 SOLID MEDIA Liquid media can be solidified with agar. For minimal plates, dissolve the agar in water and autoclave separately from the minimal medium; autoclaving the two together will give rise to an insoluble precipitate. For rich plates, autoclave the agar together with the other ingredients of the medium. Cool the agar to about 50°C and add other ingredients if necessary. At this temperature, the medium will stay liquid indefinitely, but it will rapidly solidify if its temperature falls much below 45°C. Finally, pour the medium into sterile disposable petri dishes (plates) and allow to solidify. Freshly poured plates are wet and unable to absorb liquid spread onto them. Moreover, plates that are even slightly wet tend to exude moisture underneath bacteria streaked on them, which can cause the freshly streaked bacteria to float away. So for most applications, dry the plates by leaving them out at room temperature for 2 or 3 days, or by leaving them with the lids off for 30 min in a 37°C incubator or in a laminar flow hood. Store dry plates at 4°C, wrapped in the original bags used to package the empty plates. Minimal Plates Autoclave 15 g agar in 800 ml water for 15 min. Add sterile concentrated minimal medium and carbon source. After medium has cooled to about 50°C, add supplements and antibiotics. Pouring 32 to 40 ml medium into each plate, expect about 25 to 30 plates per liter. Techniques for Bacterial Cell Culture
A.3E.4 Supplement 11
Current Protocols in Cytometry
Rich Plates To ingredients listed below, add water to 1 liter and autoclave 25 min. Pour LB and H plates with 32 to 40 ml medium, in order to get 25 to 30 plates per liter. Pour lambda plates with about 45 ml medium for about 20 plates per liter. H plates, per liter 10 g tryptone 8 g NaCl 15 g agar Lambda plates, per liter 10 g tryptone 2.5 g NaCl 10 g agar LB plates, per liter 10 g tryptone 5 g yeast extract 5 g NaCl 1 ml 1 N NaOH 15 g agar or agarose Additives Antibiotics (if required): Ampicillin to 50 µg/ml Tetracycline to 12 µg/ml Other antibiotics, see Table A.3E.1 Galactosides (if required): Xgal to 20 µg/ml IPTG to 0.1 mM For other galactosides, see Raleigh et al., 1989.
Figure A.3E.1 Making an inoculating loop.
Commonly Used Techniques
A.3E.5 Current Protocols in Cytometry
Supplement 11
foil-covered beaker
sterile toothpicks broad side down
Figure A.3E.2 Storage of sterile toothpicks.
TOP AGAR Top agar is used to distribute phage or cells evenly in a thin layer over the surface of a plate. In a typical application, molten top agar is mixed with bacteria and the mixture poured onto a plate to make a thin layer that is allowed to solidify. This layer of cells then grows denser, forming the opaque lawn of cells. Top agar contains less agar than plates, and so stays molten for days when it is kept at 45° to 50°C. Top agarose is sometimes used when DNA is to be prepared directly from phage, and is also used when libraries are plated out to be screened by plaque lifting (Quertermous, 1996). Prepare top agar in 1-liter batches, autoclave 15 min at 15 lb/in2, 121°C to melt, cool to 50°C, swirl to mix, pour into separate 100-ml bottles, reautoclave, cool, and store at room temperature. Before use, melt the agar by heating in a water bath or microwave oven (Lech and Brent, 1988), then cool to and hold at 45° to 50°C. H top agar, per liter 10 g tryptone 8 g NaCl 7 g agar LB top agar, per liter 10 g tryptone 5 g yeast extract 5 g NaCl 7 g agar Lambda top agar, per liter 10 g tryptone 2.5 g NaCl 7 g agar
Techniques for Bacterial Cell Culture
Top agarose, per liter 10 g tryptone 8 g NaCl 6 g agarose
A.3E.6 Supplement 11
Current Protocols in Cytometry
STAB AGAR Stab agar is used for storing bacterial strains. Stab agar, per liter 10 g nutrient broth 5 g NaCl 6 g agar 10 mg cysteine⋅Cl 10 mg thymine Thymine is included so that thy− bacteria can grow. Cysteine is thought to increase the amount of time bacteria can survive in stabs.
TOOLS Inoculating Loops Inoculating loops are used to move small numbers of bacteria or phage to a plate or to a new container of liquid medium. Inoculating loops may be purchased from any general scientific supply company. However, most researchers prefer to use loops made in the laboratory. These are made by inserting both ends of a 10-in. piece of 28-G platinum wire into an inoculating loop holder (also widely available) and twirling the holder while tugging on the middle of the wire with the point of a pencil (Fig. A.3E.1). Sterilize the loop by holding it in a Bunsen burner flame until it is red hot. Cool the loop by touching it to a sterile portion of the surface of an agar plate until it stops sizzling.
Figure A.3E.3 Making a spreader.
Commonly Used Techniques
A.3E.7 Current Protocols in Cytometry
Supplement 11
Sterile Toothpicks The broad sides of flat wooden toothpicks may also be used for streaking out bacteria. Round wooden toothpicks, or the pointed ends of flat toothpicks, are sometimes used to pick individual colonies or phage plaques. To sterilize, place toothpicks in a small beaker, cover the beaker with foil, and autoclave. Alternatively, simply autoclave the whole box of toothpicks and hold them in the middle when picking them up out of the opened box. It is convenient to put used toothpicks into another smaller beaker which, when full, is covered with foil and autoclaved. Used toothpicks can be saved, reautoclaved, and used again (see Fig. A.3E.2). Spreaders Spreaders are used to distribute liquid containing bacterial cells evenly over a plate. They are made by heating and bending a piece of 4-mm glass tubing (Fig. A.3E.3). Less durable spreaders can be made from a Pasteur pipet. Before each use, sterilize the spreader by dipping the triangular part into a container of ethanol, passing the spreader through a gas flame to ignite the ethanol, and letting the flame burn out. Be careful not to ignite the ethanol in the container. Cool the spreader by touching it to the surface of an agar plate that has not yet been spread with cells. LITERATURE CITED Foster, T.J. 1983. Plasmid-determined resistance to antimicrobial drugs and toxic metal ions in bacteria. Microbiol. Rev. 47:361-409. Gottlieb, D. and Shaw, P.D. 1967. Antibiotics. I. Mechanism of Action. Springer-Verlag, New York. Lech, K. and Brent, R. 1988. Plating lambda phage to generate plaques. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 1.11.1-1.11.4. John Wiley & Sons, New York. Moazed, D. and Noller, H.F. 1987. Interaction of antibiotics with functional sites in 16S ribosomal RNA. Nature 327:389-394. Quertermous, T. 1996. Plating and transferring bacteriophage libraries. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 6.1.1-6.1.4. John Wiley & Sons, New York. Raleigh, E.A., Lech, K., and Brent, R. 1989. Selected topics from classical genetics. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 1.4.1-1.4.14. John Wiley & Sons, New York.
Contributed by Karen Lech and Roger Brent Massachusetts General Hospital and Harvard Medical School Boston, Massachusetts
Techniques for Bacterial Cell Culture
A.3E.8 Supplement 11
Current Protocols in Cytometry
Growing Bacteria in Liquid Media
APPENDIX 3F
GROWING AN OVERNIGHT CULTURE
BASIC PROTOCOL 1
Small freshly saturated cultures of bacteria are called overnights. To make an overnight, remove the cap from a sterile 16- or 18-mm culture tube. Working quickly to minimize contact of the tube with the possibly contaminated air, use a sterile pipet to transfer 5 ml of liquid medium into the tube. Inoculate the liquid with a single bacterial colony by touching a sterile inoculating loop to the colony, making certain that some of the cells have been transferred to the loop, and then dipping the loop into the liquid and shaking it a bit. Replace the tube’s cap, and place the tube on a roller drum at 60 rpm, 37°C. Grow until the culture is freshly saturated (at a density of 1 to 2 × 109 cells/ml, which typically takes at least 6 hr). GROWING LARGER CULTURES Larger cultures are generally inoculated with overnight cultures diluted 1:100. Use an Erlenmeyer or baffle flask whose volume is at least 5 times the volume of the culture. Grow the culture at 37°C with vigorous agitation (∼300 rpm) to ensure proper aeration. If it is necessary to grow a culture without shaking (for example, if the strain is temperature-sensitive for growth and no low-temperature shaker is available), then, to ensure that the cells get adequate aeration, grow the culture in an Erlenmeyer flask whose volume is at least 20 times that of the culture. MONITORING GROWTH
BASIC PROTOCOL 2
With a Count Slide Take a clean count slide (or hemacytometer) and cover it with a clean cover slip. Dip a 0.1- or 1-ml pipet into the culture medium, allow a small drop of liquid to form on the end of the pipet, and touch it lightly to the surface of the slide at the periphery of the cover slip. The liquid will quickly spread under the cover slip. Put the slide on the stage of a phase-contrast microscope set to 400×, and focus on the cells. Each cell in a small square is equivalent to 2 × 107 cells/ml (see Fig. A.3F.1).
Figure A.3F.1 Culture of E. coli at 2.4 × 108 cells/ml viewed on a count slide. Commonly Used Techniques Contributed by Karen Lech and Roger Brent Current Protocols in Cytometry (2000) A.3F.1-A.3F.2 Copyright © 2000 by John Wiley & Sons, Inc.
A.3F.1 Supplement 12
With a Spectrophotometer The concentration of cells in a culture can also be determined with a spectrophotometer by measuring the amount of 600-nm light scattered by the culture. The level of absorbance (A) at 600 nm will depend on the distance between the cuvette and the detector and will vary among spectrophotometers, often by a factor of 2. It is thus wise to calibrate each instrument by recording the OD600 (sometimes expressed as A600) of a culture that contains a known number of cells determined by some other method, such as observation → on a count slide or titering for viable colonies (APPENDIX 3G). If the culture is visibly turbid, also measure a 10-fold dilution of it. For a culture grown in rich medium, a good rule of thumb is that each 0.1 OD unit is roughly equivalent to 108 cells/ml. Calculate the number of cells/ml from whichever suspension (the undiluted or the diluted) has an OD600 <1.
Contributed by Karen Lech and Roger Brent Massachusetts General Hospital and Harvard Medical School Boston, Massachusetts
Growing Bacteria in Liquid Media
A.3F.2 Supplement 12
Current Protocols in Cytometry
Growing Bacteria on Solid Media
APPENDIX 3G
TITERING AND ISOLATING BACTERIAL COLONIES BY SERIAL DILUTIONS
BASIC PROTOCOL 1
Bacteria are grown from single colonies to ensure that each cell in a population is descended from a single founder cell, and thus to help ensure that each cell in the culture has the same genetic makeup. One way to generate single colonies is to titer a culture with serial dilutions and to pick colonies from one of the dilution plates. In this procedure, a small, measured amount of a bacterial culture is diluted into fresh liquid in another tube. A small amount of liquid is taken from this tube and diluted into another fresh tube. This process is repeated several times. Equal volumes of liquid are then taken from each of the dilution tubes and plated on petri plates. The plates are incubated overnight at 37°C; well-separated single colonies will arise on some of the dilution plates. The number of living bacteria in the culture is calculated from the number of colonies formed on the dilution plates. A typical saturated culture contains 109 cells/ml. Phage suspensions can also be titered; a concentrated phage stock might typically contain 1011 phage/ml. Titering by serial dilutions is a good way to determine the number of any kind of living organism present in a suspension. The organisms do not even need to be able to grow into colonies—i.e., the concentration of living bacteriophage in a tube can be determined by titering with serial dilutions and counting the number of plaques made when an aliquot of each dilution is plated on a lawn of phage-sensitive bacteria (see UNIT 3.3). It is sometimes useful to use smaller dilution factors. Mixing 50 µl culture into 5 ml medium will give dilutions of 100×. Mixing 100 µl into 900 µl will give dilutions of 10×. NOTE: Those protocols should be performed under aseptic conditions, using sterile materials and preferably working in a biological safety cabinet. Materials LB medium (APPENDIX 3E) LB plates (APPENDIX 3E) Sterile 16- or 18-mm-diameter culture tubes Bacterial suspension to be titered 1. Using sterile pipets, introduce 5 ml LB medium into each of three sterile culture tubes. Line the tubes up, or label them so that they can be distinguished. 2. Using a pipettor, transfer 5 µl from the bacterial suspension into the first tube of LB medium. Set the vortexer to a mild setting and agitate the tube for 5 sec. 3. Put a new sterile tip on the pipettor and transfer 5 µl from the first tube of LB medium into the second tube. Vortex the second tube. Repeat step 3 to transfer 5 µl from the second tube to the third tube. The first dilution tube now contains a 103-fold dilution, generated by diluting the culture by a factor of one thousand (i.e., it contains 10−3 as many cells/ml as were present in the original culture). The second tube contains a 106-fold dilution, generated by diluting the original culture by a factor of one million (i.e., it contains 10−6 as many cells/ml as the original culture), etc. Many investigators prefer to perform serial dilutions with different volumes and different factors of dilution. These parameters can be modified in steps 1 to 3.
Contributed by Karen Lech and Roger Brent Current Protocols in Cytometry (2000) A.3G.1-A.3G.5 Copyright © 2000 by John Wiley & Sons, Inc.
Commonly Used Techniques
A.3G.1 Supplement 12
4. Spread 100 µl of liquid from the culture and from each dilution tube onto separate, labeled, dry LB plates (see Basic Protocol 3). Incubate overnight at 37°C. During this incubation, each living bacterial cell will grow into a separate colony on the plate.
5. Count the colonies from these plates. Since only 100 ìl was plated from the undiluted culture and from each dilution tube, each plate has 1⁄10 as many colonies on it as were present in each milliliter of liquid in the corresponding tube. Therefore, one can determine the number of cells per milliliter in the original culture by counting the number of colonies on a plate, and then multiplying that number by 10 times the factor of dilution. For example, if 22 colonies were observed on the plate corresponding to the 106-fold dilution, then the number of living cells in each milliliter of the original culture was 22 × 10 × 106, or 2.2 × 108 cells/ml.
6. Store plates at 4°C wrapped in Parafilm or in the plastic sleeve in which the plates were supplied. Any of the single colonies may be saved for further use. BASIC PROTOCOL 2
ISOLATING SINGLE COLONIES BY STREAKING A PLATE
BASIC PROTOCOL 3
ISOLATING SINGLE COLONIES BY SPREADING A PLATE
Another way to isolate single colonies is called streaking or streaking for single colonies. This method is easier and faster than serial dilutions for isolating single colonies, but it cannot be used to count the number of cells in a culture. An inoculum of bacteria is streaked across one side of an agar plate with an inoculating loop or sterile toothpick. The resterilized loop or a fresh toothpick is then passed once through the first streak and streaked across a fresh part of the plate (see Fig. A.3G.1). This process is repeated at least once more, and the plate is incubated at 37°C until colonies become visible. If single colonies must be isolated from many bacteria, it is convenient to divide a plate into 4, 6, or 8 sectors and to streak for single colonies in each sector.
It is sometimes necessary to distribute a liquid culture of bacteria evenly over the surface of a plate. This is usually done with a glass spreader. From 0.05 ml to 1 ml of liquid is pipetted onto a dry plate (see APPENDIX 3E) and spread using a circular motion as shown in Figure A.3G.2. Alternatively, the edge of the spreader can be used to make a raster pattern on the plate’s surface. The plate can be turned at right angles and the process repeated. Evenly spread plates should be placed in the incubator with the lids ajar until they are completely dry.
Figure A.3G.1 Streaking a plate.
Growing Bacteria on Solid Media
A.3G.2 Supplement 12
Current Protocols in Cytometry
Figure A.3G.2 Spreading a plate.
REPLICA PLATING Replica plating is a convenient way to test many colonies for their ability to grow under different conditions. In this technique, bacterial colonies are transferred from one plate to another in a way that maintains the original pattern of colonies. This technique has many applications to recombinant DNA work. As an example, consider the plasmid pBR322, which contains two antibiotic resistance genes, encoding resistance to ampicillin and tetracycline. A piece of foreign DNA cloned into the tetracycline resistance gene inactivates it; cells that carry such a plasmid are ampicillin resistant but tetracycline sensitive. These cells can be identified by replica-plating colonies from ampicillin-containing master plates onto plates containing tetracycline. Tetracycline-sensitive colonies can be identified by their inability to grow on the tetracycline plates, rescued from the master plate, and analyzed further.
SUPPORT PROTOCOL 1
This procedure requires two specialty items: a replica block and sterile velvets. The replica block is a wooden or metal cylinder that fits snugly inside a petri plate (see Fig A.3G.3). One method for constructing these has been described by Adams (1965). A metal ring is used to secure the velvets to the block. Squares of velvet should be cut so as to cover the base (a diameter of 14 cm is suggested). These velvets can be washed, autoclaved, and reused. If velvets are not available, pieces of sterile filter paper or disposable replica plates can be used. Replica plating also requires a master plate composed of well-separated colonies. The master plate can be a fresh plate onto which 50 to 100 colonies have been gridded (using toothpicks and the grid in Fig. A.3G.4), or it can be a plate on which were spread bacteria that have now grown up into well-separated colonies. Mark the top of the master plate to enable alignment with the grid. Press the plate down lightly onto the velvet. Do not bear down hard on the plate; pressing too hard will cause the colonies to run together on the velvet or may even cause the plate to collapse. Press new plates, oriented like the master plate, lightly onto the imprinted velvet to transfer the colonies. As many as 10 plates per velvet can often be replica plated. STRAIN STORAGE AND REVIVAL Most strains of E. coli can be stored for years in stab vials, or indefinitely if frozen at −70°C. It is prudent to check the genetic markers of a strain revived from storage. Ways to verify the presence of other selective markers are described in Raleigh et al. (1989).
SUPPORT PROTOCOL 2
Commonly Used Techniques
A.3G.3 Current Protocols in Cytometry
Supplement 12
Figure A.3G.3 Replica block with velvet.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Figure A.3G.4 Grid for replica plating.
Stabs
Growing Bacteria on Solid Media
Use airtight, autoclavable vials with rubber or Teflon caps (not cardboard). These are available from Wheaton Glassware and John’s Scientific (1⁄4-oz. Bijoux bottles, #15690001). Fill the vials 2⁄3 full with stab agar (see recipe, APPENDIX 3E). Inoculate them with a single colony (see Fig. A.3G.5) by collecting most of the cells in the colony with an inoculating loop, then repeatedly poking the loop deeply into the agar. Leave the cap of the stab vial slightly loose and incubate 8 to 12 hr at 37°C, or until cloudy tracks of bacterial growth are evident. Seal the vials tightly and store them in a cool (15° to 22°C), dark place. To revive a stored strain, flame sterilize an inoculating loop (APPENDIX 3E), allow it to cool, insert it into the stab agar, and move the loop around until a gobbet of bacteria-laden agar is stuck onto the loop. Smear the gobbet onto one section of an LB plate and streak for single colonies (Fig. A.3G.1).
A.3G.4 Supplement 12
Current Protocols in Cytometry
Figure A.3G.5 Inoculating a stab vial.
Frozen Stocks Add 2 ml of a mid-log culture or 1 ml of a freshly saturated culture to a stab vial or a Nunc vial (Nunc #1087) containing 1 ml glycerol solution or DMSO solution. Vials can be stored at −20° to −70°C, but most strains remain viable longer if stored at −70°C. Revive stored cells by scraping off splinters of solid ice with a toothpick or sterile pipet and streaking these splinters onto an LB plate. Do not allow the contents of the vial to thaw. Glycerol solution 65% glycerol (vol/vol) 0.1 M MgSO4 0.025 M Tris⋅Cl, pH 8 (APPENDIX 2A) DMSO solution 7% dimethylsulfoxide (vol/vol) The only advantage DMSO seems to have over glycerol for frozen stocks is that it is easier to pipet because it is less viscous. Use a bottle of reagent- or spectrophotometric-grade DMSO that has been kept tightly sealed.
LITERATURE CITED Adams, J.N. 1965. Automotive pistons for use as bases in velveteen replication. J. Bacteriol. 89:1627. Raleigh, E.A., Lech, K., and Brent, R. 1989. Selected topics from classical bacterial genetics. In Current Protocols in Molecular Biology (F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds.) pp. 1.4.1-1.4.13. John Wiley & Sons, New York.
Contributed by Karen Lech and Roger Brent Massachusetts General Hospital and Harvard Medical School Boston, Massachusetts
Commonly Used Techniques
A.3G.5 Current Protocols in Cytometry
Supplement 12
Importing Biological Materials Scientific cooperation between laboratories in the United States is no longer confined to those at the same institution, nor even within the boundaries of the country. More than ever, scientists are forming collaborations with their peers outside the U.S., leading not only to the exchange of knowledge and ideas, but also to the exchange of biological reagents. Thus, investigators need be aware that an import permit may be required to obtain purified proteins, chemically synthesized materials, cell lines, tissues, and other biological materials from abroad. Before requesting any biological material from a laboratory outside the U.S., it is the responsibility of the “receiver” to ascertain whether an import permit is required, to apply for the permit if one is determined to be necessary, and to provide the “shipper” with an official copy of the permit to be included with the shipped material. There are two government agencies in place to issue permits for the importation of most biologicals. All materials derived from animals are subject to regulations set forth by the United States Department of Agriculture (USDA) and must be cleared by USDA inspectors at the port of arrival before entry into the United States is authorized. In general, an import permit issued by the USDA is necessary for receipt of any animal-derived material or any biological material (including proteins) that has been in contact with materials of animal origin. The USDA does not have regulatory authority over the importation of human or nonhuman primate material unless it is produced via tissue culture, due to the use of bovine material (fetal bovine serum or calf serum) in tissue culture. The U.S. Public Health Service (UPHS) has jurisdiction over human and nonhuman primate materials. According to UPHS regulations, no etiologic agent, animal host, or vector of human disease may be imported into the United States or distributed after importation without being accompanied by a permit issued by the Director of the Centers for Disease Control. Failure to obtain the appropriate import permit will result in the shipment’s being held at the port of entry until a permit is applied for and issued (a 4- to 6-week process) or may result in shipment confiscation and destruction by the quarantine officer present.
APPENDIX 3H
IMPORTING RESTRICTED BIOLOGICAL MATERIALS The following is a list of restricted biological materials requiring an import permit: proteins, monoclonal and polyclonal antibodies, antisera, immunoglobins, recombinant products, enzymes, hormones, immunoassay components or kits, animal tissues, blood, cells, cell lines, RNA/DNA extracts, plasmids, vectors, microorganisms—e.g., fungi, bacteria, viruses—and other products that are derived from animals or have been in contact with animalderived materials. In addition, materials received from countries shown not to be free of foot-and-mouth disease (see Title 9, code of Federal Regulations, Part 94.1) require an import permit. Note that the USDA will not permit the importation of cell cultures, monoclonal antibodies, ascites fluid, or bovine serum from countries where rinderpest and foot-and-mouth disease are present unless the imported materials are determined to be virus-free. Cell cultures being imported into the U.S. must have been grown with fetal bovine serum only of U.S., Canadian, New Zealand, or Australian origin. Material exposed to bovine products originating in the United Kingdom, Switzerland, Ireland, France, or Portugal may NOT be used, either directly or indirectly, in any animals due to the presence of bovine spongiform encephalopathy (BSE) in those countries, and the material is restricted to in vitro uses only.
CRITERIA FOR HANDLING IMPORTED RESTRICTED MATERIALS An import permit is valid only for transfer of the material(s) listed on the permit (a permit can cover more than one item), and for shipment between the two laboratories listed on the permit. Therefore, for each laboratory abroad from which biological materials are to be imported, a separate import permit is required. Once biological materials have been received from abroad, the import permit still imposes regulations on how the materials are handled by the receiver. Import permits do not usually authorize direct or indirect exposure of domestic animals—work is limited to in vitro uses only. In addition, permits are valid only for work conducted or directed by the permit holder at his or her facilities. Materials cannot Commonly Used Techniques
Contributed by Paula Wolf Current Protocols in Cytometry (2001) A.3H.1-A.3H.3 Copyright © 2001 by John Wiley & Sons, Inc.
A.3H.1 Supplement 15
be removed to another location, nor distributed to others, without USDA authorization. There is one exception: monoclonal antibodies produced by hybridoma cell lines may be commercially distributed for in vitro uses.
IMPORTING NONRESTRICTED BIOLOGICAL MATERIALS Any of the restricted materials that do not contain animal- or cell culture–derived products or additives such as albumin, serum, or gelatin, and were not exposed to any infectious agents of agricultural concern, do not need an official import permit. However, an accompanying declaration is required with each shipment to indicate that the material does not contain any animal- or cell culture–derived products or additives. This information should be supplied with each shipment in a clear and concise manner, and be available for the USDA Port of Entry Inspector’s review. A separate memo or letter, such as U.S. Customs declaration and invoice, should be included with the shipping documents. Note that whether a particular biological material requires an import permit cannot be decided by the investigator; the investigator must contact the USDA for a decision well in advance of making arrangements for importing the material.
PACKAGING BIOLOGICAL MATERIALS FOR IMPORTATION
Importing Biological Materials
Imported biological materials are subject to packaging and shipping requirements of various federal and international regulations. Proper packaging is the primary consideration and of utmost importance in the safe transportation of hazardous materials. If the biological material being shipped is infectious or radioactive, this information has to be noted on the outside of the package. In addition, if dry ice is used for shipment, that must be clearly stated on the package. It is of course advantageous to the investigator that the shipped material be packaged in the most efficient way possible, as well as in a form which will ensure its stability. If proteins are to be shipped, it may be better to send the samples lyophilized or blotted on a membrane, rather than in liquid form. This will eliminate the need for dry ice or ice packs in the packaging, which will save on shipping costs as well as ensure sample stability if receipt of the package meets with unforeseen delays. If sending materials that require dry ice or ice packs, such as antibodies or antisera, there must be enough dry ice to allow the package to maintain the appropriate temperature during
shipping for approximately 5 days. This requires utilizing much larger packaging boxes than are normally used to ship the same material within the United States. When shipping cell lines, sometimes it is better to send cells in culture, rather than frozen aliquots. To send cells in culture, fill the tissue culture flask containing the live cells to the very top with the appropriate medium, cap tightly, and secure the seal with Parafilm. If the tissue culture flask is not completely filled with medium, the resulting air pockets will cause the flasks to explode in flight. Thus, the packaging of any biological material for shipping is an import consideration to ensure success of the scientific collaboration!
OBTAINING AN IMPORT PERMIT APPLICATION The Division of Veterinary Services of the USDA Animal and Plant Health Inspection Service (APHIS) administers regulatory programs to control the import/export of biological materials. Import permits can be obtained from either local USDA port offices or from APHIS directly: USDA, APHIS, VETERINARY SERVICES, National Center for Import-Export Products Program, 4700 River Road Unit 40, Riverdale, MD 20737-1231, Tel: (301) 7347830, Fax: (301) 734-8226. When submitting an import application, (1) obtain the import permit well in advance of the proposed shipping date; (2) inform the supplier of the USDA’s requirements, and do not allow shipment of materials until an import permit has been issued and the supplier has received a copy of the permit; (3) list all the potential U.S. ports of entry; (4) specify whether the material is for in vitro or in vivo use and whether it is for commercial distribution or for research in one’s own laboratory; (5) list any treatments the material has undergone, such as processing and purification steps involving pH, heat, chromatography, or other methods; and (6) indicate the source(s) of nutrient factors in culture media, cell line designation and history, whether enzymes were used, and what types of viruses (if any) are being studied in the laboratory of origin (all must be listed).
RESOURCES APHIS Web Site More information regarding APHIS programs and import/export can be obtained on their internet home page at http://www.aphis.usda.gov.
A.3H.2 Supplement 15
Current Protocols in Cytometry
Fax Service of Veterinary Services, National Center for Import-Export By calling (301) 734-4952, the investigator can access an automated document retrieval system and arrange to have an import application faxed directly, and choose from a menu of documents that describe general instructions for completing the application, as well as information regarding the importation of biological materials from 18 distinct categories.
Contributed by Paula Wolf Massachusetts Institute of Technology Cambridge, Massachusetts
Commonly Used Techniques
A.3H.3 Current Protocols in Cytometry
Supplement 15
SELECTED SUPPLIERS OF REAGENTS AND EQUIPMENT Listed below are addresses and phone numbers of commercial suppliers who have been recommended for particular items used in our manuals because: (1) the particular brand has actually been found to be of superior quality, or (2) the item is difficult to find in the marketplace. Consequently, this compilation may not include some important vendors of biological supplies. For comprehensive listings, see Linscott’s Directory of Immunological and Biological Reagents (Santa Rosa, CA), The Biotechnology Directory (Stockton Press, New York), the annual Buyers’ Guide supplement to the journal Bio/Technology, as well as various sites on the Internet. A.C. Daniels 72-80 Akeman Street Tring, Hertfordshire, HP23 6AJ, UK (44) 1442 826881 FAX: (44) 1442 826880 A.D. Instruments 5111 Nations Crossing Road #8 Suite 2 Charlotte, NC 28217 (704) 522-8415 FAX: (704) 527-5005 http://www.us.endress.com A.J. Buck 11407 Cronhill Drive Owings Mill, MD 21117 (800) 638-8673 FAX: (410) 581-1809 (410) 581-1800 http://www.ajbuck.com A.M. Systems 131 Business Park Loop P.O. Box 850 Carlsborg, WA 98324 (800) 426-1306 FAX: (360) 683-3525 (360) 683-8300 http://www.a-msystems.com Aaron Medical Industries 7100 30th Avenue North St. Petersburg, FL 33710 (727) 384-2323 FAX: (727) 347-9144 http://www.aaronmed.com Abbott Laboratories 100 Abbott Park Road Abbott Park, IL 60064 (800) 323-9100 FAX: (847) 938-7424 http://www.abbott.com ABCO Dealers 55 Church Street Central Plaza Lowell, MA 01852 (800) 462-3326 (978) 459-6101 http://www.lomedco.com/abco.htm Aber Instruments 5 Science Park Aberystwyth, Wales SY23 3AH, UK (44) 1970 636300 FAX: (44) 1970 615455 http://www.aber-instruments.co.uk ABI Biotechnologies See Perkin-Elmer ABI Biotechnology See Apotex
Access Technologies Subsidiary of Norfolk Medical 7350 N. Ridgeway Skokie, IL 60076 (877) 674-7131 FAX: (847) 674-7066 (847) 674-7131 http://www.norfolkaccess.com
Adaptive Biosystems 15 Ribocon Way Progress Park Luton, Bedsfordshire LU4 9UR, UK (44)1 582-597676 FAX: (44)1 582-581495 http://www.adaptive.co.uk
Accurate Chemical and Scientific 300 Shames Drive Westbury, NY 11590 (800) 645-6264 FAX: (516) 997-4948 (516) 333-2221 http://www.accuratechemical.com
Adobe Systems 1585 Charleston Road P.O. Box 7900 Mountain View, CA 94039 (800) 833-6687 FAX: (415) 961-3769 (415) 961-4400 http://www.adobe.com
AccuScan Instruments 5090 Trabue Road Columbus, OH 43228 (800) 822-1344 FAX: (614) 878-3560 (614) 878-6644 http://www.accuscan-usa.com AccuStandard 125 Market Street New Haven, CT 06513 (800) 442-5290 FAX: (877) 786-5287 http://www.accustandard.com Ace Glass 1430 NW Boulevard Vineland, NJ 08360 (800) 223-4524 FAX: (800) 543-6752 (609) 692-3333 ACO Pacific 2604 Read Avenue Belmont, CA 94002 (650) 595-8588 FAX: (650) 591-2891 http://www.acopacific.com Acros Organic See Fisher Scientific Action Scientific P.O. Box 1369 Carolina Beach, NC 28428 (910) 458-0401 FAX: (910) 458-0407 AD Instruments 1949 Landings Drive Mountain View, CA 94043 (888) 965-6040 FAX: (650) 965-9293 (650) 965-9292 http://www.adinstruments.com
Advanced Bioscience Resources 1516 Oak Street, Suite 303 Alameda, CA 94501 (510) 865-5872 FAX: (510) 865-4090 Advanced Biotechnologies 9108 Guilford Road Columbia, MD 21046 (800) 426-0764 FAX: (301) 497-9773 (301) 470-3220 http://www.abionline.com Advanced ChemTech 5609 Fern Valley Road Louisville, KY 40228 (502) 969-0000 http://www.peptide.com Advanced Machlning and Tooling 9850 Businesspark Avenue San Diego, CA 92131 (858) 530-0751 FAX: (858) 530-0611 http://www.amtmfg.com Advanced Magnetics See PerSeptive Biosystems Advanced Process Supply See Naz-Dar-KC Chicago Advanced Separation Technologies 37 Leslie Court P.O. Box 297 Whippany, NJ 07981 (973) 428-9080 FAX: (973) 428-0152 http://www.astecusa.com Advanced Targeting Systems 11175-A Flintkote Avenue San Diego, CA 92121 (877) 889-2288 FAX: (858) 642-1989 (858) 642-1988 http://www.ATSbio.com
Advent Research Materials Eynsham, Oxford OX29 4JA, UK (44) 1865-884440 FAX: (44) 1865-84460 http://www.advent-rm.com Advet Industrivagen 24 S-972 54 Lulea, Sweden (46) 0920-211887 FAX: (46) 0920-13773 Aesculap 1000 Gateway Boulevard South San Francisco, CA 94080 (800) 282-9000 http://www.aesculap.com Affinity Chromatography 307 Huntingdon Road Girton, Cambridge CB3 OJX, UK (44) 1223 277192 FAX: (44) 1223 277502 http://www.affinity-chrom.com Affinity Sensors See Labsystems Affinity Sensors Affymetrix 3380 Central Expressway Santa Clara, CA 95051 (408) 731-5000 FAX: (408) 481-0422 (800) 362-2447 http://www.affymetrix.com Agar Scientific 66a Cambridge Road Stansted CM24 8DA, UK (44) 1279-813-519 FAX: (44) 1279-815-106 http://www.agarscientific.com A/G Technology 101 Hampton Avenue Needham, MA 02494 (800) AGT-2535 FAX: (781) 449-5786 (781) 449-5774 http://www.agtech.com Agen Biomedical Limited 11 Durbell Street P.O. Box 391 Acacia Ridge 4110 Brisbane, Australia 61-7-3370-6300 FAX: 61-7-3370-6370 http://www.agen.com
Suppliers
1 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Agilent Technologies 395 Page Mill Road P.O. Box 10395 Palo Alto, CA 94306 (650) 752-5000 http://www.agilent.com/chem
Aldrich Chemical P.O. Box 2060 Milwaukee, WI 53201 (800) 558-9160 FAX: (800) 962-9591 (414) 273-3850 FAX: (414) 273-4979 http://www.aldrich.sial.com
Alpha Innotech 14743 Catalina Street San Leandro, CA 94577 (800) 795-5556 FAX: (510) 483-3227 (510) 483-9620 http://www.alphainnotech.com
Agouron Pharmaceuticals 10350 N. Torrey Pines Road La Jolla, CA 92037 (858) 622-3000 FAX: (858) 622-3298 http://www.agouron.com
Alexis Biochemicals 6181 Cornerstone Court East, Suite 103 San Diego, CA 92121 (800) 900-0065 FAX: (858) 658-9224 (858) 658-0065 http://www.alexis-corp.com
Altec Plastics 116 B Street Boston, MA 02127 (800) 477-8196 FAX: (617) 269-8484 (617) 269-1400
Agracetus 8520 University Green Middleton, WI 53562 (608) 836-7300 FAX: (608) 836-9710 http://www.monsanto.com AIDS Research and Reference Reagent Program U.S. Department of Health and Human Services 625 Lofstrand Lane Rockville, MD 20850 (301) 340-0245 FAX: (301) 340-9245 http://www.aidsreagent.org AIN Plastics 249 East Sanford Boulevard P.O. Box 151 Mt. Vernon, NY 10550 (914) 668-6800 FAX: (914) 668-8820 http://www.tincna.com Air Products and Chemicals 7201 Hamilton Boulevard Allentown, PA 18195 (800) 345-3148 FAX: (610) 481-4381 (610) 481-6799 http://www.airproducts.com ALA Scientific Instruments 1100 Shames Drive Westbury, NY 11590 (516) 997-5780 FAX: (516) 997-0528 http://www.alascience.com Aladin Enterprises 1255 23rd Avenue San Francisco, CA 94122 (415) 468-0433 FAX: (415) 468-5607 Aladdin Systems 165 Westridge Drive Watsonville, CA 95076 (831) 761-6200 FAX: (831) 761-6206 http://www.aladdinsys.com Alcide 8561 154th Avenue NE Redmond, WA 98052 (800) 543-2133 FAX: (425) 861-0173 (425) 882-2555 http://www.alcide.com Aldevron 3233 15th Street, South Fargo, ND 58104 (877) Pure-DNA FAX: (701) 280-1642 701 297-9256 http://www.aldevion.com
Alfa Aesar 30 Bond Street Ward Hill, MA 10835 (800) 343-0660 FAX: (800) 322-4757 (978) 521-6300 FAX: (978) 521-6350 http://www.alfa.com Alfa Laval Avenue de Ble 5 - Bazellaan 5 BE-1140 Brussels, Belgium 32(2) 728 3811 FAX: 32(2) 728 3917 or 32(2) 728 3985 http://www.alfalaval.com Alice King Chatham Medical Arts 11915-17 Inglewood Avenue Hawthorne, CA 90250 (310) 970-1834 FAX: (310) 970-0121 (310) 970-1063 Allegiance Healthcare 800-964-5227 http://www.allegiance.net Allelix Biopharmaceuticals 6850 Gorway Drive Mississauga, Ontario L4V 1V7 Canada (905) 677-0831 FAX: (905) 677-9595 http://www.allelix.com Allentown Caging Equipment Route 526, P.O. Box 698 Allentown, NJ 08501 (800) 762-CAGE FAX: (609) 259-0449 (609) 259-7951 http://www.acecaging.com Alltech Associates Applied Science Labs 2051 Waukegan Road P.O. Box 23 Deerfield, IL 60015 (800) 255-8324 FAX: (847) 948-1078 (847) 948-8600 http://www.alltechweb.com Alomone Labs HaMarpeh 5 P.O. Box 4287 Jerusalem 91042, Israel 972-2-587-2202 FAX: 972-2-587-1101 US: (800) 791-3904 FAX: (800) 791-3912 http://www.alomone.com
Alza 1900 Charleston Road P.O. Box 7210 Mountain View, CA 94043 (800) 692-2990 FAX: (650) 564-7070 (650) 564-5000 http://www.alza.com Alzet c/o Durect Corporation P.O. Box 530 10240 Bubo Road Cupertino, CA 95015 (800) 692-2990 (408) 367-4036 FAX: (408) 865-1406 http://www.alzet.com Amac 160B Larrabee Road Westbrook, ME 04092 (800) 458-5060 FAX: (207) 854-0116 (207) 854-0426 Amaresco 30175 Solon Industrial Parkway Solon, Ohio 44139 (800) 366-1313 FAX: (440) 349-1182 (440) 349-1313
American Cyanamid P.O. Box 400 Princeton, NJ 08543 (609) 799-0400 FAX: (609) 275-3502 http://www.cyanamid.com American HistoLabs 7605-F Airpark Road Gaithersburg, MD 20879 (301) 330-1200 FAX: (301) 330-6059 American International Chemical 17 Strathmore Road Natick, MA 01760 (800) 238-0001 (508) 655-5805 http://www.aicma.com American Laboratory Supply See American Bioanalytical American Medical Systems 10700 Bren Road West Minnetonka, MN 55343 (800) 328-3881 FAX: (612) 930-6654 (612) 933-4666 http://www.visitams.com American Qualex 920-A Calle Negocio San Clemente, CA 92673 (949) 492-8298 FAX: (949) 492-6790 http://www.americanqualex.com American Radiolabeled Chemicals 11624 Bowling Green St. Louis, MO 63146 (800) 331-6661 FAX: (800) 999-9925 (314) 991-4545 FAX: (314) 991-4692 http://www.arc-inc.com American Scientific Products See VWR Scientific Products
Ambion 2130 Woodward Street, Suite 200 Austin, TX 78744 (800) 888-8804 FAX: (512) 651-0190 (512) 651-0200 http://www.ambion.com
American Society for Histocompatibility and Immunogenetics P.O. Box 15804 Lenexa, KS 66285 (913) 541-0009 FAX: (913) 541-0156 http://www.swmed.edu/home pages/ ASHI/ashi.htm
American Association of Blood Banks College of American Pathologists 325 Waukegan Road Northfield, IL 60093 (800) 323-4040 FAX: (847) 8166 (847) 832-7000 http://www.cap.org
American Type Culture Collection (ATCC) 10801 University Boulevard Manassas, VA 20110 (800) 638-6597 FAX: (703) 365-2750 (703) 365-2700 http://www.atcc.org
American Bio-Technologies See Intracel Corporation American Bioanalytical 15 Erie Drive Natick, MA 01760 (800) 443-0600 FAX: (508) 655-2754 (508) 655-4336 http://www.americanbio.com
Amersham See Amersham Pharmacia Biotech Amersham International Amersham Place Little Chalfont, Buckinghamshire HP7 9NA, UK (44) 1494-544100 FAX: (44) 1494-544350 http://www.apbiotech.com
Suppliers
2 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Amersham Medi-Physics Also see Nycomed Amersham 3350 North Ridge Avenue Arlington Heights, IL 60004 (800) 292-8514 FAX: (800) 807-2382 http://www.nycomed-amersham.com Amersham Pharmacia Biotech 800 Centennial Avenue P.O. Box 1327 Piscataway, NJ 08855 (800) 526-3593 FAX: (877) 295-8102 (732) 457-8000 http://www.apbiotech.com Amgen 1 Amgen Center Drive Thousand Oaks, CA 91320 (800) 926-4369 FAX: (805) 498-9377 (805) 447-5725 http://www.amgen.com Amicon Scientific Systems Division 72 Cherry Hill Drive Beverly, MA 01915 (800) 426-4266 FAX: (978) 777-6204 (978) 777-3622 http://www.amicon.com Amika 8980F Route 108 Oakland Center Columbia, MD 21045 (800) 547-6766 FAX: (410) 997-7104 (410) 997-0100 http://www.amika.com Amoco Performance Products See BPAmoco AMPI See Pacer Scientific Amrad 576 Swan Street Richmond, Victoria 3121, Australia 613-9208-4000 FAX: 613-9208-4350 http://www.amrad.com.au Amresco 30175 Solon Industrial Parkway Solon, OH 44139 (800) 829-2805 FAX: (440) 349-1182 (440) 349-1199 Anachemia Chemicals 3 Lincoln Boulevard Rouses Point, NY 12979 (800) 323-1414 FAX: (518) 462-1952 (518) 462-1066 http://www.anachemia.com Ana-Gen Technologies 4015 Fabian Way Palo Alto, CA 94303 (800) 654-4671 FAX: (650) 494-3893 (650) 494-3894 http://www.ana-gen.com
Analox Instruments USA P.O. Box 208 Lunenburg, MA 01462 (978) 582-9368 FAX: (978) 582-9588 http://www.analox.com Analytical Biological Services Cornell Business Park 701-4 Wilmington, DE 19801 (800) 391-2391 FAX: (302) 654-8046 (302) 654-4492 http://www.ABSbioreagents.com Analytical Genetics Testing Center 7808 Cherry Creek S. Drive, Suite 201 Denver, CO 80231 (800) 204-4721 FAX: (303) 750-2171 (303) 750-2023 http://www.geneticid.com AnaSpec 2149 O’Toole Avenue, Suite F San Jose, CA 95131 (800) 452-5530 FAX: (408) 452-5059 (408) 452-5055 http://www.anaspec.com Ancare 2647 Grand Avenue P.O. Box 814 Bellmore, NY 11710 (800) 645-6379 FAX: (516) 781-4937 (516) 781-0755 http://www.ancare.com Ancell 243 Third Street North P.O. Box 87 Bayport, MN 55033 (800) 374-9523 FAX: (651) 439-1940 (651) 439-0835 http://www.ancell.com
Annovis 34 Mount Pleasant Drive Aston, PA 19014 (800) EASY-DNA FAX: (610) 361-8255 (610) 361-9224 http://www.annovis.com Apotex 150 Signet Drive Weston, Ontario M9L 1T9, Canada (416) 749-9300 FAX: (416) 749-2646 http://www.apotex.com Apple Scientific 11711 Chillicothe Road, Unit 2 P.O. Box 778 Chesterland, OH 44026 (440) 729-3056 FAX: (440) 729-0928 http://www.applesci.com Applied Biosystems See PE Biosystems Applied Imaging 2380 Walsh Avenue, Bldg. B Santa Clara, CA 95051 (800) 634-3622 FAX: (408) 562-0264 (408) 562-0250 http://www.aicorp.com Applied Photophysics 203-205 Kingston Road Leatherhead, Surrey, KT22 7PB UK (44) 1372-386537 Applied Precision 1040 12th Avenue Northwest Issaquah, Washington 98027 (425) 557-1000 FAX: (425) 557-1055 http://www.api.com/index.html
Aquarlum Systems B141 Tyler Boulevard Mentor. OH 44060 (800) 822-1100 FAX: (440) 266-8994 (440) 255-1997 http://www.aquaniumsystems.com Aquebogue Machine and Repair Shop Box 2055 Main Road Aquebogue, NY 11931 (631) 722-3635 FAX: (631) 722-3106 Archer Daniels Midland 4666 Faries Parkway Decatur, IL 62525 (217) 424-5200 http://www.admworld.com Archimica Florida P.O. Box 1466 Gainesville, FL 32602 (800) 331-6313 FAX: (352) 371-6246 (352) 376-8246 http://www.archimica.com Arcor Electronics 1845 Oak Street #15 Northfield, IL 60093 (847) 501-4848 Arcturus Engineering 400 Logue Avenue Mountain View, CA 94043 (888) 446 7911 FAX: (650) 962 3039 (650) 962 3020 http://www.arctur.com Ardals Corporation One ledgemont Center 128 Spring Street Lexington, MA 02421 (781) 274-6420 (781 274-6421 http://www.ardais.com
Anderson Instruments 500 Technology Court Smyrna, GA 30082 (800) 241-6898 FAX: (770) 319-5306 (770) 319-9999 http://www.graseby.com
Appligene Oncor Parc d’Innovation Rue Geiler de Kaysersberg, BP 72 67402 Illkirch Cedex, France (33) 88 67 22 67 FAX: (33) 88 67 19 45 http://www.oncor.com/prod-app.htm
Andreas Hettich Gartenstrasse 100 Postfach 260 D-78732 Tuttlingen, Germany (49) 7461 705 0 FAX: (49) 7461 705-122 http://www.hettich-centrifugen.de
Applikon 1165 Chess Drive, Suite G Foster City, CA 94404 (650) 578-1396 FAX: (650) 578-8836 http://www.applikon.com
Ariad Pharmaceuticals 26 Landsdowne Street Cambridge, MA 02139 (617) 494-0400 FAX: (617) 494-8144 http://www.ariad.com
Appropriate Technical Resources 9157 Whiskey Bottom Road Laurel, MD 20723 (800) 827-5931 FAX: (410) 792-2837 http://www.atrbiotech.com
Armour Pharmaceuticals See Rhone-Poulenc Rorer
Anesthetic Vaporizer Services 10185 Main Street Clarence, NY 14031 (719) 759-8490 www.avapor.com Animal Identification and Marking Systems (AIMS) 13 Winchester Avenue Budd Lake, NJ 07828 (908) 684-9105 FAX: (908) 684-9106 http://www.animalid.com
APV Gaulin 100 S. CP Avenue Lake Mills, WI 53551 (888) 278-4321 FAX: (888) 278-5329 http://www.apv.com Aqualon See Hercules Aqualon
Argonaut Technologies 887 Industrial Road, Suite G San Carlos, CA 94070 (650) 998-1350 FAX: (650) 598-1359 http://www.argotech.com
Aronex Pharmaceuticals 8707 Technology Forest Place The Woodlands, TX 77381 (281) 367-1666 FAX: (281) 367-1676 http://www.aronex.com Artisan Industries 73 Pond Street Waltham, MA 02254 (617) 893-6800 http://www.artisanind.com
Suppliers
3 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
ASI Instruments 12900 Ten Mile Road Warren, MI 48089 (800) 531-1105 FAX: (810) 756-9737 (810) 756-1222 http://www.asi-instruments.com Aspen Research Laboratories 1700 Buerkle Road White Bear Lake, MN 55140 (651) 264-6000 FAX: (651) 264-6270 http://www.aspenresearch.com Associates of Cape Cod 704 Main Street Falmouth, MA 02540 (800) LAL-TEST FAX: (508) 540-8680 (508) 540-3444 http://www.acciusa.com Astra Pharmaceuticals See AstraZeneca AstraZeneca 1800 Concord Pike Wilmington, DE 19850 (302) 886-3000 FAX: (302) 886-2972 http://www.astrazeneca.com AT Biochem 30 Spring Mill Drive Malvern, PA 19355 (610) 889-9300 FAX: (610) 889-9304 ATC Diagnostics See Vysis ATCC See American Type Culture Collection
Aurora Biosciences 11010 Torreyana Road San Diego, CA 92121 (858) 404-6600 FAX: (858) 404-6714 http://www.aurorabio.com Automatic Switch Company A Division of Emerson Electric 50 Hanover Road Florham Park, NJ 07932 (800) 937-2726 FAX: (973) 966-2628 (973) 966-2000 http://www.asco.com
BAbCO 1223 South 47th Street Richmond, CA 94804 (800) 92-BABCO FAX: (510) 412-8940 (510) 412-8930 http://www.babco.com Bacharach 625 Alpha Drive Pittsburgh, PA 15238 (800) 736-4666 FAX: (412) 963-2091 (412) 963-2000 http://www.bacharach-inc.com
Avanti Polar Lipids 700 Industrial Park Drive Alabaster, AL 35007 (800) 227-0651 FAX: (800) 229-1004 (205) 663-2494 FAX: (205) 663-0756 http://www.avantilipids.com
Bachem Bioscience 3700 Horizon Drive King of Prussia, PA 19406 (800) 634-3183 FAX: (610) 239-0800 (610) 239-0300 http://www.bachem.com
Aventis BP 67917 67917 Strasbourg Cedex 9, France 33 (0) 388 99 11 00 FAX: 33 (0) 388 99 11 01 http://www.aventis.com
Bachem California 3132 Kashiwa Street P.O. Box 3426 Torrance, CA 90510 (800) 422-2436 FAX: (310) 530-1571 (310) 539-4171 http://www.bachem.com
Aventis Pasteur 1 Discovery Drive Swiftwater, PA 18370 (800) 822-2463 FAX: (570) 839-0955 (570) 839-7187 http://www.aventispasteur.com/usa
Baekon 18866 Allendale Avenue Saratoga, CA 95070 (408) 972-8779 FAX: (408) 741-0944 Baker Chemical See J.T. Baker Bangs Laboratories 9025 Technology Drive Fishers, IN 46038 (317) 570-7020 FAX: (317) 570-7034 www.bangslabs.com
Athens Research and Technology P.O. Box 5494 Athens, GA 30604 (706) 546-0207 FAX: (706) 546-7395
Avery Dennison 150 North Orange Grove Boulevard Pasadena, CA 91103 (800) 462-8379 FAX: (626) 792-7312 (626) 304-2000 http://www.averydennison.com
Atlanta Biologicals 1425-400 Oakbrook Drive Norcross, GA 30093 (800) 780-7788 or (770) 446-1404 FAX: (800) 780-7374 or (770) 446-1404 http://www.atlantabio.com
Avestin 2450 Don Reid Drive Ottawa, Ontario K1H 1E1, Canada (888) AVESTIN FAX: (613) 736-8086 (613) 736-0019 http://www.avestin.com
Barnstead/Thermolyne P.O. Box 797 2555 Kerper Boulevard Dubuque, IA 52004 (800) 446-6060 FAX: (319) 589-0516 http://www.barnstead.com
Atomergic Chemical 71 Carolyn Boulevard Farmingdale, NY 11735 (631) 694-9000 FAX: (631) 694-9177 http://www.atomergic.com
AVIV Instruments 750 Vassar Avenue Lakewood, NJ 08701 (732) 367-1663 FAX: (732) 370-0032 http://www.avivinst.com
Barrskogen 4612 Laverock Place N Washington, DC 20007 (800) 237-9192 FAX: (301) 464-7347
Atomic Energy of Canada 2251 Speakman Drive Mississauga, Ontario L5K 1B2 Canada (905) 823-9040 FAX: (905) 823-1290 http://www.aecl.ca
Axon Instruments 1101 Chess Drive Foster City, CA 94404 (650) 571-9400 FAX: (650) 571-9500 http://www.axon.com
ATR P.O. Box 460 Laurel, MD 20725 (800) 827-5931 FAX: (410) 792-2837 (301) 470-2799 http://www.atrbiotech.com
Azon 720 Azon Road Johnson City, NY 13790 (800) 847-9374 FAX: (800) 635-6042 (607) 797-2368 http://www.azon.com
Bard Parker See Becton Dickinson
BAS See Bioanalytical Systems
Bausch & Lomb One Bausch & Lomb Place Rochester, NY 14604 (800) 344-8815 FAX: (716) 338-6007 (716) 338-6000 http://www.bausch.com Baxter Fenwal Division 1627 Lake Cook Road Deerfield, IL 60015 (800) 766-1077 FAX: (800) 395-3291 (847) 940-6599 FAX: (847) 940-5766 http://www.powerfulmedicine.com Baxter Healthcare One Baxter Parkway Deerfield, IL 60015 (800) 777-2298 FAX: (847) 948-3948 (847) 948-2000 http://www.baxter.com Baxter Scientific Products See VWR Scientific Bayer Agricultural Division Animal Health Products 12707 Shawnee Mission Pkwy. Shawnee Mission, KS 66201 (800) 255-6517 FAX: (913) 268-2803 (913) 268-2000 http://www.bayerus.com Bayer Diagnostics Division (Order Services) P.O. Box 2009 Mishiwaka, IN 46546 (800) 248-2637 FAX: (800) 863-6882 (219) 256-3390 http://www.bayer.com Bayer Diagnostics 511 Benedict Avenue Tarrytown, NY 10591 (800) 255-3232 FAX: (914) 524-2132 (914) 631-8000 http://www.bayerdiag.com Bayer Plc Diagnostics Division Bayer House, Strawberry Hill Newbury, Berkshire RG14 1JA, UK (44) 1635-563000 FAX: (44) 1635-563393 http://www.bayer.co.uk
BASF Specialty Products 3000 Continental Drive North Mt. Olive, NJ 07828 (800) 669-2273 FAX: (973) 426-2610 http://www.basf.com
BD Immunocytometry Systems 2350 Qume Drive San Jose, CA 95131 (800) 223-8226 FAX: (408) 954-BDIS http://www.bdfacs.com
Baum, W.A. 620 Oak Street Copiague, NY 11726 (631) 226-3940 FAX: (631) 226-3969 http://www.wabaum.com
BD Labware Two Oak Park Bedford, MA 01730 (800) 343-2035 FAX: (800) 743-6200 http://www.bd.com/labware
Suppliers
4 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
BD PharMingen 10975 Torreyana Road San Diego, CA 92121 (800) 848-6227 FAX: (858) 812-8888 (858) 812-8800 http://www.pharmingen.com BD Transduction Laboratories 133 Venture Court Lexington, KY 40511 (800) 227-4063 FAX: (606) 259-1413 (606) 259-1550 http://www.translab.com BDH Chemicals Broom Road Poole, Dorset BH12 4NN, UK (44) 1202-745520 FAX: (44) 1202- 2413720 BDH Chemicals See Hoefer Scientific Instruments BDIS See BD Immunocytometry Systems Beckman Coulter 4300 North Harbor Boulevard Fullerton, CA 92834 (800) 233-4685 FAX: (800) 643-4366 (714) 871-4848 http://www.beckman-coulter.com Beckman Instruments Spinco Division/Bioproducts Operation 1050 Page Mill Road Palo Alto, CA 94304 (800) 742-2345 FAX: (415) 859-1550 (415) 857-1150 http://www.beckman-coulter.com Becton Dickinson Immunocytometry & Cellular Imaging 2350 Qume Drive San Jose, CA 95131 (800) 223-8226 FAX: (408) 954-2007 (408) 432-9475 http://www.bdfacs.com Becton Dickinson Labware 1 Becton Drive Franklin Lakes, NJ 07417 (888) 237-2762 FAX: (800) 847-2220 (201) 847-4222 http://www.bdfacs.com Becton Dickinson Labware 2 Bridgewater Lane Lincoln Park, NJ 07035 (800) 235-5953 FAX: (800) 847-2220 (201) 847-4222 http://www.bdfacs.com Becton Dickinson Primary Care Diagnostics 7 Loveton Circle Sparks, MD 21152 (800) 675-0908 FAX: (410) 316-4723 (410) 316-4000 http://www.bdfacs.com
Behringwerke Diagnostika Hoechster Strasse 70 P-65835 Liederback, Germany (49) 69-30511 FAX: (49) 69-303-834 Bellco Glass 340 Edrudo Road Vineland, NJ 08360 (800) 257-7043 FAX: (856) 691-3247 (856) 691-1075 http://www.bellcoglass.com Bender Biosystems See Serva Beral Enterprises See Garren Scientific Berkeley Antibody See BAbCO Bernsco Surgical Supply 25 Plant Avenue Hauppague, NY 11788 (800) TIEMANN FAX: (516) 273-6199 (516) 273-0005 http://www.bernsco.com Beta Medical and Scientific (Datesand Ltd.) 2 Ferndale Road Sale, Manchester M33 3GP, UK (44) 1612 317676 FAX: (44) 1612 313656 Bethesda Research Laboratories (BRL) See Life Technologies Biacore 200 Centennial Avenue, Suite 100 Piscataway, NJ 08854 (800) 242-2599 FAX: (732) 885-5669 (732) 885-5618 http://www.biacore.com Bilaney Consultants St. Julian’s Sevenoaks, Kent TN15 0RX, UK (44) 1732 450002 FAX: (44) 1732 450003 http://www.bilaney.com Binding Site 5889 Oberlin Drive, Suite 101 San Diego, CA 92121 (800) 633-4484 FAX: (619) 453-9189 (619) 453-9177 http://www.bindingsite.co.uk
Biocell 2001 University Drive Rancho Dominguez, CA 90220 (800) 222-8382 FAX: (310) 637-3927 (310) 537-3300 http://www.biocell.com Biocoat See BD Labware BioComp Instruments 650 Churchill Road Fredericton, New Brunswick E3B 1P6 Canada (800) 561-4221 FAX: (506) 453-3583 (506) 453-4812 http://131.202.97.21
Biomeda 1166 Triton Drive, Suite E P.O. Box 8045 Foster City, CA 94404 (800) 341-8787 FAX: (650) 341-2299 (650) 341-8787 http://www.biomeda.com BioMedic Data Systems 1 Silas Road Seaford, DE 19973 (800) 526-2637 FAX: (302) 628-4110 (302) 628-4100 http://www.bmds.com
BioDesign P.O. Box 1050 Carmel, NY 10512 (914) 454-6610 FAX: (914) 454-6077 http://www.biodesignofny.com
Biomedical Engineering P.O. Box 980694 Virginia Commonwealth University Richmond, VA 23298 (804) 828-9829 FAX: (804) 828-1008
BioDiscovery 4640 Admiralty Way, Suite 710 Marina Del Rey, CA 90292 (310) 306-9310 FAX: (310) 306-9109 http://www.biodiscovery.com
Biomedical Research Instruments 12264 Wilkins Avenue Rockville, MD 20852 (800) 327-9498 (301) 881-7911 http://www.biomedinstr.com
Bioengineering AG Sagenrainstrasse 7 CH8636 Wald, Switzerland (41) 55-256-8-111 FAX: (41) 55-256-8-256 Biofluids Division of Biosource International 1114 Taft Street Rockville, MD 20850 (800) 972-5200 FAX: (301) 424-3619 (301) 424-4140 http://www.biosource.com BioFX Laboratories 9633 Liberty Road, Suite S Randallstown, MD 21133 (800) 445-6447 FAX: (410) 498-6008 (410) 496-6006 http://www.biofx.com BioGenex Laboratories 4600 Norris Canyon Road San Ramon, CA 94583 (800) 421-4149 FAX: (925) 275-0580 (925) 275-0550 http://www.biogenex.com
Bio Image See Genomic Solutions
Bioline 2470 Wrondel Way Reno, NV 89502 (888) 257-5155 FAX: (775) 828-7676 (775) 828-0202 http://www.bioline.com
Bioanalytical Systems 2701 Kent Avenue West Lafayette, IN 47906 (800) 845-4246 FAX: (765) 497-1102 (765) 463-4527 http://www.bioanalytical.com
Bio-Logic Research & Development 1, rue de l-Europe A.Z. de Font-Ratel 38640 CLAIX, France (33) 76-98-68-31 FAX: (33) 76-98-69-09
BIO 101 See Qbiogene
Biological Detection Systems See Cellomics or Amersham
Bio/medical Specialties P.O. Box 1687 Santa Monica, CA 90406 (800) 269-1158 FAX: (800) 269-1158 (323) 938-7515 BioMerieux 100 Rodolphe Street Durham, North Carolina 27712 (919) 620-2000 http://www.biomerieux.com BioMetallics P.O. Box 2251 Princeton, NJ 08543 (800) 999-1961 FAX: (609) 275-9485 (609) 275-0133 http://www.microplate.com Biomol Research Laboratories 5100 Campus Drive Plymouth Meeting, PA 19462 (800) 942-0430 FAX: (610) 941-9252 (610) 941-0430 http://www.biomol.com Bionique Testing Labs Fay Brook Drive RR 1, Box 196 Saranac Lake, NY 12983 (518) 891-2356 FAX: (518) 891-5753 http://www.bionique.com Biopac Systems 42 Aero Camino Santa Barbara, CA 93117 (805) 685-0066 FAX: (805) 685-0067 http://www.biopac.com Bioproducts for Science See Harlan Bioproducts for Science
Suppliers
5 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Bioptechs 3560 Beck Road Butler, PA 16002 (877) 548-3235 FAX: (724) 282-0745 (724) 282-7145 http://www.bioptechs.com BIOQUANT-R&M Biometrics 5611 Ohio Avenue Nashville, TN 37209 (800) 221-0549 (615) 350-7866 FAX: (615) 350-7282 http://www.bioquant.com
Biosoft P.O. Box 10938 Ferguson, MO 63135 (314) 524-8029 FAX: (314) 524-8129 http://www.biosoft.com Biosource International 820 Flynn Road Camarillo, CA 93012 (800) 242-0607 FAX: (805) 987-3385 (805) 987-0086 http://www.biosource.com
Bio-Rad Laboratories 2000 Alfred Nobel Drive Hercules, CA 94547 (800) 424-6723 FAX: (800) 879-2289 (510) 741-1000 FAX: (510) 741-5800 http://www.bio-rad.com
BioSpec Products P.O. Box 788 Bartlesville, OK 74005 (800) 617-3363 FAX: (918) 336-3363 (918) 336-3363 http://www.biospec.com
Bio-Rad Laboratories Maylands Avenue Hemel Hempstead, Herts HP2 7TD, UK 800-4-BIORAD (424-6723) http://www.bio-rad.com
Biostatus Ltd. 56 Charnwood Road Shepshed Leicestershire LE12 9NP, UK 44-7957-575402 FAX: 44-1509-651061 e-mail: [email protected] http://www.Biostatus.co.uk
Bioreclamation 492 Richmond Road East Meadow, NY 11554 (516) 483-1196 FAX: (516) 483-4683 http://www.bioreclamation.com BioRobotics 3-4 Bennell Court Comberton, Cambridge CB3 7DS, UK (44) 1223-264345 FAX: (44) 1223-263933 http://www.biorobotics.co.uk BIOS Laboratories See Genaissance Pharmaceuticals Biosearch Technologies 81 Digital Drive Novato, CA 94949 (800) GENOME1 FAX: (415) 883-8488 (415) 883-8400 http://www.biosearchtech.com BioSepra 111 Locke Drive Marlborough, MA 01752 (800) 752-5277 FAX: (508) 357-7595 (508) 357-7500 http://www.biosepra.com Bio-Serv 1 8th Street, Suite 1 Frenchtown, NJ 08825 (908) 996-2155 FAX: (908) 996-4123 http://www.bio-serv.com BioSignal 1744 William Street, Suite 600 Montreal, Quebec H3J 1R4 Canada (800) 293-4501 FAX: (514) 937-0777 (514) 937-1010 http://www.biosignal.com
Biosure See Riese Enterprises Biosym Technologies See Molecular Simulations Biosys 21 quai du Clos des Roses 602000 Compiegne, France (33) 03 4486 2275 FAX: (33) 03 4484 2297 Bio-Tech Research Laboratories NIAID Repository Rockville, MD 20850 http://www.niaid.nih.gov/ncn/repos.htm Biotech Instruments Biotech House 75A High Street Kimpton, Hertfordshire SG4 8PU, UK (44) 1438 832555 FAX: (44) 1438 833040 http://www.biotinst.demon.co.uk Biotech International 11 Durbell Street Acacia Ridge, Queensland 4110 Australia 61-7-3370-6396 FAX: 61-7-3370-6370 http://www.avianbiotech.com Biotech Source Inland Farm Drive South Windham, ME 04062 (207) 892-3266 FAX: (207) 892-6774
Bio-Tek Instruments Highland Industrial Park P.O. Box 998 Winooski, VT 05404 (800) 451-5172 FAX: (802) 655-7941 (802) 655-4040 http://www.biotek.com Biotecx Laboratories 6023 South Loop East Houston, TX 77033 (800) 535-6286 FAX: (713) 643-3143 (713) 643-0606 http://www.biotecx.com BioTherm 3260 Wilson Boulevard Arlington, VA 22201 (703) 522-1705 FAX: (703) 522-2606 Bioventures P.O. Box 2561 848 Scott Street Murfreesboro, TN 37133 (800) 235-8938 FAX: (615) 896-4837 http://www.bioventures.com BioWhittaker 8830 Biggs Ford Road P.O. Box 127 Walkersville, MD 21793 (800) 638-8174 FAX: (301) 845-8338 (301) 898-7025 http://www.biowhittaker.com Biozyme Laboratories 9939 Hibert Street, Suite 101 San Diego, CA 92131 (800) 423-8199 FAX: (858) 549-0138 (858) 549-4484 http://www.biozyme.com Bird Products 1100 Bird Center Drive Palm Springs, CA 92262 (800) 328-4139 FAX: (760) 778-7274 (760) 778-7200 http://www.birdprod.com/bird B & K Universal 2403 Yale Way Fremont, CA 94538 (800) USA-MICE FAX: (510) 490-3036 BLS Ltd. Zselyi Aladar u. 31 1165 Budapest, Hungary (36) 1-407-2602 FAX: (36) 1-407-2896 http://www.bls-ltd.com Blue Sky Research 3047 Orchard Parkway San Jose, CA 95134 (408) 474-0988 FAX: (408) 474-0989 http://www.blueskyresearch.com Blumenthal Industries 7 West 36th Street, 13th floor New York, NY 10018 (212) 719-1251 FAX: (212) 594-8828
Boblin Instruments 11 Harts Lane, Suite 1 East Brunswick, NJ 08816 (732) 254-7742 FAX: (732) 254-1577 http://www.boblin.co.uk BOC Edwards One Edwards Park 301 Ballardvale Street Wilmington, MA 01887 (800) 848-9800 FAX: (978) 658-7969 (978) 658-5410 http://www.bocedwards.com Boehringer Ingelheim 900 Ridgebury Road P.O. Box 368 Ridgefield, CT 06877 (800) 243-0127 FAX: (203) 798-6234 (203) 798-9988 http://www.boehringer-ingelheim.com Boehringer Mannheim Biochemicals Division See Roche Diagnostics Boekel Scientific 855 Pennsylvania Boulevard Feasterville, PA 19053 (800) 336-6929 FAX: (215) 396-8264 (215) 396-8200 http://www.boekelsci.com Bohdan Automation 1500 McCormack Boulevard Mundelein, IL 60060 (708) 680-3939 FAX: (708) 680-1199 BPAmoco 4500 McGinnis Ferry Road Alpharetta, GA 30005 (800) 328-4537 FAX: (770) 772-8213 (770) 772-8200 http://www.bpamoco.com Brain Research Laboratories Waban P.O. Box 88 Newton, MA 02468 (888) BRL-5544 FAX: (617) 965-6220 (617) 965-5544 http://www.brainresearchlab.com Braintree Scientific P.O. Box 850929 Braintree, MA 02185 (781) 843-1644 FAX: (781) 982-3160 http://www.braintreesci.com Brandel 8561 Atlas Drive Gaithersburg, MD 20877 (800) 948-6506 FAX: (301) 869-5570 (301) 948-6506 http://www.brandel.com Branson Ultrasonics 41 Eagle Road Danbury, CT 06813 (203) 796-0400 FAX: (203) 796-9838 http://www.plasticsnet.com/branson
Suppliers
6 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
B. Braun Biotech 999 Postal Road Allentown, PA 18103 (800) 258-9000 FAX: (610) 266-9319 (610) 266-6262 http://www.bbraunbiotech.com B. Braun Biotech International Schwarzenberg Weg 73-79 P.O. Box 1120 D-34209 Melsungen, Germany (49) 5661-71-3400 FAX: (49) 5661-71-3702 http://www.bbraunbiotech.com B. Braun-McGaw 2525 McGaw Avenue Irvine, CA 92614 (800) BBRAUN-2 (800) 624-2963 http://www.bbraunusa.com B. Braun Medical Thorncliffe Park Sheffield S35 2PW, UK (44) 114-225-9000 FAX: (44) 114-225-9111 http://www.bbmuk.demon.co.uk Brenntag P.O. Box 13788 Reading, PA 19612-3788 (610) 926-4151 FAX: (610) 926-4160 http://www.brenntagnortheast.com Bresatec See GeneWorks Bright/Hacker Instruments 17 Sherwood Lane Fairfield, NJ 07004 (973) 226-8450 FAX: (973) 808-8281 http://www.hackerinstruments.com Brinkmann Instruments Subsidiary of Sybron 1 Cantiague Road P.O. Box 1019 Westbury, NY 11590 (800) 645-3050 FAX: (516) 334-7521 (516) 334-7500 http://www.brinkmann.com Bristol-Meyers Squibb P.O. Box 4500 Princeton, NJ 08543 (800) 631-5244 FAX: (800) 523-2965 http://www.bms.com Broadley James 19 Thomas Irvine, CA 92618 (800) 288-2833 FAX: (949) 829-5560 (949) 829-5555 http://www.broadleyjames.com Brookhaven Instruments 750 Blue Point Road Holtsville, NY 11742 (631) 758-3200 FAX: (631) 758-3255 http://www.bic.com
Brownlee Labs See Applied Biosystems Distributed by Pacer Scientific Bruel & Kjaer Division of Spectris Technologies 2815 Colonnades Court Norcross, GA 30071 (800) 332-2040 FAX: (770) 847-8440 (770) 209-6907 http://www.bkhome.com Bruker Analytical X-Ray Systems 5465 East Cheryl Parkway Madison, WI 53711 (800) 234-XRAY FAX: (608) 276-3006 (608) 276-3000 http://www.bruker-axs.com Bruker Instruments 19 Fortune Drive Billerica, MA 01821 (978) 667-9580 FAX: (978) 667-0985 http://www.bruker.com BTX Division of Genetronics 11199 Sorrento Valley Road San Diego, CA 92121 (800) 289-2465 FAX: (858) 597-9594 (858) 597-6006 http://www.genetronics.com/btx Buchler Instruments See Baxter Scientific Products Buckshire 2025 Ridge Road Perkasie, PA 18944 (215) 257-0116 Burdick and Jackson Division of Baxter Scientific Products 1953 S. Harvey Street Muskegon, MI 49442 (800) 368-0050 FAX: (231) 728-8226 (231) 726-3171 http://www.bandj.com/mainframe.htm Burleigh Instruments P.O. Box E Fishers, NY 14453 (716) 924-9355 FAX: (716) 924-9072 http://www.burleigh.com Burns Veterinary Supply 1900 Diplomat Drive Farmer’s Branch, TX 75234 (800) 92-BURNS FAX: (972) 243-6841 http://www.burnsvet.com
Butterworth Laboratories 54-56 Waldegrave Road Teddington, Middlesex TW11 8LG, UK (44)(0)20-8977-0750 FAX: (44)(0)28-8943-2624 http://www.butterworth-labs.co.uk Buxco Electronics 95 West Wood Road #2 Sharon, CT 06069 (860) 364-5558 FAX: (860) 364-5116 http://www.buxco.com C/D/N Isotopes 88 Leacock Street Pointe-Claire, Quebec H9R 1H1 Canada (800) 697-6254 FAX: (514) 697-6148 C.M.A./Microdialysis AB 73 Princeton Street North Chelmsford, MA 01863 (800) 440-4980 FAX: (978) 251-1950 (978) 251-1940 http://www.microdialysis.com Calbiochem-Novabiochem P.O. Box 12087-2087 La Jolla, CA 92039 (800) 854-3417 FAX: (800) 776-0999 (858) 450-9600 http://www.calbiochem.com California Fine Wire 338 South Fourth Street Grover Beach, CA 93433 (805) 489-5144 FAX: (805) 489-5352 http://www.calinewire.com Caligor 846 Pelham Parkway Pelham Manor, NY 10803 (800) 225-9906 Fax: 914-738-9538 http://www.caligor.com Calorimetry Sciences 155 West 2050 North Spanish Fork, UT 84660 (801) 794-2600 FAX: (801) 794-2700 http://www.calscorp.com Caltag Laboratories 1849 Bayshore Highway, Suite 200 Burlingame, CA 94010 (800) 874-4007 FAX: (650) 652-9030 (650) 652-0468 http://www.caltag.com
Cambridge Isotope Laboratories 50 Frontage Road Andover, MA 01810 (800) 322-1174 FAX: (978) 749-2768 (978) 749-8000 http://www.isotope.com Cambridge Research Biochemicals See Zeneca/CRB Cambridge Technology 109 Smith Place Cambridge, MA 02138 (6l7) 441-0600 FAX: (617) 497-8800 http://www.camtech.com Camlab Nuffield Road Cambridge CB4 1TH, UK (44) 122-3424222 FAX: (44) 122-3420856 http://www.camlab.co.uk/home.htm Campden Instruments Park Road Sileby Loughborough Leicestershire LE12 7TU, UK (44) 1509-814790 FAX: (44) 1509-816097 http://www.campdeninst.com/home.htm Captiol Equipment Corp. 900 Middlesex Turnpike Billerica, MA 01821 (978) 663-2002 FAX: (978) 663-2626 Cappel Laboratories See Organon Teknika Cappel Carl Roth GmgH & Company Schoemperlenstrasse 1-5 76185 Karlsrube Germany (49) 72-156-06164 FAX: (49) 72-156-06264 http://www.carl-roth.de Carl Zeiss One Zeiss Drive Thornwood, NY 10594 (800) 233-2343 FAX: (914) 681-7446 (914) 747-1800 http://www.zeiss.com
Burroughs Wellcome See Glaxo Wellcome
Cambrex Corporation 1 Meadowlands Plaza East Rulherford, NJ 07073 (201) 804-3000 FAX: (201) 804-9852 http://www.cambrex.com
Carlo Erba Reagenti Via Winckelmann 1 20148 Milano Lombardia, Italy (39) 0-29-5231 FAX: (39) 0-29-5235-904 http://www.carloerbareagenti.com
The Butler Company 5600 Blazer Parkway Dublin, OH 43017 (800) 551-3861 FAX: (614) 761-9096 (614) 761-9095 http://www.wabutler.com
Cambridge Electronic Design Science Park, Milton Road Cambridge CB4 0FE, UK 44 (0) 1223-420-186 FAX: 44 (0) 1223-420-488 http://www.ced.co.uk
Carolina Biological Supply 2700 York Road Burlington, NC 27215 (800) 334-5551 FAX: (336) 584-76869 (336) 584-0381 http://www.carolina.com
Suppliers
7 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Carolina Fluid Components 9309 Stockport Place Charlotte, NC 28273 (704) 588-6101 FAX: (704) 588-6115 http://www.cfcsite.com Cartesian Technologies 17851 Skypark Circle, Suite C Irvine, CA 92614 (800) 935-8007 http://cartesiantech.com Cayman Chemical 1180 East Ellsworth Road Ann Arbor, MI 48108 (800) 364-9897 FAX: (734) 971-3640 (734) 971-3335 http://www.caymanchem.com CB Sciences One Washington Street, Suite 404 Dover, NH 03820 (800) 234-1757 FAX: (603) 742-2455 http://www.cbsci.com CBS Scientific P.O. Box 856 Del Mar, CA 92014 (800) 243-4959 FAX: (858) 755-0733 (858) 755-4959 http://www.cbssci.com CCR (Coriell Cell Repository) See Coriell Institute for Medical Research CE Instruments Grand Avenue Parkway Austin, TX 78728 (800) 876-6711 FAX: (512) 251-1597 http://www.ceinstruments.com Cedarlane Laboratories 5516 8th Line, R.R. #2 Hornby, Ontario L0P 1E0 Canada (905) 878-8891 FAX: (905) 878-7800 http://www.cedarlanelabs.com CEL Associates P.O. Box 721854 Houston, TX 77272 (800) 537-9339 FAX: (281) 933-0922 (281) 933-9339 http://www.cel-1.com Cel-Line Associates See Erie Scientific Celite World Minerals 130 Castilian Drive Santa Barbara, CA 93117 (805) 562-0200 FAX: (805) 562-0299 http://www.worldminerals.com/celite Cell Genesys 342 Lakeside Drive Foster City, CA 94404 (650) 425-4400 FAX: (650) 425-4457 http://www.cellgenesys.com
Cell Signaling Technology 166B Cummings Center Boverly, MA 01915 (877) 616-CELL FAX: (978) 867-2388 (978) 867-2488 http://www.cellsignal.com
Charles River Laboratories 251 Ballardvale Street Wilmington, MA 01887 (800) 522-7287 FAX: (978) 658-7132 (978) 658-6000 http://www.criver.com
Cell Systems 12815 NE 124th Street, Suite A Kirkland, WA 98034 (800) 697-1211 FAX: (425) 820-6762 (425) 823-1010
Charm Sciences 36 Franklin Street Malden, MA 02148 (800) 343-2170 FAX: (781) 322-3141 (781) 322-1523 http://www.charm.com
Cellmark Diagnostics 20271 Goldenrod Lane Germantown, MD 20876 (800) 872-5227 FAX: (301) 428-4877 (301) 428-4980 http://www.cellmark-labs.com Cellomics 635 William Pitt Way Pittsburgh, PA 15238 (888) 826-3857 FAX: (412) 826-3850 (412) 826-3600 http://www.cellomics.com CellPoint Scientific P.O. Box 10757 Gaithersburg, MD 20898 (301) 208-0696 Fax: (301) 590-1290 (800) 424-2984 Celltech 216 Bath Road Slough, Berkshire SL1 4EN, UK (44) 1753 534655 FAX: (44) 1753 536632 http://www.celltech.co.uk Cellular Products 872 Main Street Buffalo, NY 14202 (800) CPI-KITS FAX: (716) 882-0959 (716) 882-0920 http://www.zeptometrix.com CEM P.O. Box 200 Matthews, NC 28106 (800) 726-3331
Chase-Walton Elastomers 29 Apsley Street Hudson, MA 01749 (800) 448-6289 FAX: (978) 562-5178 (978) 568-0202 http://www.chase-walton.com ChemGenes Ashland Technology Center 200 Homer Avenue Ashland, MA 01721 (800) 762-9323 FAX: (508) 881-3443 (508) 881-5200 http://www.chemgenes.com Chemglass 3861 North Mill Road Vineland, NJ 08360 (800) 843-1794 FAX: (856) 696-9102 (800) 696-0014 http://www.chemglass.com Chemicon International 28835 Single Oak Drive Temecula, CA 92590 (800) 437-7500 FAX: (909) 676-9209 (909) 676-8080 http://www.chemicon.com Chem-Impex International 935 Dillon Drive Wood Dale, IL 60191 (800) 869-9290 FAX: (630) 766-2218 (630) 766-2112 http://www.chemimpex.com
Centers for Disease Control 1600 Clifton Road NE Atlanta, GA 30333 (800) 311-3435 FAX: (888) 232-3228 (404) 639-3311 http://www.cdc.gov
Chem Service P.O. Box 599 West Chester, PA 19381-0599 (610) 692-3026 FAX: (610) 692-8729 http://www.chemservice.com
CERJ Centre d’Elevage Roger Janvier 53940 Le Genest Saint Isle France
Chemsyn Laboratories 13605 West 96th Terrace Lenexa, Kansas 66215 (913) 541-0525 FAX: (913) 888-3582 http://www.tech.epcorp.com/ChemSyn/ chemsyn.htm
Cetus See Chiron Chance Propper Warly, West Midlands B66 1NZ, UK (44)(0)121-553-5551 FAX: (44)(0)121-525-0139
Chemunex USA 1 Deer Park Drive, Suite H-2 Monmouth Junction, NJ 08852 (800) 411-6734 http://www.chemunex.com
Cherwell Scientific Publishing The Magdalen Centre Oxford Science Park Oxford OX44GA, UK (44)(1) 865-784-800 FAX: (44)(1) 865-784-801 http://www.cherwell.com ChiRex Cauldron 383 Phoenixville Pike Malvern, PA 19355 (610) 727-2215 FAX: (610) 727-5762 http://www.chirex.com Chiron Diagnostics See Bayer Diagnostics Chiron Mimotopes Peptide Systems See Multiple Peptide Systems Chiron 4560 Horton Street Emeryville, CA 94608 (800) 244-7668 FAX: (510) 655-9910 (510) 655-8730 http://www.chiron.com Chrom Tech P.O. Box 24248 Apple Valley, MN 55124 (800) 822-5242 FAX: (952) 431-6345 http://www.chromtech.com Chroma Technology 72 Cotton Mill Hill, Unit A-9 Brattleboro, VT 05301 (800) 824-7662 FAX: (802) 257-9400 (802) 257-1800 http://www.chroma.com Chromatographie ZAC de Moulin No. 2 91160 Saulx les Chartreux France (33) 01-64-54-8969 FAX: (33) 01-69-0988091 http://www.chromatographie.com Chromogenix Taljegardsgatan 3 431-53 Mlndal, Sweden (46) 31-706-20-70 FAX: (46) 31-706-20-80 http://www.chromogenix.com Chrompack USA c/o Varian USA 2700 Mitchell Drive Walnut Creek, CA 94598 (800) 526-3687 FAX: (925) 945-2102 (925) 939-2400 http://www.chrompack.com Chugai Biopharmaceuticals 6275 Nancy Ridge Drive San Diego, CA 92121 (858) 535-5900 FAX: (858) 546-5973 http://www.chugaibio.com Ciba-Corning Diagnostics See Bayer Diagnostics
Suppliers
8 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Ciba-Geigy See Ciba Specialty Chemicals or Novartis Biotechnology Ciba Specialty Chemicals 540 White Plains Road Tarrytown, NY 10591 (800) 431-1900 FAX: (914) 785-2183 (914) 785-2000 http://www.cibasc.com CIBA Vision Division of Novartis AG 11460 Johns Creek Parkway Duluth, GA 30097 (770) 476-3937 http://www.cvworld.com Cidex Advanced Sterillzation Products 33 Technology Drive Irvine, CA 92618 (800) 595-0200 (949) 581-5799 http://www.cidex.com/ASPnew.htm Cinna Scientific Subsidiary of Molecular Research Center 5645 Montgomery Road Cincinnati, OH 45212 (800) 462-9868 FAX: (513) 841-0080 (513) 841-0900 http://www.mrcgene.com Cistron Biotechnology 10 Bloomfield Avenue Pine Brook, NJ 07058 (800) 642-0167 FAX: (973) 575-4854 (973) 575-1700 http://www.cistronbio.com Clark Electromedical Instruments See Harvard Apparatus Clay Adam See Becton Dickinson Primary Care Diagnostics CLB (Central Laboratory of the Netherlands) Blood Transfusion Service P.O. Box 9190 1006 AD Amsterdam, The Netherlands (31) 20-512-9222 FAX: (31) 20-512-3332 Cleveland Scientific P.O. Box 300 Bath, OH 44210 (800) 952-7315 FAX: (330) 666-2240 http:://www.clevelandscientific.com Clonetics Division of BioWhittaker http://www.clonetics.com Also see BioWhittaker Clontech Laboratories 1020 East Meadow Circle Palo Alto, CA 94303 (800) 662-2566 FAX: (800) 424-1350 (650) 424-8222 FAX: (650) 424-1088 http://www.clontech.com
Closure Medical Corporation 5250 Greens Dairy Road Raleigh, NC 27616 (919) 876-7800 FAX: (919) 790-1041 http://www.closuremed.com
Colorado Serum 4950 York Street Denver, CO 80216 (800) 525-2065 FAX: (303) 295-1923 http://www.colorado-serum.com
CMA Microdialysis AB 73 Princeton Street North Chelmsford, MA 01863 (800) 440-4980 FAX: (978) 251-1950 (978) 251 1940 http://www.microdialysis.com
Columbia Diagnostics 8001 Research Way Springfield, VA 22153 (800) 336-3081 FAX: (703) 569-2353 (703) 569-7511 http://www.columbiadiagnostics.com
Cocalico Biologicals 449 Stevens Road P.O. Box 265 Reamstown, PA 17567 (717) 336-1990 FAX: (717) 336-1993
Columbus Instruments 950 North Hague Avenue Columbus, OH 43204 (800) 669-5011 FAX: (614) 276-0529 (614) 276-0861 http://www.columbusinstruments.com
Coherent Ltd. 28 St. Thomas The Cambridgeshire Business Park Ely, CB7 4EX, UK 44-1353-658833 FAX: 44-1353-659107 e-mail: [email protected] Coherent Laser 5100 Patrick Henry Drive Santa Clara, CA 95056 (800) 227-1955 FAX: (408) 764-4800 (408) 764-4000 http://www.cohr.com Cohu P.O. Box 85623 San Diego, CA 92186 (858) 277-6700 FAX: (858) 277-0221 http://www.COHU.com/cctv Cole-Parmer Instrument 625 East Bunker Court Vernon Hills, IL 60061 (800) 323-4340 FAX: (847) 247-2929 (847) 549-7600 http://www.coleparmer.com Collaborative Biomedical Products and Collaborative Research See Becton Dickinson Labware Collagen Aesthetics 1850 Embarcadero Road Palo Alto, CA 94303 (650) 856-0200 FAX: (650) 856-0533 http://www.collagen.com Collagen Corporation See Collagen Aesthetics College of American Pathologists 325 Waukegan Road Northfield, IL 60093 (800) 323-4040 FAX: (847) 832-8000 (847) 446-8800 http://www.cap.org/index.cfm Colonial Medical Supply 504 Wells Road Franconia, NH 03580 (603) 823-9911 FAX: (603) 823-8799 http://www.colmedsupply.com
Compu Cyte Corp. 12 Emily Street Cambridge, MA 02139 (800) 840-1303 FAX: (617) 577-4501 (617) 492-1300 http://www.compucyte.com Compugen 25 Leek Crescent Richmond Hill, Ontario L4B 4B3 Canada 800-387-5045 FAX: (905) 707-2020 (905) 707-2000 http://www.compugen.com/ locations.htm Computer Associates International One Computer Associates Plaza Islandia, NY 11749 (631) 342-6000 FAX: (631) 342-6800 http://www.cai.com Connaught Laboratories See Aventis Pasteur Connectix 2955 Campus Drive, Suite 100 San Mateo, CA 94403 (800) 950-5880 FAX: (650) 571-0850 (650) 571-5100 http://www.connectix.com Contech 99 Hartford Avenue Providence, RI 02909 (401) 351-4890 FAX: (401) 421-5072 http://www.iol.ie/∼burke/contech.html Continental Laboratory Products 5648 Copley Drive San Diego, CA 92111 (800) 456-7741 FAX: (858) 279-5465 (858) 279-5000 http://www.conlab.com ConvaTec Professional Services P.O. Box 5254 Princeton, NJ 08543 (800) 422-8811 http://www.convatec.com
Cooper Instruments & Systems P.O. Box 3048 Warrenton, VA 20188 (800) 344-3921 FAX: (540) 347-4755 (540) 349-4746 http://www.cooperinstruments.com Cooperative Human Tissue Network (866) 462-2486 http://www.chin.ims.nci.nih.gov Cora Styles Needles ’N Blocks 56 Milton Street Arlington, MA 02474 (781) 648-6289 FAX: (781) 641-7917 Coriell Cell Repository (CCR) See Coriell Institute for Medical Research Coriell Institute for Medical Research Human Genetic Mutant Repository 401 Haddon Avenue Camden, NJ 08103 (856) 966-7377 FAX: (856) 964-0254 http://arginine.umdnj.edu Corion 8 East Forge Parkway Franklin, MA 02038 (508) 528-4411 FAX: (508) 520-7583 (800) 598-6783 http://www.corion.com Corning and Corning Science Products P.O. Box 5000 Corning, NY 14831 (800) 222-7740 FAX: (607) 974-0345 (607) 974-9000 http://www.corning.com Costar See Corning Coulbourn Instruments 7462 Penn Drive Allentown, PA 18106 (800) 424-3771 FAX: (610) 391-1333 (610) 395-3771 http://www.coulbourninst.com Coulter Cytometry See Beckman Coulter Covance Research Products 465 Swampbridge Road Denver, PA 17517 (800) 345-4114 FAX: (717) 336-5344 (717) 336-4921 http://www.covance.com Coy Laboratory Products 14500 Coy Drive Grass Lake, MI 49240 (734) 475-2200 FAX: (734) 475-1846 http://www.coylab.com
Suppliers
9 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
CPG 3 Borinski Road Lincoln Park, NJ 07035 (800) 362-2740 FAX: (973) 305-0884 (973) 305-8181 http://www.cpg-blotech.com CPL Scientific 43 Kingfisher Court Hambridge Road Newbury RG14 5SJ, UK (44) 1635-574902 FAX: (44) 1635-529322 http://www.cplscientific.co.uk CraMar Technologies 8670 Wolff Court, #160 Westminster, CO 80030 (800) 4-TOMTEC http://www.cramar.com Crescent Chemical 1324 Motor Parkway Hauppauge, NY 11788 (800) 877-3225 FAX: (631) 348-0913 (631) 348-0333 http://www.creschem.com Crist Instrument P.O. Box 128 10200 Moxley Road Damascus, MD 20872 (301) 253-2184 FAX: (301) 253-0069 http://www.cristinstrument.com Cruachem See Annovis http://www.cruachem.com CS Bio 1300 Industrial Road San Carlos, CA 94070 (800) 627-2461 FAX: (415) 802-0944 (415) 802-0880 http://www.csbio.com CS-Chromatographie Service Am Parir 27 D-52379 Langerwehe, Germany (49) 2423-40493-0 FAX: (49) 2423-40493-49 http://www.cs-chromatographie.de Cuno 400 Research Parkway Meriden, CT 06450 (800) 231-2259 FAX: (203) 238-8716 (203) 237-5541 http://www.cuno.com Curtin Matheson Scientific 9999 Veterans Memorial Drive Houston, TX 77038 (800) 392-3353 FAX: (713) 878-3598 (713) 878-3500 CWE 124 Sibley Avenue Ardmore, PA 19003 (610) 642-7719 FAX: (610) 642-1532 http://www.cwe-inc.com
Cybex Computer Products 4991 Corporate Drive Huntsville, AL 35805 (800) 932-9239 FAX: (800) 462-9239 http://www.cybex.com Cygnus Technology P.O. Box 219 Delaware Water Gap, PA 18327 (570) 424-5701 FAX: (570) 424-5630 http://www.cygnustech.com Cymbus Biotechnology Eagle Class, Chandler’s Ford Hampshire SO53 4NF, UK (44) 1-703-267-676 FAX: (44) 1-703-267-677 http://www.biotech.cymbus.com Cytogen 600 College Road East Princeton, NJ 08540 (609) 987-8200 FAX: (609) 987-6450 http://www.cytogen.com Cytogen Research and Development 89 Bellevue Hill Road Boston, MA 02132 (617) 325-7774 FAX: (617) 327-2405 CytRx 154 Technology Parkway Norcross, GA 30092 (800) 345-2987 FAX: (770) 368-0622 (770) 368-9500 http://www.cytrx.com Dade Behring Corporate Headquarters 1717 Deerfield Road Deerfield, IL 60015 (847) 267-5300 FAX: (847) 267-1066 http://www.dadebehring.com Dagan 2855 Park Avenue Minneapolis, MN 55407 (612) 827-5959 FAX: (612) 827-6535 http://www.dagan.com Dako 6392 Via Real Carpinteria, CA 93013 (800) 235-5763 FAX: (805) 566-6688 (805) 566-6655 http://www.dakousa.com Dako A/S 42 Produktionsvej P.O. Box 1359 DK-2600 Glostrup, Denmark (45) 4492-0044 FAX: (45) 4284-1822 Dakopatts See Dako A/S Dalton Chemical Laboratoris 349 wildcat Road Toronto, Ontario M3J 253 Canada (416) 661-2102 FAX: (416) 661-2108 (800) 567-5060 (in Canada only) http://www.dallon.com
Damon, IEC See Thermoquest Dan Kar Scientific 150 West Street Wilmington, MA 01887 (800) 942-5542 FAX: (978) 658-0380 (978) 988-9696 http://www.dan-kar.com DataCell Falcon Business Park 40 Ivanhoe Road Finchampstead, Berkshire RG40 4QQ, UK (44) 1189 324324 FAX: (44) 1189 324325 http://www.datacell.co.uk In the US: (408) 446-3575 FAX: (408) 446-3589 http://www.datacell.com DataWave Technologies 380 Main Street, Suite 209 Longmont, CO 80501 (800) 736-9283 FAX: (303) 776-8531 (303) 776-8214 Datex-Ohmeda 3030 Ohmeda Drive Madison, WI 53718 (800) 345-2700 FAX: (608) 222-9147 (608) 221-1551 http://www.us.datex-ohmeda.com DATU 82 State Street Geneva, NY 14456 (315) 787-2240 FAX: (315) 787-2397 http://www.nysaes.cornell.edu/datu
Degussa Precious Metals Division 3900 South Clinton Avenue South Plainfield, NJ 07080 (800) DEGUSSA FAX: (908) 756-7176 (908) 561-1100 http://www.degussa-huls.com Deneba Software 1150 NW 72nd Avenue Miami, FL 33126 (305) 596-5644 FAX: (305) 273-9069 http://www.deneba.com Deseret Medical 524 West 3615 South Salt Lake City, UT 84115 (801) 270-8440 FAX: (801) 293-9000 Devcon Plexus 30 Endicott Street Danvers, MA 01923 (800) 626-7226 FAX: (978) 774-0516 (978) 777-1100 http://www.devcon.com Developmental Studies Hybridoma Bank University of Iowa 436 Biology Building Iowa City, IA 52242 (319) 335-3826 FAX: (319) 335-2077 http://www.uiowa.edu/∼dshbwww DeVilbiss Division of Sunrise Medical Respiratory 100 DeVilbiss Drive P.O. Box 635 Somerset, PA 15501 (800) 338-1988 FAX: (814) 443-7572 (814) 443-4881 http://www.sunrisemedical.com
David Kopf Instruments 7324 Elmo Street P.O. Box 636 Tujunga, CA 91043 (818) 352-3274 FAX: (818) 352-3139
Dharmacon Research 1376 Mlners Drive #101 Lafayette, CO 80026 (303) 604-9499 FAX: (303) 604-9680 http://www.dharmacom.com
Decagon Devices P.O. Box 835 950 NE Nelson Court Pullman, WA 99163 (800) 755-2751 FAX: (509) 332-5158 (509) 332-2756 http://www.decagon.com
DiaCheM Triangle Biomedical Gardiners Place West Gillibrands, Lancashire WN8 9SP, UK (44) 1695-555581 FAX: (44) 1695-555518 http://www.diachem.co.uk
Decon Labs 890 Country Line Road Bryn Mawr, PA 19010 (800) 332-6647 FAX: (610) 964-0650 (610) 520-0610 http://www.deconlabs.com Decon Laboratories Conway Street Hove, Sussex BN3 3LY, UK (44) 1273 739241 FAX: (44) 1273 722088
Diagen Max-Volmer Strasse 4 D-40724 Hilden, Germany (49) 2103-892-230 FAX: (49) 2103-892-222 Diagnostic Concepts 6104 Madison Court Morton Grove, IL 60053 (847) 604-0957 Diagnostic Developments See DiaCheM
Suppliers
10 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Diagnostic Instruments 6540 Burroughs Sterling Heights, MI 48314 (810) 731-6000 FAX: (810) 731-6469 http://www.diaginc.com Diamedix 2140 North Miami Avenue Miami, FL 33127 (800) 327-4565 FAX: (305) 324-2395 (305) 324-2300 DiaSorin 1990 Industrial Boulevard Stillwater, MN 55082 (800) 328-1482 FAX: (651) 779-7847 (651) 439-9719 http://www.diasorin.com Diatome US 321 Morris Road Fort Washington, PA 19034 (800) 523-5874 FAX: (215) 646-8931 (215) 646-1478 http://www.emsdiasum.com Difco Laboratories See Becton Dickinson Digene 1201 Clopper Road Gaithersburg, MD 20878 (301) 944-7000 (800) 344-3631 FAX: (301) 944-7121 www.digene.com Digi-Key 701 Brooks Avenue South Thief River Falls, MN 56701 (800) 344-4539 FAX: (218) 681-3380 (218) 681-6674 http://www.digi-key.com Digitimer 37 Hydeway Welwyn Garden City, Hertfordshire AL7 3BE, UK (44) 1707-328347 FAX: (44) 1707-373153 http://www.digitimer.com Dimco-Gray 8200 South Suburban Road Dayton, OH 45458 (800) 876-8353 FAX: (937) 433-0520 (937) 433-7600 http://www.dimco-gray.com Dionex 1228 Titan Way P.O. Box 3603 Sunnyvale, CA 94088 (408) 737-0700 FAX: (408) 730-9403 http://dionex2.promptu.com Display Systems Biotech 1260 Liberty Way, Suite B Vista, CA 92083 (800) 697-1111 FAX: (760) 599-9930 (760) 599-0598 http://www.displaysystems.com
Diversified Biotech 1208 VFW Parkway Boston, MA 02132 (617) 965-8557 FAX: (617) 323-5641 (800) 796-9199 http://www.divbio.com DNA ProScan P.O. Box 121585 Nashville, TN 37212 (800) 841-4362 FAX: (615) 292-1436 (615) 298-3524 http://www.dnapro.com DNAStar 1228 South Park Street Madison, WI 53715 (608) 258-7420 FAX: (608) 258-7439 http://www.dnastar.com DNAVIEW Attn: Charles Brenner http://www.wco.com ∼cbrenner/dnaview.htm Doall NYC 36-06 48th Avenue Long Island City, NY 11101 (718) 392-4595 FAX: (718) 392-6115 http://www.doall.com Dojindo Molecular Technologies 211 Perry Street Parkway, Suite 5 Gaitherbusburg, MD 20877 (877) 987-2667 http://www.dojindo.com
Dremel 4915 21st Street Racine, WI 53406 (414) 554-1390 http://www.dremel.com Drummond Scientific 500 Parkway P.O. Box 700 Broomall, PA 19008 (800) 523-7480 FAX: (610) 353-6204 (610) 353-0200 http://www.drummondsci.com Duchefa Biochemie BV P.O. Box 2281 2002 CG Haarlem, The Netherlands 31-0-23-5319093 FAX: 31-0-23-5318027 http://www.duchefa.com Duke Scientific 2463 Faber Place Palo Alto, CA 94303 (800) 334-3883 FAX: (650) 424-1158 (650) 424-1177 http://www.dukescientific.com Duke University Marine Laboratory 135 Duke Marine Lab Road Beaufort, NC 28516-9721 (252) 504-7503 FAX: (252) 504-7648 http://www.env.duke.edu/rnarinelab
Dolla Eastern See Doall NYC
DuPont Biotechnology Systems See NEN Life Science Products
Dolan Jenner Industries 678 Andover Street Lawrence, MA 08143 (978) 681-8000 (978) 682-2500 http://www.dolan-jenner.com
DuPont Medical Products See NEN Life Science Products
Dow Chemical Customer Service Center 2040 Willard H. Dow Center Midland, MI 48674 (800) 232-2436 FAX: (517) 832-1190 (409) 238-9321 http://www.dow.com Dow Corning Northern Europe Meriden Business Park Copse Drive Allesley, Coventry CV5 9RG, UK (44) 1676 528 000 FAX: (44) 1676 528 001 Dow Corning P.O. Box 994 Midland, MI 48686 (517) 496-4000 http://www.dowcorning.com Dow Corning (Lubricants) 2200 West Salzburg Road Auburn, MI 48611 (800) 248-2481 FAX: (517) 496-6974 (517) 496-6000
DuPont Merck Pharmaceuticals 331 Treble Cove Road Billerica, MA 01862 (800) 225-1572 FAX: (508) 436-7501 http://www.dupontmerck.com
Dynarex 1 International Boulevard Brewster, NY 10509 (888) DYNAREX FAX: (914) 279-9601 (914) 279-9600 http://www.dynarex.com Dynatech See Dynex Technologies Dynex Technologies 14340 Sullyfield Circle Chantilly, VA 22021 (800) 336-4543 FAX: (703) 631-7816 (703) 631-7800 http://www.dynextechnologies.com Dyno Mill See Willy A. Bachofen E.S.A. 22 Alpha Road Chelmsford, MA 01824 (508) 250-7000 FAX: (508) 250-7090 E.W. Wright 760 Durham Road Guilford, CT 06437 (203) 453-6410 FAX: (203) 458-6901 http://www.ewwright.com E-Y Laboratories 107 N. Amphlett Boulevard San Mateo, CA 94401 (800) 821-0044 FAX: (650) 342-2648 (650) 342-3296 http://www.eylabs.com Eastman Kodak 1001 Lee Road Rochester, NY 14650 (800) 225-5352 FAX: (800) 879-4979 (716) 722-5780 FAX: (716) 477-8040 http://www.kodak.com ECACC See European Collection of Animal Cell Cultures
DuPont NEN Products See NEN Life Science Products
EC Apparatus See Savant/EC Apparatus
Dynal 5 Delaware Drive Lake Success, NY 11042 (800) 638-9416 FAX: (516) 326-3298 (516) 326-3270 http://www.dynal.net
Ecogen, SRL Gensura Laboratories Ptge. Dos de Maig 9(08041) Barcelona, Spain (34) 3-450-2601 FAX: (34) 3-456-0607 http://www.ecogen.com
Dynal AS Ullernchausen 52, 0379 Oslo, Norway 47-22-06-10-00 FAX: 47-22-50-70-15 http://www.dynal.no
Ecolab 370 North Wabasha Street St. Paul, MN 55102 (800) 35-CLEAN FAX: (651) 225-3098 (651) 352-5326 http://www.ecolab.com
Dynalab P.O. Box 112 Rochester, NY 14692 (800) 828-6595 FAX: (716) 334-9496 (716) 334-2060 http://www.dynalab.com
ECO PHYSICS 3915 Research Park Drive, Suite A-3 Ann Arbor, MI 48108 (734) 998-1600 FAX: (734) 998-1180 http://www.ecophysics.com
Suppliers
11 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Edge Biosystems 19208 Orbit Drive Gaithersburg, MD 20879-4149 (800) 326-2685 FAX: (301) 990-0881 (301) 990-2685 http://www.edgebio.com
EM Science 480 Democrat Road Gibbstown, NJ 08027 (800) 222-0342 FAX: (856) 423-4389 (856) 423-6300 http://www.emscience.com
ESA 22 Alpha Road Chelmsford, MA 01824 (800) 959-5095 FAX: (978) 250-7090 (978) 250-7000 http://www.esainc.com
Edmund Scientific 101 E. Gloucester Pike Barrington, NJ 08007 (800) 728-6999 FAX: (856) 573-6263 (856) 573-6250 http://www.edsci.com
EM Separations Technology See R & S Technology
Ethicon Route 22, P.O. Box 151 Somerville, NJ 08876 (908) 218-0707 http://www.ethiconinc.com
EG&G See Perkin-Elmer Ekagen 969 C Industry Road San Carlos, CA 94070 (650) 592-4500 FAX: (650) 592-4500 Elcatech P.O. Box 10935 Winston-Salem, NC 27108 (336) 544-8613 FAX: (336) 777-3623 (910) 777-3624 http://www.elcatech.com
Endogen 30 Commerce Way Woburn, MA 01801 (800) 487-4885 FAX: (617) 439-0355 (781) 937-0890 http://www.endogen.com ENGEL-Loter HSGM Heatcutting Equipment & Machines 1865 E. Main Street, No. 5 Duncan, SC 29334 (888) 854-HSGM FAX: (864) 486-8383 (864) 486-8300 http://www.engelgmbh.com
Electro Impulse, Inc. 1805 Rt. 33 (Corles Ava.) Neptune, NJ 07754 (732) 776-5800 FAX: 732-776-6793 http://www.electro-impuise.com
Enzo Diagnostics 60 Executive Boulevard Farmingdale, NY 11735 (800) 221-7705 FAX: (516) 694-7501 (516) 694-7070 http://www.enzo.com
Electron Microscopy Sciences 321 Morris Road Fort Washington, PA 19034 (800) 523-5874 FAX: (215) 646-8931 (215) 646-1566 http://www.emsdiasum.com
Enzogenetics 4197 NW Douglas Avenue Corvallis, OR 97330 (541) 757-0288
Electron Tubes 100 Forge Way, Unit F Rockaway, NJ 07866 (800) 521-8382 FAX: (973) 586-9771 (973) 586-9594 http://www.electrontubes.com Elicay Laboratory Products, (UK) Ltd. 4 Manborough Mews Crockford Lane Basingstoke, Hampshire RG 248NA, England (256) 811-118 FAX: (256) 811-116 http://www.elkay-uk.co.uk Eli Lilly Lilly Corporate Center Indianapolis, IN 46285 (800) 545-5979 FAX: (317) 276-2095 (317) 276-2000 http://www.lilly.com ELISA Technologies See Neogen Elkins-Sinn See Wyeth-Ayerst EMBI See European Bioinformatics Institute
The Enzyme Center See Charm Sciences Enzyme Systems Products 486 Lindbergh Avenue Livermore, CA 94550 (888) 449-2664 FAX: (925) 449-1866 (925) 449-2664 http://www.enzymesys.com Epicentre Technologies 1402 Emil Street Madison, WI 53713 (800) 284-8474 FAX: (608) 258-3088 (608) 258-3080 http://www.epicentre.com Erie Scientific 20 Post Road Portsmouth, NH 03801 (888) ERIE-SCI FAX: (603) 431-8996 (603) 431-8410 http://www.eriesci.com ES Industries 701 South Route 73 West Berlin, NJ 08091 (800) 356-6140 FAX: (856) 753-8484 (856) 753-8400 http://www.esind.com
Ethicon Endo-Surgery 4545 Creek Road Cincinnati, OH 45242 (800) 766-9534 FAX: (513) 786-7080 Eurogentec Parc Scientifique du Sart Tilman 4102 Seraing, Belgium 32-4-240-76-76 FAX: 32-4-264-07-88 http://www.eurogentec.com European Bioinformatics Institute Wellcome Trust Genomes Campus Hinxton, Cambridge CB10 1SD, UK (44) 1223-49444 FAX: (44) 1223-494468 European Collection of Animal Cell Cultures (ECACC) Centre for Applied Microbiology & Research Salisbury, Wiltshire SP4 0JG, UK (44) 1980-612 512 FAX: (44) 1980-611 315 http://www.camr.org.uk Evergreen Scientific 2254 E. 49th Street P.O. Box 58248 Los Angeles, CA 90058 (800) 421-6261 FAX: (323) 581-2503 (323) 583-1331 http://www.evergreensci.com Evotec Technologies GmbH Hamburg Schnackenburgallee 114 D-22525 Hamburg, Germany (49) 40 580810 FAX: (49) 40 56081488 http://www.evotec-technologies.com Exalpha Biologicals 20 Hampden Street Boston, MA 02205 (800) 395-1137 FAX: (617) 969-3872 (617) 558-3625 http://www.exalpha.com Exciton P.O. Box 31126 Dayton, OH 45437 (937) 252-2989 FAX: (937) 258-3937 http://www.exciton.com
Extrasynthese ZI Lyon Nord SA-BP62 69730 Genay, France (33) 78-98-20-34 FAX: (33) 78-98-19-45 Factor II 1972 Forest Avenue P.O. Box 1339 Lakeside, AZ 85929 (800) 332-8688 FAX: (520) 537-8066 (520) 537-8387 http://www.factor2.com Falcon See Becton Dickinson Labware Febit AG Kafertaler Strasse 190 D-66167 Mannheim Germany (49) 621-3804-0 FAX: (49) 621-3804-400 http://www.febit.com Fenwal See Baxter Healthcare Filemaker 5201 Patrick Henry Drive Santa Clara, CA 95054 (408) 987-7000 (800) 325-2747 Fine Science Tools 202-277 Mountain Highway North Vancouver, British Columbia V7J 3P2 Canada (800) 665-5355 FAX: (800) 665 4544 (604) 980-2481 FAX: (604) 987-3299 Fine Science Tools 373-G Vintage Park Drive Foster City, CA 94404 (800) 521-2109 FAX: (800) 523-2109 (650) 349-1636 FAX: (630) 349-3729 Fine Science Tools Fahrtgasse 7-13 D-69117 Heidelberg, Germany (49) 6221 905050 FAX: (49) 6221 600001 http://www.finescience.com Finn Aqua AMSCO Finn Aqua Oy Teollisuustiez, FIN-04300 Tuusula, Finland 358 025851 FAX: 358 0276019 Finnigan 355 River Oaks Parkway San Jose, CA 95134 (408) 433-4800 FAX: (408) 433-4821 http://www.finnigan.com Dr. L. Fischer Lutherstrasse 25A D-69120 Heidelberg Germany (49) 6221-16-0368 http://home.eplus-online.de/ electroporation
Suppliers
12 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Fisher Chemical Company Fisher Scientific Limited 112 Colonnade Road Nepean Ontario K2E 7L6, Canada (800) 234-7437 FAX: (800) 463-2996 http://www.fisherscientific.com Fisher Scientific 2000 Park Lane Pittsburgh, PA 15275 (800) 766-7000 FAX: (800) 926-1166 (412) 562-8300 http://www3.fishersci.com W.F. Fisher & Son 220 Evans Way, Suite #1 Somerville, NJ 08876 (908) 707-4050 FAX: (908) 707-4099 Fitzco 5600 Pioneer Creek Drive Maple Plain, MN 55359 (800) 367-8760 FAX: (612) 479-2880 (612) 479-3489 http://www.fitzco.com 5 Prime → 3 Prime See 2000 Eppendorf-5 Prime http://www.5prime.com
Fluka Chemical See Sigma-Aldrich FMC BioPolymer 1735 Market Street Philadelphia, PA 19103 (215) 299-6000 FAX: (215) 299-5809 http://www.fmc.com FMC BioProducts 191 Thomaston Street Rockland, ME 04841 (800) 521-0390 FAX: (800) 362-1133 (207) 594-3400 FAX: (207) 594-3426 http://www.bioproducts.com Forma Scientific Milcreek Road P.O. Box 649 Marietta, OH 45750 (800) 848-3080 FAX: (740) 372-6770 (740) 373-4765 http://www.forma.com Fort Dodge Animal Health 800 5th Street NW Fort Dodge, IA 50501 (800) 685-5656 FAX: (515) 955-9193 (515) 955-4600 http://www.ahp.com
Flambeau 15981 Valplast Road Middlefield, Ohio 44062 (800) 232-3474 Fax: (440) 632-1581 (400) 632-1631 http://www.flambeau.com
Fotodyne 950 Walnut Ridge Drive Hartland, WI 53029 (800) 362-3686 FAX: (800) 362-3642 (262) 369-7000 FAX: (262) 369-7013 http://www.fotodyne.com
Fleisch (Rusch) 2450 Meadowbrook Parkway Duluth, GA 30096 (770) 623-0816 FAX: (770) 623-1829 http://ruschinc.com
Fresenius HemoCare 6675 185th Avenue NE, Suite 100 Redwood, WA 98052 (800) 909-3872 (425) 497-1197 http://www.freseniusht.com
Flow Cytometry Standards P.O. Box 194344 San Juan, PR 00919 (800) 227-8143 FAX: (787) 758-3267 (787) 753-9341 http://www.fcstd.com Flow Labs See ICN Biomedicals Flow-Tech Supply P.O. Box 1388 Orange, TX 77631 (409) 882-0306 FAX: (409) 882-0254 http://www.flow-tech.com Fluid Marketing See Fluid Metering Fluid Metering 5 Aerial Way, Suite 500 Sayosett, NY 11791 (516) 922-6050 FAX: (516) 624-8261 http://www.fmipump.com Fluorochrome 1801 Williams, Suite 300 Denver, CO 80264 (303) 394-1000 FAX: (303) 321-1119
Fresenius Hemotechnology See Fresenius HemoCare Fuji Medical Systems 419 West Avenue P.O. Box 120035 Stamford, CT 06902 (800) 431-1850 FAX: (203) 353-0926 (203) 324-2000 http://www.fujimed.com Fujisawa USA Parkway Center North Deerfield, IL 60015-2548 (847) 317-1088 FAX: (847) 317-7298 Ernest F. Fullam 900 Albany Shaker Road Latham, NY 12110 (800) 833-4024 FAX: (518) 785-8647 (518) 785-5533 http://www.fullam.com Gallard-Schlesinger Industries 777 Zechendorf Boulevard Garden City, NY 11530 (516) 229-4000 FAX: (516) 229-4015 http://www.gallard-schlessinger.com
Gambro Box 7373 SE 103 91 Stockholm, Sweden (46) 8 613 65 00 FAX: (46) 8 611 37 31 In the US: COBE Laboratories 225 Union Blvd. Lakewood, CO 80215 (303) 232-6800 FAX: (303) 231-4915 http://www.gambro.com
Gemini BioProducts 5115-M Douglas Fir Road Calabasas, CA 90403 (818) 591-3530 FAX: (818) 591-7084
Garner Glass 177 Indian Hill Boulevard Claremont, CA 91711 (909) 624-5071 FAX: (909) 625-0173 http://www.garnerglass.com
Genaissance Pharmaceuticals 5 Science Park New Haven, CT 06511 (800) 678-9487 FAX: (203) 562-9377 (203) 773-1450 http://www.genaissance.com
Garon Plastics 16 Byre Avenue Somerton Park, South Australia 5044 (08) 8294-5126 FAX: (08) 8376-1487 http://www.apache.airnet.com. au/∼garon Garren Scientific 9400 Lurline Avenue, Unit E Chatsworth, CA 91311 (800) 342-3725 FAX: (818) 882-3229 (818) 882-6544 http://www.garren-scientific.com GATC Biotech AG Jakob-Stadler-Platz 7 D-78467 Constance, Germany (49) 07531-8160-0 FAX: (49) 07531-8160-81 http://www.gatc-biotech.com Gaussian Carnegie Office Park Building 6, Suite 230 Carnegie, PA 15106 (412) 279-6700 FAX: (412) 279-2118 http://www.gaussian.com G.C. Electronics/A.R.C. Electronics 431 Second Street Henderson, KY 42420 (270) 827-8981 FAX: (270) 827-8256 http://www.arcelectronics.com GDB (Genome Data Base, Curation) 2024 East Monument Street, Suite 1200 Baltimore, MD 21205 (410) 955-9705 FAX: (410) 614-0434 http://www.gdb.org GDB (Genome Data Base, Home) Hospital for Sick Children 555 University Avenue Toronto, Ontario M5G 1X8 Canada (416) 813-8744 FAX: (416) 813-8755 http://www.gdb.org Gelman Sciences See Pall-Gelman
Gen Trak 5100 Campus Drive Plymouth Meeting, PA 19462 (800) 221-7407 FAX: (215) 941-9498 (215) 825-5115 http://www.informagen.com
GENAXIS Biotechnology Parc Technologique ` 10 Avenue Ampere Montigny le Bretoneux 78180 France (33) 01-30-14-00-20 FAX: (33) 01-30-14-00-15 http://www.genaxis.com GenBank National Center for Biotechnology Information National Library of Medicine/NIH Building 38A, Room 8N805 8600 Rockville Pike Bethesda, MD 20894 (301) 496-2475 FAX: (301) 480-9241 http://www.ncbi.nlm.nih.gov Gene Codes 640 Avis Drive Ann Arbor, MI 48108 (800) 497-4939 FAX: (734) 930-0145 (734) 769-7249 http://www.genecodes.com Genemachines 935 Washington Street San Carlos, CA 94070 (650) 508-1634 FAX: (650) 508-1644 (877) 855-4363 http://www.genemachines.com Genentech 1 DNA Way South San Francisco, CA 94080 (800) 551-2231 FAX: (650) 225-1600 (650) 225-1000 http://www.gene.com General Scanning/GSI Luminomics 500 Arsenal Street Watertown, MA 02172 (617) 924-1010 FAX: (617) 924-7327 http://www.genescan.com General Valve Division of Parker Hannifin Pneutronics 19 Gloria Lane Fairfield, NJ 07004 (800) GVC-VALV FAX: (800) GVC-1-FAX http://www.pneutronics.com
Suppliers
13 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Genespan 19310 North Creek Parkway, Suite 100 Bothell, WA 98011 (800) 231-2215 FAX: (425) 482-3005 (425) 482-3003 http://www.genespan.com Gene Therapy Systems 10190 Telesis Court San Diego, CA 92122 (858) 457-1919 FAX: (858) 623-9494 http://www.genelherapysystems.com ´ ethon ´ Gen Human Genome Research Center 1 bis rue de l’Internationale 91000 Evry, France (33) 169-472828 FAX: (33) 607-78698 http://www.genethon.fr Genetic Microsystems 34 Commerce Way Wobum, MA 01801 (781) 932-9333 FAX: (781) 932-9433 http://www.genticmicro.com Genetic Mutant Repository See Coriell Institute for Medical Research Genetic Research Instrumentation Gene House Queenborough Lane Rayne, Braintree, Essex CM7 8TF, UK (44) 1376 332900 FAX: (44) 1376 344724 http://www.gri.co.uk Genetics Computer Group 575 Science Drive Madison, WI 53711 (608) 231-5200 FAX: (608) 231-5202 http://www.gcg.com Genetics Institute/American Home Products 87 Cambridge Park Drive Cambridge, MA 02140 (617) 876-1170 FAX: (617) 876-0388 http://www.genetics.com Genetix 63-69 Somerford Road Christchurch, Dorset BH23 3QA, UK (44) (0) 1202 483900 FAX: (44)(0) 1202 480289 In the US: (877) 436 3849 US FAX: (888) 522 7499 http://www.genetix.co.uk Gene Tools One Summerton Way Philomath, OR 97370 (541) 9292-7840 FAX: (541) 9292-7841 http://www.gene-tools.com
Geneva Bloinformatics (GeneBio) S.A. 25 Avenue de Champel CH--1206 Geneva, Switzerland (41) 22-702-9900 FAX: (41) 22-702-9999 http://www.genebio.com GeneWorks P.O. Box 11, Rundle Mall Adelaide, South Australia 5000, Australia 1800 882 555 FAX: (08) 8234 2699 (08) 8234 2644 http://www.geneworks.com Genome Systems (INCYTE) 4633 World Parkway Circle St. Louis, MO 63134 (800) 430-0030 FAX: (314) 427-3324 (314) 427-3222 http://www.genomesystems.com Genomic Solutions 4355 Varsity Drive, Suite E Ann Arbor, MI 48108 (877) GENOMIC FAX: (734) 975-4808 (734) 975-4800 http://www.genomicsolutions.com Genomyx See Beckman Coulter Genosys Biotechnologies 1442 Lake Front Circle, Suite 185 The Woodlands, TX 77380 (281) 363-3693 FAX: (281) 363-2212 http://www.genosys.com Genotech 92 Weldon Parkway St. Louis, MO 63043 (800) 628-7730 FAX: (314) 991-1504 (314) 991-6034 GENSET 876 Prospect Street, Suite 206 La Jolla, CA 92037 (800) 551-5291 FAX: (619) 551-2041 (619) 515-3061 http://www.genset.fr Gensia Laboratories Ltd. 19 Hughes Irvine, CA 92718 (714) 455-4700 FAX: (714) 855-8210 Genta 99 Hayden Avenue, Suite 200 Lexington, MA 02421 (781) 860-5150 FAX: (781) 860-5137 http://www.genta.com GENTEST 6 Henshaw Street Woburn, MA 01801 (800) 334-5229 FAX: (888) 242-2226 (781) 935-5115 FAX: (781) 932-6855 http://www.gentest.com
Gentra Systems 15200 25th Avenue N., Suite 104 Minneapolis, MN 55447 (800) 866-3039 FAX: (612) 476-5850 (612) 476-5858 http://www.gentra.com
Glen Research 22825 Davis Drive Sterling, VA 20166 (800) 327-4536 FAX: (800) 934-2490 (703) 437-6191 FAX: (703) 435-9774 http://www.glenresearch.com
Genzyme 1 Kendall Square Cambridge, MA 02139 (617) 252-7500 FAX: (617) 252-7600 http://www.genzyme.com See also R&D Systems
Glo Germ P.O. Box 189 Moab, UT 84532 (800) 842-6622 FAX: (435) 259-5930 http://www.glogerm.com
Genzyme Genetics One Mountain Road Framingham, MA 01701 (800) 255-7357 FAX: (508) 872-9080 (508) 872-8400 http://www.genzyme.com
Glyco 11 Pimentel Court Novato, CA 94949 (800) 722-2597 FAX: (415) 382-3511 (415) 884-6799 http://www.glyco.com
George Tiemann & Co. 25 Plant Avenue Hauppauge, NY 11788 (516) 273-0005 FAX: (516) 273-6199
Gould Instrument Systems 8333 Rockside Road Valley View, OH 44125 (216) 328-7000 FAX: (216) 328-7400 http://www.gould13.com
GIBCO/BRL A Division of Life Technologies 1 Kendall Square Grand Island, NY 14072 (800) 874-4226 FAX: (800) 352-1968 (716) 774-6700 http://www.lifetech.com Gilmont Instruments A Division of Barnant Company 28N092 Commercial Avenue Barrington, IL 60010 (800) 637-3739 FAX: (708) 381-7053 http://barnant.com
Gralab Instruments See Dimco-Gray GraphPad Software 5755 Oberlin Drive #110 San Diego, CA 92121 (800) 388-4723 FAX: (558) 457-8141 (558) 457-3909 http://www.graphpad.com Graseby Anderson See Andersen Instruments http://www.graseby.com
Gilson 3000 West Beltline Highway P.O. Box 620027 Middletown, WI 53562 (800) 445-7661 (608) 836-1551 http://www.gilson.com
Grass Instrument A Division of Astro-Med 600 East Greenwich Avenue W. Warwick, RI 02893 (800) 225-5167 FAX: (877) 472-7749 http://www.grassinstruments.com
Glas-Col Apparatus P.O. Box 2128 Terre Haute, IN 47802 (800) Glas-Col FAX: (812) 234-6975 (812) 235-6167 http://www.glascol.com
Graticules Ltd. Division of SPI Supplies 569 East Gay Street West Chester, PA 19380 800-242-4SPI/610-436-5400 FAX: 610-436-5755
Glaxo Wellcome Five Moore Drive Research Triangle Park, NC 27709 (800) SGL-AXO5 FAX: (919) 248-2386 (919) 248-2100 http://www.glaxowellcome.com
Greenacre and Misac Instruments Misac Systems 27 Port Wood Road Ware, Hertfordshire SF12 9NJ, UK (44) 1920 463017 FAX: (44) 1920 465136
Glen Mills 395 Allwood Road Clifton, NJ 07012 (973) 777-0777 FAX: (973) 777-0070 http://www.glenmills.com
Greer Labs 639 Nuway Circle Lenois, NC 28645 (704) 754-5237 http://greerlabs.com
Suppliers
14 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Greiner Maybachestrasse 2 Postfach 1162 D-7443 Frickenhausen, Germany (49) 0 91 31/80 79 0 FAX: (49) 0 91 31/80 79 30 http://www.erlangen.com/greiner GSI Lumonics 130 Lombard Street Oxnard, CA 93030 (805) 485-5559 FAX: (805) 485-3310 http://www.gsilumonics.com GTE Internetworking 150 Cambridge Park Drive Cambridge, MA 02140 (800) 472-4565 FAX: (508) 694-4861 http://www.bbn.com GW Instruments 35 Medford Street Somerville, MA 02143 (617) 625-4096 FAX: (617) 625-1322 http://www.gwinst.com H & H Woodworking 1002 Garfield Street Denver, CO 80206 (303) 394-3764 Hacker Instruments 17 Sherwood Lane P.O. Box 10033 Fairfield, NJ 07004 800-442-2537 FAX: (973) 808-8281 (973) 226-8450 http://www.hackerinstruments.com Haemenetics 400 Wood Road Braintree, MA 02184 (800) 225-5297 FAX: (781) 848-7921 (781) 848-7100 http://www.haemenetics.com Halocarbon Products P.O. Box 661 River Edge, NJ 07661 (201) 242-8899 FAX: (201) 262-0019 http://halocarbon.com Hamamatsu Photonic Systems A Division of Hamamatsu 360 Foothill Road P.O. Box 6910 Bridgewater, NJ 08807 (908) 231-1116 FAX: (908) 231-0852 http://www.photonicsonline.com Hamilton Company 4970 Energy Way P.O. Box 10030 Reno, NV 89520 (800) 648-5950 FAX: (775) 856-7259 (775) 858-3000 http://www.hamiltoncompany.com Hamilton Thorne Biosciences 100 Cummings Center, Suite 102C Beverly, MA 01915 http://www.hamiltonthorne.com
Hampton Research 27631 El Lazo Road Laguna Niguel, CA 92677 (800) 452-3899 FAX: (949) 425-1611 (949) 425-6321 http://www.hamptonresearch.com
Hellma Cells 11831 Queens Boulevard Forest Hills, NY 11375 (718) 544-9166 FAX: (718) 263-6910 http://www.helmaUSA.com
Harlan Bioproducts for Science P.O. Box 29176 Indianapolis, IN 46229 (317) 894-7521 FAX: (317) 894-1840 http://www.hbps.com
Hellma Postfach 1163 ¨ D-79371 Mullheim/Baden, Germany (49) 7631-1820 FAX: (49) 7631-13546 http://www.hellma-worldwide.de
Harlan Sera-Lab Hillcrest, Dodgeford Lane Belton, Loughborough Leicester LE12 9TE, UK (44) 1530 222123 FAX: (44) 1530 224970 http://www.harlan.com
Henry Schein 135 Duryea Road, Mail Room 150 Melville, NY 11747 (800) 472-4346 FAX: (516) 843-5652 http://www.henryschein.com
Harlan Teklad P.O. Box 44220 Madison, WI 53744 (608) 277-2070 FAX: (608) 277-2066 http://www.harlan.com Harrlck Scientific Corporation 88 Broadway Ossining. NY 10562 (914) 762-0020 FAX: (914) 762-0914 http://www.hamicksci.com Harrison Research 840 Moana Court Palo Alto, CA 94306 (650) 949-1565 FAX: (650) 948-0493 Harvard Apparatus 84 October Hill Road Holliston, MA 01746 (800) 272-2775 FAX: (508) 429-5732 (508) 893-8999 http://harvardapparatus.com Harvard Bioscience See Harvard Apparatus Haselton Biologics See JRH Biosciences Hazelton Research Products See Covance Research Products Health Products See Pierce Chemical Heat Systems-Ultrasonics 1938 New Highway Farmingdale, NY 11735 (800) 645-9846 FAX: (516) 694-9412 (516) 694-9555 Heidenhain Corp 333 East State Parkway Schaumberg, IL 60173 (847) 490-1191 FAX: (847) 490-3931 http://www.heidenhain.com HEKA Instruments 33 Valley Rd. Soulhboro, MA 01960 (866) 742-0606 FAX: (508) 481-8945 http://www.heka.com
Hitachi Scientific Instruments Nissei Sangyo America 8100 N. First Street San Elsa, CA 95314 (800) 548-9001 FAX: (408) 432-0704 (408) 432-0520 http://www.hii.hitachi.com Hi-Tech Scientific Brunel Road Salisbury, Wiltshire, SP2 7PU UK (44) 1722-432320 (800) 344-0724 (US only) http://www.hi-techsci.co.uk Hoechst AG See Aventis Pharmaceutical
Heraeus Amersil Ouartzstrasse 8 Hanau D-63450, Germany (49) 61 8135 50 (49) 61 813535 50 http://www.heraeus.amersil.com
Hoefer Scientific Instruments Division of Amersham-Pharmacia Biotech 800 Centennial Avenue Piscataway, NJ 08855 (800) 227-4750 FAX: (877) 295-8102 http://www.apbiotech.com
Heraeus Kulzer 4315 South Lafayette Boulevard South Bend, IN 46614 (800) 343-5336 (219) 291-0661 http://www.kulzer.com
Hoffman-LaRoche 340 Kingsland Street Nutley, NJ 07110 (800) 526-0189 FAX: (973) 235-9605 (973) 235-5000 http://www.rocheUSA.com
Heraeus Sepatech See Kendro Laboratory Products Hercules Aqualon Aqualon Division Hercules Research Center, Bldg. 8145 500 Hercules Road Wilmington, DE 19899 (800) 345-0447 FAX: (302) 995-4787 http://www.herc.com/aqualon/pharma Heto-Holten A/S Gydevang 17-19 DK-3450 Allerod, Denmark (45) 48-16-62-00 FAX: (45) 48-16-62-97 Distributed by ATR
Holborn Surgical and Medical Instruments Westwood Industrial Estate Ramsgate Road Margate, Kent CT9 4JZ UK (44) 1843 296666 FAX: (44) 1843 295446 Honeywell 101 Columbia Road Morristown, NJ 07962 (973) 455-2000 FAX: (973) 455-4807 http://www.honeywell.com
Hettich-Zentrifugen See Andreas Hettich
Honeywll Specialty Films P.O. Box 1039 101 Columbia Road Morristown, NJ 07962 (800) 934-5679 FAX: (973) 455-6045 http://www.honeywell-specialtyfilms. com
Hewlett-Packard 3000 Hanover Street Mailstop 20B3 Palo Alto, CA 94304 (650) 857-1501 FAX: (650) 857-5518 http://www.hp.com
Hood Thermo-Pad Canada Comp. 20, Site 61A, RR2 Summerland, British Columbia V0H 1Z0 Canada (800) 665-9555 FAX: (250) 494-5003 (250) 494-5002 http://www.thermopad.com
HGS Hinimoto Plastics 1-10-24 Meguro-Honcho Megurouko Tokyo 152, Japan 3-3714-7226 FAX: 3-3714-4657
Horiba Instruments 17671 Armstrong Avenue Irvine, CA 92714 (949) 250-4811 FAX: (949) 250-0924 http://www.horiba.com
Suppliers
15 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Hoskins Manufacturing 10776 Hall Road P.O. Box 218 Hamburg, MI 48139 (810) 231-1900 FAX: (810) 231-4311 http://www.hoskinsmfgco.com Hosokawa Micron Powder Systems 10 Chatham Road Summit, NJ 07901 (800) 526-4491 FAX: (908) 273-7432 (908) 273-6360 http://www.hosokawamicron.com HT Biotechnology Unit 4 61 Ditton Walk Cambridge CB5 8QD, UK (44) 1223-412583 Hugo Sachs Electronik Postfach 138 7806 March-Hugstetten, Germany D-79229(49) 7665-92000 FAX: (49) 7665-920090 Human Biologics International 7150 East Camelback Road, Suite 245 Scottsdale, AZ 85251 (480) 990-2005 FAX: (480)-990-2155 http://www.humanbiological.com Human Genetic Mutant Cell Repository See Coriell Institute for Medical Research HVS Image P.O. Box 100 Hampton, Middlesex TW12 2YD, UK FAX: (44) 208 783 1223 In the US: (800) 225-9261 FAX: (888) 483-8033 http://www.hvsimage.com Hybaid 111-113 Waldegrave Road Teddington, Middlesex TW11 8LL, UK (44) 0 1784 42500 FAX: (44) 0 1784 248085 http://www.hybaid.co.uk Hybaid Instruments 8 East Forge Parkway Franklin, MA 02028 (888)4-HYBAID FAX: (508) 541-3041 (508) 541-6918 http://www.hybaid.com Hybridon 155 Fortune Boulevard Milford, MA 01757 (508) 482-7500 FAX: (508) 482-7510 http://www.hybridon.com HyClone Laboratories 1725 South HyClone Road Logan, UT 84321 (800) HYCLONE FAX: (800) 533-9450 (801) 753-4584 FAX: (801) 750-0809 http://www.hyclone.com
Hyseq 670 Almanor Avenue Sunnyvale, CA 94086 (408) 524-8100 FAX: (408) 524-8141 http://www.hyseq.com IBA GmbH 1508 South Grand Blvd. St Louis, MO 63104 (877) 422-4624 FAX: (888) 531-6813 http://www.iba-go.com IBF Biotechnics See Sepracor IBI (International Biotechnologies) See Eastman Kodak For technical service (800) 243-2555 (203) 786-5600 ICN Biochemicals See ICN Biomedicals ICN Biomedicals 3300 Hyland Avenue Costa Mesa, CA 92626 (800) 854-0530 FAX: (800) 334-6999 (714) 545-0100 FAX: (714) 641-7275 http://www.icnbiomed.com ICN Flow and Pharmaceuticals See ICN Biomedicals ICN Immunobiochemicals See ICN Biomedicals ICN Radiochemicals See ICN Biomedicals ICONIX 100 King Street West, Suite 3825 Toronto, Ontario M5X 1E3 Canada (416) 410-2411 FAX: (416) 368-3089 http://www.iconix.com ICRT (Imperial Cancer Research Technology) Sardinia House Sardinia Street London WC2A 3NL, UK (44) 1712-421136 FAX: (44) 1718-314991
Ikegami Electronics 37 Brook Avenue Maywood, NJ 07607 (201) 368-9171 FAX: (201) 569-1626 Ikemoto Scientific Technology 25-11 Hongo 3-chome, Bunkyo-ku Tokyo 101-0025, Japan (81) 3-3811-4181 FAX: (81) 3-3811-1960 Imagenetics See ATC Diagnostics Imaging Research c/o Brock University 500 Glenridge Avenue St. Catharines, Ontario L2S 3A1 Canada (905) 688-2040 FAX: (905) 685-5861 http://www.imaging.brocku.ca Imclone Systems 180 Varick Street New York, NY 10014 (212) 645-1405 FAX: (212) 645-2054 http://www.imclone.com IMCO Corporation LTD., AB P.O. Box 21195 SE-100 31 Stockholm, Sweden 46-8-33-53-09 FAX: 46-8-728-47-76 http://www.imcocorp.se Imgenex Corporation 11175 Flintkole Avenue Suite E San Diego, CA 92121 (888) 723-4363 FAX: (850) 642-0937 (858) 642.0978 http://www.Imgenex.com
Immunolex Therapeutics ApS Grusbakken 6-8 DK-2820 Gentotte Denmark (45) 43-62-63-32 FAX: (56) 43-62-63-31 http://www.immunolex.com Immunotech 130, av. Delattre de Tassigny B.P. 177 13276 Marseilles Cedex 9 France (33) 491-17-27-00 FAX: (33) 491-41-43-58 http://www.immunotech.fr Imperial Chemical Industries Imperial Chemical House Millbank, London SW1P 3JF, UK (44) 171-834-4444 FAX: (44)171-834-2042 http://www.ici.com Inceltech See New Brunswick Scientific Incstar See DiaSorin Incyte 6519 Dumbarton Circle Fremont, CA 94555 (510) 739-2100 FAX: (510) 739-2200 http://www.incyte.com Incyte Pharmaceuticals 3160 Porter Drive Palo Alto, CA 94304 (877) 746-2983 FAX: (650) 855-0572 (650) 855-0555 http://www.incyte.com
IMICO Calle Vivero, No. 5-4a Planta E-28040, Madrid, Spain (34) 1-535-3960 FAX: (34) 1-535-2780
Individual Monitoring Systems 6310 Harford Road Baltimore, MD 21214
Idea Scientific Company P.O. Box 13210 Minneapolis, MN 55414 (800) 433-2535 FAX: (612) 331-4217 http://www.ideascientific.com
Immunex 51 University Street Seattle, WA 98101 (206) 587-0430 FAX: (206) 587-0606 http://www.immunex.com
Indo Fine Chemical P.O. Box 473 Somerville, NJ 08876 (888) 463-6346 FAX: (908) 359-1179 (908) 359-6778 http://www.indofinechemical.com
IEC See International Equipment Co.
Immunochemistry Technologies 9401 James Avenue, South Suite 155 Bloomington, MN 55431 (800) 829-3194 FAX: (952) 888-8988 (952) 888-8788 http://www.immunochemistry.com
Industrial Acoustics 1160 Commerce Avenue Bronx, NY 10462 (718) 931-8000 FAX: (718) 863-1138 http://www.industrialacoustics.com
IITC 23924 Viclory Boulevard Woodland Hills, CA 91387 (888) 414-4482 (818) 710-1556 FAX: (818) 992-5165 http://www.itcinc.com IKA Works 2635 N. Chase Parkway, SE Wilmington, NC 28405 (910) 452-7059 FAX: (910) 452-7693 http://www.ika.net
Immunocorp 1582 W. Deere Avenue Suite C Irvine, CA 92606 (800) 446-3063 http://www.immunocorp.com
Inex Pharmaceuticals 100-8900 Glenlyon Parkway Glenlyon Business Park Burnaby, British Columbia V5J 5J8 Canada (604) 419-3200 FAX: (604) 419-3201 http://www.inexpharm.com
Suppliers
16 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Ingold, Mettler, Toledo 261 Ballardvale Street Wilmington, MA 01887 (800) 352-8763 FAX: (978) 658-0020 (978) 658-7615 http://www.mt.com Innogenetics N.V. Technologie Park 6 B-9052 Zwijnaarde Belgium (32) 9-329-1329 FAX: (32) 9-245-7623 http://www.innogenetics.com Innovative Medical Services 1725 Gillespie Way El Cajon, CA 92020 (619) 596-8600 FAX: (619) 596-8700 http://www.imspure.com Innovative Research 3025 Harbor Lane N, Suite 300 Plymouth, MN 55447 (612) 519-0105 FAX: (612) 519-0239 http://www.inres.com Innovative Research of America 2 N. Tamiami Trail, Suite 404 Sarasota, FL 34236 (800) 421-8171 FAX: (800) 643-4345 (941) 365-1406 FAX: (941) 365-1703 http://www.innovrsrch.com Inotech Biosystems 15713 Crabbs Branch Way, #110 Rockville, MD 20855 (800) 635-4070 FAX: (301) 670-2859 (301) 670-2850 http://www.inotechintl.com INOVISION 22699 Old Canal Road Yorba Linda, CA 92887 (714) 998-9600 FAX: (714) 998-9666 http://www.inovision.com Instech Laboratories 5209 Militia Hill Road Plymouth Meeting, PA 19462 (800) 443-4227 FAX: (610) 941-0134 (610) 941-0132 http://www.instechlabs.com Instron 100 Royall Street Canton, MA 02021 (800) 564-8378 FAX: (781) 575-5725 (781) 575-5000 http://www.instron.com Instrumentarium P.O. Box 300 00031 Instrumentarium Helsinki, Finland (10) 394-5566 http://www.instrumentarium.fi
Instruments SA Division Jobin Yvon 16-18 Rue du Canal 91165 Longjumeau, Cedex, France (33)1 6454-1300 FAX: (33)1 6909-9319 http://www.isainc.com Instrutech 20 Vanderventer Avenue, Suite 101E Port Washington, NY 11050 (516) 883-1300 FAX: (516) 883-1558 http://www.instrutech.com Integrated DNA Technologies 1710 Commercial Park Coralville, Iowa 52241 (800) 328-2661 FAX: (319) 626-8444 http://www.idtdna.com Integrated Genetics See Genzyme Genetics Integrated Scientific Imaging Systems 3463 State Street, Suite 431 Santa Barbara, CA 93105 (805) 692-2390 FAX: (805) 692-2391 http://www.imagingsystems.com Integrated Separation Systems (ISS) See OWL Separation Systems IntelliGenetics See Oxford Molecular Group Interactiva BioTechnologie Sedanstrasse 10 D-89077 Ulm, Germany (49) 731-93579-290 FAX: (49) 731-93579-291 http://www.interactiva.de Interchim 213 J.F. Kennedy Avenue B.P. 1140 Montlucon 03103 France (33) 04-70-03-83-55 FAX: (33) 04-70-03-93-60 Interfocus 14/15 Spring Rise Falcover Road Haverhill, Suffolk CB9 7XU, UK (44) 1440 703460 FAX: (44) 1440 704397 http://www.interfocus.ltd.uk Intergen 2 Manhattanville Road Purchase, NY 10577 (800) 431-4505 FAX: (800) 468-7436 (914) 694-1700 FAX: (914) 694-1429 http://www.intergenco.com Intermountain Scientific 420 N. Keys Drive Kaysville, UT 84037 (800) 999-2901 FAX: (800) 574-7892 (801) 547-5047 FAX: (801) 547-5051 http://www.bioexpress.com
International Biotechnologies (IBI) See Eastman Kodak International Equipment Co. (IEC) See Thermoquest International Institute for the Advancement of Medicine 1232 Mid-Valley Drive Jessup, PA 18434 (800) 486-IIAM FAX: (570) 343-6993 (570) 496-3400 http://www.iiam.org International Light 17 Graf Road Newburyport, MA 01950 (978) 465-5923 FAX: (978) 462-0759 International Market Supply (I.M.S.) Dane Mill Broadhurst Lane Congleton, Cheshire CW12 1LA, UK (44) 1260 275469 FAX: (44) 1260 276007
ISC BioExpress 420 North Kays Drive Kaysville, UT 84037 (800) 999-2901 FAX: (800) 574-7892 (801) 547-5047 http://www.bioexpress.com ISCO P.O. Box 5347 4700 Superior Lincoln, NE 68505 (800) 228-4373 FAX: (402) 464-0318 (402) 464-0231 http://www.isco.com Isis Pharmaceuticals Carlsbad Research Center 2292 Faraday Avenue Carlsbad, CA 92008 (760) 931-9200 http://www.isip.com Isolabs See Wallac
International Marketing Services See International Marketing Ventures
ISS See Integrated Separation Systems
International Marketing Ventures 6301 Ivy Lane, Suite 408 Greenbelt, MD 20770 (800) 373-0096 FAX: (301) 345-0631 (301) 345-2866 http://www.imvlimited.com
J & W Scientific See Agilent Technologies
International Products 201 Connecticut Drive Burlington, NJ 08016 (609) 386-8770 FAX: (609) 386-8438 http://[email protected] Intracel Corporation Bartels Division 2005 Sammamish Road, Suite 107 Issaquah, WA 98027 (800) 542-2281 FAX: (425) 557-1894 (425) 392-2992 http://www.intracel.com Invitrogen 1600 Faraday Avenue Carlsbad, CA 92008 (800) 955-6288 FAX: (760) 603-7201 (760) 603-7200 http://www.invitrogen.com In Vivo Metric P.O. Box 249 Healdsburg, CA 95448 (707) 433-4819 FAX: (707) 433-2407 IRORI 9640 Towne Center Drive San Diego, CA 92121 (858) 546-1300 FAX: (858) 546-3083 http://www.irori.com Irvine Scientific 2511 Daimler Street Santa Ana, CA 92705 (800) 577-6097 FAX: (949) 261-6522 (949) 261-7800 http://www.irvinesci.com
J.A. Webster 86 Leominster Road Sterling, MA 01564 (800) 225-7911 FAX: (978) 422-8959 http://www.jawebster.com J.T. Baker See Mallinckrodt Baker 222 Red School Lane Phillipsburg, NJ 08865 (800) JTBAKER FAX: (908) 859-6974 http://www.jtbaker.com Jackson ImmunoResearch Laboratories P.O. Box 9 872 W. Baltimore Pike West Grove, PA 19390 (800) 367-5296 FAX: (610) 869-0171 (610) 869-4024 http://www.jacksonimmuno.com The Jackson Laboratory 600 Maine Street Bar Harbor, ME 04059 (800) 422-6423 FAX: (207) 288-5079 (207) 288-6000 http://www.jax.org Jaece Industries 908 Niagara Falls Boulevard North Tonawanda, NY 14120 (716) 694-2811 FAX: (716) 694-2811 http://www.jaece.com Jandel Scientific See SPSS Janke & Kunkel See Ika Works Janssen Life Sciences Products See Amersham
Suppliers
17 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Janssen Pharmaceutica 1125 Trenton-Harbourton Road Titusville, NJ 09560 (609) 730-2577 FAX: (609) 730-2116 http://us.janssen.com
Jolley Consulting and Research 683 E. Center Street, Unit H Grayslake, IL 60030 (847) 548-2330 FAX: (847) 548-2984 http://www.jolley.com
Jasco 8649 Commerce Drive Easton, MD 21601 (800) 333-5272 FAX: (410) 822-7526 (410) 822-1220 http://www.jascoinc.com
Jordan Scientific See Shelton Scientific
Jeffers P.O. Box 100 Dothan, AL 36302 (800) 533-3377 FAX: 1-334-793-5179 http://www.1800jeffers.com/ssc/ Jena Bioscience Loebstedter Str. 78 07749 Jena, Germany (49) 3641-464920 FAX: (49) 3641-464991 http://www.jenabioscience.com Jencons Scientific 800 Bursca Drive, Suite 801 Bridgeville, PA 15017 (800) 846-9959 FAX: (412) 257-8809 (412) 257-8861 http://www.jencons.co.uk JEOL Instruments 11 Dearborn Road Peabody, MA 01960 (978) 535-5900 FAX: (978) 536-2205 http://www.jeol.com/index.html Jewett 750 Grant Street Buffalo, NY 14213 (800) 879-7767 FAX: (716) 881-6092 (716) 881-0030 http://www.JewettInc.com John’s Scientific See VWR Scientific
Jorgensen Laboratories 1450 N. Van Buren Avenue Loveland, CO 80538 (800) 525-5614 FAX: (970) 663-5042 (970) 669-2500 http://www.jorvet.com JRH Biosciences and JR Scientific 13804 W. 107th Street Lenexa, KS 66215 (800) 231-3735 FAX: (913) 469-5584 (913) 469-5580 Jule Bio Technologies 25 Science Park, #14, Suite 695 New Haven, CT 06511 (800) 648-1772 FAX: (203) 786-5489 (203) 786-5490 http://hometown.aol.com/precastgel/ index.htm K.R. Anderson 2800 Bowers Avenue Santa Clara, CA 95051 (800) 538-8712 FAX: (408) 727-2959 (408) 727-2800 http://www.kranderson.com Kabi Pharmacia Diagnostics See Pharmacia Diagnostics Kanthal H.P. Reid 1 Commerca Boulevard P.O. Box 352440 Palm Coast, FL 32135 (904) 445-2000 FAX: (904) 446-2244 http://www.kanthal.com
John Weiss and Sons 95 Alston Drive Bradwell Abbey Milton Keynes, Buckinghamshire MK1 4HF UK (44) 1908-318017 FAX: (44) 1908-318708
Kapak 5305 Parkdale Drive St. Louis Park, MN 55416 (800) KAPAK-57 FAX: (612) 541-0735 (612) 541-0730 http://www.kapak.com
Johnson & Johnson Medical 2500 Arbrook Boulevard East Arlington, TX 76004 (800) 423-4018 http://www.jnjmedical.com
Karl Hecht Stettener Str. 22-24 D-97647 Sondheim ¨ Germany Rhon, (49) 9779-8080 FAX: (49) 9779-80888
Johnston Matthey Chemicals Orchard Road Royston, Hertfordshire SG8 5HE, UK (44) 1763-253000 FAX: (44) 1763-253466 http://www.chemicals.matthey.com
Karl Storz ¨ Koningin-Elisabeth Str. 60 D-14059 Berlin, Germany (49) 30-30 69 09-0 FAX: (49) 30-30 19 452 http://www.karlstorz.de
KaVo EWL P.O. Box 1320 ¨ Germany D-88293 Leutkirch im Allgau, (49) 7561-86-0 FAX: (49) 7561-86-371 http://www.kavo.com/english/ startseite.htm
Keystone Scientific Penn Eagle Industrial Park 320 Rolling Ridge Drive Bellefonte, PA 16823 (800) 437-2999 FAX: (814) 353-2305 (814) 353-2300 Ext 1 http://www.keystonescientific.com
Keithley Instruments 28775 Aurora Road Cleveland, OH 44139 (800) 552-1115 FAX: (440) 248-6168 (440) 248-0400 http://www.keithley.com
Kimble/Kontes Biotechnology 1022 Spruce Street P.O. Box 729 Vineland, NJ 08360 (888) 546-2531 FAX: (856) 794-9762 (856) 692-3600 http://www.kimble-kontes.com
Kemin 2100 Maury Street, Box 70 Des Moines, IA 50301 (515) 266-2111 FAX: (515) 266-8354 http://www.kemin.com Kemo 3 Brook Court, Blakeney Road Beckenham, Kent BR3 1HG, UK (44) 0181 658 3838 FAX: (44) 0181 658 4084 http://www.kemo.com Kendall 15 Hampshire Street Mansfield, MA 02048 (800) 962-9888 FAX: (800) 724-1324 http://www.kendallhq.com Kendro Laboratory Products 31 Pecks Lane Newtown, CT 06470 (800) 522-SPIN FAX: (203) 270-2166 (203) 270-2080 http://www.kendro.com Kendro Laboratory Products P.O. Box 1220 Am Kalkberg D-3360 Osterod, Germany (55) 22-316-213 FAX: (55) 22-316-202 http://www.heraeus-instruments.de Kent Laboratories 23404 NE 8th Street Redmond, WA 98053 (425) 868-6200 FAX: (425) 868-6335 http://www.kentlabs.com Kent Scientific 457 Bantam Road, #16 Litchfield, CT 06759 (888) 572-8887 FAX: (860) 567-4201 (860) 567-5496 http://www.kentscientific.com Kentek Corp. 1 Elm Street Pittsfield, NH 03263 (800) 432-2323 FAX: 603-435-7441 http://www.kentek-laser.com Keuffel & Esser See Azon
Kinematica AG Luzernerstrasse 147a CH-6014 Littau-Luzern, Switzerland (41) 41 2501257 FAX: (41) 41 2501460 http://www.kinematica.ch Kinetic Imaging Ltd. One Orchard Place Nottingham Business Park Nottingham NG8 6PX, UK 44-115-973-9027 FAX: 44-115-973-9021 http://www.Kineticimaging.com Kin-Tek 504 Laurel Street LaMarque, TX 77568 (800) 326-3627 FAX: (409) 938-3710 http://www.kin-tek.com Kipp & Zonen 125 Wilbur Place Bohemia, NY 11716 (800) 645-2065 FAX: (516) 589-2068 (516) 589-2885 http://www.kippzonen.thomasregister. com/olc/kippzonen Kirkegaard & Perry Laboratories 2 Cessna Court Gaithersburg, MD 20879 (800) 638-3167 FAX: (301) 948-0169 (301) 948-7755 http://www.kpl.com Kodak See Eastman Kodak Kontes Glass See Kimble/Kontes Biotechnology Kontron Instruments AG Postfach CH-8010 Zurich, Switzerland 41-1-733-5733 FAX: 41-1-733-5734 David Kopf Instruments P.O. Box 636 Tujunga, CA 91043 (818) 352-3274 FAX: (818) 352-3139 Kraft Apparatus See Glas-Col Apparatus
Suppliers
18 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Kramer Scientific Corporation 711 Executive Boulevard Valley Cottage, NY 10989 (845) 267-5050 FAX: (845) 267-5550 Kulite Semiconductor Products 1 Willow Tree Road Leonia, NJ 07605 (201) 461-0900 FAX: (201) 461-0990 http://www.kulite.com Lab-Line Instruments 15th & Bloomingdale Avenues Melrose Park, IL 60160 (800) LAB-LINE FAX: (708) 450-5830 FAX: (800) 450-4LAB http://www.labline.com Lab Products 742 Sussex Avenue P.O. Box 639 Seaford, DE 19973 (800) 526-0469 FAX: (302) 628-4309 (302) 628-4300 http://www.labproductsinc.com LabRepco 101 Witmer Road, Suite 700 Horsham, PA 19044 (800) 521-0754 FAX: (215) 442-9202 http://www.labrepco.com Lab Safety Supply P.O. Box 1368 Janesville, WI 53547 (800) 356-0783 FAX: (800) 543-9910 (608) 754-7160 FAX: (608) 754-1806 http://www.labsafety.com Lab-Tek Products See Nalge Nunc International Labconco 8811 Prospect Avenue Kansas City, MO 64132 (800) 821-5525 FAX: (816) 363-0130 (816) 333-8811 http://www.labconco.com Labindustries See Barnstead/Thermolyne Labnet International P.O. Box 841 Woodbridge, NJ 07095 (888) LAB-NET1 FAX: (732) 417-1750 (732) 417-0700 http://www.nationallabnet.com LABO-MODERNE 37 rue Dombasle Paris 75015 France (33) 01-45-32-62-54 FAX: (33) 01-45-32-01-09 http://www.labomoderne.com/fr
Laboratory of Immunoregulation National Institute of Allergy and Infectious Diseases/NIH 9000 Rockville Pike Building 10, Room 11B13 Bethesda, MD 20892 (301) 496-1124 Laboratory Supplies 29 Jefry Lane Hicksville, NY 11801 (516) 681-7711 Labscan Limited Stillorgan Industrial Park Stillorgan Dublin, Ireland (353) 1-295-2684 FAX: (353) 1-295-2685 http://www.labscan.ie Labsystems See Thermo Labsystems Labsystems Affinity Sensors Saxon Way, Bar Hill Cambridge CB3 8SL, UK 44 (0) 1954 789976 FAX: 44 (0) 1954 789417 http://www.affinity-sensors.com
Lancer 140 State Road 419 Winter Springs, FL 32701 (800) 332-1855 FAX: (407) 327-1229 (407) 327-8488 http://www.lancer.com
Leica 111 Deer Lake Road Deerfield, IL 60015 (800) 248-0123 FAX: (847) 405-0147 (847) 405-0123 http://www.leica.com
LaVision GmbH Gerhard-Gerdes-Str. 3 D-37079 Goettingen, Germany (49) 551-50549-0 FAX: (49) 551-50549-11 http://www.lavision.de
Leica Microsystems Imneuenheimer Feld 518 D-69120 Heidelberg, Germany (49) 6221-41480 FAX: (49) 6221-414833 http://www.leica-microsystems.com
Lawshe See Advanced Process Supply
Leinco Technologies 359 Consort Drive St. Louis, MO 63011 (314) 230-9477 FAX: (314) 527-5545 http://www.leinco.com
Lexotan 20, rue Leon Blum 26000 Valence, France (33) 4-75-41-91-91 FAX: (33) 4-75-41-91-98 http://www.latoxan.com LC Laboratories 165 New Boston Street Woburn, MA 01801 (781) 937-0777 FAX: (781) 938-5420 http://www.lclaboratories.com
Labtronics 546 Governors Road Guelph, Ontario N1K 1E3, Canada (519) 763-4930 FAX: (519) 836-4431 http://www.labtronics.com
LC Packings 80 Carolina Street San Francisco, CA 94103 (415) 552-1855 FAX: (415) 552-1859 http://www.lcpackings.com
Labtronix Manufacturing 3200 Investment Boulevard Hayward, CA 94545 (510) 786-3200 FAX: (510) 786-3268 http://www.labtronix.com
LC Services See LC Laboratories
Lafayette Instrument 3700 Sagamore Parkway North P.O. Box 5729 Lafayette, IN 47903 (800) 428-7545 FAX: (765) 423-4111 (765) 423-1505 http://www.lafayetteinstrument.com Lambert Instruments Turfweg 4 9313 TH Leutingewolde The Netherlands (31) 50-5018461 FAX: (31) 50-5010034 http://www.lambert-instruments.com
LECO 3000 Lakeview Avenue St. Joseph, MI 49085 (800) 292-6141 FAX: (616) 982-8977 (616) 985-5496 http://www.leco.com Lederle Laboratories See Wyeth-Ayerst Lee Biomolecular Research Laboratories 11211 Sorrento Valley Road, Suite M San Diego, CA 92121 (858) 452-7700
Lampire Biological Laboratories P.O. Box 270 Pipersville, PA 18947 (215) 795-2538 Fax: (215) 795-0237 http://www.lampire.com
The Lee Company 2 Pettipaug Road P.O. Box 424 Westbrook, CT 06498 (800) LEE-PLUG FAX: (860) 399-7058 (860) 399-6281 http://www.theleeco.com
Lancaster Synthesis P.O. Box 1000 Windham, NH 03087 (800) 238-2324 FAX: (603) 889-3326 (603) 889-3306 http://www.lancastersynthesis-us.com
Lee Laboratories 1475 Athens Highway Grayson, GA 30017 (800) 732-9150 FAX: (770) 979-9570 (770) 972-4450 http://www.leelabs.com
Leitz U.S.A. See Leica LenderKing Metal Products 8370 Jumpers Hole Road Millersville, MD 21108 (410) 544-8795 FAX: (410) 544-5069 http://www.lenderking.com Letica Scientific Instruments Panlab s.i., c/Loreto 50 08029 Barcelona, Spain (34) 93-419-0709 FAX: (34) 93-419-7145 www.panlab-sl.com Lexel Laster 853 Brown Road Fremont, CA 94539 (510)-651-0110 FAX: 510-651-1690 http://www.lexeliaser.com Leybold-Heraeus Trivac DZA 5700 Mellon Road Export, PA 15632 (412) 327-5700 LI-COR Biotechnology Division 4308 Progressive Avenue Lincoln, NE 68504 (800) 645-4267 FAX: (402) 467-0819 (402) 467-0700 http://www.licor.com Life Science Laboratories See Adaptive Biosystems Life Science Resources Two Corporate Center Drive Melville, NY 11747 (800) 747-9530 FAX: (516) 844-5114 (516) 844-5085 http://www.astrocam.com Life Sciences 2900 72nd Street North St. Petersburg, FL 33710 (800) 237-4323 FAX: (727) 347-2957 (727) 345-9371 http://www.lifesci.com
Suppliers
19 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Life Technologies 9800 Medical Center Drive P.O. Box 6482 Rockville, MD 20849 (800) 828-6686 FAX: (800) 331-2286 http://www.lifetech.com
Lomir Biochemical 99 East Main Street Malone, NY 12953 (877) 425-3604 FAX: (518) 483-8195 (518) 483-7697 http://www.lomir.com
Lifecodes 550 West Avenue Stamford, CT 06902 (800) 543-3263 FAX: (203) 328-9599 (203) 328-9500 http://www.lifecodes.com
LSL Biolafitte 10 rue de Temara 7810C St.-Germain-en-Laye, France (33) 1-3061-5260 FAX: (33) 1-3061-5234
Lightnin 135 Mt. Read Boulevard Rochester, NY 14611 (888) MIX-BEST FAX: (716) 527-1742 (716) 436-5550 http://www.lightnin-mixers.com
Ludl Electronic Products 171 Brady Avenue Hawthorne, NY 10532 (888) 769-6111 FAX: (914) 769-4759 (914) 769-6111 http://www.ludl.com
Linear Drives Luckyn Lane, Pipps Hill Basildon, Essex SS14 3BW, UK (44) 1268-287070 FAX: (44) 1268-293344 http://www.lineardrives.com
Lumigen 24485 W. Ten Mile Road Southfield, MI 48034 (248) 351-5600 FAX: (248) 351-0518 http://www.lumigen.com
Linscott’s Directory 4877 Grange Road Santa Rosa, CA 95404 (707) 544-9555 FAX: (415) 389-6025 http://www.linscottsdirectory.co.uk
Luminex 12212 Technology Boulevard Austin, TX 78727 (888) 219-8020 FAX: (512) 258-4173 (512) 219-8020 http://www.luminexcorp.com
Linton Instrumentation Unit 11, Forge Business Center Upper Rose Lane Palgrave, Diss, Norfolk IP22 1AP, UK (44) 1-379-651-344 FAX: (44) 1-379-650-970 http://www.lintoninst.co.uk
LYNX Therapeutics 25861 Industrial Boulevard Hayward, CA 94545 (510) 670-9300 FAX: (510) 670-9302 http://www.lynxgen.com
List Biological Laboratories 501-B Vandell Way Campbell, CA 95008 (800) 726-3213 FAX: (408) 866-6364 (408) 866-6363 http://www.listlabs.com LKB Instruments See Amersham Pharmacia Biotech Lloyd Laboratories 604 West Thomas Avenue Shenandoah, IA 51601 (800) 831-0004 FAX: (712) 246-5245 (712) 246-4000 http://www.lloydinc.com Loctite 1001 Trout Brook Crossing Rocky Hill, CT 06067 (860) 571-5100 FAX: (860)571-5465 http://www.loctite.com Lofstrand Labs 7961 Cessna Avenue Gaithersburg, MD 20879 (800) 541-0362 FAX: (301) 948-9214 (301) 330-0111 http://www.lofstrand.com
Lyphomed 3 Parkway North Deerfield, IL 60015 (847) 317-8100 FAX: (847) 317-8600 M.E.D. Associates See Med Associates
Mallinckrodt Baker 222 Red School Lane Phillipsburg, NJ 08865 (800) 582-2537 FAX: (908) 859-6974 (908) 859-2151 http://www.mallbaker.com
Mathsoft 1700 Westlake Avenue N., Suite 500 Seattle, WA 98109 (800) 569-0123 FAX: (206) 283-8691 (206) 283-8802 http://www.mathsoft.com
Mallinckrodt Chemicals 16305 Swingley Ridge Drive Chesterfield, MD 63017 (314) 530-2172 FAX: (314) 530-2563 http://www.mallchem.com
Matreya 500 Tressler Street Pleasant Gap, PA 16823 (814) 359-5060 FAX: (814) 359-5062 http://www.matreya.com
Malven Instruments Enigma Business Park Grovewood Road Malven, Worchestershire WR 141 XZ, United Kingdom Marinus 1500 Pier C Street Long Beach, CA 90813 (562) 435-6522 FAX: (562) 495-3120 Markson Science c/o Whatman Labs Sales P.O. Box 1359 Hillsboro, OR 97123 (800) 942-8626 FAX: (503) 640-9716 (503) 648-0762 Marsh Biomedical Products 565 Blossom Road Rochester, NY 14610 (800) 445-2812 FAX: (716) 654-4810 (716) 654-4800 http://www.biomar.com Marshall Farms USA 5800 Lake Bluff Road North Rose, NY 14516 (315) 587-2295 e-mail: [email protected] Martek 6480 Dobbin Road Columbia, MD 21045 (410) 740-0081 FAX: (410) 740-2985 http://www.martekbio.com
Macherey-Nagel 6 South Third Street, #402 Easton, PA 18042 (610) 559-9848 FAX: (610) 559-9878 http://www.macherey-nagel.com
Martin Supply Distributor of Gerber Scientific 2740 Loch Raven Road Baltimore, MD 21218 (800) 282-5440 FAX: (410) 366-0134 (410) 366-1696
Macherey-Nagel Valencienner Strasse 11 P.O. Box 101352 D-52313 Dueren, Germany (49) 2421-969141 FAX: (49) 2421-969199 http://www.macherey-nagel.ch
Mast Immunosystems 630 Clyde Court Mountain View, CA 94043 (800) 233-MAST FAX: (650) 969-2745 (650) 961-5501 http://www.mastallergy.com
Mac-Mod Analytical 127 Commons Court Chadds Ford, PA 19317 800-441-7508 FAX: (610) 358-5993 (610) 358-9696 http://www.mac-mod.com
Matheson Gas Products P.O. Box 624 959 Route 46 East Parsippany, NJ 07054 (800) 416-2505 FAX: (973) 257-9393 (973) 257-1100 http://www.mathesongas.com
Matrigel See Becton Dickinson Labware Matrix Technologies 22 Friars Drive Hudson, NH 03051 (800) 345-0206 FAX: (603) 595-0106 (603) 595-0505 http://www.matrixtechcorp.com MatTek Corp. 200 Homer Ave. Ashland, Massachusetts 01721 (508) 881-6771 FAX: (508) 879-1532 http://www.mattek.com Maxim Medical 89 Oxford Road Oxford OX2 9PD United Kingdom 44 (0)1865-865943 FAX: 44 (0)1865-865291 http://www.maximmed.com Mayo Clinic Section on Engineering Project #ALA-1, 1982 200 1st Street SW Rochester, MN 55905 (507) 284-2511 FAX: (507) 284-5988 McGaw See B. Braun-McGaw McMaster-Carr 600 County Line Road Elmhurst, IL 60126 (630) 833-0300 FAX: (630) 834-9427 http://www.mcmaster.com McNeil Pharmaceutical See Ortho McNeil Pharmaceutical MCNC 3021 Cornwallis Road P.O. Box 12889 Research Triangle Park, NC 27709 (919) 248-1800 FAX: (919) 248-1455 http://www.mcnc.org MD Industries 5 Revere Drive, Suite 415 Northbrook, IL 60062 (800) 421-8370 FAX: (847) 498-2627 (708) 339-6000 http://www.mdindustries.com
Suppliers
20 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
MDS Nordion 447 March Road P.O. Box 13500 Kanata, Ontario K2K 1X8, Canada (800) 465-3666 FAX: (613) 592-6937 (613) 592-2790 http://www.mds.nordion.com
Megazyme Bray Business Park Bray, County Wicklow Ireland (353) 1-286-1220 FAX: (353) 1-286-1264 http://www.megazyme.com
MDS Sciex 71 Four Valley Drive Concord, Ontario Canada L4K 4V8 (905) 660-9005 FAX: (905) 660-2600 http://www.sciex.com
Melles Griot 4601 Nautilus Court South Boulder, CO 80301 (800) 326-4363 FAX: (303) 581-0960 (303) 581-0337 http://www.mellesgriot.com
Mead Johnson See Bristol-Meyers Squibb Med Associates P.O. Box 319 St. Albans, VT 05478 (802) 527-2343 FAX: (802) 527-5095 http://www.med-associates.com Medecell 239 Liverpool Road London N1 1LX, UK (44) 20-7607-2295 FAX: (44) 20-7700-4156 http://www.medicell.co.uk Media Cybernetics 8484 Georgia Avenue, Suite 200 Silver Spring, MD 20910 (301) 495-3305 FAX: (301) 495-5964 http://www.mediacy.com Mediatech 13884 Park Center Road Herndon, VA 20171 (800) cellgro (703) 471-5955 http://www.cellgro.com Medical Systems See Harvard Apparatus Medifor 647 Washington Street Port Townsend, WA 98368 (800) 366-3710 FAX: (360) 385-4402 (360) 385-0722 http://www.medifor.com MedImmune 35 W. Watkins Mill Road Gaithersburg, MD 20878 (301) 417-0770 FAX: (301) 527-4207 http://www.medimmune.com MedProbe AS P.O. Box 2640 St. Hanshaugen N-0131 Oslo, Norway (47) 222 00137 FAX: (47) 222 00189 http://www.medprobe.com
Menzel-Glaser Postfach 3157 D-38021 Braunschweig, Germany (49) 531 590080 FAX: (49) 531 509799 E. Merck Frankfurterstrasse 250 D-64293 Darmstadt 1, Germany (49) 6151-720 Merck See EM Science Merck & Company Merck National Service Center P.O. Box 4 West Point, PA 19486 (800) NSC-MERCK (215) 652-5000 http://www.merck.com Merck Research Laboratories See Merck & Company Merck Sharpe Human Health Division 300 Franklin Square Drive Somerset, NJ 08873 (800) 637-2579 FAX: (732) 805-3960 (732) 805-0300 Merial Limited 115 Transtech Drive Athens, GA 30601 (800) MERIAL-1 FAX: (706) 548-0608 (706) 548-9292 http://www.merial.com Meridian Instruments P.O. Box 1204 Kent, WA 98035 (253) 854-9914 FAX: (253) 854-9902 http://www.minstrument.com Meta Systems Group 32 Hammond Road Belmont, MA 02178 (617) 489-9950 FAX: (617) 489-9952 Metachem Technologies 3547 Voyager Street, Bldg. 102 Torrance, CA 90503 (310) 793-2300 FAX: (310) 793-2304 http://www.metachem.com
Metallhantering Box 47172 100-74 Stockholm, Sweden (46) 8-726-9696
Micro Filtration Systems 7-3-Chome, Honcho Nihonbashi, Tokyo, Japan (81) 3-270-3141
Metersystems GmGH Robert-Bosch Strasse 6 68804 Althusseim, Germany 49-6205-39610 FAX: 49-6205-32270
Micro-Metrics P.O. Box 13804 Atlanta, GA 30324 (770) 986-6015 FAX: (770) 986-9510 http://www.micro-metrics.com
MethylGene 7220 Frederick-Banting, Suite 200 Montreal, Quebec H4S 2A1, Canada http://www.methylgene.com
Micro-Tech Scientific 140 South Wolfe Road Sunnyvale, CA 94086 (408) 730-8324 FAX: (408) 730-3566 http://www.microlc.com
Metro Scientific 475 Main Street, Suite 2A Farmingdale, NY 11735 (800) 788-6247 FAX: (516) 293-8549 (516) 293-9656 Metrowerks 980 Metric Boulevard Austin, TX 78758 (800) 377-5416 (512) 997-4700 http://www.metrowerks.com Mettler Instruments Mettler-Toledo 1900 Polaris Parkway Columbus, OH 43240 (800) METTLER FAX: (614) 438-4900 http://www.mt.com Miami Serpentarium Labs 34879 Washington Loop Road Punta Gorda, FL 33982 (800) 248-5050 FAX: (813) 639-1811 (813) 639-8888 http://www.miamiserpentarium.com Michrom BioResources 1945 Industrial Drive Auburn, CA 95603 (530) 888-6498 FAX: (530) 888-8295 http://www.michrom.com Mickle Laboratory Engineering Gomshall, Surrey, UK (44) 1483-202178 Micra Scientific A division of Eichrom Industries 8205 S. Cass Ave, Suite 111 Darien, IL 60561 (800) 283-4752 FAX: (630) 963-1928 (630) 963-0320 http://www.micrasci.com
Microbix Biosystems 341 Bering Avenue Toronto, Ontario M8Z 3A8 Canada 1-800-794-6694 FAX: 416-234-1626 1-416-234-1624 http://www.microbix.com MicroCal 22 Industrial Drive East Northampton, MA 01060 (800) 633-3115 FAX: (413) 586-0149 (413) 586-7720 http://www.microcalorimetry.com Microfluidics 30 Ossipee Road P.O. Box 9101 Newton, MA 02164 (800) 370-5452 FAX: (617) 965-1213 (617) 969-5452 http://www.microfluidicscorp.com Microgon See Spectrum Laboratories Microlase Optical Systems West of Scotland Science Park Kelvin Campus, Maryhill Road Glasgow G20 0SP, UK (44) 141-948-1000 FAX: (44) 141-946-6311 http://www.microlase.co.uk Micron Instruments 4509 Runway Street Simi Valley, CA 93063 (800) 638-3770 FAX: (805) 522-4982 (805) 552-4676 http://www.microninstruments.com Micron Separations See MSI
MicroBrightField 74 Hegman Avenue Colchester, VT 05446 (802) 655-9360 FAX: (802) 655-5245 http://www.microbrightfield.com
Micro Photonics 4949 Liberty Lane, Suite 170 P.O. Box 3129 Allentown, PA 18106 (610) 366-7103 FAX: (610) 366-7105 http://www.microphotonics.com
Micro Essential Laboratory 4224 Avenue H Brooklyn, NY 11210 (718) 338-3618 FAX: (718) 692-4491
MicroTech 1420 Conchester Highway Boothwyn, PA 19061 (610) 459-3514
Suppliers
21 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Midland Certified Reagent Company 3112-A West Cuthbert Avenue Midland, TX 79701 (800) 247-8766 FAX: (800) 359-5789 (915) 694-7950 FAX: (915) 694-2387 http://www.mcrc.com
Mirus Corporation 506 S. Rosa Road Suite 104 Madison, WI 53719 (608) 441-2852 FAX: (608) 441-2849 http://www.genetransfer.com
Molecular Simulations 9685 Scranton Road San Diego, CA 92121 (800) 756-4674 FAX: (858) 458-0136 (858) 458-9990 http://www.msi.com
MWG-Biotech Anzinger Str. 7 D-85560 Ebersberg, Germany (49) 8092-82890 FAX: (49) 8092-21084 http://www.mwg biotech.com
Midwest Scientific 280 Vance Road Valley Park, MO 63088 (800) 227-9997 FAX: (636) 225-9998 (636) 225-9997 http://www.midsci.com
Misonix 1938 New Highway Farmingdale, NY 11735 (800) 645-9846 FAX: (516) 694-9412 http://www.misonix.com Mitutoyo (MTI) See Dolla Eastern
Monoject Disposable Syringes & Needles/Syrvet 16200 Walnut Street Waukee, IA 50263 (800) 727-5203 FAX: (515) 987-5553 (515) 987-5554 http://www.syrvet.com
Myriad Industries 3454 E Street San Diego, CA 92102 (800) 999-6777 FAX: (619) 232-4819 (619) 232-6700 http://www.myriadindustries.com
MJ Research Waltham, MA 02451 (800) PELTIER FAX: (617) 923-8080 (617) 923-8000 http://www.mjr.com
Monsanto Chemical 800 North Lindbergh Boulevard St. Louis, MO 63167 (314) 694-1000 FAX: (314) 694-7625 http://www.monsanto.com
Miles See Bayer Miles Laboratories See Serological Miles Scientific See Nunc Millar Instruments P.O. Box 230227 6001-A Gulf Freeway Houston, TX 77023 (713) 923-9171 FAX: (713) 923-7757 http://www.millarinstruments.com MilliGen/Biosearch See Millipore Millipore 80 Ashbury Road P.O. Box 9125 Bedford, MA 01730 (800) 645-5476 FAX: (781) 533-3110 (781) 533-6000 http://www.millipore.com Miltenyi Biotec 251 Auburn Ravine Road, Suite 208 Auburn, CA 95603 (800) 367-6227 FAX: (530) 888-8925 (530) 888-8871 http://www.miltenyibiotec.com Miltex 6 Ohio Drive Lake Success, NY 11042 (800) 645-8000 FAX: (516) 775-7185 (516) 349-0001 Milton Roy See Spectronic Instruments Mini-Instruments 15 Burnham Business Park Springfield Road Burnham-on-Crouch, Essex CM0 8TE, UK (44) 1621-783282 FAX: (44) 1621-783132 http://www.mini-instruments.co.uk Mini Mitter P.O. Box 3386 Sunriver, OR 97707 (800) 685-2999 FAX: (541) 593-5604 (541) 593-8639 http://www.minimitter.com
Modular Instruments 228 West Gay Street Westchester, PA 19380 (610) 738-1420 FAX: (610) 738-1421 http://www.mi2.com Molecular Biology Insights 8685 US Highway 24 Cascade, CO 80809-1333 (800) 747-4362 FAX: (719) 684-7989 (719) 684-7988 http://www.oligo.net Molecular Biosystems 10030 Barnes Canyon Road San Diego, CA 92121 (858) 452-0681 FAX: (858) 452-6187 http://www.mobi.com Molecular Devices 1312 Crossman Avenue Sunnyvale, CA 94089 (800) 635-5577 FAX: (408) 747-3602 (408) 747-1700 http://www.moldev.com Molecular Designs 1400 Catalina Street San Leandro, CA 94577 (510) 895-1313 FAX: (510) 614-3608 Molecular Dynamics 928 East Arques Avenue Sunnyvale, CA 94086 (800) 333-5703 FAX: (408) 773-1493 (408) 773-1222 http://www.apbiotech.com Molecular Probes 4849 Pitchford Avenue Eugene, OR 97402 (800) 438-2209 FAX: (800) 438-0228 (541) 465-8300 FAX: (541) 344-6504 http://www.probes.com Molecular Research Center 5645 Montgomery Road Cincinnati, OH 45212 (800) 462-9868 FAX: (513) 841-0080 (513) 841-0900 http://www.mrcgene.com
Moravek Biochemicals 577 Mercury Lane Brea, CA 92821 (800) 447-0100 FAX: (714) 990-1824 (714) 990-2018 http://www.moravek.com Moss P.O. Box 189 Pasadena, MD 21122 (800) 932-6677 FAX: (410) 768-3971 (410) 768-3442 http://www.mosssubstrates.com
Nacalai Tesque Nijo Karasuma, Nakagyo-ku Kyoto 604, Japan 81-75-251-1723 FAX: 81-75-251-1762 http://www.nacalai.co.jp Nalge Nunc International Subsidiary of Sybron International 75 Panorama Creek Drive P.O. Box 20365 Rochester, NY 14602 (800) 625-4327 FAX: (716) 586-8987 (716) 264-9346 http://www.nalgenunc.com Nanogen 10398 Pacific Center Court San Diego, CA 92121 (858) 410-4600 FAX: (858) 410-4848 http://www.nanogen.com
Motion Analysis 3617 Westwind Boulevard Santa Rosa, CA 95403 (707) 579-6500 FAX: (707) 526-0629 http://www.motionanalysis.com
Nanoprobes 95 Horse Block Road Yaphank, NY 11980 (877) 447-6266 FAX: (631) 205-9493 (631) 205-9490 http://www.nanoprobes.com
Mott Farmington Industrial Park 84 Spring Lane Farmington, CT 06032 (860) 747-6333 FAX: (860) 747-6739 http://www.mottcorp.com
Narishige USA 1710 Hempstead Turnpike East Meadow, NY 11554 (800) 445-7914 FAX: (516) 794-0066 (516) 794-8000 http://www.narishige.co.jp
MSI (Micron Separations) See Osmonics
Nasco-Fort Atkinson P.O. Box 901 901 Janesville Ave. Fort Alkinson, WI 53538-0901 (800) 558-9595 FAX: (920) 563-8296 http://www.enasco.com
Multi Channel Systems Markwiesenstrasse 55 72770 Reutlingen, Germany (49) 7121-503010 FAX: (49) 7121-503011 http://www.multichannelsystem.com Multiple Peptide Systems 3550 General Atomics Court San Diego, CA 92121 (800) 338-4965 FAX: (800) 654-5592 (858) 455-3710 FAX: (858) 455-3713 http://www.mps-sd.com Murex Diagnostics 3075 Northwoods Circle Norcross, GA 30071 (707) 662-0660 FAX: (770) 447-4989
National Bag Company 2233 Old Mill Road Hudson, OH 44236 (800) 247-6000 FAX: (330) 425-9800 (330) 425-2600 http://www.nationalbag.com National Band and Tag Department X 35, Box 72430 Newport, KY 41032 (606) 261-2035 FAX: (800) 261-8247 https://www.nationalband.com National Biosciences See Molecular Biology Insights
Suppliers
22 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
National Diagnostics 305 Patton Drive Atlanta, GA 30336 (800) 526-3867 FAX: (404) 699-2077 (404) 699-2121 http://www.nationaldiagnostics.com National Disease Research Exchange 1880 John F. Kennedy Blvd., 11th Fl. Philadelphia, PA 19103 (800) 222-6374 http://www.ndri.com National Institute of Standards and Technology 100 Bureau Drive Gaithersburg, MD 20899 (301) 975-NIST FAX: (301) 926-1630 http://www.nist.gov National Instruments 11500 North Mopac Expressway Austin, TX 78759 (512) 794-0100 FAX: (512) 683-8411 http://www.ni.com
NEN Research Products, Dupont (UK) Diagnostics and Biotechnology Systems Wedgewood Way Stevenage, Hertfordshire SG1 4QN, UK 44-1438-734831 44-1438-734000 FAX: 44-1438-734836 http://www.dupont.com Neogen 628 Winchester Road Lexington, KY 40505 (800) 477-8201 FAX: (606) 255-5532 (606) 254-1221 http://www.neogen.com Neosystems 380, 11012 Macleod Trail South Calgary, Alberta T2J 6A5 Canada (403) 225-9022 FAX: (403) 225-9025 http://www.neosystems.com
National Labnet See Labnet International
Neuralynx 2434 North Pantano Road Tucson, AZ, 85715 (520) 722-6144 FAX: (520) 722-8163 http://www.neuraiynx.com
National Scientific Instruments 975 Progress Circle Lawrenceville, GA 300243 (800) 332-3331 FAX: (404) 339-7173 http://www.nationalscientific.com
Neuro Probe 16008 Industrial Drive Gaithersburg, MD 20877 (301) 417-0014 FAX: (301) 977-5711 http://www.neuroprobe.com
National Scientific Supply 1111 Francisco Bouldvard East San Rafael, CA 94901 (800) 525-1779 FAX: (415) 459-2954 (415) 459-6070 http://www.nat-sci.com Naz-Dar-KC Chicago Nazdar 1087 N. North Branch Street Chicago, IL 60622 (800) 736-7636 FAX: (312) 943-8215 (312) 943-8338 http://www.nazdar.com NB Labs 1918 Avenue A Denison, TX 75021 (903) 465-2694 FAX: (903) 463-5905 http://www.nblabslarry.com NEB See New England Biolabs NEN Life Science Products 549 Albany Street Boston, MA 02118 (800) 551-2121 FAX: (617) 451-8185 (617) 350-9075 http://www.nen.com
Neurocrine Biosciences 10555 Science Center Drive San Diego, CA 92121 (619) 658-7600 FAX: (619) 658-7602 http://www.neurocrine.com Nevtek HCR03, Box 99 Burnsville, VA 24487 (540) 925-2322 FAX: (540) 925-2323 http://www.nevtek.com New Brunswick Scientific 44 Talmadge Road Edison, NJ 08818 (800) 631-5417 FAX: (732) 287-4222 (732) 287-1200 http://www.nbsc.com New England Biolabs (NEB) 32 Tozer Road Beverly, MA 01915 (800) 632-5227 FAX: (800) 632-7440 http://www.neb.com
Newark Electronics 4801 N. Ravenswood Avenue Chicago, IL 60640 (800) 4-NEWARK FAX: (773) 907-5339 (773) 784-5100 http://www.newark.com Newell Rubbermaid 29 E. Stephenson Street Freeport, IL 61032 (815) 235-4171 FAX: (815) 233-8060 http://www.newellco.com Newport Biosystems 1860 Trainor Street Red Bluff, CA 96080 (530) 529-2448 FAX: (530) 529-2648 Newport 1791 Deere Avenue Irvine, CA 92606 (800) 222-6440 FAX: (949) 253-1800 (949) 253-1462 http://www.newport.com Nexin Research B.V. P.O. Box 16 4740 AA Hoeven, The Netherlands (31) 165-503172 FAX: (31) 165-502291 NIAID See Bio-Tech Research Laboratories Nichiryo 230 Route 206 Building 2-2C Flanders, NJ 07836 (877) 458-6667 FAX: (973) 927-0099 (973) 927-4001 http://www.nichiryo.com Nichols Institute Diagnostics 33051 Calle Aviador San Juan Capistrano, CA 92675 (800) 286-4NID FAX: (949) 240-5273 (949) 728-4610 http://www.nicholsdiag.com Nichols Scientific Instruments 3334 Brown Station Road Columbia, MO 65202 (573) 474-5522 FAX: (603) 215-7274 http://home.beseen.com technology/nsi technology
New England Nuclear (NEN) See NEN Life Science Products
Nicolet Biomedical Instruments 5225 Verona Road, Building 2 Madison, WI 53711 (800) 356-0007 FAX: (608) 441-2002 (608) 273-5000 http://nicoletbiomedical.com
New MBR Gubelstrasse 48 CH8050 Zurich, Switzerland (41) 1-313-0703
N.I.G.M.S. (National Institute of General Medical Sciences) See Coriell Institute for Medical Research
Nikon Science and Technologies Group 1300 Walt Whitman Road Melville, NY 11747 (516) 547-8500 FAX: (516) 547-4045 http://www.nikonusa.com Nippon Gene 1-29, Ton-ya-machi Toyama 930, Japan (81) 764-51-6548 FAX: (81) 764-51-6547 Noldus Information Technology 751 Miller Drive Suite E-5 Leesburg, VA 20175 (800) 355-9541 FAX: (703) 771-0441 (703) 771-0440 http://www.noldus.com Nonllnear Dynamics See MDS NovoDynamics Nordion International See MDS Nordion North American Biologicals (NABI) 16500 NW 15th Avenue Miami, FL 33169 (800) 327-7106 (305) 625-5305 http://www.nabi.com North American Reiss See Reiss Northwestern Bottle 24 Walpole Park South Walpole, MA 02081 (508) 668-8600 FAX: (508) 668-7790 NOVA Biomedical Nova Biomedical 200 Prospect Street Waltham, MA 02454 (800) 822-0911 FAX: (781) 894-5915 http://www.novabiomedical.com Novagen 601 Science Drive Madison, WI 53711 (800) 526-7319 FAX: (608) 238-1388 (608) 238-6110 http://www.novagen.com Novartis 59 Route 10 East Hanover, NJ 07936 (800)526-0175 FAX: (973) 781-6356 http://www.novartis.com Novartis Biotechnology 3054 Cornwallis Road Research Triangle Park, NC 27709 (888) 462-7288 FAX: (919) 541-8585 http://www.novartis.com Nova Sina AG Subsidiary of Airflow Lufttechnik GmbH Kleine Heeg 21 52259 Rheinbach, Germany (49) 02226 920-0 FAX: (49) 02226 9205-11
Suppliers
23 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Novex/Invitrogen 1600 Faraday Carlsbad, CA 92008 (800) 955-6288 FAX: (760) 603-7201 http://www.novex.com Novo Nordisk Biochem 77 Perry Chapel Church Road Franklington, NC 27525 (800) 879-6686 FAX: (919) 494-3450 (919) 494-3000 http://www.novo.dk Novo Nordisk BioLabs See Novo Nordisk Biochem Novocastra Labs Balliol Business Park West Benton Lane Newcastle-upon-Tyne Tyne and Wear NE12 8EW, UK (44) 191-215-0567 FAX: (44) 191-215-1152 http://www.novocastra.co.uk
NYCOMED AS Pharma c/o Accurate Chemical & Scientific 300 Shames Drive Westbury, NY 11590 (800) 645-6524 FAX: (516) 997-4948 (516) 333-2221 http://www.accuratechemical.com Nycomed Amersham Health Care Division 101 Carnegie Center Princeton, NJ 08540 (800) 832-4633 FAX: (800) 807-2382 (609) 514-6000 http://www.nycomed-amersham.com Nyegaard Herserudsvagen 5254 S-122 06 Lidingo, Sweden (46) 8-765-2930 Ohmeda Catheter Products See Datex-Ohmeda
NovoDynamics 123 North Ashley Street Sulte 210 Ann Arbor, MI 48104 (734) 205-9100 FAX: (734) 205-9101 http://www.novodynamics.com
Ohwa Tsusbo Hiby Dai Building 1-2-2 Uchi Saiwai-cho Chiyoda-ku Tokyo 100, Japan 03-3591-7348 FAX: 03-3501-9001
Novus Biologicals P.O. Box 802 Littleton, CO 80160 (888) 506-6887 FAX: (303) 730-1966 http://www.novus-biologicals.com/ main.html
Oligos Etc. 9775 S.W. Commerce Circle, C-6 Wilsonville, OR 97070 (800) 888-2358 FAX: (503) 6822D1635 (503) 6822D1814 http://www.oligoetc.com
NPI Electronic Hauptstrasse 96 D-71732 Tamm, Germany (49) 7141-601534 FAX: (49) 7141-601266 http://www.npielectronic.com NSG Precision Cells 195G Central Avenue Farmingdale, NY 11735 (516) 249-7474 FAX: (516) 249-8575 http://www.nsgpci.com Nu Chek Prep 109 West Main P.O. Box 295 Elysian, MN 56028 (800) 521-7728 FAX: (507) 267-4790 (507) 267-4689
Olis Instruments 130 Conway Drive Bogart, GA 30622 (706) 353-6547 (800) 852-3504 http://www.olisweb.com Olympus America 2 Corporate Center Drive Melville, NY 11747 (800) 645-8160 FAX: (516) 844-5959 (516) 844-5000 http://www.olympusamerica.com
Nuclepore See Costar
Omega Engineering One Omega Drive P.O. Box 4047 Stamford, CT 06907 (800) 848-4286 FAX: (203) 359-7700 (203) 359-1660 http://www.omega.com
Numonics 101 Commerce Drive Montgomeryville, PA 18936 (800) 523-6716 FAX: (215) 361-0167 (215) 362-2766 http://www.interactivewhiteboards.com
Omega Optical 3 Grove Street P.O. Box 573 Brattleboro, VT 05302 (802) 254-2690 FAX: (802) 254-3937 http://www.omegafilters.com
Omnetics Connector Corporation 7280 commerce Circle East Minneapolls, MN 55432 (800) 343-0026 (763) 572-0656 Fax: (783) 752-3925 http://www.omnetics.com/rhain.htm
Orbigen 6827 Nancy Ridge Drive San Diego, CA 92121 (866) 672-4436 (858) 362-2030 (858) 362-2026 http://www.orbigen.com
Omni International 6530 Commerce Court Warrenton, VA 20187 (800) 776-4431 FAX: (540) 347-5352 (540) 347-5331 http://www.omni-inc.com
Oread BioSaftey 1501 Wakarusa Drive Lawrence, KS 66047 (800) 447-6501 FAX: (785) 749-1882 (785) 749-0034 http://www.oread.com
Omnion 2010 Energy Drive P.O. Box 879 East Troy, WI 53120 (262) 642-7200 FAX: (262) 642-7760 http://www.omnion.com
Organomation Associates 266 River Road West Berlin, MA 01503 (888) 978-7300 FAX: (978)838-2786 (978) 838-7300 http://www.organomation.com
Omnitech Electronics See AccuScan Instruments
Organon 375 Mount Pleasant Avenue West Orange, NJ 07052 (800) 241-8812 FAX: (973) 325-4589 (973) 325-4500 http://www.organon.com
Oncogene Research Products P.O. Box Box 12087 La Jolla, CA 92039-2087 (800) 662-2616 FAX: (800) 766-0999 http://www.apoptosis.com Oncogene Science See OSI Pharmaceuticals Oncor See Intergen
Organon Teknika (Canada) 30 North Wind Place Scarborough, Ontario M1S 3R5 Canada (416) 754-4344 FAX: (416) 754-4488 http://www.organonteknika.com
Online Instruments 130 Conway Drive, Suites A & B Bogart, GA 30622 (800) 852-3504 (706) 353-1972 (708) 353-6547 http://www.olisweb.com
Organon Teknika Cappel 100 Akzo Avenue Durham, NC 27712 (800) 682-2666 FAX: (800) 432-9682 (919) 620-2000 FAX: (919) 620-2107 http://www.organonteknika.com
Operon Technologies 1000 Atlantic Avenue Alameda, CA 94501 (800) 688-2248 FAX: (510) 865-5225 (510) 865-8644 http://www.operon.com
Oriel Corporation of America 150 Long Beach Boulevard Stratford, CT 06615 (203) 377-8282 FAX: (203) 378-2457 http://www.oriel.com
Optiscan P.O. Box 1066 Mount Waverly MDC, Victoria Australia 3149 61-3-9538 3333 FAX: 61-3-9562 7742 http://www.optiscan.com.au Optomax 9 Ash Street P.O. Box 840 Hollis, NH 03049 (603) 465-3385 FAX: (603) 465-2291 Opto-Line Associates 265 Ballardvale Street Wilmington, MA 01887 (978) 658-7255 FAX: (978) 658-7299 http://www.optoline.com
OriGene Technologies 6 Taft Court, Suite 300 Rockville, MD 20850 (888) 267-4436 FAX: (301) 340-9254 (301) 340-3188 http://www.origene.com OriginLab One Roundhouse Plaza Northhampton, MA 01060 (800) 969-7720 FAX: (413) 585-0126 http://www.originlab.com Orion Research 500 Cummings Center Beverly, MA 01915 (800) 225-1480 FAX: (978) 232-6015 (978) 232-6000 http://www.orionres.com
Suppliers
24 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Ortho Diagnostic Systems Subsidiary of Johnson & Johnson 1001 U.S. Highway 202 P.O. Box 350 Raritan, NJ 08869 (800) 322-6374 FAX: (908) 218-8582 (908) 218-1300
Oxford Instruments Old Station Way Eynsham Witney, Oxfordshire OX8 1TL, UK (44) 1865-881437 FAX: (44) 1865-881944 http://www.oxinst.com
Ortho McNeil Pharmaceutical Welsh & McKean Road Spring House, PA 19477 (800) 682-6532 (215) 628-5000 http://www.orthomcneil.com
Oxford Labware See Kendall
Oryza 200 Turnpike Road, Unit 5 Chelmsford, MA 01824 (978) 256-8183 FAX: (978) 256-7434 http://www.oryzalabs.com OSI Pharmaceuticals 106 Charles Lindbergh Boulevard Uniondale, NY 11553 (800) 662-2616 FAX: (516) 222-0114 (516) 222-0023 http://www.osip.com Osmonics 135 Flanders Road P.O. Box 1046 Westborough, MA 01581 (800) 444-8212 FAX: (508) 366-5840 (508) 366-8212 http://www.osmolabstore.com
Oxford Molecular Group Oxford Science Park The Medawar Centre Oxford OX4 4GA, UK (44) 1865-784600 FAX: (44) 1865-784601 http://www.oxmol.co.uk Oxford Molecular Group 2105 South Bascom Avenue, Suite 200 Campbell, CA 95008 (800) 876-9994 FAX: (408) 879-6302 (408) 879-6300 http://www.oxmol.com OXIS International 6040 North Cutter Circle Suite 317 Portland, OR 97217 (800) 547-3686 FAX: (503) 283-4058 (503) 283-3911 http://www.oxis.com
Oster Professional Products 150 Cadillac Lane McMinnville, TN 37110 (931) 668-4121 FAX: (931) 668-4125 http://www.sunbeam.com
Oxoid 800 Proctor Avenue Ogdensburg, NY 13669 (800) 567-8378 FAX: (613) 226-3728 http://www.oxoid.ca
Out Patient Services 1260 Holm Road Petaluma, CA 94954 (800) 648-1666 FAX: (707) 762-7198 (707) 763-1581
Oxoid Wade Road Basingstoke, Hampshire RG24 8PW, UK (44) 1256-841144 FAX: (4) 1256-814626 http://www.oxoid.ca
Owans and Minor 4800 Cox Road Glen Allen, VA 23060 (804) 747-9794 Fax: 804-270-7281 http://www.owens-minor.com OWL Scientific Plastics See OWL Separation Systems OWL Separation Systems 55 Heritage Avenue Portsmouth, NH 03801 (800) 242-5560 FAX: (603) 559-9258 (603) 559-9297 http://www.owlsci.com Oxford Biochemical Research P.O. Box 522 Oxford, MI 48371 (800) 692-4633 FAX: (248) 852-4466 http://www.oxfordbiomed.com Oxford GlycoSystems See Glyco
Oxyrase P.O. Box 1345 Mansfield, OH 44901 (419) 589-8800 FAX: (419) 589-9919 http://www.oxyrase.com Ozyme ` 10 Avenue Ampere Montigny de Bretoneux 78180 France (33) 13-46-02-424 FAX: (33) 13-46-09-212 http://www.ozyme.fr PAA Laboratories 2570 Route 724 P.O. Box 435 Parker Ford, PA 19457 (610) 495-9400 FAX: (610) 495-9410 http://www.paa-labs.com
Pacer Scientific 5649 Valley Oak Drive Los Angeles, CA 90068 (323) 462-0636 FAX: (323) 462-1430 http://www.pacersci.com Pacific Bio-Marine Labs P.O. Box 1348 Venice, CA 90294 (310) 677-1056 FAX: (310) 677-1207 Packard Instrument 800 Research Parkway Meriden, CT 06450 (800) 323-1891 FAX: (203) 639-2172 (203) 238-2351 http://www.packardinst.com
Pel-Freez Biologicals 219 N. Arkansas P.O. Box 68 Rogers, AR 72757 (800) 643-3426 FAX: (501) 636-3562 (501) 636-4361 http://www.pelfreez-bio.com Pel-Freez Clinical Systems Subsidiary of Pel-Freez Biologicals 9099 N. Deerbrook Trail Brown Deer, WI 53223 (800) 558-4511 FAX: (414) 357-4518 (414) 357-4500 http://www.pelfreez-bio.com
Padgett Instrument 1730 Walnut Street Kansas City, MO 64108 (816) 842-1029
Peninsula Laboratories 601 Taylor Way San Carlos, CA 94070 (800) 650-4442 FAX: (650) 595-4071 (650) 592-5392 http://www.penlabs.com
Pall Filtron 50 Bearfoot Road Northborough, MA 01532 (800) FILTRON FAX: (508) 393-1874 (508) 393-1800
Pentex 24562 Mando Drive Laguna Niguel, CA 92677 (800) 382-4667 FAX: (714) 643-2363 http://www.pentex.com
Pall-Gelman 25 Harbor Park Drive Port Washington, NY 11050 (800) 289-6255 FAX: (516) 484-2651 (516) 484-3600 http://www.pall.com PanVera 545 Science Drive Madison, WI 53711 (800) 791-1400 FAX: (608) 233-3007 (608) 233-9450 http://www.panvera.com Parke-Davis See Warner-Lambert Parr Instrument 211 53rd Street Moline, IL 61265 (800) 872-7720 FAX: (309) 762-9453 (309) 762-7716 http://www.parrinst.com Partec Otto Hahn Strasse 32 D-48161 Munster, Germany (49) 2534-8008-0 FAX: (49) 2535-8008-90 PCR See Archimica Florida PE Biosystems 850 Lincoln Centre Drive Foster City, CA 94404 (800) 345-5224 FAX: (650) 638-5884 (650) 638-5800 http://www.pebio.com
PeproTech 5 Crescent Avenue P.O. Box 275 Rocky Hill, NJ 08553 (800) 436-9910 FAX: (609) 497-0321 (609) 497-0253 http://www.peprotech.com Peptide Institute 4-1-2 Ina, Minoh-shi Osaka 562-8686, Japan 81-727-29-4121 FAX: 81-727-29-4124 http://www.peptide.co.jp Peptide Laboratory 4175 Lakeside Drive Richmond, CA 94806 (800) 858-7322 FAX: (510) 262-9127 (510) 262-0800 http://www.peptidelab.com Peptides International 11621 Electron Drive Louisville, KY 40299 (800) 777-4779 FAX: (502) 267-1329 (502) 266-8787 http://www.pepnet.com Perceptive Science Instruments 2525 South Shore Blvd., Suite 100 League City, TX 77573 (281) 334-3027 FAX: (281) 538-2222 http://www.persci.com Perimed 4873 Princeton Drive North Royalton, OH 44133 (440) 877-0537 FAX: (440) 877-0534 http://www.perimed.se
Suppliers
25 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Perkin-Elmer 761 Main Avenue Norwalk, CT 06859 (800) 762-4002 FAX: (203) 762-6000 (203) 762-1000 http://www.perkin-elmer.com See also PE Biosystems PerSeptive Bioresearch Products See PerSeptive BioSystems PerSeptive BioSystems 500 Old Connecticut Path Framingham, MA 01701 (800) 899-5858 FAX: (508) 383-7885 (508) 383-7700 http://www.pbio.com PerSeptive Diagnostic See PE Biosystems (800) 343-1346 Pettersson Elektronik AB Tallbacksvagen 51 S-756 45 Uppsala, Sweden (46) 1830-3880 FAX: (46) 1830-3840 http://www.bahnhof.se/∼pettersson Pfanstiehl Laboratories, Inc. 1219 Glen Rock Avenue Waukegan, IL 60085 (800) 383-0126 FAX: (847) 623-9173 http://www.pfanstiehl.com PGC Scientifics 7311 Governors Way Frederick, MD 21704 (800) 424-3300 FAX: (800) 662-1112 (301) 620-7777 FAX: (301) 620-7497 http://www.pgcscientifics.com Pharmacia Biotech See Amersham Pharmacia Biotech Pharmacia Diagnostics See Wallac Pharmacia LKB Biotech See Amersham Pharmacia Biotech Pharmacia LKB Biotechnology See Amersham Pharmacia Biotech Pharmacia LKB Nuclear See Wallac Pharmaderm Veterinary Products 60 Baylis Road Melville, NY 11747 (800) 432-6673 http://www.pharmaderm.com Pharmed (Norton) Norton Performance Plastics See Saint-Gobain Performance Plastics
PHLS Centre for Applied Microbiology and Research See European Collection of Animal Cell Cultures (ECACC) Phoenix Flow Systems 11575 Sorrento Valley Road, Suite 208 San Diego, CA 92121 (800) 886-3569 FAX: (619) 259-5268 (619) 453-5095 http://www.phnxflow.com Phoenix Pharmaceutical 4261 Easton Road, P.O. Box 6457 St. Joseph, MO 64506 (800) 759-3644 FAX: (816) 364-4969 (816) 364-5777 http://www.phoenixpharmaceutical.com Photometrics See Roper Scientific Photon Technology International 1 Deerpark Drive, Suite F Monmouth Junction, NJ 08852 (732) 329-0910 FAX: (732) 329-9069 http://www.pti-nj.com Physik Instrumente Polytec PI 23 Midstate Drive, Suite 212 Auburn, MA 01501 (508) 832-3456 FAX: (508) 832-0506 http://www.polytecpi.com Physitemp Instruments 154 Huron Avenue Clifton, NJ 07013 (800) 452-8510 FAX: (973) 779-5954 (973) 779-5577 http://www.physitemp.com Pico Technology The Mill House, Cambridge Street St. Neots, Cambridgeshire PE19 1QB, UK (44) 1480-396-395 FAX: (44) 1480-396-296 www.picotech.com Pierce Chemical P.O. Box 117 3747 Meridian Road Rockford, IL 61105 (800) 874-3723 FAX: (800) 842-5007 FAX: (815) 968-7316 http://www.piercenet.com
PharMingen See BD PharMingen
Pierce & Warriner 44, Upper Northgate Street Chester, Cheshire CH1 4EF, UK (44) 1244 382 525 FAX: (44) 1244 373 212 http://www.piercenet.com
Phenomex 2320 W. 205th Street Torrance, CA 90501 (310) 212-0555 FAX: (310) 328-7768 http://www.phenomex.com
Pilling Weck Surgical 420 Delaware Drive Fort Washington, PA 19034 (800) 523-2579 FAX: (800) 332-2308 www.pilling-weck.com
PixelVision A division of Cybex Computer Products 14964 NW Greenbrier Parkway Beaverton, OR 97006 (503) 629-3210 FAX: (503) 629-3211 http://www.pixelvision.com
Polymer Laboratories Amherst Research Park 160 Old Farm Road Amherst, MA 01002 (800) 767-3963 FAX: (413) 253-2476 http://www.polymerlabs.com
P.J. Noyes P.O. Box 381 89 Bridge Street Lancaster, NH 03584 (800) 522-2469 FAX: (603) 788-3873 (603) 788-4952 http://www.pjnoyes.com
Polymicro Technologies 18019 North 25th Avenue Phoenix, AZ 85023 (602) 375-4100 FAX: (602) 375-4110 http://www.polymicro.com
Plas-Labs 917 E. Chilson Street Lansing, MI 48906 (800) 866-7527 FAX: (517) 372-2857 (517) 372-7177 http://www.plas-labs.com Plastics One 6591 Merriman Road, Southwest P.O. Box 12004 Roanoke, VA 24018 (540) 772-7950 FAX: (540) 989-7519 http://www.plastics1.com Platt Electric Supply 2757 6th Avenue South Seattle, WA 98134 (206) 624-4083 FAX: (206) 343-6342 http://www.platt.com Plexon 6500 Greenville Avenue Suite 730 Dallas, TX 75206 (214) 369-4957 FAX: (214) 369-1775 http://www.plexoninc.com
Polyphenols AS Hanabryggene Technology Centre Hanaveien 4-6 4327 Sandnes, Norway (47) 51-62-0990 FAX: (47) 51-62-51-82 http://www.polyphenols.com Polysciences 400 Valley Road Warrington, PA 18976 (800) 523-2575 FAX: (800) 343-3291 http://www.polysciences.com Polyscientific 70 Cleveland Avenue Bayshore, NY 11706 (516) 586-0400 FAX: (516) 254-0618 Polytech Products 285 Washington Street Somerville, MA 02143 (617) 666-5064 FAX: (617) 625-0975 Polytron 8585 Grovemont Circle Gaithersburg, MD 20877 (301) 208-6597 FAX: (301) 208-8691 http://www.polytron.com
Polaroid 784 Memorial Drive Cambridge, MA 01239 (800) 225-1618 FAX: (800) 832-9003 (781) 386-2000 http://www.polaroid.com
Popper and Sons 300 Denton Avenue P.O. Box 128 New Hyde Park, NY 11040 (888) 717-7677 FAX: (800) 557-6773 (516) 248-0300 FAX: (516) 747-1188 http://www.popperandsons.com
Polyfiltronics 136 Weymouth St. Rockland, MA 02370 (800) 434-7659 FAX: (781) 878-0822 (781) 878-1133 http://www.polyfiltronics.com
Porphyrin Products P.O. Box 31 Logan, UT 84323 (435) 753-1901 FAX: (435) 753-6731 http://www.porphyrin.com
Polylabo Paul Block Parc Tertiare de la Meinau 10, rue de la Durance B.P. 36 67023 Strasbourg Cedex 1 Strasbourg, France 33-3-8865-8020 FAX: 33-3-8865-8039 PolyLC 9151 Rumsey Road, Suite 180 Columbia, MD 21045 (410) 992-5400 FAX: (410) 730-8340
Portex See SIMS Portex Limited Powderject Vaccines 585 Science Drive Madison, WI 53711 (608) 231-3150 FAX: (608) 231-6990 http://www.powderject.com Praxair 810 Jorie Boulevard Oak Brook, IL 60521 (800) 621-7100 http://www.praxair.com
Suppliers
26 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Precision Dynamics 13880 Del Sur Street San Fernando, CA 91340 (800) 847-0670 FAX: (818) 899-4-45 http://www.pdcorp.com Precision Scientific Laboratory Equipment Division of Jouan 170 Marcel Drive Winchester, VA 22602 (800) 621-8820 FAX: (540) 869-0130 (540) 869-9892 http://www.precisionsci.com Primary Care Diagnostics See Becton Dickinson Primary Care Diagnostics Primate Products 1755 East Bayshore Road, Suite 28A Redwood City, CA 94063 (650) 368-0663 FAX: (650) 368-0665 http://www.primateproducts.com 5 Prime → 3 Prime See 2000 Eppendorf-5 Prime http://www.5prime.com Princeton Applied Research PerkinElmer Instr.: Electrochemistry 801 S. Illinois Oak Ridge, TN 37830 (800) 366-2741 FAX: (423) 425-1334 (423) 481-2442 http://www.eggpar.com Princeton Instruments A division of Roper Scientific 3660 Quakerbridge Road Trenton, NJ 08619 (609) 587-9797 FAX: (609) 587-1970 http://www.prinst.com Princeton Separations P.O. Box 300 Aldephia, NJ 07710 (800) 223-0902 FAX: (732) 431-3768 (732) 431-3338 Prior Scientific 80 Reservoir Park Drive Rockland, MA 02370 (781) 878-8442 FAX: (781) 878-8736 http://www.prior.com PRO Scientific P.O. Box 448 Monroe, CT 06468 (203) 452-9431 FAX: (203) 452-9753 http://www.proscientific.com Professional Compounding Centers of America 9901 South Wilcrest Drive Houston, TX 77099 (800) 331-2498 FAX: (281) 933-6227 (281) 933-6948 http://www.pccarx.com
Progen Biotechnik Maass-Str. 30 69123 Heidelberg, Germany (49) 6221-8278-0 FAX: (49) 6221-8278-23 http://www.progen.de
Purina Mills LabDiet P. O. Box 66812 St. Louis, MO 63166 (800) 227-8941 FAX: (314) 768-4894 http://www.purina-mills.com
Prolabo A division of Merck Eurolab 54 rue Roger Salengro 94126 Fontenay Sous Bois Cedex France 33-1-4514-8500 FAX: 33-1-4514-8616 http://www.prolabo.fr
Qbiogene 2251 Rutherford Road Carlsbad, CA 92008 (800) 424-6101 FAX: (760) 918-9313 http://www.qbiogene.com
Proligo 2995 Wilderness Place Boulder, CO 80301 (888) 80-OLIGO FAX: (303) 801-1134 http://www.proligo.com Promega 2800 Woods Hollow Road Madison, WI 53711 (800) 356-9526 FAX: (800) 356-1970 (608) 274-4330 FAX: (608) 277-2516 http://www.promega.com Protein Databases (PDI) 405 Oakwood Road Huntington Station, NY 11746 (800) 777-6834 FAX: (516) 673-4502 (516) 673-3939 Protein Polymer Technologies 10655 Sorrento Valley Road San Diego, CA 92121 (619) 558-6064 FAX: (619) 558-6477 http://www.ppti.com Protein Solutions 391 G Chipeta Way Salt Lake City, UT 84108 (801) 583-9301 FAX: (801) 583-4463 http://www.proteinsolutions.com Prozyme 1933 Davis Street, Suite 207 San Leandro, CA 94577 (800) 457-9444 FAX: (510) 638-6919 (510) 638-6900 http://www.prozyme.com PSI See Perceptive Science Instruments
Qiagen 28159 Avenue Stanford Valencia, CA 91355 (800) 426-8157 FAX: (800) 718-2056 http://www.qiagen.com Quality Biological 7581 Lindbergh Drive Gaithersburg, MD 20879 (800) 443-9331 FAX: (301) 840-5450 (301) 840-9331 http://www.qualitybiological.com Quantitative Technologies P.O. Box 470 Salem Industrial Park, Bldg, 5 Whltehouse, NJ 08888 (908) 534-4445 FAX: 534-1054 http://www.qtionline.com Quantum Appligene Parc d’Innovation Rue Geller de Kayserberg 67402 Illkirch, Cedex, France (33) 3-8867-5425 FAX: (33) 3-8867-1945 http://www.quantum-appligene.com Quantum Biotechnologies See Qbiogene Quantum Soft Postfach 6613 CH-8023 ¨ Zurich, Switzerland FAX: 41-1-481-69-51 [email protected] Quest Scientific 1177 West 21st Street North Vancouver, BC V7R 2C7 Canada (804) 984-7878 http://www.quest-sci.com
Pulmetrics Group 82 Beacon Street Chestnut Hill, MA 02167 (617) 353-3833 FAX: (617) 353-6766
Questcor Pharmaceuticals 26118 Research Road Hayward, CA 94545 (510) 732-5551 FAX: (510) 732-7741 http://www.questcor.com
Purdue Frederick 100 Connecticut Avenue Norwalk, CT 06850 (800) 633-4741 FAX: (203) 838-1576 (203) 853-0123 http://www.pharma.com
Quidel 10165 McKellar Court San Diego, CA 92121 (800) 874-1517 FAX: (858) 546-8955 (858) 552-1100 http://www.quidel.com
R-Biopharm 7950 Old US 27 South Marshall, MI 49068 (616) 789-3033 FAX: (616) 789-3070 http://www.r-biopharm.com R. C. Electronics 6464 Hollister Avenue Santa Barbara, CA 93117 (805) 685-7770 FAX: (805) 685-5853 http://www.rcelectronics.com R & D Systems 614 McKinley Place NE Minneapolis, MN 55413 (800) 343-7475 FAX: (612) 379-6580 (612) 379-2956 http://www.rndsystems.com R & S Technology 350 Columbia Street Peacedale, RI 02880 (401) 789-5660 FAX: (401) 792-3890 http://www.septech.com RACAL Health and Safety See 3M 7305 Executive Way Frederick, MD 21704 (800) 692-9500 FAX: (301) 695-8200 Radiometer America 811 Sharon Drive Westlake, OH 44145 (800) 736-0600 FAX: (440) 871-2633 (440) 871-8900 http://www.rameusa.com Radiometer A/S The Chemical Reference Laboratory kandevej 21 DK-2700 Brnshj, Denmark 45-3827-3827 FAX: 45-3827-2727 Radionics 22 Terry Avenue Burlington, MA 01803 (781) 272-1233 FAX: (781) 272-2428 http://www.radionics.com Radnoti Glass Technology 227 W. Maple Avenue Monrovia, CA 91016 (800) 428-l4l6 FAX: (626) 303-2998 (626) 357-8827 http://www.radnoti.com Rainin Instrument Rainin Road P.O. Box 4026 Woburn, MA 01888 (800)-4-RAININ FAX: (781) 938-1152 (781) 935-3050 http://www.rainin.com Rank Brothers 56 High Street Bottisham, Cambridge CB5 9DA UK (44) 1223 811369 FAX: (44) 1223 811441 http://www.rankbrothers.com
Suppliers
27 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Rapp Polymere Ernst-Simon Strasse 9 ¨ D 72072 Tubingen, Germany (49) 7071-763157 FAX: (49) 7071-763158 http://www.rapp-polymere.com Raven Biological Laboratories 8607 Park Drive P.O. Box 27261 Omaha, NE 68127 (800) 728-5702 FAX: (402) 593-0995 (402) 593-0781 http://www.ravenlabs.com Razel Scientific Instruments 100 Research Drive Stamford, CT 06906 (203) 324-9914 FAX: (203) 324-5568 RBI See Research Biochemicals Reagents International See Biotech Source Receptor Biology 10000 Virginia Manor Road, Suite 360 Beltsville, MD 20705 (888) 707-4200 FAX: (301) 210-6266 (301) 210-4700 http://www.receptorbiology.com Regis Technologies 8210 N. Austin Avenue Morton Grove, IL 60053 (800) 323-8144 FAX: (847) 967-1214 (847) 967-6000 http://www.registech.com Reichert Ophthalmic Instruments P.O. Box 123 Buffalo, NY 14240 (716) 686-4500 FAX: (716) 686-4545 http://www.reichert.com Reiss 1 Polymer Place P.O. Box 60 Blackstone, VA 23824 (800) 356-2829 FAX: (804) 292-1757 (804) 292-1600 http://www.reissmfg.com Remel 12076 Santa Fe Trail Drive P.O. Box 14428 Shawnee Mission, KS 66215 (800) 255-6730 FAX: (800) 621-8251 (913) 888-0939 FAX: (913) 888-5884 http://www.remelinc.com Reming Bioinstruments 6680 County Route 17 Redfield, NY 13437 (315) 387-3414 FAX: (315) 387-3415 RepliGen 117 Fourth Avenue Needham, MA 02494 (800) 622-2259 FAX: (781) 453-0048 (781) 449-9560 http://www.repligen.com
Research Biochemicals 1 Strathmore Road Natick, MA 01760 (800) 736-3690 FAX: (800) 736-2480 (508) 651-8151 FAX: (508) 655-1359 http://www.resbio.com Research Corporation Technologies 101 N. Wilmot Road, Suite 600 Tucson, AZ 85711 (520) 748-4400 FAX: (520) 748-0025 http://www.rctech.com Research Diagnostics Pleasant Hill Road Flanders, NJ 07836 (800) 631-9384 FAX: (973) 584-0210 (973) 584-7093 http://www.researchd.com Research Diets 121 Jersey Avenue New Brunswick, NJ 08901 (877) 486-2486 FAX: (732) 247-2340 (732) 247-2390 http://www.researchdiets.com Research Genetics 2130 South Memorial Parkway Huntsville, AL 35801 (800) 533-4363 FAX: (256) 536-9016 (256) 533-4363 http://www.resgen.com Research Instruments Kernick Road Pernryn Cornwall TR10 9DQ, UK (44) 1326-372-753 FAX: (44) 1326-378-783 http://www.research-instruments.com Research Organics 4353 E. 49th Street Cleveland, OH 44125 (800) 321-0570 FAX: (216) 883-1576 (216) 883-8025 http://www.resorg.com Research Plus P.O. Box 324 Bayonne, NJ 07002 (800) 341-2296 FAX: (201) 823-9590 (201) 823-3592 http://www.researchplus.com Research Products International 410 N. Business Center Drive Mount Prospect, IL 60056 (800) 323-9814 FAX: (847) 635-1177 (847) 635-7330 http://www.rpicorp.com Research Triangle Institute P.O. Box 12194 Research Triangle Park, NC 27709 (919) 541-6000 FAX: (919) 541-6515 http://www.rti.org
Restek 110 Benner Circle Bellefonte, PA 16823 (800) 356-1688 FAX: (814) 353-1309 (814) 353-1300 http://www.restekcorp.com Rheodyne P.O. Box 1909 Rohnert Park, CA 94927 (707) 588-2000 FAX: (707) 588-2020 http://www.rheodyne.com Rhone Merieux See Merial Limited Rhone-Poulenc 2 T W Alexander Drive P.O. Box 12014 Research Triangle Park, NC 08512 (919) 549-2000 FAX: (919) 549-2839 http://www.Rhone-Poulenc.com Also see Aventis Rhone-Poulenc Rorer 500 Arcola Road Collegeville, PA 19426 (800) 727-6737 FAX: (610) 454-8940 (610) 454-8975 http://www.rp-rorer.com Rhone-Poulenc Rorer Centre de Recherche de Vitry-Alfortville 13 Quai Jules Guesde, BP14 94403 Vitry Sur Seine, Cedex, France (33) 145-73-85-11 FAX: (33) 145-73-81-29 http://www.rp-rorer.com Ribi ImmunoChem Research 563 Old Corvallis Road Hamilton, MT 59840 (800) 548-7424 FAX: (406) 363-6129 (406) 363-3131 http://www.ribi.com RiboGene See Questcor Pharmaceuticals Ricca Chemical 448 West Fork Drive Arlington, TX 76012 (888) GO-RICCA FAX: (800) RICCA-93 (817) 461-5601 http://www.riccachemical.com Richard-Allan Scientific 225 Parsons Street Kalamazoo, MI 49007 (800) 522-7270 FAX: (616) 345-3577 (616) 344-2400 http://www.rallansci.com Richelieu Biotechnologies 11 177 Hamon Montral, Quebec H3M 3E4 Canada (802) 863-2567 FAX: (802) 862-2909 http://www.richelieubio.com
Richter Enterprises 20 Lake Shore Drive Wayland, MA 01778 (508) 655-7632 FAX: (508) 652-7264 http://www.richter-enterprises.com Riese Enterprises BioSure Division 12301 G Loma Rica Drive Grass Valley, CA 95945 (800) 345-2267 FAX: (916) 273-5097 (916) 273-5095 http://www.biosure.com Ritz Safety Equipment. Inc. 1851 North Powerline Road Pompano beach, FL 33069 (954) 971-3176 Fax: 954-971-1272 www.vandynecrotty.com/ first aid safety/index safetyproducts.asp Robbins Scientific 1250 Elko Drive Sunnyvale, CA 94086 (800) 752-8585 FAX: (408) 734-0300 (408) 734-8500 http://www.robsci.com Roboz Surgical Instruments 9210 Corporate Boulevard, Suite 220 Rockville, MD 20850 (800) 424-2984 FAX: (301) 590-1290 (301) 590-0055 Roche Diagnostics 9115 Hague Road P.O. Box 50457 Indianapolis, IN 46256 (800) 262-1640 FAX: (317) 845-7120 (317) 845-2000 http://www.roche.com Roche Molecular Systems See Roche Diagnostics Rocklabs P.O. Box 18-142 Auckland 6, New Zealand (64) 9-634-7696 FAX: (64) 9-634-7696 http://www.rocklabs.com Rockland P.O. Box 316 Gilbertsville, PA 19525 (800) 656-ROCK FAX: (610) 367-7825 (610) 369-1008 http://www.rockland-inc.com Rockwell Laser Industries P.O. Box 43010 7754 Camargo Road Cincinnati, OH 45243 (513) 271-1568 Fax: 513-271-1598 www.rli.com
Suppliers
28 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Rohm Chemische Fabrik Kirschenallee D-64293 Darmstadt, Germany (49) 6151-1801 FAX: (49) 6151-1802 http://www.roehm.com Roper Scientific 3440 East Brittania Drive, Suite 100 Tucson, AZ 85706 (520) 889-9933 FAX: (520) 573-1944 http://www.roperscientific.com Rosetta Inpharmatics 12040 115th Avenue NE Kirkland, WA 98034 (425) 820-8900 FAX: (425) 820-5757 http://www.rii.com ROTH-SOCHIEL 3 rue de la Chapelle Lauterbourg 67630 France (33) 03-88-94-82-42 FAX: (33) 03-88-54-63-93 Rotronic Instrument 160 E. Main Street Huntington, NY 11743 (631) 427-3898 FAX: (631) 427-3902 http://www.rotronic-usa.com Roundy’s 23000 Roundy Drive Pewaukee, WI 53072 (262) 953-7999 FAX: (262) 953-7989 http://www.roundys.com RS Components Birchington Road Weldon Industrial Estate Corby, Northants NN17 9RS, UK (44) 1536 201234 FAX: (44) 1536 405678 http://www.rs-components.com Rubbermaid See Newell Rubbermaid SA Instrumentation 1497 Tzena Way Encinitas, CA 92024 (858) 453-1776 FAX: (800) 268-1776 http://www.sainst.com Safe Cells See Bionique Testing Labs Sage Instruments 240 Airport Boulevard Freedom, CA 95076 831-761-1000 FAX: 831-761-1008 http://www.sageinst.com Sage Laboratories 11 Huron Drive Natick, MA 01760 (508) 653-0844 FAX: 508-653-5671 http://www.sagelabs.com
Saint-Gobain Performance Plastics P.O. Box 3660 Akron, OH 44309 (330) 798-9240 FAX: (330) 798-6968 http://www.nortonplastics.com San Diego Instruments 7758 Arjons Drive San Diego, CA 92126 (858) 530-2600 FAX: (858) 530-2646 http://www.sd-inst.com Sandown Scientific Beards Lodge 25 Oldfield Road Hampden, Middlesex TW12 2AJ, UK (44) 2089 793300 FAX: (44) 2089 793311 http://www.sandownsci.com Sandoz Pharmaceuticals See Novartis Sanofi Recherche Centre de Montpellier 371 Rue du Professor Blayac 34184 Montpellier, Cedex 04 France (33) 67-10-67-10 FAX: (33) 67-10-67-67 Sanofi Winthrop Pharmaceuticals 90 Park Avenue New York, NY 10016 (800) 223-5511 FAX: (800) 933-3243 (212) 551-4000 http://www.sanofi-synthelabo.com/us Santa Cruz Biotechnology 2161 Delaware Avenue Santa Cruz, CA 95060 (800) 457-3801 FAX: (831) 457-3801 (831) 457-3800 http://www.scbt.com
Savant/EC Apparatus A ThermoQuest company 100 Colin Drive Holbrook, NY 11741 (800) 634-8886 FAX: (516) 244-0606 (516) 244-2929 http://www.savec.com Savillex 6133 Baker Road Minnetonka, MN 55345 (612) 935-5427 Scanalytics Division of CSP 8550 Lee Highway, Suite 400 Fairfax, VA 22031 (800) 325-3110 FAX: (703) 208-1960 (703) 208-2230 http://www.scanalytics.com Schering Laboratories See Schering-Plough Schering-Plough 1 Giralda Farms Madison, NJ 07940 (800) 222-7579 FAX: (973) 822-7048 (973) 822-7000 http://www.schering-plough.com Schleicher & Schuell 10 Optical Avenue Keene, NH 03431 (800) 245-4024 FAX: (603) 357-3627 (603) 352-3810 http://www.s-und-s.de/englishindex.html
Sarasep (800) 605-0267 FAX: (408) 432-3231 (408) 432-3230 http://www.transgenomic.com
Science Technology Centre 1250 Herzberg Laboratories Carleton University 1125 Colonel Bay Drive Ottawa, Ontario, Canada K1S 5B6 (613) 520-4442 FAX: (613) 520-4445 http://www.carleton.ca/universities/stc
Sarstedt P.O. Box 468 Newton, NC 28658 (800) 257-5101 FAX: (828) 465-4003 (828) 465-4000 http://www.sarstedt.com
Scientific Instruments 200 Saw Mill River Road Hawthorne, NY 10532 (800) 431-1956 FAX: (914) 769-5473 (914) 769-5700 http://www.scientificinstruments.com
Sartorius 131 Heartsland Boulevard Edgewood, NY 11717 (800) 368-7178 FAX: (516) 254-4253 http://www.sartorius.com SAS Institute Pacific Telesis Center One Montgomery Street San Francisco, CA 94104 (415) 421-2227 FAX: (415) 421-1213 http://www.sas.com
Scientific Solutions 9323 Hamilton Mentor, OH 44060 (440) 357-1400 FAX: (440) 357-1416 www.labmaster.com Scion 82 Worman’s Mill Court, Suite H Frederick, MD 21701 (301) 695-7870 FAX: (301) 695-0035 www.scioncorp.com
Scott Specialty Gases 6141 Easton Road P.O. Box 310 Plumsteadville, PA 18949 (800) 21-SCOTT FAX: (215) 766-2476 (215) 766-8861 http://www.scottgas.com Scripps Clinic and Research Foundation Instrumentation and Design Lab 10666 N. Torrey Pines Road La Jolla, CA 92037 (800) 992-9962 FAX: (858) 554-8986 (858) 455-9100 http://www.scrippsclinic.com SDI Sensor Devices 407 Pilot Court, 400A Waukesha, WI 53188 (414) 524-1000 FAX: (414) 524-1009 Sefar America 111 Calumet Street Depew, NY 14043 (716) 683-4050 FAX: (716) 683-4053 http://www.sefaramerica.com Seikagaku America Division of Associates of Cape Cod 704 Main Street Falmouth, MA 02540 (800) 237-4512 FAX: (508) 540-8680 (508) 540-3444 http://www.seikagaku.com Sellas Medizinische Gerate Hagener Str. 393 Gevelsberg-Vogelsang, 58285 Germany (49) 23-326-1225 Sensor Medics 22705 Savi Ranch Parkway Yorba Linda, CA 92887 (800) 231-2466 FAX: (714) 283-8439 (714) 283-2228 http://www.sensormedics.com Sensor Systems LLC 2800 Anvil Street, North Saint Petersburg, FL 33710 (800) 688-2181 FAX: (727) 347-3881 (727) 347-2181 http://www.vsensors.com SenSym/Foxboro ICT 1804 McCarthy Boulevard Milpitas, CA 95035 (800) 392-9934 FAX: (408) 954-9458 (408) 954-6700 http://www.sensym.com Separations Group See Vydac Sepracor 111 Locke Drive Marlboro, MA 01752 (877)-SEPRACOR (508) 357-7300 http://www.sepracor.com
Suppliers
29 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Sera-Lab See Harlan Sera-Lab Sermeter 925 Seton Court, #7 Wheeling, IL 60090 (847) 537-4747 Serological 195 W. Birch Street Kankakee, IL 60901 (800) 227-9412 FAX: (815) 937-8285 (815) 937-8270 Seromed Biochrom Leonorenstrasse 2-6 D-12247 Berlin, Germany (49) 030-779-9060 Serotec 22 Bankside Station Approach Kidlington, Oxford OX5 1JE, UK (44) 1865-852722 FAX: (44) 1865-373899 In the US: (800) 265-7376 http://www.serotec.co.uk Serva Biochemicals Distributed by Crescent Chemical S.F. Medical Pharmlast See Chase-Walton Elastomers SGE 2007 Kramer Lane Austin, TX 78758 (800) 945-6154 FAX: (512) 836-9159 (512) 837-7190 http://www.sge.com Shandon/Lipshaw 171 Industry Drive Pittsburgh, PA 15275 (800) 245-6212 FAX: (412) 788-1138 (412) 788-1133 http://www.shandon.com Sharpoint P.O. Box 2212 Taichung, Taiwan Republic of China (886) 4-3206320 FAX: (886) 4-3289879 http://www.sharpoint.com.tw Shelton Scientific 230 Longhill Crossroads Shelton, CT 06484 (800) 222-2092 FAX: (203) 929-2175 (203) 929-8999 http://www.sheltonscientific.com Sherwood-Davis & Geck See Kendall Sherwood Medical See Kendall
Shimadzu Scientific Instruments 7102 Riverwood Drive Columbia, MD 21046 (800) 477-1227 FAX: (410) 381-1222 (410) 381-1227 http://www.ssi.shimadzu.com Sialomed See Amika Siemens Analytical X-Ray Systems See Bruker Analytical X-Ray Systems Sievers Instruments Subsidiary of Ionics 6060 Spine Road Boulder, CO 80301 (800) 255-6964 FAX: (303) 444-6272 (303) 444-2009 http://www.sieversinst.com SIFCO 970 East 46th Street Cleveland, OH 44103 (216) 881-8600 FAX: (216) 432-6281 http://www.silco.com Sigma-Aldrich 3050 Spruce Street St. Louis, MO 63103 (800) 358-5287 FAX: (800) 962-9591 (800) 325-3101 FAX: (800) 325-5052 http://www.sigma-aldrich.com Silenus/Amrad 34 Wadhurst Drive Boronia, Victoria 3155 Australia (613)9887-3909 FAX: (613)9887-3912 http://www.amrad.com.au Silicon Genetics 2601 Spring Street Redwood City, CA 94063 (866) SIG SOFT FAX: (650) 365 1735 (650) 367 9600 http://www.sigenetics.com SIMS Deltec 1265 Grey Fox Road St. Paul, Minnesota 55112 (800) 426-2448 FAX: (615) 628-7459 http://www.deltec.com SIMS Portex 10 Bowman Drive Keene, NH 03431 (800) 258-5361 FAX: (603) 352-3703 (603) 352-3812 http://www.simsmed.com SIMS Portex Limited Hythe, Kent CT21 6JL, UK (44)1303-260551 FAX: (44)1303-266761 http://www.portex.com
SLM-AMINCO Instruments See Spectronic Instruments
Spectramed See BOC Edwards
Small Parts 13980 NW 58th Court P.O. Box 4650 Miami Lakes, FL 33014 (800) 220-4242 FAX: (800) 423-9009 (305) 558-1038 FAX: (305) 558-0509 http://www.smallparts.com
SpectraSource Instruments 31324 Via Colinas, Suite 114 Westlake Village, CA 91362 (818) 707-2655 FAX: (818) 707-9035 http://www.spectrasource.com
Smith & Nephew 11775 Starkey Road P.O. Box 1970 Largo, FL 33779 (800) 876-1261 http://www.smith-nephew.com SmithKline Beecham 1 Franklin Plaza, #1800 Philadelphia, PA 19102 (215) 751-4000 FAX: (215) 751-4992 http://www.sb.com Solid Phase Sciences See Biosearch Technologies SOMA Scientific Instruments 5319 University Drive, PMB #366 Irvine, CA 92612 (949) 854-0220 FAX: (949) 854-0223 http://somascientific.com Somatix Therapy See Cell Genesys Sonics & Materials 53 Church Hill Road Newtown, CT 06470 (800) 745-1105 FAX: (203) 270-4610 (203) 270-4600 http://www.sonicsandmaterials.com Sonosep Biotech See Triton Environmental Consultants Sorvall See Kendro Laboratory Products Southern Biotechnology Associates P.O. Box 26221 Birmingham, AL 35260 (800) 722-2255 FAX: (205) 945-8768 (205) 945-1774 http://SouthernBiotech.com SPAFAS 190 Route 165 Preston, CT 06365 (800) SPAFAS-1 FAX: (860) 889-1991 (860) 889-1389 http://www.spafas.com
Skatron Instruments See Molecular Devices
Specialty Media Division of Cell & Molecular Technologies 580 Marshall Street Phillipsburg, NJ 08865 (800) 543-6029 FAX: (908) 387-1670 (908) 454-7774 http://www.specialtymedia.com
SLM Instruments See Spectronic Instruments
Spectra Physics See Thermo Separation Products
Siris Laboratories See Biosearch Technologies
Spectronic Instruments 820 Linden Avenue Rochester, NY 14625 (800) 654-9955 FAX: (716) 248-4014 (716) 248-4000 http://www.spectronic.com Spectrum Medical Industries See Spectrum Laboratories Spectrum Laboratories 18617 Broadwick Street Rancho Dominguez, CA 90220 (800) 634-3300 FAX: (800) 445-7330 (310) 885-4601 FAX: (310) 885-4666 http://www.spectrumlabs.com Spherotech 1840 Industrial Drive, Suite 270 Libertyville, IL 60048 (800) 368-0822 FAX: (847) 680-8927 (847) 680-8922 http://www.spherotech.com SPSS 233 S. Wacker Drive, 11th floor Chicago, IL 60606 (800) 521-1337 FAX: (800) 841-0064 http://www.spss.com SS White Burs 1145 Towbin Avenue Lakewood, NJ 08701 (732) 905-1100 FAX: (732) 905-0987 http://www.sswhiteburs.com Stag Instruments 16 Monument Industrial Park Chalgrove, Oxon OX44 7RW, UK (44) 1865-891116 FAX: (44) 1865-890562 Standard Reference Materials Program National Institute of Standards and Technology Building 202, Room 204 Gaithersburg, MD 20899 (301) 975-6776 FAX: (301) 948-3730 Starna Cells P.O. Box 1919 Atascandero, CA 93423 (805) 468-8855 FAX: (805) 461-1575 (800) 228-4482 http://www.stamacells.com Starplex Scientific 50 Steinway Etobieoke, Ontario M9W 6Y3 Canada (800) 665-0954 FAX: (416) 674-6067 (416) 674-7474 http://www.starplexscientific.com
Suppliers
30 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
State Laboratory Institute of Massachusetts 305 South Street Jamaica Plain, MA 02130 (617) 522-3700 FAX: (617) 522-8735 http://www.state.ma.us/dph Stedim Labs 1910 Mark Court, Suite 110 Concord, CA 94520 (800) 914-6644 FAX: (925) 689-6988 (925) 689-6650 http://www.stedim.com Steinel America 9051 Lyndate Avenue Bloomingron, MN 55420 (800) 852 4343 FAX: (952) 888-5132 http://www.stelnelamerica.com Stem Cell Technologies 777 West Broadway, Suite 808 Vancouver, British Columbia V5Z 4J7 Canada (800) 667-0322 FAX: (800) 567-2899 (604) 877-0713 FAX: (604) 877-0704 http://www.stemcell.com Stephens Scientific 107 Riverdale Road Riverdale, NJ 07457 (800) 831-8099 FAX: (201) 831-8009 (201) 831-9800
Stovall Lifescience 206-G South Westgate Drive Greensboro, NC 27407 (800) 852-0102 FAX: (336) 852-3507 http://www.slscience.com
Sunox 1111 Franklin Boulevard, Unit 6 Cambridge, ON N1R 8B5, Canada (519) 624-4413 FAX: (519) 624-8378 http://www.sunox.ca
Stratagene 11011 N. Torrey Pines Road La Jolla, CA 92037 (800) 424-5444 FAX: (888) 267-4010 (858) 535-5400 http://www.stratagene.com
Supelco See Sigma-Aldrich
Strategic Applications 530A N. Milwaukee Avenue Libertyville, IL 60048 (847) 680-9385 FAX: (847) 680-9837 Strem Chemicals 7 Mulliken Way Newburyport, MA 01950 (800) 647-8736 FAX: (800) 517-8736 (978) 462-3191 FAX: (978) 465-3104 http://www.strem.com StressGen Biotechnologies Biochemicals Division 120-4243 Glanford Avenue Victoria, British Columbia V8Z 4B9 Canada (800) 661-4978 FAX: (250) 744-2877 (250) 744-2811 http://www.stressgen.com
Steraloids P.O. Box 689 Newport, RI 02840 (401) 848-5422 FAX: (401) 848-5638 http://www.steraloids.com
Structure Probe/SPI Supplies (Epon-Araldite) P.O. Box 656 West Chester, PA 19381 (800) 242-4774 FAX: (610) 436-5755 http://www.2spi.com
Steris Corporation 5960 Heistey Road Mentor, Ohio 44060 (800) 548-4873 440-354-2600 http://www.steris.com
¨ Sud-Chemie Performance Packaging 101 Christine Drive Belen, NM 87002 (800) 989-3374 FAX: (505) 864-9296 http://www.uniteddesiccants.com
Sterling Medical 2091 Springdale Road, Ste. 2 Cherry Hill, NJ 08003 (800) 229-0900 FAX: (800) 229-7854 http://www.sterlingmedical.com
Sumitomo Chemical Sumitomo Building 5-33, Kitahama 4-chome Chuo-ku, Osaka 541-8550, Japan (81) 6-6220-3891 FAX: (81)-6-6220-3345 http://www.sumitomo-chem.co.jp
Sterling Winthrop 90 Park Avenue New York, NY 10016 (212) 907-2000 FAX: (212) 907-3626 Sternberger Monoclonals 10 Burwood Court Lutherville, MD 21093 (410) 821-8505 FAX: (410) 821-8506 http://www.sternbergermonoclonals. com Stoelting 502 Highway 67 Kiel, WI 53042 (920) 894-2293 FAX: (920) 894-7029 http://www.stoelting.com
Sun Box 19217 Orbit Drive Gaithersburg, MD 20879 (800) 548-3968 FAX: (301) 977-2281 (301) 869-5980 http://www.sunboxco.com Sunbrokers See Sun International Sun International 3700 Highway 421 North Wilmington, NC 28401 (800) LAB-VIAL FAX: (800) 231-7861 http://www.autosamplervial.com
SuperArray P.O. Box 34494 Bethesda, MD 20827 (888) 503-3187 FAX: (301) 765-9859 (301) 765-9888 http://www.superarray.com Surface Measurement Systems 3 Warple Mews, Warple Way London W3 ORF, UK (44) 20-8749-4900 FAX: (44) 20-8749-6749 http://www.smsuk.co.uk/index.htm SurgiVet N7 W22025 Johnson Road, Suite A Waukesha, WI 53186 (262) 513-8500 (888) 745-6562 FAX: (262) 513-9069 http://www.surgivet.com Sutter Instruments 51 Digital Drive Novato, CA 94949 (415) 883-0128 FAX: (415) 883-0572 http://www.sutter.com Swiss Precision Instruments 1555 Mittel Boulevard, Suite F Wooddale, IL 60191 (800) 221-0198 FAX: (800) 842-5164 Synaptosoft 3098 Anderson Place Decatur, GA 30033 (770) 939-4366 FAX: 770-939-9478 http://www.synaptosoft.com SynChrom See Micra Scientific Synergy Software 2457 Perkiomen Avenue Reading, PA 19606 (800) 876-8376 FAX: (610) 370-0548 (610) 779-0522 http://www.synergy.com Synteni See Incyte Synthetics Industry Lumite Division 2100A Atlantic Highway Gainesville, GA 30501 (404) 532-9756 FAX: (404) 531-1347 Systat See SPSS
Systems Planning and Analysis (SPA) 2000 N. Beauregard Street Suite 400 Alexandria, VA 22311 (703) 931-3500 http://www.spa-inc.net 3M Bioapplications 3M Center Building 270-15-01 St. Paul, MN 55144 (800) 257-7459 FAX: (651) 737-5645 (651) 736-4946 T Cell Diagnostics and T Cell Sciences 38 Sidney Street Cambridge, MA 02139 (617) 621-1400 TAAB Laboratory Equipment 3 Minerva House Calleva Park Aldermaston, Berkshire RG7 8NA, UK (44) 118 9817775 FAX: (44) 118 9817881 Taconic 273 Hover Avenue Germantown, NY 12526 (800) TAC-ONIC FAX: (518) 537-7287 (518) 537-6208 http://www.taconic.com Tago See Biosource International TaKaRa Biochemical 719 Alliston Way Berkeley, CA 94710 (800) 544-9899 FAX: (510) 649-8933 (510) 649-9895 http://www.takara.co.jp/english Takara Shuzo Biomedical Group Division Seta 3-4-1 Otsu Shiga 520-21, Japan (81) 75-241-5100 FAX: (81) 77-543-9254 http://www.Takara.co.jp/english Takeda Chemical Products 101 Takeda Drive Wilmington, NC 28401 (800) 825-3328 FAX: (800) 825-0333 (910) 762-8666 FAX: (910) 762-6846 http://takeda-usa.com TAO Biomedical 73 Manassas Court Laurel Springs, NJ 08021 (609) 782-8622 FAX: (609) 782-8622 Tecan US P.O. Box 13953 Research Triangle Park, NC 27709 (800) 33-TECAN FAX: (919) 361-5201 (919) 361-5208 http://www.tecan-us.com
Suppliers
31 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Techne University Park Plaza 743 Alexander Road Princeton, NJ 08540 (800) 225-9243 FAX: (609) 987-8177 (609) 452-9275 http://www.techneusa.com Technical Manufacturing 15 Centennial Drive Peabody, MA 01960 (978) 532-6330 FAX: (978) 531-8682 http://www.techmfg.com Technical Products International 5918 Evergreen St. Louis, MO 63134 (800) 729-4451 FAX: (314) 522-6360 (314) 522-8671 http://www.vibratome.com Technicon See Organon Teknika Cappel Techno-Aide P.O. Box 90763 Nashville, TN 37209 (800) 251-2629 FAX: (800) 554-6275 (615) 350-7030 http://www.techno-aid.com Ted Pella 4595 Mountain Lakes Boulevard P.O. Box 492477 Redding, CA 96049 (800) 237-3526 FAX: (530) 243-3761 (530) 243-2200 http://www.tedpella.com Tekmar-Dohrmann P.O. Box 429576 Cincinnati, OH 45242 (800) 543-4461 FAX: (800) 841-5262 (513) 247-7000 FAX: (513) 247-7050 Tektronix 142000 S.W. Karl Braun Drive Beaverton, OR 97077 (800) 621-1966 FAX: (503) 627-7995 (503) 627-7999 http://www.tek.com Tel-Test P.O. Box 1421 Friendswood, TX 77546 (800) 631-0600 FAX: (281)482-1070 (281)482-2672 http://www.isotex-diag.com TeleChem International 524 East Weddell Drive, Suite 3 Sunnyvale, CA 94089 (408) 744-1331 FAX: (408) 744-1711 http://www.gst.net/∼telechem Terrachem Mallaustrasse 57 D-68219 Mannheim, Germany 0621-876797-0 FAX: 0621-876797-19 http://www.terrachem.de
Terumo Medical 2101 Cottontail Lane Somerset, NJ 08873 (800) 283-7866 FAX: (732) 302-3083 (732) 302-4900 http://www.terumomedical.com Tetko 333 South Highland Manor Briarcliff, NY 10510 (800) 289-8385 FAX: (914) 941-1017 (914) 941-7767 http://www.tetko.com TetraLink 4240 Ridge Lea Road Suite 29 Amherst, NY 14226 (800) 747-5170 FAX: (800) 747-5171 http://www.tetra-link.com TEVA Pharmaceuticals USA 1090 Horsham Road P.O. Box 1090 North Wales, PA 19454 (215) 591-3000 FAX: (215) 721-9669 http://www.tevapharmusa.com Texas Fluorescence Labs 9503 Capitol View Drive Austin, TX 78747 (512) 280-5223 FAX: (512) 280-4997 http://www.teflabs.com The Nest Group 45 Valley Road Southborough, MA 01772 (800) 347-6378 FAX: (508) 485-5736 (508) 481-6223 http://world.std.com/∼nestgrp ThermoCare P.O. Box 6069 Incline Village, NV 89450 (800) 262-4020 (775) 831-1201 Thermo Labsystems 8 East Forge Parkway Franklin, MA 02038 (800) 522-7763 FAX: (508) 520-2229 (508) 520-0009 http://www.finnpipette.com Thermometric Spjutvagen 5A S-175 61 Jarfalla, Sweden (46) 8-564-72-200 Thermoquest IEC Division 300 Second Avenue Needham Heights, MA 02194 (800) 843-1113 FAX: (781) 444-6743 (781) 449-0800 http://www.thermoquest.com
Thermo Separation Products Thermoquest 355 River Oaks Parkway San Jose, CA 95134 (800) 538-7067 FAX: (408) 526-9810 (408) 526-1100 http://www.thermoquest.com
Tissue-Tek A Division of Sakura Finetek USA 1750 West 214th Street Torrance, CA 90501 (800) 725-8723 FAX: (310) 972-7888 (310) 972-7800 http://www.sakuraus.com
Thermo Shandon 171 Industry Drive Pittsburgh, PA 15275 (800) 547-7429 FAX: (412) 899-4045 http://www.thermoshandon.com
Tocris Cookson 114 Holloway Road, Suite 200 Ballwin, MO 63011 (800) 421-3701 FAX: (800) 483-1993 (636) 207-7651 FAX: (636) 207-7683 http://www.tocris.com
Thertno Spectronic 820 Linden Avenue Rochester, NY 14625 (585) 248-4000 FAX: (585) 248-4200 http://www.thermo.com Thomas Scientific 99 High Hill Road at I-295 Swedesboro, NJ 08085 (800) 345-2100 FAX: (800) 345-5232 (856) 467-2000 FAX: (856) 467-3087 http://www.wheatonsci.com/html/nt/ Thomas.html Thomson Instrument 354 Tyler Road Clearbrook, VA 22624 (800) 842-4752 FAX: (540) 667-6878 (800) 541-4792 FAX: (760) 757-9367 http://www.hplc.com Thorn EMI See Electron Tubes
Tocris Cookson Northpoint, Fourth Way Avonmouth, Bristol BS11 8TA, UK (44) 117-982-6551 FAX: (44) 117-982-6552 http://www.tocris.com Tomtec See CraMar Technologies TopoGen P.O. Box 20607 Columbus, OH 43220 (800) TOPOGEN FAX: (800) ADD-TOPO (614) 451-5810 FAX: (614) 451-5811 http://www.topogen.com Toray Industries, Japan Toray Building 2-1 Nihonbash-Muromach 2-Chome, Chuo-Ku Tokyo, Japan 103-8666 (03) 3245-5115 FAX: (03) 3245-5555 http://www.toray.co.jp
Thorlabs 435 Route 206 Newton, NJ 07860 (973) 579-7227 FAX: (973) 383-8406 http://www.thorlabs.com
Toray Industries, U.S.A. 600 Third Avenue New York, NY 10016 (212) 697-8150 FAX: (212) 972-4279 http://www.toray.com
Tiemann See Bernsco Surgical Supply
Toronto Research Chemicals 2 Brisbane Road North York, Ontario M3J 2J8, Canada (416) 665-9696 FAX: (416) 665-4439 http://www.trc-canada.com
TILL Photonics GmbH Lochhamer Schlag 19 D-82166 Graleling Germany (49) 89-895-662-0 FAX: (49) 89-895-662-101 http://www.till-photonics.com/ Timberline Instruments 1880 South Flatiron Court, H-2 P.O. Box 20356 Boulder, CO 80308 (800) 777-5996 FAX: (303) 440-8786 (303) 440-8779 http://www.timberlineinstruments.com Tissuelnformatics 711 Bingham Street, Suite 202 Pittsburgh, PA 15203 (418) 488-1100 FAX: (418) 488-6172 http://www.tissuelnformatics.com
TosoHaas 156 Keystone Drive Montgomeryville, PA 18036 (800) 366-4875 FAX: (215) 283-5035 (215) 283-5000 http://www.tosohaas.com Towhill 647 Summer Street Boston, MA 02210 (617) 542-6636 FAX: (617) 464-0804 Toxin Technology 7165 Curtiss Avenue Sarasota, FL 34231 (941) 925-2032 FAX: (9413) 925-2130 http://www.toxintechnology.com Toyo Soda See TosoHaas
Suppliers
32 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Trace Analytical 3517-A Edison Way Menlo Park, CA 94025 (650) 364-6895 FAX: (650) 364-6897 http://www.traceanalytical.com
Tyler Research 10328 73rd Avenue Edmonton, Alberta T6E 6N5 Canada (403) 448-1249 FAX: (403) 433-0479
Transduction Laboratories See BD Transduction Laboratories
UBI See Upstate Biotechnology
Transgenomic 2032 Concourse Drive San Jose, CA 95131 (408) 432-3230 FAX: (408) 432-3231 http://www.transgenomic.com Transonic Systems 34 Dutch Mill Road Ithaca, NY 14850 (800) 353-3569 FAX: (607) 257-7256 http://www.transonic.com Travenol Lab See Baxter Healthcare Tree Star Software 20 Winding Way San Carlos, CA 94070 800-366-6045 http://www.treestar.com Trevigen 8405 Helgerman Court Gaithersburg, MD 20877 (800) TREVIGEN FAX: (301) 216-2801 (301) 216-2800 http://www.trevigen.com Trilink Biotechnologies 6310 Nancy Ridge Drive San Diego, CA 92121 (800) 863-6801 FAX: (858) 546-0020 http://www.trilink.biotech.com Tripos Associates 1699 South Hanley Road, Suite 303 St. Louis, MO 63144 (800) 323-2960 FAX: (314) 647-9241 (314) 647-1099 http://www.tripos.com Triton Environmental Consultants 120-13511 Commerce Parkway Richmond, British Columbia V6V 2L1 Canada (604) 279-2093 FAX: (604) 279-2047 http://www.triton-env.com Tropix 47 Wiggins Avenue Bedford, MA 01730 (800) 542-2369 FAX: (617) 275-8581 (617) 271-0045 http://www.tropix.com
Ugo Basile Biological Research Apparatus Via G. Borghi 43 21025 Comerio, Varese, Italy (39) 332 744 574 FAX: (39) 332 745 488 http://www.ugobasile.com UltraPIX See Life Science Resources Ultrasonic Power 239 East Stephenson Street Freeport, IL 61032 (815) 235-6020 FAX: (815) 232-2150 http://www.upcorp.com Ultrasound Advice 23 Aberdeen Road London N52UG, UK (44) 020-7359-1718 FAX: (44) 020-7359-3650 http://www.ultrasoundadvice.co.uk UNELKO 14641 N. 74th Street Scottsdale, AZ 85260 (480) 991-7272 FAX: (480)483-7674 http://www.unelko.com Unifab Corp. 5260 Lovers Lane Kalamazoo, MI 49002 (800) 648-9569 FAX: (616) 382-2825 (616) 382-2803 Union Carbide 10235 West Little York Road, Suite 300 Houston, TX 77040 (800) 568-4000 FAX: (713) 849-7021 (713) 849-7000 http://www.unioncarbide.com United Desiccants ¨ See Sud-Chemie Performance Packaging United States Biochemical See USB
Upchurch Scientific 619 West Oak Street P.O. Box 1529 Oak Harbor, WA 98277 (800) 426-0191 FAX: (800) 359-3460 (360) 679-2528 FAX: (360) 679-3830 http://www.upchurch.com Upjohn Pharmacia & Upjohn http://www.pnu.com Upstate Biotechnology (UBI) 1100 Winter Street, Suite 2300 Waltham, MA 02451 (800) 233-3991 FAX: (781) 890-7738 (781) 890-8845 http://www.upstatebiotech.com
Valco Instruments P.O. Box 55603 Houston, TX 77255 (800) FOR-VICI FAX: (713) 688-8106 (713) 688-9345 http://www.vici.com Valpey Fisher 75 South Street Hopkin, MA 01748 (508) 435-6831 FAX: (508) 435-5289 http://www.valpeyfisher.com Value Plastics 3325 Timberline Road Fort Collins, CO 80525 (800) 404-LUER FAX: (970) 223-0953 (970) 223-8306 http://www.valueplastics.com
USA/Scientific 346 SW 57th Avenue P.O. Box 3565 Ocala, FL 34478 (800) LAB-TIPS FAX: (352) 351-2057 (3524) 237-6288 http://www.usascientific.com
Vangard International P.O. Box 308 3535 Rt. 66, Bldg. #4 Neptune, NJ 07754 (800) 922-0784 FAX: (732) 922-0557 (732) 922-4900 http://www.vangard1.com
USB 26111 Miles Road P.O. Box 22400 Cleveland, OH 44122 (800) 321-9322 FAX: (800) 535-0898 FAX: (216) 464-5075 http://www.usbweb.com
Varian Analytical Instruments 2700 Mitchell Drive Walnut Creek, CA 94598 (800) 926-3000 FAX: (925) 945-2102 (925) 939-2400 http://www.varianinc.com
USCI Bard Bard Interventional Products 129 Concord Road Billerica, MA 01821 (800) 225-1332 FAX: (978) 262-4805 http://www.bardinterventional.com Uvex Safety 10 Thurber Blvd. Smithiiled, RI 02917 (800) 434-3411 Fax: 401-232-1830 http://www.uvex.com UVP (Ultraviolet Products) 2066 W. 11th Street Upland, CA 91786 (800) 452-6788 FAX: (909) 946-3597 (909) 946-3197 http://www.uvp.com
TSI Center for Diagnostic Products See Intergen
United States Biological (US Biological) P.O. Box 261 Swampscott, MA 01907 (800) 520-3011 FAX: (781) 639-1768 http://www.usbio.net
V & P Scientific 9823 Pacific Heights Boulevard, Suite T San Diego, CA 92121 (800) 455-0644 FAX: (858) 455-0703 (858) 455-0643 http://www.vp-scientific.com
2000 Eppendorf-5 Prime 5603 Arapahoe Avenue Boulder, CO 80303 (800) 533-5703 FAX: (303) 440-0835 (303) 440-3705
Universal Imaging 502 Brandywine Parkway West Chester, PA 19380 (610) 344-9410 FAX: (610) 344-6515 http://www.image1.com
V-A Optical Labs 60 Red Hill Ave. San Anselmo, CA 94960 (415) 459-1919 FAX: (415) 459-7216 http://www.vaoptical.com
Varian Associates 3050 Hansen Way Palo Alto, CA 94304 (800) 544-4636 FAX: (650) 424-5358 (650) 493-4000 http://www.varian.com Vector Core Laboratory/National Gene Vector Labs University of Michigan 3560 E MSRB II 1150 West Medical Center Drive Ann Arbor, MI 48109 (734) 936-5843 FAX: (734) 764-3596 Vector Laboratories 30 Ingold Road Burlingame, CA 94010 (800) 227-6666 FAX: (650) 697-0339 (650) 697-3600 http://www.vectorlabs.com Vedco 2121 S.E. Bush Road St. Joseph, MO 64504 (888) 708-3326 FAX: (816) 238-1837 (816) 238-8840 http://database.vedco.com Ventana Medical Systems 3865 North Business Center Drive Tucson, AZ 85705 (800) 227-2155 FAX: (520) 887-2558 (520) 887-2155 http://www.ventanamed.com
Suppliers
33 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29
Verity Software House P.O. Box 247 45A Augusta Road Topsham, ME 04086 (207) 729-6767 FAX: (207) 729-5443 http://www.vsh.com Vernitron See Sensor Systems LLC Vertex Pharmaceuticals 130 Waverly Street Cambridge, MA 02139 (617) 577-6000 FAX: (617) 577-6680 http://www.vpharm.com Vetamac Route 7, Box 208 Frankfort, IN 46041 (317) 379-3621 Vet Drug Unit 8 Lakeside Industrial Estate Colnbrook, Slough SL3 0ED, UK VetEquip, Inc. P.O. Box 10785 Pleasanton, CA 94588 (800)-466-6463 Fax: 925-463-1943 Vetus Animal Health See Burns Veterinary Supply Viamed 15 Station Road Cross Hills, Keighley W. Yorkshire BD20 7DT, UK (44) 1-535-634-542 FAX: (44) 1-535-635-582 http://www.viamed.co.uk Vical 9373 Town Center Drive, Suite 100 San Diego, CA 92121 (858) 646-1100 FAX: (858) 646-1150 http://www.vical.com Victor Medical 2349 North Watney Way, Suite D Fairfield, CA 94533 (800) 888-8908 FAX: (707) 425-6459 (707) 425-0294 Virion Systems 9610 Medical Center Drive, Suite 100 Rockville, MD 20850 (301) 309-1844 FAX: (301) 309-0471 http://www.radix.net/∼virion VirTis Company 815 Route 208 Gardiner, NY 12525 (800) 765-6198 FAX: (914) 255-5338 (914) 255-5000 http://www.virtis.com Visible Genetics 700 Bay Street, Suite 1000 Toronto, Ontario M5G 1Z6, Canada (888) 463-6844 (416) 813-3272 http://www.visgen.com
Vitrocom 8 Morris Avenue Mountain Lakes, NJ 07046 (973) 402-1443 FAX: (973) 402-1445 VTI 7650 W. 26th Avenue Hialeah, FL 33106 (305) 828-4700 FAX: (305) 828-0299 http://www.vticorp.com VWR Scientific Products 200 Center Square Road Bridgeport, NJ 08014 (800) 932-5000 FAX: (609) 467-5499 (609) 467-2600 http://www.vwrsp.com Vydac 17434 Mojave Street P.O. Box 867 Hesperia, CA 92345 (800) 247-0924 FAX: (760) 244-1984 (760) 244-6107 http://www.vydac.com Vysis 3100 Woodcreek Drive Downers Grove, IL 60515 (800) 553-7042 FAX: (630) 271-7138 (630) 271-7000 http://www.vysis.com ¨ W&H Dentalwerk Burmoos P.O. Box 1 ¨ A-5111 Burmoos, Austria (43) 6274-6236-0 FAX: (43) 6274-6236-55 http://www.wnhdent.com Wako BioProducts See Wako Chemicals USA Wako Chemicals USA 1600 Bellwood Road Richmond, VA 23237 (800) 992-9256 FAX: (804) 271-7791 (804) 271-7677 http://www.wakousa.com Wako Pure Chemicals 1-2, Doshomachi 3-chome Chuo-ku, Osaka 540-8605, Japan 81-6-6203-3741 FAX: 81-6-6222-1203 http://www.wako-chem.co.jp/egaiyo/ index.htm Wallac See Perkin-Elmer Wallac A Division of Perkin-Elmer 3985 Eastern Road Norton, OH 44203 (800) 321-9632 FAX: (330) 825-8520 (330) 825-4525 http://www.wallac.com Waring Products 283 Main Street New Hartford, CT 06057 (800) 348-7195 FAX: (860) 738-9203 (860) 379-0731 http://www.waringproducts.com
Warner Instrument 1141 Dixwell Avenue Hamden, CT 06514 (800) 599-4203 FAX: (203) 776-1278 (203) 776-0664 http://www.warnerinstrument.com Warner-Lambert Parke-Davis 201 Tabor Road Morris Plains, NJ 07950 (973) 540-2000 FAX: (973) 540-3761 http://www.warner-lambert.com Washington University Machine Shop 615 South Taylor St. Louis, MO 63310 (314) 362-6186 FAX: (314) 362-6184 Waters Chromatography 34 Maple Street Milford, MA 01757 (800) 252-HPLC FAX: (508) 478-1990 (508) 478-2000 http://www.waters.com Watlow 12001 Lackland Road St. Louis, MO 63146 (314) 426-7431 FAX: (314) 447-8770 http://www.watlow.com Watson-Marlow 220 Ballardvale Street Wilmington, MA 01887 (978) 658-6168 FAX: (978) 988 0828 http://www.watson-marlow.co.uk Waukesha Fluid Handling 611 Sugar Creek Road Delavan, WI 53115 (800) 252-5200 FAX: (800) 252-5012 (414) 728-1900 FAX: (414) 728-4608 http://www.waukesha-cb.com WaveMetrics P.O. Box 2088 Lake Oswego, OR 97035 (503) 620-3001 FAX: (503) 620-6754 http://www.wavemetrics.com Weather Measure P.O. Box 41257 Sacramento, CA 95641 (916) 481-7565 Weber Scientific 2732 Kuser Road Hamilton, NJ 08691 (800) FAT-TEST FAX: (609) 584-8388 (609) 584-7677 http://www.weberscientific.com Weck, Edward & Company 1 Weck Drive Research Triangle Park, NC 27709 (919) 544-8000 Wellcome Diagnostics See Burroughs Wellcome
Wellington Laboratories 398 Laird Road, Guelph Ontario, N1G 3X7, Canada (800) 578-6985 FAX: (519) 822-2849 http://www.well-labs.com Wesbart Engineering Daux Road Billingshurst, West Sussex RH14 9EZ, UK (44) 1-403-782738 FAX: (44) 1-403-784180 http://www.wesbart.co.uk Whatman 9 Bridewell Place Clifton, NJ 07014 (800) 631-7290 FAX: (973) 773-3991 (973) 773-5800 http://www.whatman.com Wheaton Science Products 1501 North 10th Street Millville, NJ 08332 (800) 225-1437 FAX: (800) 368-3108 (856) 825-1100 FAX: (856) 825-1368 http://www.algroupwheaton.com Whittaker Bioproducts See BioWhittaker Wild Heerbrugg Juerg Dedual Gaebrisstrasse 8 CH 9056 Gais, Switzerland (41) 71-793-2723 FAX: (41) 71-726-5957 http://www.homepage.swissonline.net/ dedual/wild heerbrugg Willy A. Bachofen AG Maschinenfabrik Utengasse 15/17 CH4005 Basel, Switzerland (41) 61-681-5151 FAX: (41) 61-681-5058 http://www.wab.ch Winthrop See Sterling Winthrop Wolfram Research 100 Trade Center Drive Champaign, IL 61820 (800) 965-3726 FAX: (217) 398-0747 (217) 398-0700 http://www.wolfram.com World Health Organization Microbiology and Immunology Support 20 Avenue Appia 1211 Geneva 27, Switzerland (41-22) 791-2602 FAX: (41-22) 791-0746 http://www.who.org World Precision Instruments 175 Sarasota Center Boulevard International Trade Center Sarasota, FL 34240 (941) 371-1003 FAX: (941) 377-5428 http://www.wpiinc.com
Suppliers
34 CPCY Supplement 29
Current Protocols Selected Suppliers of Reagents and Equipment
Worthington Biochemical Halls Mill Road Freehold, NJ 07728 (800) 445-9603 FAX: (800) 368-3108 (732) 462-3838 FAX: (732) 308-4453 http://www.worthington-biochem.com
Xeragon 19300 Germantown Road Germantown, MD 20874 (240) 686-7860 FAX: (240)686-7861 http://www.xeragon.com
WPI See World Precision Instruments
Xillix Technologies 300-13775 Commerce Parkway Richmond, British Columbia V6V 2V4 Canada (800) 665-2236 FAX: (604) 278-3356 (604) 278-5000 http://www.xillix.com
Wyeth-Ayerst 2 Esterbrook Lane Cherry Hill, NJ 08003 (800) 568-9938 FAX: (858) 424-8747 (858) 424-3700 Wyeth-Ayerst Laboratories P.O. Box 1773 Paoli, PA 19301 (800) 666-7248 FAX: (610) 889-9669 (610) 644-8000 http://www.ahp.com
Xomed Surgical Products 6743 Southpoint Drive N Jacksonville, FL 32216 (800) 874-5797 FAX: (800) 678-3995 (904) 296-9600 FAX: (904) 296-9666 http://www.xomed.com
Xenotech 3800 Cambridge Street Kansas City, KS 66103 (913) 588-7930 FAX: (913) 588-7572 http://www.xenotechllc.com
Yakult Honsha 1-19, Higashi-Shinbashi 1-chome Minato-ku Tokyo 105-8660, Japan 81-3-3574-8960
Yamasa Shoyu 23-8 Nihonbashi Kakigaracho 1-chome, Chuoku Tokyo, 103 Japan (81) 3-479 22 0095 FAX: (81) 3-479 22 3435 Yeast Genetic Stock Center See ATCC Yellow Spring Instruments See YSI YMC YMC Karasuma-Gojo Building 284 Daigo-Cho, Karasuma Nisihiirr Gojo-dori Shimogyo-ku Kyoto, 600-8106, Japan (81) 75-342-4567 FAX: (81) 75-342-4568 http://www.ymc.co.jp YSI 1725-1700 Brannum Lane Yellow Springs, OH 45387 (800) 765-9744 FAX: (937) 767-9353 (937) 767-7241 http://www.ysi.com Zeneca/CRB See AstraZeneca (800) 327-0125 FAX: (800) 321-4745
Zivic-Miller Laboratories 178 Toll Gate Road Zelienople, PA 16063 (800) 422-LABS FAX: (724) 452-4506 (800) MBM-RATS FAX: (724) 452-5200 http://zivicmiller.com Zymark Zymark Center Hopkinton, MA 01748 (508) 435-9500 FAX: (508) 435-3439 http://www.zymark.com Zymed Laboratories 458 Carlton Court South San Francisco, CA 94080 (800) 874-4494 FAX: (650) 871-4499 (650) 871-4494 http://www.zymed.com Zymo Research 625 W. Katella Avenue, Suite 30 Orange, CA 92867 (888) 882-9682 FAX: (714) 288-9643 (714) 288-9682 http://www.zymor.com Zynaxis Cell Science See ChiRex Cauldron
Suppliers
35 Current Protocols Selected Suppliers of Reagents and Equipment
CPCY Supplement 29